Professional Documents
Culture Documents
handbook
seventeenth edition
William I. Orr, W6SAI
radio
handbook
seventeenth edition
www.americanradiohistory.com
SEVENTEENTH EDITION
FIRST PRINTING
SECOND PRINTING
-- 1967
1968
THIRD PRINTING -1969
RADIO HANDBOOK
www.americanradiohistory.com
Preface
Vast changes have taken place in the Amateur Radio Service throughout the past
two decades. The amateur station has literally moved from the garage or home work-
shop into the living room, and ham gear has forsaken the black -crackle panel and the
six -foot relay rack for the streamlined, sleek, miniaturized desk -top cabinet. This
progression has been brought about by the sweeping change from amplitude modu-
lation (a -m) to single -sideband transmission, heralding a whole new field of technical
advances in the communication art. Bandswitching linear amplifiers, compact solid -
state power supplies, and highly stable variable-frequency oscillators (all of which
were practically unknown twenty years ago) are modern counterparts of the bulky
plug -in coil class-C amplifier, the cumbersome modulators, and the weighty power
supplies that identified the amateur station of the late "forties."
The gradual eclipse of amplitude modulation has also been stimulated by the advent
of the SSB transceiver and its unique VOX -operated break-in ability to make use of
a single communications channel for local and long- distance contacts. In addition,
the elimination of the interstation heterodyne and selective a -m fading by the wide-
spread use of SSB has permitted more efficient occupancy of the high- frequency
amateur bands by double the number of stations compared to twenty years ago.
Today's radio equipment bears little resemblance to the rough- and -ready ham gear
of the pre -TVI, pre -SSB era of the relay rack and the breadboard. Today's radio
amateur, moreover, is a more proficient, sophisticated operator than his counterpart
of twenty or thirty years ago. The horizons of the Amateur Radio Service have been
greatly expanded as a result of this worthwhile revolution in communication tech-
niques and practices. It is hoped that this trend will be evident in the years to come.
The author is pleased to note that the RADIO HANDBOOK has been a force in
advancing the state of the art of these various and diversified radio amateur develop-
ments, many of which are reflected in this new edition of the handbook.
Over thirty years ago the historic first edition of the RADIO HANDBOOK was
published as a unique, independent communications manual written especially for the
advanced radio amateur and electronics engineer. Since that early time, each succeed-
ing edition of the RADIO HANDBOOK has -led the rapidly advancing field of
communications electronics. This new seventeenth edition typifies the modern trend
in amateur radio today toward more advanced and sophisticated communication
techniques and equipment.
The preparation of this edition of the RADIO HANDBOOK would have been
impossible without the help that was tended the author by fellow radio amateurs and
sympathetic electronics organizations. To those individuals and companies whose
unselfish support made the compilation and publication of this Handbook an
interesting and inspired task, I extend my thanks.
WILLIAM I. ORR, W6SAI
Acknowledgments
www.americanradiohistory.com
Contents
Chapter One. INTRODUCTION TO RADIO 1
1 -1 Amateur Radio 1
2 -1 The Atom 3
2 -2 Fundamental Electrical Units and Relationships 24
2 -3 Electrostatics- Capacitors 33
2 -4 Magnetism and Electromagnetism 39
2 -5 RC and RL Transients 3
Chapter
ter Three. ALTERNATING -CURRENT CIRCUITS 46
3 -1 Alternating Current 46
3 -2 Resonant Circuits 60
3 -3 Nonsinusoidal Waves and Transients 5
3 -4 Transformers 69
3 -5 Electric Filters 1
4 -1 Thermionic Emission 75
4 -2 The Diode 80
4 -3 The Triode 81
4 -4 Tetrode or Screen -Grid Tubes 86
4 -5 Mixer and Converter Tubes 89
4 -6 Electron Tubes at Very -High Frequencies 90
4 -7 Special Microwave Electron Tubes 91
4 -8 The Cathode -Ray Tube 94
4 -9 Gas Tubes 97
4 -10 Miscellaneous Tube Types 99
www.americanradiohistory.com
6 -3 Biasing Methods 123
6 -4 Distortion in Amplifiers 123
6 -5 Resistance -Capacitance Coupled Audio -
Frequency Amplifiers 125
6 -6 Video -Frequency Amplifiers 127
6 -7 Other Interstage Coupling Methods 128
6 -8 Phase Inverters 130
6-9 D -C Amplifiers 132
6 -10 Single -Ended Triode Amplifiers 133
6 -11 Single -Ended Pentode Amplifiers 135
6 -12 Push -Pull Audio Amplifiers 137
6 -13 Class -B Audio -Frequency Power Amplifiers 138
6 -14 Cathode- Follower Power Amplifiers 143
6 -15 Feedback Amplifiers 145
www.americanradiohistory.com
10 -5 R -F Stages 201
10 -6 Signal- Frequency Tuned Circuits 203
10 -7 I -F Tuned Circuits 206
10 -8 Detector, Audio, and Control Circuits 213
10 -9 Noise Suppression 216
10 -10 Special Considerations in UHF Receiver Design 220
10 -11 Receiver Adjustment 224
10 -12 Receiving Accessories 226
www.americanradiohistory.com
Chapter Fifteen. RADIO TELETYPE SYSTEMS 325
15 -1 Radioteletype Systems 325
15 -2 RTTY Reception 326
15 -3 Frequency -Shift Keying 328
17 -1 Resistors 358
17 -2 Capacitors 360
17 -3 Wire and Inductors 362
17 -4 Grounds 364
17 -5 Holes, Leads, and Shafts 365
17 -6 Parasitic Resonances 366
17 -7 Parasitic Oscillation in R -F Amplifiers 367
17 -8 Elimination of VHF Parasitic Oscillations 369
17 -9 Checking for Parasitic Oscillations 372
www.americanradiohistory.com
20 -4 Antenna Directivity 418
20 -5 Bandwidth 421
20 -6 Propagation of Radio Waves 421
20 -7 Ground -Wave Communication 422
20 -8 Ionospheric Propagation 425
20 -9 Transmission Lines 429
20 -10 Nonresonant Transmission Lines 430
20 -11 Tuned or Resonant Lines 433
20 -12 Line Discontinuities 434
20 -13 A Broadband 50 -Ohm Balun 435
www.americanradiohistory.com
24 -4 Unidirectional Driven Arrays 523
24 -5 Bidirectional Rotatable Arrays 525
24 -6 Construction of Rotatable Arrays 525
24 -7 Tuning the Array 528
24 -8 Indication of Direction 530
24 -9 Three -Band Beams 531
www.americanradiohistory.com
29 -3 General- Purpose Triode Class -B Modulator 702
29 -4 A 15 -Watt Clipper -Amplifier 704
29 -5 Zero -Bias Tetrode Modulators 706
www.americanradiohistory.com
CHAPTER ONE
Introduction to Radio
www.americanradiohistory.com
HANDBOOK Amateur Bands 13
ileges, is valid for a period of only one year been made. As a usual rule, however, 160 -
(as contrasted to all other classes of amateur meter amateur operation is confined to
licenses which run for a term of five years), ground -wave contacts or single -skip con-
and is not renewable. tacts of 1000 miles or less. Popular before
All Novice- and Technician -Class ex- World War II, the 160 -meter band is now
aminations are given by volunteer examin- only sparsely occupied since many areas of
ers, as regular examinations for these two the country are blanketed by the megawatt
classes are not given in FCC offices. Ama- pulses of the Loran chains.
teur radio clubs in the larger cities have
established examining committees to assist 80 Meters The 80 -meter band
would -be amateurs of the area in obtaining (3500 kHz -4000 kHz) is the most popular
their Novice and Technician licenses. amateur band in
the continental United States for local "rag
chewing" and traffic nets. During the years
1 -3 The Amateur Bands of minimum sunspot activity the iono-
spheric absorption on this band may be
Certain small segments of the radio -fre- quite low, and long distance DX contacts
quency spectrum between 1500 kHz and are possible during the winter night hours.
10,000 MHz are reserved for operation of Daytime operation, in general, is limited to
amateur radio stations. These segments are contacts of 500 miles or less. During the
in general agreement throughout the world, summer months, local static and high iono-
although certain parts of different amateur spheric absorption limit long distance con-
bands may be used for other purposes in tacts on this band. As the sunspot cycle ad-
various geographic regions. In particular, vances and the MUF rises, increased iono-
the 40 -meter amateur band is used legally spheric absorption will tend to degrade the
(and illegally) for short -wave broadcasting long distance possibilities of this band. At
by many countries in Europe, Africa and the peak of the sunspot cycle, the 80 -meter
Asia. Parts of the 80 -meter band are used band becomes useful only for short -haul
for short distance marine work in Europe, communication.
and for broadcasting in South America. The
amateur bands available to United States 40 Meters The 40 -meter band
radio amateurs are: (7000 kHz-7300 kHz) is high enough in
frequency to be se-
160 Meters The 160 -meter band verely affected by the 11 -year sunspot cycle.
(1800 kHz -2000 kHz) is divided into 25- During years of minimum solar activity,
kHz segments on a the MUF may drop below 7 MHz, and the
regional basis, with day and night power band will become very erratic, with signals
limitations, and is available for amateur use dropping completely out during the night
provided no interference is caused to the hours. Ionospheric absorption of signals is
Loran (Long Range Navigation) stations not as large a problem on this band as it is
operating in this band. This band is least on 80 and 160 meters. As the MUF grad-
affected by the 11 -year solar sunspot cycle. ually rises, the skip distance will increase on
The maximum usable frequency (MUF) 40 meters, especially during the winter
even during the years of decreased sunspot months. At the peak of the solar cycle, the
activity does not usually drop below 4 MHz, daylight skip distance on 40 meters will be
therefore this band is not subject to the quite long, and stations within a distance of
violent fluctuations found on the higher - 500 miles or so of each other will not be
frequency bands. DX contacts on this band able to hold communication. DX operation
are limited by the ionospheric absorption of on the 40 -meter band is considerably ham-
radio signals, which is quite high. During pered by broadcasting stations, propaganda
winter nighttime hours the absorption is stations, and jamming transmitters. In
often of a low enough value to permit trans- Europe and Asia the band is in a chaotic
oceanic contacts on this band. On rare oc- state, and amateur operation in this region
casions, contacts up to 10,000 miles have is severely hampered.
www.americanradiohistory.com
14 Introduction to Radio THE RADIO
20 meters At the present model TV receivers having a 21 -MHz i -f
(14,000 kHz- 14,350 kHz) time, the 20 -me- channel, which falls directly in the 1 5 -meter
ter band is by band. The interference problem may be al-
far the most popular band for long- distance leviated by retuning the i -f system to a
contacts. High enough in frequency to be frequency outside the amateur assignment.
almost obliterated at the bottom of the
10 Meters During the peak
solar cycle, the band nevertheless provides
good DX contacts during years of minimal
(28,000 kHz- 29,700 kHz) of the sunspot
sunspot activity. At the present time, the cycle, the 10-
band is open to almost all parts of the world meter band is without doubt the most pop-
at some time during the year. During the ular amateur band. The combination' of
summer months, the band is active until the long skip and low ionospheric absorption
late evening hours, but during the winter make reliable DX contacts with low -powered
months the band is only good for a few equipment possible. The great width of the
hours during daylight. Extreme DX contacts band (1700 kHz) provides room for a
are usually erratic, but the 20 -meter band is large number of amateurs. The long skip
the only band available for DX operation (1500 miles or so) prevents nearby amateurs
the year around during the bottom of the from hearing each other, thus dropping the
sunspot cycle. As the sunspot count increases interference level. During the winter
and the MUF rises, the 20 -meter band will months, sporadic -E (short -skip) signals up
become open for longer hours during the to 1200 miles or so will be heard. The 10-
winter. The maximum skip distance in- meter band is poorest in the summer months,
creases, and DX contacts are possible over even during a sunspot maximum. Extreme-
paths other than the Great Circle route. ly long daylight skip is common on this
Signals can be heard via the "long path," 180 band, and in years of high MUF the 10-
degrees opposite the Great Circle path. Dur- meter band will support intercontinental
ing daylight hours, absorption may become DX contacts during daylight hours.
apparent on the 20 -meter band, and all The second harmonic of stations operating
signals except very short skip may disappear. in the 10 -meter band falls directly into
On the other hand, the band will be open television channel 2, and the higher har-
for worldwide DX contacts all night long. monics of 10 -meter transmitters fall into
The 20 -meter band is very susceptible to the higher TV channels. This harmonic
"fadeouts" caused by solar disturbances, and problem seriously curtailed amateur 10 -meter
all except local signals may completely dis- operation during the late 40's. However,
appear for periods of a few hours to a day with new circuit techniques and the TVI
or so. precautionary measures stressed in this Hand-
book, 10 -meter operation should cause little
15 Meters
or no interference to nearby television re-
This is a rela- ceivers of modern design.
(21,000 kHz-21,450 kHz) tively new band
for radio ama- Six Meters At the peak of the sun -
teurs since it has only been available for (50 MHz -54 MHz) spot cycle, the MUF
amateur operation since 1952. It has charac- occasionally rises high
teristics similar to both the 20- and 10- enough to permit DX contacts up to 10,000
meter amateur bands. During a period of miles or so on 6 meters. Activity on this
low sunspot activity, the MUF will rarely band during such a period is often quite high.
rise as high as 15 meters, so this band will Interest in this band wanes during a period
be "dead" for a large part of the sunspot of lesser solar activity, since contacts, as a
cycle. During the next few years, 15 -meter rule, are restricted to short -skip work. The
activity should pick up rapidly, and the proximity of the 6 -meter band to television
band should support extremely long DX channel 2 often causes interference prob-
contacts. The band will remain open 24 lems to amateurs located in areas where
hours a day in Equatorial areas of the world. channel 2 is active. As the sunspot cycle
Fifteen -meter operation may be hampered increases, activity on the 6 -meter band will
in some cases when neighbors possess older- increase.
www.americanradiohistory.com
HANDBOOK Learning the Code 15
The VHF Bands The vhf bands are send very slowly when working other sta-
(Two Meters and "Up ") the least affected by tions at great distances. Stations repeat their
the vagaries of the calls many times when calling other stations
sunspot cycle and Heaviside layer. Their before contact is established, and one need
predominant use is for reliable communica- not have achieved much code proficiency to
tion over distances of 150 miles or less. make out their calls and thus determine
These bands are sparsely occupied in the their location.
rural sections of the United States, but are The Code The applicant for any class of
quite heavily congested in the urban areas amateur operator license must be
of high population.
able to send and receive the Continental
In recent years it has been found that Code (sometimes called the International
vhf signals are propagated by other means Morse Code). The speed required for the
than by line -of -sight transmission. "Scatter
sending and receiving test may be either 5,
signals," Aurora reflection, and air-mass
13, or 20 words per minute, depending on
boundary bending are responsible for vhf the class of license assuming an average of
communication up to 1200 miles or so. five characters to the word in each case.
Weather conditions will often affect long -
The sending and receiving tests run for
distance communication on the 2 -meter
five minutes, and one minute of errorless
band, and all the vhf bands are particularly
transmission or reception must be accom-
sensitive to this condition.
plished within the five -minute interval.
In recent years the vhf bands have been
If the code test is failed, the applicant
used for experimental "moonbounce" (earth -
must wait at least one month before he
moon- earth) transmissions and for repeater -
may again appear for another test. Approxi-
satellite experiments (Project Oscar). The
vhf bands hold great promise for serious mately 30% of amateur applicants fail to
pass the test. It should be expected that
experimenters as radio amateurs forge into
nervousness and excitement will, at least to
the microwave region.
some degree, temporarily lower the appli-
cant's code ability. The best insurance
1 -4 Starting Your Study against this is to master the code at a little
greater than the required speed under ordi-
When you start to prepare yourself for nary conditions. Then if you slow down a
the amateur examination you will find that little due to nervousness during a test the
the circuit diagrams, tube characteristic result will not prove fatal.
curves, and formulas appear confusing and
difficult to understand. But after a few Memorizing There is no shortcut to code
study sessions one becomes sufficiently fa- the Code proficiency. To memorize the
miliar with the notation of the diagrams alphabet entails but a few eve-
and the basic concepts of theory and opera- nings of diligent application, but consider-
tion so that the acquisition of further able time is required to build up speed. The
knowledge becomes easier and even fascinat- exact time required depends on the individ-
ing. ual's ability and the regularity of practice.
Since it takes a considerable time to be- While the speed of learning will naturally
come proficient in sending and receiving vary greatly with different individuals, about
code, it is a good idea to intersperse techni- 70 hours of practice (no practice period to
cal study sessions with periods of code prac- be over 30 minutes) will usually suffice to
tice. Many short code -practice sessions ben - bring a speed of about 13 w.p.m.; 16 w.p.m.
fit one more than a small number of longer requires about 120 hours; 20 w.p.m., 175
sessions. Alternating between one study and hours.
the other keeps the student from getting Since code reading requires that individual
"stale" since each type of study serves as a letters be recognized instantly, any memor-
sort of respite from the other. izing scheme which depends on orderly se-
When you have practiced the code long quence, such as learning all "doh" letters and
enough you will be able to follow the gist of all "dit" letters in separate groups, is to be
the slower- sending stations. Many stations discouraged. Before beginning with a code
16 Introduction to Radio THE RADIO
A im N .8. IM ss
a ss
i
>..
s
f3 1=9 2
a.
O IMS MI
C a= =ID p I= MD 3 OM
am
.
D Q INEDMIS 4
E R 5
F S 6
G OM MI T 7 .,OM.
H U
.,. 8 ss sa a
J
I
wss V
W ss
9 ss ssm
0
s
ess
L a. X am
ss
0 MEANS ZERO. AND IS WRITTEN IN THIS
WAY TO DISTINGUISH IT FROM THE LETTER `O"
,
Y
M Z s IT OFTEN IS TRANSMITTED INSTEAD AS ONE
LONG DASH (EQUIVALENT TO S DOTS)
ss
PERIOD (.)
COMMA (,)
INTERROGATION (7
QUOTATION MARK ("
sss .
MIIM
ss
MID
ssNM=1
IMP
WAIT SIGN (AS)
DOUBLE DASH (BREAK)
ERROR (ERASE SIGN)
FRACTION BAR (/)
Mss
CM
COLON (i
SEMICOLON ())
PARENTHESIS (
ssss NMI
IMss1WO
END OF MESSAGE (AR)
END OF TRANSMISSION (SK)
INTERNAT. DISTRESS SIG. (SOS)
.1.IMP
IMO
MI/ Inatf.
Figure 1
The Continental (or International Morse) Code is used for substantially all non -automatic
radio
communication. DO NOT memorize from the printed pane; code is a language of SOUND, and
must not be learned visually; learn by listening as explained in the text.
practice set it is necessary to memorize the period, comma, and question mark have been
whole alphabet perfectly. A good plan is to memorized so thoroughly that any one can
study only two or three letters a day and be sounded without the slightest hesitation.
to drill with those letters until they become Do not bother with other punctuation or
part of your consciousness. Mentally trans- miscellaneous signals until later.
late each day's letters into their sound equiv-
alent wherever they are seen, on signs, in
papers, indoors and outdoors. Tackle two
Sound - Each letter and figure must be
Not Sight memorized by its sound rather
additional letters in the code chart each day, than its appearance. Code is a
at the same time reviewing the characters system of sound communication, the same
already learned. as is the spoken word. The letter A, for ex-
Avoid memorizing by routine. Be able
to sound out any letter immediately with-
out so much as hestitating to think about all MI
the letters preceding or following the one
in question. Know C, for example, apart
MO OMMI
I111 1111 OM GM
from the sequence ABC. Skip about among
all the characters learned, and before very IM
long sufficient letters will have been acquired OM MOMIMD
to enable you to spell out simple words to
yourself in "dit dahs." This is interesting
OM Mil OM
exercise, and for that reason it is good to 1=1 MD
memorize all the vowels first and the most Figure 2
common consonants next.
Actual code practice should start only These code characters are used in languages
other than English. They may occasionally
when the entire alphabet, the numerals, be encountered so it is well to know them.
www.americanradiohistory.com
HANDBOOK Learning the Code 17
ample, is one short and one long sound in manner until a speed of about 16 words
combination sounding like dit dah, and it per minute is attained if the object is to pass
must be remembered as such, and not as the amateur 13 -word per minute code test.
"dot dash." The margin of 3 w.p.m. is recommended to
overcome a possible excitement factor at
Practice Time, patience, and regularity are examination time. Then when you take the
required to learn the code properly. test you don't have to worry about the
Do not expect to accomplish it within a few "jitters" or an "off day."
days. Speed should not be increased to a new
Don't practice too long at one stretch; it level until the student finally makes solid
does more harm than good. Thirty minutes copy with ease for at least a five-minute
at a time should be the limit. period at the old level. How frequently in-
Lack of regularity in practice is the creases of speed can be made depends on in-
most common cause of lack of progress. dividual ability and the amount of practice.
Irregular practice is very little better than Each increase is apt to prove disconcerting,
no practice at all. Write down what you but remember "you are never learning when
have heard; then forget it; do not look back. you are comfortable."
If your mind dwells even for an instant on A number of amateurs are sending code
a signal about which you have doubt, you practice on the air on schedule once or
will miss the next few characters while your twice each week; excellent practice can be
attention is diverted. obtained after you have bought or con-
While various automatic code machines, structed your receiver by taking advantage
phonograph records, etc., will give you prac- of these sessions.
tice, by far the best practice is to obtain a If you live in a medium -size or large city.
study companion who is also interested in the chances are that there is an amateur -
learning the code. When you have both radio club in your vicinity which offers
memorized the alphabet you can start send- free code -practice lessons periodically.
ing to each other. Practice with a key and
oscillator or key and buzzer generally proves Skill When you listen to someone speaking
superior to all automatic equipment. Two you do not consciously think how his
such sets operated between two rooms are words are spelled. This is also true when you
-or
fine between your house and his will read. In code you must train your ears to
be just that much better. Avoid talking to read code just as your eyes were trained in
your partner while practicing. If you must school to read printed matter. With enough
ask him a question, do it in code. It makes practice you acquire skill, and from skill,
more interesting practice than confining speed. In other words, it becomes a habit,
yourself to random practice material. something which can be done without con-
When two co- learners have memorized scious effort. Conscious effort is fatal to
the code and are ready to start sending to speed; we can't think rapidly enough; a
each other for practice, it is a good idea to speed of 25 words a minute, which is a com-
enlist the aid of an experienced operator for mon one in commercial operations, means
the first practice session or two so that they 125 characters per minute or more than two
will get an idea of how properly formed per second, which leaves no time for con-
characters sound. scious thinking.
During the first practice period the speed
should be such that substantially solid copy Perfect Formation When transmitting on the
can be made without strain. Never mind if of Characters code practice set to your
this is only two or three words per minute. partner, concentrate on the
In the next period the speed should be in- quality of your sending, not on your speed.
creased slightly to a point where nearly all Your partner will appreciate it and he could
of the characters can be caught only through not copy you if you speeded up anyhow.
conscious effort. When the student becomes If you want to get a reputation as hav-
proficient at this new speed, another slight ing an excellent "fist" on the air, just re-
increase may be made, progressing in this member that speed alone won't do the
www.americanradiohistory.com
18 Introduction to Radio THE RADIO
trick. Proper execution of your letters and oqo;oocod.
spacing will make much more of an im-
pression. Fortunately, as you get so that
you can send evenly and accurately, your B C
sending speed will automatically increase.
Remember to try to see how evenly you can
send, and how fast you can receive. Con-
centrate on making signals properly with
ei * i
A
- O. Ni
your key. Perfect formation of characters Figure 3
is paramount to everything else. Make every
signal right no matter if you have to prac- Diagram illustrating relative lengths of
dashes and spaces referred to the duration
tice it hundreds or thousands of times. of a dot. A dash is exactly equal in duration
Never allow yourself to vary the slightest to three dots; spaces between parts of a
letter equal one dot; those between letters,
from perfect formation once you have three dots; spare between words, five dots.
learned it. Note that a slight increase between two parts
of a letter will make it sound like two letters.
If possible, get a good operator to listen
to your sending for a short time, asking him
to criticize even the slightest imperfections. dashes. At the greater speed he will merely
have to learn the identification of the same
Timing It is of the utmost importance to sound without taking as long to do so.
maintain uniform spacing in charac- Be particularly careful of letters like B.
ters and combinations of characters. Lack of Many beginners seem to have a tendency to
uniformity at this point probably causes be- leave a longer space after the dash than
ginners more trouble than any other single that which they place between succeeding
factor. Every dot, every dash, and every dots, thus making it 'sound like TS. Simi-
space must be correctly timed. In other larly, make sure that you do not leave
words, accurate timing is absolutely essen- a longer space after the first dot in the
tial to intelligibility, and timing of the letter C than you do between other parts
spaces between the dots and dashes is just as of the same letter: otherwise it will sound
important as the lengths of the dots and like NN.
dashes themselves.
The characters are timed with the dot as Sending vs. Once you have memorized the
a "yardstick." A standard dash is three times Receiving code thoroughly you should con-
as long as a dot. The spacing between parts centrate on increasing your re-
of the same letter is equal to one dot, the ceiving speed. True, if you have to practice
space between letters is equal to three dots, with another newcomer who is learning the
and that between words equal to five dots. code with you, you will both have to do
The rule for spacing between letters and some sending. But don't attempt to prac-
words is not strictly observed when sending tice sending just for the sake of increasing
slower than about 10 words per minute for your sending speed.
the benefit of someone learning the code When transmitting on the code practice
and desiring receiving practice. When send- set to your partner so that he can get re-
ing at, say, f w.p.m., the individual letters ceiving practice, concentrate on the quality
should be made the same as if the sending of your sending, not on your speed.
rate were about 10 w.p.m., except that the Because it is comparatively easy to learn
spacing between letters and words is greatly to send rapidly, especially when no particu-
exaggerated. The reason for this is obvious. lar care is given to the quality of sending,
The letter L, for instance, will then sound many operators who have just received their
exactly the same at 10 w.p.m. as at 5 licenses get on the air and send mediocre (or
w.p.m., and when the speed is increased worse) code at 20 w.p.m. when they can
above f w.p.m. the student will not have barely receive good code at 13. Most old -
to become familiar with what may seem timers remember their own period of initia-
to him like a new sound, although it is in tion and are only too glad to be patient and
reality only a faster combination of dots and considerate if you tell them that you are
www.americanradiohistory.com
HANDBOOK Learning the Code 19
www.americanradiohistory.com
20 Introduction to Radio THE RADIO
Figure 5
practice), you will occasionally miss a sig- Copying Behind All good operators copy sev-
nal or a small group of them. When you do, eral words behind, that is,
leave a blank space; do not spend time fu- while one word is being received, they are
tilely trying to recall it; dismiss it, and writing down or typing, say the fourth or
center attention on the next letter; other- fifth previous word. At first this is very
wise you'll miss more. Do not ask the sender difficult, but after sufficient practice it will
any questions until the transmission is fin- be found actually to be easier than copying
ished. close up. It also results in more accurate
To prevent guessing and get equal prac- copy and enables the receiving operator to
tice on the less common letters, depart oc- capitalize and punctuate copy as he goes
casionally from plain language material and along. It is not recommended that the be-
use a jumble of letters in which the usually ginner attempt to do this until he can send
less commonly used letters predominate. and receive accurately and with ease at a
As mentioned before, many students put speed of at least 12 words a minute.
a greater space after the dash in the letter It requires a considerable amount of train-
B, than between other parts of the same ing to disassociate the action of the subcon-
letter so it sounds like TS. C, F, Q,V, X, Y, scious mind from the direction of the con-
and Z often give similar trouble. Make a list scious mind. It may help some in obtaining
of words or arbitrary combinations in which
these letters predominate and practice them,
both sending and receiving until they no GE-2 2N406.
longer give you trouble. Stop everything else 2N108.oR 2N2614
and stick to them. So long as these characters E C
www.americanradiohistory.com
HANDBOOK Learning the Code 21
this training to write down two columns of can use a set of phonograph code-practice
short words. Spell the first word in the first records. The records are of use only if you
column out loud while writing down the have a phonograph whose turntable speed is
first word in the second column. At first readily adjustable. The tape machine can
this will be a bit awkward, but you will be rented by the month for a reasonable fee.
rapidly gain facility with practice. Do the Once you can copy about 10 w.p.m. you
same with all the words, and then reverse can also get receiving practice by listening
columns. to slow- sending stations on your receiver.
Next try speaking aloud the words in the Many amateur stations send slowly par-
one column while writing those in the ticularly when working far distant stations.
other column; then reverse columns. When receiving conditions are particularly
After the foregoing can be done easily, poor many commercial stations also send
try sending with your key the words in one slowly, sometimes repeating every word. Un-
column while spelling those in the other. til you can copy around 10 w.p.m. your
It won't be easy at first, but it is well worth receiver isn't much use, and either another
keeping after if you intend to develop any operator or a machine or records is neces-
real code proficiency. Do not attempt to sary for getting receiving practice after you
catch up. There is a natural tendency to have once memorized the code.
close up the gap, and you must train your-
self to overcome this. Code Practice If you don't feel too foolish
Next have your code companion send you Sets doing it, you can secure a
a word either from a list or from straight measure of code practice with
text; do not write it down yet. Now have the help of a partner by sending "dit -dah"
him send the next word; after receiving this messages to each other while riding to work,
second word, write down the first word. eating lunch, etc. It is better, however, to
After receiving the third word, write the use a buzzer or code- practice oscillator in
second word; and so on. Never mind how conjunction with a regular telegraph key.
slowly you must go, even if it is only two As a good key may be considered an in-
or three words per minute. Stay behind. vestment it is wise to make a well -made key
It will probably take quite a number of your first purchase. Regardless of what type
practice sessions before you can do this with code- practice set you use, you will need a
any facility. After it is relatively easy, then key, and later on you will need one to key
try staying two words behind; keep this up your transmitter. If you get a good key to
until it is easy. Then try three words, four begin with, you won't have to buy another
words, and five words. The more you prac- one later.
tice keeping received material in mind, the The key should be rugged and have
easier it will be to stay behind. It will be fairly heavy contacts. Not only will the
found easier at first to copy material with key stand up better, but such a key will
which one is fairly familiar, then gradually contribute to the "heavy" type of sending
switch to less familiar material. so desirable for radio work. Morse (tele-
graph) operators use a "light" style of
Automatic Code The two practice sets which sending and can send somewhat faster when
Machines are described in this chapter using this light touch. But, in radio work
are of most value when you static and interference are often present,
have someone with whom to practice. Auto- and a slightly heavier dot is desirable. If
matic code machines are not recommended you use a husky key, you will find yourself
to anyone who can possibly obtain a com- automatically sending in this manner.
panion with whom to practice, someone who To generate a tone simulating a code
is also interested in learning the code. If signal as heard on a receiver, either a me-
you are unable to enlist a code partner and chanical buzzer or an audio oscillator may
have to practice by yourself, the best way be used. Figure S shows a simple code -prac-
to get receiving practice is by the use of tice set using a buzzer which may be used
a tape machine (automatic code -sending directly simply by mounting the buzzer on
machine) with several practice tapes. Or you a sounding board, or the buzzer may be
22 Introduction to Radio
9V.
100K
OIBIo--s KEY
GE-3, 2N2148,
OR 2N2869
GE-10, 2N2923, PM
OR 2N339I SPEAKER
.04
Figure 8
www.americanradiohistory.com
CHAPTER TWO
www.americanradiohistory.com
24 Direct- Current Circuits THE RADIO
filled are those of the inert gases mentioned attract free electrons from the wire in order
before; all other elements have one or more for the positive terminal to become neutral.
uncompleted rings of electrons. If the un- The attracting of electrons continues
completed ring is nearly empty, the element through the wire, and finally the excess elec-
is metallic in character, being most metallic trons at the negative terminal of the battery
when there is only one electron in the outer are attracted by the positively charged atoms
ring. If the incomplete ring lacks only one at the end of the wire. Other sources of
or two electrons, the element is usually non- electrical potential (in addition to a bat-
metallic. Elements with a ring about half tery) are: an electrical generator (dynamo),
completed will exhibit both nonmetallic and a thermocouple, an electrostatic generator
metallic characteristics; carbon, silicon, (static machine), a photoelectric cell, and
germanium, and arsenic are examples. Such a crystal or piezoelectric generator.
elements are called semiconductors. Thus it is seen that a potential dif-
In metallic elements these outer ring elec- ference is the result of a difference in the
trons are rather loosely held. Consequently, number of electrons between the two (or
there is a continuous helter -skelter move- more) points in question. The force or pres-
ment of these electrons and a continual sure due to a potential difference is termed
shifting from one atom to another. The the electromotive force, usually abbreviated
electrons which move about in a substance e.m.f. or E.M.F. It is expressed in units
are called free electrons, and it is the ability called volts.
of these electrons to drift from atom to It should be noted that for there to be a
atom which makes possible the electric cur- potential difference between two bodies or
rent. points it is not necessary that one have a
positive charge and the other a negative
Conductors and If the free electrons are nu- charge. If two bodies each have a negative
Insulators merous and loosely held, charge, but one more negative than the
the element is a good con - other, the one with the lesser negative
ductor. On the other hand, if there are few charge will act as though it were positively
free electrons (as is the case when the elec- charged with respect to the other body. It
trons in an outer ring are tightly held), the is the algebraic potential difference that
element is a poor conductor. If there are determines the force with which electrons
virtually no free electrons, the element is a are attracted or repulsed, the potential of
good insulator. the earth being taken as the zero reference
point.
2 -2 Fundamental Electrical
Units and Relationships The Electric The flow of electrons along a
Current conductor due to the applica-
Electromotive Force: The free electrons in a tion of an electromotive force
Potential Difference conductor move con- constitutes an electric current. This drift is
stantly about and in addition to the irregular movements of
change their position in a haphazard man- the electrons. However, it must not be
ner. To produce a drift of electrons, or elec- thought that each free electron travels from
tric current, along a wire it is necessary that one end of the circuit to the other. On the
there be a difference in "pressure" or poten- contrary, each free electron travels only a
tial between the two ends of the wire. This short distance before colliding with an atom;
potential difference can be produced by con- this collision generally knocks off one or
necting a source of electrical potential to more electrons from the atom, which in
the ends of the wire. turn move a short distance and collide with
As will be explained later, there is an ex- other atoms, knocking off other electrons.
cess of electrons at the negative terminal of Thus, in the general drift of electrons along
a battery and a deficiency of electrons at the a wire carrying an electric current, each
positive terminal, due to chemical action. electron travels only a short distance and
When the battery is connected to the wire, the excess of electrons at one end and the
the deficient atoms at the positive terminal deficiency at the other are balanced by the
www.americanradiohistory.com
HANDBOOK Resistance 25
source of the e.m.f. When this source is Resistance The flow of current in a materi-
removed the state of normalcy returns; al depends on the ease with
there is still the rapid interchange of free which electrons can be detached from the
electrons between atoms, but there is no atoms of the material and on its molec-
general trend or "net movement" in either ular structure. In other words, the easier
one direction or the other -in other words, it is to detach electrons from the atoms the
no current flows. more free electrons there will be to contrib-
ute to the flow of current, and the fewer
Ampere and There are two units of meas- collisions that occur between free electrons
Coulomb urement associated with cur- and atoms the greater will be the total
rent, and they are often con- electron flow.
fused. The rate of flow of electricity is The opposition to a steady electron flow
stated in amperes. The unit of quantity is is called the resistance of a material, and is
the coulomb. A coulomb is equal to 6.28 one of its physical properties.
X 10'" electrons, and when this quantity of The unit of resistance is the ohm. Every
electrons flows by a given point in every substance has a specific resistance, usually
second, a current of one ampere is said to expressed as ohms per mil -foot, which is de-
be flowing. An ampere is equal to one termined by the material's molecular struc-
coulomb per second; a coulomb is, converse- ture and temperature. A mil -foot is a piece
ly, equal to one ampere- second. Thus we of material one circular mil in area and one
see that coulomb indicates amount and foot long. Another measure of resistivity
ampere indicates rate of flow of electric frequently used is expressed in the units
current. microhms per centimeter cube. The resist-
ance of a uniform length of a given sub-
Current and Older textbooks speak of cur- stance is directly proportional to its length
Electron Flow rent flow as being from the and specific resistance, and inversely pro-
positive terminal of the e.m.f. portional to its cross -sectional area. A wire
source through the conductor to the nega- with a certain resistance for a given length
tive terminal. Nevertheless, it has long been will have twice as much resistance if the
an established fact that the current flow length of the wire is doubled. For a given
in a metallic conductor is the electron drift length, doubling the cross -sectional area of
from the negative terminal of the source the wire will halve the resistance, while
of voltage through the conductor to the doubling the diameter will reduce the re-
positive terminal. The only exceptions to sistance to one fourth. This is true since
the electronic direction of flow occur in the cross -sectional area of a wire varies as
gaseous and electrolytic conductors where the square of the diameter. The relationship
the flow of positive ions toward the cathode between the resistance and the linear dimen-
or negative electrode constitutes a positive sions of a conductor may be expressed by
flow in the opposite direction to the elec- the following equation:
tron flow. (An ion is an atom, molecule, or rl
particle which either lacks one or more R
electrons, or else has an excess of one or
more electrons.)
where,
In radio work the terms "electron flow" R equals resistance in ohms,
and "current" are becoming accepted as r equals resistivity in ohms per mil -foot,
being synonymous, but the older terminology l equals length of conductor in feet,
is still accepted in the electrical (industrial) A equals cross -sectional area in circular
field. Because of the confusion this some- mils.
times causes, it is often safer to refer to the The resistance also depends on tempera-
direction of electron flow rather than to the ture, rising with an increase in tempera-
direction of the "current." Since electron ture for most substances (including most
flow consists actually of a passage of nega- metals), due to increased electron accelera-
tive charges, current flow and algebraic tion and hence a greater number of impacts
electron flow do pass in the same direction. between electrons and atoms. However, in
www.americanradiohistory.com
26 Direct-Current Circuits THE RADIO
Conductors and In the molecular structure
TABLE 1. TABLE OF RESISTIVITY
Insulators of many materials such as
Resistivity in glass, porcelain, and mica all
Ohms per Temp. Coeff. of
Circular resistance per C. electrons are tightly held within their orbits
Material Mil -Feet at 20 C.
and there are comparatively few free elec-
Aluminum 17 0.0049 trons. This type of substance will conduct
Brass 45 0.003 to 0.007
an electric current only with great difficulty
Cadmium 46 0.0038
Chromium 16 0.00 and is known as an insulator. An insulator
Copper 10.4 0.0039 is said to have a high electrical resistance.
iron 59 0.006 On the other hand, materials that have a
Silver 9.8 0.004 large number of free electrons are known'as
Zinc 36 0.0005
Nichrome 650 0.0002 conductors. Most metals (those elements
Constantan 295 0.00001 which have only one or two electrons in
Manganin 290 0.00001 their outer ring) are good conductors. Silver,
Monet 255 0.0019 copper, and aluminum, in that order, are
the best of the common metals used as con-
the case of some substances such as carbon ductors and are said to have the greatest
and glass the temperature coefficient is nega- conductivity, or lowest resistance to the flow
tive and the resistance decreases as the tem- of an electric current.
perature increases. This is also true of elec-
trolytes. The temperature may be raised by Fundamental These units are the volt, the
the external application of heat, or by the Electrical Unitsampere, and the ohm. They
flow of the current itself. In the latter case, were mentioned in the pre-
the temperature is raised by the heat gen- ceding paragraphs, but were not completely
erated when the electrons and atoms collide. defined in terms of fixed, known quantities.
yri
11111111111111M1111
.
Figure 1
TYPICAL RESISTORS
Shown above are various types of resistors used in electronic circuits. The larger
units are power resistors. On the left is a variable power resistor. Three precision -
type resistors are shown in the canter with two small composition resistors beneath
them. At the right is a composition -type potentiometer, used for audio circuitry.
www.americanradiohistory.com
HANDBOOK Resistive Circuits 27
RESISTANCE RI -- R2
W
A
Figure 3
--
CONDUCTORS -4.
SIMPLE PARALLEL
E
CIRCUIT
--
The two resistors R and R, are said to be in
parallel since the flow of is offered
Figure 2 two parallel paths. An electron leaving point
A will pass either through R, or R,, but not
SIMPLE SERIES CIRCUITS through both. to reach the positive terminal
of the battery. If a large number of electreat
At (A) the battery Is in series with o single are considered, the greater number will pass
resistor. At (I) the battery is in series with through whichever of the two resistors has
two resistors, the resistors themselves being the lower resistance.
in series. The arrows indicate the direction of
electron Row.
found. If the voltage and current are
The fundamental unit of current, or rate known, and the resistance is unknown, the
of flow of electricity is the ampere. A cur- E
rent of one ampere will deposit silver from resistance (R) is equal to .When the
a specified solution of silver nitrate at a
rate of 1.118 milligrams per second. voltage is the unknown quantity, it can be
The international standard for the ohm is found by multiplying 1 X R. These three
the resistance offered by a uniform column equations are all secured from the original
of mercury at 0 C., 14.4521 grams in mass, by simple transposition. The expressions are
of constant cross- sectional area and 106.300 here repeated for quick reference:
centimeters in length. The expression meg-
ohm (1,000,000 ohms) is also sometimes
used when speaking of very large values of
I= R
R= E =IR
resistance. where,
A volt is the e.m.f. that will produce a
current of one ampere through a resistance 1 is the current in amperes,
of one ohm. The standard of electromotive R is the resistance in ohms,
force is the Weston cell which at 20 C. E is the electromotive force in volts.
has a potential of 1.0183 volts across its
terminals. This cell is used only for reference Application of All electrical circuits fall into
purposes in a bridge circuit, since only an Ohm's Law one of three classes: series
infinitesimal amount of current may be circuits, parallel circuits, and
drawn from it without disturbing its char- series - parallel circuits. A series circuit is
acteristics. one in which the current flows in a single
continuous path and is of the same value at
Ohm's Low The relationship between the every point in the circuit (figure 2). In a
electromotive force (voltage), parallel circuit there are two or more cur-
the flow of current (amperes), and the re- rent paths between two points in the circuit,
sistance which impedes the flow of current as shown in figure 3. Here the current di-
(ohms), is very clearly expressed in a simple vides at A, part going through R, and part
but highly valuable law known as Ohm's through R2, and combines at B to return
Law. This law states that the current in am-
peres is equal to the voltage in volts divided
by the resistance in ohms. Expressed as an Figure 4
equation: SERIES -PARALLEL
CIRCUIT
I = R
In this type of circuit the resistors are ar-
If the voltage (E) and resistance (R) ranged in series groups, and those groups are
are known, the current (I) can be readily then placed in parallel.
www.americanradiohistory.com
28 Direct -Current Circuits THE RADIO
to the battery. Figure 4 shows a series - Resistances Consider two resistors, one of
parallel circuit. There are two paths between in Parallel 100 ohms and one of 10 ohms,
points A and B as in the parallel circuit, and connected in parallel as in fig-
in addition there are two resistances in series ure 3, with a potential of 10 volts applied
in each branch of the parallel combination. across each resistor, so the current through
Two other examples of series -parallel ar- each can be easily calculated.
rangements appear in figure S. The way in
which the current splits to flow through E
1
the parallel branches is shown by the arrows.
In every circuit, each of the parts has
some resistance: the batteries or generator,
= 10 volts
E
= 10
= 0.1 ampere
the connecting conductors, and the appa- R, = 100 ohms I' 100
ratus itself. Thus, if each part has some re-
sistance, no matter how little, and a current
is flowing through it, there will be a volt-
E = lO volts
R. = 10 ohms
12 - 10
10
= 1.0 ampere
age drop across it. In other words, there will Total current = 1, + I2 = 1.1 ampere
be a potential difference between the two
ends of the circuit element in question. This Until it divides at A, the entire current
drop in voltage is equal to the product of of 1.1 amperes is flowing through the con-
the current and the resistance hence it is ductor from the battery to A, and again
called the IR drop. from B through the conductor to the bat-
tery. Since this is more current than flows
Internal The source of voltage has an in- through the smaller resistor it is evident
Resistance ternal resistance, and when con- that the resistance of the parallel combina-
nected into a circuit so that tion must be less than 10 ohms, the re-
current flows, there will be an IR drop sistance of the smaller resistor. We can find
in the source just as in every other part this value by applying Ohm's Law.
of the circuit. Thus, if the terminal voltage
of the source could be measured in a way RT =
that would cause no current to flow, it
would be found to be more than the voltage E = 10 volts 10
measured when a current flows by the I = 1.1 amperes RT- 1.1 = 9.09 ohms
amount of the IR drop in the source. The
voltage measured with no current flowing The resistance of the parallel combination
is termed the no load voltage; that measured is 9.09 ohms.
with current flowing is the load voltage. Mathematically, we can derive a simple
It is apparent that a voltage source having formula for finding the effective resistance
a low internal resistance is most desirable. of two resistors connected in parallel.
R, X R_
Resistances The current flowing in a series RT = R, + R2
in Series circuit is equal to the voltage
impressed divided by the total where,
resistance across which the voltage is im- RT is the unknown resistance,
pressed. Since the same current flows through R, is the resistance of the first resistor,
every part of the circuit, it is merely nec- R_ is the resistance of the second resistor.
essary to add all the individual resistances
to obtain the total resistance. Expressed as If the effective value required is known,
a formula:
and it is desired to connect one unknown
resistor in parallel with one of known value,
RTo1a1= R, +R2 +R1 +... +R,. the following transposition of the above
formula will simplify the problem of ob-
Of course, if the resistances happened to be taining the unknown value:
all the same value, the total resistance would
be the resistance of one multiplied by the
number of resistors in the circuit.
R - R, X RT
R, -RT
www.americanradiohistory.com
HANDBOOK Voltage Divider 29
Rs
Ri R
Re R
R2 Rs R1
Figure 5 Figure 6
www.americanradiohistory.com
HANDBOOK Kirchhoff's Laws 31
-2 AMP3
ySly
lo +z+S+zoM 1 !
50 VOLTS DROP { Ps
I
A r -2 AMPS
10 +2 +5
MA. R2
{
150 VOLTS DROP AMP.;
l
300 VOLTS - +
10+2 MA. 111111
25 VOLTS DROP 20 VOLTS
Figure 9
BLEEDER CURRENT,10 MA.
75 VOLTS DROP
ILLUSTRATING KIRCHHOFF'S
FIRST LAW
--POWER SUPPLY LOAD The carrent flowing toward point "A" is equal
to the current flowing away from point "A."
Figure 8
www.americanradiohistory.com
32 Direct -Current Circuits THE RADIO
algebraic sum of all toltage drops around a which was drawn originally. This is illus-
closed path or loop in a circuit is zero. The trated in the example of figure 10, where
applied e.m.f.'s (voltages) are considered the direction of flow of I, is opposite to the
positive, while IR drops taken in the direc- direction assumed in the sketch.
tion of current flow (including the internal
drop of the sources of voltage) are consid- Power in In order to cause electrons
ered negative. Resistive Circuits to flow through a conduc-
Figure 10 shows an example of the appli- tor, constituting a current
cation of Kirchhoff's laws to a comparative- flow, it necessary to apply an electromotive
is
ly simple circuit consisting of three resistors force (voltage) across the circuit. Less pow-
and two batteries. First assume an arbitrary er is expended in creating a small current
direction of current flow in each closed loop flow through a given resistance than in
of the circuit, drawing an arrow to indicate creating a large one; so it is necessary to
the assumed direction of current flow. Then have a unit of power as a reference.
equate the sum of all IR drops plus battery The unit of electrical power is the watt,
drops around each loop to zero. You will which is the rate of energy consumption
need one equation for each unknown to be when an e.m.f. of 1 volt forces a current
determined. Then solve the equations for the of 1 ampere through a circuit. The power
unknown currents in the general manner in a resistive circuit is equal to the product
indicated in figure 10. If the answer comes of the voltage applied across, and the cur-
out positive the direction of current flow rent flowing in, a given circuit. Hence: P
you originally assumed was correct. If the (watts) = E (volts) X I (amperes).
answer comes out negative, the current flow Since it is often convenient to express
is in the opposite direction to the arrow power in terms of the resistance of the cir-
cuit and the current flowing through it,
a substitution of IR for E (E =
IR) in the
above formula gives: P =
IR X I or P =
2 OHMS
12--'')11,
S OHMS /2R. in terms of voltage and resistance, P
= E2 /R. Here, I =
E/R and when this is
substituted for 1 the original formula be-
3
+
VOLTS
Iz
3 VOLTS
comes P =
E X E /R, or P=
E2 /R. To
repeat these three expressions:
1. SET VOLTAGE DROPS AROUND EACH LOOP EQUAL TO ZERO. P =EI,P =12R, and P =E2 /R
1 1 .
2 (0 . s ) + 2 ( 1 1 - 1 2) +3= 0 (FIRST Loop)
-6 +2 (12- I1)+312 =0 (SECOND LOOP) where,
2. SIMPLIFY P is the power in watts,
211 +211 -212 +3 =0 212-2I1+312-6=0 E is the electromotive force in volts, and
411+3 51z-211-6 =0 I is the current in amperes.
2 tz 211+6 I2
5
To apply the above equations to a typical
3. EQUATE
41, +3 2I1+6 problem: The voltage drop across a cathode
2 - 5 resistor in a power amplifier stage is SO
volts; the plate current flowing through the
4 SiMPLIFV
201,+15= 411 +12 resistor is 150 milliamperes. The number of
AMPERE watts the resistor will be required to dissi-
pate is found from the formula: P = EI, or
5. RE
Z+3
-SUBSTITUTE
24- 50 X .1S0 = 7.S watts (.150 ampere is
12= 2 = 2 - 1 a
I
AMPERE
equal to 150 milliamperes). From the fore-
Figure 10 going it is seen that a 7.5 -watt resistor will
safely carry the required current, yet a 10-
ILLUSTRATING KIRCHHOFF'S or 20 -watt resistor would ordinarily be used
SECOND LAW to provide a safety factor.
voltage drop a d any closed loop in a
In another problem, the conditions being
Tho
network is equal to zero. similar to those above, but with the resist-
www.americanradiohistory.com
I
HANDBOOK , Cpacitance 33
\\"46\1
pressed in coulombs.
Capacitance and Two metallic plates sep-
Capacitors arated from each other by
a thin layer of insulating
material (called a dielectric, in this case)
Figure 12 becomes a capacitor. When a source of d -c
potential is momentarily applied across these
TYPICAL CAPACITORS
plates, they may be said to become charged.
The two large units are high value filter ca- If the same two plates are then joined to-
pacitors. Shown beneath these are various gether momentarily by means of a switch,
types of by -pass capacitors for r -f and audio
application. the capacitor will discharge.
www.americanradiohistory.com
i
www.americanradiohistory.com
HANDBOOK Capacitive Circuits 35
III\
weight, and the power factor is good. The
CIRCULAR PLATE CAPACITORS
life of such a capacitor is dependent on the CAPACITANCE POR A GIVEN SPACING
moisture penetration of the paper dielectric,
and on the level of the applied d -c voltage. 11111
Air-dielectric capacitors are used in trans- IIIIM
mitting and receiving circuits, principally
..
where a variable capacitor of high resetabil- MEE
ity is required. The dielectric strength is
high, though somewhat less at radio fre-
1111
RE O
\\E
quencies than at 60 Hz. In addition,
ii..
:iii
corona discharge at high frequencies will
cause ionization of the air dielectric causing HEMP
an increase in power loss. Dielectric strength MATME5511 OMEN
may be increased by increasing the air pres-
sure, as is done in hermetically sealed radar 1 2 3 S s 7 f 10 11 12 13 It
units. In some units, dry nitrogen gas may CAPACITANCE IN PICOFARAOS
www.americanradiohistory.com
HANDBOOK Capacitive Circuits 37
This formula indicates that the capaci- Capacitors in Equations for calculating ca-
tance is directly proportional to the area of Parallel and pacitances of capacitors in par -
the plates and inversely proportional to the in Series allel connections are the same
thickness of the dielectric (spacing between as those for resistors in series.
the plates). This simply means that when
the area of the plate is doubled, the spacing
CT= C1 +C2 +... +C
between plates remaining constant, the ca- Capacitors in series connection are cal-
pacitance will be doubled. Also, if the area culated in the same manner as are resistors
of the plates remains constant, and the in parallel connection.
plate spacing is doubled the capacitance will The formulas are repeated: (1) For two
be reduced to half. or more capacitors of unequal capacitance
The above equation also shows that ca- in series:
pacitance is directly proportional to the di-
electric constant of the spacing material.
An air -spaced capacitor that has a capaci-
CT - I
1
1
+
tance of 100 pf in air would have a ca- Cl C2 Cq
pacitance of 467 pf when immersed in castor
1 1 1 1
oil, because the dielectric constant of castor or,
oil is 4.67 times as great as the dielectric CT Cl C2 Cs
constant of air. (2) Tim capacitors of unequal capacitance
Where the area of the plate is definitely in series:
set, when it is desired to know the spacing
needed to secure a required capacitance, C1 X Cz
CT = Ci +C_
AX0.2248XK
t
(3) Three capacitors of equal capacitance
in series:
where all units are expressed just as in the
preceding formula. This formula is not con-
fined to capacitors having only square or CT = 3 1
www.americanradiohistory.com
HANDBOOK Magnetism 39
trolytic capacitors are connected in series, these fields to build up into a resultant
the positive terminal is always connected external field which acts in a plane at right
to the positive lead of the power supply; angles to the direction in which the cur-
the negative terminal of the capacitor con- rent is flowing. This field is known as the
nects to the positive terminal of the next magnetic field.
capacitor in the series combination. The The magnetic field around a current -car-
method of connection for electrolytic ca- rying conductor is illustrated in figure 17.
pacitors in series is shown in figure 16. Elec- The direction of this magnetic field depends
trolytic capacitors have very low cost per entirely on the direction of electron drift or
microfarad of capacitance, but also have a current flow in the conductor. When the
large power factor and high leakage; both flow is toward the observer, the field about
dependent on applied voltage, temperature, the conductor is clockwise; when the flow
and the age of the capacitor. The modern is away from the observer, the field is
electrolytic capacitor uses a dry paste elec- counterclockwise. This is easily remembered
trolyte embedded in a gauze or paper dielec- if the left hand is clenched, with the thumb
tric. Aluminum foil and the dielectric are outstretched and pointing in the direction
wrapped in a circular bundle and are of electron flow. The fingers then indicate
mounted in a cardboard or metal box. the direction of the magnetic field around
Etched electrodes may be employed to in- the conductor.
crease the effective anode area, and the total Each electron adds its field to the total
capacitance of the unit. external magnetic field, so that the greater
The capacitance of an electrolytic ca- the number of electrons moving along the
pacitor is affected by the applied voltage, conductor, the stronger will be the resulting
the usage of the capacitor, the temperature field.
and the humidity of the environment. The One of the fundamental laws of magnet-
capacitance usually drops with the aging ism is that like poles repel one another and
of the unit. The leakage current and power unlike poles attract one another. This is
factor increase with age. At high frequen- true of current -carrying conductors as well
cies the power factor becomes so poor that as of permanent magnets. Thus, if two
the electrolytic capacitor acts as a series conductors are placed side by side and the
resistance rather than as a capacitance. current in each is flowing in the same di-
rection, the magnetic fields will also be in
2 -4 Magnetism the same direction and will combine to form
a larger and stronger field. If the current
and Electromagnetism flow in adjacent conductors is in opposite
The common bar or horseshoe magnet is directions, the magnetic fields oppose each
familiar to most people. The magnetic field other and tend to cancel.
The magnetic field around a conductor
which surrounds it causes the magnet to at-
tract other magnetic materials, such as iron may be considerably increased in strength
nails or tacks. Exactly the same kind of
by winding the wire into a coil. The field
magnetic field is set up around any conduc-
tor carrying a current, but the field exists
only while the current is flowing.
Magnetic Fields Before a potential, or volt-
age, is applied to a conduc- ELECTRON OR FT
www.americanradiohistory.com
HANDBOOK Inductance 41
is the m.m.f. in ampere -turns per inch. Residual Magnetism; The magnetism remain -
These relations may also be stated as follows: Retentivity ing in a material after
the magnetizing force
H =-or H = -B and B= Hp. orB= H is removed called residual magnetism. Re-
is
P. tentivity the property which causes a
is
It can be seen from the foregoing that magnetic material to have residual magne-
permeability is inversely proportional to the tism after having been magnetized.
specific reluctance of a material.
Hysteresis; Hysteresis is the characteristic
Coercive Force of a magnetic system which
Saturation Permeability is similar to electric
conductivity. This is, however, causes a loss of power due to
one important difference: the permeability
the fact that a negative magnetizing force
of magnetic materials is not independent of must be applied to reduce the residual mag-
the magnetic current (flux) flowing through netism to zero. This negative force is termed
coercive force. By "negative" magnetizing
it, although electrical conductivity is sub-
force is meant one which is of the opposite
stantially independent of the electric cur-
polarity with respect to the original magne-
rent in a wire. When the flux density of a
magnetic conductor has been increased to tizing force. Hysteresis loss is apparent in
the saturation point, a further increase in
transformers and chokes by the heating of
the core.
the magnetizing force will not produce a
corresponding increase in flux density. Inductance If the switch shown in figure 17
is opened and closed, a pulsating
B -H Curve To simplify magnetic circuit
calculations, a magnetization
direct current will be produced. When it is
first closed, the current does not instanta-
curve may be drawn for a given unit of
neously rise to its maximum value, but
material. Such a curve is termed a B -H
curve, and may be determined by experi- builds up to it. While it is building up, the
magnetic field is expanding around the con-
ment. When the current in an iron -core
coil is first applied, the relation between the
ductor. Of course, this happens in a small
winding current and the core flux is shown fraction of a second. If the switch is then
opened, the current stops and the magnetic
at A -B in figure 18. If the current is then
field contracts quickly. This expanding and
reduced to zero, reversed, brought back
again to zero and reversed to the original
contracting field will induce a current in
any other conductor that is part of a contin-
direction the flux passes through a typical uous circuit which it cuts. Such a field can
hysteresis loop as shown.
be obtained in the way just mentioned by
means of a vibrator interruptor, or by ap-
plying a. c. to the circuit in place of the
battery. Varying the resistance of the circuit
will also produce the same effect. This in-
ducing of a current in a conductor due to
MAGNETIZ I NG FORCE
a varying current in another conductor not
H -
in actual contact is called electromagnetic
induction.
is more commonly used, in the following Inductors in Inductors in parallel are com-
manner: Porallel bined exactly as are resistors
in parallel, provided that they
1 henry = 1000 millihenrys, or 10' are far enough apart so that the mutual
millihenrys. inductance is entirely negligible.
1 millihenry = 1/1000 henry, .001
Inductors in Inductors in series are additive,
henry, or 10 -9 henry.
Series just as are resistors in series,
1 microhenry = 1 /1,000,000 henry, again provided that no mutual
.000001 henry, or 10 -e henry. inductance exists. In this case, the total in-
ductance L is:
microhenry = 1 /1000 millihenry,
1
www.americanradiohistory.com
HANDBOOK Time Constant 43
f~ -s I INDUCTANCE OF
SINGLE-LAVER
SOLENOID COILS
Iron -Core Inductors Iron -core inductors may
at Radio Frequencies be used at radio fre-
quencies if the iron is
L MICRONENRYS in a very finely divided form, as in the case
9R+qS
of the powdered -iron cores used in some
types of r -f coils and i -f transformers. These
WHERE R RADIUS OF COIL TO CENTER WIRE cores are made of extremely small particles
S LENGTH OF COIL
N NUMBER CF TURNS of iron. The particles are treated with an in-
sulating material so that each particle will
Figure 20 be insulated from the others, and the treated
powder is molded with a binder into cores.
FORMULA FOR Eddy current losses are greatly reduced,
CALCULATING INDUCTANCE with the result that these special iron cores
Through the use of the equation and the are entirely practical in circuits which op-
sketch shown above the inductance of single - erate up to 100 MHz in frequency.
layer solenoid coils can be calculated with an
accuracy of about one p t for the types
of coils normally used ln the hf and vhf range. 2 -5 RC and RL Transients
Core Material Ordinary magnetic cores can- A voltage divider may be constructed as
not be used for radio frequen- shown in figure 21. Kirchhoff's and Ohm's
cies because the eddy current and hysteresis Laws hold for such a divider. This circuit
losses in the core material become enormous is known as an RC circuit.
as the frequency is increased. The principal
use for conventional magnetic cores is in the
audio- frequency range below approximately
15,000 Hertz, whereas at very low frequen-
cies (50 to 60 Hertz) their use is manda-
T E
tory if an appreciable value of inductance
isdesired.
An air -core inductor of only 1 henry in-
ductance would be quite large in size, yet
values as high as 500 henrys are commonly
available in small iron -core chokes. The in-
ductance of a coil with a magnetic core will
vary with the amount of current (both a -c
and d -c) which passes through the coil.
For this reason, iron -core chokes that are used TIM[t, IN TERMS OS TIME CONSTANT RC3
in power supplies have a certain inductance
rating at a predetermined value of direct
current.
The permeability of air does not change
with flux density; so the inductance of iron -
core coils often is made less dependent on
flux density by making part of the magnetic z 3
e 2.* 1, C
path air, instead of utilizing a closed loop of TIME IN TERMS OS TIME CON
www.americanradiohistory.com
44 Direct -Current Circuits T H E R A D I O
Figure 22
TYPICAL INDUCTANCES
The large inductance is a 1000 -watt transmitting coil. To the right and left of this coil are
small r -f chokes. Several varieties of low power capability coils are shown below, along
with various types of r -f chokes intended for high- frequency operation.
www.americanradiohistory.com
HANDBOOK Time Constant 45
age across the capacitor (e) could be ex- maining difference in voltage, or 86.5 per
pressed in the following manner: cent of the applied voltage (E).
e =E (1 - tl") RL CircuitIn the case of a series combina-
tion of a resistor and an inductor,
where e,E,R, and C have the values discussed as shown in figure 23, the current through
above, r = 2.716 (the base of Naperian or the combination follows a very similar law
natural logarithms), and t represents the to that given above for the voltage appear-
time which has elapsed since the closing of ing across the capacitor in an RC series cir-
the switch. With t expressed in seconds, R cuit. The equation for the current through
and C may be expressed in farads and ohms, the combination is:
or R and C may be expressed in microfarads
and megohms. The product RC is called the i = (1-F -il")
time constant of the circuit, and is expressed R
in seconds. As an example, if R is one meg- where i represents the current at any instant
ohm and C is one microfarad, the time through the series circuit, E represents the
constant RC will be equal to the product of applied voltage, and R represents the total
the two, or one second. resistance of the resistor and the d -c resist-
When the elapsed time (t) is equal to the ance of the inductor in series. Thus the time
time constant of the RC network under constant of the RL circuit is L /R, with R
consideration, the exponent of F becomes expressed in ohms and L expressed in henrys.
-1. Now F-1 is equal to 1/r, or 1/2.716,
which is 0.368. The quantity (1- 0.368) Voltage Decay When the switch in figure
then is equal to 0.632. Expressed as percent- 21 is moved to position 3
age, the above means that the voltage across after the capacitor has been charged, the
the capacitor will have increased to 63.2 per- capacitor voltage will drop in the manner
cent of the battery voltage in an interval shown in figure 21 -C. In this case the volt-
equal to the time constant or RC product age across the capacitor will decrease to 36.8
of the circuit. Then, during the next period percent of the initial voltage (will make
equal to the time constant of the RC com- 63.2 per cent of the total drop) in a period
bination, the voltage across the capacitor of time equal to the time constant of the
will have risen to 63.2 per cent of the re- RC circuit.
Figure 24
The previous chapter has been devoted to Frequency Spectrum At present the usable
a discussion of circuits and circuit elements frequency range for al-
upon which is impressed a current consisting ternating electrical currents extends over
of a flow of electrons in one direction. This the electromagnetic spectrum from about
type of unidirectional current flow is called 15 cycles per second to perhaps 30,000,-
direct current (abbreviated d -c or d.c.). 000,000 cycles per second. It is obviously
Equally as important in radio and communi- cumbersome to use a frequency designation
cations work and power practice is a type of in c.p.s. for enormously high frequencies,
current whose direction of electron flow so three common units which are multiples
reverses periodically. The reversal of flow of one cycle per second were established and
may take place at a low rate, in the case of are still used by many engineers.
power systems, or it may take place millions
of times per second, in the case of communi-
cations frequencies. This type of current
flow is called alternating current (abbrevi-
-
I-
ated a -c or a.c.). W
+
!r
www.americanradiohistory.com
The Sine Wave 47
1
II
II
I
IMI
.
LINES OF FONCE
111111j111IIIII;.T_-__
II II III11 I
ILII_
-
A s c o c
-- - ,
'MAX
I
IIIIIIIII1111 11111111
LINES Of PONCE
TIME
(uunroott. OENsITT)
Figure 3 I CYCLE = 3-
ZCYCLE =
OUTPUT OF THE ALTERNATOR
WHERE F . FREQUENCY IN CYCLES OR HERTZ
Graph showing sine -wave output current of
the alternator of figure 2. Figure 4
www.americanradiohistory.com
HANDBOOK A -C Relationships 49
When the arrow has traveled from point ing- current circuits. However, trigonometric
A to point E, it has traveled 90 degrees or tables are much more readily available in
one quarter cycle. The other three quadrants terms of degrees than radians, so the follow-
are not shown because their complementary ing simple conversions are useful.
or mirror relationship to the first quadrant
is obvious.
2a radians = 1 cycle = 360
It is important to note that time units are 7r radians = 1/2 cycle = 180
represented by degrees or quadrants. The
fact that AB, BC, CD, and DE are equal radians = 1/4 cycle = 90
chords (forming equal quadrants) simply 2
means that the arrow (conductor or vector)
is traveling at a constant speed, because
radians = 1/6 cycle = 60
3
these points on the radius represent the pas-
sage of equal units of time.
The whole picture can be represented in radians = 1/8 cycle = 45
4
another way, and its derivation from the
foregoing is shown in figure 3. The time
bise is represented by a straight line rather 1 radian = 2a cycle = 57.3
than by angular rotation. Points A, B, C,
etc., represent the same units of time as be- When the conductor in the simple alter-
fore. When the voltage corresponding to nator of figure 2 has made one complete
each point is projected to the corresponding revolution it has generated one cycle and
time unit, the familiar sine curve is the re- has rotated through 2rr radians. The expres-
sult. sion 2irf then represents the number of
The frequency of the generated voltage is radians in one cycle multiplied by the num-
proportional to the speed of rotation of the ber of cycles per second (the frequency) of
alternator, and to the number of magnetic the alternating voltage or current. The ex-
poles in the field. Alternators may be built pression then represents the number of ra-
to produce radio frequencies up to 30 kHz, dians per second through which the con-
and some such machines are still used for ductor has rotated. Hence 274 represents the
low- frequency communication purposes. By angular velocity of the rotating conductor,
means of multiple windings, three -phase or of the rotating vector, which represents
output may be obtained from large in- any alternating current or voltage, expressed
dustrial alternators. in radians per second.
In technical literature the expression 274
Radian Notation From figure 1 we see that is often replaced by m, the lower-case Greek
the value of an a -c wave letter omega. Velocity multiplied by time
varies continuously. It is often of impor- gives the distance travelled, so 2ir f t (or
tance to know the amplitude of the wave in (4) represents the angular distance through
terms of the total amplitude at any instant which the rotating conductor or the rotat-
or at any time within the cycle. To be able ing vector has travelled since the reference
to establish the instant in question we must time t = 0. In the case of a sine wave the
be able to divide the cycle into parts. We reference time t = 0 represents the instant
could divide the cycle into eighths, hun- when the voltage or the current, whichever
dredths, or any other ratio that suited our is under discussion, also is equal to zero.
fancy. However, it is much more convenient
mathematically to divide the cycle either Instantaneous Value The instantaneous volt -
into electrical degrees (360 represent one of Voltage or age or current is propor-
cycle) or into radians. A radian is an arc of Current tional to the sine of the
a circle equal to the radius of the circle; angle through which the
-or
hence there are 2irr radians per cycle per rotating vector has travelled since reference
circle (since there are zr diameters per cir- time t = 0. Hence, when the peak value of
cumference, there are 27r radii). the a -c wave amplitude (either voltage or
Both radian notation and electrical- degree current amplitude) is known, and the angle
notation are used in discussions of alternat- through which the rotating vector has
50 Alternating- Current Circuits THE RADIO
WHERE
when O = 90
e (THETA). PHASE ANGLE= 2 Fr FT sin 0 = 1.0
A = RADIANS OR RO
D=rr RADIANS OR 1eO soe =Erna=
G. AIL RADIANS OR 270
277 RADIANS OR SeO
D =
sin O = 0.8415
Figure 5 so e = 0.8415 E.a,
ILLUSTRATING RADIAN NOTATION Effective Value The instantaneous value
The radian is a unit of phase angle, equal te of an of an alternating cur -
57.324 degrees. It is commonly used in math- Alternating Current rent or voltage varies
ematical relationships involving phase angles
since such relationships are simplified when continuously throughout
radian notation is used. the cycle, so some value of an a -c wave
must be chosen to establish a relationship
travelled is established, the amplitude of the between the effectiveness of an a -c and a
wave at this instant can be determined d -c voltage or current. The heating value
through use of the following expression: of an alternating current has been chosen
to establish the reference between the ef-
e = Emaz Sin 271 t fective values of a.c. and d.c. Thus an
where, alternating current will have an effective
e equals the instantaneous voltage,
value of 1 ampere when it produces the
E equals maximum crest value of voltage,
same heat in a resistor as does 1 ampere of
f equals frequency in Hertz, direct current.
t equals period of time which has elapsed The effective value is derived by taking
since t = 0(expressed as a fraction of the instantaneous values of current over a
one second). cycle of alternating current, squaring these
values, taking an average of the squares,
The instantaneous current can be found and then taking the square root of the
from the same expression by substituting i average. By this procedure, the effective
for e and /mar for E0100. value becomes known as the root mean
It is often easier to visualize the process square, or rms, value. This is the value that
of determining the instantaneous amplitude is read on a -c voltmeters and a -c ammeters.
by ignoring the frequency and considering The rms value is 70.7 percent of the peak
only one cycle of the a -c wave. In this case, or maximum instantaneous value (for sine
for a sine wave, the expression becomes: waves only) and is expressed as follows:
e = E. sin O
EFrr or E,.,,,.1 = 0.707 X E,,,,,s,or
where O represents the angle through which lrrr or /rm., = 0.707 X Inns
the vector has rotated since time (and am-
plitude) were zero. As examples: The following relations are extremely use-
ful in radio and power work:
when O = 30
E,.n,, = 0.707 X Emex, and
sin O = 0.5
E,,,,,s = 1.414 X Enos
so e = 0.5 E,naz
Rectified Alternating If an alternating cur -
when 0=60 Current or Pulsat- rent is passed through a
ing Direct Current rectifier, it emerges in
sin = 0.866
O
the form of a current
so e = 0.866 E. of varying amplitude which flows in one
direction only. Such a current is known as
www.americanradiohistory.com
HANDBOOK Reactance 51
Figure 6
peak = 1i 7 X average
shown in figure 7. Xo
27rfC
XL = 2TrfL
where,
where,
Xe equals capacitive reactance in ohms,
XL equals inductive reactance expressed a equals 3.1416,
in ohms, f equals frequency in Hertz,
ITequals 3.1416 (27r = 6.283), C equals capacitance in farads.
f equals frequency in Hertz,
L equals inductance in henrys.
Capacitive Re- Here again, as in the case
Inductive Reactance It is very often neces- actance at of inductive reactance,
at Radio Frequencies sary to compute induc- Radio Frequencies the units of capacitance
tive reactance at radio and frequency can be
frequencies. The same formula may be used, converted into smaller units for practical
but to make it less cumbersome the induct- problems encountered in radio work. The
ance is expressed in millihenrys and the equation may be written:
frequency in kilohertz. For higher frequen-
cies and smaller values of inductance, fre-
1,000,000
Xc =
quency is expressed in megahertz and induct- 2zrfC
ance in microhenrys. The basic equation where,
need not be changed, since the multiplying
factors for inductance and frequency appear f equals frequency in megahertz,
in numerator and denominator, and hence C equals capacitance in picofarads.
are cancelled out. However, it is not pos- In the audio range it is often convenient to
sible in the same equation to express L in express frequency (f) in Hertz and capac-
millihenrys and f in Hertz without con- itance (C) in microfarads, in which event
version factors. the same formula applies.
E
Phase When an alternating current flows
through a purely resistive circuit, it
will be found that the current will go
TIME through maximum and minimum in perfect
step with the voltage. In this case the cur-
rent is said to be in step, or in phase with
the voltage. For this reason, Ohm's Law will
apply equally well for a.c. or d.c. where
CURRENT LEADING VOLTAGE BY 90
(CIRCUIT CONTAINING PURE CAPACITANCE ONLY)
pure resistances are concerned, provided that
the same values of the wave (either peak or
Figure 8 rms) for both voltage and current are used
LEADING PHASE ANGLE in the calculations.
However, in calculations involving alter-
Showing the manner in which the nt
leads the voltage in an a-c circuit containing nating currents the voltage and current are
pure capacitance only. The lead is equal to not necessarily in phase. The current
one -quarter cycle or 90 degrees. through the circuit may lag behind the
voltage, in which case the current is said to
Capacitive It has been explained that induc- have lagging phase. Lagging phase is caused
Reactance tive reactance is the measure of by inductive reactance. If the current
the ability of an inductor to of- reaches its maximum value ahead of the
fer impedance to the flow of an alternating voltage (figure 8) the current is said to
current. Capacitors have a similar property have a leading phase. A leading phase angle
although in this case the opposition is to is caused by capacitive reactance.
any change in the voltage across the capaci- In an electrical circuit containing re-
tor. This property is called capacitive react- actance only, the current will either lead or
ance and is expressed as follows: lag the voltage by 90 . If the circuit con-
www.americanradiohistory.com
HANDBOOK Impedance 53
tains inductive reactance only, the current combination the resulting phase angle of
will lag the voltage by 90 . If only capaci- current flow with respect to the impressed
tive reactance is in the circuit, the current voltage lies somewhere between plus or
will lead the voltage by 90 . minus 90 and 0 depending on the relative
Reactances
magnitudes of the reactance and the resist-
Inductive and capacitive re- ance.
in Combination actance have exactly op- The term impedance is a general term
posite effects on the phase which can be applied to any electrical entity
relation between current and voltage in a which impedes the flow of current. Hence
circuit. Hence when they are used in com- the term may be used to designate a resist-
bination their effects tend to neutralize. The ance, a pure reactance, or a complex corn -
combined effect of a capacitive and an in- bination of both reactance and resistance.
ductive reactance is often called the net re- The designation for impedance is Z. An im-
actance of a circuit. The net reactance (X) pedance must be defined in such a manner
is found by subtracting the capacitive that both its magnitude and its phase angle
reactance from the inductive reactance are established. The designation may be ac-
(X = X1, - X0)
The result of such a combination of pure
complished in either of two ways -one of
which is convertible into the other by
reactances may be either positive, in which simple mathematical operations.
case the positive reactance is greater so that
the net reactance is inductive, or it may be The j Operator The first method of des-
negative in which case the capacitive react- ignating an impedance is
ance is greater so that the net reactance is actually to specify both the resistive and the
capacitive. The net reactance may also be reactive component in the form R + X.
zero in which case the circuit is said to be In this form R represents the resistive com-
resonant. The condition of resonance will be ponent in ohms and X represents the re-
discussed in a later section. Note that in- active component. The j merely means that
ductive reactance is always taken as being the X component is reactive and thus can-
positive while capacitive reactance is always not be added directly to the R component.
taken as being negative. Plus jX means that the reactance is positive
Impedance; Circuits
or inductive, while if minus jX were given
Pure reactances intro - it would mean that the reactive component
Containing Reactance duce a phase angle of was negative or capacitive.
and Resistance 90 between voltage In figure 9 we have a vector (+A) lying
and current; pure re- along the positive X -axis of the usual X -Y
sistance introduces no phase shift between coordinate system. If this vector is multi-
voltage and current. Hence we cannot add plied by the quantity ( -1) , it becomes
a reactance and a resistance directly. When
a reactance and a resistance are used in
(- A) and its position now lies along the
X -axis in the negative direction. The opera-
tor (-1) has caused the vector to rotate
Y-sxIa
`f
(+A)* (-
VECTOII
ROTATES
TNWGM
(- 1 ) is equal to ( X Ji)
through an angle of 180 degrees. Since
, the
same result may be obtained by operating on
IRO* ` \ the vector with the operator ( X
X AMR
11-1) However if the vector is operated
.
www.americanradiohistory.com
54 Alternating- Current Circuits THE RADIO
Polar Notation The second method of repre- Y-A%IS
senting an impedance is to
specify its absolute magnitude and the phase
angle of current with respect to voltage, in
(+A)
'
C TO rNRWGM
v 90
Es
X Z' 4+J3
tan B =
R
Ji<
+ IzI 4Le.^-'i
x" 2 LSe.e! IZ I` 5 ran
One common problem is that of determining
the scalar magnitude of the impedance, IZI, 0I RESISTANCE -R IZI 5 se.es
L R=ONMS
and the phase angle B, when resistance and
reactance are known; hence, of converting
from the Z = R + jX to the IZI L B form. Figure 11
In this case we use two of the expressions THE IMPEDANCE TRIANGLE
just given:
Showing the graphical construction of a tri-
angle for obtaining the net (scalar) impedance
= R2 + X2 resulting from the connection of a resistance
and a reactance in series. Shown also along-
side is the alternative mathematical pro-
tan = cedure for obtaining the values associated
O
R,(or8=tatR )
with the triangle.
www.americanradiohistory.com
HANDBOOK Vector Algebra 55
ohm; hence, the ohm may be thought of as Multiplication and It is often necessary in
a unit of opposition to current flow, with- Division of solving certain types of
out reference to the relative phase angle be- Complex Quantities circuits to multiply or
tween the applied voltage and the current divide two complex
which flows. quantities. It is a much simpler mathemati-
Further, since both capacitive and in- cal operation to multiply or divide complex
ductive reactance are functions of fre- quantities if they are expressed in the polar
quency, impedance will vary with fre- form. Hence if they are given in the rec-
quency. Figure 12 shows the manner in tangular form they should be converted to
which IZI will vary with frequency in an the polar form before multiplication or
RL series circuit and in an RC series circuit. division is begun. Then the multiplication is
Series RLC Circuits
accomplished by multiplying the IZI terms
In a series circuit con- together and adding algebraically the L O
taining R, L, and C, the terms, as:
impedance is determined as discussed before
except that the reactive component in the (1Z11 Le,) (1Z2! Lee) =
expressions defines the net reactance -that IZ1I 1221 (Le, + Le2)
is,the difference between XL and Xc. Hence
( XL- Xe) may be substituted for X in For example, suppose that the two imped-
the equations. Thus: ances 1201 L43 and 1321 L -23 are to be
multiplied. Then:
1ZI=VR2+ ()CL -Xc)2 (1201 L43 ) (1321 L -23 ) =120.321
(L43 + L -23 )
O =tan - 1(X1 -Xc) = 640 L20
Division is accomplished by dividing the
A series RLC circuit thus may present denominator into the numerator, and sub-
an impedance which is capacitively reactive
tracting the angle of the denominator from
if the net reactance is capacitive, inductively that of the numerator, as:
reactive if the net reactance is inductive, or
resistive if the capacitive and inductive re-
actances are equal.
Addition of The addition of complex
Complex Quantities quantities (for example,
impedances in series) is
quite simple if the quantities are in the rec-
tangular form. If they are in the polar form
they only can be added graphically, unless
they are converted to the rectangular form
by the relationships previously given. As an
example of the addition of complex quanti-
ties in the rectangular form, the equation
for the addition impedance is: o
www.americanradiohistory.com
HANDBOOK Vector Algebra 57
ZTotat
1
Z1 Z,
1 1
Z:,
= 1.85 - j 2.77
or when only two impedances are involved: Equivalent Series Through the series of op-
Circuit erations in the previous
ZTotal - Z,
Z,Z2
+ Z,
paragraph we have con-
verted a circuit composed of two imped-
As an example, using the two -impedance ances in parallel into an equivalent series cir-
relation, take the simple case, illustrated in cuit composed of impedances in series. An
figure 15, of a resistance of 6 ohms in paral- equivalent series circuit is one which, as
lel with a capacitive reactance of 4 ohms. far as the terminals are concerned, acts iden-
To simplify the first step in the computa- tically to the original parallel circuit; the
tion it is best to put the impedances in the current through the circuit and the power
polar form for the numerator, since multi- dissipation of the resistive elements are the
plication is involved, and in the rectangular same for a given voltage at the specified
form for the addition in the denominator. frequency.
We can check the equivalent series
Zrotal - (6 L0 ) (4 L -90 ) circuit of figure 15 with respect to the
6 -j4 original circuit by assuming that one volt
a -c (at the frequency where the capacitive
24 L -90 reactance in the parallel circuit is 4 ohms)
6 -j4 is applied to the terminals of both the series
and parallel circuits.
In the parallel circuit the current through
the resistor will be % ampere (0.166 amp)
1.11511 while the current through the capacitor will
-J 40. be j 1/4 ampere (+ j 0.25 amp) . The total
-J 2.nn current will be the sum of these two cur-
o T rents, or 0.166 + j 0.25 amp. Adding these
vectorially we obtain:
PARALLEL EQUIVALENT SERIES
CIRCUIT CIRCUIT
= \/0.1662 + 0.252 = 0.09
Figure 15 = 0.3 amp.
THE EQUIVALENT SERIES CIRCUIT The dissipation in the resistor will be 12/6
= 0.166 watts.
Showing a parallel RC circuit and the equiv-
alent series RC circuit which represents the In the case of the equivalent series cir-
same net impedance as the parallel circuit. cuit the current will be:
www.americanradiohistory.com
HANDBOOK Voltage Dividers 59
0.3 amp
III IZ 3.33 =
And the dissipation in the resistor will be:
W= 12R= 0.32 X1.85
EzE. EzE. XCz Ea E. L:
L.+Lz
= 0.09 X 1.85 RI. Rz XC. +XCa
C.
Ez E.
tCz
= 0.166 watts C.
C)
So we see that the equivalent series circuit
checks exactly with the original parallel cir- Figure 16
cuit.
SIMPLE A -C VOLTAGE DIVIDERS
Parallel RLC In solving a more complicated
Circuits circuit made up of more than
two impedances in parallel we in phase with the input voltage. By using
may elect to use either of two methods of combinations of different types of imped-
solution. These methods are called the ad- ances, the phase angle of the output may be
mittance method and the assumed- voltage shifted in relation to the input phase angle
method. However, the two methods are at the same time the amplitude is reduced.
equivalent since both use the sum-of- recip- Several dividers of this type are shown in
rocals equation: figure 17. Note that the ratio of output
voltage is equal to the ratio of the output
1 1
+
1 1 impedance to the total divider impedance.
ZTotal Z1 Z2 Z:4 This relationship is true only if negligible
current is drawn by a load on the output
In the admittance method we use the rela-
termina Is.
tion Y = I /Z, where Y =
G + jB; Y is
called the admittance, defined above, G is
the conductance or R /Z2 and B is the sus -
cebtance or -X
/Z2. Then Ytotal = 1 /Ztotal
= Y1 + Y2 + Y3 ....
In the assumed -
voltage method we multiply both sides of
the equation above by E, the assumed volt-
age, and add the currents, as:
E xc XL
EEE...=1Z1+122+IZ3... Ea. E. E: E.
ZTotal Z1 Z2 Z3 R2+xCa Rz +XLz
Figure 18
www.americanradiohistory.com
HANDBOOK Resonant Circuits 61
f
25,330
LC
or L = 25,330 or C
f2C
- 25,330
f -L
The complete equations are:
E
where,
I- 1 r2 + (XL - Xc) 2
a'?,
resonant frequency, the difference between El'`
the reactances will be a definite quantity FREQUENCY
and will add with the resistance to make ohe
impedance higher and higher as the circuit Figure 20
is tuned off the resonant frequency. RESONANCE CURVE
If Xe should be greater than X,,, then the
term (X1, -
X0) will give a negative num- Showing the increase in impedance at reson-
ance for a parallel -resonant circuit, and
ber. However, when the difference is squared similarly, the increase in t at resonance
the product is always positive. This means for a series -resonant circuit. She sharpness of
once is determined by the Q of the cir-
that the smaller reactance is subtracted cuit, as illustrated by a comparison between
from the larger, regardless of whether it be the three curves.
62 Alternating- Current Circuits THE RADIO
current against the frequency either side Skin Effect The actual resistance in a wire
of resonance, the resultant curve becomes or an inductor can be far greater
what is known as a resonance curve. Such a than the d -c value when the coil is used in a
curve is shown in figure 20, the frequency radio -frequency circuit; this is because the
being plotted against current in the series - current does not travel through the entire
resonant circuit. cross section of the conductor, but has a
Several factors will have an effect on the tendency to travel closer and closer to the
shape of this resonance curve, of which re- surface of the wire as the frequency is in-
sistance and L -to -C ratio are the important creased. This is known as the skin effect.
considerations. The lower curves in figure The actual current carrying portion of the
20 show the effect of adding increasing val- wire is decreased as a result of the skin
ues of resistance to the circuit. It will be effect so that the ratio of a -c to d -c resist-
seen that the peaks become less and less ance of the wire, called the resistance ratio,
prominent as the resistance is increased; is increased. The resistance ratio of wires to
thus, it can be said that the selectivity of be used at frequencies below about 500 kHz
the circuit is thereby decreased. Selectivity may be materially reduced through the use
in this case can be defined as the ability of of litz wire. Litz wire, of the type common-
a circuit to discriminate against frequencies ly used to wind the coils of 45 5 - kHz i -f
transformers, may consist of 3 to 10 strands
adjacent to (both above and below) the
resonant frequency. of insulated wire, about No. 40 in size, with
the individual strands connected together
Voltage Across Coil Because the a -c or r -f only at the ends of the coils.
and Capacitor in voltage across a coil and Variation of Q Examination of the equation
Series Circuit capacitor is proportional with Frequency for determining Q might
to the reactance (for a
given current), the actual voltages across give rise to the thought that
the coil and across the capacitor may be even though the resistance of an inductor
many times greater than the terminal volt- increases with frequency, the inductive re-
age of the circuit. At resonance, the voltage
actance does likewise, so that the Q might
be a constant. Actually, however, it works
across the coil (or the capacitor) is Q
times the applied voltage. Since the Q (or
out in practice that the Q of an inductor
merit factor) of a series circuit can be in will reach a relatively broad maximum at
some particular frequency. Hence, coils nor-
the neighborhood of 100 or more, the volt-
age across the capacitor, for example, may
mally are designed in such a manner that the
peak in their curve of Q versus frequency
be high enough to cause flashover, even
will occur at the normal operating frequen-
though the applied voltage is of a value con-
cy of the coil in the circuit for which it is
siderably below that at which the capacitor
designed.
is rated.
Circuit Q - Sharp.
ness of Resonance
An extremely important
property of a capacitor
The Q of a capacitor ordinarily is much
higher than that of the best coil. Therefore,
it usually is the merit of the coil that limits
the over -all Q of the circuit.
or an inductor is its fac- At audio frequencies the core losses in an
tor-of-merit, more generally called its Q. iron -core inductor greatly reduce the Q
It is this factor, Q, which primarily deter- from the value that would be obtained sim-
mines the sharpness of resonance of a tuned
ply by dividing the reactance by the resist-
circuit. This factor can be expressed as the ance. Obviously the core losses also represent
ratio of the reactance to the resistance, as
circuit resistance, just as though the loss
follows:
occurred in the wire itself.
27rfL Parallel
Q In radio circuits, parallel reso-
R Resonance nance (more correctly termed
where, antiresonance) is more frequent-
ly encountered than series resonance; in fact,
R equals total resistance. it is the basic foundation of receiver and
www.americanradiohistory.com
HANDBOOK Circuit Q 63
T LOOlC CONrUnG
NiGN a
CI% MAL COUPLING
LOW O
O
Figue 22
EFFECT OF COUPLING ON CIRCUIT IMPEDANCE AND Q
www.americanradiohistory.com
HANDBOOK Nonsinusoidal Waves 65
The plate current is supplied in short which ordinarily are very low. The unloaded
pulses which do not begin to resemble a sine Q of a good quality large diameter tank coil
wave, even though the grid may be excited in the high- frequency range may be as high
by a sine -wave voltage. These spurts of as 500 to 800, and values greater than 300
plate current are converted into a sine wave are quite common.
in the plate tank circuit by virtue of the
Q or flywheel effect of the tank. Tank- Circuit Since the unloaded Q of a tank
If a tank did not have some resistance Efficiency circuit is determined by the
losses, it would, when given a "kick" with a minimum losses in the tank,
single pulse, continue to oscillate indefinitely. while the loaded Q is determined by useful
With a moderate amount of resistance or loading of the tank circuit from the external
"friction" in the circuit the tank will still load in addition to the internal losses in the
have inertia, and continue to oscillate with tank circuit, the relationship between the
decreasing amplitude for a time after being two Q values determines the operating effi-
given a "kick." With such a circuit, almost ciency of the tank circuit. Expressed in the
pure sine -wave voltage will be developed form of an equation, the loaded efficiency of
across the tank circuit even though power a tank circuit is:
is supplied to the tank in short pulses or
spurts, so long as the spurts are evenly
spaced with respect to time and have a fre-
Tank efficiency = 1 - Q.,
' X 100
SQUARE WAVE
O ` \ I
FUND. PLUS 2M0. 3RO
AND TM HARYONIC4
FUND. PLUS 2NC IRO, sTN.
15TH. /TH. AND TTN HARMS.
\ !
FUNDAMENTAL PLUS
HARMONIC
FUNDAMENTAL
3RD HARMONIC
\y/
Figure 24
Figure 25
7TH HARMONIC
called harmonics, and are always a whole
number of times higher than the funda-
mental. For example, the frequency twice as
high as the fundamental is called the second
harmonic.
The Square Wave Figure 23 compares a Figure 27
square wave with a sine
wave (A) of the same frequency. If another COMPOSITION OF A PEAKED WAVE
www.americanradiohistory.com
HAN D BO O K Nonsinusoidal Waves 67
rs0v
A -C Transient CircuitsIf an a -c voltage is eft o
substituted for the d -c -50V.
input voltage in the RC transient circuits
discussed in Chapter 2, the same principles
may be applied in the analysis of the tran- i i
I
INTEGRATOR
F100V eE INTEGRATOR OUTPUT 100V OUTPUT (ec
(PEAK) (PEAK)
-
100HERTZ 1000 HERTZ R= 35011. DIFFERENTIATOR
eR OIRERENTIATOR OUTPUT OUTPUT (e0)
+IDO
0o v.
-00
+125V.
T5 V.
CO / /r pU- U10
EIITEIIdTAT011'(R)
OUTPUT OF
DIFFERENTIATOR (es)
-- -125V.
e0 OUTPUT OP
INTEGRATOR (ec)
+25 V.
ED 1TC MOT ec)
-25 V. 11
Figure 30 Figure 31
www.americanradiohistory.com
HANDBOOK Transformers 69
O
Figure 33
Amplifier deficient in low -frequency response will distort square wave applied to the input
circuit, as shown. A 60-Hz square wave may be used.
A: Drop in gain at low frequencies
a: Leading phase shift at low frequencies
C: Lagging phase shift at low frequencies
D: Accentuated low- frequency gain
Figure 34 Figure 35
Output waveshape of amplifier having defi- Output waveshape of amplifier having limited
ciency in high -frequency response. Tested low- frequency and high -frequency response.
with 10 -kHz square wave. Tested with I kHz s wave.
ing current is constant. The output voltage Low -frequency and high- frequency re-
of the differentiator, therefore, is constant sponse, as well as transient response can be
during each half of the sawtooth input. examined easily.
If the amplifier is deficient in low -fre-
Miscellaneous Various voltage waveforms quency response, the flat top of the square
Inputs other than those represented wave will be canted, as in figure 33. If the
here may be applied to short - high- frequency response is inferior, the rise
time- constant RC circuits for the purpose of time of the output wave will be retarded
producing across the resistor an output volt- (figure 34).
age with an amplitude proportional to the An amplifier with a limited high- and
rate of change of the input signal. The low- frequency response will turn the square
shorter the RC time constant is made with wave into the approximation of a sawtooth
respect to the period of the input wave, the wave (figure 3 f) .
more nearly the voltage across the capacitor
conforms to the input voltage. Thus, the Transformers
3 -4
differentiator output becomes of particular
importance in very short - time- constant RC When two coils are placed in such induc-
circuits. Differentiator outputs for various tive relation to each other that the lines of
types of input waves are shown in figure 32. force from one cut across the turns of the
Square -Wave Test The application of a other inducing a current, the combination
for Audio Equipment square -wave input sig- can be called a transformer. The name is
nal to audio equipment, derived from the fact that energy is trans-
and the observation of the reproduced out- formed from one winding to another. The
put signal on an oscilloscope will provide inductance in which the original flux is
a quick and accurate check of the over -all produced is called the primary; the induct-
operation of audio equipment. ance which receives the induced current is
www.americanradiohistory.com
70 Alternating- Current Circuits THE RADIO
Types of Transformers are used in al- Leakage Reactance Since unity coupling does
Transformers ternating- current circuits to not exist in a practical
transfer power at one voltage transformer, part of the flux passing from
and impedance to another circuit at another the primary circuit to the secondary circuit
voltage and impedance. There are three main follows a magnetic circuit acted on by the
classifications of transformers: those made primary only. The same is true of the sec-
for use in power- frequency circuits, those ondary flux. These leakage fluxes cause leak-
made for audio -frequency applications, and age reactance in the transformer, and tend
those made for radio frequencies. to cause the transformer to have poor volt-
age regulation. To reduce such leakage re-
The Transformation In a perfect transformer actance, the primary and secondary windings
Ratio all the magnetic flux should be in close proximity to each other.
lines produced by the The more expensive transformers have inter-
primary winding link every turn of the leaved windings to reduce inherent leakage
secondary winding. For such a transformer, reactance.
the ratio of the primary and secondary volt-
ages is exactly the same as the ratio of the Impedance In the ideal transformer, the
number of turns in the two windings: Transformation impedance of the secondary
load is reflected back into
Np_ Ep
the primary winding in the following rela-
Ns Es tionship:
where, Zr = NzZs, or N = Zp/Zs
Np equals number of turns in the primary,
Ns equals number of turns in the sec- where,
ondary, Zr equals reflected primary impedance,
Er equals voltage across the primary, N equals turns ratio of transformer,
Es equals voltage across the secondary. Zs equals impedance of secondary load.
www.americanradiohistory.com
HANDBOOK Electric Filters 71
ELEMENTARY FILTER SECTIONS Filter Sections The most common filter sec-
L-iLCT1ON4 7-NETWORK
tion is one in which the two
impedances ZA and ZB are so related that
their arithmetical product is a constant: ZA
X ZB = k2 at all frequencies. This type of
:. filter section is called a constant -k section.
T A section having a sharper cutoff fre-
Po NCTWORR
quency than a constant -k section, but less
attenuation at frequencies far removed from
cutoff is the m- derived section, so called
because the shunt or series element is reso-
nated with a reactance of the opposite sign.
If the complementary reactance is added
to the series arm, the section is said to be
Figure 38 shunt derived; if added to the shunt arm,
Complex Alters may for made up from those series derived. Each impedance of the m-
basic Alter sections. derived section is related to a corresponding
impedance in the constant -k section by some
illustrated in figure 38. A definite number of factor which is a function of the constant
L sections may be combined into basic filter m. In turn, m is a function of the ratio
sections, called T networks or 7r networks, between the cutoff frequency and the fre-
also shown in figure 38. Both the T and Tr quency of infinite attenuation, and will
networks may be divided in two to form have some value between zero and one. As
half -sections. the value of m approaches zero, the sharp-
LOW -PASS SHUNT- DERIVED FILTER HON -PASS SERIES -DERIVED FILTER
(SERIES ARM RESONATED) (SHUNT ARM RESONATED)
CI
2
2C, 2C1
T2
f2 1,C 14
FREQUENCY FREQUENCY
L2
LMnf2 R
m2 1
CM
2
LM.
4n11
R
m2 11
CM
4n11R
Figure 39
TYPICAL LOW -PASS AND HIGH -PASS FILTERS, ILLUSTRATING SHUNT AND SERIES
DERIVATIONS
www.americanradiohistory.com
HANDBOOK Electric Filters 73
R LOAD RESISTANCE
' :Flo
ICT- -T
12CUTOFF FREQUENCY I O p O '
6 I
LI
fFREQUENCY OF VERY fC2 C2 }C2 /CT ti iC2 '
HIGH ATTENUATION
o T T o a o ; o
Ly R
LILy LI 0.6Lymly
f2
C2 Cy CI 0.267 Cy
I<m'y SAME MOLUES AS m 0.6
LOW PASS n
C2 0.6 CymCy
C
y nf2R
I
4
r f4
6
m I- W
z j SAME CURVE AS m 0 6
.0.6 W
IQ ) 12
< 12
FREOUENCY FREQUENCY
I
I
1 iREOUEyKr OF VERY
HIGH ATTENUATION
2L2 2L2 2L2- - 2L2 2L2 I
m
I 2L2
O O
Lk
R LI 3.75 Ly 2%
9m
m
Ly
I I
CICy I
Figure 40
Through the use of the curves and equations which accompany the diagrams in the illustra-
tion above it is possible to determine the c lues of inductance and capacitance for
the usual typos of pi- section Alters.
ness of cutoff increases, but the less will be loss. This attenuation is usually uniform to
the attenuation at several times cutoff fre- all frequencies within the passband. The
quency. A value of 0.6 may be used for in insertion loss varies with the type of filter,
in most applications. The "notch" frequency the Q of the components, and the type of
is determined by the resonant frequency of termination employed.
the tuned filter element. The amount of
attenuation obtained at the "notch" when Electric Filter Electric wave filters have long
a derived section is used is determined by Design been used in some amateur sta-
the effective Q of the resonant arm (fig- tions in the audio channel to
ure 39). reduce the transmission of unwanted high
frequencies and hence to reduce the band-
Filter Assembly Constant -k sections and m- width occupied by a radiophone signal. The
derived sections may be cas- effectiveness of a properly designed and
caded to obtain the combined characteristics properly used filter circuit in reducing QRM
of sharp cutoff and good remote frequency and sideband splatter should not be under-
attenuation. Such a filter is known as a estimated.
composite filter. The amount of attenuation In recent years, high- frequency filters
will depend on the number of filter sec- have become commonplace in TVI reduction.
tions used, and the shape of the transmission High -pass type filters are placed before the
curve depends on the type of filter sec- input stage of television receivers to reject
tions used. All filters have some insertion the fundamental signal of low- frequency
74 Alternating- Current Circuits
transmitters. Low -pass filters are used in the commonly used type, figure 40 gives design
output circuits of low- frequency transmit- data and characteristics for this type of
ters to prevent harmonics of the transmitter filter.
from being radiated in the television chan- The image -parameter technique of filter
nels. design outlined in this section is being super -
The chart of figure 40 gives design data ceded by modern network synthesis, which
and procedure on the 7r section type of takes advantage of the digital computer as a
filter. The in- derived sections with an m of tool for multisection filter design. Filters
0.6 will be found to be most satisfactory as designed by this new technique provide su-
the input section (or half- section) of the perior performance with less components
usual filter since the input impedance of such than equivalent filters designed by the image -
a section is most constant over the passband parameter scheme. Design tables for synthesis
of the filter section. systems may be found in Simplified Modern
Simple filters may use either L, T, or 7r Filter Design by Geffe, published by John F.
sections. Since the 7r section is the more Rider Publisher, Inc., New York.
www.americanradiohistory.com
CHAPTER FOUR
In the previous chapters we have seen the electrons which have been emitted. The
manner in which an electric current flows electron source is called the cathode, and the
through a metallic conductor as a result of electron collector is usually called the anode.
an electron drift. This drift, which takes Some external source of energy must be ap-
place when there is a difference in potential plied to the cathode in order to impart suffi-
between the ends of the metallic conductor, cient velocity to the electrons within the
is in addition to the normal random electron cathode material to enable them to overcome
motion between the molecules of the con- the surface forces and thus escape into the
ductor. surrounding medium. In the usual types of
The electron may be considered as a min- electron tubes the cathode energy is applied
ute negatively charged particle, having a in the form of heat; electron emission from
mass of 9 X 10 -28 gram, and a charge of a heated cathode is called thermionic emis-
1.59 X 10-19 coulomb. Electrons are always sion. In another common type of electron
identical, regardless of the source from tube, the photoelectric cell, energy in the
which they are obtained. form of light is applied to the cathode to
An electric current can be caused to cause photoelectric emission.
flow through other media than a metallic
conductor. One such medium is an ionized
solution, such as the sulfuric acid electrolyte 4 -1 Thermionic Emission
in a storage battery. This type of current
flow is called electrolytic conduction. Fur- Electron Emission of electrons from the
ther, it was shown at about the turn of the Emission cathode of a thermionic electron
century that an electric current can be tube takes place when the cath-
carried by a stream of free electrons in an ode of the tube is heated to a tempera-
evacuated chamber. The flow of a current ture sufficiently high that the free elec-
in such a manner is said to take place by trons in the emitter have sufficient velocity
electronic conduction. The study of elec- to overcome the restraining forces at the
tron tubes (also called vacuum tubes, or surface of the material. These surface forces
valves) is actually the study of the control vary greatly with different materials. Hence
and use of electronic currents within an different types of cathodes must be raised
evacuated or partially evacuated chamber. to different temperatures to obtain adequate
Since the current flow in an electron tube quantities of electron emission. The several
takes place in an evacuated chamber, there types of emitters found in common types of
must be located within the enclosure both transmitting and receiving tubes will be
a source of electrons and a collector for the described in the following paragraphs.
76 Vacuum -Tube Principles THE RADIO
Figure 1
Cathode Types The emitters or cathodes as tubes. Tungsten is not adversely affected by
used in present -day thermi- such bombardment.
onic electron tubes may be classified into
The Thoriated- In the course of experi-
two groups; the directly heated or fila-
Tungsten Filament ments made upon tung-
ment type and the indirectly heated or
heater- cathode type. Directly heated emit- sten emitters, it was found
ters may be further subdivided into three that filaments made from tungsten having a
important groups, all of which are common- small amount of thoria (thorium oxide) as
ly used in modern vacuum tubes. These an impurity had much greater emission than
classifications are: the pure- tungsten fila- those made from the pure metal. Subsequent
ment, the thoriated- tungsten filament, and development has resulted in the highly effi-
the oxide -coated filament. cient carburized thoriated- tungsten filament
as used in many medium -power trans-
The Pure-Tung- Pure -tungsten wire was used mitting tubes today.
sten Filament as the filament in nearly all Thoriated - tungsten emitters consist of a
the earlier transmitting and tungsten wire containing from 1% to 2%
receiving tubes. However, the thermionic thoria. The activation process varies between
efficiency of tungsten wire as an emitter different manufacturers of vacuum tubes,
(the number of milliamperes emission per but it is essentially as follows: (1) the tube
watt of filament -heating power) is quite is evacuated; (2) the filament is burned for
low; the filaments become fragile after use; a short period at about 2860 Kelvin to
their life is rather short, and they are sus- clean the surface and reduce some of the
ceptible to burnout at any time. Pure -tung- thoria within the filament to metallic thor-
sten filaments must be run at bright white ium; (3) the filament is burned for a longer
heat (about 2500 Kelvin). For these rea- period at about 2100 Kelvin to form a
sons, tungsten filaments have been replaced layer of thorium on the surface of the
in all applications where another type of tungsten; (4) the temperature is reduced
filament could be used. They are, however, to about 1600 Kelvin and some pure hy-
still often employed in large water -cooled drocarbon gas is admitted to form a layer
tubes and in certain large, high -power of tungsten carbide on the surface of the
air -cooled triodes where another filament tungsten. This layer of tungsten carbide re-
type would be unsuitable. Tungsten fila- duces the rate of thorium evaporation from
ments are the most satisfactory for high - the surface at the normal operating temper-
power, high -voltage tubes where the emitter ature of the filament and thus increases the
is subjected to positive ion bombardment operating life of the vacuum tube. Tho-
caused by the residual gas content of the rium evaporation from the surface is a natu-
www.americanradiohistory.com
HANDBOOK Types of Emitters 77
ONE INCH
Figure 2
www.americanradiohistory.com
78 Vacuum -Tube Principles THE RADIO
voltage for a few more hours. This should
bring it back almost to normal. If the tube
checks still very low after the first attempt
at reactivation, the complete process can
be repeated as a last effort.
The Oxide- The most efficient of all
Coated Filament modern filaments is the
oxide- coated type which
consists of a mixture of barium and stron-
tium oxides coated on a nickel alloy wire or
strip. This type of filament operates at a Figure 3
dull -red to orange -red temperature (1050 CUTAWAY DRAWING OF A .5C4 TRIODE
to 1170 K) at which temperature it will
emit large quantities of electrons. The oxide -
would be destroyed in an exceedingly short
coated filament is somewhat more efficient
than the thoriated- tungsten type in small period of time.
The activation of oxide -coated filaments
sizes and it is considerably less expensive to
also varies with tube manufacturers but
manufacture. For this reason all receiving
consists essentially in heating the wire which
tubes and quite a number of the low -pow-
has been coated with a mixture of barium
ered transmitting tubes use the oxide- coated
and strontium carbonates to a temperature
filament. Another advantage of the oxide -
coated emitter is its extremely long life
the average tube can be expected to run
- of about 1500 Kelvin for a time and then
applying a potential of 100 to 200 volts
through a protective resistor align limit the
from 3000 to 5000 hours, and when loaded
very lightly, tubes of this type have been emission current. This process thermally
known to give 50,000 hours of life before reduces the carbonates to oxides, cleans the
filament surface of foreign materials, and
their characteristics changed to any great
extent. activates the cathode surface.
Oxide filaments are unsatisfactory for use
Reactivation of oxide -coated filaments is
at very high plate voltage because: (1)
not possible since there is always more than
sufficient reduction of the oxides and diffu-
their activity is seriously impaired by the
sion of the metals to the surface of the
high temperature necessary to de -gas the
filament to meet the emission needs of the
high -voltage tubes and, (2) the positive ion
cathode.
bombardment which takes place even in the
best evacuated high -voltage tube causes de- The Heater The heater -type cathode was de-
struction of the oxide layer on the surface of Cathode veloped as a result of the re-
the filament. quirement for a type of emitter
Oxide -coated emitters have been found which could be operated from alternating
capable of emitting an enormously large current and yet would not introduce a -c
current pulse with a high applied voltage ripple modulation even when used in low -
for a very short period of time without level stages. It consists essentially of a small
damage. This characteristic has proved to nickel -alloy cylinder with a coating of
be of great value in radar work. For example, strontium and barium oxides on its surface
the relatively small cathode in a microwave similar to the coating used on the oxide -
magnetron may be called on to deliver 25 to coated filament. Inside the cylinder is an
50 amperes at an applied voltage of perhaps insulated heater element consisting usually
25,000 volts for a period in the order of one of a double spiral of tungsten wire. The
microsecond. After this large current pulse heater may operate on any voltage from 2
has been passed, plate voltage normally will to 117 volts, although 6.3 is the most com-
be removed for 1000 microseconds or more mon value. The heater is operated at quite
so that the cathode surface may recover a high temperature so that the cathode itself
in time for the next pulse of current. If the usually may be brought to operating tem-
cathode were to be subjected to a contin- perature in a matter of 15 to 30 seconds.
uous current drain of this magnitude, it Heat -coupling between the heater and the
www.americanradiohistory.com
HANDBOOK Thermionic Emission 79
OXIDE COATED
AMONT
10 20 a0 40 SO PLATE VOLTAGE
DC PLATE VOLTS
Figure 6
Figure S
MAXIMUM SPACE -CHARGE -LIMITED
AVERAGE PLATE CHARACTERISTICS
EMISSION FOR DIFFERENT
OF A POWER DIODE
TYPES OF EMITTERS
www.americanradiohistory.com
HANDBOOK Triode Characteristics 81
Plate- Current As plate voltage is raised to Electron Energy The current flowing in the
Saturation the potential where the Dissipation plate- cathode space of a con-
cathode space charge is neu- ducting electron tube repre-
tralized, all the electrons that the cathode is sents the energy required to accelerate elec-
capable of emitting are being attracted to trons from the zero potential of the cathode
the plate. The electron tube is said then to space charge to the potential of the anode.
have reached saturation plate current. Then, when these accelerated electrons strike
Further increase in plate voltage will cause the anode, the energy associated with their
only a relatively small increase in plate cur- velocity is immediately released to the anode
rent. The initial point of plate- current sat- structure. In normal electron tubes this
uration is sometimes called the point of energy release appears as heating of the plate
Maximum Space -Charge- Limited Emission or anode structure.
(MSCLE) .
4 -3 The Triode
If an element consisting of a mesh or
O spiral of wire is inserted concentric with the
plate and between the plate and the cathode,
such an element will be able to control by
electrostatic action the cathode-to -plate cur-
rent of the tube. The new element is called
a grid, and a vacuum tube containing a cath-
ode, grid, and plate is commonly called a
triode.
3
,s
12
.
I'.1
Vacuum -Tube Principles
...r...............
.... IIi.I
TYPE
Er
6J5
.S VOLTS
THE RADIO
_ . IIIIiIIGI
lIIIIlNI
IIIIIIII'I
IIlIIIIIIl I
ducing a field at the cathode.
It would then be expected that the cath-
ode current in a triode would be proportion-
_
/II/IIIIIIIIII
IIIIII,IIIIIII
al to the three -halves power of (Eg +
+
III,/IIIIII:
IIIIIIAII%I/I
I.%I,III.II./%I.II/%/
EP /). The cathode current of a triode can
be represented with fair accuracy by the
expression:
0
/DIG!!IMCM2IZII
,00 200 300 400 SOO cathode current - K +
PLATE VOLTS (Er) (E,
Figure 8 where,
NEGATIVE -GRID CHARACTERISTICS (IP K is a constant determined by element
VS. E,, CURVES) OF A TYPICAL geometry within the triode.
TRIODE
Average plate characteristics of this form Plate ResistanceThe plate resistance of a
are most commonly used in determining the vacuum tube is the ratio of
class -A operating characteristics of a triode
amplifier stage. a change in plate voltage to the change in
plate current which the change in plate
a small change in grid bias has a consider- voltage produces. To be accurate, the
ably greater effect than a similar change in changes should be very small with respect
plate voltage. The ratio between the change to the operating values. Expressed as an
in grid bias and the change in plate current equation:
which will cause the same small change in
plate current is called the amplification fac-
tor or tt of the electron tube. Expressed as
P - oEP
0tP
an equation: where,
AE, Eg is held constant,
DEg equals small increment.
with i, constant (0 represents a small incre- The plate resistance can also be determined
ment) .
by the experiment mentioned above. By
The can be determined experimentally noting the change in plate current as it oc-
by making a small change in grid bias, thus curs when the plate voltage is changed (grid
slightly changing the plate current. The voltage held constant), and by dividing the
plate current is then returned to the original latter by the former, the plate resistance
value by making a change in the plate volt- can be determined. Plate resistance is ex-
age. The ratio of the change in plate voltage pressed in ohms.
to the change in grid voltage is the p. of the
tube under the operating conditions chosen Transconductonce The mutual conductance,
for the test. also referred to as trans -
Current Flow In a diode it was shown that conductance, is the ratio of a change in the
in o Triode the electrostatic field at the plate current to the change in grid voltage
cathode was proportional to which brought about the plate- current
the plate potential (En) and that the total change, the plate voltage being held con-
cathode current was proportional to the stant. Expressed as an equation:
three- halves power of the plate voltage.
Similiarly, in a triode it can be shown that SIP
the field at the cathode space charge is pro- gm
- DEg
www.americanradiohistory.com
a
HANDBOOK
.,,
400
aso
300
aa0
ROO
'
t 00
JD
ao
F
J
O
f
530
300
2!o
200 \\`
-.,,
.".,.
-,.,.,
-50
Triode Load Line
-IS -0 -S
GRID VOLTS
0
(Ea)
45 IO 11S
83
410
Figure 10
?0 -4 -a 0 10 30 40 00 00 70 b0 00 100
RO
GRID VOLTAGE (Eg)
CONSTANT CURRENT (E7 vs. E0)
CHARACTERISTICS OF A
Figure 9 TYPICAL TRIODE TUBE
POSITIVE -GRID CHARACTERISTICS This type of graphical representation is used
for class -C amplifier calculations since the
(Ip vs. E6) OF A TYPICAL TRIODE operating characteristic of a class -C amplifier
is a straight line when drawn on a constant
Plate characteristics of this type are most current graph.
commonly used in determining the pulse -
signal operating characteristics of a triode
amplifier stage. Note the large emission cap- from the Ip vs. Eg curve, and the amplifica-
ability of the oxide -coated heater cathode in
tubes of the g type of the 6J5.
1
tion factor (to) may be determined from the
Ep vs. E curve.
where, The Load Line A load line is a graphical
E held constant,
is representation of the voltage
A equals small increment. on the plate of a vacuum tube and the cur-
rent passing through the plate circuit of the
The transconductance is also numerically tube for various values of plate load resist-
equal to the amplification factor divided by ance and plate supply voltage. Figure 11
the plate resistance. gm = p. /Rp. illustrates a triode tube with a resistive plate
Transconductance is most commonly ex- load, and a supply voltage of 300 volts. The
pressed in microreciprocal -ohms or micro -
voltage at the plate of the tube (er) may be
mhos. However, since transconductance ex-
expressed as:
presses change in plate current as a function
of a change in grid voltage, a tube is often
said to have a transconductance of so many eP=E - ( IpXRL )
milliamperes per volt. If the transconduct-
ance in milliamperes per volt is multiplied where,
by 1000 it will then be expressed in mi- the plate supply voltage,
Ep is
cromhos. Thus the transconductance of a the plate current,
ip is
6A3 could be called either 5.25 ma /volt or RL is the load resistance in ohms.
5250 micromhos. Assuming various values of ip flowing in
the circuit, controlled by the internal resist-
Characteristic Curves The operating charac- ance of the tube, (a function of the grid
of a Triode Tube teristics of a triode bias) values of plate voltage may be plotted
tube may be summa- as shown for each value of plate current
rized in three sets of curves: The Ip vs. Ep (i7). The line connecting these points is
curve (figure 8) , the Ip vs. ER curve (figure called the load line for the particular value
9) and the Ep vs. Eg curve (figure 10). of plate load resistance used. The slope of
The plate resistance (Rp) of the tube may the load line is equal to the ratio of the
be observed from the Ip vs. Ep curve, the lengths of the vertical and horizontal pro-
transconductance (gm) may be observed jections of any segment of the load line.
84 Vacuum -Tube Principles THE RADIO
Er
o 300
s 250
10 200 RL=sK
Is 150
20 100
25 So
30 0
3
Figure 12
DA l
STATIC LOAD LINE
25 IOR RL= 10000L TRIODE TUBE CONNECTED FOR DETER-
IPZ
0 MINATION OF PLATE- CIRCUIT LOAD
LINE AND OPERATING PARAMETERS
OF THE CIRCU :T
5 I
I
loo
Er
200 300 nal voltage drop, or if the tube is short -cir-
cuited from cathode to plate. Point B on the
Figure 11 load line corresponds to the cutoff point of
The static load line for a typical triode tube
the tube, where no plate current is flowing.
with a plate load resistance of 10,000 ohms. The operating range of the tube lies between
these two extremes. For additional informa-
For this example it is: tion regarding dynamic load lines, the reader
is referred to the Radiotrnz Designer's
= .01 - .02
-( 100 Handbook distributed by Radio Corporation
slope
- 200 of America.
1
= -.0001 = Application of Tube As an example of the
10,000
Characteristics application of tube char-
The slope of the load line is equal to acteristics, the constants
-1 /RL. At point A on the load line, the of the triode amplifier circuit shown in fig-
voltage across the tube is zero. This would ure 12 may be considered. The plate supply
be true for a perfect tube with zero inter- is 300 volts, and the plate load is 8000 ohms.
40
37.5
35
o'
ao
OAD LINE 1000A
6
AJ
25 INSTANTA/LOUS
GID SWING
Figure 13
2
IPNAK..11.25
20
-- APPLICATION OF Ir vs. E
CHARACTERISTICS OF
W IS
-
A VACUUM TUBE
J
::'A%iI
0.
10 11111/211,
31LIA.
IulN.y 7.5
s
o 112
100
3 zoo 3 300 A00 S00
www.americanradiohistory.com
HANDBOOK Triode Load Line 85
T -
Figure 15
SCHEMATIC REPRESENTATION
STEADY-STATE
PLATE CURRENT(t )
OF INTERELECTRODE
CAPACITANCE
www.americanradiohistory.com
86 Vacuum -Tube Principles THE RADIO
10
TYPE 24 -A
esc so v.
eG
i eG -)
0
O 100 200 300 400 300
VOLTS (Er)
Figure 16 Figure 17
I
TYPICAL vs. E TETRODE I
TYPICAL vs. E,, PENTODE
CHARACTERISTIC CURVES CHARACTERISTIC CURVES
www.americanradiohistory.com
HANDBOOK Tetrodes and Pentodes 87
and on to the plate. Due also to the screen, ner that the suppressor allows high voltage
the plate current is largely independent of gain, at the same time permitting fairly
plate voltage, thus making for high ampli- high gain at low plate voltage. This holds
fication. When the screen voltage is held at true even if the plate voltage is the same or
a constant value, it is possible to make large slightly lower than the screen voltage.
changes in plate voltage without appreciably
Remote -Cutoff Remote -cutoff tubes (vari-
affecting the plate current, (figure 16).
When the electrons from the cathode ap-
Tubes able -) are screen grid
tubes in which the control
proach the plate with sufficient velocity,
they dislodge electrons on striking the plate. grid structure has been physically modified
so as to cause the plate current of the tube
This effect of bombarding the plate with
high- velocity electrons, with the consequent to drop off gradually, rather than to have a
well-defined cutoff point (figure 18) . A non-
dislodgement of other electrons from the
plate, gives rise to the condition of second- uniform control -grid structure is used, so
ary emission which has been discussed in a that the amplification factor is different for
previous paragraph. This effect can cause no different parts of the control grid.
particular difficulty in a triode because the Remote -cutoff tubes are used in circuits
secondary electrons so emitted are eventually where it is desired to control the amplifica-
attracted back to the plate. In the screen - tion by varying the control -grid bias. The
grid tube, however, the screen is close to the characteristic curve of an ordinary screen -
plate and is maintained at a positive poten- grid tube has considerable curvature near
tial. Thus, the screen will attract these elec- the plate- current cutoff point, while the
trons which have been knocked from the curve of a remote -cutoff tube is much more
plate, particularly when the plate voltage linear (figure 19). The remote -cutoff tube
falls to a lower value than the screen volt- minimizes cross -talk interference that would
age, with the result that the plate current is otherwise be produced. Examples of remote
lowered and the amplification is decreased. cutoff tubes are: 6BD6, 6BA6, 6SG7 and
In the application of tetrodes, it is neces- 6SK7.
sary to operate the plate at a high voltage in Beam -Power A beam -power tube makes use
relation to the screen in order to overcome Tubes of another method of suppres-
these effects of secondary emission. sing secondary emission. In this
The Pentode The undesirable effects of sec- tube there are four electrodes: a cathode, a
ondary emission from the plate grid, a screen, and a plate, so spaced and
can be greatly reduced if yet another ele- placed that secondary emission from the
ment is added between the screen and plate. plate is suppressed without actual power loss.
This additional element is called a suppres- Because of the manner in which the elec-
sor, and tubes in which it is used are called trodes are spaced, the electrons which travel
pentodes. The suppressor grid is sometimes to the plate are slowed down when the plate
connected to the cathode within the tube; voltage is low, almost to zero velocity in a
sometimes it is brought out to a connecting certain region between screen and plate. For
pin on the tube base, but in any case it is this reason the electrons form a stationary
established negative with respect to the cloud, or space charge. The effect of this
minimum plate voltage. The secondary elec- space charge is to repel secondary electrons
trons that would travel to the screen if emitted from the plate and thus cause them
there were no suppressor are diverted back to return to the plate. In this way, secondary
to the plate. The plate current is, therefore, emission is suppressed.
not reduced and the amplification possibili- Another feature of the beam-power tube
ties are increased (figure 17). is the low current drawn by the screen. The
Pentodes for audio applications are de- screen and the grid are spiral wires wound
signed so that the suppressor increases the so that each turn in the screen is shaded
limits to which the plate voltage may swing; from the cathode by a grid turn. This align-
therefore the consequent power output and ment of the screen and the grid causes the
gain can be very great. Pentodes for radio - electrons to travel in sheets between the
frequency service function in such a man- turns of the screen so that very few of them
88 Vacuum -Tube Principles THE RADIO
GRID
strike the screen itself. This formation of or pentode tube. The relationship between
the electron stream into sheets or beams in- control -grid potential and screen potential
creases the charge density in the screen -plate determines the plate current of the tube as
region and assists in the creation of the space well as the screen current since the plate
charge in this region. current is essentially independent of the
Because of the effective suppressor action plate voltage in tubes of this type. In other
provided by the space charge, and because of words, when the tube is operated at cutoff
the low current drawn by the screen, the bias as determined by the screen voltage and
beam -power tube has the advantages of high the grid- screen p. factor (determined in
power output, high power sensitivity, and the same way as with a triode, by dividing
high efficiency. The 6AQ5 is such a beam - the operating voltage by the factor) the
power tube, designed for use in the power - plate current will be substantially at cutoff,
amplifier stages of receivers and speech am- as will be the screen current. The grid- screen
plifiers or modulators. Larger tubes employ- p. factor is numerically equal to the am-
ing the beam -power principle are being made plification factor of the same tetrode or
by various manufacturers for use in the pentode tube when it is triode connected.
radio- frequency stages of transmitters. These
tubes feature extremely high power sensitiv- Current Flow The following equation is the
ity (a very small amount of driving power in Tetrodes expression for total cathode
is required for a large output), good plate and Pentodes current in a triode tube. The
efficiency, and low grid -to -plate capacitance. expression for the total cathode
Examples of these tubes are 813, 4-250A, current of a tetrode and a pentode tube is
4CX2 f 0B, etc. the same, except that the screen -grid voltage
Grid -Screen The grid- screen p. factor (,,.) and the grid- screen factor are used in
Mu Factor is analogous to the amplification place of the plate voltage and of the
factor in a triode, except that triode.
the screen of a pentode or tetrode is sub- //
stituted for the plate of a triode. p,, denotes cathode current = K (Ec + E_c)312
the ratio of a change in grid voltage to 'a sR
change in screen voltage, each of which will Cathode current, of course, is the sum of
produce the same change in screen current. the screen and plate currents plus control -
Expressed as an equation: grid current in the event that the control
grid is positive with respect to the cathode.
- AE,
DEAR
It will be noted that total csthode current
is independent of plate voltage in a tetrode
where,
L, is held constant, or pentode. Also, in the usual tetrode or
.1 equals small increment.
pentode the plate current is substantially
independent of plate voltage over the usual
The grid -screen factor is important in operating range -which means simply that
determining the operating bias of a tetrode the effective plate resistance of such tubes
www.americanradiohistory.com
HANDBOOK Mixer and Converter Tubes 89
www.americanradiohistory.com
HANDBOOK The Klystron 91
Tubes especially designed for vhf and velocity will be increased. However, due to
uhf use have had their lead inductances the law of electron motion in an electric
minimized. The usual procedures for reduc- field, transit time is increased only as the
ing lead inductance are: (1) using heavy square root of the ratio of operating poten-
lead conductors or several leads in parallel tial increase; therefore this expedient is of
(examples are the 6J4 and 6AK5), (2) limited value due to other limitations on
scaling down the tube in all dimensions to operating voltages of small electron tubes.
reduce both lead inductances and interelec-
trode capacitances (examples are the 6CW4,
6F4, and other nuvistor and miniature
4 -7 Special Microwave
tubes), and (3) the use of very low -in- Electron Tubes
ductance extensions of the elements them-
selves as external connections (examples are Due primarily to the limitation imposed
lighthouse tubes such as the 2C40, planar by transit time, conventional negative-grid
tubes such as the 2C29, and many types of electron tubes are capable of affording
vhf transmitting tubes) .
worthwhile amplification and power output
only up to a definite upper frequency. This
Effect of When an electron tube is op- upper frequency limit varies from perhaps
Transit Time erated at a frequency high
100 MHz for conventional tube types to
enough that electron transit about 4000 MHz for specialized types such
time between cathode and plate is an ap- as the lighthouse tube. Above the limiting
preciable fraction of a cycle at the input frequency, the conventional negative -grid
frequency, several undesirable effects take tube no longer is practicable and recourse
place. First, the grid takes power from the must be taken to totally different types of
input signal even though the grid is negative electron tubes in which electron transit
at all times. This comes about since the grid time is not a limitation to operation. Three
will have changed its potential during the of the most important of such microwave
time required for an electron to pass from tube types are the klystron, the magnetron,
cathode to plate. Due to interaction, and a and the traveling -wave tube.
resulting phase difference between the field
associated with the grid and that associated The Power Klystron The klystron is a type
with a moving electron, the grid presents a of electron tube in which
resistance to an input signal in addition to electron transit time is used to advantage.
its normal "cold" capacitance. Further, as Such tubes comprise, as shown in figure 22,
a result of this action, plate current no a cathode, a focusing electrode, a resonator
longer is in phase with grid voltage. connected to a pair of grids which afford
An amplifier stage operating at a fre- velocity modulation of the electron beam
quency high enough that transit time is (called the "buncher "), a drift space, and
appreciable: another resonator connected to a pair of
(a) Is difficult to excite as a result of grids (called the "catcher "). A collector
grid loss from the equivalent input grid re- for the expended electrons may be included
sistance, at the end of the tube, or the catcher may
(b) Is capable of less output since trans - also perform the function of electron col-
conductance is reduced and plate current is lection.
not in phase with grid voltage. The tube operates in the following man-
The effects of transit time increase with ner: The cathode emits a stream of electrons
the square of the operating frequency, and which is focused into a beam by the focus-
they increase rapidly as frequency is in- ing electrode. The stream passes through the
creased above the value where they become buncher where it is acted upon by any field
just appreciable. These effects may be re- existing between the two grids of the
duced by scaling down tube dimensions; a buncher cavity. When the potential between
procedure which also reduces lead induct- the two grids is zero, the stream passes
ance. Further, transit -time effects may be through without change in velocity. But
reduced by the obvious procedure of in- when the potential between the two grids of
creasing electrode potentials so that electron the buncher is increasingly positive in the
92 Vacuum -Tube Principles THE RADIO
Figure 22
" Figure 23
REFLEX KLYSTRON OSCILLATOR
TWO- CAVITY KLYSTRON OSCILLATOR
A conventional reflex klystron oscillator of
A e tional two-cavity klystron is shown the type commonly used as a local oscillator
with a feedback loop connected between the in superheterodyne receivers operating above
two cavities so that the tube may be used as about 2000 MHz is shown above. Frequency
an oscillator. modulation of the output frequency of the
oscillator, or afc operation in a receiver, may
be obtained by varying the negative voltage
direction of electron motion, the velocity of on the repeller electrode.
the electrons in the beam is increased. Con-
versely, when the field becomes increasingly voltage is rich in harmonics. Klystron oscil-
negative in the direction of the beam (cor- lators of various types afford power outputs
responding to the other half -cycle of the ranging from less than 1 watt to many
exciting voltage from that which produced thousand watts. Operating efficiency varies
electron acceleration) the velocity of the between I and 50 percent. Frequency may
electrons in the beam is decreased.
be shifted to some extent by varying the
When the velocity -modulated electron beam voltage. Tuning is carried on mechan-
beam reaches the drift space where there is
ically in some klystrons by altering (by
no field, those electrons which have been
means of knob settings) the shape of the
sped up on one half -cycle overtake those
resonant cavity.
immediately ahead which were slowed down
on the other half -cycle. In this way, the The Reflex Klystron The multicavity kly-
beam electrons become bunched together. stron as described in the
As the bunched groups pass through the two preceding paragraphs is primarily used as a
grids of the catcher cavity, they impart transmitting device since quite reasonable
pulses of energy to these grids. The catcher - amounts of power are made available in its
grid space is charged to different voltage output circuit. However, for applications
--
levels by the passing electron bunches, and where a much smaller amount of power is
a corresponding oscillating field is set up in required power levels in the milliwatt
the catcher cavity. The catcher is designed range for low -power transmitters, re-
to resonate at the frequency of the velocity - ceiver local oscillators, etc., another type of
modulated beam, or at a harmonic of this klystron having only a single cavity is
frequency. more frequently used.
In the klystron amplifier, energy delivered The theory of operation of the single -
by the buncher to the catcher grids is greater cavity klystron is essential the same as the
than that applied to the buncher cavity by multicavity type with the exception that the
the input signal. In the klystron oscillator a velocity- modulated electron beam, after hav-
feedback loop connects the two cavities. ing left the buncher cavity is reflected
Coupling to either buncher or catcher is back into the area of the buncher again by
provided by small loops which enter the a repeller electrode as illustrated in figure
cavities by way of concentric lines. 23. The potentials on the various electrodes
The klystron is an electron- coupled de- are adjusted to a value sch that proper
vice. 'When used as an oscillator, its output bunching of the electron beam will take
www.americanradiohistory.com
HANDBOOK The Magnetron 93
1 MAGNET COIL
GMD ANODE ANODE
TERMINAI TERMINAL GLASS
ANODE
SEAL
CATHODE ANODE
GRID NEATER ANODE
FIL
FILAMENT
MAGNET
BATTERY
!i lF
EYELET
GLASS
ErrEEE
40 l FILAMENT PLATE
GETTER
SFAI VOLTAGE VOLTAGE
LEAD
TERMINAI ETELEI TUEULA LION
Figure 25
www.americanradiohistory.com
94 Vacuum -Tube Principles THE RADIO
WAVE GUIDE
INPUT
Figure 27
Figure 26 THE TRAVELING -WAVE TUBE
MODERN MULTICAVITY MAGNETRON Operation of this tube is the result of inter-
action between the electron beam and wave
Illustrated is an external -anode strapped traveling along the helix.
magnetron of the type commonly used in
radar equipment for the 10-cm. range. An
integral permanent magnet is shown in the lower. The electrons emitted by the cathode
right hand portion of the drawing, with the
magnetron in place between the pole pieces of gun pass axially through the helix to the
the magnet. collector, located at the output end of the
helix. The average velocity of the electrons
the electron rotation, no external tuned cir- depends on the potential of the collector
cuits being employed. Wavelengths of less with respect to the cathode. When the
than 1 centimeter have been produced with average velocity of the electrons is greater
such circuits. than the velocity of the helix wave, the ele-
More complex magnetron tubes employ no trons become crowded together in the var-
external tuned circuit, but utilize instead ious regions of retarded field, where they im-
one or more resonant cavities which are in- part energy to the helix wave. A power gain
tegral with the anode structure. Figure 26 of 100 or more may be produced by this
shows a magnetron of this type having a tube.
multicellular anode of eight cavities. It will
be noted, also, that alternate cavities (which 4 -8 The Cathode -Ray Tube
would operate at the same polarity when the The Cathode -Roy Tube The cathode -ray tube
tube is oscillating) are strapped together. type of
is a special
Strapping was found to improve the effi- electron tube which permits the visual obser-
ciency and stability of high -power radar vation of electrical signals. It may be in-
magnetrons. In most radar applications of corporated into an oscilloscope for use as a
magnetron oscillators, a powerful permanent test instrument or it may be the display de-
magnet of controlled characteristics is em- vice for radar equipment or television.
ployed to supply the magnetic field, rather
than the use of an electromagnet. Operation of A cathode -ray tube always in-
the CRT eludes an electron gun for pro-
The Traveling- The Traveling -Wave Tube ducing a stream of electrons, a
Wave Tube (figure 27) consists of a grid for controlling the intensity of the elec-
helix located within an evac- tron beam, and a luminescent screen for con-
uated envelope. Input and output termina- verting the impinging electron beam into
tions are affixed to each end of the helix. An visible light. Such a tube always operates in
electron beam passes through the helix and conjunction with either a built -in or an ex-
interacts with a wave traveling along the ternal means for focusing the electron stream
helix to produce broadband amplification at into a narrow beam, and a means for deflect-
microwave frequencies. ing the electron beam in accordance with an
When the input signal is applied to the electrical signal.
gun end of the helix, it travels along the The main electrical difference between
helix wire at approximately the speed of types of cathode -ray tubes lies in the means
light. However, the signal velocity measured employed for focusing and deflecting the
along the axis of the helix is considerably electron beam. The beam may be focused
www.americanradiohistory.com
HAN DBOOK The Cathode -Ray Tube 95
www.americanradiohistory.com
96 Vacuum -Tube Principles THE RADIO
plates. This is done for two reasons: First,
FOCUS COIL TERMINAL
RASE FIRST DEFLECTION COILS
the amount of deflection voltage required
ANODE
by the high -voltage tubes is so great that
IECOND ANDO(
(ADUADAG)
a transmitting tube operating from a
high -voltage supply would be required to
attain this voltage without distortion. By
GRID 41
I FLUORESCENT SCREEN
using push-pull deflection with two tubes
CATNODE (R) feeding the deflection plates, the necessary
plate - supply voltage for the deflection am-
Figure 29 plifier is halved. Second, a certain amount of
defocusing of the electron stream is always
TYPICAL ELECTROMAGNETIC
present on the extreme excursions in deflec-
CATHODE -RAY TUBE tion voltage when this voltage is applied
only to one deflecting plate. When the de-
Standard oscilloscope practice with small flecting voltage is fed in push -pull to both
cathode -ray tubes calls for connecting one deflecting plates in each plane, there is no
of the B plates and one of the C plates to- defocusing because the average voltage act-
gether and to the high -voltage accelerating ing on the electron stream is zero, even
anode. With the newer three -inch tubes and though the net voltage (which causes the
with five -inch tubes and larger, all four deflection) acting on the stream is twice
deflection plates are commonly used for that on either plate.
deflection. The positive high voltage is The fact that the beam is deflected by a
grounded, instead of the negative as is com- magnetic field is important even in an oscil-
mon practice in amplifiers, etc., in order to loscope which employs a tube using electro-
permit operation of the deflecting plates at static deflection, because it means that pre-
a d -c potential at or near ground. cautions must be taken to protect the tube
An Aquadag coating is applied to the in- from the transformer fields and sometimes
side of the envelope to attract any secondary even the earth's magnetic field. This nor-
electrons emitted by the fluorescent screen. mally is done by incorporating a magnetic
In the average electrostatic- deflection CR shield around the tube and by placing any
tube the spot will be fairly well centered if transformers as far from the tube as pos-
all four deflection plates are returned to the sible, oriented to the position which pro-
potential of the second anode (ground). duces minimum effect on the electron stream.
However, for accurate centering and to
permit moving the entire trace either hori- Construction of Electro- The electromagnetic
zontally or vertically to permit display of a magnetic CRT cathode -ray tube al-
particular waveform, horizontal- and verti- lows greater defini-
cal- centering controls usually are provided tion than does the electrostatic tube. Also,
on the front of the oscilloscope. electromagnetic definition has a number of
After the spot is once centered, it is nec- advantages when a rotating radial sweep is
essary only to apply a positive or negative required to give polar indications.
voltage (with respect to ground) to one of The production of the electron beam in
the ungrounded or "free" deflector plates an electromagnetic tube is essentially the
in order to move the spot. If the voltage is same as in the electrostatic tube. The grid
positive with respect to ground, the beam structure is similar, and controls the electron
will be attracted toward that deflector plate, beam in an identical manner. The elements
If it is negative, the beam and spot will of a typical electromagnetic tube 'are shown
be repulsed. The amount of deflection is in figure 29. The focus coil is wound on an
directly proportional to the voltage (with iron core which may be moved along the
respect to ground) that is applied to the neck of the tube to focus the electron beam.
free electrode. For final adjustment, the current flowing in
With the larger- screen higher -voltage tubes the coil may be varied. A second pair of
it becomes necessary to place deflecting volt- coils, the deflection coils, are mounted at
age on both horizontal and both vertical right angles to each other around the neck
www.americanradiohistory.com
HANDBOOK Gas Tubes 97
ments. If the temperature is too high, the of these three groups may also be further
vapor pressure is too high, and the voltage classified for military duty (MIL spec. or
at which the tube will "flash back" is low- JAN classification). To qualify for MIL
ered to the point where destruction of the classification, sample lots of the particular
tube may take place. Since the ambient tem- tube must have passed special qualification
perature range specified above is within the tests at the factory. It should not be con-
normal room temperature range, no trouble strued that a MIL -type tube is better than
will be encountered under normal operating a commercial tube, since some commercial
conditions. However, by the substitution of tests and specifications are more rigid than
xenon gas for mercury it is possible to pro- the corresponding MIL specifications. The
duce a rectifier with characteristics compar- MIL stamped tube has merely been accepted
able to those of the mercury-vapor tube ex- under a certain set of conditions for mili-
cept that the tube is capable of operating tary service.
over the range from approximately -70 Ruggedized or Radio tubes are being used
to + 90 C. The 3B25 rectifier is an ex- Premium Tubes in increasing numbers for
ample of this type of tube. industrial applications, such
Thyratron If as computing and control machinery, and
a grid is inserted between the
Tubes cathode and plate of a mercury- in aviation and marine equipment. When
a tube fails in a home radio receiver, it is
vapor gaseous- conduction recti-
fier, a negative potential placed on the added merely inconvenient, but a tube failure in
element will increase the plate -to- cathode industrial applications may bring about stop-
voltage drop required before the tube will page of some vital process, resulting in fi-
ionize or "fire." The potential on the control nancial loss, or even danger to life.
grid will have no effect on the plate -to- To meet the demands of these industrial
cathode drop after the tube has ionized. applications, a series of tubes was evolved
However, the grid voltage may be adjusted incorporating many special features designed
to such a value that conduction will take to ensure a long and predetermined operat-
place only over the desired portion of the ing life, and uniform characteristics among
cycle of the a -c voltage being impressed on similar tubes. Such tubes are known as rug -
the plate of the rectifier. gedized or premium tubes. Early attempts to
select reliable specimens of tubes from ordi-
Voltage -Regulator In a glow- discharge gas nary stock tubes proved that in the long run
Tubes tube the voltage drop the selected tubes were no better than tubes
across the electrodes re- picked at random. Long life and ruggedness
mains constant over a wide range of current had to be built into the tubes by means of
www.americanradiohistory.com
HANDBOOK Miscellaneous Tube Types 99
100
Figure 31 L 40
SCHEMATIC REPRESENTATION
20
OF "MAGIC EYE" TUBE
o
o 10 20 30 40 SO 60
proper choice and 100% inspection of all EP VOLTS)
materials used in the tube, by critical proc-
Figure 32
essing inspection and assembling, and by
conservative ratings of the tube. AMPLIFICATION FACTOR OF TYPICAL
Pure tungsten wire is used for heaters in STANDARD TUBE DROPS RAPIDLY AS
preference to alloys of lower tensile strength. PLATE VOLTAGE IS DECREASED
Nickel tubing is employed around the heater BELOW 20 VOLTS
wires at the junction to the stem wires to
reduce breakage at this point. Element struc-
tures are given extra supports and bracing. or give off light, so that the deflection caused
Finally, all tubes are given a 50 -hour test by the ray -control electrode, which prevents
run under full operating conditions to elimi- electrons from striking part of the anode,
nate early failures. When operated within produces a wedge- shaped electrical shadow
their ratings, ruggedized or premium tubes on the fluorescent anode. The size of this
should provide a life well in excess of 10,000 shadow is determined by the voltage on
hours. the ray electrode. When this electrode is at
Ruggedized tubes will withstand severe the same potential as the fluorescent anode,
impact shocks for short periods, and wil! the shadow disappears; if the ray electrode
is less positive than the anode, a shadow ap-
operate under conditions of vibration for
many hours. The tubes may be identified in pears the width of which is proportional to
many cases by the fact that their nomen- the voltage on the ray electrode.
clature includes a "W" in the type number, Controlled Series heater strings are employed
as in 807W, 5U4W, etc. Some ruggedized Warmup in a -c /d -c radio receivers and
tubes are included in the "5000" series Tubes television sets to reduce the cost,
nomenclature. The 5654 is a ruggedized size, and weight of the equip-
version of the 6AK5, the 5692 is a rugged - ment. Voltage surges of great magnitude
ized version of the 6SN7, etc.
occur in series -operated filaments because
4 -10 Miscellaneous of variations in the rate of warm -up of the
various tubes. As the tubes warm up, the
Tube Types heater resistance changes. This change is not
Electron- The electron -ray tube or magic the same between tubes of various types, or
Ray Tubes eye contains two sets of ele- even between tubes of the same type made
ments, one of which is a triode by different manufacturers. Some 6 -volt
amplifier and the other a cathode -ray indi- tubes show an initial surge as high as 9 volts
cator. The plate of the triode section is during warm -up, while slow- heating tubes
internally connected to the ray -control elec- such as the 25BQ6 are underheated during
trode (figure 31) , so that as the plate volt- the voltage surge on the 6 -volt tubes.
ge varies in accordance with the applied Standardization of heater characteristics
signal the voltage on the ray -control elec- in a new group of tubes designed for series
trode also varies. The ray- control electrode heater strings has eliminated this trouble.
is a metal cylinder so placed relative to the The new tubes have either 600 ma or 400
cathode that it deflects some of the electrons ma heaters, with a controlled warm -up time
emitted from the cathode. The electrons of approximately 11 seconds. The 5U8,
which strike the anode cause it to fluoresce, 6CG7, and 12BH7 -A are examples.
www.americanradiohistory.com
CHAPTER FIVE
Semiconductor Devices
One of the earliest detection devices used formed on a pure copper disk. This film offers
in radio was the galena crystal, a crude ex- low resistance for positive voltages, and high
ample of a semiconductor. More modern ex- resistance for negative voltages. The same
amples of semiconductors are the copper - action is observed in selenium rectifiers,
oxide rectifier, the selenium and silicon recti- where a film of selenium is deposited on an
fiers, and the germanium diode. All of these iron surface.
devices offer the interesting property of
greater resistance to the flow of electrical 5 -1 Atomic Structure of
current in one direction than in the opposite Germanium and Silicon
direction. Typical conduction curves for
these semiconductors are shown in figure 1. It has been previously stated that the
The copper -oxide rectifier action results from electrons in an element having a large atom-
the function of a thin film of cuprous oxide ic number are grouped into rings, each ring
having a definite number of electrons. Atoms
in which these rings are completely filled
are called Merl gases, of which helium and
30
argon are examples. All other elements have
1N34 one or more incomplete rings of electrons.
40
CIIYiT4L MOOT
If the incomplete ring is loosely bound, the
electrons may be easily removed, the element
TYPICAL STATIC CHARACTERISTICS
is called metallic, and is a conductor of
.o
electric current. If the incomplete ring is
.0
tightly bound, with only a few missing elec-
trons, the element is called nonmetallic and
is an insulator to electric current. Ger-
lo
manium and silicon fall between these two
sharply defined groups, and exhibit both
o
0.1- metallic and nonmetallic characteristics.
es
es
o.
Pure germanium or silicon may be consid-
-so -30 -TO -0 s s ered to be a good insulator. The addition of
VOLTS certain impurities in carefully controlled
Figure 1
amounts to the pure germanium will alter
the conductivity of the material. In addi-
TYPICAL CHARACTERISTIC CURVE tion, the choice of the impurity can change
OF A SEMICONDUCTOR DIODE the direction of conductivity through the
www.americanradiohistory.com
Mechanism of Conduction 101
SCHEMATIC REPRESENTATION
field with a velocity which is proportional
to the field itself:
Vdh = hE
where,
Van equals drift velocity of hole,
E equals magnitude of electric field,
h equals mobility of hole
In an electric field the holes will drift in a
direction opposite to that of the electron
and with about one -half the velocity, since
the hole mobility is about one-half the elec-
ANODES CATHODES tron mobility. A sample of a semiconductor,
such as germanium or silicon, which is both
chemically pure and mechanically perfect
will contain in it approximately equal num-
bers of holes and electrons and is called an
intrinsic semiconductor. The intrinsic resis-
tivity of the semiconductor depends strongly
e- Color Bua"." on the temperature, being about 50 ohm /cm
for germanium at room temperature. The
intrinsic resistivity of silicon is about 65,000
ohm /cm at the same temperature.
If, in the growing of the semiconductor
crystal, a small amount of an impurity, such
as phosphorous, arsenic, or antimony is in-
TUBE. GEPMANIUM. SILICON cluded in the crystal, each atom of the im-
AND SELENIUM DIODES
purity contributes one free electron. This
Figure 2 electron is available for conduction. The
COMMON DIODE COLOR CODES crystal is said to be doped and has become
AND MARKINGS electron -conducting in nature and is called
N (negative) -type germanium. The impuri-
ties which contribute electrons are called
crystal, some impurities increasing conduc- donors. N -type germanium has better con-
tivity to positive voltages, and others in- ductivity than pure germanium in one di-
creasing conductivity to negative voltages. rection, and a continuous stream of electrons
will flow through the crystal in this direc-
5 -2 Mechanism of tion as long as an external potential of the
Conduction correct polarity is applied across the crystal.
Other impurities, such as aluminum, gal -
As indicated by their name, semiconduc- lium,or indium add one hole to the semi-
tors are substances which have a conduc- conducting crystal by accepting one electron
tivity intermediate between the high values for each atom of impurity, thus creating ad-
observed for metals and the low values ob- ditional holes in the semiconducting crystal.
served for insulating materials. The mech- The material is now said to be hole- conduct-
anism of conduction in semiconductors is ing, or P (positive) -type germanium. The
different from that observed in metallic impurities which create holes are called ac-
conductors. There exist in semiconductors ceptors. P -type germanium has better con-
both negatively charged electrons and posi- ductivity than pure germanium in one di-
tively charged particles, called holes, which rection. This direction is opposite to that of
behave as though they had a positive electri- the N -type material. Either the N -type or
cal charge equal in magnitude to the nega- the P-type germanium is called extrinsic
tive electrical charge on the electron. These conducting type. The doped materials have
holes and electrons drift in an electrical lower resistivities than the pure materials,
www.americanradiohistory.com
102 Semiconductor Devices THE RADIO
and doped semiconductors in the resistivity 10
www.americanradiohistory.com
HANDBOOK The PN Junction Diode 103
MA-4060A
m.10
288 MHz
IDLER LOOP
Figure 6
I IS
- -j
,
FORWARD
CHARACTERISTIC 10
illlCy
REVERSE VOLTAGE
20 10
/20 20
I 2ENER KNEE O S 1 1 S
If AT S MA. VI (VOLTS)
OSI-
CONSTANT
VOLTAGE 2
PLATEAU
www.americanradiohistory.com
HANDBOOK Transistors 105
IOV
DIODE materials, including the compounds indium,
f antimony, and lead sulfide have been used
2 VOLT
UNREG.
VOLT
REG.
VOLT. experimentally in the production of transis-
tors.
Types of Transistors There are two basic
FIGURE 9 types of transistors, the
A -ZENER DIODE FUNCTIONS AS
VOLTAGE REGULATOR OVER
point- contact type and the junction type
RANGE OF CONSTANT VOLTAGE
PLATEAU.
(figure 10). Typical construction detail of a
5 -TWO ZENER DIODES OF DIFFER - point- contact transistor is shown in figure
'N ti y OUT ENT VOLTAGE CAN PROVIDE
SMALL REGULATED VOLTAGE.
11, and the electrical symbol is shown in fig-
C- OPPOSED ZENER DIODES CLIP ure 12. The emitter and collector electrodes
BOTH HALVES OF CYCLE OF A -C
WAVE. make contact with a small block of germani-
um, called the base. The base may be either
Figure 9 N -type or P-type germanium, and is ap-
proximately .05" long and .03" thick. The
ZENER DIODES USED AS
emitter and collector electrodes are fine
VOLTAGE REGULATORS
wires, and are spaced about .005" apart on
AND CLIPPERS
the germanium base. The complete assembly
is usually encapsulated in a small, plastic
used to protect meter movements since they case to provide ruggedness and to avoid
provide a very- low -resistance shunt across contaminating effects of the atmosphere.
the movement when the applied voltage ex- The polarity of emitter and collector volt-
ceeds a certain critical level. ages depends on the type of germanium em-
ployed in the base, as illustrated in figure 12.
5 -4 The Transistor The junction transistor consists of a piece
of either N -type or P -type germanium be-
In the past few years an entire new tech- tween two wafers of germanium of the op-
nology has been developed for the applica- posite type. Either NPN or PNP transistors
tion of certain semiconducting materials in may be made. In one construction called the
production of devices having gain properties. grown -crystal process, the original crystal,
These gain properties were previously found grown from molten germanium or silicon,
only in vacuum tubes. The elements ger- is created in such a way as to have the two
manium and silicon are the principal ma- closely spaced junctions imbedded in it. In
terials which exhibit the proper semicon- the other construction called the fusion
ducting properties permitting their applica- process, the crystals are grown so as to make
tion in the new amplifying devices called them a single- conductivity type. The junc-
transistors. However, other semiconducting tions are then produced by fusing small
PLASTIC CASE
N- TYPE
GERMANIUM
CRYSTAL LAYER
SASE CONNECTION
SMALL 3-PIN
RASE
P - TYPE GERMANIUM
CRYSTAL LAYER
COLLECTOR
LOW Z
-
PE - Nb JUNCTION
Pe
le
e
N b-Pc JUNCTION
PC-
NIGH Z
ASE CONNECTION 1- t
COLLECTOR
Figure 10
www.americanradiohistory.com
106 Semiconductor Devices THE RADIO
EMITTER COLLECTOR
EMITTER COLLECTOR EMITTER COLLECTOR
MEE
N TYPE
OERUANIYU
us--.I
Figure 12
wlrra - SASE
COLLECTOR
ELECTRICAL SYMBOLS
FOR TRANSISTORS
Figure 11
www.americanradiohistory.com
HANDBOOK Transistors 107
Values of beta run to 100 or so in inex- The Transition Region A useful rule common
pensive junction transistors. The static d -c to both PNP and NPN
forward current gain of a transistor in the transistors is: moving the base potential to-
common -emitter mode is termed the d -c wards the collector voltage point turns the
beta and may be designated F or hFE. transistor on, while moving the base poten-
tial away from the collector voltage point
Cutoff Frequencies The alpha cutoff frequen- turns the transistor off. When fully on, the
cy (fun) of a transistor transistor is said to be saturated. When fully
is that frequency at which the grounded - off, the transistor is said to be cutoff. The
base current gain has decreased to 0.7 of the region between these two extremes is termed
gain obtainable at 1kHz. For audio transis- the transition region. A transistor may be
tors the alpha cutoff frequency is about 1 used as a switch by simply biasing the base -
MHz. For r -f and switching transistors the emitter circuit on and off. Adjusting the
alpha cutoff frequency may be 50 MHz or base -emitter bias to some point in the tran-
higher. The upper frequency limit of opera- sition region will permit the transistor to
tion of the transistor is determined by the act as a signal amplifier. For such operation,
small but finite time it takes the majority base- emitter d -c bias will be about 0.3 volt
carriers to move from one electrode to the for many common germanium transistors,
other. and about 0.6 volt for many common sili-
The beta cutoff frequency (fhre) is that con transistors.
frequency at which the grounded- emitter
current gain has decreased to 0.7 of the Transistor Transistors are usually classi-
gain obtainable at 1 kHz. Transconductance Classifications fled according to the manu-
cutoff frequency (f g,,,) is that frequency at facturing techniques used to
which the transconductance falls to 0.7 of construct the two junctions. Figure 14
that value obtainable at 1 kHz. The maxi- lists some of the many classifications of
mum frequency of oscillation (f,,, x) is that junction transistors. Manufacturing tech-
frequency at which the maximum power niques, transistor end -use and patent restric-
gain of the transistor drops to unity. tions result in a multitude of transistor
Various internal time constants and tran- types, most of which fall in these general
sit times limit the high- frequency response classifications. Transistors, moreover, may
of the transistor and these limitations are be grouped in families where each member
summarized in the gain- bandwidth product
(f e) , which is identified by the frequency at PROCESS
COMMERCIAL TYPICAL
which the beta current gain drops to unity. DESIGNATION CETYPE5S1 TRANSISTOR
These various cutoff frequencies and the
IRATE GROWN NPN - 2N167 Z
gain- bandwidth products are shown in fig-
-
s
ure 13. MELT -BACK NPN 2NI 289 0
M
GROWN NPN 2N335
-
DIFFUSED PNP
I..
M111111111111111110 ff - fhfe xhf< NPN 2N1149
I0M11
DOUBLE-DOPED
10IBI11IBIBIIEIB
II1l11eimem' 6 oar R OCTA% C
"iMr-____ll\1
iiaaEscs
SLOPC
NNE ALLOY PNP
NPN
PNP
--
NPN 2N525
ao
MIuMIIii
111111V E\NIM\
f0llllll1lllllIeIII aum. hfb
imurnammomo max
DRIFT
MESA
PNP
PNP
-
2N247
2N695
z
O
J1
o
lo
'1l`!'1N
001
0y//,,I
0.1 1 0
F (M141
10 100 1000
SURFACE
BARRIER
MICRO -ALLOY
MA DT
PNP
PNP
PNP
PNP
NPN
2N344
2N393
2N501
n
11.
Figure 13 Figure 14
www.americanradiohistory.com
108 Semiconductor Devices THE RADIO
of the family is a unique type, but subtle input resistance. The input resistance of a
differences exist between members in the typical transistor is low, in the neighbor-
matter of end -use, gain, capacitance, mount-
..
hood of 300 ohms, while the output resist-
i M
ing, case, leads, breakdown voltage charac- ance is relatively high, usually over 20,000
teristics, etc. The differences are important ohms. For a point- contact transistor, the
enough to warrant individual type identifi- resistance gain is usually over 60.
__.
cation of each member. In addition, the
state of the art permits transistor parameters
to be economically designed to fit the var-
\
ious equipments, rather than designing the
equipment around available transistor types.
1/.Mi_
s
This situation results in a great many tran- J
I%ET
sistor types having nearly identical general
characteristics. Finally, improved manufac-
I/n -
turing techniques may "obsolete" a whole J '
family of transistors with a newer, less -ex- ,
.I
pensive family. It is recommended, therefore,
.
that the reader refer to one of the various 10 20 30 40 so
COLLECTOR VOLTS
transistor substitution manuals for up -to-
OA
date guidance in transistor classification and
substitution.
5 -5 Transistor
so El I
0
Characteristics
,
The transistor produces results that may
be comparable to a vacuum tube, but there o 20 ,'_A
is a basic difference between the two devices.
._1.
The vacuum tube is a voltage- controlled
device whereas the transistor is a current -
controlled device. A vacuum tube normally
+
ABER
25 50 75 100
PLATE VOLTS
123 uo 175 200
www.americanradiohistory.com
HAN DBOOK Transistor Characteristics 109
BASE
-R- CURRENT
AMPLIFICATION
.0 0.7
Figure 18
Figure 17
LOW- FREQUENCY EQUIVALENT
EMITTER CHARACTERISTIC CURVE (Common Base) CIRCUIT FOR POINT -
FOR TYPICAL POINT-CONTACT CONTACT AND JUNCTION
TRANSISTOR TRANSISTOR
www.americanradiohistory.com
110 Semiconductor Devices THE RADIO
Figure 19
COMPARISON OF BASIC VACUUM -TUBE AND TRANSISTOR CONFIGURATIONS
grounded -plate circuits in vacuum -tube ter- low output impedance, and no phase reversal
minology (figure 19). of signal from input to output circuit.
The grounded -base circuit has a low input Power and voltage gain are both low.
impedance and high output impedance, and Figure 20 illustrates some practical vac-
no phase reversal of signal occurs from input uum -tube circuits, as applied to transistors.
to output circuit. The grounded-emitter cir-
cuit has a higher input impedance and a 5 -6 Transistor Circuitry
lower output impedance than the grounded -
base circuit, and a reversal of phase be- To establish the correct operating para-
tween the input and output signal occurs. meters of the transistor, a bias voltage must
This usually provides maximum voltage gain be established between the emitter and the
from a transistor. The grounded -collector base. Since transistors are temperature- sensi-
circuit has relatively high input impedance. tive devices, and since some variation in
Figure 20
www.americanradiohistory.com
HANDBOOK Transistor Circuitry 111
-E
E
LOAD BIAS LOAO LOAD
BIAS RESISTOR RESISTOR RESISTOR
RESISTOR RESISTOR
Rz= to Re
Re = soo-+000 n
2
LT
t so (REVERSE
FOR NPN
POLARITY
TRANS /STR)
O O
Figure 21
The voltage divider system of C is recommended for general transistor use. Ratio of RI /RI
establishes base bias, and emitter bias is provided by voltage drop
a R. Battery polarity is reversed for NPN transistors.
characteristics usually exists between tran- is low, the operating point moves upward
sistors of a given type, attention must be along the load line, thus providing auto-
given to the bias system to overcome these matic control of the base bias voltage. This
difficulties. The simple self -bias system is circuit is sensitive to changes in ambient
shown in figure 21A. The base is simply temperature, and may permit transistor fail-
connected to the power supply through a ure when the transistor is operated near
large resistance which supplies a fixed value maximum dissipation ratings.
of base current to the transistor. This bias A better bias system is shown in figure
system is extremely sensitive to the current - 21C, where the base bias is obtained from a
transfer ratio of the transistor, and must be voltage divider, (R1, R2), and the emitter
adjusted for optimum results with each -
is forward biased. To prevent signal degen-
transistor. eration, the emitter bias resistor is bypassed
When the supply voltage is fairly high with a large capacitance. A high degree of
and wide variations in ambient temperature circuit stability is provided by this form of
do not occur, the bias system of figure 21B bias, providing the emitter capacitance is of
may be used, with the bias resistor connected the order of IO td for audio -frequency
from base to collector. When the collector applications.
voltage is high, the base current is increased,
moving the operating point of the transistor Audio Circuitry A simple voltage amplifier
down the load line. If the collector voltage is shown in figure 22. Di-
-12V
Figure 22
Figure 23
TWO-STAGE TRANSISTOR AUDIO AMPLIFIER
Tho feedback loop of A may be added to the RC amplifier to reduce distortion, or to control
the audio response. A direct-coupled amplifier is shown in C.
rect- current stabilization is employed in the in figure 23B. A direct- coupled version of
emitter circuit. Operating parameters for the resistance -coupled amplifier is shown in
the amplifier are given in the drawing. figure 23C. The input impedance is of the
In this case, the input impedance of the order of 15,000 ohms, and an over -all voltage
amplifier is quite low. When used with a gain of 80 may be obtained with a supply
high- impedance driving source such as a potential of 12 volts.
crystal microphone a step -down input trans- It is possible to employ NPN and PNP
former should be employed as shown in transistors in complementary- symmetry cir-
figure 22B. The grounded- collector circuit cuits which have no equivalent in vacuum-
of figure 22C provides a high input im- tube design. Figure 24A illustrates such a
pedance and a low output impedance, much circuit. A symmetrical push-pull circuit is
as in the manner of a vacuum -tube cathode shown in figure 24B. This circuit may be
follower. used to directly drive a high- impedance
The circuit of a two-stage resistance- speaker, eliminating the output transformer.
coupled amplifier is shown in figure 23A. A direct -coupled three -stage amplifier hav-
The input impedance is approximately 1100 ing a gain figure of 80 db is shown in figure
ohms. Feedback may be placed around this 24C.
amplifier from the emitter of the second The transistor may also be used as a class -
stage to the base of the first stage, as shown A power amplifier, as shown in figure 25A.
S
101JF
Figure 24
COMPLEMENTARY - SYMMETRY AMPLIFIERS
NPN and PNP transistors may be combined in circuits which have no equivalent in
vacuum-tube design. Direst coupling between cascaded stages using a single
power-supply source may be employed, as in C. impedance of power
supply should be extremely low.
www.americanradiohistory.com
HANDBOOK Transistor Circuitry 113
Figure 25 2N404
I MAX
MAXIMUM COLLECTOR
TYPICAL CLASS -A DISSIPATION (Ic X Ec)
AUDIO POWER
OPERATING POINT
TRANSISTOR CIRCUIT
The correct operating point
is chosen so that output
signal can swing equally in Ec 2Ec
a positive or negative di- COLLECTOR VOLTAGE
rection, without exceeding
maximum collector dissipa-
tion.
Commercial transistors are available that point is thus approximately one -half the
will provide five or six watts of audio power supply voltage.
when operating from a 12 -volt supply. The The circuit of a typical push -pull class -B
smaller units provide power levels of a few transistor amplifier is shown in figure 26A.
milliwatts. The correct operating point is Push -pull operation is desirable for transistor
chosen so that the output signal can swing operation, since the even -order harmonics
equally in the positive and negative direc- are largely eliminated. This permits transis-
tions, as shown in the collector curves of tors to be driven into high collector current
figure 25B. regions without distortion normally caused
The proper primary impedance of the out- by nonlinearity of the collector. Crossover
put transformer depends on the amount of distortion is reduced to a minimum by pro-
power to be delivered to the load: viding a slight forward base bias in addition
E,'
E
to the normal emitter bias. The base bias
is usually less than 0.5 volt in most cases.
R" 2P Excessive base bias will boost the quiescent
The collector current bias is: collector current and thereby lower the
2P over -all efficiency of the stage.
I,. _ E,
The operating point of the class -B ampli-
fier is set on the Ie =O axis at the point
In a class -A output stage, the maximum where the collector voltage equals the sup-
a -c power output obtainable is limited to ply voltage. The collector-to- collector im-
0.5 the allowable dissipation of the transistor. pedance of the output transformer is:
The product ICE,. determines the maximum 2E,2
collector dissipation, and a plot of these R,_,. = P
values is shown in figure 25B. The load line
should always lie under the dissipation curve, In the class -B circuit, the maximum a -c
and should encompass the maximum possible power input is approximately equal to three
area between the axes of the graph for max- times the allowable collector dissipation of
imum output condition. In general, the load each transistor. Power transistors, such as the
line is tangent to the dissipation curve and 2N514 have collector dissipation ratings of
passes through the supply voltage point at 80 watts and operate with class -B efficiency
zero collector current. The d -c operating of about 67%. To achieve this level of
.7N
12 V
Figure 26
NO S/GNAL
OPERATING
CIRCUITRY J
POINT
O
2N109
COLLECTOR VOLTAGE Ec
www.americanradiohistory.com
114 Semiconductor Devices THE RADIO
2N242 2N44I P.P. the alpha cutoff frequency of the units is
T
sufficiently higher than the operating fre-
PROM
quency. Shown in figure 28A is a typical i -f
SPEECH AMP amplifier employing an NPN transistor. The
collector current is determined by a voltage
divider on the base circuit and by a bias re-
sistor in the emitter leg. Input and output
are coupled by means of tuned i -f trans-
formers. Bypass capacitors are placed across
Figuro 27 the bias resistors to prevent signal- frequency
35-WATT MODULATOR degeneration. The base is connected to a
low- impedance untuned winding of the in-
T,-48 -ohm CT to 3/8/16 sec. use 3 -ohm tap put transformer, and the collector is con-
as CT (Thordarson TR -61). T,-6-ohm CT to 6k
(Triad TT -66A), Transistors mounted on heat nected to a tap on the output transformer
sink with mica washers. to provide proper matching, and also to
make the performance of the stage relative-
operation the heavy -duty transistor relies on ly independent of variations between tran-
efficient heat transfer from the transistor sistors of the same type. With a rate -grown
case to the chassis, using the large thermal NPN transistor such as the G.E. 2N293, it
capacity of the chassis as a heal sink. An in- is unnecessary to use neutralization to obtain
finite heat sink may be approximated by circuit stability. When PNP 'alloy tran-
mounting the transistor in the center of a sistors are used, it is necessary to neutralize
6" X 6" copper or aluminum sheet. This the circuit to obtain stability (figure 28B).
area may be part of a larger chassis. The gain of a transistor i -f amplifier will
The collector of most power transistors is decrease as the emitter current is decreased.
electrically connected to the case. For appli- This transistor property can be used to
cations where the collector is not grounded control the gain of an i -f amplifier so that
a thin sheet of mica may be used between weak and strong signals will produce the
the case of the transistor and the chassis. same audio output. A special i-f strip incor-
Large, inexpensive power transistors such porating this automatic volume control ac-
as the 2N441 may be used as modulators tion is shown in figure 29.
for medium power a -m mobile equipment. R -f transistors may be used as mixers or
Such a modulator is shown in figure 27. It autodyne converters much in the same man-
is capable of a power output of about 35 ner as vacuum tubes. The autodyne circuit
watts and is capable of plate modulating a is shown in figure 30. Transformer T1 feeds
70 -watt transmitter. back a signal from the collector to the
emitter causing oscillation. Capacitor C,
R-F Circuitry Transistors may be used for tunes the oscillator circuit to a frequency
radio-frequency work provided 45 5 kHz higher than that of the incoming
2N293
NPN y
TO
MIXER
OR
CONVERTER
-
bOVT.
MIXER
TO
OR
CONVERTER
ION !21{
+9V
Figurg 28
TRANSISTOR I -F AMPLIFIERS
Typical PNP transistor must be neutralized because of high collector capacitance.
Rate -grown NPN transistor does not usually require external neutralising
cirealt.
www.americanradiohistory.com
HANDBOOK Transistor Circuitry 115
~ 2N292 T-71
- 10 .0T
1 10
K
TO MIXER I
.05 .02=
.01 R1
\I-1(
.05 .05 R2 2.2 10
470 30 4.7K K K 50 470 '1.51(
A VC
62K
AUDIO OUT
-9 V.
Figure 29
signal. The local -oscillator signal is induc- currents of 0.6 ma to 0.8 ma are common,
tively coupled into the emitter circuit of and the local-oscillator injection voltage at
the transistor. The incoming signal is reso- the emitter is in the range of 0.15 to 0.25
nated in T2 and coupled via a low- imped- volt, rms.
ance winding to the base circuit. Notice that A receiver "front end" capable of opera-
the base is biased by a voltage- divider cir- tion through the 10 -meter band is shown
cuit much the same as is used in audio -fre- in figure 31. The inexpensive RCA type
quency operation. The two signals are mixed 2N1177 or 2N1180 transistors are used. If
in this stage and the desired beat frequency proper shielding is employed between the
of 455 kHz is selected by i -f transformer tuned circuits of the r -f stage and the
T, and passed to the next stage. Collector mixer, no neutralization of the r-f stage is
required. The complete assembly obtains
power from a 3 -volt battery. The base of
the r-f transistor is link- coupled to the r -f
coil to achieve proper impedance match.
The oscillator operates on its third harmonic
to produce an intermediate frequency of 1.6
MHz.
The epitaxial planar transistors of the
silicon NPN family have characteristics
TOI -F
which make them useful as general -purpose
AMPLIFIER
r-f amplifiers at frequencies up to 450 MHz
Ta or so. These characteristics include low noise
figure, low leakage current and a high gain -
www.americanradiohistory.com
116 Semiconductor Devices THE RADIO
14 MHz 2N1177 (R-F) 14 MHz 2N1177 (MIX.) 1.6 MHz
2N1177 (osc.)
50
22
47 K Too1
_ _ +3V.
Figure 31
Transistor Sufficient coupling of the proper transistor oscillators are shown in figure 33.
Oscillators phase between input and output A simple grounded-emitter Hartley oscil-
circuits of the transistor will per- lator having positive feedback between the
mit oscillation up to and slightly above the base and the collector (33A) is compared
alpha cutoff frequency. Various forms of to a grounded -base Hartley oscillator (33B).
In each case the resonant tank circuit is
common to the input and output circuits of
2N3478
the transistor. Self -bias of the transistor is
employed in both these circuits.
ANT.
A typical transistor crystal oscillator and
frequency -multiplier circuit are shown in
figure 34. The 2N707 NPN transistor op-
erates at 25 MHz, driving a 2N2218 doubler
to 50 MHz and a 2N2786 amplifier. Crystal
CR1 is for bias stabilization.
The point- contact transistor exhibits neg-
K T .001 ative input and output resistances over part
6 V.
of its operating range, due to its unique
Figure 32 ability to multiply the input current. Th;s
characteristic affords the use of oscillator
VHF LOW -NOISE TRANSISTOR PREAMPLIFIER circuitry having no external feedback paths
Collector voltage and base bias are adjusted for (figure 35). A high- impedance resonant
best noise fleure. circuit in the base lead produces circuit in-
www.americanradiohistory.com
HANDBOOK Transistor Circuitry 117
RFC
1-3 K
RFC
O O
Figure 33
/ WATT
1-2F
R OUTPUT
12 v
Figure 34
30
-9V.
270
O O
Figure 35 Figure 36
www.americanradiohistory.com
118 Semiconductor Devices THE RADIO
-E
+E
z
POSITIVE
TRIGGER
PULSE
II IOR 10
OuT
NPN
10R
__ e PNP P NP
0 10 R
Figuro 37
highest alpha cutoff frequency among the ploying surface -barrier transistors may be
group of "alpha -less- than -unity" transistors. driven between `off" and "on" positions by
Relaxation oscillators used for high -speed an exciting pulse as shown in figure 37B.
counting require transistors capable of op- The first pulse drives the "on" transistor into
eration at repetition rates off to 10 MHz. saturation. This transistor remains in a
A simple emitter -controlled relaxation os- highly conductive state until the second
cillator is shown in figure 36, together with exciting pulse arrives. The transistor does
its operating characteristic. The emitter of not immediately return to the cutoff state,
the transistor is biased to cutoff at the start since a time lapse occurs before the output
of the cycle (point 1). The charge on the waveform starts to decrease, This storage
emitter capacitor slowly leaks to ground time is caused by the transit lag of the
through emitter resistor R,. Discharge time minority carriers in the base of the transis-
is determined by the time constant of R,C,. tor. Proper circuit design and the use of
When the emitter voltage drops sufficiently high -alpha transistors can reduce the effects
low to permit the transistor to reach the of storage time to a minimum. Driving
negative- resistance region (point 2) the pulses may be coupled to the multivibrator
emitter and collector resistances drop to a through steering diodes as shown in the
low value, and the collector current is lim-
illustration.
ited only by collector resistor R2. The col-
lector current is abruptly reduced by the
charging action of emitter capacitor C, 5 -7 The Field
(point 3) , bringing the circuit back to the Effect Transistor
original operating point. The "spike" of
collector current is produced during the The Field Effect Transistor (FET) , or
charging period of C1. The duration of the unipolar transistor, is an N- or P- channel
pulse and the pulse- repetition frequency amplifying device that modulates the flow
(p.r.f.) are controlled by the values of C,, of current in a semiconductor channel by
R1, R2, and R3.
Transistors may also be used as blocking establishing regions of depletion (lack of
oscillators (figure 37A). The oscillator may current carriers: holes or electrons) between
be synchronized by coupling the locking the electron source and the drain. Depletion
signal to the base circuit of the transistor. control is exercised by a gate consisting of
An oscillator of this type may be used to a junction of opposite intrinsic material sand-
drive a flip -flop circuit as a counter. An wiching part of the conducting path (fig-
Eccles- Jordan bistable flip -flop circuit em- ure 38).
www.americanradiohistory.com
HANDBOOK Field Effect Transistor 119
2N3823
ANT.
+9V. +9V.
Figure 39
TYPICAL FET R -F AMPLIFIER CIRCUITS
www.americanradiohistory.com
CHAPTER SIX
6 -1 Vacuum -Tube
Parameters Ebb - Electrode Potentials
d -c plate-supply voltage (a positive
quantity)
The ability of the control grid of a vacu- E -d -c grid- supply voltage (a negative
um tube to control large amounts of plate
power with a small amount of grid energy -
Eg,,,
quantity)
peak grid excitation voltage (V2 total
allows the vacuum tube to be used as an
amplifier. It is this ability of vacuum tubes
to amplify an extremely small amount of
Epm -
peak -to -peak grid swing)
peak plate voltage (V2 total peak -to-
peak plate swing)
e,-
energy up to almost any level, without
change in anything except amplitude, which
makes the vacuum tube such an extremely
e5--
epmi
instantaneous plate potential
instantaneous grid potential
minimum instantaneous plate volt-
p-
valuable adjunct to modern electronics and age
communication. eg, maximum positive instantaneous grid
voltage
As an assistance in simplify - E,-
Symbols for
Vacuum -Tube ing and shortening expressions
Parameters involving vacuum -tube pa-
Eg
Et.
-- static plate voltage
static grid voltage
cutoff bais
rameters, the following sym-
bols will be used throughout this book: Electrode Currents
lb-
-- Tube Constants
amplification factor -- average plate current
1 average grid current
RP
-- plate resistance
g, transconductance
Ipm
max -peak fundamental plate current
maximum instantaneous plate cur-
- -- -
ag grid- screen mu factor rent
ge conversion transconductance (mixer igmax maximum instantaneous grid current
tube) Ip static plate current
C59
-- grid -plate capacitance
Co plate -cathode capacitance
Ci input capacitance (tetrode or pen-
Pi
P.-
- Other Symbols
plate power input
plate power output
P,-
C t- tode)
tode)
output capacitance (tetrode or pen- Pit- plate dissipation
grid- driving power (grid plus bias
losses)
www.americanradiohistory.com
CoP:T:
r--
1
TRIODE
CPK
Ciw
---
Cw,-
:
-
r--1
L._ _
.
- -,
PENTODE OR TETRODE
I
`
i
CouT
Vacuum -Tube Parameters
Cgkldynamic)
where,
Cgk is the
- Cgk(static) + (A +
Co
--
working into a resistance plate load in the
Pg, grid dissipation middle audio range.
plate efficiency (expressed as a deci- The more complete expression for the in-
Bp --
mal)
one -half angle of plate- current flow
put admittance (vector sum of capacitance
and resistance) of an amplifier operating into
--
19g one -half angle of grid-current flow any type of plate load is as follows:
RL load resistance
ZL load impedance input capacitance = Cgk+ (1 + A cos B) Cgp
www.americanradiohistory.com
122 Vacuum -tube Amplifiers THE RADIO
tion of low -noise triodes, this practice has linear r -f amplifiers. Gain is lower and dis-
been largely superseded through the use of tortion higher than for class -A amplifiers.
tetrode and pentode tubes in which the Cr;, Class -AB,A class -AB2 amplifier is operated
or feedback capacitance has been reduced to Amplifier under essentially the same condi-
such a low value that neutralization of its
tions of grid bias as the class-AB,
effects is not necessary to prevent oscillation
amplifier mentioned above, but the exciting
and instability.
voltage is of such amplitude that grid cur-
rent flows over an appreciable portion of
6 -2 Classes and Types of the input wave cycle.
Vacuum -Tube Amplifiers
Class -B A class -B amplifier is biased sub -
Vacuum -tube amplifiers are grouped into Amplifier stantially to cutoff of plate cur-
various classes and subclasses according to rent (without exciting voltage) so
the type of work they are intended to per- that plate current flows essentially over one -
form. The difference between the various half the input voltage cycle. The operating
classes is determined primarily by the angle angle of plate- current flow is 180 . The
of plate- current flow, of average grid bias class -B amplifier is usually excited to the
employed, and the maximum value of the extent that grid current flows.
exciting signal to be impressed on the grid.
Class -C A class -C amplifier is biased to a
Class -A A class -A amplifier is an amplifier Amplifier value greater than the value re-
Amplifier biased and supplied with excitation quired for plate-current cutoff
of such amplitude that plate cur- and is excited with a signal of such ampli-
rent flows continuously (360 of the excit- tude that grid current flows over an appreci-
ing voltage waveshape) and grid current able period of the input -voltage waveshape.
does not flow at any time. Such an amplifier The angle of plate- current flow in a class -C
is normally operated in the center of the amplifier is appreciably less than 180 , or
grid- voltage plate -current transfer charac- in other words, plate current flows appreci-
teristic and gives an output waveshape which ably less than one -half the time. Class -C
is a substantial replica of the input wave- amplifiers are not capable of linear amplifica-
shape. tion as their output waveform is not a rep-
lica of the input voltage for all signal ampli-
Class -A operation is employed in most tudes.
small -signal applications such as in receivers
and exciters. This mode of operation is char- Types of There are three general types of
acterized by high gain, low distortion, and Amplifiers amplifier circuits in use. These
low efficiency. Class -A mode may be further types are classified on the basis
subdivided into A, and A2 operation signify- of the return for the input and output cir-
ing the degree of grid drive on the stage, cuits. Conventional amplifiers are called
with the A2 mode signifying grid drive ap- grid-driven amplifiers, with the cathode ac-
proaching the class -AB, mode. ting as the common return for both the in-
put and output circuits. The second type is
Class -AB, This is an amplifier operated known as a plate- return amplifier or cathode
Amplifier under such conditions of grid
bias and exciting voltage that
plate current flows for more than one -half
the input voltage cycle but for less than
the complete cycle. In other words the
operating angle of plate current flow is ap-
preciably greater than 180 but less than
360 . The suffix , indicates that grid current Figure 2
does not flow over any portion of the input
TYPES OF BIAS SYSTEMS
cycle.
A - Grid bias
Class-AB,operation is utilized in most high - Cathode Has
quality, medium -power audio amplifiers and C - Grid-leak bias
www.americanradiohistory.com
HANDBOOK Classes of Amplifiers 123
follower since the plate circuit is effectively drops from an infinite value to a very low
at ground for the input and output signai value (on the order of 1000 ohms or so)
voltages and the output voltage or power is making the charging time constant of the
taken between cathode and plate. The third capacitor very short. This enables Cg to
type is called a cathode -driven or grounded - charge up to essentially the full value of the
grid amplifier since the grid is effectively at positive input voltage and results in the grid
ground potential for input and output sig- (which is connected to the low -potential
nals and output is taken between grid and plate of the capacitor) being held essentially
plate. at ground potential. During the negative
swing of the input signal no grid current
6 -3 Biasing Methods flows and the discharge path of Cg is through
the grid resistance which has a value of
The difference in potential between grid 500,000 ohms or so. The discharge time con-
and cathode is called the grid bias of a vacu- stant for Cg is, therefore, very long in com-
um tube. There are three general methods parison to the period of the input signal and
of providing this bias voltage. In each of only a small part of the charge on Cg is lost.
these methods the purpose is to establish the Thus, the bias voltage developed by the dis-
grid at a potential with respect to the cath- charge of Cg is substantially constant and
ode which will place the tube in the desired the grid is not permitted to follow the posi-
operating condition as determined by its tive portions of the input signal.
characteristics.
Grid bias may be obtained from a source
of voltage specially provided for this pur-
6 -4 Distortion in Amplifiers
pose, such as a battery or other d -c power
supply. This method is illustrated in figure There are three main types of distortion
2A, and is known as fixed bias. that may occur in amplifiers: frequency dis-
A second biasing method is illustrated in tortion, phase distortion and amplitude dis-
figure 2B which utilizes a cathode resistor tortion.
across which an IR drop is developed as a Frequency Frequency distortion may occur
result of plate current flowing through it. Distortion when some frequency compo-
The cathode of the tube is held at a positive nents of a signal are amplified
potential with respect to ground by the more than others. Frequency distortion oc-
amount of the IR drop because the grid is curs at low frequencies if coupling capaci-
at ground potential. Since the biasing volt- tors between stages are too small, or it may
age depends on the flow of plate current the
occur at high frequencies as a result of the
tube cannot be held in a cutoff condition shunting effects of the distributed capacities
by means of the cathode bias voltage devel- in the circuit.
oped across the cathode resistor. The value
of this resistor is determined by the bias Phase In figure an input signal con-
3
required and the plate current which flows Distortion sisting of a fundamental and a
at this value of bias, as found from the third harmonic is passed through
tube characteristic curves. A capacitor is a two -stage amplifier. Although the ampli-
shunted across the bias resistor to provide a tudes of both components are amplified by
low -impedance path to ground for the a -c identical ratios, the output waveshape is
component of the plate current which re- considerably different from the input signal
sults from an a -c input signal on the grid. because the phase of the third -harmonic
The third method of providing a biasing signal has been shifted with respect to the
voltage is shown in figure 2C, and is called fundamental signal. This phase shift is
grid-leak bias. During the portion of the known as phase distortion, and is caused
input cycle which causes the grid to be posi- principally by the coupling circuits between
tive with respect to the cathode, grid cur- the stages of the amplifier. Most coupling
rent flows from cathode to grid, charging circuits shift the phase of a sine wave, but
capacitor Cg. When the grid draws current, this has no effect on the shape of the out-
the grid -to- cathode resistance of the tube put wave. However, when a complex wave
www.americanradiohistory.com
124 Vacuum -Tube Amplifiers THE RADIO
INPUT
SIGNAL
Figure 4
OUTPUT
SIGNAL
STANDARD CIRCUIT FOR RESISTANCE -
CAPACITANCE COUPLED TRIODE
Figure 3 AMPLIFIER STAGE
RL RG
G A=
Re (RL+RG)+RL RG
E 11EG
A HIGH FRED.
COK
(DYNAMIC. A WO FRED.
11--LEG NEXT STAGE) (I+ (REQ /XS)2
Ri.
REP-
RL RL
HIGH- FREQUENCY RANGE 1t
RG Re
Xs -
277r (CPK +CGK (DYNAMIC)
A LOW FRED. I
G
A MID FRED.
E=-JJEG J 1 +(XC /R)2
Xc = I
2 nrcc
LOW- FREQUENCY RANGE
RL RP
R = RG+ RL+ RP
Figure 5
Equivalent circuits and gain equations for a triode RC- coupled amplifier stage. In using these
equations, be sure the values of and Rp are proper for the static and voltages with
which the tube will operate. These values may be obtained from curves published in the RCA
Receiving Tube Manual (series RC).
www.americanradiohistory.com
HANDBOOK RC- Coupled Amplifiers 125
amplifier is excited with a signal that over- that is low with respect to the cathode re-
drives the tubes, the resultant signal is dis- sistor at the lowest frequency to be passed
torted in amplitude, since the tubes are then by the amplifier stage.
operating over a nonlinear portion of their
characteristic.
6 -5 Resistance -
Capacitance Coupled
Audio- Frequency Amplifiers
Present practice in the design of audio -
frequency voltage amplifiers is almost ex- Figure 6
clusively to use resistance-capacitance
coupling between the low -level stages. Both STANDARD CIRCUIT FOR RESISTANCE -
CAPACITANCE COUPLED PENTODE
triodes and pentodes are used; triode ampli-
fier stages will be discussed first. AMPLIFIER STAGE
www.americanradiohistory.com
126 Vacuum -Tube Amplifiers THE RADIO
dropping resistor Rd at the lowest frequency
to be passed by the amplifier stage. 100 1. RL= 500000 OHMS
I= -GMEc
A = GM Rep
y
G
RED = RC
+R+
e Re
Figure 7
MID -FREQUENCY RANGE
Equivalent circuits and
gain equations for a
pentode RC- coupled am- I= -GMEG
plifier stage. In using A HIGH FRED
these equations be sure A MID reED +(REQ /X5)2
to select the values of r 1
g R which are
proper for the static REQ=
+ HL
currents and voltages 1
RD + Re
with which the tube will HIGH FREQUENCY RANGE
operate. These values Xs - 1
XC -
27rr Cc
LOW -FREQUENCY RANGE RR
R = RG + RL
RL+Rv
www.americanradiohistory.com
HANDBOOK Video -Frequency Amplifiers 127
Grid -Leak Bias The correct operating uniform amplification for frequencies in the
for High -Mu Triodes bias for a high -mu tri- audio range of about 100 to 20,000 Hz.
ode such as the 12AT7, Changes in the values of coupling capacitors
is fairly critical, and will be found to be and load resistors can extend this frequency
highly variable from tube to tube because range to cover the very wide range required
of minute variations in contact potential for video service. However, extension of the
within the tube itself. A satisfactory bias range can only be obtained at the cost of
method is to use grid -leak bias, with a grid reduced over -all amplification. Thus RC cou-
resistor of one to ten megohms connected pling allows good frequency response with
directly between grid and cathode of the minimum distortion, but low amplification.
tube with the cathode grounded. Grid cur- Phase distortion is less with RC coupling
rent flows at all times, and the effective in- than with other types, except direct coup-
put resistance is about one -half the resis- ling. The RC amplifier may exhibit tenden-
tance value of the grid leak. This circuit is cies to motorboat or oscillate if it is used
particularly well suited as a high -gain amp- with a high- impedance plate supply.
lifier following low- output devices, such as
crystal, or dynamic microphones. 6 -6 Video -Frequency
RC Amplifier A resistance - capaci- Amplifiers
General Characteristicstance coupled ampli-
fier can be designed A video- frequency amplifier is one which
to provide a good frequency response for has been designed to pass frequencies from
almost any desired range. For instance, such the lower audio range (lower limit perhaps
SO Hz) to the middle r -f range (upper
an amplifier can be built to provide a fairly
limit perhaps 4 to 6 MHz) . Such amplifiers,
in addition to passing such an extremely
wide frequency range, must be capable of
amplifying this range with a minimum of
amplitude, phase, and frequency distortion.
Video amplifiers are commonly used in tele-
vision, pulse communication, and radar
work.
Tubes used in video amplifiers must have
a high ratio of g,0 to capacitance if a usable
MIO-rNtOUtNCT GAIN GMVI RL
gain per stage is to be obtained. Commonly
NIGH-rNCDUCNCTGAIN Gm TI ICOUALINGNCTWONH
available tubes which have been designed
C CouTVI.C,NV?.CDISTRIBUTED for or are suitable for use in video amplifiers
FOR c0rPRorlst HIG.rI1COUtNC. cowLl:wTION are: 6AU6, 6AGS, 6AKS, 6CB6, 6BC5,
XL,.. 0 AL AT rc
6DE6, and 6AH6. Since, at the upper fre-
RL xc rc
AT
quency limits of a video amplifier the input
WHORL rc CuTOrr.RCDUCHCT Or AMPLIFIER
and output shunting capacitances of the
LL REARING INDUCTOR
amplifier tubes have rather low values of re-
FOR COMPROMISE LOW[wCOUCNCT EQUALIZATION
actance, low values of coupling resistance,
R, RN (Grvl RL)
along with peaking coils or other special in-
RO C. RR Ce
terstage coupling impedances, are usually
CR if TO 50 LID IN WITH 001 NOCA
www.americanradiohistory.com
128 Vacuum -Tube Amplifiers THE RADIO
www.americanradiohistory.com
HANDBOOK I nterstage Coupling Methods 129
+e
PUSH -PULL TRANSFORMER COUPLING QD IMPEDANCE COUPLING
+8 +8
IMPEDANCE- TRANSFORMER COUPLING FO RESISTANCE- TRANSFORMER COUPLING
113
Figure 10
Cathode The cathode -coupling arrangement aient triode obtained through the use of a
Coupling of figure 10G has been widely used pair of similar triodes connected in the cath-
only comparatively recently. One ode- coupled circuit shown. With these equiv-
outstanding characteristic of such a circuit alent triode factors it is possible to use the
is that there is no phase reversal between the expressions shown in figure 5 to determine
grid and the plate circuit. All other common the gain of the stage at different frequencies.
types of interstage coupling are accompanied The input capacitance of such a stage is less
Dy a 180 phase reversal between the grid than that of one of the triodes, the effective
circuit and the plate circuit of the tube. grid -to -plate capacitance is very much less
Figure 11 gives the expressions for deter- (it is so much less that such a stage may be
mining the appropriate factors for an equiv- used as an r-f amplifier without neutraliza-
www.americanradiohistory.com
130 Vacuum -Tube Amplifiers THE RADIO
www.americanradiohistory.com
HANDBOOK Phase Inverters 131
www.americanradiohistory.com
132 Vacuum -Tube Amplifiers THE RADIO
.01 Rs Re
Figure 14
DIRECT-COUPLED D -C AMPLIFIER
Figure 13
then of the same amplitude as the output of
VOLTAGE -DIVIDER PHASE Vi, but of opposite phase.
INVERTER
6 -9 D -C Amplifiers
output grids. The phase characteristics are
such that the circuit is commonly used in Direct -current amplifiers are special types
deriving push -pull deflection voltage for a used where amplification of very slow varia-
cathode -ray tube from a single -ended input tions in voltage, or of d -c voltages is desired.
signal. A simple d -c amplifier consists of a single
The first section of the 12AU7 is used as tube with a grid resistor across the input
an amplifier to increase the amplitude of the terminals, and the load in the plate circuit.
applied signal to the desired level. The sec-
ond section of the 12AU7 is used as an in- Basic D -C A simple d -c amplifier cir-
verter and amplifier to produce a signal of Amplifier Circuit cuit is shown in figure 14,
the same amplitude but of opposite polarity. wherein the grid of one
Since the common cathode resistor (R11) is tube is connected directly to the plate of the
not bypassed the voltage across it is the alge- preceding tube in such a manner that volt-
braic sum of the two plate currents and has age changes on the grid of the first tube will
the same shape and polarity as the voltage ap- be amplified by the system. The voltage drop
plied to the input grid of the first half of across the plate coupling resistor is impressed
the 12AU7. When a signal (e) is applied to directly on the grid of the second tube,
the input circuit, the effective grid -cathode which is provided with enough negative grid
voltage of the first section is Ae /2, when A bias to balance out the excessive voltage drop
is the gain of the first section. Since the grid across the coupling resistor. The grid of the
of the second section of the 12AU7 is second tube is thus maintained in a slightly
grounded, the effect of the signal voltage negative position.
across Rk (equal to e/2 if Rk is the proper The d -c amplifier will provide good low -
value) is the same as though a signal of the frequency response, with negligible phase
same amplitude but of opposite polarity were distortion. High- frequency response is lim-
applied to the grid. The output of the sec- ited by the shunting effect of the tube ca-
ond section is equal to -Ae /2 if the plate pacitances, as in the normal resistance -
load resistors are the same for both tube coupled amplifier.
sections. A common fault with d -c amplifiers of all
types is static instability. Small changes in
Voltage- Divider A commonly used phase in- the filament, plate, or grid voltages cannot
Phase Inverter verter is shown in figure 13. be distinguished from the exciting voltage.
The input section (V,) is Regulated power supplies and special balanc-
connected as a conventional amplifier. The ing circuits have been devised to reduce the
output voltage from V, is impressed on the effects of supply variations on these ampli-
voltage divider R5 -R6. The values of R and fiers. A successful system is to apply the
R,; are in such a ratio that the voltage im- plate potential in phase to two tubes, and to
pressed on the grid of V2 is 1 /A times the apply the exciting signal to a push -pull grid -
output voltage of V1, where A is the ampli - circuit configuration. If the two tubes are
fication factor of V,. The output of V2 is identical, any change in electrode voltage is
www.americanradiohistory.com
HANDBOOK D -C Amplifiers 133
Figure 15
Figure 16
LOFTIN -WHITE D-C AMPLIFIER
PUSH -PULL D -C AMPLIFIER
balanced out. The use of negative feedback WITH EITHER SINGLE -ENDED
can also greatly reduce drift problems. OR PUSH -PULL INPUT
Q IMPEDANCE COUPLING
Eg
-
(0.68 X Ebb)
=
where,
P
Ebb is the actual plate voltage of the
class -A stage,
'is the amplification factor of the
K
tube.
+B 3. Locate the Eg bias point on the I
TRANSFORMER COUPLING
versus E graph where the Eg bias line
crosses the plate -voltage line, as shown
in figure 18. Call this point P.
4. Locate on the plate family of curves
the value of zero -signal plate current,
(1 )
corresponding to operating point
P.
+2 S. Locate 2 X 1 (twice the value of 1 )
IMPEDANCE -TRANSFORMER COUPLING on the plate- current axis (Y axis).
This point corresponds to the value of
maximum- signal plate current (i,,,nx)
6. Locate point x on the d -c bias curve
at zero volts (Es = 0), corresponding
to the value of im,x
-BIAS= +S
7. Draw a straight line (x -
y) through
TRANSFORMER COUPLING FOR A2 OPERATION
points x and P. This line is the load -
resistance line. Its slope corresponds to
the value of the load resistance.
AUTO- 8. Load resistance, (in ohms) equals:
TKANSrOPAKA
L - e,,,nx emIn
In
BIAS + TO CLASS-C
LOAD
where,
CLASS Ag MODULATOR
FORMER COUPLING
WITH AUTOTRANS-
e isin volts,
Figure 17 i isin amperes,
Output coupling arrangements for single -ended
R,, is in ohms.
class -A triode audio -free y power 9. Check: Multiply the zero -signal plate
amplifiers. current (1) by the operating plate
voltage, (E ).
If the plate dissipation
amplification factor, high plate dissipation, rating of the tube is exceeded, it is
and relatively high filament emission. necessary to increase the bias (Eg) on
The operating characteristics of a class -A the tube so that the plate dissipation
triode amplifier employing an output- trans- falls within the maximum rating of
former coupled load may be calculated from the tube. If this step is taken, opera-
the plate family of curves for the particular tions 2 through 8 must be repeated
tube in question by employing the following with the new value of Eg.
steps: 10. For maximum power output, the peak
a -c grid voltage on the tube should
1. The load resistance should be approxi- swing to 2E, on the negative cycle,
mately twice the plate resistance of and to zero -bias on the positive cycle.
the tube for maximum undistorted At the peak of the negative swing,
www.americanradiohistory.com
HANDBOOK
w
!oo
p
_
Ipp
'
r
I..
API
yy117
V_I
_ IIIII
1/
Iw/
:'1/IIVINIMIIIIMON
M MNIMww
r4111wIr-
f NrIIF111111111
IIIw
NI
Pentode Amplifiers
max
%min)
- imIn
135
X 100
N/I .
MM. IN ' plied to a single class -A 2A3 amplifier stage.
i.' IhvIY
tI!11IP `II11I
IIMIIIIIallg,.: r.,Nl
Nwuaa.a
Iu.NM
Mi II a 6 -11 Single -Ended Pentode
Amplifiers
LI
o
-
ErIN.
Ioo
Ki
PLATE VOLTS
n/7r./L/
Eus
Zoo
f'o
noo I .00
Figure 19 illustrates the conventional cir-
cuit for a single -ended tetrode or pentode
.
where,
i is in amperes,
e is in volts. Figure 19
11. The second -harmonic distortion gen- C fional single -ended pentode or beam
erated in a single-ended class -A triode tetrode audio-frequency power -output stage.
www.americanradiohistory.com
i,/
Y
iw
136
rq.}m
t11!
MINEM
rs-ia,1
'
wuMs.
iaEuailiEM
r-
WAIIIIIle
/
JmlN-L
e
=is
..020511=
/Yiii!R!
lIfrilC
,iIMIE
e.= w_..
:-
Vacuum -Tube Amplifiers
--r
4A
5l3!F,>r
YIN
LOAD LINE 50 THAT
CHOOSE
AP
i7
/110"'
(STATIC VALUE) eYAx
PLATE VOLTS
MN
EG. v.
1.
2.
THE RADIO
pentode amplifiers, the following steps are
taken:
Figure 21
www.americanradiohistory.com
HANDBOOK Push -Pull Amplifiers 137
NSssm!s
sMms, Nmo
so lllllll111111Nlliiiiiiliiiiiliiiiiii
300
EMmIMI
mummmmsommoomonwramommommum
i
oomossm/Y M
iiGmiiiiGiiii
Mm ms // .. oM
iilGiiaiiii
`almo
IMNA11111I1N1u111111111111111111
M
IM NIIIII
NI/Im
m
mMMM oi
MN Mm/ aANMw M
osss o\ /I.sQ
' INN NNI
IMIN MINI M IN
MIN MNI IM IM
Il
I11IMN IMU 250
NN 11111! IM MIAI
!I! IN1IpNI1 rI N
1gM1q'I'
;;11111/Al :"
M sIM
MN .smmMNM NIN NNI NN1 INN N1I I 11I'd N N1 NNI W
ssM
MMMM1 EMM r.sss'MMr. IMM MA MNI NMI M N 111 II11IM NNI 150 `
INN MsI NN11M NI1I r41111111111111
a
MsMM srIMl"!!!!!!C
1Emr/ApM Amm 19!!!!SI!!!!!!N!1I!!!!'1111I IIIM 11111
MMMM Im
MMMMri..m //H'EM'M
,m rw.M
1111111111111N11111111111/41111111111N111
NIIININNIN1N11IV
1u1I1MNININUr.nNMI
IUMIII
NNMEMAMr/
MMMS//M/.sm ,MMI'eeleJ
/M I MUMWA.MU !MN4!lLENIPB!!II!a IIIIIIIIIIMmm
III II NNI MI1N .11111111111 MI INN <J
100 W
Figure 22
where, eg is equal to: E,, + 0.7 E,. output transformer; (2) even harmonics of
7. The percentage harmonic distortion is: the input signal (second and fourth harmon-
'% 2nd harmonic distortion = ics primarily) generated in the push -pull
stage are cancelled when the tubes are bal-
imax mIn -219 X 100 anced.
max -rmin+ 1.41 (I. -ly) The cancellation of even harmonics gener-
where, ated in the stage allows the tubes to be oper-
11, is the static plate current of the ated class AB-in other words the tubes may
tube. be operated with bias and input signals of
3rd harmonic distortion = such amplitude that the plate current of
1r:1ax -1min -1.41 -1y) X (1x
alternate tubes may be cut off during a por-
100 tion of the input voltage cycle. If a tube
Jmnx- Jn, 1, +1.41,(1. -1y)
were operated in such a manner in a single -
ended amplifier the second -harmonic ampli-
tude generated would be prohibitively high.
6 -12 Push -Pull Audio Push -pull class -AB operation allows a
Amplifiers plate circuit efficiency of from 45 to 60
percent to be obtained in an amplifier stage
A number of advantages are obtained depending on whether or not the exciting
through the use of the push -pull connection voltage is of such amplitude that grid cur-
of two or four tubes in an audio- frequency rent is drawn by the tubes. If grid current
power amplifier. Two conventional circuits is taken on input voltage peaks the amplifier
for the use of triode and tetrode tubes in the is said to be operating class -AB2 and the
push -pull connection are shown in figure 21. plate- circuit efficiency can be as high as the
The two main advantages of the push -pull upper value just mentioned. If grid current
circuit arrangement are: (1) the magnetiz- is not taken by the stage it is said to be
ing effect of the plate currents of the output operating class -AB, and the plate- circuit
tubes is cancelled in the windings of the efficiency will be toward the lower end of
138 Vacuum -Tube Amplifiers THE RADIO
the range just quoted. In all class-AB ampli- output should be subtracted from this
fiers the plate current will increase from 40 value to obtain the total operating
to 150 percent over the no- signal value plate dissipation of the two tubes. If
when full excitation voltage is applied. the plate dissipation is excessive, a
Operating Characteristics The operating char - slightly higher value of RL should be
of Push -Pull Class -A chosen to limit the plate dissipation.
acteristics of push - The correct value of operating bias,
Triode Power Amplifier 5.
pull class-A ampli-
and the static plate current for the
fiers may also be
push -pull tubes may be determined
determined from the plate family of curves
from the Eg versus Ip curves, which
for a particular triode tube by the following
are a derivation of the Ep versus I,
steps:
curves for various values of Eg.
1. Erect a vertical line from the plate - 6. The Eg versus Ip curve may be con-
voltage axis (x -axis) at 0.6 E (figure structed in this manner: Values of
22) , which intersects the Eg = O grid bias are read from the intersection
curve. This point of intersection (P), of each grid -bias curve with the load
interpolated to the plate current axis line. These points are transferred to
(y- axis), may be taken as 1, ax. It is as- the Eg versus Ip graph to produce a
sumed for simplification that i,,,ax oc- curved line, A -B. If the grid bias
curs at the point of the zero -bias curves of the Ep versus I, graph were
2.
curve corresponding to 0.6 E
The power output obtainable from the
. straight lines, the lines of the Eg versus
I, graph would also be straight. This
two tubes is: is usually not the case. A tangent to
this curve is therefore drawn, starting
power output (P ) = max X Ep
f
at point A', and intersecting the grid -
voltage abscissa (x- axis). This inter-
section (C) is the operating -bias point
where,
for fixed -bias operation.
P is expressed in watts, 7. This operating -bias point may now be
i,,,ax in amperes, plotted on the original Eg versus Ip
E is the applied plate voltage. family of curves (C'), and the zero -
3. Draw a preliminary load line through signal current produced by this bias is
point P to the Ep point located on the determined. This operating bias point
x -axis (the zero plate- current line). (C'), does not fall on the operating
This load line represents %4 of the load line, as in the case of a single -
actual plate -to-plate load of the class - ended amplifier.
A tubes. Therefore: 8. Under conditions of maximum power
XEp -0.6 Ep
output, the exciting signal voltage
RL (plate -to- plate) =4 swings from zero -bias voltage to zero -
max
bias voltage for each of the tubes on
1.6 Ep
each half of the signal cycle. Second-
max harmonic distortion is largely cancelled
where, out.
RL is expressed in ohms,
Ep is expressed in volts, 6 -13 Class -B Audio -
imax is expressed in amperes.
Frequency Power
Figure 22 illustrates the above steps ap- Amplifiers
plied to a push -pull class -A amplifier using
two 2A3 tubes.
The class -B audio -frequency power ampli-
4. The average plate current is 0.636 imax, fier (figure 23) operates at a higher plate -
and multiplied by plate voltage E9, circuit efficiency than any of the previously
will give the average watts input to described types of audio power amplifiers.
the plates of the two tubes. The power Full- signal plate -circuit efficiencies of 60 to
www.americanradiohistory.com
HANDBOOK Class -B Audio Amplifiers 139
not specified in the tube operating charac- other words, when nonsinusoidal waves such
teristics published by the tube manufacturer. as plain speech or speech that has passed
The same procedure can be used with equal through a clipper are concerned, we are no
effectiveness for the calculation of the oper- longer concerned with average power output
ating conditions of beam tetrodes as class- of the modulator as far as its capability of
AB., amplifiers or modulators when the rest- modulating a class -C amplifier is concerned;
ing plate current of the tubes (no- signal con- we are concerned with its peak power out-
dition) is less than 25 or 30 percent of the put capability.
maximum -signal plate current. Under these conditions we call on other,
more general relationships. The first of these
1. With the average plate characteristics is: it requires a peak power output equal to
of the tube as published by the manu- the class -C stage input to modulate that in-
facturer before you, select a point on put fully.
the Ep = ER (diode bend) line at The second relationship is: the average
about twice the plate current you ex- power output required of the modulator is
pect the tubes to draw under modu- equal to the shape factor of the modulating
lation peaks. If beam tetrode tubes are wave multiplied by the input to the class -C
concerned, select a point at about the stage. The shape factor of unclipped speech
same amount of plate current men- is approximately 0.25. The shape factor of
tioned above, just to the right of the a sine wave is 0.5. The shape factor of a
region where the 1,, line takes a sharp speech wave that has been passed through a
curve downward. This will be the first clipper -filter arrangement is somewhere be-
trial point, and the plate voltage at tween 0.25 and 0.9 depending on the amount
the point chosen should be not more of clipping that has taken place. With 15 or
than about 20 percent of the d -c volt- 20 db of clipping the shape factor may be as
age applied to the tubes if good plate - high as the figure of 0.9 mentioned above.
circuit efficiency is desired. This means that the audio power output of
2. Note down the value of i, x and , the modulator will be 90% of the input to
ep min at this point.
the class -C stage. Thus with a kilowatt input
3. Subtract the value of ep from the ,I we would be putting 900 watts of audio
d -c plate voltage on the tubes. into the class -C stage for 100 percent modu-
4. Substitute the values obtained in the lation as contrasted to perhaps 250 watts for
following equations: unclipped speech modulation of 100 percent.
RI, (2 tubes) = 4
(Ebb - ep min) type 811A tube with a
load line for class -B operation. Figure 25
ip max
lists a sample calculation for determining the
Full signal efficiency (Np) = proper operating conditions for obtaining
approximately 185 watts output from a pair
(ep min) of the tubes with 1000 volts d -c plate po-
78.5 I 1
Ebb tential. Also shown in figure 25 is the meth-
od of determining the proper ratio for the
Effects of Speech All the above equations are modulation transformer to couple between
Clipping true for sine -wave operating the 811's or 811A's and the anticipated final
condition of the tubes con- amplifier which is to operate at 2000 plate
cerned. However, if a speech clipper is being volts and 175 ma plate current.
used in the speech amplifier, or if it is de-
sired to calculate the operating conditions Modulation Transformer The method illus-
on the basis of the fact that the ratio of Calculation trated in figure 25
peak power to average power in a speech can be used in general
wave is approximately 4 to 1 as contrasted
to the ratio of 2 to 1 in a sine wave in - for the determination of the proper trans-
former ratio to couple between the modula-
www.americanradiohistory.com
HANDBOOK Parameters 141
800
Figure 24 +W
s Ec
Typical class -8 a -f ampli-
fier load fine. Tho load lino
has Mon drawn on the
average characteristics of a
type 811 tube.
skiars
0 400 800 1200 1800
E c' o
2000 2400
PLATE VOLTS (Est.)
AVERAGE PLATE CHARACTERISTICS TYPE 811 AND 811 -A
tor tube and the amplifier to be modulated. for this condition is that the ratio between
The procedure can be stated as follows: (1) the total primary voltage and the d -c plate -
Determine the proper plate -to -plate load im- supply voltage on the modulated stage is
pedance for the modulator tubes either by equal to the turns ratio of the transformer,
the use of the type of calculation shown in since a peak secondary voltage equal to the
figure 25, or by reference to the published plate voltage on the modulated stage is re-
characteristics on the tubes to be used. (2) quired to modulate this stage 100 percent.
Determine the load impedance which will be
presented by the class-C amplifier stage to SAMPLE CALCULATION
be modulated by dividing the operating CONDITION 2 TYPE 811 TUBES. Eco... I0Oo
plate voltage on that stage by the operating INPUT TO FINAL STAGE, 350 W.
PEAR POWER OUTPUT NEEDED = 350 +5%' 370 W
value of plate current in amperes. (3) Di- FINAL AMPLIFIER Ebb 2000 V.
'' .175 A. 1140011
vide the class -C load impedance determined
in (2) above by the plate -to -place load im-
FINAL AMPLIFIER Ib
FINAL AMPLIFIER ZL 2_
.175
'
pedance for the modulator tubes determined EXAMPLE. CHOSE POINT ON 211 CHARACTERISTICS JUST
Ebb= Ecc (POINT X, FIG.24)
TO RIGHT OF
in (1) above. The ratio determined in this
.
ratio. (4) Take the square root of this ratio PEAK PO ' .410 X (1000 -100) ' .10 X 800' 388 W.
to determine the secondary-to- primary turns RL' 4X ' 88000.
ratio. If the turns ratio is greater than unity,
the use of a step -up transformer is required. NP = 78.5 (1 - 1000 )' 78.5 (s)'pp770
-
l
.55 mi,,,,
I184SW
WO (AVERAGE WITH SINE WAVE)
If the turns ratio as determined in this way
is less than unity, a step -down transformer is
WIN = AR-
Lb (MAXIMUM WITH SINE WAVE) ' 260 MA
called for. WG PEAK ' . 100 X '
80 S W.
If the procedure shown in figure 25 has DRIVING POWER ' WZ PK
- 4 W.
www.americanradiohistory.com
142 Vacuum -Tube Amplifiers THE RADIO
Use of Clipper Speech When a clipper speech the maximum drive requirements of the
Amplifier with Tetrode amplifier is used in class -B stage. If the step -down ratio of the
Modulator Tubes conjunction with a driver transformer is too large, the driver
class -B modulator plate load will be so high that the power re-
stage, the plate current on that stage will quired to drive the class-B stage to full out-
rise to a higher value with modulation (due put cannot be developed. If the step -down
to the greater average power output and ratio is too small the regulation of the driver
input) but the plate dissipation on the tubes stage will be impaired.
will ordinarily be less than with sine -wave Driver -Stage The parameters for the driver
modulation. However, when tetrode tubes Calculations stage may be calculated from
are used as modulators, the screen dissipation
the plate characteristic curve,
will bemuch greater than with sine -wave
a sample of which is shown in figure 24.
modulation. Care must be taken to ensure
The required positive grid voltage (eg max)
that the screen dissipation rating on the
modulator tubes is not exceeded under full
for the 811A tubes used in the sample cal-
culation is found at point X, the intersection
modulation conditions with a clipper speech
of the load line and the peak plate current
amplifier. The screen dissipation is equal to
as found on the y -axis. This is + 80 volts.
screen voltage times screen current.
If a vertical line is dropped from point X
Practical Aspects of As stated previously, a
to intersect the dotted grid- current curves,
Class -B Modulators class -B audio amplifier it will be found that the grid current for a
requires the driving single 811A at this value of grid voltage is
stage to supply well -regulated audio power 100 milliamperes (point Y). The peak grid -
to the grid circuit of the class -B stage. Since driving power is therefore 80 X 0.100 8 =
the performance of a class -B modulator may watts. The approximate average driving
easily be impaired by an improperly designed power is 4 watts. This is an approximate
driver stage, it is well to study the problems figure because the grid impedance is not
incurred in the design of the driver stage. constant over the entire audio cycle.
The grid circuit of a class -B modulator A pair of 2A3 tubes will be used as
may be compared to a variable resistance drivers, operating class -A, with the maxi-
which decreases in value as the exciting grid mum excitation to the drivers occurring just
voltage is increased. This variable resistance below the point of grid- current flow in the
appears across the secondary terminals of the 2A3 tubes. The driver plate voltage is 300
driver transformer so that the driver stage is volts, and the grid bias is -62 volts. The
called on to deliver power to a varying load. peak power (Pr) developed in the primary
For best operation of the class -B stage, the winding of the driver transformer is:
grid excitation voltage should not drop as
the power taken by the grid circuit increases. (Pr) = 2RI .
E
Rv+R / l 2
www.americanradiohistory.com
HANDBOOK Cathode -Follower Amplifiers 143
ROQCwrwoOt)=
RL (RKt+R,u) R[
11 RK, +R2+
PENTODE:
RomArHoot)= Rp R
Gtr 1+RL G,,,
A = G, Rep
Figure 27
Equivalent faders fer pentode (or tetrode)
cathode- follower power amplifiers
www.americanradiohistory.com
HANDBOOK Feedback Amplifiers 145
systems for feeding a coaxial cable of rela- point where the feedback is taken off and
tively low impedance from a vacuum -tube the point where the feedback energy is in-
amplifier. A pentode or tetrode tube with a serted are said to be included within the
cathode impedance as a cathode follower feedback loop. An amplifier containing a
(1 /g,,,) of approximately the same impe- feedback loop is said to be a feedback ampli-
dance as the cable should be chosen. The fier. One stage or any number of stages may
12BY7A and 6CL6 have cathode impedances be included within the feedback loop. How-
of the same order as the surge impedances of ever, the difficulty of obtaining proper oper-
certain types of low- capacitance coaxial ation of a feedback amplifier increases with
cable. An arrangement such as 26D is also the bandwidth of the amplifier, and with the
usable for feeding coaxial cable with audio or number of stages and circuit elements in-
r -f energy where it is desired to transmit the cluded within the feedback loop.
output signal over moderate distances. The Gain and Phase Shift The gain and phase
resistor Rk is added to the circuit as shown in Feedback Amplifiers shift of any amplifier
if the cathode impedance of the tube used is are functions of fre-
lower than the characteristic impedance of quency. For any amplifier containing a feed-
the cable. If the output impedance of the back loop to be completely stable, the gain of
stage is higher than the cable impedance, a such an amplifier, as measured from the in-
resistance of appropriate value is sometimes put back to the point where the feedback
placed in parallel with the input end of the circuit connects to the input, must be less
cable. The values of Cd and R,, should be than unity at the frequency where the feed-
chosen with the same considerations in mind back voltage is in phase with the input volt-
as mentioned in the discussion of the circuit
age of the amplifier. If the gain is equal to
of figure 26C. or more than unity at the frequency where
The Cathode Follower The cathode follower the feedback voltage is in phase with the in-
in R -F Stages may conveniently be put, the amplifier will oscillate. This fact im-
used as a method of poses a limitation on the amount of feedback
coupling r -f or i -f energy between two which may be employed in an amplifier
units separated a considerable distance. In which is to remain stable. If the reader is
such an application a coaxial cable should be
used to carry the r -f or i -f energy. One such
application would be for carrying the out-
put of a vfo to a transmitter located a con-
siderable distance from the operating posi-
e
INPUT SIGNAL ES AMPLIFIER
GAINA
OUTPUT E
/ECDBACK OR E PATH
tion. Another application would be where it
is desired to feed a single -sideband demodu- VOLTAGE AMPLIFICATION WITH FEEDBACK A
1 -A L
lator, an f -m adaptor, or another accessory FEEDBACK
A GAIN IN ABSENCE OF
with an intermediate -frequency signal from
a communications receiver. A tube such as a E FRACTION OF OUTPUT VOLTAGE FED BACK
-
OR I[[DOACM 00 LOG I RO) 1
Il R2 J amplifier without feedback to the load im-
Ra +RAIYOL7AG[GAIMOrVi)1
10 LOG I
I. Ra 1
pedance. The reduction in noise and hum in
R RA
those stages included within the feedback
GAM OF .OTM STAGES ( GYV, ( )) [ (Goya Ro)
Re+RA loop is proportional to the reduction in gain.
GM[AG
RA -R + However, due to the reduction in gain of
R
a
the output section of the amplifier some-
Re
Grva RD what increased gain is required of the stages
Ro a[rL[CT[D LOAD iY0/10AMC[ OM Va preceding the stages included within the
Ra r[[DGACO 000i[TO0 (USUALLY AsauT 00M) feedback loop. Therefore the noise and hum
OuTPUT irnDAMC[ R. Ra output of the entire amplifier may or may
lt R a + RA (G v Ya RO)/X(1+
1 -rit- ) not be reduced dependent on the relative
O
www.americanradiohistory.com
CHAPTER SEVEN
Radio -Frequency
Vacuum -Tube Amplifiers
Tuned r -f voltage amplifiers are used in First Tuned It is obvious that the highest
receivers for the amplification of the incom- Circuit ratio of signal to noise be im-
ing r -f signal and for the amplification of pressed on the grid of the first
intermediate- frequency signals after the in- r -f amplifier tube. Attaining the optimum
coming frequency has been converted to the ratio is a complex problem since noise will
intermediate frequency by the mixer stage. be generated in the antenna due to its equiv-
Signal- frequency stages are normally called alent radiation resistance (this noise is in
tuned r-f amplifiers and intermediate-fre - addition to any noise of atmospheric origin)
quency stages are called i -f amplifiers. Both and in the first tuned circuit due to its
tuned r -f and i -f amplifiers are operated equivalent coupled resistance at resonance.
class A and normally operate at signal levels The noise voltage generated due to antenna
from a fraction of a microvolt to amplitudes radiation resistance and to equivalent tuned
as high as 10 to 50 volts at the plate of the circuit resistance is similar to that generated
last i -f stage in a receiver. in a resistor due to thermal agitation and is
expressed by the following equation:
E.2 = 4kTRof
7 -1 Grid Circuit where,
Considerations E. = rms value of noise voltage over the
interval Of,
Since the full amplification of a receiver k = Boltzman's constant (1.380 X 10 -23
follows the first tuned circuit, the operating joule per K),
conditions existing in that circuit and in its T = Absolute temperature K,
coupling to the antenna on one side and to R = Resistive component of impedance
the grid of the first amplifier stage on the across which thermal noise is devel-
other are of greater importance in determin- oped,
ing the signal -to -noise ratio of the receiver Af = Frequency band across which volt-
on weak signals. age is measured.
148 R -F Vacuum -Tube Amplifiers THE RADIO
In the above equation At is essentially the the frequency in question over conventional
frequency band passed by the intermediate - types.
frequency amplifier of the receiver under Noise Factor The limiting condition for sen-
consideration. This equation can be greatly sitivity in any receiver is the
simplified for the conditions normally en- thermal noise generated in the antenna and
countered in communications work. If we in the first tuned circuit. However, with
assume the following conditions: T = 300 proper coupling between the antenna and
K or 27 C or 80J F, room temperature; the grid of the tube, through the first tuned
Of = 8000 Hertz (the average passband of circuit, the noise contribution of the first
a communications receiver or speech ampli- tuned circuit can be made quite small. Un-
fier) , the equation reduces to: E = fortunately, though, the major noise contri-
0.0115 \TA microvolts. Accordingly, the bution in a properly designed receiver is that
thermal- agitation voltage appearing in the of the first tube. The noise contribution due
center of a half -wave antenna (assuming to electron flow and due to losses in the tube
effective temperature to be 300 K) having can be lumped into an equivalent value of
a radiation resistance of 73 ohms is approxi- resistance which, if placed in the grid circuit
mately 0.096 microvolts. Also, the thermal - of a perfect tube having the same gain but
agitation voltage appearing across a 500,- no noise would give the same noise voltage
000 -ohm grid resistor in the first stage of a output in the plate load. The equivalent
speech amplifier is approximately 8 micro- noise resistance of tubes such as the 6B6,
volts under the conditions cited above. 6DC6, etc., runs from 500 to 1000 ohms.
Further, the voltage due to thermal agitation Very high gin tubes such as the 6BZ6 and
being impressed on the grid of the first 6EH7 have equivalent noise resistances as
r -f stage in a receiver by a first tuned circuit low as 300 to 700 ohms. The lower the
whose resonant resistance is 50,000 ohms is value of equivalent noise resistance, the
approximately 2J microvolts. Suffice to say, lower will be the noise output under a fixed
however, that the value of thermal- agita- set of conditions.
tion voltage appearing across the first tuned The equivalent noise resistance of a tube
circuit when the antenna is properly coupled must not be confused with the actual input
to this circuit will be very much less than loading resistance of a tube. For highest sig-
this value. nal -to -noise ratio in an amplifier the input
It is common practice to match the im- loading resistance should be as high as pos-
pedance of the antenna transmission line to sible so that the amount of voltage that can
the input impedance of the grid of the first be developed from grid to ground by the
r -f amplifier stage in a receiver. This is the antenna energy will be as high as possible.
condition of antenna coupling which gives The equivalent noise resistance should be as
maximum gain in the receiver. However, low as possible so that the noise generated
when vhf tubes such as nuvistors and minia- by this resistance will be lower than that
tures are used at frequencies somewhat less attributable to the antenna and first tuned
than their maximum capabilities, a signifi- circuit, and the losses in the first tuned cir-
cant improvement in signal -to-noise ratio cuit should be as low as possible.
can be attained by increasing the coupling The absolute sensitivity of receivers has
between the antenna and first tuned circuit been designated in recent years in government
to a value greater than that which gives and commercial work by an arbitrary dimen-
greatest signal amplitude out of the receiver. sionless number known as "noise factor" or
In other words, in the 10 -, 6 -, and 2 -meter N. The noise factor is the ratio of noise
bands it is possible to attain somewhat im- output of a "perfect" receiver having a
proved signal- to-noise ratio by increasing given amount of gain with a dummy anten-
antenna coupling to the point where the na matched to its input, to the noise output
gain of the receiver is slightly reduced. of the receiver having the same amount of
It is always possible, in addition, to obtain gain with an injected signal, and the dummy
improved signal -to-noise ratio in a vhf re- antenna matched to its input. Although a
ceiver through the use of tubes which have perfect receiver is not a physically realizable
improved input impedance characteristics at thing, the noise factor of a receiver under
www.americanradiohistory.com
HANDBOOK Circuit Considerations 149
measurement can be determined by calcula- ages. Transit time varies inversely as the
tion from the amount of additional noise square root of the applied plate voltage.
(from a temperature -limited diode or other Cathode lead inductance is an additional
calibrated noise generator) required to in- cause of reduced input resistance at high
crease the noise -power output of a receiver frequencies. This effect has been reduced in
by a predetermined amount. certain tubes such as the 6EAS and the
6BCS by providing two cathode leads on
Tube Input As has been mentioned in a pre -
the tube base. One cathode lead should be
Loading vious paragraph, greatest gain connected to the input circuit of the tube
in a receiver is obtained when and the other lead should be connected to
the antenna is matched, through the r -f the bypass capacitor for the plate return of
coupling transformer, to the input resistance the tube.
of the r -f tube. However, the higher the The reader is referred to the Radiation
ratio of tube input resistance to equivalent Laboratory Series, Volume 23: Microwave
noise resistance of the tube the higher will
be the signal -to -noise ratio of the stage
and of course, the better will be the noise
- Receivers (McGraw -Hill, publishers) for ad-
ditional information on noise factor and in-
put loading of vacuum tubes.
factor of the over -all receiver. The input re-
sistance of a tube is very high at frequencies
in the broadcast band and gradually decreases 7 -2 Plate- Circuit
as the frequency increases. Tube input re- Considerations
sistance of conventional tube types begins to
become an important factor at frequencies of Noise is generated in a vacuum tube by
about 25 MHz and above. At frequencies the fact that the current flow within the
above about 100 MHz the use of conven- tube is not a smooth flow but rather is made
tional tube types becomes impractical up of the continuous arrival of particles
since the input resistance of the tube has (electrons) at a very high rate. This shot
become so much lower than the equivalent effect is a source of noise in the tube, but
noise resistance that it is impossible to attain its effect is referred back to the grid circuit
reasonable signal-to -noise ratio on any but of the tube since it is included in the equiv-
very strong signals. Hence, special vhf tube alent noise resistance discussed in the preced-
types such as the 6BCS, 6CW4, and 6EH7 ing paragraphs.
must be used.
Plate- Circuit For the purpose of this section,
The lowering of the effective input re-
sistance of a vacuum tube at higher fre- Coupling it will be considered that the
quencies is brought about by a number of function of the plate load cir-
factors. The first, and most obvious, is the cuit of a tuned vacuum -tube amplifier is to
fact that the dielectric loss in the internal deliver energy to the next stage with the
insulators, and in the base and press of the greatest efficiency over the required band of
tube increases with frequency. The second frequencies. Figure 1 shows three methods
factor is due to the fact that a finite time is of interstage coupling for tuned r -f voltage
required for an electron to move from the amplifiers. In figure lA omega (w) is 27r
space charge in the vicinity of the cathode, times the resonant frequency of the circuit
pass between the grid wires, and travel on in the plate of the amplifier tube, and L and
to the plate. The fact that the electrostatic Q are the inductance and Q of the inductor
effect of the grid on the moving electron L. In figure 1B the notation is the same and
acts over an appreciable portion of a Cycle M is the mutual inductance between the
at these high frequencies causes a current primary coil and the secondary coil. In fig-
flow in the grid circuit which appears to the ure 1C the notation is again the same and k
input circuit feeding the grid as a resistance. is the coefficient of coupling between the
The decrease in input resistance of a tube two tuned circuits. As the coefficient of
due to electron transit time varies as the coupling between the circuits is increased
square of the frequency. The undesirable the bandwidth becomes greater but the re-
effect of transit time can be reduced in sponse over the band becomes progressively
certain cases by the use of higher plate volt- more double- humped. The response over the
150 R -F Vacuum -Tube Amplifiers THE RADIO
band is the flattest when the Q's of pri- a large number of spurious signals in the
mary and secondary are approximately the output of the receiver -in most cases these
same and the value of each Q is equal to signals will carry the modulation of both the
1.75/k. carriers which have been cross -modulated to
Variable- Tubes
produce the spurious signal.
It
is common practice
The undesirable effect of cross- modulation
in R -F Stages to control the gain of a can be eliminated in most cases and greatly
succession of r -f or i -f reduced in the balance through the use
amplifier stages by varying the average bias of a variable -p. tube in all stages which
on their control grids. However, as the bias have avc voltage or other large negative
is raised above the operating value on a
bias applied to their grids. The variable -p.
conventional sharp -cutoff tube the tube be- tube has a characteristic which causes the
comes increasingly nonlinear in operation as
cutoff of plate current is approached. The cutoff of plate current to be gradual with
an increase in grid bias, and the reduction in
effect of such nonlinearity is to cause cross -
modulation between strong signals which plate current is accompanied by a decrease
appear on the grid of the tube. When a tube in the effective amplification factor of the
operating in such a manner is in one of the tube. Variable -p. tubes ordinarily have some-
first stages of a receiver a number of signals what reduced gm as compared to a sharp -
are appearing on its grid simultaneously and cutoff tube of the same group. Hence the
cross -modulation between them will take sharp -cutoff tube will perform best in stages
place. The result of this effect is to produce to which avc voltage is not applied.
All modern transmitters in the medium - power output. Nevertheless, the class -C am-
frequency range and an increasing percent- plifier gives less power gain than either a
age of those in the vhf and uhf ranges con- class -A or class -B amplifier under similar
sist of a comparatively low -level source of conditions since the grid of a class -C stage
radio- frequency energy which is multiplied must be driven highly positive over the por-
in frequency and successively amplified to tion of the cycle of the exciting wave when
the desired power level. Microwave trans- the plate voltage on the amplifier is low, and
mitters may be of the self -excited oscil- must be at a large negative potential over a
lator type, but when it is possible to use large portion of the cycle so that no plate
r -f amplifiers in uhf transmitters the flex- current will flow except when plate voltage
ibility of their application is increased. is very low. This, in fact, is the fundamental
The following portion of this chapter will
be devoted, however, to the method of
operation and calculation of operating char-
class -C amplifier stage can be made high
plate current is cut off at all times except
-
reason why the plate- circuit efficiency of a
acteristics of r -f power amplifiers for op- when the plate -to- cathode voltage drop
eration in the range of approximately 3.5 to across the tube is at its lowest value. Class -
500 MHz. C amplifiers almost invariably operate into
a tuned tank circuit as a load, and as a result
7 -3 Class -C R -F are used as amplifiers of a single frequency
or of a comparatively narrow band of fre-
Power Amplifiers quencies.
The majority of r-f power amplifiers fall Relationships in Figure2 shows the relation -
in the class -B and -C modes since such stages Class-C Stage shipsbetween the various
can be made to give the best plate -circuit voltages and currents over
efficiency of any present type of vacuum - one cycle of the exciting grid voltage for a
tube amplifier. Hence, the cost of tubes for class -C amplifier stage. The notation given
such a stage and the cost of the power to in figure 2 and in the discussion to follow
supply that stage is least for any given is the same as given at the first of Chapter
www.americanradiohistory.com
HANDBOOK Class -C R -F Amplifiers 151
EPM
La EBB
o
ePMIN
I
--t--L.--I--1------r
I
I
I I I
I I
I
PEAK
I 1
I I PLATE
IPMAX. 1
I I
CURRENT
If--I
-
---L-I. --1--
I IeP-4.6-e1.-01 I
I
f-
II 1 I
FUNDAMENTAL COMPONENT
OF PLATE CURRENT
AMPLIFICATION AT RESONANCE(APPROIkGMK
QP s
1
CENTER FREQUENCY
a K FI- CURRENT
MAXIMUM AMPLITUDE OCCURS AT CRITICAL COUPLING -
WNEN K
-i -I 111 I
-- t`
QP Qs
111 I
Figura I III I I
www.americanradiohistory.com
HAN DBOOK Class -C Operating Characteristics 153
0
The first determining factor in selecting
a tube or tubes for a particular application
is the amount of plate dissipation which will
0 be required of the stage. The total plate dis-
sipation rating for the tube or tubes to be
used in the stage must be equal to or greater
than that calculated from: Pp = P10 Pout. -
O 1.0
After selecting a tube or tubes to meet
the power output and plate dissipation re-
quirements it becomes necessary to determine
from the tube characteristics whether the
0 tube selected is capable of the desired opera-
tion and, if so, to determine the driving
power, grid bias, and grid dissipation.
The complete procedure necessary to de-
30
termine a set of class-C amplifier operating
-t 0 -t s -20 -30
conditions is given in the following steps:
RATIO
rg-
Fiyun 4 1. Select the plate voltage, power output,
Relationship between the ratio of the peak
value of the fundamental component of the
grid excitation voltage, and the average grid
2.
and efficiency.
Determine plate input from: P;,, -
U t /NP
'
bias; as compared to the ratio between in- 3. Determine plate dissipation from:
stantaneous peak grid
grid current
and average
Pp = Pt -
Pout. Pp must not exceed
E MAC 250TH
CONSTANT CURRENT
CHARACTERISTICS
r
5.
Ib
Determine approximate
= Pin /Ebb
ip max from: L
- egmp cos 0 - Ezl12J
ip max = 4.9 1b for Np = 0.85
for tetrodes, where p.12 is the grid -
ip max = 4.5 1b for Np = 0.80
= screen amplification factor, and E,.2 is
ip max = 4.0 lb for Np 0.75
for Np = the d -c screen voltage.
ip max = 3.5 Ib 0.70
14. Calculate the peak fundamental grid
6. Locate the point on constant- current
characteristics where the constant excitation voltage from:
plate -current line corresponding to the Egm = egmp - Ecc
appropriate ip max determined in step 5 15. Calculate the ratio Egm/E, for the
crosses the line of equal plate and grid values of E,.,, and Egm found in steps
voltages (diode line). Read ep min at 13 and 14.
this point. In a few cases the lines of 16. Read ig max/Ic from figure 4 for the
constant plate current will inflect ratio Egm /ECC found in step 15.
sharply upward before reaching the 17. Calculate the average grid current
diode line. In these cases ep min should from the ratio found in step 16, and
not be read at the diode line but at the value of ig max found in step 11:
the point where the plate current line
ig max
intersects a line drawn from the ori-
gin through these points of inflection. Ratio from step 16
7. Calculate Epm from: 18. Calculate approximate grid driving
Epm = Ebb - ep min power from:
8. Calculate the ratio Ipm /li, from: Pa = 0.9 Egmlc
19. Calculate grid dissipation from:
fpm 2 Np Ebb
Ih Epm
Pg = Pa + Eccic
Pg must not exceed the maximum
9. From the ratio of /p./lb calculated in rated grid dissipation for the tube se-
step 8 determine the ratio ip mx /li,
lected.
from figure 3. Sample A typical example of class -C
10. Calculate a new value for ip max from
Calculation amplifier calculation is shown in
the ratio found in step 9.
the example below. Reference is
ip max = (ratio from step 9) 1,,
made to figures 3, 4 and 5 in the calcula-
11. Read egmp and ig max from the constant -
tion.
current characteristics for the values
1. Desired power output -800 watts.
of ep min and ip max determined in steps
2. Desired plate voltage-3500 volts.
6 and 10.
Desired plate efficiency -80 percent
12. Calculate the cosine of one -half the
angle of plate -current flow from:
(Np = 0.80)
Pin= 800/0.80 = 1000 watts
3. Pp = 1000 - 800 = 200 watts
cos Op = 2.32 j" - 1.57
J 4.
Use 250TH; max Pp = 250w; = 37.
Ib = 1000/3500 = 0.285 ampere
13. Calculate the grid bias voltage from:
(285 ma) Max. lb for 250TH is 350
ma.
1 5. Approximate ip max = 0.285 X 4.5
1 - cos Bp
=1.28 ampere
6. ep min =
260 volts (see figure 5 first
trial point)
cos Bp
egmp
Ebb 1
7. Epm =
3500 260 -
3240 volts =
8. Ipm /Ib =
2 X 0.80 X 3500/3240 =
for triodes. 5600/3240 1.73 =
www.americanradiohistory.com
HANDBOOK Class -C Operating Characteristics 155
9. iv max /In =
4.1 (from figure 3) ing compromise values of Q for class -C am-
10. 1p max =
0.285 X 4.1 1.17 = plifiers are given in the chapter, Generation
11. egmp =
240 volts of R -F Energy. However, the amount of in-
ig max =
0.430 amperes ductance required for a special tank -circuit
(Both above from final point on Q under specified operating conditions can
figure 5 ) be calculated from the following expression:
12. cos Bp = 2.32 (1.73 1.57) = 0.37 -
,
(Op = 68.3 ) ,L =
Q
1
13. Ecc=
1 - 0.37 where,
0, equals 2 7r X operating frequency,
[ ( L equals tank inductance,
0.37 3370 -240 I - 33OJ RL equals required tube load impedance,
Q equals effective tank circuit Q.
\\
= -240 volts A tank circuit Q of 12 to 20 is recom-
14. Eg,,, = 240 -
( -240) = 480 volts mended for all normal conditions. However,
if a balanced push -pull amplifier is employed
grid swing
15. Egm /Ecc = 480/ 240 = - -2 the tank receives two impulses per cycle and
16. ig O,ax /Ic = 5.75 (from figure 4) the circuit Q may be lowered somewhat
m,
17. le = 0.430/5.75 = 0.075 amp (75 from the above values.
ma) grid current Quick Method of The plate- circuit effi-
18. Pa = 0.9 X 480 X 0.075 = 32.5 Calculating Amplifierciency of a class -B or
watts driving power
19. Pg = 32.5 -(
-240 X 0.75) =
Plate Efficiency class -C r -f amplifier
can be determined
14.5 watts grid dissipation
from the following facts. The plate- circuit
Max Pg for 250TH is 40 watts
efficiency of such an amplifier is equal to
The power output of any type of r -f am-
the product of two factors: F1, which is
NWI
plifier is equal to: equal to the ratio of Epm to Ebb (F1 =
IpmEpm /2 =P
.,-'i.......
N,
Ip, can be determined, of course, from
the ratio determined in step 8 above (in this o.
type of calculation) by multiplying this
ratio times In.
It is frequently of importance to know
the value of load impedance into which a
class -C amplifier operating under a certain
set of conditions should operate. This is
Fz
ON
o.w
o.
I ku 0E.. ,...llllll1l,,llllll1..
simply RL = /I pm. In the case of the
operating conditions just determined for a
250TH amplifier stage the value of load
impedance is:
,-.lllll,.
082
a7s
0.711 1El
lllll'.
0.74
0.72
Ip11I = II111
In 0.70 NI
0 10 20 30 40 50 N 70 90 90 100 1)0 1 o
n I I o-... ri cercles, nraorre 1 I I
Er,,,
I
I
O.W
of Amplifier Figure 6
Q In order to obtain proper plate
Tank Circuit tank -circuit tuning and low Relationship between factor F, and the half -
radiation of harmonics from angle of plate -carrent Row in an amplifier
with sine-wave Input and output voltage,
an amplifier it is necessary that the plate operating at a grid -bias voltage greater than
tank circuit have the correct Q. Charts giv- cutoff
156 R -F Vacuum -Tube Amplifiers THE RA D I O
www.americanradiohistory.com
HANDBOOK
1.0
.7
.s
.3
.2
IMO
NNE
L'c
WWI=
e
,
100
_
200
=cz,E=+ol
300
'
-,
=+100 l
rr-+
400 500
--
e00 700 600 900
Class-B R -F Amplifiers
Ecz= +400
Eca=0v.
1000 1100
V.
1200 1300
Eu
Eu=+60
1400
=
Eco. +4e
Eci
1
M1
r eo
lem 171
157
PLATE VOLTS
Figure 7
and plate efficiency will drop to approxi- 21 milliamperes willproduce this fig-
mately one -half the values at the 100 per- ure. Referring to figure 7, a grid bias
cent positive modulation peak and the power of -45 volts is approximately cor-
output of the stage will drop to one - fourth rect.
the peak -modulation value. On the negative 2. A practical class -B linear r -f amplifier
modulation peak the input, efficiency, and runs at an efficiency of about 66% at
output all drop to zero. full ouput, the carrier efficiency drop-
In general, the proper plate voltage, bias ping to about 33% with a modulated
voltage, load resistance, and power output exciting signal. In the case of single -
listed in the tube tables for class -B audio sideband suppressed -carrier excitation,
work will also apply to class -B linear r -f ap- the linear amplifier runs at the resting
plication. or quiescent input of 42 watts with
no exciting signal. The peak allow-
Calculation of Oper- Figure 7 illustrates able power input to the 813 is:
ating Parameters for o the characteristic
Clau -B Linear Amplifier curves for an 813 Input peak power (WO =
tube. Assume the plate dissipation X 100
plate supply to be 2000 volts, and the screen
supply to be 400 volts. To determine the
(100 -%
plate efficiency)
operating parameters of this tube as a class- -125
33
X 100 = 379 watts
B linear SSB r -f amplifier, the following
steps should be taken: 3. The maximum d -c signal plate current
is:
1. The grid bias is chosen so that the
resting plate current will produce ap-
proximately 1/3 of the maximum 1p max - Wp
Ep 2000
379
s
4
so
Ecu=+ioov.
Ec2=ov.
C
-inductance
106
6.28 X 4.0 X 445 - 90 pf
140
11. The required to resonate at
c1=4110
LEcu 440 V.
4.0 MHz with this value of capacity
3
+ zo 06
'Illt
u
E=+4ov is:
445
Eu +xo L = 6.28 X 4.0
0 00 200 300 400
PLATE VOLTS En
= 17.8 microhenries
Figure 8
Grid -Circuit The maximum positive
1.
E,, VERSUS E, CHARACTERISTICS OF 813 Considerations grid potential is 60
TUBE volts, and the peak r-f
grid voltage is 105 volts. Required
ip ,,,ax =3.14 X 0.189 = 0.595 ampere driving power is 0.525 watt. The
5. Referring to figure 7, a current of equivalent grid resistance of this stage
about 0.595 ampere (Point A) will is:
flow at a positive grid potential of 60 (eg)2 1052
volts and a minimum plate potential Rg = 2 X Pg 2 X 0.525
of 420 volts. The grid is biased at
-45 volts, so a peak r -f grid voltage 10,000 ohms
of 60 + 45 volts, or 105 volts, swing
2. As in the case of the class -B audio am-
is required.
plifier the grid resistance of the linear
6. The grid driving power required for
amplifier varies from infinity to a low
the class -B linear stage may be found
value when maximum grid current is
by the aid of figure 8. It is one-third
the product of the peak grid current drawn. To decrease the effect of this
times the peak grid swing. resistance excursion, a swamping resis-
tor should be placed across the grid -
0.015 X 105 tank circuit. The value of the resistor
Pp = 3
= 0.525 watt
should be dropped until a shortage of
7. The single-tone power output of the driving power begins to be noticed.
813 stage is: For this example, a resistor of 3000
Pp = 78.5 (E1, -ep a) X Ip ohms is used. The grid circuit load for
Pp = 78.5 (2000 -420) X .189 = no grid current is now 3000 ohms in-
235 watts stead of infinity, and drops to 2300
8. The plate load resistance is: ohms when maximum grid current is
Ep - ep mio1580
3.
drawn.
A circuit Q of 15 is chosen for the
R''- O.5tpmax 0.5 X .595 grid tank. The capacitive reactance
= 5320 ohms required is:
9. If a loaded plate tank circuit Q of 12
is desired, the reactance of the plate
Xo = 2100 = 154 ohms
tank capacitor at the resonant fre-
quency should be:
4. At 4.0 MHz the effective capacitance
R,,
reactance (ohms) = is:
Q
108
5320 C
6.28 X 4.0 X 154
= 259 pf
12
= 445 ohms 5. The inductive reactance required to
www.americanradiohistory.com
HANDBOOK Special R-F Amplifiers 159
resonate the grid circuit at 4.0 MHz cathode leads of the tube and in the leads to
is: the neutralizing capacitor. In other words
the bandwidth of neutralization decreases as
= 154 = 6.1 microhenries the presence of the neutralizing capacitor
6.28 X 4.0 adds additional undesirable capacitive load-
6. By substituting the loaded -grid resist- ing to the grid and plate tank circuits of
the tube or tubes. To look at the problem in
ance figure in the formula in the first
paragraph, the grid driving power is another way, an amplifier that may be per-
now found to be approximately 2.4
fectly neutralized at a frequency of 30 MHz
watts. may be completely out of neutralization at a
frequency of 120 MHz. Therefore, if there
Screen -Circuit By reference to the plate are circuits in both the grid and plate cir-
Considerations characteristic curve of the cuits which offer appreciable impedance at
813 tube, it can be seen that
this high frequency it is quite possible that
at a minimum plate potential of 420 volts, the stage may develop a parasitic oscilla-
and a maximum plate current of 0.6 ampere, tion in the vicinity of 120 MHz.
the screen current will be approximately 30 Grounded -Grid This condition of restricted -
milliamperes, dropping to one or two milli- R-F Amplifiers range neutralization of r -f
amperes in the quiescent state. It is necessary power amplifiers can be great-
to use a well - regulated screen supply to hold ly alleviated through the use of a cathode -
the screen voltage at the correct potential driven or grounded -grid r -f stage. The
over this range of current excursion. The use grounded -grid amplifier has the following
of an electronically regulated screen supply advantages:
is recommended.
1. The output capacitance of a stage is
reduced to approximately one -half the
7-5 Special R -F Power value which would be obtained if the
Amplifier Circuits same tube or tubes were operated as a
conventional neutralized amplifier.
The r -f power amplifier discussions of 2. The tendency toward parasitic oscilla-
Sections 7 -3 and 7 -4 have been based on the tions in such a stage is greatly reduced
assumption that a conventional grounded - since the shielding effect of the control
cathode or cathode- return type of amplifier grid between the filament and the
was in question. It is possible, however, as in plate is effective over a broad range of
the case of a -f and low -level r -f amplifiers frequencies.
3. The feedback capacitance within the
to use circuits in which electrodes other than
the cathode are returned to ground insofar stage is the plate -to- cathode capaci-
as the signal potential is concerned. Both the
tance which is ordinarily very much
less than the grid -to -plate capacitance.
plate- return or cathode- follower amplifier
and the grid- return or grounded -grid am- Hence neutralization is ordinarily not
plifier are effective in certain circuit applica- required. If neutralization is required
tions as tuned r -f power amplifiers. the neutralizing capacitors are very
small in value and are cross -connected
Disadvantages of An undesirable aspect of between plates and cathodes in a push -
Grounded -Cathode the operation of cathode - pull stage, or between the opposite end
Amplifiers return r -f power ampli- of a split plate tank and the cathode
fiers using triode tubes is in a single -ended stage.
that such amplifiers must be neutralized. The disadvantages of a grounded -grid am-
Principles and methods of neutralizing r -f plifier are:
power amplifiers are discussed in the chapter
Generation of R -F Energy. As the frequency 1. A large amount of excitation energy
of operation of an amplifier is increased the is required. However, only the normal
stage becomes more and more difficult to amount of energy is lost in the grid
neutralize due to inductance in the grid and circuit of the amplifier tube; all addi-
160 R -F Vacuum -Tube Amplifiers THE RADIO
tional energy over this amount is de-
livered to the load circuit as useful
output. LOAD
former must be of the low capacitance PoWER OUTPUT TO LOOM (E'Elw) Ew a, EwElw EGr [...
type and adequately insulated for the
POWER DELIVERED ST OUTPUT TY4E 1112Pr
r -f voltage which will be present.
rPO.. DRIVER Ew IPu
3. A grounded -grid r -f amplifier cannot
POWER TO LOAD
,
be plate modulated 100 percent unless TOTAL POWER DEIIVERED 4, OR VCR
EGr (=M+ IVA)
the output of the exciting stage is .olE4ric
modulated also. Approximately 70 per- POWER RuoRUO RT OUTPUT TUDE GRID .n0 SIPS /uPPL
cent modulation of the exciter stage, EON 14r o.l EGr Ic
while the final stage is modulated oR
www.americanradiohistory.com
HANDBOOK Grounded -Grid Amplifiers 161
-0.51 X
Plate- Return or
Cathode- Follower R -F
Circuit diagram, elec-
trode potentials and
Power Amplifier currents, and operat-
0.51 (2560_140 - 20 ing conditions for a
12
- volts
= 385
cathode- follower r -f power amplifier are
given in figure 10. This circuit can be used,
14. Egm = 140- -385) = ( volts 525 in addition to the grounded -grid circuit just
15. Egm /Ecc = - 1.36
/Ic = approx. 8.25 (extrapolated
IPM
16. ig max
from figure 4)
17. le = 0.480/8.25 = 0.058 (58 ma d -c
grid current)
18. Pd = 0.9 X 525 X 0.058 = 27.5
watts
19. Pg= 27.5 -( -385 X 0.058) = 5.2
watts
Max. Pg for 304TL is 50 watts
( PM+ IGM)
We can check the operating plate efficien- POwes Output TO LOAD -
Epa.
2
Figure 10
Ipn, = 1.79 X 0.370 = 0.660 amperes
(from 4 and 8 above) CATHODE -FOLLOWER R -F POWER
AMPLIFIER
The total average power required of the
driver (from figure 9) is equal to Exmlpm /2 Showing the relationships between the tube
potentials and e is and the input and out-
(since the grid is grounded and the grid put power of the stage. The approximate grid
swing appears also as cathode swing) plus impedance also is given.
www.americanradiohistory.com
162 R -F Vacuum -Tube Amplifiers THE RADIO
2504,4- 400A,ETC
discussed, as an r -f amplifier with a triode 4-
www.americanradiohistory.com
N
HANDBOOK Class -A131 R -F Power Amplifiers 163
M
M1-
7 -6 Class -AIL Radio-
WE
MENEN.i,
Frequency Power Amplifiers M MOST LINEAN
:PORTION
I,GG!=!!GG!!lGGG
ME
IR//NENN
13
EWEE
Class-AB' r -f amplifiers operate under
611
such conditions of bias and excitation that !!lGI:.S!=IG:GCIGGG!=GGG
grid current does not flow over any portion 1111111-41111111111111111111111111111113
NEMNaq
of the input cycle. This is desirable, since
distortion caused by grid -current loading is [fBZRiE.J
absent, and also because the stage is capable
of high power gain. Stage efficiency is about
58 percent when a plate current operating
angle of 210 is chosen, as compared to 62
Figure 12
percent for class -B operation.
The level of static (quiescent) plate cur- Er -1 CURVE
rent for lowest distortion is quite critical for
Amplifier operation is confined to the mod
class -AB, tetrode operation. This value is linear portion of the characteristic curve.
determined by the tube characteristics, and
is not greatly affected by the circuit para-
meters or operating voltages. The maximum is usually of exponential shape, and the sig-
d -c plate potential is therefore limited by the nal distortion is held to a small value by
static dissipation of the tube, since the rest- operating the tube well below its maximum
ing plate current figure is fixed. The static output, and centering operation over the
plate current of a tetrode tube varies as the most linear portion of the characteristic
3/2 power of the screen voltage. For ex- curve.
ample, raising the screen voltage from 300 The relationship between exciting voltage
to 500 volts will double the plate current. in a class -AB, amplifier and the r -f plate
The optimum static plate current for mini- circuit voltage is shown in figure 13. With a
mum distortion is also doubled, since the small value of static plate current the lower
shape of the Eg -1p curve does not change. portion of the line is curved. Maximum un-
In actual practice, somewhat lower static distorted output is limited by the point on
plate current than optimum may be em- the line (A) where the instantaneous plate
ployed without raising the distortion appre- voltage drops down to the screen voltage.
ciably, and values of static plate current of This "hook" in the line is caused by current
0.6 to 0.8 of optimum may be safely used, diverted from the plate to the grid and screen
depending on the amount of nonlinearity elements of the tube. The characteristic plot
that can be tolerated. of the usual linear amplifier takes the shape
As with the class -B linear stage, the mini- of an S -curve. The lower portion of the
mum plate voltage swing of the class -AB, curve is straightened out by using the proper
amplifier must be kept above the d -c screen value of static plate current, and the upper
potential to prevent operation in the non- portion of the curve is avoided by limiting
linear portion of the characteristic curve. minimum plate voltage swing to a point
A low value of screen voltage allows greater substantially above the value of the screen
r -f plate voltage swing, resulting in im- voltage.
provement in plate efficiency of the tube. A
balance between plate dissipation, plate effi- Operating Parameters The approximate oper-
ciency, and plate -voltage swing must be for the Class-AB, ating parameters may
achieved for best linearity of the amplifier. Linear Amplifier be obtained from the
constant - current
The S -Curve The perfect linear amplifier de- curves (Eg -E) or the Eg Ip curves of the
livers a signal that is a replica tube in question. An operating load line is
of the input signal. Inspection of the plate - first approximated. One end of the load line
characteristic curve of a typical tube will is determined by the d -c operating voltage
disclose the tube linearity under class -AB, of the tube, and the required static plate
operating conditions (figure 12) The curve
. current. As a starting point, let the product
164 R -F Vacuum-Tube Amplifiers THE RADIO
!0{
2sPOINT B
1000 40
...... 2000
E
I
3000
0
V
W
2! `_ -
.k
.4
M1115\
-
!O
LOAO LINE
.2
POINT
A
I0 (
o
I I
123 li
VALUE OF VALUE OF
EP MIN. SOOV.. MAX.
ISO ilO.SeA DISSIPATION
0000 Y. X 0.12 A .340 N'4r77)
173
1001 I 1 1
Figure 14
www.americanradiohistory.com
HANDBOOK Grounded -Grid Amplifiers 165
4
(3000-600) X .58
348 watts
or,
' 4
RFC
The plate current -angle efficiency factor C20 PF RFC vv
for this class of operation is 0.73, and the PER
METER
88W
NO 800
.001
aV
actual plate-circuit efficiency is: 8+
2000V
EP-ep min X0.73
Emax 01 .01
(3000 - 600)
X 0.73 = 58.4% 8-
3000
The power input to the stage is therefore
Figure 15
348
P X 100 or, 595 watts SIMPLE GROUNDED -GRID
Np 5 8.4
LINEAR AMPLIFIER
The plate dissipation is: Tuned cathode (L-C) is required to pr
distortion of driving -signal wave form
595 -348 = 247 watts.
It can be seen that the limiting factor for voltage, or cause excessive screen dissipation.
this class of operation is the static plate dis- Good regulation of the screen supply is there-
sipation, which is quite a bit higher than the fore required. Screen dissipation is highly re-
operating dissipation level. It is possible, at sponsive to plate loading conditions and the
the expense of a higher level of distortion, to plate circuit should always be adjusted so as
drop the static plate dissipation and to in- to keep the screen current below the maxi-
crease the screen voltage to obtain greater mum dissipation level as established by the
power output. If the screen voltage is set at applied voltage.
800, and the bias increased sufficiently to
drop the static plate current to 90 ma, the 7 -7 Grounded -Grid
single -tone d -c plate current may rise to 300
ma, for a power input of 900 watts. The Linear Amplifiers
plate circuit efficiency is 55.6 percent, and
the power output is 500 watts. Static plate A typical grounded -grid amplifier is shown
dissipation is 270 watts. in figure 15. The driving signal is applied
At a screen potential of 500 volts, the between the grid and the cathode, with the
maximum screen current is less than 1 ma, grid held at r -f ground potential. The con-
and under certain loading conditions may be trol grid serves as a shield between the
negative. When the screen potential is raised cathode and the plate, thus making neutral-
to 800 volts maximum screen current is 18 ization unnecessary at medium and high
ma. The performance of the tube depends on frequencies. High -p. triodes and triode -
the voltage fields set up in the tube by the connected tetrodes may be used in this con-
cathode, control grid, screen grid, and plate. figuration. Care must be taken to monitor
The quantity of current flowing in the the #1 -grid current of the tetrode tubes as
screen circuit is only incidental to the fact it may run abnormally high in some types
that the screen is maintained at a positive (4X150A family) and damage to the tube
potential with respect to the electron stream may possibly result unless a protective cir-
surrounding it. cuit of the form shown in figure 11 is used.
The tube will perform as expected as long "Zero- bias" triodes (811 -A, 3-400Z and
as the screen current, in either direction, does 3- 1000Z) and certain triode-connected te-
not create undesirable changes in the screen trodes (813 and 4 -400A, for example) re-
166 R -F Vacuum -Tube Amplifiers THE RADIO
www.americanradiohistory.com
HANDBOOK Grounded -Grid Linears 167
W4
Figure 17
Waveform distortion caused by half -cycle loading at cathode of grounded -grid amplifier may be
observed (right whereas undistorted waveform is observed with tuned cathode circuit (left).
Two -tone tests at 2.0 MHz proved the necessity of using a cathode tank circuit for lowest inter -
modulation distortion.
A B
INPUT
01
(-INPUT
01
ti x V X X 'L X
Figure 18
Tuned cathode network for cathode -driven circuit may take form of bifilar coil (A), pi- network (B),
or shunt LC circuit (C). Circuit Q of at least 2 is recommended. Capacitor C, may be a 3 -gang
broadcast -type unit. Coils L,, L,, or L. are adjusted to resonate to the operating frequency with
C, set to approximately 13 pf -per meter wavelength. Capacitor C. is approximately 1.5 times the
value of C,. The input taps on coils L. and L, or the capacitance of C. are adjusted for minimum
SWR on coaxial line to the exciter.
168 R -F Vacuum -Tube Amplifiers
RANDOM LENGTH
SSB EXCITER -e" COAXIAL LINE
G G LINEAR AMPLIFIER
OUTPUT
Figure 19
Untuned cathode circuit of grounded-grid amplifier offers high- impedance path to the r -f current
flowing between plate and cathode of the amplifier tube. The alternative path Is via the inter-
connecting coaxial line and tank circuit of the exciter. Waveform distortion of the driving signal
and high intermodulation distortion may result from use of alternative input circuit.
ties as a result of the harmonic content of a complex signal having more than one
the wave. Use of a tuned cathode circuit in frequency. A voice signal (made up of a
the grounded -grid stage will preserve the multiplicity of tones) will become blurred
waveform as shown in the photographs. The or distorted by IMD when amplified by a
tuned cathode circuit need have only a Q of nonlinear device. As practical linear ampli-
2 or more to do the job, and should be fiers have some degree of IMD (depending
resonated to the operating frequency of the on design and operating parameters) this
amplifier. Various versions of cathode tank disagreeable form of distortion exists to a
circuits are shown in figure 18. greater or lesser extent on most SSB signals.
In addition to reduction of waveform A standard test to determine the degree
distortion, the tuned cathode circuit pro- of IMD is the two -tone test, wherein two
vides a short r -f return path for plate cur- radio- frequency signals of equal amplitude
rent pulses from plate to cathode (figure are applied to the linear equipment, and the
19). When the tuned circuit is not used, resulting output signal is examined for spuri-
the r -f return path is via the outer shield of ous signals, or unwanted products. These
the coaxial line, through the output capaci- unwanted signals fall in the fundamental -
tor of the exciter plate -tank circuit and signal region and in the various harmonic
back to the cathode of the linear amplifier regions of the amplifier. Signals falling out-
tube via the center conductor of the coaxial side the fundamental- frequency region are
line. This random, uncontrolled path varies termed even -order products, and may be at-
with the length of interconnecting coaxial tenuated by high -Q tuned circuits in the
line, and permits the outer shield of the line amplifier. The spurious products falling close
to be "hot" compared to r -f ground. to the fundamental -frequency region are
termed odd -order products. These unwanted
7 -8 Intermodulation products cannot be removed from the
wanted signal by tuned circuits and show
Distortion up on the signal as "splatter," which can
If the output signal of a linear amplifier cause severe interference to communication
is an exact replica of the exciting signal there in an adjacent channel. Nonlinear operation
will be no distortion of the original signal of a so- called "linear" amplifier will generate
and no distortion products will be generated these unwanted products. Amateur practice
in the amplifier. Amplitude distortion of the calls for suppression of these spurious prod-
signal exists when the output signal is not ucts to better than 30 decibels below peak
strictly proportional to the driving signal power level of one tone of a two -tone test
and such a change in magnitude may result signal. Commercial practice demands sup-
in intermodulation distortion (IMD). IMD pression to be better than 40 decibels below
occurs in any nonlinear device driven by this peak level.
www.americanradiohistory.com
CHAPTER EIGHT
The Oscilloscope
The cathode-ray oscilloscope (also called which can be converted into electrical po-
oscillograph) is an instrument which permits tentials. This conversion is made possible by
visual examination of various electrical phe- the use of some type of transducer, such as a
nomena of interest to the electronic engineer. vibration pickup unit, pressure pickup unit,
Instantaneous changes in voltage, current photoelectric cell, microphone, or a variable
and phase are observable if they take place impedance. The use of such a transducer
slowly enough for the eye to follow, or if makes the oscilloscope a valuable tool in
they are periodic for a long enough time so fields other than electronics.
that the eye can obtain an impression from
the screen of the cathode -ray tube. In addi- 8 -1 A Modern Oscilloscope
tion, the cathode -ray oscilloscope may be
used to study any variable (within the limits For the purpose of analysis, the operation
of its frequency- response characteristic) of a modern oscilloscope will be described.
N/-LO GAIN
VIDEO
Y -AXIS INPUT 0 CATHODE
SIGNAL O-- ATTENU- FOLLOWER -
INPUT I ATORI
VIDEO AMP.
_ye- V2 VERTICAL
AMPLIFIER -Lo CATHODE
RAY TUBE
-
VERTICAL
VI GAIN V3 V4
1rVS
INPUT
TRIGGER BLANKING
o AMPLIFIER
--
AMPLIFIER
Vet VT VIS
4
POWER
SUPPLIES
Figure 1
INPUT
- -i
ATTENUATOP
i
TO
DEFLECTION
PLATES
=001
. 7 n
0.1
VERTICAL
GAIN
300
- BIAS +SCREEN
REG.
Figure 2
VERTICAL AMPLIFIER
The vertical amplifier is capable of passing sine waves from 10 Hz to 10 MHz. Compensated
input attenuator and peaking circuits provide gain that is essentially independent of
frequency. Deflection amplifier serves as phase inverter to provide push -pull signal to
deflection plates of cathode-ray tube. Deflection polarity switch permits greater upward or
downward deflection of pattern to accommodate reversed polarity of input wave.
The simplified block diagram of the instru- preamplifier depending on the amount of
ment is shown in figure 1. This oscilloscope amplification needed. The preamplifier is de-
is capable of reproducing sine waves from signed to pass the wide frequency band
10 Hz to 10 MHz and pulses as short as 0.1 desired by the use of peaking coils in the
microsecond may be observed. The sweep plate circuit, which enhance the high -fre-
speed is continuously variable, and the elec- quency response, in addition to large value
tron beam of the cathode -ray tube can be coupling capacitors which ensure good low -
moved vertically or horizontally, or the frequency response (see chapter 6, section 6
movements may be combined to produce Video Frequency Amplifiers). The signal
composite patterns on the screen. As shown then passes through a cathode -follower stage
in the diagram, the cathode-ray tube re- (V2) to the vertical amplifier. The cathode
ceives signals from two sources: the vertical follower serves as an impedance transformer
(Y -axis) the sweep (X -axis) amplifiers, and so that a low- impedance vertical gain con-
also receives blanking pulses that remove un- trol may be used. It is necessary that the po-
wanted return -trace signals from the screen. tentiometer have a low value so that stray
The operation of the cathode -ray tube has capacitances do not appreciably affect the
been covered in an earlier chapter and the frequency response as the control is rotated.
auxiliary circuits pertaining to signal pre- The original deflection polarity of the signal
sentation will be discussed here. is reversed when two stages of amplification
are used, resulting in a downward deflection
The Vertical The incoming signal to be dis- of the oscilloscope pattern for positive input
Amplifier played is applied to the vertical polarity. A deflection polarity switch is used
amplifier (figure 2) . An input to change the operating bias and screen volt-
attenuator (compensated to provide attenu- age on the cathode- coupled push -pull vertical
ation that is essentially independent of signal amplifier tubes (V3, V4) permitting greater
frequency) permits the gain of the amplifier undistorted upward or downward deflection.
to be adjusted in calibrated steps. The signal The amplified signal is coupled from the
is then amplified by the wideband (video) plate circuit of the vertical amplifier through
preamplifier (V1), or is shunted around the a peaking circuit that affords optimum
www.americanradiohistory.com
HANDBOOK A Modern Oscilloscope 171
transient response rather than best frequency in synchronism with the trigger impulse. A
response, which has been previously deter- trigger amplifier (V6, V7) enhances the
mined in the preamplifier stages. trigger pulse and selects the proper polarity
of the pulse. To convert the various shapes
The Time -Base Investigation of electrical of trigger impulses into square waves of
Circuitry waveforms by the use of a controllable duration suitable for operating
cathode-ray tube requires the sweep generator and unblanking the
that some means be readily available to de- cathode -ray tube, a flip-flop multivibrator
termine the variation in these waveforms type of pulse generator is used (figure 3) .
with respect to time. An X -axis time base on The frequency of pulse generation of the
the screen of the cathode -ray tube shows the multivibrator is controlled by the external
variation in amplitude of the input signal negative trigger signal. The multivibrator
with respect to time. This display is made consists of two tubes (V0, V9) with one
possible by a time -base generator (sweep gen- tube in a conducting state and the other
erator) which moves the spot across the nonconducting. When a trigger impulse is
screen at a constant rate from left to right received, the negative pulse lowers the plate
between selected points, returns the spot al- potential of the nonconducting tube (V8)
most instantaneously to its original position, and also decreases the grid bias of V9 via
and repeats this procedure at a specified rate the switchable coupling capacitor (sweep -
(referred to as the sweep frequency). speed control). The first tube conducts and
The Sweep -Trigger Circuit -An external the second tube is driven toward cutoff by
synchronizing impulse (which may be the the buildup of voltage in the coupling ca-
presented signal) initiates the horizontal pacitor between the two tubes. This condi-
sweep circuits of the oscilloscope, deflecting tion is maintained until the switchable sweep -
the beam of the cathode -ray tube across the speed capacitor is discharged, thus raising the
screen at uniform rate, starting each sweep grid voltage of V9 to such a point that the
tube starts to conduct. This lowers the plate
potential of V,,, carrying with it the direct -
coupled grid of Vs and starting a regenera-
tive cycle which ends with V9 conducting
NEGATIVE and Vs cut off -the condition which existed
PULSE FROM
TRIGGER before the trigger pulse occurred. Thus the
AMPL IF /ER
plate of Vs produces a square negative pulse
and simultaneously the plate of V9 produces
a square positive pulse. The negative pulse is
used to control the operation of the sweep
generator and the unblanking circuit of the
cathode -ray tube. The positive pulse may be
used to furnish gate voltage available at the
panel of the instrument to trigger auxiliary
POSITIVE
UNLLANK-
circuits.
ING PULSE
TOCRT The Blanking Circuit -During the wait -
B+
period between trigger pulses, the bias on the
SWEEP SPEED
cathode -ray tube is such that the tube is
NEGATIVE TRIGGER PULSE
completely cut off. As soon as a trigger ap-
TO SWEEP GENERATOR pears and the sweep starts, it is necessary to
Figure 3 provide a positive pulse on the grid of the
TRIGGER AND UNBLANKING CIRCUIT
cathode -ray tube and thus turn on the elec-
tron beam. This pulse must have extremely
Flip -flop multivibrator (V,, V) is triggered rapid rise time and a very flat top so that the
externally and generates negative trigger
pulse to start sweep generator. Impulse rate brightness of the image is uniform. To se-
is controlled by switchable sweep speed ca- cure a pulse of this nature, the negative
pacitor bank. Positive trigger pubis unblanks pulse from the multivibrator is passed
cathode -ray tube by reducing cutoff bias on
the grid of the cathode toy tube through a cathode- follower blanking ampli-
172 The Osc i loscope l THE RADIO
B+ REG
270 K
the beam returns to the left of the screen to
V1 wait another trigger signal. It is this vari-
VOL TAGE
NEGATIVE
REGULATOR SWEEP SPEED able waiting period which makes the sweep
TRIGGER ADJUSTMENT
PULSE FROM time independent of the signal period, per-
MOLT/VIBRATOR
SWEEP SPEED
820 mitting the oscilloscope to view pulses and
PAUL T /PL /ER
B+ REG. TO CLAMP CIRCUITS other short duration signals where the length
N-I ANO SWEEP AMPL /F /ER of the pulse is very short compared to the
10K
i b n space between the pulses.
+SCR
SWEEP SPEED
TIMING CAPACITORS Some inexpensive oscilloscopes employ a
W 7 120K recurrent or sawtooth sweep such as that
which is generated by a gas tube or other
00 K VII
SWEEP GENERATOR similar device that synchronizes the sweep
-BIAS with the input signal. The sweep time is
thus equal to, or a multiple of, the signal
Figure 4
period. The circuit of figure 4 may be modi-
SWEEP GENERATOR fied to produce a sawtooth sweep by the
Each swoop of this triggered swoop circuit is omission of the trigger signal and adjustment
started independently of the preceding sweep of the multivibrator frequency to synchro-
by a trigger pulse received from the multi - nize with the period of the observed signal.
vibrator circuit. Sweep -speed timing capacitors
are ganged with sweep-speed circuit of mul- The sweep voltage necessary to produce the
tivibrator. Timing voltage is derived from sawtooth sweep is shown in figure S. The
voltage regulator to ensure sweep accuracy.
sweep occurs as the voltage varies from A to
B, and the return trace as the voltage varies
fier (V15) which provides a pulse of proper from B to C. At high sweep frequencies, the
amplitude and rise time. return trace is an appreciable portion of the
sweep time.
The Sweep The voltage necessary to obtain Operation of the Sweep Generator-The
Generator a linear time base may be gen- sweep generator (V11, figure 4) is held in
erated by the circuit of figure 4. a conducting state by the positive grid bias
In this representative triggered sweep cir- derived from the voltage divider in the grid
cuit each sweep is started independently of circuit. The plate voltage of the sweep gen-
the preceding sweep by a trigger, or syn- erator is low, and the switchable sweep -speed
chronizing, pulse received from the mul- timing capacitor is essentially uncharged.
tivibrator circuit. When no trigger is The negative trigger pulse from the multi -
received the cathode -ray tube potentials vibrator rapidly cuts off V11 allowing the
position the beam at the left end of the timing capacitor to charge exponentially
horizontal trace. When the trigger signal through the 1- megohm sweep -speed multi-
arrives, the beam goes linearly to the right in plier control, approaching the voltage at the
a time interval determined by the length of
cathode of regulator tube V1o. This voltage
the trigger pulse. At the end of each sweep, is adjusted by the sweep -speed control in the
grid circuit of the regulator tube. The tim-
ing capacitor is charged from a constant
voltage supply having a low impedance to
ensure sweep -speed accuracy. Sweep linearity
is enhanced by using only 10 percent or less
of the charging voltage. The linear sweep
voltage is taken from the plate of the sweep
Figure 5 generator, clamped and impressed on the
following sweep amplifier.
SAWTOOTH WAVEFORM When the multivibrator trigger pulse
Recurrent or sawtooth sweep waveform is ends, the grid of the sweep generator tube
used in inexpensive oscilloscopes. Sawtooth returns to a positive potential and the heavy
may be generated by gas swoop tube, such as
the 884, and is usually synchronized with plate current reduces the plate voltage of
input signal. V,1 to near zero, discharging the timing
www.americanradiohistory.com
HANDBOOK Waveform Displays 173
B+
2S K
To achieve proper focus on the screen of
the cathode -ray tube it is necessary that the
SIGNAL CLAMP
final anode and both pairs of deflection
plates have approximately the same average
.001 potential. Since it is necessary to have the
V13
1.0 22
SWEEP vertical deflection plates at ground potential
AMPLIFIER
-BIAS so a direct connection may be made if de-
0+
sired, the average potential of the horizontal
200
660K
plates must also be near ground. The mean
SWEEP
potential of the sweep amplifier plate circuit
AMPLIFIER is about +250 volts. This is moved down
120K V14
100 K to ground by means of the groups of neon
HORIZON.
POSITION
glow lamps (N1 -N,0) which produce a
CONTROL, - 1500 V. constant voltage drop. A steady current of
about 200 microamperes keeps the lamps
2SK ionized so that any change in plate potential
B+ of the sweep amplifier tubes (such as caused
by signals) appears on the deflection plates
Figure 6 unchanged in amplitude, but moved down
SWEEP AMPLIFIER
in potential about 250 volts. The ionizing
current is obtained from the -1500 volt
Cathode- coupled swoop amplifier provides bal-
anced push -pull signal to deflection plates of cathode -ray tube power supply through a
cathode-ray tube. Two groups of neon glow high- resistance network. Since the imped-
lamps pass sweep signal but move average ance of the neon glow lamps is rather high
signal potential close to ground by virtue of
constant voltage drop across lamps. Ionising at frequencies involving the faster sweeps,
voltage for glow lamps is taken from high - small capacitors are shunted across the lamps
voltage cathode -ray tube power supply.
to pass these frequencies.
capacitor and leaving it ready to receive the The Power The low- voltage power supply
next sweep pulse from the multivibrator. Supply provides positive and negative
regulated voltages for the var-
The Sweep Since the amplitude of the sweep ious stages of the oscilloscope. The accel-
Amplifier waveform at the output of the erating potential for the cathode -ray tube
sweep generator is not large is obtained from an oscillator operating from
enough to drive the horizontal deflection the low -voltage supply (figure 7). The os-
plates of the cathode -ray tube, further am- cillator is a conventional Hartley circuit,
plification is needed. The signal from the with a high- voltage secondary winding on
sweep amplitude is impressed on the grid of the oscillator transformer which supplies
a cathode -coupled sweep amplifier (V1,, about 1200 volts rms to the rectifier tubes.
V14, figure 6) which inverts the phase and Filament voltages for these tubes are also
operates as a push -pull stage. Balanced sweep obtained from windings on the oscillator
voltage is necessary to maintain the average transformer. The frequency of oscillation is
potential of the deflection plates constant about 2000 Hertz.
over the entire sweep to prevent defocussing.
The horizontal position control varies the
bias on one amplifier tube and thus deter- 8 -2 Display of Waveforms
mines the position from which the sweep
starts. To ensure that the sweep will always Together with a working knowledge of
start at the same position on the screen each the controls of the oscilloscope, an under-
time (for a given setting of the position standing of how the patterns are traced on
control) a diode clamp (V12) is placed be- the screen must be obtained for a thorough
tween the grid of the opposite amplifier knowledge of oscilloscope operation. With
tube and ground to remove any charge that this in mind a careful analysis of two funda-
the input coupling capacitor may have mental waveform patterns is discussed under
gained during the previous sweep cycle. the following headings:
174 The Oscilloscope THE RADIO
SABP1
Br
.05-r 20.5
FROM SWEEP UNBLANN /NC
C/RCU /r PULSE
Figure 7
1. Patterns plotted against time (using wave is amplified by the vertical amplifier
the sweep generator for horizontal de- and impressed on the vertical (Y -axis) de-
flection) . flection plates of the cathode-ray tube. Si-
2. Lissajous figures (using a sine wave for multaneously the sawtooth wave from the
horizontal deflection). time -base generator is amplified and im-
Patterns Plotted
pressed on the horizontal (X -axis) deflection
A sine wave is typical of
Against Time plates.
such a pattern and is con-
venient for this study. This
TIME
Figure 8
PROJECTION DRAWING OF A SINE
Figure 9
WAVE APPLIED TO THE VERTICAL
AXIS AND 1 SAWTOOTH WAVE OF THE PROJECTION DRAWING SHOWING THE
SAME FREQUENCY APPLIED SIMUL- RESULTANT PATTERN WHEN THE FRE-
TANEOUSL'" ON THE HORIZONTAL QUENCY OF THE SAWTOOTH IS ONE -
AXIS HALF OF THAT EMPLOYED IN FIGURE 8
www.americanradiohistory.com
HANDBOOK Lissajous Figures 175
The electron beam moves in accordance form plotted against time can be displayed
with the resultant of the sine and sawtooth and analyzed by the same procedure as used
signals. The effect is shown in figure 8 where in these examples.
the sine and sawtooth waves are graphically The sine -wave problem just illustrated is
represented on time and voltage axes. Points typical of the method by which any wave-
on the two waves that occur simultaneously form can be displayed on the screen of the
are numbered similarly. For example, point 2 cathode -ray tube. Such waveforms as square
on the sine wave and point 2 on the saw - wave, sawtooth wave, and many more ir-
tooth wave occur at the same instant. There- regular recurrent waveforms can be observed
fore the position of the beam at instant 2 is by the same method explained in the preced-
the resultant of the voltages on the hori- ing paragraphs.
zontal and vertical deflection plates at in-
stant 2. Referring to figure 8, by projecting 8 -3 Lissajous Figures
lines from the two point -2 positions, the
position of the electron beam at instant 2
can be located. If projections were drawn Another fundamental pattern is the Lissa -
from every other instantaneous position of jous figures, named after the 19th- century
each wave to intersect on the circle repre- French scientist. This type of pattern is of
senting the tube screen, the intersections of particular use in determining the frequency
similarly timed projects would trace out a ratio between two sine -wave signals. If one
sine wave. of these signals is known, the other can be
In summation, figure 8 illustrates the easily calculated from the pattern made by
principles involved in producing a sine - the two signals on the screen of the cathode -
wave trace on the screen of a cathode -ray ray tube. Common practice is to connect the
tube. Each intersection of similarly timed known signal to the horizontal channel and
projections represents the position of the the unknown signal to the vertical channel.
electron beam acting under the influence of The presentation of Lissajous figures can
be analyzed by the same method as previ-
the varying voltage waveforms on each pair
of deflection plates. Figure 9 shows the effect ously used for sine -wave presentation. A
simple example is shown in figure 10. The
on the pattern of decreasing the frequency
of the sawtooth wave. Any recurrent wave- frequency ratio of the signal on the hori-
zontal axis to the signal on the vertical axis
is 3 to 1. If the known signal on the hori-
zontal axis is 180 Hertz, the signal on the
vertical axis is 60 Hertz.
Obtaining a Lissajous 1. The horizontal am-
Pattern on the Screen; plifier should be dis -
Oscilloscope Settings connected from the
sweep oscillator. The
signal to be examined should be connected to
the horizontal amplifier of the oscilloscope.
Figure 10
PHASE DIFFERENCE =0 PHASE DIFFERENCE =4S PHASE DIFFERENCE .90. PHASE DIFFERENCE=133
PHASE DIFFERENCE 150 PHASE DIFFERENCE =223 PHASE DIFFERENCE 270 PHASE DIFFERENCE 315
Figure 13
www.americanradiohistory.com
HANDBOOK Lissajous Figures 177
Y MAXIMUM' I Y MAXIMUM'I
Y INTERCEPT '.7
''
Y INTERCEPT Y INTERCEPT ..5
SINE .T SINE +
. 0 45
MAXIMUM.'
Figure 15
Figure 16
TRAPEZOIDAL MODULATION PATTERN
Figure 18
Figure 17 PROJECTION DRAWING SHOWING TRAPE-
MODULATED CARRIER -WAVE PATTERN ZOIDAL PATTERN
gain controls in the oscillograph, it should cal channel of the oscilloscope, and follow
not be changed for the duration of the meas- the procedure set forth in this discussion to
urement. observe the various phase -shift patterns.
www.americanradiohistory.com
HANDBOOK I -F Al ignment 179
R -F POWER AMPLIFIER
MODULATED
CARRIER
TO ANTENNA
TIME
MODULATOR 500PF IM
STAGE 10,000 - VOLT
TV CAPACITOR
SAW 700TH
SWEEP
LC TUNES TO OP-
ERATING FREQUENCY
Figure 19
C L
PROJECTION DRAWING SHOWING s+ NOTE' IF R-F PICKUP /S INSUFFICIENT,
A TUNEO CIRCUIT MAY RE USED
MODULATED- CARRIER WAVE PATTERN As THE OSCILLOSCOPE AS SHOWN.
Figure 20
deflection. The modulated -wave pattern also MONITORING CIRCUIT FOR TRAPEZOIDAL
can be used for analyzing waveforms. Figure MODULATION PATTERN
19 shows how the two signals cause the elec-
tron beam to trace out the pattern.
where,
The Trapezoidal The oscilloscope connections
and Eln n are defined as in figure 21.
Emax
Pattern for obtaining a trapezoidal 1
with an Oscilloscope
Modulation percentage =
Emax - Em1n
X 100
The alignment of the i -f amplifiers of a
receiver consists of adjusting all the tuned
Emax + Emin
circuits to resonance at the intermediate fre-
180 The Oscilloscope THE RADIO
TRAPEZOIDAL PATTERNS
T T
E MIN EIMAN
J-
_i
Figure 21 Figure 22 Figure 23
quency and at the same time permitting pas- As a rule medium- and low- priced receiv-
sage of a predetermined number of side- ers use i -f transformers whose bandwidth is
bands. The best indication of this adjustment about 5 kHz on each side of the funda-
is a resonance curve representing the response mental frequency. The response curve of
of the i -f circuit to its particular range of these i -f transformers is shown in figure 28.
frequencies. High- fidelity receivers usually contain i -f
transformers which have a broader band-
width which is usually 10 kHz on each side
R -F POWER AMPLIFIER CRO of the fundamental. The response curve for
this type transformer is shown in figure 29.
TO ANTENNA Resonance curves such as these can be dis-
USE INTERNAL
played on the screen of an oscilloscope. For
SWEE
a complete understanding of the procedure
it is important to know how the resonance
curve is traced.
!r FROM
MODULATOR The Resonance To present a resonance curve
LC TUNES TO OP-
ERATING FREQUENCY
Curve on the on the screen, .a frequency -
Screen modulated signal source must
Figure 24 be available. This signal
source is a signal generator whose output is
MONITORING CIRCUIT FOR the fundamental i -f frequency which is fre-
MODULATED-WAVE PATTERN quency- modulated S to 10 kHz each side of
EMAIL
NAin
IAA
Figure 25 Figure 26 Figure 27
www.americanradiohistory.com
HANDBOOK I -F Alignment 181
--
the fundamental frequency.
When using sawtooth sweep, the two re-
sponse curves can also be superimposed. If
the sawtooth signal is generated at exactly R -i INPUT ._,,,___IPOWER AMPLIFIER
UNDER TEST
TO LOAD
www.americanradiohistory.com
HANDBOOK SSB Applications 183
block diagram of a typical linearity test set- It is convenient to build the detector in a
up. A two -tone test signal is normally em- small shielded enclosure such as an i -f trans-
ployed to supply an SSB modulation envelope, former can fitted with coaxial input and
but any modulating signal that provides an output connectors. Voltage dividers can be
envelope that varies from zero to full ampli- similarly constructed so that it is easy to in-
tude may be used. Speech modulation gives a sert the desired amount of voltage attenua-
satisfactory trace, so that this instrument tion from the various sources. In some cases
may be used as a visual monitor of transmit- it is convenient to use a pickup loop on the
ter linearity. It is particularly useful for end of a short length of coaxial cable.
monitoring the signal level and clearly shows The phase shift of the amplifiers in the os-
when the amplifier under observation is over- cilloscope should be the same and their fre-
loaded. The linearity trace will be a straight quency response should be flat out to at least
line regardless of the envelope shape if the twenty times the frequency difference of the
amplifier has no distortion. Overloading two test tones. Excellent high- frequency
causes a sharp break in the linearity curve. characteristics are necessary because the
Distortion due to too much bias is also easily rectified SSB envelope contains harmonics
observed and the adjustment for low dis- extending to the limit of the envelope de-
tortion can easily be made. tector's response. Inadequate frequency re-
Another feature of the linearity detector sponse of the vertical amplifier may cause a
is that the distortion of each individual little "foot" to appear on the lower end of
stage can be observed. This is helpful in the trace, as shown in figure 37. If it is
troubleshooting. By connecting the input small, it may be safely neglected.
envelope detector to the output of the SSB Another spurious effect often encountered
generator, the over-all distortion of the en- is a double trace, as shown in figure 38. This
tire r -f circuit beyond this point is observed. can usually be corrected with an RC net-
The unit can also serve as a voltage indicator work placed between one detector and the
which is useful in making tuning adjust- oscilloscope. The best method of testing the
ments. detectors and the amplifiers is to connect the
The circuit of a typical envelope detector input of the envelope detectors in parallel.
is shown in figure 36. Two matched ger- A perfectly straight line trace will result
manium diodes are used as detectors. The when everything is working properly. One
detectors are not linear at low signal levels, detector is then connected to the other r -f
but if the nonlinearity of the two detectors source through a voltage divider adjusted so
:s matched, the effect of their nonlinearity on that no appreciable change in the setting of
184 The Oscilloscope
es
ourPUT
SIGNAL
LEVEL
i
Figure 37
EFFECT OF INADEQUATE
RESPONSE OF VERTICAL INPUT SIGNAL LEVEL
AMPLIFIER
Figure 40
Figure 39
www.americanradiohistory.com
CHAPTER NINE
A whole new concept of vacuum -tube ap- positive potential may be placed on the cath-
plications has been developed in recent years. ode, but the tube will not conduct until the
No longer are vacuum tubes chained to the voltage on the plate rises above an equally
field of communication. This chapter is de- positive value. As the plate becomes more
voted to some of the more common circuits positive with respect to the cathode, the
encountered in industrial and military appli- diode conducts and passes that portion of the
cations of the vacuum tube. wave which is more positive than the cath-
ode voltage. Diodes may be used as either
9 -1 Limiting Circuits series or parallel limiters, as shown in figure
1. A diode may be so biased that only a cer-
tain portion of the positive or negative cycle
The term limiting refers to the removal or is removed.
suppression, by electronic means, of the ex-
tremities of an electronic signal. Circuits Audio Peak
which perform this function are referred to An audio peak clipper consist -
Limiting ing of two diode limiters may
as limiters or clippers. Limiters are useful in
waveshaping circuits where it is desirable to be used to limit the amplitude
square off the extremities of the applied sig- of an audio signal to a predetermined value
nal. A sine wave may be applied to a limiter
to provide a high average level of modula-
tion without danger of overmodulation. An
circuit to produce a rectangular wave. A
peaked wave may be applied to a limiter cir- effective limiter for this service is the series -
diode gate clipper. A circuit of this clipper
cuit to eliminate either the positive or nega-
is shown in figure 2. The audio signal to be
tive peaks from the output. Limiter circuits
are employed in f -m receivers where it is clipped is coupled to the clipper through C1.
necessary to limit the amplitude of the signal R, and R2 are the clipper input and output
load resistors. The clipper plates are tied
applied to the detector. Limiters may be
used to reduce automobile ignition noise in
together and are connected to the clipping
level control (R,) through series resistor Rg.
short -wave receivers, or to maintain a high
average level of modulation in a transmitter.
R, acts as a voltage divider between the
high -voltage supply and ground. The exact
They may also be used as protective devices
to limit input signals to special circuits. point at which clipping will occur is set by
R which controls the positive potential ap-
Diode LimitersThe characteristics of a diode plied to the diode plates.
tube are such that the tube Under static conditions, a d -c voltage is
conducts only when the plate is at a positive obtained from R, and applied through R3
potential with respect to the cathode. A to both plates of the 6AL5 tube. Current
186 Special Vacuum -Tube Circuits THE RADIO
e OUT
eiN
1\1\
eour
E=VOLTAGE DROP
ACROSS DIODE
E
_e our
nn nn
E= VOLTAGE DROP
ACROSS DIODE
E11
e IN
end
Figure 1
flows through R,, R3, and divides through is lessthan 4 volts, both halves of the tube
the the two diode sections of the 6AL5 and conduct at all times. As long as the tube
the two load resistors (R1 and R2). All parts conducts, its resistance is very low compared
of the clipper circuit are maintained at a with plate resistor R3. Whenever a voltage
positive potential above ground. The voltage change occurs across input resistor R1, the
drop between the plate and cathode f each voltage at all of the tube elements increases
diode is very small compared to the drop or decreases by the same amount as the input
across the 300,000 -ohm resistor (R3) in voltage changes, and the voltage drop across
series with the diode plates. The plate and R3 changes by an equal amount. As long as
cathode of each diode are therefore main- the peak input voltage is less than 4 volts,
tained at approximately equal potentials as the 6AL5 acts merely as a conductor, and
long as there is plate- current flow. Clipping the output cathode is permitted to follow
does not occur until the peak audio -input all voltage changes at the input cathode.
voltage reaches a value greater than the If, under static conditions, 4 volts appear
static voltages at the plates of the diode. at the diode plates, then twice this voltage (8
Assume that R, has been set to a point volts) will appear if one of the diode circuits
that will give 4 volts at the plates of the is opened, thus removing its d -c load from
6AL5. When the peak audio -input voltage the circuit. As long as only one of the diodes
www.americanradiohistory.com
HANDBOOK Limiting Circuits 187
GALS R3
CLIPPING
BOOK LEVEL
CONTROL IM
e1N
CI C2
o.I 0.1 R4
100
e Dur 11
eIN
Bf
R2
20011 EGRIDCArHOOE RESISTANCE
WHEN GRID IS DRIVEN MIME
Figure 2 Figure 3
continues to conduct, the voltage at the potential at the output cathode will follow
diode plates cannot rise above twice the the input cathode voltage and decrease from
voltage selected by R*. In this example, the its normal value of 4 volts until it reaches
voltage cannot rise above 8 volts. Now, if zero potential. As the input cathode voltage
the input audio voltage applied through C, decreases to less than zero, the plates will
is increased to any peak value between zero follow. However, the output cathode,
and + 4 volts, the first cathode of the grounded through R2, will stop at zero
6AL5 will increase in voltage by the same potential as the plate becomes negative. Con-
amount to the proper value between 4 and duction through the second diode is impos-
8 volts. The other tube elements will assume sible under these conditions. The output
the same potential as the first cathode. How- cathode remains at zero potential until the
ever, the 6AL5 plates cannot increase more voltage at the input cathode swings back to
than 4 volts above their original 4 -volt static zero.
level. When the input voltage to the first The voltage developed across output resis-
cathode of the 6AL5 increases to more than tor R_ follows the input voltage variations
+ 4 volts, the cathode potential increases as long as the input voltage does not swing
to more than 8 volts. Since the plate circuit to a peak value greater than the static volt-
potential remains at 8 volts, the first diode age at the diode plates, which is determined
section ceases to conduct until the input by R,. Effective clipping may thus be ob-
voltage across R, drops below 4 volts. tained at any desired level.
When the input voltage swings in a nega- The square- topped audio waves generated
tive direction, it will subtract from the 4- by this clipper are high in harmonic content,
volt drop across R, and decrease the voltage but these higher -order harmonics may be
on the input cathode by an amount equal greatly reduced by a low -level speech filter.
to the input voltage. The plates and the out-
put cathode will follow the voltage level at Grid Limiters A triode grid limiter is shown
the input cathode as long as the input volt- in figure 3. On positive peaks
age does not swing below 4 volts. If - of the input signal, the triode grid attempts
to swing positive, and the grid -cathode re-
the input voltage does not change more than
4 volts in a negative direction, the plates of sistance drops to about 1000 ohms or so. The
the 6AL5 will also become negative. The voltage drop across the series grid resistor
e1N eour
o
DIODE CONDUCTS
Figure 4
elw--- i iDCrLECTION
COIL
L _
:
1 1
-100v. CI CHARGE PATH Cz DISCHARGE PATH
Figure 5 Figure 7
1+ s+ 5+
SY AONIZING
SIGNAL
O
DIRECT-COUPLED CATHODE ELECTRON - COUPLED MULTI VIBRATOR WITH SINE -WAVE
MULTI VIBRATOR MULTIVIBRATOR SYNCHRONIZING SIGNAL APPLIED
TO ONE TUBE
Figure 8
www.americanradiohistory.com
HANDBOOK Multivibrators 189
e+ I+
0
BASIC ECCLES-JORDAN TRIGGER ONE -SHOT MULTIVIBRATOR
CIRCUIT
Figure 9
but the inherent frequency stability is poor. The Eccles- Jordan The Eccles- Jordan trigger
The multivibrator may be stabilized by the Circuit circuit is shown in figure
introduction of synchronizing voltages of 9A. This is not a true
harmonic or subharmonic frequency. multivibrator, but rather a circuit that pos-
In its simplest form, the multivibrator is sesses two conditions of stable equilibrium.
a simple two -stage RC- coupled amplifier One condition is when V, is conducting and
with the output of the second stage coupled V2 is cutoff; the other when V2 is conduct-
through a capacitor to the grid of the first ing and V, is cutoff. The circuit remains in
tube, as shown in figure 6. Since the output one or the other of these two stable condi-
of the second stage is of the proper polarity tions with no change in operating potentials
to reinforce the input signal applied to the until some external action occurs which
first tube, oscillations can readily take place, causes the nonconducting tube to conduct.
started by thermal- agitation and miscellane- The tubes then reverse their functions and re-
ous tube noise. Oscillation is maintained by main in the new condition as long as no
the process of building up and discharging plate current flows in the cut -off tube. This
the store of energy in the grid -coupling type of circuit is known as a flip-flop circuit.
capacitors of the two tubes. The charging Figure 9B illustrates a modified Eccles -
and discharging paths are shown in figure 7. Jordan circuit which accomplishes a com-
Various types of multivibrators are shown in plete cycle when triggered by a positive
figure 8. pulse. Such a circuit is called a one -shot
The output of a multivibrator may be used multivibrator. For initial action, V, is cut
as a source of square waves, as an electronic off and V2 is conducting. A large positive
switch, or as a means of obtaining frequency
division. Submultiple frequencies as low as
one -tenth of the injected synchronizing fre- R -F
PULSE
R -F
PULSE
R -F
PULSE
quency may easily be obtained.
Intl I Nil
I.--I
CUTOFF
I---I
CUTOFF
TIME TIM[
eouT
II u
TIM[ Figure 11
www.americanradiohistory.com
190 Special Vacuum -Tube Circuits THE RADIO
POSITIVE COUNTING CIRCUIT NEGATIVE COUNTING CIRCUIT POSITIVE COUNTING CIRCUIT WITH
METER INDICATION
Figure 12
pulse applied to the grid of V1 causes this currence is determined by the RC time con-
tube to conduct, and the voltage at its plate stant of the grid circuit. A single -swing
decreases by virtue of the IR drop through blocking oscillator is shown in figure 10,
R3. Capacitor C2 is charged rapidly by this wherein the tube is cut off before the comple-
abrupt change in V1 plate voltage, and V2 tion of one cycle. The tube produces single
becomes cut off while V1 conducts. This con- pulses of energy, the time between the
dition exists until C2 discharges, allowing V2 pulses being regulated by the discharge time
to conduct, raising the cathode bias of V1 of the grid RC network. The self - pulsing
until it is once again cut off. blocking oscillator is shown in figure 11, and
-
A direct cathode- coupled multivibrator is used to produce pulses of r -f energy, the
is shown in figure SA. RR is a common number of pulses being determined by the
cathode resistor for the two tubes, and cou- timing network in the grid circuit of the
pling takes place across this resistor. It is oscillator. The rate at which these pulses
impossible for a tube in this circuit to com- occur is know as the pulse- repetition fre-
pletely cut off the other tube, and a circuit quency, or p.r.f.
of this type is called a free-running multi -
vibrator in which the condition of one tube
temporarily cuts off the other.
9 -5 Counting Circuits
A counting circuit, or frequency divider,
9 -4 The Blocking Oscillator isone which receives uniform pulses (repre-
senting units to be counted) and produces a
A blocking oscillator is any oscillator voltage that is proportional to the frequency
which cuts itself off after one or more cycles of the pulses. A counting circuit may be
caused by the accumulation of a negative
charge on the grid capacitor. This negative v:
charge may gradually be drained off through eIN
the grid resistor of the tube, allowing the
circuit to oscillate once again. The process
is repeated and the tube becomes an inter-
mittent oscillator. The rate of such an oc-
Vx
7e 1N
Figure 14
www.americanradiohistory.com
HANDBOOK Resistance -Capacitance Oscillators 191
Re
Rs 7
Figure 15 Figure 16
THE WIEN- BRIDGE AUDIO OSCILLATOR THE PHASE -SHIFT OSCILLATOR
R,C,
when,
ure 13) . A blocking oscillator may be con-
nected to a step counter, as shown in figure
R, X C, equals R2 X C2
14. The oscillator is triggered into operation A lamp (Lu) is used for the cathode resistor
when the voltage across C2 reaches a point of VI as a thermal stabilizer of the oscillator
sufficiently positive to raise the grid of V, amplitude. The variation of the resistance
above cutoff. Circuit parameters may be with respect to the current of the lamp
chosen so that a count division up to 1/20 bulb holds the oscillator output voltage at a
may be obtained with reliability. nearly constant amplitude.
www.americanradiohistory.com
192 Special Vacuum -Tube Circuits THE RADIO
+
TK CISOK
6AU6 S
6CL6 G
-..+ IGL/
330
LOOP I
I
-11+
R3 LP, 2010 K
GOO iW Ca
I
Figure 17
..5,10.
www.americanradiohistory.com
HANDBOOK Feedback 193
ROOM
FURNACE TEMPERATURE PHASE SHIFT INPUT SIGNAL
OF SYSTEM ,OUTPUT SIGNAL
(F) (T1
THERMOSTAT
..TH) /
FEEDBACK
TIME-.
(ERROR S!CNAL ) FEEDBACK SIGNAL OUTPUT SIGNAL
NO PHASE SHIFT
FUEL
SUPPLY
3
II,k 2. A FEEDBACK
SIGNAL
Figure 20 I
WITH 1/10.
PHASE
a SHIFT
SIMPLE CLOSED -LOOP
TIME-.
FEEDBACK SYSTEM
Room temperature (T) controls fuel supply to Figure 21
furnace (F) by feedback loop through thermo-
stat (TH) control. PHASE SHIFT OF ERROR
SIGNAL MAY CAUSE OSCILLA-
Mechanical feedback has been used for many TION IN CLOSED LOOP SYSTEM
years in such forms as engine-speed gover- To p t oscillation, the gain of the feed-
nors and servo steering engines on ships. back loop must be less than unity when the
phase shift of the system reaches 180 degrees.
A simple feedback system for temperature
control is shown in figure 20. This is a cause-
and- effect system. The furnace (F) raises cause the feedback control to overcorrect in
the room temperature (T) to a predeter- the opposite direction, resulting in hunting
mined value at which point the sensing or oscillation of the closed -loop system about
thermostat (TH) reduces the fuel flow to the correct operating point.
the furnace. When the room temperature Negative- feedback control would tend to
drops below the predetermined value the damp out spurious system oscillation if it
fuel flow is increased by the thermostat con- were not for the time lag or phase shift in the
trol. An interdependent control system is system. If the over -all phase shift is equal to
created by this arrangement: the room tem- one -half cycle of the operating frequency of
perature depends on the thermostat action, the system, the feedback will maintain a
and the thermostat action depends on the steady state of oscillation when the circuit
room temperature. This sequence of events gain is sufficiently high (figure 21) . In
may be termed a closed -loop feedback system. order to prevent oscillation, the gain figure
of the feedback loop must be less than unity
Error Cancellation A feedback control sys- when the phase shift of the system reaches
tem is dependent on a de- 180 degrees. In an ideal control system the
gree of error in the output signal, since this gain of the loop would be constant through-
error component is used to bring about the out the operating range of the device, and
correction. This component is called the would drop rapidly outside the range to
error signal. The error, or deviation from the reduce the bandwidth of the control system
desired signal is passed through the feedback to a minimum.
loop to cause an adjustment to reduce the The time lag in a closed -loop system may
value of the error signal. Care must be taken be reduced by using electronic circuits in
in the design of the feedback loop to reduce place of mechanical devices, or by the use of
over- control tendencies wherein the correc- special circuit elements having a phase -lead
tion signal would carry the system past the characteristic. Such devices make use of the
point of correct operation. Under certain properties of a capacitor, wherein the current
circumstances the new error signal would leads the voltage applied to it.
CHAPTER TEN
www.americanradiohistory.com
Detection or Demodulation 195
1-
Lt L2 PLATE- TICKLER REGENERATION WITH 'THROTTLE"
GROUND
CAPACITOR REGENERATION CONTROL.
www.americanradiohistory.com
HANDBOOK Superheterodyne Receivers 197
TANTENNA AUDIO
12AT7 OUTPUT
I RP
r----
IINTCRMCD SECOND
r -
I
AUDIO
-T
REWENCT 1
----i
AMPLIr1ER DETECTOR IAMPLIrICRI
1
AMPLIFIER
_ ---
I
L _ -- ` -'
r 1
IPREOUENCTI
OSCILLATORI
! (roRCw)I
Figure 5
ESSENTIAL UNITS OF A
SUPERHETERODYNE RECEIVER
The basic portions of the receiver are shown
in solid blocks Practicable receivers employ
the dotted blocks and also usually include such
additional circuits as a noise limiter, an ave
circuit, and a crystal Alter in the i-1 amplifier.
www.americanradiohistory.com
HANDBOOK Superheterodyne Receivers 199
6S7, 65B7-Y,
TO 1./
6BE6. 6BA7 AMP.
2!O V.
1!O V.
Figure 7
use two tubes, or two sets of elements in ever, the equivalent input noise resistance of
the same envelope, in an oscillator -mixer ar- such tubes is of the order of 200,000 ohms,
rangement. Figure 7 shows a group of cir- which is a rather high value indeed; so such
cuits of both types to illustrate present prac- tubes are not suited for operation without an
tice with regard to types of converter stages. r -f stage in the high- frequency range if
Converter -tube combinations such as shown weak -signal reception is desired.
in figures 7A and 7B are relatively simple The 6L7 mixer circuit shown in figure 7C,
and inexpensive, and they do an adequate and the 6BA7 circuit of figure 7D, also are
job for most applications. With a converter characterized by an equivalent input noise
tube such as the 6SB7 -Y or the 6BA7 quite resistance of several hundred thousand ohms,
satisfactory performance may be obtained so that these also must be preceded by one or
for the reception of relatively strong signals more r -f stages with a fairly high gain per
(as for example f -m broadcast reception) up stage if a low noise factor is desired of the
to frequencies in excess of 100 MHz. How- complete receiver.
www.americanradiohistory.com
200 Radio Receiver Fundamentals THE RADIO
However, the circuit arrangements shown having a high signal -to -noise ratio ahead of
at figures 7E and 7F are capable of low - the mixer. This remedy causes the signal out-
noise operation within themselves, so that put from the mixer to be large in proportion
these circuits may be fed directly from the to the noise generated in the mixer stage. In-
antenna without an r -f stage and still pro- creasing the gain after the mixer will be of
vide a fair noise factor to the complete re- no advantage in eliminating mixer noise dif-
ceiver. Note that both these circuits use ficulties; greater selectivity after the mixer
control-grid injection of both the incoming will help to a certain extent, but cannot be
signal and the local- oscillator voltage. Hence. carried too far, since this type of selectivity
paradoxically, circuits such as these should decreases the i -f bandpass and if carried too
be preceded by an r -f stage if local- oscillator far will not pass the sidebands that are an
radiation is to be held to any reasonable essential part of a voice -modulated signal.
value of field intensity.
Triode Mixers A triode having a high trans -
Diode Mixers As the frequency of operation conductance is the quietest
of a superheterodyne receiver is mixer tube, exhibiting somewhat less gain
increased above a few hundred megahertz but a better signal -to-noise ratio than a
the signal-to -noise ratio appearing in the comparable multigrid mixer tube. However,
plate circuit of the mixer tube when triodes below 30 MHz it is possible to construct a
or pentodes are employed drops to a prohibi- receiver that will get down to the atmos-
tively low value. At frequencies above the pheric noise level without resorting to a
upper frequency limit for conventional mix- triode mixer. The additional difficulties ex-
er stages, mixers of the diode type are most perienced in avoiding pulling, undesirable
commonly employed. The diode may be feedback, etc., when using a triode with
either a vacuum -tube heater diode of a spe- control -grid injection tend to make multi -
cial uhf design such as the 9005, or it may grid tubes the popular choice for this ap-
be a crystal diode of the general type of the plication on the lower frequencies.
1N21 through 1N28 series. On very-high frequencies, where set noise
rather than atmospheric noise limits the
10 -4 Mixer Noise weak -signal response, triode mixers are more
and Images widely used. A 6J6 miniature twin triode
with grids in push -pull and plates in parallel
makes an excellent mixer up to about 150
The effects of mixer noise and images are
MHz.
troubles common to all superheterodynes.
Since both these effects can largely be ob- Injection The amplitude of the injection
viated by the same remedy, they will be con- Voltage voltage will affect the conversion
sidered together. transconductance of the mixer.
and therefore should be made optimum if
Mixer Noise Mixer noise of the shot -effect maximum signal -to -noise ratio is desired. If
type, which is evidenced by a fixed bias is employed on the injection grid,
hiss in the audio output of the receiver, is the optimum injection voltage is quite criti-
caused by small irregularities in the plate cal. If cathode bias is used, the optimum
current in the mixer stage and will mask voltage is not so critical; and if grid -leak
weak signals. Noise of an identical nature is bias is employed, the optimum injection volt-
generated in an amplifier stage, but due to age is not at all critical -just so it is ade-
the fact that the conductance in the mixer quate. Typical optimum injection voltages
stage is considerably lower than in an amp- will run from 1 to 3 volts for control -grid
lifier stage using the same tube, the propor- injection, and 20 volts or so for screen- or
tion of inherent noise present in a mixer suppressor -grid injection.
usually is considerably greater than in an
amplifier stage using a comparable tube. Images There always are two signal fre-
Although this noise cannot be eliminated, quencies which will combine with a
its effects can be greatly minimized by plac- given frequency to produce the same differ-
ing sufficient signal- frequency amplification ence frequency. For example: assume a super-
www.americanradiohistory.com
HANDBOOK Mixer Noise and Images 201
PEN root
frequency, greater selectivity in the mixer
grid circuit through the use of additional
tuned circuits between the mixer and the
antenna is necessary if a good image ratio is
to be maintained.
10 -5 R -F Stages
Figure 8
TYPICAL PENTODE R -F AMPLIFIER STAGE Since the necessary tuned circuits between
the mixer and the antenna can be combined
with tubes to form r -f amplifier stages, the
heterodyne with its oscillator operating on a reduction of the effects of mixer noise and
higher frequency than the signal (which is the increasing of the image ratio can be ac-
common practice in many superhetero- complished in a single section of the receiver.
dynes) tuned to receive a signal at 14,100 When incorporated in the receiver, this sec-
kHz. Assuming an i-f amplifier frequency of tion is known simply as an r -f amplifier;
450 kHz, the mixer input circuit will be when it is a separate unit with a separate
tuned to 14,100 kHz, and the oscillator to tuning control it is often known as a pre -
14,100 plus 450, or 14,550 kHz. Now, a selector. Either one or two stages are com-
strong signal at the oscillator frequency plus monly used in the preselector or r -f ampli-
the intermediate frequency (14,550 plus fier. Some preselectors use regeneration to
450, or 15,000 kHz) will also give a differ- obtain still greater amplification and selec-
ence frequency of 450 kHz in the mixer tivity. An r -f amplifier or preselector em-
output and will be heard also. Note that the bodying more than two stages rarely ever is
image is always twice the intermediate fre- employed since two stages will ordinarily
quency away from the desired signal. Images give adequate gain to override mixer noise.
cause repeat points on the tuning dial.
The only way that the image could be R -FStages in Generally speaking, atmos-
eliminated in this particular case would be to the VHF Range pheric noise in the frequency
make the selectivity of the mixer input cir- range above 30 MHz is
cuit, and any circuits preceding it, great quite low-so low, in fact, that the noise
enough so that the 15,000 -kHz signal never generated within the receiver itself is greater
reaches the mixer grid in sufficient amplitude than the noise received on the antenna.
to produce interference. Hence it is of the greatest importance that
For any particular intermediate frequency, internally generated noise be held to a mini-
image interference troubles become increas- mum in a receiver. At frequencies above 500
ingly greater as the frequency (to which the MHz there is not much that can be done
signal-frequency portion of the receiver is in the direction of reducing receiver noise
tuned) is increased. This is due to the fact below that generated in the converter stage,
that the percentage difference between the aside from the use of specialized parametric
desired frequency and the image frequency amplifiers. But in the vhf range, between 30
decreases as the receiver is tuned to a higher and 500 MHz, the receiver noise factor in a
frequency. The ratio of strength between a well- designed unit is determined by the char-
signal at the image frequency and a signal at acteristics of the first r -f stage.
the frequency to which the receiver is tuned The usual vhf receiver, whether for com-
producing equal output is known as the munications or for f -m or TV reception,
image ratio. The higher this ratio is, the uses a miniature pentode or triode for the
better the receiver will be in regard to image first r -f amplifier stage. The nuvistors
interference troubles . (6CW4 and 6DS4) are the best of presently
With but a single tuned circuit between available types, with the 6EH7 (pentode)
the mixer grid and the antenna, and with and the cascode -style amplifier approaching
400- to 500 -kHz i -f a?1 plifiers, image ratios nuvistor performance in the lower vhf
of 60 db and over are easily obtainable up to region. However, when gain in the first r -f
frequencies around 2000 kHz. Above this stage is not so important, and the best noise
202 Radio Receiver Fundamentals THE RADIO
DA GROUNDED GRID factory when the first r -f stage is to be fed
6CW4 directly from a low- impedance coaxial trans-
.001
mission line. Figure 9 (B) gives somewhat
more gain than (A), but requires an input
matching circuit. The effective gain of this
circuit is somewhat reduced when it is being
used to amplify a broad band of frequencies
since the effective gm of the cathode -cou-
pled dual tube is somewhat less than half the
g,,, of either of the two tubes taken alone.
The CascadeThe Cascode r-f amplifier, de-
Amplifier veloped at the MIT Radiation
Laboratory during World War
II, is a low -noise circuit employing a
grounded -cathode triode driving a grounded -
grid triode, as shown in figure 9C. The stage
CATHODE COUPLED gain of such a circuit is about equal to that
of a pentode tube, while the noise figure re-
mains at the low level of a triode tube. Neu-
+150v tralization of the first triode tube is usually
unnecessary below SO MHz. Above this fre-
001 LN 6CW4
6CW4
quency, a definite improvement in the noise
figure may be obtained through the use of
neutralization. The neutralizing coil (LN)
should resonate at the operating frequency
with the grid -plate capacity of the first tri-
ode tube.
b0 Ool
The TV -type double triodes such as the
c
-I- LOW-NOISE 00 6DJ8 (and older style 6BQ7 and 6BZ7) may
NUVISTOR be used to good advantage up to 144 MHz
CASCODE
or so. At 2 meters and above, however, the
0 DUAL- TRIODE
CASCODE LN
+120V
6CW4 nuvistor family is recommended for
use.
60.16,6807, Etc
Double Conversion As previously mentioned,
the use of a higher inter-
mediate frequency will also improve the
100
001
image ratio, at the expense of i -f selectivity,
1001 by placing the desired signal and the image
Figure 9 J +250V farther apart. To give both good image
ratio at the higher frequencies and good
TYPICAL TRIODE VHF selectivity in the i -f amplifier, a system
AMPLIFIER STAGES
R -F known as double conversion is sometimes
Triode r -f stages contribute the least amount employed. In this system, the incoming signal
of noise output for a given signal level, hence is first converted to a rather high intermed-
their frequent use in the vhf range.
iate frequency, and then amplified and again
converted, this time to a much lower fre-
factor must be obtained, the first r -f stage quency. The first intermediate frequency
usually uses a triode or a low -noise transistor. supplies the necessary wide separation be-
Shown in figure 9 are four commonly used tween the image and the desired signal,
types of triode r -f stages for use in the vhf while the second one supplies the bulk of the
range. The circuit at (A) uses few com- i-f selectivity.
ponents and gives a moderate amount of The double-conversion system, as illus-
gain with very low noise. It is most satis- trated in figure 10, is receiving two general
www.americanradiohistory.com
HANDBOOK R-F Stages 203
t 14 MHz
TUNABLE
114
AMPLIFIER
MIXER
4 MHz
I
TUNABLE
AMPLIFIER
MIXER
455 kHz
ILUSLQ
1I
AMPLIFIER
DEMODULATOR
ANO
AUDIO
CRYSTAL VARIARLE
10MHZ
OSCILLATOR
3545 kHz
OSCILLATOR
r-
I 14 MHz 455 kHz
-11-
SO kHz
1 I
I
TUNABLE FIXED I I
FIXED OEMOOULATOR
I
R.F --1 MIXER MIXER AND
AMPLIFIER AMPLIFIER I I
AMPLIFIER AUDIO
1
IQ I
I I
VARIABLE I! 1X
14.445kHz 505 kHz
OSCILLATOR OSCILLATOR
I!
I I
Figure 10
selectivity from a c tional communications receiver through the use of the double-conversion
superheterodyne principle.
types of application at the present time. The usually is in the 450- to 915 -kHz range) in-
first application is for the purpose of attain- stead of being fed to a demodulator and
ing extremely good stability in a communi- then to the audio system, is alternatively fed
cations receiver through the use of crystal to a fixed -tuned mixer stage and then into a
control of the first oscillator. In such an much lower intermediate- frequency ampli-
arrangement, as used in several types of Col- fier before the signal is demodulated and fed
lins receivers, the first oscillator is crystal to the audio system. The accessory i -f amp-
controlled and is followed by a tunable i -f lifier system (sometimes called a Q5'er)
amplifier which then is followed by a mixer normally is operated on a frequency of 175
stage and a fixed -tuned i -f amplifier on a kHz, 85 kHz, or 50 kHz.
much lower frequency. Through such a cir-
cuit arrangement the stability of the com-
plete receiver is equal to the stability of the
10-6 Signal- Frequency
oscillator which feeds the second mixer, Tuned Circuits
while the selectivity is determined by the
bandwidth of the second fixed i -f amplifier. The signal- frequency tuned circuits in
The second common application of the high- frequency superheterodynes and tuned -
double -conversion principle is for the pur- radio- frequency types of receivers consist of
pose of obtaining a very high degree of se- coils of either the solenoid or universal-
lectivity in the complete communications wound types shunted by variable capacitors.
receiver. In this type of application, as illus- It is in these tuned circuits that the causes
trated in figure 10 B, a conventional com- of success or failure of a receiver often lie.
munications receiver is modified in such a The universal -wound type coils usually are
manner that its normal i -f amplifier (which used at frequencies below 2000 kHz; above
204 Radio Receiver Fundamentals THE RADIO
R -r INPUT C
Ec conventional r -f amplifier tubes is high
enough so that it is not bothersome. But as
the frequency is increased, the input resist-
ance becomes lower and lower, until it ulti-
mately reaches a value so low that no ampli-
fication can be obtained from the r -f stage.
Figure 11
The two contributing factors to the de-
crease in input resistance with increasing
ILLUSTRATING "COMMON- POINT" frequency are the transit time required by
BYPASSING an electron traveling between the cathode
To reduce the detrimental effects of cathode and grid, and the inductance of the cathode
circuit inductance In vhf stages, all bypass lead common to both the plate and grid cir-
capacitors should be returned to the cathode
terminal at the socket. Tubes with two cathode cuits. As the frequency becomes higher, the
leads can give improved performance if the transit time can become an appreciable por-
grid return is made to one cathode terminal tion of the time required by an r-f cycle of
while the plate and screen bypass returns are
made to the cathode terminal which is con- the signal voltage, and current will actually
nected to the suppressor within the tube. flow into the grid. The result of this effect
is similar to that which would be obtained
this frequency the single-layer solenoid type by placing a resistance between the grid and
of coil is more satisfactory. cathode of the tube.
Impedance The two factors of greatest sig- Superheterodyne Because the oscillator in a
and Q nificance in determining the Tracking superheterodyne operates
gain -per-stage and selectivity, re- "offset" from the other
spectively, of a tuned amplifier are tuned- front -end circuits, it is
circuit impedance and tuned -circuit Q. Since necessary to make special provisions to allow
the resistance of modern capacitors is low at the oscillator to track when similar tuning
ordinary frequencies, the resistance usually capacitor sections are ganged. The usual
can be considered to be concentrated in the
coil. The resistance to be considered in mak-
ing Q determinations is the r -f resistance, MIXER
not the d -c resistance of the wire in the coil.
The latter ordinarily is low enough that it PADDING CAPACITOR
may be neglected. The increase in r -f resist- TUNING CAPACITOR
ance over d -c resistance primarily is due to
skin effect and is influenced by such factors
as wire size and type, and the proximity of OSCILLATOR
www.americanradiohistory.com
HANDBOOK Signal- Frequency Tuned Circuits 205
www.americanradiohistory.com
HANDBOOK I -F Tuned Circuits 207
Figure 16
ELECTRICAL EQUIVALENT OF
QUARTZ FILTER CRYSTAL
The crystal is equivalent fo a very largo value
of inductance in series with small values of
capacitance and resistance across the whole
circuit (rep ing holder capacitance plus
stray capacitances).
www.americanradiohistory.com
HANDBOOK I -F Filters 209
CRYSTAL
O -
7 SELECTIVITY
CONTROL
PH SIING
CONTROL
Figure 17 Figure 18
SELECTIVITY
Figure 19
CONTROL
r CRYSTAL NOTCH
VARIABLE SELECTIVITY
CRYSTAL FILTER 8
J
40
www.americanradiohistory.com
HANDBOOK I -F Filters 211
desired phone signal which is very close in ical vibration travels through the resonant
frequency to a desired phone signal. The mechanical section to the output transducer,
filter is adjusted to "broad" so as to permit where it is converted by magnetostriction to
telephone reception, and the receiver tuned an electrical signal which appears at the out-
so that the carrier frequency of the un- put terminals.
desired signal falls in the rejection notch. In order to provide the most efficient elec-
The modulation sidebands of the undesired tromechanical coupling, a small magnet in
signal still will come through, but the car- the mounting above each transducer applies
rier heterodyne will be effectively eliminated a magnetic bias to the nickel transducer core.
and interference greatly reduced. The electrical impulses then add to or sub-
A typical crystal selectivity curve for a tract from this magnetic bias, causing vibra-
communications receiver is shown in fig- tion of the filter elements which corresponds
ure 20. to the exciting signal. There is no mechani-
cal motion except for the imperceptible
Crystal Filter A crystal filter, especially vibration of the metal discs.
Considerations when adjusted for single - Magnetostrictively driven mechanical fil-
signal reception, greatly re- ters have several advantages over electrical
duces interference and background noise, the equivalents. In the region from 100 kHz to
latter feature permitting signals to be copied 500 kHz, the mechanical elements are ex-
that would ordinarily be too weak to be heard tremely small, and a mechanical filter having
above the background hiss. However, when better selectivity than the best of conven-
the filter is adjusted for maximum selectiv- tional i -f systems may be enclosed in a
ity, the passband is so narrow that the re- package smaller than one i -f transformer.
ceived signal must have a high order of sta- Since mechanical elements with Q's of
bility in order to stay within the passband. 5000 or more are readily obtainable, mechan-
Likewise, the local oscillator in the receiver ical filters may be designed in accord with
must be highly stable, or constant retuning the theory for lossless elements. This permits
will be required. Another effect that will be characteristics to be achieved that are unob-
noticed with the filter adjusted too "sharp" tainable with electrical circuits because of
is a tendency for code characters to produce the relatively high losses in electrical ele-
a ringing sound, and have a hangover or
"tails." This effect limits the code speed that
can be copied satisfactorily when the filter
is adjusted for extreme selectivity.
The Mechanical The Collins Mechanical Fil-
Filter ter (figure 21) is a new con-
cept in the field of selectivi-
ty. It is an electromechanical bandpass filter
about half the size of a cigarette package.
As shown in figure 22, it consists of an input
transducer, a resonant mechanical section
comprised of a number of metal discs, and y
an output transducer.
The frequency characteristics of the reso-
nant mechanical section provide the almost
rectangular selectivity curves shown in figure
23. The input and output transducers serve
`'/, G$/m/'%9v ,"
Figure 21
S'''"
0.., ,
only as electrical -to- mechanical coupling de- COLLINS MECHANICAL FILTERS
vices and do not affect the selectivity char-
The Collins Mechanical filter is an electro-
acteristics which are determined by the mechanical bandpass Alter which surpasses, In
metal discs. An electrical signal applied to one small unit, the selectivity of conventional,
the input terminals is converted into a me- space-consuming Alters. At the loft is the
miniaturized Alter, less than 21/4" long. A
chanical vibration at the input transducer vertical design is next, and two horizontal
by means of magnetostriction. This mechan- mounting types are at right.
212
DISC AT
EACH END
Adiek.L,
MAGNETOSTRICTIVE
DRIVING ROD
ELECTRICAL SIGNAL
(INPUT OR OUTPUT)
r
Radio Receiver Fundamentals
ONE SUPPORTING
AT/AV/WANWW1'
1
/
COIL
COUPLING RODS
RIAS MAGNET
RANSDUCER
ELECTRICAL SIGNAL
(INPUT OR OUTPUT)
THE RADIO
Figure 24
www.americanradiohistory.com
HANDBOOK Detector, Audio, and Control Circuits 213
1F STAGE
(- AUDIO changed, as the latter circuits usually change
the frequency of the bfo at the same time
they change the strength, making it nec-
essary to reset the trimmer each time the
5+ e+ output is adjusted.
The bfo usually is provided with a small
OA GRID -LEAK DETECTOR
trimmer which is adjustable from the front
IF STAGE DET. panel to permit adjustment over a range of S
or 10 kHz. For single -signal reception the
bfo always is adjusted to the high- frequency
side, in order to permit placing the hetero-
dyne image in the rejection notch.
In order to reduce the bfo signal output
voltage to a reasonable level which will pre-
AUDIO vent blocking the second detector, the signal
voltage is delivered through a low- capaci-
tance (high- reactance) capacitor having a
OB DIODE DETECTOR value of 1 to 10 -pf.
Care must be taken with the bfo to pre-
vent harmonics of the oscillator from being
AUDIO
picked up at multiples of the bfo frequency.
The complete bfo together with the coupling
circuits to the second detector, should be
thoroughly shielded to prevent pickup of the
harmonics by the input end of the receiver.
If bfo harmonics still have a tendency to
PLATE DETECTOR give trouble after complete shielding and
isolation of the bfo circuit has been accom-
plished, the passage of these harmonics from
I -P STAGE DET.
the bfo circuit to the rest of the receiver can
be stopped through the use of a low -pass
filter in the lead between the output of the
bfo circuit and the point on the receiver
where the bfo signal is to be injected.
Figure 26
Typical circuits for grid -leak, diode, plate Even with the circuit shown, a -c loading
and infinite- impedance detectors are shown can occur unless a very high (S megohms, or
in figure 25. more) value of grid resistor is used in the
Automatic Vol- following audio amplifier stage.
The elements of an automat-
ume Control ic volume control (avc) sys- AVC in In receivers having a beat -
tem are shown in figure 26. BFO Equipped frequency oscillator for the
A dual -diode tube is used as a combination Receivers reception of c -w or SSB
diode detector and avc rectifier. The left - signals, the use of avc can
hand diode operates as a simple rectifier in result in a great loss in sensitivity when the
the manner described earlier in this chapter. bfo is switched on. This is because the beat -
Audio voltage, superimposed on a d -c volt- oscillator output acts exactly like a strong
age, appears across the 500,000-ohm potenti- received signal, and causes the avc circuit to
ometer (the volume control) and the .0001- put high bias on the r -f and i -f stages, thus
pfd capacitor, and is passed on to the audio greatly reducing the receiver's sensitivity.
amplifier. The right -hand diode receives sig- Due to the above effect, it is necessary to
nal voltage directly from the primary of the either isolate the avc voltage or make the avc
last i -f amplifier, and acts as the avc recti- circuit inoperative when the bfo is being
fier. The pulsating d -c voltage across the used. The simplest method of eliminating the
1- megohm avc -diode load resistor is filtered avc action is to short the avc line to ground
by a 500,000-ohm resistor and a .0 S - tfd when the bfo is turned on. A two -circuit
capacitor, and is applied as bias to the grids of switch may be used for the dual purpose of
the r -f and i -f amplifier tubes; an increase or turning on the beat oscillator and shorting
decrease in signal strength will cause a corre- out the avc if desired.
sponding increase or decrease in avc bias
voltage, and thus the gain of the receiver is Signal- StrengthVisual means for determin-
automatically adjusted to compensate for Indicators ing whether or not the re-
changes in signal strength. ceiver is properly tuned, as
well as an indication of the relative signal
A -C Loading of By disassociating the avc and strength, are both provided by means of
Second Detector detecting functions through tuning indicators (S meters) of the meter
the use of separate diodes, as or vacuum -tube type. A d -c milliammeter
shown, most of the ill effects of a -c shunt can be connected in the plate -supply circuit
loading on the detector diode are avoided. of one or more r -f or i -f amplifiers, as shown
This type of loading causes serious distor- in figure 27A, so that the change in plate
tion, and the additional components required current, due to the action of the avc volt-
to eliminate it are well worth their cost. age, will be indicated on the instrument.
www.americanradiohistory.com
HANDBOOK AVC Circuits 215
Lac
commercial receivers using this type of
circuit. Alternatively, the meter may be
mounted upside down, so that the needle
moves toward the right with increased
0-1 D
strength.
The circuit of figure 27B can frequently
be used to advantage in a receiver where the
R. F or I. F cathode of one of the r -f or i -f amplifier
stages runs directly to ground through the
cathode -bias resistor instead of running
O-1 D.C- through a cathode -voltage gain control. In
this case a 0 -1 d -c milliammeter in con-
junction with a resistor of 1000 to 3000
ohms can be used as shown as a signal -
470
strength meter. With this circuit the meter
will read backwards with increasing signal
6U5/6G5
Q OR 6E5 strength as in the circuit previously dis-
cussed.
TO AVC
Figure 27C is the circuit of a forward -
reading S meter often used in communi-
+20v cations receivers. The instrument is used in
an unbalanced bridge circuit with the d -c
Figure 27 plate resistance of one i -f tube as one leg
of the bridge and with resistors for the other
SIGNAL -STRENGTH METER CIRCUITS three legs. The value of resistor R must be
Shown above are four circuits for obtaining a determined by trial and error and will be
signal -strength reading which is a function of somewhere in the vicinity of 50,000 ohms.
incoming carrier amplitude. The circuits are
di d in the accompanying text. Sometimes the screen circuits of the r -f and
i -f stages are taken from this point along
The d -c instrument (MA) should have a with the screen- circuit voltage divider.
full -scale reading approximately equal to Electron -ray tubes (sometimes called mag-
the total plate current taken by the stage or ic eyes) can also be used as indicators of
stages whose plate current passes through relative signal strength in a circuit similar to
the instrument. The value of this current that shown in figure 27D. A 6U5/6G5 tube
can be estimated by assuming a plate current should be used where the avc voltage will be
on each stage (with no signal input to the from 5 to 20 volts and a type 6E5 tube
receiver) of about 6 ma. However, it will should be used when the avc voltage will run
be found to be more satisfactory to measure from 2 to 8 volts.
the actual plate current on the stages with a Audio Amplifiers Audio amplifiers are em-
milliammeter of perhaps 0 -100 ma full scale ployed in nearly all radio
before purchasing an instrument for use receivers. The audio amplifier stage or stages
as an S meter. The 50-ohm potentiometer are usually of the class -A type, although
216 Radio Receiver Fundamentals THE RADIO
class -AB push -pull stages are used in some There are two principal methods for re-
receivers. The purpose of the audio amplifier ducing this noise. They are:
is to bring the relatively weak signal from
the detector up to a strength sufficient to (1) line filters at the source of inter-
a -c
operate a pair of headphones or a loudspeak- ference, if the noise is created by an
er. Either triodes, pentodes, or beam tetrodes electrical appliance; and
may be used, the pentodes and beam tetrodes (2) noise -limiting circuits for the reduc-
usually giving greater output. In some re- tion, in the receiver itself, of inter-
ceivers, particularly those employing grid ference of the type caused by auto-
leak detection, it is possible to operate the mobile ignition systems.
headphones directly from the detector, with-
Power Line Many household appliances, such
out audio amplification. In such receivers, a
Filters as electric mixers, heating pads,
single audio stage with a beam tetrode or
pentode tube is ordinarily used to drive the vacuum cleaners, refrigerators,
loudspeaker. oil burners, sewing machines, doorbells, etc.,
Most communications receivers, either create an interference of an intermittent na-
home -constructed or factory -made, have a ture. The insertion of a line filter near the
single -ended beam tetrode such as a 6V6 or source of interference often will effect a
6AQS in the audio output stage feeding the complete cure. Filters for small appliances
speaker. If precautions are not taken such a can consist of a 0.1 -fd capacitor connected
stage will actually bring about a decrease in across the 120 -volt a -c line. Two capacitors
the effective signal -to -noise ratio of the re- in series across the line, with the midpoint
ceiver due to the rising high- frequency char- connected to ground, can be used in con-
acteristic of such a stage when feeding a junction with industrial heating machines,
speaker. One way of improving this condi- refrigerators, oil- burner furnaces, and other
tion is to place a mica or paper capacitor of more stubborn offenders. In severe cases of
approximately 0.003 /lid capacitance across interference, additional filters in the form of
the primary of the output transformer. The heavy -duty r -f choke coils must be con-
use of a capacitor in this manner tends to nected in series with the 120 -volt a -c line on
make the load impedance seen by the plate both sides of the line right at the interfering
of the output tube more constant over the appliance.
audio- frequency range. The speaker and
transformer will tend to present a rising im-
Peak Noise Numerous noise- limiting circuits
Limiters which are beneficial in overcom-
pedance to the tube as the frequency in-
creases, and the parallel capacitor will tend
ing key clicks, automobile igni-
to make the total impedance more constant tion interference, and similar noise impulses
since it will tend to present a decreasing have become popular. They operate on the
impedance with increasing audio frequency. principle that each individual noise pulse is
A still better way to improve the fre-
of very short duration, yet of very high
quency characteristic of the output stage, amplitude. The popping or clicking type of
noise from electrical ignition systems may
and at the same time reduce the harmonic
produce a signal having a peak value ten to
distortion, is to use shunt feedback from the
plate of the output tube to the plate of a twenty times as great as the incoming radio
tube such as a 6AU6 acting as an audio - signal, but an average power much less than
the signal.
amplifier stage ahead of the output stage.
As the duration of this type of noise peak
10-9 Noise Suppression is short, the receiver can be made inoperative
during the noise pulse without the human
The problem of noise suppression confronts ear detecting the total loss of signal. Some
the listener who is located in places where noise limiters actually punch a hole in the
interference from power lines, electrical ap- signal while others merely limit the maxi-
pliances, and automobile ignition systems is mum peak signal which reaches the head -
troublesome. This noise is often of such in- phones or speaker.
tensity as to swamp out signals from desired The noise peak is of such short duration
stations. that it would not be objectionable except for
www.americanradiohistory.com
HANDBOOK Noise Suppression 217
iST DET. IST I. i. 2ND I.T. silencer circuit. The i -f signal is fed into a
6BE6
double -grid tube, such as a 6BE6, and thence
into the i -f chain. A 6BJ6 high -gain pent-
ode is capacity coupled to the input of the
i -f system. This auxiliary tube amplifies
both signal and noise that is fed to it. It has
a maximum of selectivity ahead of it so that
it receives the true noise pulse before it is
degraded by the i -f strip. A broadly tuned
i -f transformer is used to couple the noise
amplifier to a 6AL5 noise rectifier. The gain
of the noise amplifier is controlled by a po-
tentiometer in the cathode of the 6BJ6
noise amplifier. This potentiometer controls
the gain of the noise amplifier stage and in
addition sets the bias level on the 6AL5
diode so that the incoming signal will not be
rectified. Only noise peaks louder than the
signal can overcome the resting bias of the
6AL5 and cause it to conduct. A noise pulse
Figure 28 rectified by the 6AL5 is applied as a nega-
tive voltage to the control grid of the 6BE6
THE LAMB I -F NOISE SILENCER
i -f tube, disabling the tube, and punching
www.americanradiohistory.com
HANDBOOK Noise Suppression 219
VI V2
LAST -/TUBE
1 irr 6146,6AL5,ETC. o,r AUDIO
C,-0.1 -fd paper
C, -50-pf mica
Ci-1 00-pf mica
C,, C6 -0.01 -pfd paper
R,, R= 1 meg, t/2 watt
R,, R.-220,000 ohms,
1/2 watt
miss,
1/2 watt
R, -2 -meg potentiometer
=c4
Figure 30
It is important that the exact resistance Note that the return for the volume con-
values shown be used, for best results, and trol must be made to the cathode of the de-
that 10- percent tolerance resistors be used tector diode (and not to ground) when a
for R, and R4. Also, the rectified carrier dual tube is used as combined second -de-
voltage developed across C, should be at least tector first -audio. This means that in the cir-
f volts for good limiting. cuit shown in figure 31 a connection will
The limiter will work well on c -w and exist across the points where the "X" is
SSB if the amplitude of beat- frequency shown on the diagram since a common
oscillator injection is not too high. Variable cathode lead is brought out of the tube for
injection is to be preferred, adjustable from D, and V,. If desired, of course, a single
the front panel. If this feature is not pro- dual diode may be used for D, and D_ in
vided, the bfo injection should be reduced to this circuit as well as in the circuit of figure
the lowest value that will give a satisfactory 30. Switching the limiter in and out with
beat. When this is done, effective limiting the switch S brings about no change in vol-
and a good beat can be obtained by proper ume.
adjustment of the r -f and a -f gain controls. In any diode limiter circuit such as the
It is assumed, of course, that the avc is cut ones shown in these two figures it is im-
out of the circuit for c -w telegraphy re- portant that the mid -point of the heater
ception. potential for the noise -limiter diode be as
close to ground potential as possible. This
Alternative The circuit of figure 31 is means that the center -tap of the heater sup-
Limiter Circuit more effective than that
ply for the tubes should be grounded wher-
shown in figure 30 under cer-
ever possible rather than grounding one side
tain conditions and requires the addition of of the heater supply as is often done. Diffi-
only one more resistor and one more capaci- culty with hum pickup in the limiter circuit
tor than the other circuit. Also, this circuit may be encountered when one side of the
involves a smaller loss in output level than
heater is grounded due to the high values of
the circuit of figure 30. This circuit can be
resistance necessary in the limiter circuit.
used with equal effectiveness with a com-
The circuit of figure 31 has been used
bined diode- triode or diode- pentode tube
(6AT6, 6BN8, 6FM8, or similar diode -tri- with excellent success in several home -con-
odes, or 6AS8, 6CR6, 6BW8, or similar structed receivers. It is also used in certain
diode -pentodes) as diode detector and first manufactured receivers.
audio stage. However, a separate diode must An excellent check on the operation of
be used for the noise limiter (D_). This the noise limiter in any communications re-
diode may be one-half of a 6H6, or 6AL5, ceiver can be obtained by listening to the
etc.; it may be a triode connected 6J5, 6C4, Loran signals in the 160 -meter band. With
or similar type, or it may be a high back - the limiter out a sharp rasping buzz will be
resistance diode (1N65 8) , or equivalent. obtained when one of these stations is tuned
220 Radio Receiver Fundamentals THE RADIO
This circuit is of the self -ad-
justing type and gives less
distortion for a given degree
of modulation than the more
common limiter circuits.
R1, R,-070K, '/2 watt
R3-100K, /2 watt
Ro, R,-1 meg, /2 watt
5.-2 -meg potentiometer
C,- 0.00025 mica (approx.)
C, -0.01 -fd paper
C, -0.01 -fd paper
C, -0.01 -fd paper
D,, D, -6116, 6AL5, diode
sections of a 6T6, or
crystal diodes.
Figure 31
in. With the noise limiter switched into the pedance from an ordinary coil and capacitor
circuit the buzz should be greatly reduced used as a resonant circuit. On the other
and a low- pitched hum should be heard. hand, quarter -wavelength sections of parallel
conductors or concentric transmission line
The Full -Wave The most satisfactory diode are not only more efficient but also become
Limiter noise limiter is the series full - of practical dimensions.
wave limiter, shown in figure
32. The positive noise peaks are clipped by Tuning Tubes and tuning capacitors
diode A, the clipping level of which may be Short Lines connected to the open end of a
adjusted to clip at any modulation level be- transmission line provide a ca-
tween 25 and 100 percent. The negative pacitance that makes the resonant length less
noise peaks are clipped by the right -hand than a quarter wavelength. The amount of
diode at a fixed level. shortening for a specified capacitive react-
The TNS Limiter The Twin Noise Squelch, ance is determined by the surge impedance
is a combination of a diode of the line section. It is given by the equa-
noise clipper and an audio squelch tube. The tion for resonance:
squelch circuit is useful in eliminating the
grinding background noise that is the resid-
1
27rfC
- Z tan I
ual left by the diode clipper. In figure 33,
the setting of the 470K potentiometer deter- where,
mines the operating level of the squelch ac-
tion and should be set to eliminate the resid- 7r equals 3.1416,
ual background noise. Because of the low in- f equals the frequency,
herent distortion of the TNS, it may be left C equals the capacitance,
in the circuit at all times. As with other Z equals the surge impedance of the line,
limiters, the TNS requires a high signal level tan l equals the tangent of the electrical
at the second detector for maximum limiting
length in degrees.
effect. The capacitive reactance of the capaci-
tance across the end is 1/ 27rfC ohms.
10-10 Special For resonance, this must equal the surge im-
Considerations pedance of the line times the tangent of its
in UHF Receiver Design electrical length (in degrees, where 90
equals a quarter wave). It will be seen that
Transmission At increasingly higher fre- twice the capacitance will resonate a line if
Line Circuits quencies, it becomes progres- its surge impedance is halved; also that a
sively more difficult to obtain given capacitance has twice the loading ef-
a satisfactory amount of selectivity and im- fect when the frequency is doubled.
www.americanradiohistory.com
HANDBOOK UHF Receiver Design 221
2ND DET.
These should be soldered at only one end and
grounded.
An unbalanced tap on a coaxial resonant
circuit can be made directly on the inner
conductor at the point where it is properly
matched (figure 34). For low impedances,
such as a concentric -line feeder, a small one -
AUDIO
half turn loop can be inserted through a hole
in the outer conductor of the coaxial circuit,
being in effect a half of the hairpin type
recommended for coupling balanced feeders
to coaxial resonant lines. The size of the loop
Figure 32 and closeness to the inner conductor deter-
mines the impedance matching and loading.
THE FULL-WAVE SERIES AUDIO
Such loops coupled in near the shorting disc
NOISE LIMITER
do not alter the tuning appreciably, if they
are not overcoupled.
Coupling Into It is possible to couple into Resonant A cavity is a closed resonant
Lines and a parallel -rod line by tap - Cavities chamber made of metal. The cav-
Coaxial Circuits ping directly on one or both ity, having both inductance and
rods, preferably through capacitance, supersedes coil- capacitor and ca-
blocking capacitors if any d.c. is present. pacitance loaded transmission -line tuned cir-
More commonly, however, a hairpin is induc- cuits at extremely high fequencies where
tively coupled at the shorting -bar end, either conventional L and C components, of even
to the bar or to the two rods, or both. This the most refined design, prove impractical
normally will result in a balanced load. because of the tiny electrical and physical
Should a loop unbalanced to ground be cou- dimensions they must have. Microwave cav-
pled in, any resulting unbalance reflected in- ities have high Q factors and are superior to
to the rods can be reduced with a simple conventional tuned circuits. They may be
Faraday screen, made of a few parallel wires employed in the manner of an obsorption
placed between the hairpin loop and the rods. wavemeter or as the tuned circuit in other
r-f test instruments, and in microwave trans-
mitters and receivers.
I -E STAGE 2ND DE T. -AUDIO
(- AUDIO
Resonant cavities usually are closed on all
sides and all of their walls are made of elec-
trical conductor. However, in some forms,
small openings are present for the purpose of
excitation.
Cavities have been produced in several
shapes including the plain sphere, dimpled
sphere, sphere with re- entrant cones of
various sorts, cylinder, prism (including
cube), ellipsoid, ellipsoid -hyperboloid, dough-
nut- shape, and various re- entrant types. In
appearance, they resemble in their simpler
forms metal boxes or cans.
The cavity actually is a linear circuit, but
one which is superior to a conventional co-
axial resonator in the uhf range. The cavity
resonates in much the same manner as does a
barrel or a closed room with reflecting walls.
Figure 33 Because electromagnetic energy (and the
associated electrostatic energy) oscillates to
THE TNS AUDIO NOISE LIMITER and fro inside them in one mode or another,
222 Radio Receiver Fundamentals THE RADIO
CAVITY CAVITY
LOOP
CONCENTRIC LINE
LINE
O O
Figure 34
GRIDS
HOLE ELECTAMRON
COUPLING AN ANTENNA TO A BE
resonant cavities resemble waveguides. The Wavelength and cavity dimensions (in centi-
mode of operation in a cavity is affected by meters) are related by the following simple
the manner in which microwave energy is in- resonance formulas:
jected. A cavity will resonate to a large
number of frequencies, each being associated for cylinder Ar = 2.6 X radius;
with a particular mode or standing -wave for cube Ar = 2.83 X half of 1 side;
pattern. The lowest mode (lowest frequency for sphere Ar = 2.28 X radius.
of operation) of a cavity resonator normally Butterfly Unlike the cavity resonator, which
is the one used. Circuit in its conventional form is a device
The resonant frequency of a cavity may which can tune over a relatively
be varied, if desired, by means of movable narrow band, the butterfly circuit is a tun-
plungers or plugs, as shown in figure 36A, able resonator which permits coverage of a
or a movable metal disc (figure 36B). A fairly wide uhf band. The butterfly circuit
cavity that is too small for a given wave- is very similar to a conventional coil /vari-
length will not oscillate. able- capacitor combination, except that both
The resonant frequencies of simple spheri- inductance and capacitance are provided by
cal, cylindrical, and cubical cavities may be
calculated simply for one particular mode.
TUNING
SLUGS
Figure 37
www.americanradiohistory.com
HANDBOOK UHF Receiver Design 223
what appears to be a variable capacitor MHz to perhaps 3500 MHz, and above
alone. The Q of this device is somewhat less about 3500 MHz resonant cavities are used
than that of a concentric -line tuned circuit almost exclusively. The resonant cavity is
but is entirely adequate for numerous appli- also quite generally employed in the 2000 -
cations. to 3500 -MHz range.
Figure 37A shows construction of a In a properly designed receiver, thermal
single butterfly section. The butterfly- shaped agitation in the first tuned circuit is ampli-
rotor, from which the device derives its fied by subsequent tubes and predominates
name, turns in relation to the unconvention- in the output. For good signal -to- set -noise
al stator. The two groups of stator "fins" or ratio, therefore, one must strive for a high -
sectors are in effect joined together by a gain low -noise r -f stage. Hiss can be held
semicircular metal band, integral with the down by giving careful attention to this
sectors, which provides the circuit induc- point. A mixer has about 0.3 the gain of an
tance. When the rotor is set to fill the loop r -f tube of the same type; so it is advisable
opening (the position in which it is shown to precede a mixer by an efficient r -f stage.
in figure 37A), the circuit inductance and It is also of some value to have good r -f
capacitance are reduced to minimum. When selectivity before the first detector in order
the rotor occupies the position indicated by to reduce noises produced by beating noise
the dotted lines, the inductance and capaci- at one frequency against noise at another, to
tance are at maximum. The tuning range of produce noise at the intermediate frequency
practical butterfly circuits is in the ratio of in a superheterodyne.
1.5:1 to 3.5:1. The frequency limit of a tube is reached
Direct circuit connections may be made to when the shortest possible external connec-
points A and B. If balanced operation is de- tions are used as the tuned circuit, except
sired, either point C or D will provide the for abnormal types of oscillation. Wires or
electrical midpoint. Coupling may be ef- sizeable components are often best con-
fected by means of a small single -turn loop sidered as sections of transmission lines rather
placed near point E or F. The butterfly thus than as simple resistances, capacitances, or
permits continuous variation of both capaci- inductances.
tance and inductance, as indicated by the So long as small triodes and pentodes will
equivalent circuit in figure 37B, while at the operate normally, they are generally pre-
same time eliminating all pigtails and wiping ferred as vhf tubes over other receiving
contacts. methods that have been devised. However,
Several butterfly sections may be stacked the input capacitance, input conductance,
in parallel in the same way that variable ca- and transit time of these tubes limit the
pacitors are built up. In stacking these sec- upper frequency at which they may be
tions, the effect of adding inductances in operated. The input resistance, which drops
parallel is to lower the total circuit induct- to a low value at very short wavelengths,
ance, while the addition of stators and ro- limits the stage gain and broadens the tun-
tors raises the total capacitance, as well as ing.
the ratio of maximum to minimum capaci-
tance. VHF The first tube in a vhf receiver is
Butterfly circuits have been applied spe- Tubes most important in raising the signal
cifically to oscillators for transmitters, super- above the noise generated in succes-
heterodyne receivers, and heterodyne fre- sive stages, for which reason small vhf types
quency meters in the 100- to 1000 -MHz are definitely preferred.
frequency range. Tubes employing the conventional grid -
controlled and diode rectifier principles have
Receiver The types of resonant circuits de- been modernized, through various expedients,
Circuits scribed in the previous paragraphs for operation at frequencies as high, in some
have largely replaced conventional new types, as 4000 MHz. Beyond that fre-
coil- capacitor circuits in the range above 100 quency, electron transit time becomes the
MHz. Tuned short lines and butterfly cir- limiting factor and new principles must be
cuits are used in the range from about 100 enlisted. In general, the improvements em-
224 Radio Receiver Fundamentals THE RADIO
bodied in existing tubes have consisted of
(1) reducing electrode spacing to cut down SYMIOL i BRASS BASE CONNECTOR
www.americanradiohistory.com
HANDBOOK Receiver Adjustment 225
adjusted. The most practical technique for "neutralizing tools' to adjust the various i-f
making these adjustments is given below. and r-f trimmer capacitors; and (3) some
convenient type of tuning indicator, such as
Instruments A very small number of instru- a copper -oxide or electronic voltmeter.
ments will suffice to check and Throughout the alignment process, unless
align a communications receiver, the most specifically stated otherwise, the r-f gain
important of these testing units being a control must be set for maximum output,
modulated oscillator and a d -c and a -c volt- the beat oscillator switched off, and the avc
meter. The meters are essential in checking turned off or shorted out. When the signal
the voltage applied at each circuit point output of the receiver is excessive, either the
from the power supply. If the a -c voltmeter attenuator or the a -f gain control may be
is of the oxide- rectifier type, it can be used, turned down, but never the r -f gain control.
in addition, as an output meter when con-
nected across the receiver output when tun- I -F Alignment After the receiver has been
ing to a modulated signal. If the signal is a given a rigid electrical and
steady tone, such as from a test oscillator, mechanical inspection, and any faults which
the output meter will indicate the value of may have been found in wiring or the selec-
the detected signal. In this manner, align- tion and assembly of parts are corrected,
ment results may be visually noted on the the i -f amplifier may be aligned as the first
meter. step in the checking operations.
With the signal generator set to give a
TRF Receiver Alignment procedure in a modulated signal on the frequency at which
Alignment multistage trf receiver is ex- the i -f amplifier is to operate, clip the "hot"
actly the same as aligning a output lead from the generator through a
single stage. If the detector is regenerative, small fixed capacitor to the control grid of
each preceding stage is successively aligned the last i -f tube. Adjust both trimmer ca-
while keeping the detector circuit tuned to pacitors in the last i -f transformer (the one
the test signal, the latter being a station between the last i -f amplifier tube and the
signal or one locally generated by a test second detector) to resonance as indicated by
oscillator loosely coupled to the antenna maximum deflection of the output meter.
lead. During these adjustments, the r -f am- Each i -f stage is adjusted in the same
plifier gain control is adjusted for maximum manner, moving the hot lead, stage by stage,
sensitivity, assuming that the r -f amplifier back toward the front end of the receiver
is stable and does not oscillate. Often a sen- and backing off the attenuator as the signal
sitive receiver can be roughly aligned by strength increases in each new position. The
tuning for maximum noise pickup. last adjustment will be made to the first i -f
transformer, with the hot signal generator
Superheterodyne Aligning a superhet is a lead connected to the control grid of the
Alignment detailed task requiring a mixer. Occasionally it is necessary to discon-
great amount of care and nect the mixer grid lead from the coil,
patience. It should never be undertaken grounding it through a 1000- or 5000 -ohm
without a thorough understanding of the resistor, and then couple the signal generator
involved job to be done and then only when through a small capacitor to the grid.
there is abundant time to devote to the When the last i -f adjustment has been
operation. There are no shortcuts; every completed, it is good practice to go back
circuit must be adjusted individually and through the i -f channel, re- peaking all of
accurately if the receiver is to give peak per- the transformers. It is imperative that this
formance. The precision of each adjustment recheck be made in sets which do not include
is dependent on the accuracy with which a crystal filter, and where the simple align-
the preceding one was made. ment of the i -f amplifier to the generator is
Superhet alignment requires (1) a good final.
signal generator (modulated oscillator) cov-
ering the radio and intermediate frequencies I-F with There are several ways of align-
and equipped with an attenuator; (2) the Crystal Filter ing an i -f channel which con-
necessary socket wrenches, screwdrivers, or tains a crystal -filter circuit.
www.americanradiohistory.com
226 Radio Receiver Fundamentals THE RADIO
I-F SIGNAL IN F SIGNAL OUT not be set to zero beat when the receiver is
tuned to resonance with the signal, as this
will cause an equally strong beat to be ob-
tained on both sides of resonance.
Front -End Alignment of the front end of a
Alignment home -constructed receiver is a
relatively simple process, consist-
'r- 1 -F ALONE ing of first getting the oscillator to cover
the desired frequency range and then of
I F PLUS Q MULTIPLIER
peaking the various r -f circuits for maxi-
mum gain. However, if the frequency range
covered by the receiver is very wide a fair
amount of cut and try will be required to
453 kHz
obtain satisfactory tracking between the r -f
FREQUENCY circuits and the oscillator. Manufactured
communications receivers should always be
Figure 39
tuned in accordance with the instructions
The Q- MULTIPLIER given in the maintenance manual.
The loss resistance of a high -Q circuit is neu-
tralized by regeneration in a simple feedback 10-12 Receiving
amplifier. A highly selective passband is pro-
duced which is coupled to the i -f circuit of the Accessories
receiver.
The Q- Multiplier
The selectivity of a receiv-
However, the following method is one which er may be increased by rais-
has been found to give satisfactory results in ing the Q of the tuned circuits of the i -f
every case: An unmodulated signal generator strip. A simple way to accomplish this is to
capable of tuning to the frequency of the add a controlled amount of positive feed-
filter crystal in the receiver is coupled to back to a tuned circuit, thus increasing its
the grid of the stage which precedes the Q. This is done in the Q- multiplier, whose
crystal filter in the receiver. Then, with the
crystal filter switched in, the signal genera-
tor is tuned slowly to find the frequency I -F SIGNAL IN I -F SIGNAL OUT
where the crystal peaks. The receiver "S"
meter may be used as the indicator, and the
sound heard from the speaker will be of as-
sistance in finding the point. When the fre-
quency at which the crystal peaks has been
found, all the i -f transformers in the receiv-
er should be touched up to peak at that
frequency. O MULTIPLIER "NULL"
www.americanradiohistory.com
HANDBOOK Receiving Accessories 227
12AX7
TO PLATE TERMINAL PRODUCT AUDIO OUTPUT
OF FIRST I-F TUOE THRV 6 R -P SIGNAL
DETECTOR
2' OF COAXIAL UNE .001
L2 II
.00s L, 003
DOS
NULL.--. PEAK LOCAL
OSCILLATOR
10 MN
RFC
Figure 42
212AX7 SELECTIVITY
CONTROLS
THE PRODUCT DETECTOR
1.5 R i2 MEG. 0R
Audio output signal is developed only when
local oscillator is on.
- e +2oo-3oo V. - 6.3 V.
VI Vz V3 12AU7 .01
12AU7 + 12AU7 .01 VI + Vz
AUDIO OUT E---AUDIO
100 OUT
47K I-FSIC. 47K
100 K
BEAT - OSC. +
BEAT - OSC. SIGNAL
SIGNAL
Figure 44 Figure 45
6BE6 10K
Figure 46
www.americanradiohistory.com
CHAPTER ELEVEN
Generation of
Radio -Frequency Energy
pA SHUNT -FED HARTLEY OB SHUNT -FED COLPITTS TUNED PLATE TUNED GRID
R R
250
GRID
COIL
Figure 1
divided into two general classes: self -con- Negative -Grid A negative -grid oscillator is
trolled and crystal-controlled. Oscillators essentially a vacuum - tube
There are a great many types of self -con- amplifier with a sufficient
trolled oscillators, each of which is best portion of the output energy coupled back
suited to a particular application. They can into the input circuit to sustain oscillation.
further be subdivided into the classifications The control grid is biased negatively with
of: negative -grid oscillators, electron -orbit respect to the cathode. Common types of
oscillators, negative - resistance oscillators, negative -grid oscillators, which can be used
velocity -modulation oscillators, and mag- as fundamental or harmonic oscillators, are
netron oscillators. diagrammed in figure 1.
www.americanradiohistory.com
HANDBOOK Oscillators 231
The Hartley Illustrated in figure 1 (A) is For best operation of the Hartley and Col -
the oscillator circuit which pitts oscillators, the voltage from grid to
finds the most general application at the cathode, determined by the tap on the coil
present time; this circuit is commonly called or the setting of the two capacitors, normal-
the Hartley. The operation of this oscillator ly should be from 1/3 to 1/5 that appearing
will be described as an index to the opera- between plate and cathode.
tion of all negative -grid oscillators; the only
real difference between the various circuits is The T.P.T.G. The tuned -plate tuned -grid os-
the manner in which energy for excitation is cillator illustrated at (C) has
coupled from the plate to the grid circuit. a tank circuit in both the plate and grid cir-
When plate voltage is applied to the Hart- cuits. The feedback of energy from the plate
ley oscillator shown at (A), the sudden flow to the grid circuits is accomplished by the
of plate current accompanying the applica- plate - to - grid interelectrode capacitance
tion of plate voltage will cause an electro- within the tube. The necessary phase re-
magnetic field to be set up in the vicinity of versal in feedback voltage is provided by
the coil. The building -up of this field will tuning the grid tank capacitor to the low
cause a potential drop to appear from turn side of the desired frequency and the plate
to turn along the coil. Due to the inductive capacitor to the high side. A broadly reso-
coupling between the portion of the coil in nant coil may be substituted for the grid
which the plate current is flowing and the tank to form the T.N.T. (tuned -not tuned)
grid portion, a potential will be induced in oscillator shown at D.
the grid portion.
Since the cathode tap is between the grid Electron -Coupled In any of the oscillator cir-
and plate ends of the coil, the induced grid Oscillators cuits just described it is
voltage acts in such a manner as to increase possible to take energy
further the plate current to the tube. This from the oscillator circuit by coupling an
action will continue for a short period of external load to the tank circuit. Since the
time determined by the inductance and ca- tank circuit determines the frequency of os-
pacitance of the tuned circuit, until the fly- cillation of the tube, any variations in the
wheel effect of the tuned circuit causes this conditions of the external circuit will be
action to come to a maximum and then to coupled back into the frequency -determining
reverse itself. The plate current then de- portion of the oscillator. These variations
creases (the magnetic field around the coil will result in frequency instability.
also decreasing) until a minimum is reached, The frequency- determining portion of
when the action starts again in the original an oscillator may be coupled to the load cir-
direction and at a greater amplitude than be- cuit only by an electron stream, as illustrated
fore. The amplitude of these oscillations, the in (E) and (F) of figure 1. When it is con-
frequency of which is determined by the sidered that the screen of the tube acts as
coil-capacitor circuit, will increase in a very the plate to the oscillator circuit, the plate
short period of time to a limit determined by merely acting as a coupler to the load, then
the plate voltage of the oscillator tube. the similarity between the cathode -grid-
screen circuit of these oscillators and the
The Colpitts Figure 1 (B) shows a version cathode -grid -plate circuits of the corre-
of the Colpilts oscillator. It sponding prototype can be seen.
can be seen that this is essentially the same The electron- coupled oscillator has good
circuit as the Hartley except that the ratio stability with respect to load and voltage
of a pair of capacitances in series determines variation. Load variations have a relatively
the effective cathode tap, instead of actually small effect on the frequency, since the only
using a tap on the tank coil. Also, the net coupling between the oscillating circuit and
capacitance of these two capacitors com- the load is through the electron stream flow-
prises the tank capacitance of the tuned cir- ing through the other elements to the plate.
cuit. This oscillator circuit is somewhat less The plate is electrostatically shielded from
susceptible to parasitic (spurious) oscilla- the oscillating portion by the bypassed
tions than the Hartley. screen.
232 Generation of R -F Energy THE RADIO
The stability of the e.c.o. with respect to tor stability with respect to tube variations.
variations in supply voltages is explained as High -g,,, tubes such as the 6AH6, 5763, and
follows: The frequency will shift in one 6CB6 will permit the use of larger values of
direction with an increase in screen voltage, capacitance at C, and C, than will more
while an increase in plate voltage will cause conventional tubes such as the 6BA6, 6AQ5,
it to shift in the other direction. By a proper and such types. In general it may be said
proportioning of the resistors that comprise that the reactance of capacitors C, and C,
the voltage divider supplying screen voltage,
it is possible to make the frequency of the
oscillator substantially independent of sup-
should be on the order of 40 to 120 ohms at
the operating frequency of the oscillator
with the lower values of reactance going
-
ply voltage variations. with high -gm tubes and the higher values
being necessary to permit oscillation with
The Gait/ A relatively new type of oscilla- tubes having gm in the range of 2000 mi-
Oscillator tor circuit which is capable of cromhos.
giving excellent frequency sta- It will be found that the Clapp oscillator
bility is illustrated in figure 1G. Comparison will have a tendency to vary in power out-
between the more standard circuits of figure put over the frequency range of tuning ca-
IA through IF and the Clapp oscillator cir- pacitor C. The output will be greatest where
cuits of figures 1G and 1H will immediately C is at its largest setting, and will tend to
show one marked difference: the tuned cir- fall off with C at minimum capacitance. In
cuit which controls the operating frequency fact, if capacitors C, and C_2 have too large
in the Clapp oscillator is series resonant, a value the circuit will stop oscillation near
while in all the more standard oscillator cir- the minimum capacitance setting of capaci-
cuits the frequency -controlling circuit is tor C.
parallel resonant. Also, the capacitors C, and Hence it will be necessary to use a slightly
Co are relatively large in terms of the usual smaller value of capacitance at C, and C_
values for a Colpitts oscillator. In fact, the (to provide an increase in the capacitive re-
value of capacitors C, and Co will be in the actance at this point), or else the frequency
vicinity of 0.001 pfd. to 0.0025 pfd. for range of the oscillator must be restricted by
an oscillator which is to be operated in the paralleling a fixed capacitor across C so that
1.8 -MHz band. its effective capacitance at minimum setting
The Clapp oscillator operates in the fol- will be increased to a value which will sus-
lowing manner: at the resonant frequency of tain oscillation.
the oscillator tuned circuit (L, C) the im- In the triode Clapp oscillator, such as
pedance of this circuit is at minimum (since shown at figure 1G, output voltage for ex-
it operates in series resonance) and maxi- citation of an amplifier, doubler, or isolation
mum current flows through it. Note how- stage normally is taken from the cathode of
ever, that C, and C_ also are included within the oscillator tube by capacitive coupling to
the current path for the series -resonant cir- the grid of the next tube. However, where
cuit, so that at the frequency of resonance greater isolation of succeeding stages from
an appreciable voltage drop appears across the oscillating circuit is desired, the electron -
these capacitors. The voltage drop appearing coupled Clapp oscillator diagrammed in fig-
across C, is applied to the grid of the oscilla- ure 1H may be used. Output then may be
tor tube as excitation, while the amplified taken from the plate circuit of the tube by
output of the oscillator tube appears across capacitive coupling with either a tuned cir-
Co as the driving power to keep the circuit cuit, as shown, or with an r -f choke or a
in oscillation. broadly resonant circuit in the plate return.
Capacitors C, and C, should be made as Alternatively, energy may be coupled from
large in value as possible, while still permit- the output circuit L_ -C:, by link coupling
ting the circuit to oscillate over the full tun- to L2.
ing range of C. The larger these capacitors The considerations with regard to C, C2,
are made, the smaller will be the coupling and the grid tuned circuit are the same as
between the oscillating circuit and the tube, for the triode oscillator arrangement of fig-
and consequently the better will be oscilla- ure 1G.
www.americanradiohistory.com
HANDBOOK Oscillators 233
6SK 7
Figure 2
TRANSITRON OSCILLATOR
OA TRANSITRON OSCILLATOR C,- 0.01 -fd mica for r.f. 10-fd elect. for a.f.
C2- 0.00005 -fd mica for r.f. 0.1 -fd paper for a.f.
C3 -0.003 -fd mica for r.f. 0.5 -fd paper for a.f.
C4- 0.01 -fd mica for r.f. 8-fd elect. for a.f.
12U7 R1-220K 1/2 -watt carbon
R2 -1800 ohms 1/2-watt carbon
R3-22K 2 -watt carbon
R4 -22K 2-watt carbon
Negative- Resist- Negative - resistance oscilla- effect similar to the dynatron is obtained in
once Oscillators tors often are used when un- the transitron circuit, which uses a pentode
usually high frequency sta- with the suppressor coupled to the screen.
bility is desired, as in a frequency meter. The The negative resistance in this case is ob-
dynatron of a few years ago and the newer tained from a combination of secondary
transitron are examples of oscillator circuits emission and interelectrode coupling, and is
which make use of the negative- resistance considerably more stable than that obtained
characteristic between different elements in from uncontrolled secondary emission alone
some multigrid tubes. in the dynatron. A representative transitron
In the dynatron, the negative resistance is oscillator circuit is shown in figure 2.
a consequence of secondary emission of elec-
trons from the plate of a tetrode tube. By a
proper proportioning of the electrode volt-
age, an increase in screen voltage will cause
a decrease in screen current, since the in-
creased screen voltage will cause the screen
to attract a larger number of the secondary
electrons emitted by the plate. Since the net
screen current flowing from the screen sup-
ply will be decreased by an increase in screen
voltage, it is said that the screen circuit pre- Figure 3
sents a negative resistance.
If any type of tuned circuit, or even a re- THE FRANKLIN OSCILLATOR CIRCUIT
sistance- capacitance circuit, is connected in A separate phase -inverter tube is used in this
series with the screen, the arrangement will oscillator to feed a portion of the output bask
oscillate-provided, of course, that the ex- to the input in the proper phase to sustain
oscillation. The values of C, and C, should be
ternal circuit impedance is greater than the as small as will permit oscillations to be sus-
negative resistance. A negative- resistance tained over the desired frequency range.
234 Generation of R -F Energy THE RADIO
The chief distinction between a conven- itations on frequency tolerance, several pre-
tional negative -grid oscillator and a negative- cautions are taken to ensure that a variable -
resistance oscillator is that in the former the frequency oscillator will stay on frequency.
tank circuit must act as a phase inverter in The oscillator is fed from a voltage- regulated
order to permit the amplification of the tube power supply, uses a well- designed and tem-
to act as a negative resistance, while in the perature- compensated tank circuit, is of
latter the tube acts as its own phase inverter. rugged mechanical construction to avoid
Thus a negative- resistance oscillator requires the effects of shock and vibration, is pro-
only an untapped coil and a single capacitor tected against excessive changes in ambient
as the frequency -determining tank circuit, room temperature, and is isolated from feed-
and is classed as a two- terminal oscillator. In back or stray coupling from other portions
fact, the time constant of an RC circuit of the transmitter by shielding, filtering of
may be used as the frequency -determining voltage supply leads, and incorporation of
element and such an oscillator is rather one or more buffer -amplifier stages. In a
widely used as a tunable audio- frequency high -power transmitter a small amount of
oscillator. stray coupling from the final amplifier to the
The Franklin The Franklin oscillator makes oscillator can produce appreciable degrada-
Oscillator use of two cascaded tubes to tion of the oscillator stability if both are
obtain the negative- resistance on the same frequency. Therefore, the os-
effect (figure 3) . The tubes may be either a cillator usually is operated on a subharmonic
pair of triodes, tetrodes, or pentodes; a dual of the transmitter output frequency, with
triode; or a combination of a triode and a one or more frequency multipliers between
multigrid tube. The chief advantage of this the oscillator and final amplifier.
oscillator circuit is that the frequency -de-
termining tank only has two terminals, and 11 -2 Quartz Crystal
one side of the circuit is grounded. Oscillators
The second tube acts as a phase inverter to
give an effect similar to that obtained with Quartz is a naturally occurring crystal
the dynatron or transitron, except that the having a structure such that when plates are
effective transconductance is much higher. cut in certain definite relationships to the
If the tuned circuit is omitted or is replaced crystallographic axes, these plates will show
by a resistor, the circuit becomes a relaxa- the piezoelectric effect -the plates will be de-
tion oscillator or a multitrbrator. formed in the influence of an electric field,
Oscillator The Clapp oscillator has proved to and, conversely, when such a plate is com-
Stability be inherently the most stable of all pressed or deformed in any way a potential
difference will appear on its opposite sides.
the oscillator circuits discussed
The crystal has mechanical resonance, and
above, since minimum coupling between the
will vibrate at a very high frequency because
oscillator tube and its associated tuned cir-
of its stiffness, the natural period of vibra-
cuit is possible. However, this inherently
tion depending on the dimensions, the
good stability is with respect to tube varia-
method of electrical excitation, and crystal-
tions; instability of the tuned circuit with
respect to vibration or temperature will of lographic orientation. Because of the piezo-
electric properties, it is possible to cut a
course have as much effect on the frequency
of oscillation as with any other type of oscil- quartz plate which, when provided with
lator circuit. Solid mechanical construction suitable electrodes, will have the characteris-
tics of a series -resonant circuit with a very
of the components of the oscillating circuit,
high LC ratio and very high Q. The Q is
along with a small negative -coefficient com-
pensating capacitor included as an element several times as high as can be obtained with
an inductor -capacitor combination in con-
of the tuned circuit, usually will afford an
adequate degree of oscillator stability. ventional physical sizes. The equivalent elec-
trical circuit is shown in figure 4A, the re-
VFO Transmit- When used to control the fre- sistance component simply being an ac-
ter Controls quency of a transmitter in knowledgement of the fact that the Q, while
which there are stringent lim- high, does not have an infinite value.
www.americanradiohistory.com
HAN DBOOK Crystal Oscillators 235
C,
(SMALL)
c, i L,
(parallel -resonant) frequencies, the latter be-
ing slightly higher than the series- resonant
frequency and approaching it as C2 is in-
(
L,
(LARGE
L, ':ca creased.
SHUNT) The series- resonance characteristic is em-
R,
ployed in crystal filter circuits in receivers
and also in certain oscillator circuits wherein
(SMALL)
- the crystal is used as a selective feedback
element in such a manner that the phase of
Figure 4 the feedback is correct and the amplitude
adequate only at or very close to the series -
EQUIVALENT ELECTRICAL CIRCUIT OF resonant frequency of the crystal.
QUARTZ PLATE IN A HOLDER While quartz, tourmaline, Rochelle salts,
At A is shown the equivalent series -reso- ADP, and EDT crystals all exhibit the
nant circuit of the crystal itself, at 11 is piezoelectric effect, quartz is the material
shown how the shunt capacitance of the holder widely employed for frequency control.
electrodes and associated wiring affects the
circuit to the combination circuit of C which As the cutting and grinding of quartz
exhibits both series resonance and parallel plates has progressed to a high state of devel-
resonance (antiresonance), the separation in
frequency between the two modes being very opment and these plates may be purchased at
small and determined by the ratio of C, to C,. prices which discourage the cutting and
grinding by simple hand methods for one's
The shunt capacitance of the electrodes own use, the procedure will be only lightly
and associated wiring (crystal holder and touched on here.
socket, plus circuit wiring) is represented by The crystal blank is cut from the raw
the dotted portion of figure 4B. In a high - quartz at a predetermined orientation with
frequency crystal this will be considerably respect to the optical and electrical axes, the
greater than the capacitance component of orientation determining the activity, tem-
an equivalent series LC circuit, and unless perature coefficient, thickness coefficient,
the shunt capacitance is balanced out in a and other characteristics. Various orienta-
bridge circuit, the crystal will exhibit both tions or "cuts" having useful characteristics
resonant (series -resonant) and antiresonant are illustrated in figure S.
Figure 5
ION TfV[MTWa
rmplK'[,R
ONAiATMS
PWAMIeNI AMO
MOND ,AIMING
WIN Ta, - MT e
,IGT,NMAL
COWMAN.? NT RlieOi COTSTAL
T OIGLATOIS
A
M. KnOS oeeMeAe
See ThMIIUTYea M. T(MP IOtlPK.,iT,T
(piIPKKNT TP PaTOS
236 Generation of R -F Energy THE RADIO
The crystal blank is then rough -ground made free from spurious responses, and has
almost to frequency, the frequency increas- an excellent temperature characteristic.
ing in inverse ratio to the oscillating dimen- However, above about 6 MHz it becomes
sion (usually the thickness). It is then quite thin and a difficult production job.
finished to exact frequency either by careful Between 6 MHz and about 12 MHz, the
lapping, by etching, or plating. The latter BT -cut plate is widely used. It also works
process consists of finishing it to a frequency well between 500 kHz and 6MHz, but the
slightly higher than that desired and then AT-cut is more desirable when a high order
silver plating the electrodes right on the of stability is desired and no crystal oven is
crystal, the frequency decreasing as the de- employed.
posit of silver is increased. If the crystal is For low- frequency operation on the order
not etched, it must be carefully scrubbed of 100 kHz, such as is required in a fre-
and "baked" several times to stabilize it, or quency standard, the GT-cut crystal is rec-
otherwise the frequency and activity of the ommended, though CT-and DT-cuts also
crystal will change with time. Irradiation by are widely used for applications between 50
X -rays recently has been used in crystal and 500 kHz. The CT-, DT-, and GT-cut
finishing. plates are known as contour cuts, as these
Unplated crystals usually are mounted in plates oscillate along the long dimension of
pressure holders, in which two electrodes are the plate or bar, and are much smaller phys-
held against the crystal faces under slight ically than would be the case for a regular
pressure. Unplated crystals also are some- AT- or BT-cut crystal for the same fre-
times mounted in an air-gap holder in which quency.
there is a very small gap between the crystal
and one or both electrodes. By making this Crystal Holders Crystals normally are pur-
gap variable, the frequency of the crystal chased ready mounted. The
may be altered over narrow limits (about best type mount is determined by the type
0.3% for certain types) . crystal and its application, and usually an
The temperature coefficient of frequency optimum mounting is furnished with the
for various crystal cuts of the "T "- rotated crystal. However, certain features are desir-
family is indicated in figure 5. These angles able in all holders. One of these is exclusion
are typical, but crystals of a certain cut will of moisture and prevention of electrode oxi-
vary slightly. By controlling the orientation dization. The best means of accomplishing
and dimensioning, the turning point (point this is a metal holder, hermetically sealed,
of zero temperature coefficient) for a BT- with glass insulation and a metal -to -glass
cut plate may be made either lower or higher bond. However, such holders are more ex-
than the 75 degrees shown. Also, by careful pensive, and a ceramic or phenolic holder
control of axes and dimensions, it is possible with rubber gasket will serve where require-
to get AT -cut crystals with a very flat ments are not too exacting.
temperature- frequency characteristic.
The first quartz plates used were either Y- Temperature- Control; Where the frequency
cut or X -cut. The former had a very high Crystal Ovens tolerance requirements
temperature coefficient which was discon- are not too stringent
tinuous, causing the frequency to jump at and the ambient temperature does not in-
certain critical temperatures. The X-cut had clude extremes, an AT -cut plate, or a BT-
a moderately bad coefficient, but it was more cut plate with optimum (mean- tempera-
continuous, and by keeping the crystal in a ture) turning point, will often provide
temperature controlled oven, a high order of adequate stability without resorting to a
stability could be obtained. However, the temperature -controlled oven. However, for
X-cut crystal was considerably less active broadcast stations and other applications
than the Y- cut, especially in the case of where very close tolerances must be main-
poorly grounded plates. tained, a thermostatically controlled oven,
For frequencies between 500 kHz and adjusted for a temperature slightly higher
about 6 MHz, the AT -cut crystal now is than the highest ambient likely to be en-
the most widely used. It is active, can be countered, must of necessity be employed.
www.americanradiohistory.com
HANDBOOK Crystal Oscillators 237
tones) .
+5
,00-150 v. e +e
,00 -150V.
By grinding the crystal especially for BASIC "PIERCE" OSCILLATOR HOT -CATHODE "PIERCE'
harmonic operation, it is possible to enhance OSCILLATOR
www.americanradiohistory.com
HANDBOOK Crystal Oscillators 239
002
150
+250 v. +250V
BASIC -PLATE OSCILLATOR RECOMMENDED TUNED -PLATE SPECIAL RCUIT FOR USE WITH
C
OSCILLATOR OVERTONE CUT CRYSTAL.
`.J O
10PF
200PF
200PF
vAniAeLe
1soP
22 R
+250v +250 V.
O
Figure 7
has the disadvantage that both sides of the (1) the circuit will oscillate with crystals
crystal are above ground potential. The Tri- over a very wide frequency range with no
tet circuit of figure 7E is widely used and change other than plugging in or switching
can give excellent output with low crystal in the desired crystal; (2) crystal current is
current. However, the circuit has the dis- extremely low; (3) one side of the crystal
advantages of requiring a cathode coil, of re- is grounded, which facilitates crystal- switch-
quiring careful setting of the variable cath- ing circuits; (4) the circuit will operate
ode capacitor to avoid damaging the crystal straight through without frequency pulling,
when changing frequency ranges, and of or it may be operated with output on the
having both sides of the crystal above second, third, or fourth harmonic of the
ground potential. crystal frequency.
The Colpitts harmonic oscillator of figure
7F is recommended as being the most gener- Crystal Oscillator The tunable circuits of
ally satisfactory harmonic crystal oscillator Tuning all oscillators illustrated
circuit since it has the following advantages: should be tuned for max-
240 Generation of R -F Energy THE RADIO
5763 75 ations become especially important. Keying
-F,2F,3F, of the plate voltage (in the negative lead)
20-is-IO4F. of a crystal oscillator, with the screen volt-
age regulated at about 150 volts, has been
2
found to give satisfactory results.
150
A Versatile 5763 The 5763 tube may be
\
7T.s305
- RFC Crystal Oscillator used in a modified Tri -tet
SAW MIN/DOCTOR
(20LIN)
151( 4.7K,2W crystal oscillator, capable
+300 v. of delivering sufficient power on all bands
NOTES from 160 meters through 10 meters to fully
I. Lo =/s LIN (sf- OF BBW 30/S) drive a pentode tube, such as the 807, 2E26
2. L2' 1.eLN (/ - OFBSW It 3003)
3. FOR /.O METER OPERATION ADO SPFCAP4O7OR
or 6146. Such an oscillator is extremely use-
BETWEEN PINS 1811o90f57s3 PLATE Q7R.55 p(
(2 fOF 84W 3010)
ful for portable or mobile work, since it
4. X7 MHz CRYSTAL FOR HARMONIC OPERATION combines all essential exciter functions in
one tube. The circuit of this oscillator is
Figure 8 shown in figure 8. For 160 -, 80- and 40-
meter operation the 5763 functions as a
ALL -BAND CRYSTAL OSCILLATOR tuned -plate oscillator. Fundamental- frequen-
CAPABLE OF DRIVING BEAM -TETRODE cy crystals are used on these three bands.
TUBE. 6CL6 OR 5763 MAY BE USED For 20 -, 15- and 10 -meter operation the
5763 functions as a Tri -tet oscillator with a
imum output as indicated by maximum ex- fixed -tuned cathode circuit. The impedance
citation to the following stage, except that of this cathode circuit does not affect opera-
the oscillator tank of tuned -plate oscillators tion of the 5763 on the lower frequency
(figure 7A and figure 7B) should be backed bands so it is left in the circuit at all times.
off slightly towards the low capacitance side A 7 -MHz crystal is used for fundamental
from maximum output, as the oscillator then output on 40 meters and for harmonic out-
is in a more stable condition and sure to put on 20, 15, and 10 meters. Crystal cur-
start immediately when power is applied. rent is extremely low regardless of the out-
This is especially important when the oscil- put frequency of the oscillator. The plate
lator is keyed, as for break -in c -w operation. circuit of the 5763 is capable of tuning a
frequency range of 2:1, requiring only
Crystal Switching It is desirable to keep stray two output coils: one for 80- 40 -meter oper-
shunt capacitances in the ation, and one for 20 -, 15-, and 10 -meter
crystal circuit as low as possible, regardless operation. In some cases it may be necessary
of the oscillator circuit. If a selector switch to add 5 picofarads of external feedback ca-
is used, this means that both switch and pacity between the plate and control grid of
crystal sockets must be placed close to the the 5763 tube to sustain oscillation with
oscillator -tube socket. This is especially true sluggish 160 -meter crystals.
of overtone -cut crystals operating on a
comparatively high frequency. In fact, on Triode Overtone The recent development of
the highest frequency crystals it is prefer- Oscillators reliable overtone crystals
able to use a turret arrangement for switch- capable of operation on the
ing, as the stray capacitances can be kept third, fifth, seventh (or higher) overtones
lower. has made possible vhf output from a low -
frequency crystal by the use of a double -
Crystal- OscillatorWhen the crystal oscillator triode regenerative oscillator circuit. Some
Keying is keyed, it is necessary of the twin triode tubes such as the 12AU7,
that crystal activity and 12AV7 and 6J6 are especially satisfactory
oscillator -tube transconductance be moder- when used in this type of circuit. Crystals
ately high, and that oscillator loading and that are ground for overtone service may
crystal shunt capacitance be low. Below be made to oscillate on odd- overtone fre-
2500 kHz and above 6 MHz these consider- quencies other than the one marked on the
www.americanradiohistory.com
HANDBOOK Crystal Oscillators 241
6J6 2F 6J6
4F
5K
X 7 CI C2
F 7-BMMi
I2AU7 6J6
3F 6,9F
3F 6,9F
RFC
F 144MHi La
- +200 V.
Figure 9
crystal holder. A 24 -MHz overtone crystal, The first section of the 6J6 dual triode
for example, is a specially ground 8 -MHz comprises a regenerative oscillator, with out-
crystal operating on its third overtone. In put on either the third or fifth overtone of
the proper circuit it may be made to oscillate the crystal frequency. The regenerative loop
on 40 MHz (fifth overtone), 56 MHz (sev- of this oscillator consists of a capacitance
enth overtone), or 72 MHz (ninth over- bridge made up of C, and C_ with the ratio
tone) Even the ordinary 8 -MHz crystals
. C2 /C, determining the amount of regenera-
not designed for overtone operation may be tive feedback in the circuit. With an 8 -MHz
made to oscillate readily on 24 MHz (third crystal, output from the first section of the
overtone) in these circuits. 6J6 tube may be obtained on either 24 or
A variety of overtone oscillator circuits is 40 MHz, depending on the resonant frequen-
shown in figure 9. The oscillator of figure cy of the plate circuit inductor (L,) The .
9A is attributed to Frank Jones, W6AJF. second half of the 6J6 acts as a frequency
242 Generation of R -F Energy THE RADIO
multiplier, its plate circuit (L2) tuned to will prevent self-oscillation of the circuit.
the sixth- or ninth -harmonic frequency The use of a 144 -MHz overtone crystal is
when L, is tuned to the third overtone, or to illustrated in figure 9F. A 6AB4 or one -half
the tenth- harmonic frequency when L, is of a 12AT7 tube may be used, with output
tuned to the fifth overtone. directly in the 2 -meter amateur band. A
Figure 9B illustrates a Colpitts overtone slight amount of regeneration is provided by
oscillator employing a 6J6 tube. This is an the one turn link, (L2) which is loosely
outgrowth of the Colpitts harmonic oscil- coupled to the 144 -MHz tuned tank circuit
lator of figure 7F. The regenerative loop in (L,) in the plate circuit of the oscillator
this case consists of C,, C2, and RFC be- tube. If a 12AT7 tube and a 110 -MHz
tween the grid, cathode, and ground of the crystal are employed, direct output in the
first section of the 6J6. The plate circuit of 220 -MHz amateur band may be obtained
the first section is tuned to the second, har- from the second half of the 12AT7.
monic of the crystal, and the second section
of the 6J6 doubles to the fourth harmonic 11 -4 Radio- Frequency
of the crystal. This circuit is useful in ob-
taining 28 -MHz output from a 7 -MHz Amplifiers
crystal and is highly popular in mobile work.
The circuit of figure 9C shows a typical The output of the oscillator stage in a
regenerative overtone oscillator employing a transmitter (whether it be self -controlled or
12AU7 double- triode tube. Feedback is con- crystal controlled) must be kept down to a
trolled by the number of turns in L2, and fairly low level to maintain stability and to
the coupling between L2 and L,. Only maintain a factor of safety from fracture of
enough feedback should be employed to the crystal when one is used. The low power
maintain proper oscillation of the crystal. output of the oscillator is brought up to the
Excessive feedback will cause the first sec- desired power level by means of radio-fre-
tion of the 12AU7 to oscillate as a self - quency amplifiers. The two classes of r -f
excited T.N.T. oscillator, independent of the amplifiers that find widest application in
crystal. A variety of this circuit is shown in radio transmitters are the class -B and class -C
figure 9D, wherein a tapped coil, (L,) is types.
used in place of the two separate coils.
Operation of the circuit is the same in either The Class -B Class -B amplifiers are used in a
case, regeneration now being controlled by Amplifier radio- telegraph transmitter when
the placement of the tap on L,. maximum power gain and mini-
A cathode -follower overtone oscillator is mum harmonic output is desired in a par-
shown in figure 9E. The cathode coil (L1) ticular stage. A class -B amplifier operates
is chosen so as to resonate with the crystal with cutoff bias and a comparatively small
and tube capacities just below the third - amount of excitation. Power gains of 20 to
overtone frequency of the crystal. For ex- 200 or so are obtainable in a well -designed
ample, with an 8 -MHz crystal, L3 is tuned class -B amplifier. The plate efficiency of a
to 24 MHz, L, resonates with the circuit class -B c -w amplifier will run around 65
capacities to 23.5 MHz, and the harmonic percent.
tank circuit of the second section of the
12AT7 is tuned either to 48 MHz or 72 The Class -B Another type of class -B ampli -
MHz. If a 24 -MHz overtone crystal is used Linear fier is the class -B linear stage as
in this circuit, L, may be tuned to 72 MHz, employed in radiophone work.
L, resonates with the circuit capacities to 70 This type of amplifier is used to increase the
MHz, and the harmonic tank circuit (L_) is level of a modulated signal, and de-
tuned to 144 MHz. If there is any tendency pends for its operation on the linear relation
towards self -oscillation in the circuit, it may between excitation voltage and output volt-
he eliminated by a small amount of inductive age. Or, to state the fact in another manner,
coupling and between L., and L,. Placing the power output of a class -B linear stage
these coils near each other, with the winding varies linearly with the square of the excita-
of L., correctly polarized with respect to L3 tion voltage.
www.americanradiohistory.com
HANDBOOK R -F Amplifiers 243
The class-B linear amplifier is operated Grid -Modulated If the grid current to a class -
with cutoff bias and a small value of excita- Class-C C amplifier is reduced to a
tion, the actual value of exciting power low value, and the plate
being such that the power output under loading is increased to the point where the
carrier conditions is one - fourth of the peak plate dissipation approaches the rated value,
power capabilities of the stage. Class -B such an amplifier may be grid modulated
linears are very widely employed in broadcast for radiotelephony. If the plate voltage is
and commercial installations, and are com- raised to quite a high value and the stage is
mcn in amateur application for single -side- adjusted carefully, efficiencies as high as 40
band transmitters. Tubes with high plate to 43 percent with good modulation capa-
dissipation are required for moderate output bility and comparatively low distortion may
in this mode. The carrier efficiency of such an be obtained. Fixed bias is required. This type
amplifier will vary from 30 to 35 percent. of operation is termed class-C grid -bias mod-
ulation.
Figure 10
varied with respect to the tuned circuits to methods by which this energy feedback may
obtain the greatest grid drive for allowable be eliminated: the first, and the most com-
values of buffer or doubler plate current. mon method, is through the use of a capaci-
Slight readjustments sometimes can be made tance bridge, and the second method is
after plate voltage has been applied to the through the use of a parallel reactance of
driver tube. equal and opposite polarity to the grid -to-
Excessive grid current damages tubes by plate capacitance, to nullify the effect of
overheating the grid structure; beyond a this capacitance.
certain point of grid drive, no increase in Examples of the first method are shown
power output can be obtained for a given in figure 10. Figure l0A shows a capaci-
plate voltage. tance- neutralized stage employing a balanced
tank circuit. Phase reversal in the tank cir-
11 -5 Neutralization of cuit is obtained by grounding the center of
the tank coil to radio-frequency energy by
R -F Amplifiers capacitor C1. Points A and B are 180 degrees
out of phase with each other, and the cor-
The plate -to -grid feedback capacitance of rect amount of out -of -phase energy is
triodes makes it necessary that they be neu- coupled through the neutralizing capacitor
tralized for operation as r -f amplifiers at fre- (NC) to the grid circuit of the tube. The
quencies above about 500 kHz. Those equivalent bridge circuit of this is shown in
screen -grid tubes, pentodes, and beam figure 11A. It is seen that the bridge is not
tetrodes which have a plate -to -grid capaci- in balance, since the plate- filament capaci-
tance of 0.1 pf or less may be operated as an tance of the tube forms one leg of the bridge,
amplifier without neutralization in a well - and there is no corresponding capacitance
designed amplifier up to 30 MHz. from the neutralizing capacitor (point B)
to ground to obtain a complete balance. In
Neutralizing The object of neutralization is addition, it is mechanically difficult to obtain
Circuits to cancel or neutralize the ca- a perfect electrical balance in the tank coil,
pacitive feedback of energy and the potential between point A and
from plate to grid. There are two general ground and point B and ground, in most
www.americanradiohistory.com
HANDBOOK Neutralization 245
cases, is unequal. This circuit, therefore, decrease, and the voltage from point B to
holds neutralization over a very small oper- ground will increase, both in direct propor-
ating range and unless tubes of low inter - tion to the amount of circuit unbalance.
electrode capacitance are used the inherent The use of this circuit is not recommended
unbalance of the circuit will permit only ap- above 7 MHz, and it should be used below
proximate neutralization. that frequency only with low internal ca-
pacitance tubes.
Split- Stator Figure 10B shows the neu-
Plate Neutrali- tralization circuit which Push -Pull Two tubes of the same type
zotion is widely used in single - Neutralization can be connected for push-
ended r -f stages. The use of pull operation so as to obtain
a split- stator plate capacitor makes the elec- twice as much output as that of a single
trical balance of the circuit substantially in- tube. A push -pull amplifier, such as that
dependent of the mutual coupling within the shown in figure 12 also has an advantage
coil and also makes the balance independent in that the circuit can more easily be bal-
of the place where the coil is tapped. With anced than a single -tube r -f amplifier. The
conventional tubes this circuit will allow various interelectrode capacitances and the
one neutralization adjustment to be made on, neutralizing capacitors are connected in such
for example, 28 MHz, and this adjustment a manner that the reactances on one side of
usually will hold sufficiently close for opera-
tion on all lower- frequency bands.
Capacitor C2 is used to balance out the
plate- filament capacity of the tube to allow
a perfect neutralizing balance at all frequen-
cies. The equivalent bridge circuit is shown
in figure 11B. If the plate -filament capaci-
tance of the tube is extremely low (100TH
triode, for example) , capacitor C2 may be
omitted, or may merely consist of the
residual capacity of NC to ground.
O BRIDGE EQUIVALENT OF FIGURE 10 -A
Grid Neutralization A split grid -tank cir-
cuit may also be used C
www.americanradiohistory.com
246 Generaticn of R -F Energy THE RADIO
www.americanradiohistory.com
HANDBOOK Neutralizing Procedure 247
capacitance. But since, in all ordinary cases detuned slightly from resonance on either
with tubes operating on frequencies for side the grid current on the stage should
which they were designed, the LC ratio of decrease the same amount and without any
the tuned circuit will be very high, the sudden jumps on either side of resonance.
coil can use comparatively small wire, al- This will be found to be a very precise indi-
though it must be wound on air or very low - cation of accurate neutralization in either a
loss dielectric and must be insulated for the triode or beam-tetrode r -f amplifier stage,
sum of the plate r -f voltage and the grid r -f so long as the stage is feeding a load which
voltage. presents a resistive impedance at the oper-
ating frequency.
11 -6 Neutralizing Push -pull circuits usually can be more
completely neutralized than single -ended
Procedure circuits at very high frequencies. In the in-
termediate range of from 3 to 30 MHz,
An r -f amplifier is neutralized to prevent
single -ended circuits will give satisfactory
self -oscillation or regeneration. A neon bulb,
results.
a flashlight bulb and a loop of wire, or a di-
cde voltmeter can be used as a null indicator
for neutralizing low -power stages. The plate Neutralization of Radio- frequency amplifiers
voltage lead is disconnected from the r-f am- Screen -Grid R -F using screen -grid tubes can
plifier stage while it is being neutralized. Amplifiers be operated without any
Normal grid drive then is applied to the r -f additional provision for
stage, the neutralizing indicator is coupled neutralization at frequencies up to about 15
to the plate coil, and the plate tuning capac- MHz, provided adequate shielding has been
itor is tuned to resonance. The neutralizing provided between the input and output cir-
capacitor (or capacitors) then can be ad- cuits. Special vhf screen -grid and beam tet-
justed until minimum r.f. is indicated for rode tubes such as the 2E26, 6146, and 5516
resonant settings of both grid- and plate - in the low -power category and 4E27A,
tuning capacitors. Both neutralizing capaci- 4 -65A, 4 -125A, and 4 -250A in the medium-
tors are adjusted simultaneously and to ap- power category can frequently be operated
proximately the same value of capacitance at frequencies as high as 50 MHz without
when a physically symmetrical push -pull any additional provision for neutralization.
stage is being neutralized. Tubes such as the 807, 7094, and 813 can
A final check for neutralization should be be operated with good circuit design at fre-
made with a d -c milliammeter connected in quencies up to 30 MHz without any addi-
the grid -leak or grid -bias circuit. There will tional provision for neutralization. The 829
be no movement of the meter reading as the tube has been found to require neutralization
plate circuit is tuned through resonance in many cases above 20 MHz although the
(without plate voltage being applied) when 832A tube will operate quite stably at 100
the stage is completely neutralized. MHz without neutralization.
Plate voltage should be completely re- None of these tubes, however, has perfect
moved by actually opening the d -c plate cir- shielding between the grid and the plate, a
cuit. If there is a d -c return through the condition brought about by the inherent in-
plate supply, a small amount of plate current ductance of the screen leads within the tube
will flow when grid excitation is applied, even itself. In addition, unless "watertight" shield-
though no primary a -c voltage is being fed ing is used between the grid and plate cir-
to the plate transformer. cuits of the tube a certain amount of ex-
A further check on the neutralization of ternal leakage between the two circuits is
any r -f amplifier can be made by noting present. These difficulties may not be serious
whether maximum grid current on the stage enough to require neutralization of the stage
comes at the same point of tuning on the to prevent oscillation, but in many instances
plate- tuning capacitor as minimum plate they show up in terms of key- clicks when
current. This check is made with plate volt- the stage in question is keyed, or as parasitics
age on the amplifier and with normal anten- when the stage is modulated. Unless the
na coupling. As the plate tuning capacitor is designer of the equipment can carefully
248 Generation of R -F Energy THE RADIO
E
O
Figure 14
A c tional cross -neutralized circuit for use with push -pull beam tetrodes Is shown at A.
The neutralizing capacitors (NC) usually consist of small plates or rods mounted alongside
the plate elements of the tubes. B and C show grid -neutralized circuits for use with a single-
ended tetrode having either link coupling or capacitive coupling into the grid tank. D shows
a method of tuning the screen -lead inductance to accomplish neutralization in a single -
ended vhf tetrode amplifier, while E shows a method of neutralization by increasing the
grid-to -plate capacitance on a tetrode when the operating frequency is higher than that
frequency where the tetrode is "self- neutralized" as a result of series resonance in the
screen lead. Methods D and E normally are not practicable at frog 'es below about 50
MHz with the usual types of beam tetrode tubes.
check the tetrode stage for miscellaneous quired with beam tetrode tubes. This order
feedback between the grid and plate circuits, of capacitance is far less than can be ob-
and make the necessary circuit revisions to tained with a conventional neutralizing
reduce this feedback to an absolute mini- capacitor at minimum setting, so the neu-
mum, it is wise to neutralize the tetrode just tralizing arrangement is most commonly
as if it were a triode tube. made especially for the case at hand. Most
In most push -pull tetrode amplifiers the common procedure is to bring a conductor
simplest method of accomplishing neutrali- (connected to the opposite grid) in the vi-
zation is to use the cross- neutralized capaci- cinity of the plate itself or of the plate
tance bridge arrangement as normally em- tuning capacitor of one of the tubes. Either
ployed with triode tubes. The neutralizing one or two such capacitors may be used,
capacitances, however, must be very much two being normally used on a higher -fre-
smaller than used with triode tubes, values quency amplifier in order to maintain bal-
of the order of 0.2 pf normally being re- ance within the stage.
www.americanradiohistory.com
HANDBOOK Neutralizing Procedure 249
r-- ,;---
INTERWOi1NO
COILS
C20PF
PER METER .01
WAVELENGTH
Figure 16
www.americanradiohistory.com
HANDBOOK Grounded -Grid Amplifiers 251
nn
K N T
1 (CVTOrr) A u w LI
'
M
,
o n I
-, U\ w T
I 1
HI
N(CUTOFP)- - (
I CxC1TATION
VOLTAGE
`o
Figure 18
Figure 19 Figure 20
PUSH-PUSH FREQUENCY DOUBLER PUSH -PULL FREQUENCY TRIPLER
The output of a doubler stage may be materi- The push -pull tripler is advantageous in the
ally increased through the use of a push -push vhf range since circuit balance is maintained
circuit such as illustrated above. both in the input and output circuits. li the
circuit is neutralized it may be used either as
a straight amplifier or as a tripler. Either
have an angle of flow of 90 degrees or less, triodes or tetrodes may be used; dual -unit
tetrodes such as the 6360, 832A, and 8298 are
tripiers 60 degrees or less, and quadruplers particularly effective in the vhf range.
45 degrees or less. Under these conditions
the efficiency will be on the same order as
can be analyzed, the tripler skipping two ex-
the reciprocal of the harmonic on which the
citation pulses and the quadrupler three. In
stage operates. In other words the efficiency
each case the excitation pulse ideally should
of a doubler will be approximately %z or 50
be short enough that it does not exceed 180
percent, the efficiency of a tripler will be
degrees at the output frequency; otherwise
approximately 1/3 or 33 percent and that of
the excitation actually is bucking the output
a quadrupler will be about 25 percent. With
over a portion of the cycle.
good stage design the efficiency can be some-
In actual practice, it is found uneconomi-
what greater than these values, but as the
cal to provide sufficient excitation to run a
angle of flow is made greater than these
tripler or quadrupler in this fashion. Usually
limiting values, the efficiency falls off
the excitation pulses will be at least 90
rapidly. The reason is apparent from a study
degrees at the exciting frequency, with cor-
of figure 18.
respondingly low efficiency, but it is more
The pulses ABC, EFG, and JKL illustrate
practicable to accept the low efficiency and
180- degree excitation pulses under class -B
build up the output in succeeding amplifier
operation, the solid straight line indicating
stages. The efficiency can become quite low
cutoff bias. If the bias is increased by before the power gain becomes less than
N times, to the value indicated by the dot- unity.
ted straight line, and the excitation increased
until the peak r -f voltage with respect to Push -Push Two tubes can be connected in
ground is the same as before, then the ex- Multipliers parallel to give twice the output
citation frequency can be cut in half and the of a single -tube doubler. If the
effective excitation pulses will have almost grids are driven out of phase instead of in
the same shape as before. The only difference phase, the tubes then no longer work simul-
is that every other pulse is missing; MNO taneously, but rather one at a time. The ef-
simply shows where the missing pulse would fect is to fill in the missing pulses (figure
go. However, if the Q of the plate tank cir- 18) . Not only is the output doubled, but
cuit is high, it will have sufficient flywheel several advantages accrue which cannot be
effect to carry over through the missing obtained by straight parallel operation.
pulse, and the only effect will be that the Chief among these is the effective neutral-
plate input and r -f output at optimum load- ization of the fundamental and all odd har-
ing drop to approximately half. As the input monics, an advantage when spurious emis-
frequency is half the output frequency, an sions must be minimized. Another advantage
efficient frequency doubler is the result. is that when the available excitation is low
By the same token, a tripler or quadrupler and excitation pulses exceed 90 degrees, the
www.americanradiohistory.com
HANDBOOK Frequency Multipliers 253
output and efficiency will be greater than for lent for operation with push-pull beam
the same tubes connected in parallel. tetrodes such as the 6360 and 829B, al-
The same arrangement may be used as a though a pair of tubes such as the 2E26, or
quadrupler, with considerably better effi- 5763 could just as well be used if proper at-
ciency than for straight parallel operation, tention were given to the matter of screen -
because seldom is it practicable to supply lead inductance.
sufficient excitation to permit 45- degree ex-
citation pulses. As pointed out above, the
push -push arrangement exhibits better effi- 11 -9 Tank -Circuit
ciency than a single -ended multiplier when Capacitances
excitation is inadequate for ideal multiplier
operation. It is necessary that the proper value of Q
A typical push -push doubler is illustrated be used in the plate tank circuit of any r -f
in figure 19. When high -transconductance amplifier. The following section has been de-
tubes are employed, it is necessary to employ voted to a treatment of the subject, and
a split- stator grid -tank capacitor to prevent charts are given to assist the reader in the
self -oscillation. With well screened tetrodes determination of the proper LC ratio to be
or pentodes having medium values of trans - used in a radio- frequency amplifier stage.
conductance, a split -coil arrangement with a A class -C amplifier draws plate current in
single- section capacitor may be employed the form of very distorted pulses of short
(the center tap of the grid coil being by- duration. Such an amplifier is always oper-
passed to ground). ated into a tuned inductance -capacitance or
tank circuit which tends to smooth out these
Push -Pull Frequency It is frequently desir- pulses, by its storage or tank action, into a
Triplers able in the case of uhf sine wave of radio- frequency output. Any
and vhf transmitters waveform distortion of the carrier frequency
that frequency multiplication stages be bal- results in harmonic interference in higher -
anced with respect to ground. Further it is frequency channels.
just as easy in most cases to multiply the A class -A r -f amplifier would produce a
crystal or vfo frequency by powers of three sine wave of radio- frequency output if its
rather than multiplying by powers of two as exciting waveform were also a sine wave.
is frequently done in lower- frequency trans- However, a class -A amplifier stage converts
mitters. Hence the use of push -pull tripiers its d -c input to r -f output by acting as a
has become quite prevalent in both commer- variable resistance, and therefore heats con-
cial and amateur vhf and uhf transmitter siderably. A class -Bor -C amplifier driven
designs. Such stages are balanced with re- hard with short pulses at the peak of the
spect to ground and appear in construction exciting waveform acts more as an electronic
and on paper essentially the same as a push - switch, and therefore can convert its d -c
pull r -f amplifier stage with the exception input to r -f output with relatively good
that the output tank circuit is tuned to three efficiency. Values of plate- circuit efficiency
times the frequency of the grid -tank circuit. from 65 to 85 percent are common in class -
A circuit for a push -pull tripler stage is C amplifiers operating under optimum con-
shown in figure 20. ditions of excitation, grid bias, and loading.
A push -pull tripler stage has the further
advantage in amateur work that it can also Tank Circuit Q As stated before, the tank
be used as a conventional push -pull r -f am- circuit of a class -C amplifier
plifier merely by changing the grid and plate receives energy in the form of short pulses
coils so that they tune to the same fre- of plate current which flow in the amplifier
quency. This is of some advantage in the tube. But the tank circuit must be able to
case of operating the 50 -MHz band with 50- store enough energy so that it can deliver a
MHz excitation, and then changing the current essentially sine wave in form to the
plate coil to tune to 144 MHz for operation load. The ability of a tank to store energy in
of the stage as a tripler from excitation on this manner may be designated as the effec-
48 MHz. This circuit arrangement is excel- tive Q of the tank circuit. The effective cir-
254 Generation of R -F Energy THE RADIO
ating conditions on the class -BIC tube. This
load impedance may be obtained from the
DYNAMIC
CHARACTER IS following expression, which is true in the
T
-0+
p O
Li general case of any class -B /C amplifier:
RL - 2
E p m2
Np 1,, Ebb
Q = eRL
=
Rt,
XL
where,
RI, is the resonant impedance of the tank,
Xe
XI,
is
is
the reactance of the tank capacitor,
the reactance of the tank coil.
1111
' ,
11111111111111111111111111
1
11111111161161i:ii~..M!
1116111111111111
5 U 201025 30
TANK CIRCUIT Q
as shown in figure 21.
The value of load impedance (RI,) which Figure 22
the class -B /C amplifier tube sees may be ob-
tained, looking in the other direction from RELATIVE HARMONIC OUTPUT
the tank coil, from a knowledge of the oper- PLOTTED AGAINST TANK CIRCUIT Q
www.americanradiohistory.com
HANDBOOK Tank Circuits 255
\111111
\I
20000
Q= 2
5000
__- ME UT BALI
II
ZI NG
W
_ 1\
1\1111111M11IIIIII
BIBB.BBIBIBBfBBBBI
B)
mum
COIL
RFC
re
< lII
o 1m\11m1n\11111
<
\111!I\1\11111
il CI
II
'I19i1\\11111l
tr
10 20
H!iiIHi!!i
TOTAL CAPACITANCE ACROSS LC CIRCUIT (C1)
50 WO 200 500 1000 2000
Figure 23
Illill\1IIII
\-
o
Q1211II
\111,111\11111111011111111
10 ..
oo0i
.
BIIi
mu
w
UZ
w
i` a5
J CC
s
.
`
,'
B;;I
III.
IIII
IIII
,111III
O J
U iIh!Id;!uIiIIIMuuhuIII
< <3
w
11IIII\1IIII1\\1111111
d
2 IIIIIilI'11I1111 11IIIIII ,
2 3 5 7 10
' ,1. 1II,
11,,II 20 70 50
I
tube is approximately equal to one -half the to the tube, with the expression for tank
d -c load resistance which the class -C stage Q given in a previous paragraph we have the
presents to the power supply (and also to the following expression which relates the react-
modulator in case high -level modulation of ance of the tank capacitor or coil to the d -c
the stage is to be used). input to the class -B /C stage:
Combining the above simplified expression Rd.c.
for the r -f impedance presented by the tank Xe=Xr, 2Q
www.americanradiohistory.com
256 Generation of R -F Energy THE RADIO
The foregoing expression is the basis of the harmonic energy, so that additional harmon-
usual charts giving tank capacitance for the ic filtering circuits external to the amplifier
various bands in terms of the d -c plate volt- proper must be used if increased attenuation
age and current to the class -B /C stage, includ- of higher -order harmonics is desired. The
ing the charts of figure 23, figure 24, and curves also show that push -pull amplifiers
figure 25. may be operated at Q values of 6 or so, since
the second harmonic is cancelled to a large
Harmonic Radia- The problem of harmonic extent if there is no unbalanced coupling be-
tion versus Q radiation from transmitters tween the output tank circuit and the an-
has long been present, but tenna system.
it has become critical during the past decades
along with the extensive occupation of Capacity Charts for Figures 23, 24, and 2S
the vhf range. Television signals are particu- Optimum Tank Q lustrate the correct value
larly susceptible to interference from other of tank capacitance for
signals falling within the passband of the re- various circuit configurations. A Q value
ceiver, so that the TVI problem has received of 12 has been chosen as optimum for single -
the major emphasis of all the services in the ended circuits, and a value of 6 has been
vhf range which are susceptible to interfer- chosen for push -pull circuits. Figure 23 is
ence from harmonics of signals in the hf or used when a single -ended stage is employed,
lower -vhf range. and the capacitance values given are for the
Inspection of figure 22 will show quickly total capacitance across the tank coil. This
that the tank circuit of an r-f amplifier value includes the tube interelectrode ca-
pacitance (plate to ground), coil distributed
should have an operating Q of 12 or greater
capacitance, wiring capacitance, and the
to afford satisfactory rejection of second - value of any low- inductance plate -to- ground
harmonic energy. The curve begins to bypass capacitor as used for reducing har-
straighten out above a Q of about 15, so monic generation, in addition to the actual
that a considerable increase in Q must be "in -use" capacitance of the plate tuning ca-
made before an appreciable reduction in sec- pacitor. Total circuit stray capacitance may
ond- harmonic energy is obtained. Above a vary from perhaps S picofarads for a vhf
circuit Q of about 10 any increase will not stage to 30 picofarads for a medium -power
afford appreciable reduction in the third- tetrode h -f stage.
1,\ l'Il11\111111111
111111IIIII
; 111 ,I
W
W
. .
\ 1 imu
ll...1III
IL MIR
\\Cllil\Ilill11111
\111111\11111111111
11
I\ 1"
I11111
a sssssisssstsinImLIss.sisvssssssssinm
s , Il1llli\111I1i211111111
IIIIIiiLiM!iiiiiIIOI
2 3 S 7 10 20 30 30 CO 20D 300 1000
CORRECT VALUES OF TANK CIRCUIT CAPACITANCE (C ) FOR OPERATING O OF 8 WITH PUSH -PULL
TANK CIRCUITS
Figure 25
www.americanradiohistory.com
HANDBOOK Tank Circuits 257
When a split plate -tank coil is employed series -fed. The important thing in establish-
in the stage in question, the graph of figure ing the operating Q of the tank circuit is
24 should be used. The capacitance read the ratio of the loaded resonant impedance
from the graph is the total capacitance across its terminals to the reactance of the L
across the tank coil. If the split- stator tun- and the C which make up the tank.
ing capacitor is used, each section of the Due to the unknowns involved in deter-
capacitor should have a value of capacitance mining circuit stray capacitances it is some-
equal to twice the value indicated by the times more convenient to determine the
graph. As in the case of figure 23, the values value of L required for the proper circuit Q
of capacitance read on the graph of figure (by the method discussed earlier in this Sec-
24 include all residual circuit capacitances. tion) and then to vary the tuned -circuit
For push-pull operation, the correct values capacitance until resonance is reached. This
of tank circuit capacitance may be deter- method is most frequently used in obtaining
mined with the aid of figure 25. The capaci- proper circuit Q in commercial transmitters.
tance values obtained from figure 25 are The values of R for using the charts are
the effective values across the tank circuit, easily calculated by dividing the d -c plate -
and if a split- stator tuning capacitor is used, supply voltage by the total d -c plate current
each section of the capacitor should have a (expressed in amperes). Correct values of
value of capacitance equal to twice the value total tuning capacitance are shown in the
indicated by the graph. As in the case of chart for the different amateur bands. The
figures 23 and 24, the values of capacitance shunt stray capacitance can be estimated
read on the graph of figure 25 include all closely enough for all practical purposes.
residual circuit capacitances. The coil inductance should then be chosen
The tank circuit operates in the same man- which will produce resonance at the desired
ner whether the tube feeding it is pentode, frequency with the total calculated tuning
beam tetrode, neutralized triode, grounded - capacitance.
grid triode; whether it is single -ended or Effect of Load- The Q of a circuit depends
push -pull; or whether it is shunt -fed or ing on Q on the resistance in series
with the capacitance and in-
Figure 26
ductance. This series resistance is very low
USUAL BREAKDOWN RATINGS OF for a low -loss coil not loaded by an antenna
COMMON PLATE SPACINGS
Air -gap in Peak voltage circuit. The value of Q may be from 100 to
inches breakdown 600 under these conditions. Coupling an an-
.030 1000
.050 2000 tenna circuit has the effect of increasing the
.070 3000
.100 4000 series resistance, though in this case the pow-
.125
.150
4500
5200
er is consumed as useful radiation by the an-
.170 6000 tenna. Mathematically, the antenna increases
.200
.250
7500
9000 the value of R in the expression Q = toL /R
.350 11,000 where L is the coil inductance in micro -
.500 15,000
.700 20,000 henrys and w is the term 27r (f being in
Recommended air -gap for use when no d -e MHz) .
voltage appears a plate tank capacitor
(when plate circuit is shunt fed, or when the The coupling from the final tank circuit
plate tank capacitor is Insulated from to the antenna or antenna transmission line
ground). can be varied to obtain values of Q from
D-C plate Plate perhaps 3 at maximum coupling to a value
voltage C-W mod.
400 .030 .050
of Q equal to the unloaded Q of the circuit
600 .050 .070 at zero antenna coupling. This value of un-
750 .050 .084
loaded Q can be as high as 500 or 600, as
1000 .070 .100
1250 .070 .144 mentioned in the preceding paragraph. How-
.200
ever, the value of Q = 12 will not be ob-
1500 .078
2000 .100 .250
2500 .175 .375 tained at values of normal d -c plate current
3000 .200 .500
3500 .250 .600 in the class -C amplifier stage unless the
Spacings should be multiplied by 1.5 for same C -to-L ratio in the tank circuit is correct
safety factor when d-e voltage appears across
plate tank capacitor. for that frequency of operation.
258 Generation of R -F Energy THE RADIO
Tuning Capacitor To determine the required RP RA(Q2+1)(CxACT)
Air Gap tuning -capacitor air gap
RP Q2 RA (APPROX.)
for a particular amplifier
circuit it is first necessary to estimate the Q.
XC
L
peak r -f voltage which will appear between RI XLC
the plates of the tuning capacitor. Then, RP' APRO%. PLATC YOLTAGr
- PLATCCURRLLNT
**
using figure 26, it is possible to estimate the +e Re. 225 RA
FOR OPERATING CIRCUIT
plate spacing which will be required. Qor15; X
The instantaneous r -f voltage in the plate xL
circuit of a class -C amplifier tube varies
from nearly zero to nearly twice the d -c
plate voltage. If the d -c voltage is being 100 Figure 27
percent modulated by an audio voltage, the THE L- NETWORK IMPEDANCE
r -f peaks will reach nearly four times the TRANSFORMER
d -c voltage.
These rules apply to a loaded amplifier or The L- network is useful with a moderate
operating Q for high values of impedance
buffer stage. If either is operated without an transformation, and it may be used for appli-
r -f load, the peak voltages will be greater cations other than in the plate circuit of a
and can exceed the d -c plate supply voltage. tube with relatively low values of operating
Q for moderate impedance transformations.
For this reason no amplifier should be oper- Exact and approximate design equations are
ated without load when anywhere near nor- given.
mal d -c plate voltage is applied.
If a plate blocking capacitor is used, it to (Q2+1). The operating Q may be rela-
must be rated to withstand the d -c plate tively low (perhaps 3 to 6) in a matching
voltage plus any audio voltage. This capaci- net work between the plate tank circuit of
tor should be rated at a d -c working voltage an amplifier and a transmission line; hence
of at least twice the d-c plate supply in a impedance transformation ratios of 10 to 1
plate- modulated amplifier, and at least equal and even lower may be attained. But when
to the d -c supply in any other type of r -f the network also acts as the plate tank cir-
amplifier. cuit of the amplifier stage, as in figure 27,
11 -10 L- and Pi- Matching the operating Q should be at least 12 and
preferably 15. An operating Q of 15 repre-
Networks sents an impedance transformation of 225;
The L -and pi-networks often can be put this value normally will be too high even
to advantageous use in accomplishing an im- for transforming from the 2000- to 10,000 -
pedance match between two differing im- ohm plate impedance of a class -C amplifier
pedances. Common applications are the stage down to a 50 -ohm transmission line.
matching between a transmission line and However, the L- network is interesting
an antenna, or between the plate circuit of since it forms the basis of design for the pi-
a single -ended amplifier stage and an anten- network. Inspection of figure 27 will show
na transmission line. Such networks may be that the L- network in reality must be con-
used to accomplish a match between the sidered as a parallel- resonant tank circuit in
plate tank circuit of an amplifier and a which RA represents the coupled -in load re-
transmission line, or they may be used to sistance; only in this case the load resistance
match directly from the plate circuit of an is directly coupled into the tank circuit
amplifier to the line without the requirement rather than being inductively coupled as in
for a tank circuit -provided the network is the conventional arrangement where the load
designed in such a manner that it has suffi- circuit is coupled to the tank circuit by
cient operating Q for accomplishing har- means of a link. When R:.. is shorted, L and C
monic attenuation. comprise a conventional parallel- resonant
The L-Matching The L- network is of limited tank circuit, since for proper operation L
Network utility in impedance match- and C must be resonant in order for the net-
ing since its ratio of imped- work to present a resistive load to the class -C
ance transformation is fixed at a value equal amplifier.
www.americanradiohistory.com
H ANDBOOK L and Pi Networks 259
The Pi- Network The pi impedance -matching as the two knowns-or the actual values of
network, illustrated in figure the capacitance may be obtained for an op-
28, is much more general in its application erating Q of 12 by reference to figures 23,
than the L network since it offers greater 24 and 2S.
harmonic attenuation, and since it can be The inductive arm in the pi-network can
used to match a relatively wide range of be thought of as consisting of two induct-
impedances while still maintaining any de- ances in series, as illustrated in figure 28.
sired operating Q. The values of C, and L, The first portion of this inductance (L,)
in the pi_ network of figure 28 can be thought is that value of inductance which would
of as having the same values of the L net- resonate with C, at the operating frequency
work in figure 27 for the same operating Q, -the same as in a conventional tank circuit.
but, what is more important from the com- However, the actual value of inductance in
parison standpoint these values will be about this arm of the pi- network, Lrot will be
the sanie as in a conventional tank circuit. greater than L, for normal values of imped-
The value of the capacitance may be de- ance transformation. For high transformation
termined by calculation with the operating ratios LTot will be only slightly greater than
Q and the load impedance which should be L,; for a transformation ratio of 1.0, trot
reflected to the plate of the class -C amplifier will be twice as great as L,. The amount of
COAx
OUTPUT
Estimated Plate
Load (ohms) 1000 1500 2000 2500 3000 3500 4000 4500 5000 6000` Notes
C, in pf, 3.5 MHz 520 360 280 210 180 155 135 120 110 90 The actual capacitance setting
7 260 180 140 105 90 76 68 60 56 45 for C, equals the value in this
14 130 90 70 52 45 38 34 30 28 23 table minus the published tube
21 85 60 47 35 31 25 23 20 19 15 output capacitance. Air gap
28 65 45 35 26 23 19 17 15 14 11 approx. 10 mils/100 y E,,.
L in h, 3.5 MHz 4.5 6.5 8.5 10.5 12.5 14 15.5 18 20 25 Inductance values are for a
7 2.2 3.2 4.2 5.2 6.2 7 7.8 9 10 12.5 50-ohm load. For a 70-ohm
14 1.1 1.6 2.1 2.6 3.1 3.5 3.9 4.5 5 6.2 load, values are approx. 3%
21 0.73 1.08 1.38 1.7 2.05 2.3 2.6 3 3.3 4.1 higher.
28 0.55 0.8 1.05 1.28 1.55 1.7 1.95 2.5 2.5 3.1
C2 in pf, 3.5 MHz 2400 2100 1800 1550 1400 1250 1100 1000 900 700 For 50-ohm transmission line.
7 1200 1060 900 760 700 630 560 .500 460 350 Air gap for C2 is approx. 1
14 600 530 450 380 350 320 280 250 230 175 mil /100 v E,,.
21 400 350 300 250 230 210 185 165 155 120
28 300 265 225 190 175 160 140 125 115 90
Ca in pf, 3.5 MHz 1800 1500 1300 1100 1000 900 800 720 640 500 For 70-ohm transmission line.
7 900 750 650 560 500 450 400 360 320 250
14 450 370 320 280 250 220 200 180 160 125
21 300 250 215 190 170 145 130 120 110 85
28 225 185 160 140 125 110 100 90 80 65
Values given ore approximations All components shown in Table I are for a Q of 12. For other values of Q, use
Q. C, Q. Ln
Qn = e =
Qt,
When the estimated plate load is higher than 5000 ohms, it is recommended that the
components be selected for a circuit Q between 20 and 30.
Table 1. Components for Pi- Coupled Final Amplifiers (class AB, B, and C)
260 Generation of R -F Energy THE RADIO
power level and the impedance to be fed. If
a S0 -ohm coaxial line is to be fed from the
pi-network, receiving -type capacitors will be
satisfactory even up to the power level of a
plate -modulated kilowatt amplifier. In any
event, the peak voltage which will be im-
pressed across the output capacitor is ex-
pressed by:
EPkz = 2RBWo
Roc. Ebb XC2RA RP
lb RA(Q2+1)R where,
Rear Rpc, the peak voltage across the capacitor,
EPk is
XL2
RA2 XC2 Ra isthe value of resistive load which the
X C, Ql- RA2+XC22 network is feeding,
W is the maximum value of the average
XL,= Q- XLTOT.'XLI+XL2 power output of the stage.
should be removed from the circuit and re- C2- ONE -HALF TO TWO -THIRDS THAT
VALUE OF C2 GIVEN IN TABLE I
placed by a smaller coil which has been LI - 1.25 TIMES THAT VALUE OF L,
especially designed for the higher frequency GIVEN IN TABLE I
L2- ONE -THIRD VALUE OF L,, ABOVE
ranges.
The peak voltage rating of the output or
loading capacitor (CO is determined by the Table 2. The Pi -L Network
www.americanradiohistory.com
HANDBOOK Grid Bias 261
Figure 29
This linear amplifier makes use of a pi -L network in the plate circuit. The large vertical
coil is the main portion of the pi section, with the 10 -meter coil placed horizontally in front
of it. The L section is placed beneath the chassis in the recessed area. A three -deck
bandswitch is used one deck for the Pi coil, one deck for the L coil and the third deck for
additional 80 -meter loading capacitance. A small series -tuned circuit (adjusted to TV channel
2 or 3) is placed across the coaxial antenna receptacle to provide additional harmonic
protection at this band of frequencies.
The Pi -L Network The pi-L network shown value required for the equivalent pi- circuit
in Table II will provide 10 capacitor, and increasing the voltage rating
to 15 db more attenuation of the second by a factor of three over that minimum rat-
harmonic than will the pi, and even more ing established for the pi- capacitor. The pi-
attenuation to the higher harmonics. A pi- section coil (L1) will have an inductance
network may be converted to the pi -L con- about 1.2 times that of its pi- circuit
S
figuration by reducing the loading capacitor counterpart (coil L, Table I). The L- section
(C2) to about one -half to two -thirds that coil (L2) has no equivalent in the pi- circuit
262 Generation of R -F Energy THE RADIO
and should be about one-third the induct- voltage. This procedure will ensure that the
ance of the pi- section coil (L1) as deter- tube is operating at a bias greater than cut-
mined above. A formal calculation of the off when the plate voltage is doubled on
pi -L circuit parameters is given in the article positive modulation peaks. C -w telegraph
"The Pi -L Plate Circuit in Kilowatt Am- and f -m transmitters can be operated with
plifiers ", by Rinaudo, QST, July 1962. (A bias as low as cutoff, if only limited excita-
free reprint of this article may be obtained tion is available and moderate plate efficiency
by writing to: Amateur Service Department, is satisfactory. In a c -w transmitter, the bias
EIMAC Division of Varian, San Carlos, supply or resistor should be adjusted to the
California) . point which will allow normal grid current
to flow for the particular amount of grid
11 -11 Grid Bias driving r -f power available. This form of
adjustment will allow more output from the
Radio- frequency amplifiers require some underexcited r -f amplifier than when higher
form of grid bias for proper operation. Prac- bias is used with corresponding lower values
tically all r -f amplifiers operate in such a of grid current. In any event, the operating
manner that plate current flows in the form bias should be set at as low a value as will
of short pulses which have a duration of give satisfactory operation, since harmonic
only a fraction of an r -f cycle. To accom- generation in a stage increases rapidly as the
plish this with a sinusoidal excitation volt- bias is increased.
age, the operating grid bias must be at least
Self Bias A resistor can be connected in
sufficient to cut off the plate current. In
very high efficiency class -C amplifiers the the grid circuit of a class-C ampli-
operating bias may be many times the cutoff fier to provide self bias. This resistor (R,
value. Cutoff bias, it will be recalled, is that in figure 30), is part of the d -c path in the
value of grid voltage which will reduce the grid circuit.
plate current to zero at the plate voltage The r -f excitation applied to the grid cir-
employed. The method for calculating it has cuit of the tube causes a pulsating direct cur-
been indicated previously. This theoretical rent to flow through the bias supply lead,
value of cutoff will not reduce the plate cur- due to the rectifying action of the grid, and
rent completely to zero, due to the variable- any current flowing through R, produces a
tendency or "knee" which is characteristic voltage drop across that resistor. The grid of
of all tubes as the cutoff point is approached. the tube is positive for a short duration of
each r -f cycle, and draws electrons from the
Class -C BiasAmplitude - modulated class -C filament or cathode of the tube during that
amplifiers should be operated time. These electrons complete the circuit
with the grid bias adjusted to a value great- through the d -c grid return. The voltage
er than twice cutoff at the operating plate drop across the resistance in the grid return
provides a negative bias for the grid.
Self bias automatically adjusts itself over
fairly wide variations of r -f excitation. The
I' ROM DRIVCR value of grid resistance should be such that
normal values of grid current will flow at
the maximum available amount of r-f ex-
citation. Self bias cannot be used for grid -
Figure 30 modulated or linear amplifiers in which the
average d -c current is constantly varying
SELF BIAS with modulation.
The grid resistor on an amplifier or multiplier
stag. may also be used as the shunt food Safety Bias Self bias alone provides no pro-
impedance to the grid of the tube when a tection against excessive plate
high value of resistor (greater than perhaps
20,000 ohms) is used. When a lower value of current in case of failure of the source of r -f
grid resistor is to be employed, an r -f choke grid excitation. A C- battery or C -bias supply
should be used between the grid of the tub*
and the grid resistor to r.duce r -f losses in can be connected in series with the grid resis-
the grid resistance. tor as shown in figure 31. This fixed protec-
www.americanradiohistory.com
HANDBOOK Protective Circuits 263
PROM
DRIVER
Figure 33
www.americanradiohistory.com
HANDBOOK Interstage Coupling 265
www.americanradiohistory.com
266 Generation of R -F Energy THE RADIO
Unity Coupling If the grid- tuning capacitor
of figure 39 is removed and
the coupling increased to the maximum
practicable value by interwinding the turns
of the two coils, the circuit insofar as r.f.
is concerned, acts like that of figure 37, in
which one tank serves both as plate tank
for the driver and grid tank for the driven
stage. The interwound grid winding serves
Figure 39 simply to isolate the d -c plate voltage of the
INDUCTIVE INTERSTAGE COUPLING driver from the grid of the driven stage, and
to provide a return for d -c grid current.
This type of coupling, illustrated in figure
half. This type of coupling is shown in 40, is commonly known as unity coupling.
figure 38. Because of the high mutual inductance,
Capacitive coupling can be used to ad- both primary and secondary are resonated by
vantage in reducing the total number of the one tuning capacitor.
tuned circuits in a transmitter so as to con-
serve space and cost. It also can be used to Link Coupling A special form of inductive
advantage between stages for driving beam coupling which is widely em-
tetrode or pentode amplifier or doubler ployed in radio transmitter circuits is known
stages. as link coupling. A low impedance r-f trans-
mission line couples the two tuned circuits
Inductive Inductive coupling (figure 39) together. Each end of the line is terminated
Coupling results when two coils are electro- in one or more turns of wire, or links, wound
magnetically coupled to one an- around the coils which are being coupled to-
other. The degree of coupling is controlled gether. These links should be coupled to each
by varying the mutual inductance of the tuned circuit at the point of zero r -f po-
two coils, which is accomplished by chang- tential, or nodal point. A ground connection
ing the spacing or the relationship between to one side of the link usually is used to re-
the axes of the coils. duce harmonic coupling, or where capacitive
Inductive coupling is used extensively for coupling between two circuits must be min-
coupling r -f amplifiers in radio receivers. imized. Coaxial line is commonly used to
However, the mechanical problems involved transfer energy between the two coupling
in adjusting the degree of coupling limit the links, although twin -lead may be used where
usefulness of direct inductive coupling in harmonic attenuation is not so important.
transmitters. Either the primary or the sec-
ondary or both coils may be tuned.
INTERWOUND
LINK COUPLING
AT `COLD' ONOS.
UPPER ENOS'HOT"
Figure 41
www.americanradiohistory.com
HANDBOOK Shunt and Series Feed 267
LINK COUPLING
AT -COLD - CENTER.
ENDS "MOT'
+SG +NV +SG +MV
Figure 42
Figure 43
PUSH -PULL LINK COUPLING
ILLUSTRATING PARALLEL AND
SERIES PLATE FEED
Typical link -coupled circuits are shown in Parallel plate feed is desirable from a safety
figure 41 and 42. Some of the advantages of standpoint since the tank circuit is at ground
link coupling are the following: potential with respect to d.c. However, a
high -impedance r -f choke is required, and
the r -f choke must be able to withstand the
(1) It eliminates coupling taps on tuned peak r -f voltage output of the tube. Series
plate feed eliminates the requirement for a
circuits. high -performance r -f choke, but requires the
(2) It permits the use of series power sup- use of a relatively large value of bypass
ply connections in both tuned-grid capacitance at the bottom end of the tank
circuit, as contrasted to the moderate value
and tuned -plate circuits, and thereby of coupling capacitance which may be used
eliminates the need of shunt -feed r -f at the top of the tank circuit for parallel
plate feed.
chokes.
(3) It allows considerable separation be-
tween transmitter stages without ap- sage of r -f energy while still permitting a
preciable r-f losses or stray chassis cur- direct current or audio- frequency current to
rents. pass. They consist of inductances wound
(4) It reduces capacitive coupling and with a large number of turns, either in the
thereby makes neutralization more form of a solenoid, a series of solenoids, a
easily attainable in r -f amplifiers. single universal pie winding, or a series of
(S) It provides semiautomatic impedance pie windings. These inductors are designed
matching between plate and grid to have as much inductance and as little
tuned circuits, with the result that distributed or shunt capacitance as possible.
greater grid drive can be obtained in The unavoidable small amount of distributed
comparison to capacitive coupling. capacitance resonates the inductance, and
(6) It effectively reduces the coupling of this frequency normally should be much
harmonic energy. lower than the frequency at which the
The link -coupling line and links can be
made of No. 18 pushback wire for coupling
between low -power stages. For coupling be-
tween higher- powered stages the 150-ohm
twin -lead transmission line is quite effective
and has very low loss. Coaxial transmission
is most satisfactory between high powered
amplifier stages, and should always be used
where harmonic attenuation is important. -BIAS -BIAS
www.americanradiohistory.com
CHAPTER TWELVE
R -F Feedback
Bf
B
R.f.1N
BIAS
O
Figure 1
R. F. OUT
B AS BIAS B+
Figure 2 Figure 4
SINGLE STAGE AMPLIFIER WITH R -FAMPLIFIER WITH FEEDBACK
R -F FEEDBACK CIRCUIT AND IMPEDANCE MATCHING
OUTPUT NETWORK.
Tuning and loading are accomplished by C,
and C.. C and are tuned in unison to
establish the correct degree of feedback.
R.F.OUT
E
lIN000
R. F. I NO)
Figure 5
BASIC CIRCUIT OF TWO-STAGE AMPLIFIER WITH R -F FEEDBACK
Feedback voltage is obtained from a voltage divider across the output circuit and
applied directly to the cathode of the first tube. The input tank circuit is thus
outside the feedback loop.
www.americanradiohistory.com
HANDBOOK R -F Feedback Circuits 271
of both tubes as effectively as using individual cuit is inductive there is energy transferred
feedback loops around each stage, yet will from the plate to the grid circuit (positive
allow a higher level of over -all gain. With feedback) which will introduce negative resist-
only two tuned circuits in the feedback loop, ance in the grid circuit. When this shunt
it is possible to use 12 to 15 db of feedback negative resistance across the grid circuit is
and still leave a wide margin for stability. It lower than the equivalent positive resistance
is possible to reduce the distortion by nearly of the grid loading, circuit losses, and driving
as many db as are used in feedback. This cir- source impedance, the amplifier will oscillate.
cuit has two advantages that are lacking in the When the plate circuit is in resonance
single -stage feedback amplifier. First, the fila- (phase angle equal to zero) the input resist-
ment of the output stage can now be operated ance due to the grid -plate capacitance becomes
at r -f ground potential. Second, any conven- infinite. As the plate circuit is tuned to the
tional pi output network may be used. capacitive side of resonance, the input resist-
R -f feedback will correct several types of ance becomes positive and power is actually
distortion. It will help correct distortion caused transferred from the grid to the plate circuit.
by poor power supply regulation, too low grid This is the reason that the grid current in an
bias, and limiting on peaks when the plate unneutralized tetrode r -f amplifier varies from
voltage swing becomes too high. a low value with the plate circuit tuned on the
low- frequency side of resonance to a high value
Neutralization The purpose of neutraliza- on the high -frequency side of resonance The
and R -F Feedback tion of an r -f amplifier grid current is proportional to the r-f voltage
stage is to balance out ef- on the grid which is varying under these con-
fects of the grid -plate capacitance coupling in ditions. In a tetrode class -ABs amplifier, the
the amplifier. In a conventional amplifier us- effect of grid -plate feedback can be observed
ing a tetrode tube, the effective input capacity by placing a r -f voltmeter across the grid cir-
is given by: cuit and observing the voltage change as the
Input capacitance = Cin +Cgp (1 + A cos B) plate circuit is tuned through resonance.
If the amplifier is over -neutralized, the ef-
where, fects reverse so that with the plate circuit
Cia equals tube input capacitance, tuned to the low- frequency side of resonance,
C. equals grid -plate capacitance, the grid voltage is high, and on the high -fre-
A equals grid -to -plate voltage amplifica- quency side of resonance, it is low.
cation,
O equals angle of load.
Amplifier A useful "rule of
Neutralization Check thumb' method of
In a typical unneutralized tetrode amplifier checking neutraliza-
having a stage gain of 33, the input capaci- tion of an amplifier stage (assuming that it
tance of the tube with the plate circuit in is nearly correct to start with) is to tune both
resonance is increased by 8 pf due to the grid and plate circuits to resonance. Then, ob-
unneutralized grid -plate capacitance. This is serving the r -f grid current, tune the plate cir-
unimportant in amplifiers where the gain (A) cuit to the high - frequency side of resonance.
remains constant but if the tube gain varies, If the grid current rises, more neutralization
serious detuning and r-f phase shift may result. capacitance is required. Conversely, if the grid
A grid or screen modulated r -f amplifier is an current decreases, less capacitance is needed.
example of the case where the stage gain var- This indication is very sensitive in a neutral-
ies from a maximum down to zero. The gain ized triode amplifier, and correct neutraliza-
of a tetrode r -f amplifier operating below plate tion exists when the grid current peaks at the
current saturation varies with loading so that point of plate current dip. In tetrode power
if it drives a following stage into grid current amplifiers this indication is less pronounced.
the loading increases and the gain falls off. Sometimes in a supposedly neutralized tetrode
The input of the grid circuit is also affected amplifier, there is practically no change in
by the grid -plate capacitance, as shown in this grid voltage as the plate circuit is tuned
equation: through resonance, and in some amplifiers it
1
Input resistance = 2af X Cg, (Asia)
is unchanged on one side of resonance and
drops slightly on the other side. Another ob-
This resistance is in shunt with the grid servation sometimes made is a small dip in
current loading, grid tank circuit losses, and the center of a broad peak of grid current.
driving source impedance. When the plate cir- These various effects are probably caused by
272 R -F Feedback THE RADIO
Figure 6 Figure 7
SINGLE STAGE R -F AMPLIFIER NEUTRALIZED AMPLIFIER AND
WITH FEEDBACK RATIO OF INHERENT FEEDBACK CIRCUIT
C /C, to C:,. /C,r DETERMINES Neutralization is achieved by varying
STAGE NEUTRALIZATION
the capacity of c,.
coupling from the plate to the grid circuit tential when r.f. is impressed upon the cathode.
through other paths which are not balanced The output voltage available with capacity
out by the particular neutralizing circuit used. coupling, of course, is less than the plate -
Figure 6 shows an r -f am- cathode r -f voltage developed by the amount
Feedback and
plifier with negative feed- of feedback voltage across Cl.
Neutralization
of a One -Stage back. The voltage devel-
R -F Amplifier oped across C, due to the 12 -2 Feedback and
divider action of C :, and C, Neutralization of a
is introduced in series with the voltage devel-
Two -Stage R -F Amplifier
oped across the grid tank circuit and is in
phase -opposition to it. The feedback can be Feedback around two r -f stages has the ad-
made any value from zero to 100% by proper- vantage that more of the tube gain can be
ly choosing the values of C3 and C1. realized and nearly as much distortion reduc-
For reasons stated previously, it is necessary tion can be obtained using 12 db around two
to neutralize this amplifier, and the relation- stages as is realized using 12 db around each
ship for neutralization is: of two stages separately. Figure 9 shows a
C. Cgp basic circuit of a two -stage feedback ampli-
fier. Inductive output coupling is used, al-
C, Cgt though a pi- network configuration will also
It is often necessary to add capacitance from work well. The small feedback voltage required
plate to grid to satisfy this relationship is obtained from the voltage divider (C1-C2)
Figure 7 is identical to figure 6 except that and is applied to the cathode of the driver
it is redrawn to show the feedback inherent in tube. C, is only a few pf, so this feedback
this neutralization circuit more clearly. CN and voltage divider may be left fixed for a wide
C replace C3 and C4, and the main plate tank frequency range. If the combined tube gain is
tuning capacitance is C3. The circuit of figure 160, and 12 db of feedback is desired, the ratio
7 presents a problem in coupling to the grid of C._ to C, is about 40 to 1. This ratio in
circuit. Inductive coupling is ideal, but the practice may be 100 pf to 2.5 pf, for example.
extra tank circuits complicate the tuning of a A complication is introduced into this sim-
transmitter which uses several cascaded am- plified circuit by the cathode -grid capacitance
plifiers with feedback around each one. The
grid could be coupled to a high source imped-
ance such as a tetrode plate, but the driver R. F. OUT
www.americanradiohistory.com
HANDBOOK R -F Feedback Circuits 273
Figure 9
TWO -STAGE AMPLIFIER WITH FEEDBACK.
Included is a capacitor (C,) for neutralizing the cathode -grid capacity of the first tube. V, is neutralized
by capacitor C., and V, is neutralized by the correct ratio of C, /C,.
of the first tube which .causes an. undesired more feedback from the output stage to over-
coupling to the input grid circuit. It is neces- come.
sary to neutralize out this capacitance coupling,
as illustrated in figure 9. The relationship for
Tests For Neutralizing the circuit of
Neutralization figure 9 balances out cou-
neutralization is:
pling between the input
Ca CRf
tank circuit and the output tank circuit, but it
C4 C6 does not remove all coupling from the plate
circuit to the grid- cathode tube input. This
The input circuit may be made unbalanced latter coupling is degenerative, so applying a
by making C, five times the capacity of C.,. signal to the plate circuit will cause a signal
This will tend to reduce the voltage across to appear between grid and cathode, even
the coil and to minimize the power dissipated though the stage is neutralized. A bench test
by the coil. For proper balance in this case, for neutralization is to apply a signal to the
C,; must be five times the grid- filament capaci- plate of the tube and detect the presence of a
tance of the tube. signal in the grid coil by inductive coupling
Except for tubes having extremely small to it. No signal will be present when the stage
grid -plate capacitance, it is still necessary to is neutralized. Of course, a signal could be in-
properly neutralize both tubes. If the ratio of ductively coupled to the input and neutraliza-
C, to C, is chosen to be equal to the ratio of tion accomplished by adjusting one branch of
the grid -plate capacitance to the grid -filament the neutralizing circuit bridge (Cr, for ex-
capacitance in the second tube (V2), this tube ample) for minimum signal on the plate cir-
will be neutralized. Tubes such as a 4X -150A cuit.
have very low grid -plate capacitance and prob-
ably will not need to be neutralized when used Neutralizing the cathode -grid capacitance of
in the first (V,) stage. If neutralization is the first stage of figure 9 may be accomplished
necessary, capacitor C., is added for this pur- by applying a signal to the cathode of the tube
pose and the proper value is given by the and adjusting the bridge balance for minimum
following relationship: signal on a detector inductively coupled to the
input coil.
C6 Cgf Ca
Cs C C, Tuning a Two -Stage Tuning the two -stage
Feedback Amplifier feedback amplifier of
If neither tube requires neutralization, the figure 9 is accom-
bottom end of the interstage tank circuit may plished in an unconventional way because the
be returned to r -f ground. The screen and output circuit cannot be tuned for maximum
suppressor of the first tube should then be output signal. This is because the output cir-
grounded to keep the tank output capaci- cuit must be tuned so the feedback voltage
tance directly across this interstage circuit and applied to the cathode is in -phase with the
to avoid common coupling between the feed- input signal applied to the first grid. When
back on the cathode and the interstage circuit. the feedback voltage is not in- phase, the result-
A slight amount of degeneration occurs in the ant grid- cathode voltage increases as shown
first stage since the tube also acts as a grounded in figure 10. When the outl. t
circuit is
grid amplifier with the screen as the grounded properly tuned, the resultant grid-c,-bode volt-
grid. The p. of the screen is much lower than age on the first tube will be at a mini, um, and
that of the control grid so that this effect may the voltage on the interstage tuned circuit will
be unnoticed and would only require slightly also be at a minimum.
274 R -F Feedback THE RADIO
vOLTAGE-
INNIT GRID
TO
GROUND 1VOLTAGE -CATHODE TO GROUND
(PEEOIACEI
J
Figure 12
INTERSTAGE CIRCUIT WITH
Figure 10 SEPARATE NEUTRALIZING
AND FEEDBACK CIRCUITS.
VECTOR RELATIONSHIP OF
FEEDBACK VOLTAGE
A = Output Circuit Properly Tuned It is convenient, however, to separate these cir-
B= Output Circuit Mis -Tuned
cuits so neutralization and feedback can be
adjusted independently. Also, it may be de-
The two -stage amplifier may be tuned by sirable to be able to switch the feedback out
placing a r -f voltmeter across the interstage of the circuit. For these reasons, the circuit
tank circuit ( "hot" side to ground) and tuning shown in figure 12 is often used. Switch S1
the input and interstage circuits for maximum removes the feedback loop when it is closed.
meter reading, and tuning the output circuit A slight tendency for low- frequency para-
for minimum meter reading. If the second tube sitic oscillations still exists with this circuit.
is driven into the grid current region, the grid LI should have as little inductance as possible
current meter may be used in place of the r -f without upsetting the feedback. If the value of
voltmeter. On high powered stages where oper- L, is too low, it cancels out part of the re-
ation is well into the class -AB region, the actance of feedback capacitor C, and causes
plate current dip of the output tube indicates the feedback to increase at low values of radio
correct output circuit tuning, as in the usual frequency. In some cases, a swamping resistor
amplifier. may be necessary across L1. The value of this
resistor should be high compared to the re-
Parasitic Oscillations in Quite often low fre- actance of C, to avoid phase -shift of the r -f
the Feedback Amplifier q u e n c y parasitics feedback.
may be found in
the interstage circuit of the two -stage feedback
amplifier. Oscillation occurs in the first stage 12 -3 Neutralization
due to low frequency feedback in the cathode Procedure in
circuit. R -f chokes, coupling capacitors, and Feedback -Type Amplifiers
bypass capacitors provide the low frequency
tank circuits. When the feedback and second Experience with feedback amplifiers has
stage neutralizing circuits are combined, it is brought out several different methods of neu-
necessary to use the configuration of figure 11. tralizing. An important observation is that
This circuit has the advantage that only one when all three neutralizing adjustments are
capacitor (G) is required from the plate of correctly made the peaks and dips of various
the output tube, thus keeping the added ca- tuning meters all coincide at the point of cir-
pacitance across the output tank at a minimum. cuit resonance. For example, the coincident in-
dications when the various tank circuits are
tuned through resonance with feedback oper-
ating are:
A-When the PA plate circuit is tuned
through resonance:
1 -PA plate current dip
2-Power output peak
3 -PA r -f grid voltage dip
8 AS
4-PA grid current dip
Figure 11 (Note: The PA grid current peaks
INTERSTAGE CIRCUIT COMBINING
NEUTRALIZATION AND when feedback circuit is disabled
FEEDBACK NETWORKS. and the tube is heavily driven)
www.americanradiohistory.com
HANDBOOK Neutralization 275
(--/
c.
C. I,
R. F. IN
R. F OUT
(z, J[Cs k ;Lc
`
Lt
P
Hb
Tc9
L aF
Cto
; `CGF
IOU7
?/
St1-
Cs
r ,1
\
CN
RFCP
F
o
e.,
-1
TiT
HP =
BIAS B+ BIAS
Figure 13
TWO-STAGE AMPLIFIER WITH FEEDBACK CIRCUIT
B-When the PA grid circuit is tuned 2- Neutralize the grid -plate capaci-
through resonance: tance of the driver stage
1- Driver plate current dip 3- Neutralize the grid-plate capaci-
2 -PA r -f grid voltage peak tance of the power amplifier (PA )
3 -PA grid current peak stage
4-PA power output peak 4-Apply r -f feedback
5- Neutralize driver grid -cathode ca-
C -When the driver grid circuit is tuned pacitance
through resonance:
1- Driver r -f grid voltage peak These steps will be explained in more detail
2- Driver plate current peak in the following paragraphs:
3 -PA r -f grid current peak Step 1. The removal of r -f feedback through
4 -PA plate current peak the feedback circuit must be complete. The
5 -PA power output peak
switch (Si) shown in the feedback circuit
( figure 13 ) is one satisfactory method. Since
Four meters may be employed to measure
the most important of these parameters. The C1, is effectively across the PA plate tank cir-
meters should be arranged so that the follow- cuit it is desirable to keep it across the circuit
ing pairs of readings are displayed on meters when feedback is removed to avoid appreciable
located close together for ease of observation detuning of the plate tank circuit. Another
of coincident peaks and dips: method that can be used if properly done is
to ground the junction of CI; and C. Ground-
1-PA plate current and power output ing this common point through a switch or
2 -PA r -f grid current and PA plate relay is not good enough because of common
current coupling through the length of the grounding
3 -PA r -f grid voltage and power out- lead. The grounding method shown in figure
put 14 is satisfactory.
4- Driver plate current and PA r -f
Step 2. Plate power and excitation are applied.
grid voltage
The driver grid tank is resonated by tuning
The third pair listed above may not be for a peak in driver r -f grid voltage or driver
necessary if the PA plate current dip is pro- plate current. The power amplifier grid tank
nounced. When this instrumentation is pro- circuit is then resonated and adjusted for a
vided, the neutralizing procedure is as follows: dip in driver plate current. Driver neutraliza-
tion is now adjusted until the PA r -f grid
1- Remove the r -f feedback voltage (or PA grid current) peaks at exactly
the point of driver plate current dip. A handy
rule for adjusting grid -plate neutralization of
a tube without feedback: with all circuits in
resonance, detune the plate circuit to the high
frequency side of resonance: If grid current
to next stage (or power output of the stage
under test) increases, more neutralizing capaci-
tance is required and vice versa.
www.americanradiohistory.com
CHAPTER THIRTEEN
Amplitude Modulation
www.americanradiohistory.com
278 Amplitude Modulation THE RADIO
voice or music modulation, two sidebands
will be formed by each modulating frequen-
cy (one on each side of the carrier), and
the radiated signal will consist of a band of 1
A
C.W. OR UNMODULATED CARRIER
frequencies. The bandwidth, or channel, taken
up in the frequency spectrum by a conven-
tional double -sideband amplitude -modulated
signal, is equal to twice the highest modu-
SINE WAVE
lating frequency. For example, if the highest AUDIO SIGNAL FROM MODULATOR
saying that during modulation, the transmit- While the average r-f voltage of the mod-
ted r -f energy no longer is confined to a ulated wave over a modulation cycle is the
single radio frequency. same as for the unmodulated carrier, the
It will be noted that the average amplitude average power increases with modulation. If
of the peak r -f voltage, or modulation enve- the radio- frequency power is integrated over
lope, is the same with or without modula- the audio cycle, it will be found with 100
tion. This simply means that the modulation percent sine -wave modulation the average r -f
is symmetrical (assuming a symmetrical power has increased 50 percent. This addi-
modulating wave) and that for distortionless tional power is represented by the sidebands,
www.americanradiohistory.com
HANDBOOK Modulation 279
Modulation Figure 2
So long as the relative propor-
Percentage tion of the various sidebands GRAPHICAL DETERMINATION OF MODU-
making up voice modulation is LATION PERCENTAGE
maintained, the signal may be received and The procedure for determining modulation
detected without distortion. However, the percentage from the peak voltage points
indicated is discussed in the text.
higher the average amplitude of the side -
bands, the greater the audio signal produced
at the receiver. For this reason it is desirable
to increase the modulation percentage, or de- M
Ecar - Emin
Ecar
gree of modulation, to the point where maxi-
mum peaks just hit 100 percent. If the
modulation percentage is increased so that In the above two formulas E. is the
the peaks exceed this value, distortion is in- maximum carrier amplitude with modulation
and En, in is the minimum amplitude; Ecar
troduced, and if carried very far, bad inter-
is the steady -state amplitude of the carrier
ference to signals on nearby channels will
result. without modulation. Since the deflection of
the spot on a cathode -ray tube is linear with
respect to voltage, the relative voltages of
Modulation The amount by which a car - these various amplitudes may be determined
Measurement rier is being modulated may by measuring the deflections, as viewed on
be expressed either as a mod- the screen, with a rule calibrated in inches
ulation factor, varying from zero to 1.0 at or centimeters. The percentage of modula-
maximum modulation, or as a percentage. tion of the carrier may be found by multiply-
The percentage of modulation is equal to 100 ing the modulation factor thus obtained by
times the modulation factor. Figure 2A 100. The above procedure assumes that there
shows a carrier wave modulated by a sine - is no carrier shift, or change in average
wave audio tone. A picture such as this carrier amplitude with modulation.
might be seen on the screen of a cathode-ray If the modulating voltage is symmetrical,
oscilloscope with sawtooth sweep on the such as a sine wave, and modulation is ac-
horizontal plates and the modulated carrier complished without the introduction of dis-
impressed on the vertical plates. The same tortion, then the percentage modulation will
carrier without modulation would appear on be the same for both negative and positive
the oscilloscope screen as figure 2B. peaks. However, the distribution and phase
The percentage of modulation of the posi- relationships of harmonics in voice and music
tive peaks and the percentage of modulation waveforms are such that the percentage
of the negative peaks can be determined modulation of the negative modulation peaks
separately from two oscilloscope pictures may exceed the percentage modulation of the
such as shown. positive peaks, or vice versa. The percent-
The modulation factor of the positive age modulation when referred to without
peaks may be determined by the formula: regard to polarity is an indication of the
average of the negative and positive peaks.
M - Emu: - Ecar Modulation The modulation capability of a
Ecar Capability transmitter is the maximum
percentage to which that trans-
The factor for negative peaks may be de- mitter may be modulated before spurious
termined from the formula: sidebands are generated in the output or
280 Amplitude Modulation THE RADIO
before the distortion of the modulating removed from the desired signal and ad-
waveform becomes objectionable. The high- justing the phase-reversing switch to the
est modulation capability which any trans- position which gives the least "splatter" when
mitter may have on the negative peaks is the transmitter is modulated rather heavily.
100 percent. The maximum permissible mod- If desired, the switch then may be replaced
ulation of many transmitters is less than with permanent wiring, so long as the micro-
100 percent, especially on positive peaks. phone and speech system are not to be
The modulation capability of a transmitter changed.
may be limited by tubes with insufficient A more conclusive illustration of the lop-
filament emission, by insufficient excitation sidedness of a speech waveform may be ob-
or grid bias to a plate -modulated stage, too tained by observing the modulated waveform
light loading of any type of amplifier car- of a radiotelephone transmitter on an oscil-
rying modulated r.f., insufficient power out- loscope. A portion of the carrier energy of
put capability in the modulator, or too much the transmitter should be coupled by means
excitation to a grid -modulated stage or a of a link directly to the vertical plates of
class -B linear amplifier. In any case, the the 'scope, and the horizontal sweep should
FCC regulations specify that no transmitter be a sawtooth or similar wave occurring at
may be modulated in excess of its modulation a rate of approximately 30 to 70 sweeps
capability. Hence, it is desirable to make per second.
the modulation capability of a transmitter With the speech signal from the speech
as near as possible to 100 percent so that the amplifier connected to the transmitter with
carrier power may be used most effectively. one polarity it will be noticed that negative-
peak clipping -as indicated by bright "spots"
Speech Waveform The manner in which the in the center of the 'scope pattern whenever
Dissymmetry human voice is produced the carrier amplitude goes to zero-will
by the vocal cords gives occur at a considerably lower level of aver-
rise to a certain dissymmetry in the wave- age modulation than with the speech signal
form of voice sounds when they are picked being fed to the transmitter with the other
up by a good quality microphone. This is polarity. When the input signal to the
especially pronounced in the male voice, and transmitter is polarized in such a manner
more so on certain voice sounds than on that the "fingers" of the speech wave ex-
others. The result of this dissymmetry in tend in the direction of positive modulation
the waveform is that the voltage peaks on these fingers usually will be clipped in the
one side of the average value of the wave plate circuit of the modulator at an accept-
will be considerably greater, often two or able peak modulation level.
three times as great, as the voltage excursions The use of the proper polarity of the in-
on the other side of the zero axis. The coming speech wave in modulating a trans-
average value of voltage on both sides of mitter can afford an increase of approxi-
the wave is, of course, the same. mately two to one in the amount of speech
As a result of this dissymmetry in the audio power which may be placed on the
male voice waveform, there is an optimum carrier of an amplitude- modulated trans-
polarity of the modulating voltage that must mitter for the same amount of sideband
be observed if maximum sideband energy is splatter. More effective methods for in-
to be obtained without negative peak clip- creasing the amount of audio power on the
ping and generation of splatter on adjacent carrier of an a -m phone transmitter are
channels. discussed later in this chapter.
A double -pole double -throw phase-revers-
ing switch in the input or output leads of
any transformer in the speech amplifier sys- 13 -3 Systems of Amplitude
tem will permit switching the extended peaks Modulation
in the direction of maximum modulation cap-
ability. The optimum polarity may be de- There are many different systems and
termined easily by listening on a selective methods for amplitude -modulating a carrier,
receiver tuned to a frequency 30 to 50 kHz but most may be grouped under three gen-
www.americanradiohistory.com
HANDBOOK Modulation Systems 281
www.americanradiohistory.com
HANDBOOK Modulation Systems 283
R-F AMPLIFIER
sistor across the grid tank of the stage
serves as swamping to stabilize the r-f driv-
ing voltage. At least 50 percent of the out- RFC .:O
PANT.
put of the driving stage should be dissipated
in this swamping resistor under carrier con-
ditions. IDOA f
A comparatively small amount of audio W.W.
.9
power will be required to modulate the am-
plifier stage 100 percent. An audio amplifier
having 20 watts output will be sufficient to
modulate an amplifier with one kilowatt
input. Proportionately smaller amounts of
audio will be required for lower -powered
stages. However, the audio amplifier that is 2SR 10w
MIDGET CHORE
www.americanradiohistory.com
284 Amplitude Modulation THE RADIO
ized, and any possible tendency toward does offer other advantages and the linearity
parasitics under any condition of operation is quite adequate for communications work.
should be eliminated. Then the antenna There are two significant and worthwhile
should be coupled to the plate circuit, the advantages of screen -grid modulation for
grid bias should be run up to the maximum communications work: (1) The excitation
available value, and the plate voltage and requirements for an amplifier which is to be
excitation should be applied. The grid -bias screen modulated are not at all critical,
voltage should then be reduced until the am- and good regulation of the excitation volt-
plifier draws the approximate amount of age is not required. The normal rated grid -
plate current it is desired to run, and modu- circuit operating conditions specified for
lation corresponding to about 80 percent is class -C c -w operation are quite adequate for
then applied. If the plate current kicks up screen -grid modulation. (2) The audio mod-
when modulation is applied, the grid bias ulating power requirements for screen -grid
should be reduced; if the plate meter kicks modulation are relatively low.
down, increase the grid bias. A screen -grid modulated r -f amplifier op-
When the amount of bias voltage has been erates as an efficiency- modulated amplifier,
found (by adjusting the fine control (R2) the same as does a class -B linear amplifier and
on the bias supply) where the plate meter re- a grid -modulated stage. Hence, plate circuit
mains constant with modulation, it is more loading is relatively critical as in any effi-
than probable that the stage will be drawing ciency- modulated stage, and must be ad-
either too much or too little input. The an- justed to the correct value if normal power
tenna coupling should then be either in- output with full modulation capability is to
creased or decreased (depending on whether be obtained. As in the case of any efficiency -
the input was too little or too much, respec- modulated stage, the operating efficiency at
tively) until the input is more nearly the the peak of the modulation cycle will be
correct value. The bias should then be re- between 70 and 80 percent, with efficiency
adjusted until the plate meter remains con- at the carrier level (if the stage is operating
stant with modulation as before. By slight in the normal manner with full carrier)
jockeying back and forth of antenna cou- about half of the peak -modulation value.
pling and grid bias, a point can be reached There are two main disadvantages of
where the tubes are running at rated plate screen -grid modulation, and several factors
dissipation, and where the plate milliam- which must be considered if satisfactory op-
meter on the modulated stage remains sub- eration of the screen -grid modulated stage is
stantially constant with modulation. to be obtained. The disadvantages are: (1)
The linearity of the stage should then be As mentioned before, the linearity of modu-
checked by any of the conventional methods; lation with respect to screen -grid voltage of
the trapezoidal pattern method employing a such a stage is satisfactory only for com-
cathode -ray oscilloscope is probably the most munications work, unless carrier -rectified
satisfactory. The check with the trapezoidal degenerative feedback is employed around
pattern will allow the determination of the the modulated stage to improve the linear-
proper amount of gain to employ on the ity of modulation. (2) The impedance of
speech amplifier. Too much audio power on the screen grid to the modulating signal is
the grid of the modulated stage should not nonlinear. This means that the modulating
be used in the tuning -up process, as the plate signal must be obtained from a source of
meter will kick erratically and it will be im- quite low impedance if audio distortion of
possible to make a satisfactory adjustment. the signal appearing at the screen grid is to
Screen-Grid be avoided.
Amplitude modulation may be
Modulation accomplished by varying the Screen -Grid Instead of being linear with re-
screen -grid voltage in a class - Impedance to modulating voltage, as
spect
C amplifier which employs a pentode, beam is the plate circuit of a plate -
tetrode, or other type of screen -grid tube. modulated class -C amplifier, the screen grid
The modulation obtained in this way is not presents approximately a square -law imped-
especially linear, but screen -grid modulation ance to the modulating signal over the region
www.americanradiohistory.com
HANDBOOK Modulation Systems 285
of signal excursion where the screen is posi- cycle when the screen does draw current, it
tive with respect to ground. This nonlineari- presents approximately a square -law imped-
ty may be explained in the following man- ance.
ner: At the carrier level of a conventional Circuits for Laboratory analysis of a large
screen -modulated stage the plate -voltage Screen -Grid number of circuits for accom-
swing of the modulated tube is one -half the Modulation plishing screen modulation has
voltage swing at peak -modulation level. This
led to the conclusion that the
condition must exist in any type of conven- audio modulating voltage must be obtained
tional efficiency -modulated stage if 100 per-
from a low- impedance source if low- distor-
cent positive modulation is to be attainable.
tion modulation is to be obtained. Figure 4
Since the plate -voltage swing is at half am-
shows a group of sketches of the modulation
plitude, and since the screen voltage is at
envelope obtained with various types of
half its full modulation value, the screen cur-
modulators and also with insufficient antenna
rent is relatively low. But at the positive
modulation peak the screen voltage is ap- coupling. The result of this laboratory work
led to the conclusion that the cathode -fol-
proximately doubled, and the plate -voltage
lower modulator of the basic circuit shown
swing also is at twice the carrier amplitude.
in figure 5 is capable of giving good- quality
Due to the increase in plate -voltage swing
screen -grid modulation, and in addition the
with rising screen voltage, the screen current
increases more than linearly with rising
circuit provides convenient adjustments for
the carrier level and the output level on
screen voltage.
negative modulation peaks. This latter con-
In a test made on an amplifier with an 813
trol (P2) in figure 5, allows the amplifier to
tube, the screen current at carrier level was
be adjusted in such a manner that negative -
about 6 ma with screen potential of 190
peak clipping cannot take place, yet the
volts; but under conditions which repre-
negative modulation peaks may be adjusted
sented a positive modulation peak the screen
to a level just above that at which sideband
current measured 2S ma at a potential of splatter will occur.
400 volts. Thus instead of screen current
doubling with twice screen voltage as would The Cathode- The cathode follower is
be the case if the screen presented a resistive Follower Modulator ideally suited for use as
impedance, the screen current became about the modulator for a
four times as great with twice the screen screen -grid stage since it acts as a relatively
voltage. low- impedance source of modulating voltage
Another factor which must be considered for the screen -grid circuit. In addition the
in the design of a screen -modulated stage, if cathode- follower modulator allows the sup-
full modulation is to be obtained, is that the ply voltage both for the modulator and for
power output of a screen -grid stage with the screen grid of the modulated tube to be
zero screen voltage is still relatively large. obtained from the high- voltage supply for
Hence, if anything approaching full modu- the plate of the screen -grid tube or beam
lation on negative peaks is to be obtained, tetrode. In the usual case the plate supply
the screen potential must be made negative for the cathode follower, and hence for the
with respect to ground on negative modula- screen grid of the modulated tube, may be
tion peaks. In the usual types of beam taken from the bleeder on the high - voltage
tetrode tubes the screen potential must be power supply. A tap on the bleeder may be
20 to SO volts negative with respect to used, or two resistors may be connected in
ground before cutoff of output is obtained. series to make up the bleeder, with appro-
This condition further complicates the prob- priate values such that the voltage applied
lem of obtaining good linearity in the audio to the plate of the cathode follower is ap-
modulating voltage for the screen -modulated propriate for the tube to be modulated. It
stage, since the screen voltage must be driven is important that a bypass capacitor be used
negative with respect to ground over a from the plate of the cathode -follower mod-
portion of the cycle. Hence the screen draws ulator to ground.
no current over a portion of the modulating The voltage applied to the plate of the
cycle, and over the major portion of the cathode follower should be about 100 volts
www.americanradiohistory.com
286 Amplitude Modulation THE RADIO
Figure 4
SCREEN-MODULATION CIRCUITS
Three common screen -modulation circuits are illustrated above. All three circuits are capable
of giving intelligible voice modulation although the waveform distortion in the circuits of
A and it is likely to be rather severe. The arrangement at A is often called "clamp- tube"
screen modulation; by returning the grid resistor on the clamp tube to ground the circuit will
give controlled -carrier screen modulation. This circuit has the advantage that it is simple
and is well suited to use in mobile transmitters. 6 is an arrangement using a transformer-
coupled modulator, and offers no particular advantages. The arrangement at C is capable
of giving good modulation linearity due to the low impedance of the cathode -follower
modulator. However, due to the relatively low heater- cathode ratings on tubes suited for
use as the modulator, a separate heater supply for the modulator tube normally is required.
This limitation makes application of the circuit to the mobile transmitter a special problem,
since an isolated heater supply normally is not available. Shown at D as an assistance in the
tuning of a screen- modulated transmitter (or any efficiency- modulated transmitter for that
matter) is the type of modulation envelope which results when loading to the modulated
stage is insufficient.
greater than the rated screen voltage for the The only current taken by the cathode
tetrode tube as a c -w class -C amplifier. follower itself will be that which will flow
Hence the cathode- follower plate voltage through the 100,000 -ohm resistor between
should be about 350 volts for an 815, 2E26, the cathode of the 6L6 modulator and the
or 829B, about 400 volts for an 807 or negative supply. The current taken from
4 -125A, about 500 volts for an 813, and the bleeder on the high -voltage supply will
about 600 volts for a 4 -250A or a 4E27. be the carrier -level screen current of the
Then potentiometer (P1) in figure 5 should tube being modulated (which current passes
be adjusted until the carrier -level screen of course through the cathode follower) plus
voltage on the modulated stage is about one - that current which will pass through the
half the rated screen voltage specified for 100,000 -ohm resistor.
the tube as a class -C c -w amplifier. The The loading of the modulated stage should
current taken by the screen of the modu- be adjusted until the input to the tube is
lated tube under carrier conditions will be about 50 percent greater than the rated
about one- fourth the normal screen current plate dissipation of the tube or tubes in the
for c -w operation. stage. If the carrier -level screen voltage value
HANDBOOK Modulation Systems 287
www.americanradiohistory.com
288 Amplitude Modulation THE RADIO
ural result of using a low- impedance source ically, suppressor-grid modulation operates
of modulating signal for the stage. in the same general manner as other forms
As an example of a typical screen -modu- of efficiency modulation; carrier plate -cir-
lated stage, full output of 75 watts of car- cuit efficiency is about 35 percent, and an-
rier may be obtained from an 813 tube tenna coupling must be rather tight. How-
operating with a plate potential of only 1250 ever, suppressor -grid modulation has one
volts. No increase in output from the 813 sizeable disadvantage, in addition to the fact
may be obtained by increasing the plate that pentode tubes are not nearly so widely
voltage, since the tube may be operated with used as beam tetrodes which of course do
full rated plate dissipation of 125 watts, not have the suppressor element. This dis-
with normal plate efficiency for a screen - advantage is that the screen -grid current
modulated stage-37.5 percent, at the 1250 - to a suppressor -grid modulated amplifier is
volt potential. rather high. The high screen current is a
The operating conditions of a screen - natural consequence of the rather high neg-
modulated 813 stage are as follows: ative bias on the suppressor grid, which re-
duces the plate -voltage swing and plate cur-
Plate voltage -1250 volts rent with a resulting increase in the screen
Plate current-160 ma current.
Plate input-200 watts In tuning a suppressor -grid modulated
Grid current -11 ma amplifier, the grid bias, grid current, screen
-
Grid bias- 110 volts voltage, and plate voltage are about the
Carrier screen voltage -190 volts same as for class -C c -w operation of the
Carrier screen current -6
ma stage. But the suppressor grid is biased
Power output -approx. 75 watts negatively to a value which reduces the
plate- circuit efficiency to about one -half the
With full 100 percent modulation the plate maximum obtainable from the particular
current decreases about 2 ma and the screen amplifier, with antenna coupling adjusted
current increases about 1 ma; hence plate, until the plate input is about 1.5 times the
screen, and grid current remain essentially rated plate dissipation of the stage. It is im-
constant with modulation. Referring to portant that the input to the screen grid be
figure 5, which was the circuit used as mod- measured to make sure that the rated screen
ulator for the 813, E1 measured + 155 dissipation of the tube is not being exceeded.
volts,
+
E2 measured -
190 volts, E4 measured
50 volts, E3 measured
+ 500 volts,
and the rms swing at E5 for full modulation 4E27 CARRIER
OUTPUT
measured 210 volts, which represents a peak -33W
swing of about 296 volts. Due to the high
R -P INPUT
positive voltage, and the large audio swing,
1G
on the cathode of the 6L6 (triode con-
-
E MA.
www.americanradiohistory.com
HANDBOOK Input Modulation Systems 289
Then the audio signal is applied to the sup- trical quarter -wave lines. They will be de-
pressor grid. In the normal application the scribed later in this section.
audio voltage swing on the suppressor will
be somewhat greater than the negative bias Plate Modulation Plate modulation is the ap-
on the element. Hence suppressor-grid cur- plication of the audio pow-
rent will flow on modulation peaks, so that er to the plate circuit of an r -f amplifier.
the source of audio signal voltage must have The r-f amplifier must be operated class C
good regulation. Tubes suitable for suppres- for this type of modulation in order to ob-
sor -grid modulation are: the 4E27A /5 -I25B tain a radio- frequency output which changes
and 803. A typical suppressor -grid modu- in exact accord with the variation in
lated amplifier is illustrated in figure 6. plate voltage. The r -f amplifier is 100 per-
cent modulated when the peak a -c voltage
13 -4 Input Modulation from the modulator is equal to the d -c volt-
age applied to the r -f tube. The positive
Systems peaks of audio voltage increase the instan-
taneous plate voltage on the r -f tube to twice
Constant -efficiency variable -input modu- the d -c value, and the negative peaks reduce
lation systems operate by virtue of the addi- the voltage to zero.
tion of external power to the modulated The instantaneous plate current to the
stage to effect the modulation. There are r -f stage also varies in accord with the
two general classifications that come under modulating voltage. The peak alternating
this heading; those systems in which the current in the output of a modulator must be
additional power is supplied as audio -fre- equal to the d -c plate current of the class -C
quency energy from a modulator (usually r -f stage at the point of 100 percent modu-
called plate -modulation systems) and those lation. This combination of change in audio
systems in which the additional power to voltage and current can be most easily re-
effect modulation is supplied as direct cur- ferred to in terms of audio power in watts.
rent from the plate supply. In a sinusoidally modulated wave, the an-
Under the former classification comes tenna current increases approximately 22
Heising modulation (probably the oldest percent for 100 percent modulation with a
type of modulation to be applied to a con- pure tone input; an r -f meter in the antenna
tinuous carrier) , class -B plate modulation, circuit indicates this increase in antenna cur-
and series modulation. These types of plate rent. The average power of the r -f wave in-
modulation are by far the easiest to get into creases 50 percent for 100 percent modula-
operation, and they give a very good ratio tion, the efficiency remaining constant.
of power input to the modulated stage to This indicates that in a plate -modulated
power output; 65 to 80 percent efficiency radiotelephone transmitter, the audio -fre-
is the general rule. It is for these two im- quency channel must supply this additional
SO percent increase in average power for
portant reasons that these modulation sys-
sine -wave modulation. If the power input
tems, particularly class -B plate modulation,
to the modulated stage is 100 watts, for
are at present the most popular for a -m
example, the average power will increase
communications work. to 1 5 0 watts at 100 percent modulation, and
Modulation systems coming under the this additional 50 watts of power must be
second classification are of comparatively supplied by the modulator when plate mod-
recent development but have been widely ulation is used. The actual antenna power is
applied to broadcast work. There are quite a constant percentage of the total value of
a few systems in this class. Two of the more input power.
widely used are the Doherty linear amplifier, One of the advantages of plate (or power)
and the Terman - Woodyard high -efficiency modulation is the ease with which proper
grid- modulated amplifier. Both systems oper- adjustments can be made in the transmitter.
:.te by virtue of a carrier amplifier and a Also, there is less plate loss in the r -f ampli-
peak amplifier connected together by elec- fier for a given value of carrier power than
290 Amplitude Modulation THE RADIO
MOOULATED CLASS -C CLASS -C
R-i AMPLIFIER AMPLI FIER
R-F OR IVE
CLASS -S
MODULATOR
MODULATOR
A-F
DRIVE
UCH
Figure 7
with other forms of modulation because the Heising Heising modulation is the old -
plate efficiency is higher. Modulation est system of plate modulation,
By properly matching the plate impedance and usually consists of a class -A
of the r-f tube to the output of the modu- audio amplifier coupled to the r -f amplifier
lator, the ratio of voltage and current swing by means of a modulation choke coil, as
to d -c voltage and current is automatically shown in figure 7.
obtained. The modulator should have a peak The d -c plate voltage and plate current of
voltage output equal to the average d -c the r -f amplifier must be adjusted to a value
plate voltage on the modulated stage. The which will cause the plate impedance to
modulator should also have a peak power match the output of the modulator, since
output equal to the d -c plate input power the modulation choke gives a 1 -to -1 cou-
to the modulated stage. pling ratio. A series resistor, bypassed for
The average power output of the modula- audio frequencies by means of a capacitor,
tor will depend on the type of waveform. If must be connected in series with the plate of
the amplifier is being Heising modulated by the r -f amplifier to obtain modulation up to
a class -A stage, the modulator must have an 100 percent. The peak output voltage of a
average power output capability of one-half class -A amplifier does not reach a value equal
the input to the class -C stage. If the mod- to the d -c voltage applied to the amplifier
and, consequently, the d -c plate voltage im-
ulator is a class -B audio amplifier, the aver-
pressed across the r -f tube must be reduced
age power required of it may vary from
to a value equal to the maximum available
one -quarter to more than one -half the class - a -c peak voltage if 100% modulation is to
C input depending on the waveform. How- be obtained.
ever, the peak power output of any modu- A higher degree of distortion can be toler-
lator must be equal to the class-C input to ated in low -power emergency phone trans-
be modulated. mitters which use a pentode modulator tube,
www.americanradiohistory.com
HANDBOOK Plate Modulation 291
and the series resistor and bypass capacitor though the average power for this value of
are usually omitted in such transmitters. peak varies widely depending on the mod-
Class -B High -level class -B plate ulation waveform, being greater than 50 per-
Plate Modulation modulation is the least cent for speech that has been clipped and
expensive method of plate filtered, SO percent for a sine wave, and
modulation. Figure 8 shows a conventional about 25 percent for typical unclipped
class -B plate -modulated class -C amplifier. speech tones.
The statement that the modulator output Modulation The modulation transformer is
power must be one -half the class-C input for Transformer a device for matching the load
100 percent modulation is correct only if Calculations impedance of the class -C am-
the waveform of the modulating power is a plifier to the recommended load
sine wave. Where the modulator waveform is impedance of the class -B modulator tubes.
unclipped speech waveforms, the average Modulation transformers intended for com-
modulator power for 100 percent modulation munications work are usually designed to
is considerably less than one -half the class -C carry the class -C plate current through their
input. secondary windings, as shown in figure 8.
Power Relations in It has been determined The manufacturer's ratings should be con-
Speech Waveforms experimentally that the sulted to ensure that the d -c plate current
ratio of peak -to- average passed through the secondary winding does
power in a speech waveform is approximately not exceed the maximum rating.
4 to 1 as contrasted to a ratio of 2 to 1 in a A detailed discussion of the method of
sine wave. This is due to the high harmonic making modulation transformer calculations
content of such a waveform, and to the fact has been given in Chapter Six. However, to
that this high harmonic content manifests emphasize the method of making the calcula-
itself by making the wave unsymmetrical tion, an additional example will be given.
and causing sharp peaks or "fingers" of high Suppose we take the case of a class -C
energy content to appear. Thus for unclipped amplifier operating at a plate voltage of
speech, the average modulator plate current, 2000 volts with 225 ma of plate current.
plate dissipation, and power output are ap- This amplifier would present a load resist-
proximately one -half the sine wave values ance of 2000 divided by 0.225 ampere or
for a given peak output power. 8888 ohms. The plate power input would be
Both peak power and average power are 2000 times 0.225 or 450 watts. By reference
necessarily associated with waveform. Peak to Chapter Six we see that a pair of 811 tubes
power is just what the name implies; the operating at 1500 plate volts will deliver
power at the peak of a wave. Peak power, 225 watts of audio output. The plate -to-
although of the utmost importance in mod- plate load resistance for these tubes under
ulation, is of no great significance in a -c the specified operating conditions is 18,000
power work, except insofar as the average ohms. Hence our problem is to match the
power may be determined from the peak class -C amplifier load resistance of 8888
value of a known waveform. ohms to the 18,000 -ohm load resistance re-
There is no time element implied in the quired by the modulator tubes.
definition of peak power; peak power may be A 200- to 300 -watt modulation trans-
instantaneous-and for this reason average former will be required for the job. If the
power, which is definitely associated with taps on the transformer are given in terms
time, is the important factor in plate dissi- of impedances it will only be necessary to
pation. It is possible that the peak power of connect the secondary for 8888 ohms (or a
a given waveform be several times the aver- value approximately equal to this such as
age value; for a sine wave, the peak power is 9000 ohms) and the primary for 18,000
twice the average value and for unclipped ohms. If it is necessary to determine the
speech the peak power is approximately four proper turns ratio required of the transform-
times the average value. For 100 percent er it can be determined in the following
modulation, the peak (instantaneous) audio manner. The square root of the impedance
power must equal the class -C input, al- ratio is equal to the turns ratio, hence:
292 Amplitude Modulation THE RADIO
www.americanradiohistory.com
HANDBOOK Plate Modulation 293
e+ 8+
B+ 5.5 e+
B+ e+
Figure 9
any plate -modulated transmitter. The com- inexpensive modulator of grid modulation.
bined screen and plate current is divided into Cathode modulation consists essentially of a
the plate voltage in order to obtain the mixture of the two.
class -C amplifier load impedance. The peak The efficiency of the average well- designed
audio power required to obtain 100 percent plate -modulated transmitter is in the vicinity
modulation is equal to the d -c power input of 75 to 80 percent, with a compromise at
to the screen, screen resistor, and plate of the perhaps 77.5 percent. On the other hand, the
modulated r -f stage. efficiency of a good grid -modulated trans-
mitter may run from 28 to perhaps 40 per-
13 -5 Cathode Modulation cent with the average falling at about
34 percent. Now since cathode modulation
Cathode modulation offers a workable consists of simultaneous grid and plate mod-
compromise between the good plate efficiency ulation, in phase with each other, we can
but expensive modulator of high-level plate theoretically obtain any efficiency from
modulation, and the poor plate -efficiency but about 34 to 77.5 percent from our cathode-
294 Amplitude Modulation THE RADIO
modulated stage, depending on the relative moved for cathode modulation, as shown in
percentages of grid and plate modulation. figure 11. With low -p triodes, some extra
Since the system is a compromise between grid bias (over and above that amount sup-
the two fundamental modulation arrange- plied by the cathode modulator) may be
ments, a value of efficiency approximately needed to achieve proper linearity of the
half way between the two would seem to be modulated stage. In any case, proper oper-
the best compromise. Experience has proved ation of a cathode - modulated stage should
this to be the case. A compromise efficiency be determined by examining the modulated
of about 56.5 percent, roughly half way be- output waveform of the stage on an oscil-
tween the two limits, has proved to be opti- loscope.
mum. Calculation has shown that this value
of efficiency can be obtained from a cathode - Excitation The r -f driver for a cathode -
modulated amplifier when the audio- frequen- modulated stage should have
cy modulating power is approximately 20 about the same power output capabilities as
percent of the d -c input to the cathode - would be required to drive a c -w amplifier
modulated stage. to the same input as it is desired to drive the
cathode- modulated stage. However, some
An Economical Series cathode modulation is form of excitation control should be avail-
Series Cathode ideally suited as an economi- able since the amount of excitation power
Modulator cal modulating arrangement has a direct bearing on the linearity of a
for a high -power triode c -w cathode -modulated amplifier stage. If link
transmitter. The modulator can be con- coupling is used between the driver and the
structed quite compactly and for a minimum modulated stage, variation in the amount of
component cost since no power supply is re- link coupling will afford ample excitation
quired for it. When it is desired to change variation. If much less than 40 percent plate
over from c -w to 'phone, it is only neces- modulation is employed, the stage begins to
sary to cut the series modulator into the resemble a grid -bias modulated stage, and
cathode- return circuit of the c -w amplifier the necessity for good r -f regulation will
stage. The plate voltage for the modulator apply.
tubes and for the speech amplifier is taken
from the cathode voltage drop of the modu- Cathode Modulation Cathode modulation has
lated stage across the modulator unit. of Tetrodes not proved too satisfac-
Figure 10 shows the circuit of such a tory for use with beam
modulator, designed to cathode -modulate a tetrode tubes. This is a result of the small
class -C amplifier using push -pull 810 tubes, excitation and grid -swing requirements for
running at a supply voltage of 2500, and such tubes, plus the fact that some means
with a plate input of 660 watts. The mod- for holding the screen voltage at the po-
ulated stage runs at about 50 percent effi- tential of the cathode as far as audio is con-
ciency, giving a power output of nearly 350 cerned is usually necessary. Because of these
watts, fully modulated. The voltage drop factors, cathode modulation is not recom-
across the cathode modulator is 400 volts, mended for use with tetrode r -f amplifiers.
allowing a net plate to cathode voltage of
2100 volts on the final amplifier. The plate 13 -6 The Doherty and the
current of the 810's should be about 330 ma, Terman- Woodyard
and the grid current should be approximate-
ly 40 ma, making the total cathode current Modulated Amplifiers
of the modulated stage 370 ma. Four parallel
6L6 modulator tubes can pass this amount These two amplifiers will be described to-
of plate current without difficulty. It must gether since they operate on very similar
be remembered that the voltage drop across principles. Figure 12 shows a greatly simpli-
the cathode modulator is also the cathode fied schematic diagram of the operation of
bias of the modulated stage. In most cases, both types. Both systems operate by virtue
no extra grid bias is necessary. If a bias sup- of a carrier tube, (V1 in both figures 12 and
ply is used for c -w operation, it may be re- 13) which supplies the unmodulated carrier,
HAN DBOOK Cathode Modulation 295
TO CATHODE -
MODULATED
STAGE
6AU6 6AU6 6L6 6L6 6L6 6L6
500 K
.002
1o6,iw
e.
ALL KES /STORS 0.5 WATT UNLESS CAUTION - FILAMENTS OF 646 TUBES MUST BE AT OPERATING
OTHERWISE NOTED TEMPERATURE BEFORE PLATE VOLTAGE IS APPLIED
ALL CAPACITORS IN /IF UNLESS TO MODULATED AMPLIFIER.
OTHERWISE NOTED.
Figure 10
and whose output is reduced to supply neg- until at the peak of the modulation cycle it
ative peaks, and a peak tube, (V2) whose is contributing enough power so that the
function is to supply approximately half the impedance at the load end of the line is equal
positive peak of the modulation cycle and to R, instead of the R/2 that is presented
whose additional function is to lower the under the carrier conditions. This is true
load impedance on the carrier tube so that because at a positive modulation peak (since
it will be able to supply the other half of it is delivering full power) the peak tube
the positive peak of the modulation cycle. subtracts a negative resistance of R/2 from
The peak tube is able to increase the the load end of the line.
output of the carrier tube by virtue of an Now, since under the peak condition of
impedance - inverting line between the plate modulation the load end of the line is termi-
circuits of the two tubes. This line is de-
signed to have a characteristic impedance of R -F AMPLIFIER
one -half the value of load into which the
carrier tube operates under the carrier con-
ditions. Then a load of one-half the charac-
teristic impedance of the quarter -wave line
is coupled into the output. By experience
with quarter -wave lines in antenna -matching
circuits we know that such a line will vary
the impedance at one end of the line in such
a manner that the geometric mean between
the two terminal impedances will be equal
to the characteristic impedance of the line.
Thus, if we have a value of load of one -half
4 -6L6'5
the characteristic impedance of the line at UC 64U6 6405
one end, the other end of the line will present CATHODE
MODULATOR
a value of twice the characteristic imped-
ance of the lines to carrier tube V,.
This is the situation that exists under the
carrier conditions when the peak tube merely Figure 11
floats across the load end of the line and
contributes no power. Then as a positive CATHODE -MODULATOR INSTALLATION
peak of modulation comes along, the peak SHOWING PHONE -CW TRANSFER
tube starts to contribute power to the load SWITCH
;
296 Amplitude Modulation THE RADIO
nated in R ohms instead of R /2, the imped-
ance at the carrier -tube will be reduced from
2R ohms to R ohms. This again is due to the
impedance- inverting action of the line. Since
o - t
-JXRo
Ci
Vt
j6
ELECTRICAL A/4
LINE ZD R
2
LOAD
www.americanradiohistory.com
HANDBOOK Speech Clipping 297
-
1001b POB. MO014ATIQN
Figure 14 AV R
SPEECH -WAVEFORM
AMPLITUDE
MODULATION
11, 100% NEG MODULATION
Speech Waveform Dissymmetry earlier in that phase shift between the clipper -filter
this chapter. The effect of deriving proper stage and the final modulated amplifier be
speech polarity from the speech amplifier is kept to a minimum. However, if there is
shown in figure 14. phase shift after the clipper -filter the system
A much more desirable and effective meth- does not completely break down. The pres-
od of obtaining speech clipping is actually to ence of phase shift merely requires that the
employ a clipper circuit in the earlier stages audio gain following the clipper -filter be
of the speech amplifier, and then to filter out reduced to the point where the cant applied
the objectionable distortion components by to the clipped speech waves still cannot
means of a sharp low -pass filter having a cause overmodulation. This effect is illus-
cutoff frequency of approximately 3000 Hz. trated in figures 15 and 16.
Tests on clipper -filter speech systems have The cant appearing on the tops of the
shown that 6 db of clipping on voice is just square waves leaving the clipper -filter centers
noticeable, 12 db of clipping is quite accept- about the clipping level. Hence, as the fre-
able, and values of clipping from 20 to 25 quency being passed through the system is
db are tolerable under such conditions that a lowered, the amount by which the peak of
high degree of clipping is necessary to get the canted wave exceeds the clipping level
through heavy QRM or QRN. A signal with is increased.
12 db of clipping doesn't sound quite natur-
al but it is not unpleasant to listen to and is Phase -Shift a normal transmitter having
In
much more readable than an unclipped signal Correction moderate amount of phase
a
in the presence of strong interference. shift the cant applied to the
The use of a clipper -filter in the speech tops of the waves will cause overmodulation
amplifier, to be completely effective, requires on frequencies below those for which the
www.americanradiohistory.com
HANDBOOK Speech Clipping 299
AV RA
Figure 15
_ Nf,GQTIL( 4IPP1NQ.LEYEL ACTION OF A CLIPPER -FILTER
ON A SPEECH WAVE
Drawing A shows the incoming
speech way. before it reaches the
INCOMING SPEECH WAVE clipper stage. fi shows the output
of the clipper -filter, illustrating the
manner in which the peaks are clip-
ped and then the sharp edges of
the tipped way. removed by the
filter. C shows the effect of phase
shift in the stages following the
clipper -filter and the manner in
which the transmitter may b. ad-
NEGATIVE CLIPPING LEVEL justed for 100 percent modulation
of the "canted" peaks of the way.,
CLIPPED AND FILTERED SPEECH WAVE
the sloping top of the ware reach-
ing about 70 percent modulation.
at.
_100 -.b POSITIVE MOOUATION
AVERAGE LEVEL
gain following the clipper -filter has been into the microphone the distortion
adjusted unless remedial steps have been being introduced by the clipper -filter
taken. The following steps are advised: circuit is quite apparent but not ob-
jectionable. This amount of distortion
1. Introduce bass suppression into the will be apparent to the normal listener
speech amplifier ahead of the clipper - when 10 to 13 db of clipping is taking
filter. place.
2. Improve the low- frequency response 2. Tune a selective communications re-
characteristic insofar as it is possible in ceiver about 13kHz to one side or the
the stages following the clipper -filter. other of the frequency being transmit-
Feeding the plate current to the final ted. Use a short antenna or no antenna
amplifier through a choke rather than at all on the receiver so that the trans-
through the secondary of the modu- mitter is not blocking the receiver.
lation transformer will help materially. 3. Again, with normal talking into the
Even with the normal amount of im- microphone, adjust the gain follow-
provement which can be attained through the ing the clipper -filter to the point where
steps mentioned above there will still be an the sideband splatter is being heard,
amount of wave cant which must be com- and then slightly back -off the gain
pensated in some manner. This compensation after the clipper -filter until the splat-
can be done in either of two ways. The first ter disappears.
and simpler way is as follows: If the phase shift in the transmitter or
1. Adjust the speech gain ahead of the modulator is not excessive the adjustment
clipper -filter until with normal talking procedure given above will allow a clean
300 Amplitude Modulation THE RADIO
F 7.- 7r
MODULATOR SR4GY,1616
836 Cz
z
O o
rLi i C. 1
TO
la V PLATE- MODULATED
3000 % WAVE CLASS -C AMPLI FIER
FIL. TRANS. 7500 -10 000 OHMS
INAVTED LOAD
_ _ ML.
FOSUR
W}
O
>
W
Z
0 Ai 11 +5 MOD. 115-V A.0
Figure 17
+S R-F FINAL
O.
HIGH -LEVEL SPLATTER SUPPRESSOR
Ra_.
1000% WAVE This circuit is effective in reducing splatter
caused by negative -peak clipping in the mod-
ulated amplifier stage. The use of a two-section
Alter as shown is recommended, although
either a single m- derived or a constant -k
section may be used for greater my.
Suitable chokes, along with recommended ca-
pacitor values, are available from s I
manuf act s.
www.americanradiohistory.com
HANDBOOK Splatter Suppression 301
Figure 18
6AU6 6J6 20 H
(STANCOR 0ISI0
Figure 19
Figure 20
CLASS -C AMPLIFIER
a full 12 db of clipping is used. This means
that the Adj. Gain control may be advanced
some 12 db past the point where the clipping
action started. Clipping action should start
at 85 to 90 percent modulation when a sine
wave is used for circuit- adjustment purposes.
High -Level Even though we may have cut
Filters off all frequencies above 3000 or
3500 Hz through the use of a
filter system such as is shown in the circuits
CLASS -C STAGE
e+ MOD. 5+ R F.
Figure 21
manufacturers.
www.americanradiohistory.com
HAN D BOOK Splatter Suppression 303
of figures 19 and 20, higher frequencies may load resistance for the class -B modulator
again be introduced into the modulated wave tubes. Resistor R3 should be equal to the
by distortion in stages following the speech value of load resistance which the class -C
amplifier. Harmonics of the incoming audio modulated stage will present to the modu-
frequencies may be generated in the driver lator. The meter (V) can be any type of
stage for the modulator; they may be gen- a -c voltmeter. The indicating instrument on
erated in the plate circuit of the modulator; the secondary of the transformer can be
or they may be generated by nonlinearity in either a cathode -ray oscilloscope or a high -
the modulated amplifier itself. impedance a -c voltmeter of the vacuum -
Regardless of the point in the system fol- tube or rectifier type.
lowing the speech amplifier where the high With a setup as shown in figure 23 a plot
audio frequencies may be generated, these of output voltage against frequency is made,
frequencies can still cause a broad signal to at all times keeping the voltage across V
be transmitted even though all frequencies constant, using various values of capacitance
above 3000 or 3500 Hz have been cut off in for C, and C2 plus C3. When the proper
the speech amplifier. The effects of distortion values of capacitance have been determined
in the audio system following the speech am- which give substantially constant output up
plifier can be eliminated quite effectively to about 3000 or 3500 Hz and decreasing
through the use of a frost-modulator filter. output at all frequencies above, high -voltage
Such a filter may be used between the modu- mica capacitors can be substituted if receiv-
lator plate circuit and the r -f amplifier ing types were used in the tests and the
which is being modulated. transformer connected to the modulator and
This filter may take three general forms in class -C amplifier.
a normal case of a class -C amplifier plate With the transformer reconnected in the
modulated by a class -B modulator. The best transmitter a check of the modulated -wave
method is to use a high -level low -pass filter output of the transmitter should be made
as shown in figure 21 and discussed previous- using an audio oscillator as signal generator
ly. Another method which will give excellent and an oscilloscope coupled to the transmit-
results in some cases and poor results in ter output. With an input signal amplitude
others, dependent on the characteristics of fed to the speech amplifier of such amplitude
the modulation transformer, is to "build - that limiting does not take place, a substan-
out" the modulation transformer into a filter tially clean sine wave should be obtained on
section. This is accomplished as shown in the carrier of the transmitter at all input
figure 22 by placing mica capacitors of the frequencies up to the cutoff frequency of
correct value across the primary and second- the filter system in the speech amplifier and
ary of the modulation transformer. The of the filter which includes the modulation
proper values for capacitors C, and C2 must, transformer. Above these cutoff frequencies
in the ideal case, be determined by trial and
error. Experiment with a number of modu-
lators has shown, however, that if a 0.002 -
pfd. capacitor is used for C1, and if the sum
of C2 and C:, is made 0.004 -pfd (0.002 pfd AUDIO
OSCILLATOR
for C_ and 0.002 for C3) the ideal condition
of cutoff above 3000 Hz will be approached
in most cases with the "multiple- match"
type of modulation transformer.
If it is desired to determine the optimum Figure 23
values of the capacitors across the trans-
former this can be determined in several TEST SETUP FOR BUILDING -OUT
ways, all of which require the use of a cali- MODULATION TRANSFORMER
brated audio oscillator. One way is dia- Through the use of a test setup such as is
grammed in figure 23. The series resistors shown and the method described in the text
it is possible to determine the correct values
(R, and R2) should each be equal to 0.5 for a specified Alter characteristic in the built -
the value of the recommended plate -to -plate out modulation transformer.
304 Amplitude Modulation THE RADIO
tion or overload of the audio system. In
addition, if speech clipping is used, attenua-
tion of the lower audio frequencies before
the clipper will reduce phase shift and cant-
2 MIMMEillir ing of the clipper output.
, MMI5111/1 A simple method of bass suppression is to
reduce the size of the interstage coupling
capacitors in a resistance- coupled amplifier.
AMARECOMMENOEO CURVESfOR
Figure 24 shows the frequency characteris-
VO/C( COMMUN/CAT/ON tics caused by such a suppression circuit. A
, second simple bass -suppression circuit is to
place a small a -c /d -c type filter choke from
MIA 11
grid to ground in a speech -amplifier stage,
I_II.. 4
r =
RG= S00K
as shown in figure 25.
,
ited by the plate dissipation of the tube
under quiescent condition. The bias-shift
LINEAR PORTION
OF
CURVE
EG-1 P All PLATE
OUTPUT
SIGNAL
modulator is biased close to plate current
cutoff under no- signal condition (figure 27).
Resting plate current and plate dissipation
Ec
All! RESTING
PLATE
CURRENT
+EG
are therefore quite low. On application of an
audio signal, the bias of the stage is shifted
toward the class -A operating point, prevent-
CUTOFF
ing the negative peaks of the applied audio
BIAS voltage from cutting off the plate current of
RESTING BIAS VOLTAGE
-
the tube. As the audio voltage increases, the
-IP operating -bias point is shifted to the right
GRID INPUT on figure 27 until the class -A operating
SIGNAL
point is reached at maximum excitation.
Figure 26 The bias -shift voltage may be obtained
directly from the exciting signal by recti-
CHARACTERISTIC GRID-VOLTAGE fication, as shown in figure 28. A simple
PLATE-CURRENT CURVE FOR low -pass filter system is used that will pass
CLASS -A HEISING only the syllabic components of speech.
MODULATOR Enough negative bias is applied to the bias-
306 Amplitude Modulation THE RADIO
CLASS-C
AMPLIFIER circuit of the bias -shift modulator, and a
A
heavy -duty filter choke will serve as an im-
AUDIO BIAS -SHIFT pedance coupler for the modulated stage.
AMPLIFIER MODULATOR
RI
unison. The modulator bias, therefore varies
+ at the same rate. Excitation for the bias
ADJUST
RESTING
8/AS
control tube is obtained from the audio sig-
nal through potentiometer R1 which regu-
BIAS CONTROL
TUBE
BIAS CONTROL
RECTIFIER lates the amplitude of the control signal.
The audio signal is rectified by the bias con-
NEGATIVE
trol rectifier, and filtered by network R3 -C1
MODULATOR BIAS in the grid circuit of the bias control tube.
The "parallel" control system is illustrated
Figure 29 in figure 30. Resting bias for the modulator
SERIES CONTROL CIRCUIT is obtained from the voltage divider R2 -R4.
FOR BIAS -SHIFT MODULATOR Potentiometer R2 adjusts the resting bias
level, determining the static plate current of
Ti,. internal resistant* of the bias control tub* the modulator. Resistor R3 serves as a bias
is varied at a syllabic rat* to thong* the
operating bias of the modulator tube. resistor for the control tube, reducing its
HANDBOOK Bias -Shift Modulation 307
plate current to a low level. When an audio conjunction with bias resistor R3 form a syl-
signal is applied via R, to the grid of the labic filter for the control bias that is applied
control tube the internal resistance is low- to the modulator stage.
ered, decreasing the shunt resistance across A large value of plate dissipation is re-
R_. The negative modulator bias is therefore quired for the bias -shift modulator tube.
reduced. The bias axis of the modulator is For plate voltages below 1500, the 211
shifted from the cutoff region to a point on (VT -4C) may be used, while the 304 -TL is
the linear portion of the operating curve. suitable for voltages up to 3000. As with
The amount of bias -shift is controlled by the normal class-A amplifiers, low -p. tubes func-
setting of potentiometer R1. Capacitor C, in tion best in this circuit.
CHAPTER FOURTEEN
Frequency Modulation
Exciter systems for f -m and single -side- In this chapter various points of difference
band transmission are basically similar in between frequency- modulation and ampli-
that modification of the signal in accordance tude- modulation transmission and reception
with the intelligence to be transmitted is will be discussed and the advantages of
normally accomplished at a relatively low frequency -modulation for certain types of
level. Then the intelligence -bearing signal is communication pointed out. Since the dis-
amplified to the desired power level for ulti- tinguishing features of the two types of
mate transmission. True, amplifiers for the transmission lie entirely in the modulating
two types of signals are basically different; circuits at the transmitter and in the detec-
linear amplifiers of the class -A or class -B tor and limiter circuits in the receiver, these
type being used for SSB signals,while class -C parts of the communication system will re-
or nonlinear class-B amplifiers may be used ceive the major portion of attention.
for f -m amplification. But the principle of
low -level generation and subsequent amplifi- Modulation Modulation is the process of al-
cation is standard for both types of trans- tering a radio wave in accord
mission. with the intelligence to be transmitted. The
nature of the intelligence is of little impor-
tance as far as the process of modulation is
14 -1 Frequency concerned; it is the method, by which this
Modulation intelligence is made to give a distinguishing
characteristic to the radio wave which will
enable the receiver to convert it back into
The use of frequency modulation and the intelligence,that determines the type of mod-
allied system of phase modulation has become ulation being used.
of increasing importance in recent years. Figure 1 is a drawing of an r -f carrier
For amateur communication, frequency and amplitude- modulated by a sine-wave audio
phase modulation offer important advantages voltage. After modulation the resultant mod-
in the reduction of broadcast and TV inter- ulated r -f wave is seen still to vary about
ference and in the elimination of the costly the zero axis at a constant rate, but the
high -level modulation equipment most com- strength of the individual r -f waves is
monly employed with amplitude modulation. proportional to the amplitude of the mod-
For broadcast work frequency modulation ulation voltage.
offers an improvement in signal -to -noise Tn figure 2, the carrier of figure 1 is shown
ratio for the high field intensities available frequency- modulated by the same modula-
in the local -coverage area of f -m and TV ting voltage. Here it may be seen that mod-
broadcast stations. ulation voltage of one polarity causes the
www.americanradiohistory.com
Frequency Modulation 309
CARRIER
FREQUENCY
Figure 3
J A -M SIDE FREQUENCIES
FIGURE t FIGURE 2
For each a -m modulating frequency, a pair of
side freq less is produced. The side frequen-
cies are spaced away from the carrier by an
amount equal to the modulation frequency, and
A -M AND F -M WAVES their amplitude is directly proportional to the
amplitude of the modulation. The amplitude of
Figure 1 shows a sketch of the scope pattern the carrier does not change under modulation.
of an amplitude -modulated wave at the bot-
tom. The center sketch shows the modulating
wave and the upper sketch shows the carrier
wave. output level that they are capable of produc-
Figure 2 shows at the bottom a sketch of a ing the additional output without distortion
frequency -modulated wave. in this case the
center sketch also shows the modulating wave in the low -level system. On the other hand,
and the upper sketch shows the carrier wave. a frequency- modulated transmitter requires
Note that the carrier wave and the modulating an insignificant amount of power from the
wave are the same in either case, but that
the waveform of the modulated wave is quite modulator and needs no provision for in-
different in the two cases. creased power output on modulation peaks.
All of the stages between the oscillator and
carrier frequency to decrease, as shown by the antenna may be operated as high-efficien-
the fact that the individual r -f waves of cy class -B or class -C amplifiers or frequency
the carrier are spaced farther apart. A mod- multipliers.
ulating voltage of the opposite polarity causes
the frequency to increase, and this is shown Currier -Wove The second characteristic of
by the r -f waves being compressed together to Distortion f -m and a -m waves revealed
allow more of them to be completed in a by figures 1 and 2 is that both
given time interval. types of modulation result in distortion of
Figures 1 and 2 reveal two very important the r-f carrier. That is, after modulation, the
characteristics about amplitude- and fre- r -f waves are no longer sine waves, as they
quency- modulated waves. First, it is seen would be if no frequencies other than the
that while the amplitude (power) of the sig- fundamental carrier frequency were present.
nal is varied in a -m transmission, no such It may be shown in the amplitude- modula-
variation takes place in frequency modula- tion case illustrated, that there are only two
tion. In many cases this advantage of fre- additional frequencies present, and these are
quency modulation is probably of equal or the familiar side frequencies, one located on
greater importance than the widely publi- each side of the carrier, and each spaced
cized noise- reduction capabilities of the sys- from the carrier by a frequency interval
tem. When 100 percent amplitude modula- equal to the modulation frequency. In regard
tion is obtained, the average power output to frequency and amplitude, the situation is
of the transmitter must be increased by SO as shown in figure 3. The strength of the
percent. This additional output must be carrier itself does not vary during modula-
supplied either by the modulator itself, in tion, but the strength of the side frequencies
the high -level system, or by operating one or depends on the percentage of modulation.
more of the transmitter stages at such a low At 100 percent modulation the power in the
310 Frequency Modulation THE RADIO
The relative strengths of the f -m carrier deviation ratio, however, the noise suppres-
and the various side frequencies depend sion becomes progressively better as the
directly on the modulation index, these rela- deviation ratio is increased.
tive strengths varying widely as the modula- For high -fidelity f -m broadcasting pur-
tion index is varied. In the preceding ex- poses, a deviation ratio of S is ordinarily
ample, for instance, side frequencies occur used, the maximum audio frequency being
on the high side of 1000 kHz at 1002, 1004, 1 5,000 Hz, and the peak deviation at full
1006, 1008, 1010, 1012, etc., and on the low modulation being 75 kHz. Since a swing of
frequency side at 998, 996, 994, 992, 990, 150 kHz is covered by the transmitter, it is
988, etc. In proportion to the unmodulated obvious that wide -band f -m transmission
carrier strength (100 percent) , these side must necessarily be confined to the vhf range
frequencies have the following strengths, as or higher, where room for the signals is avail-
indicated by a modulation index of S: 1002 able.
and 998 -33 percent; 1004 and 996 -5 per- In the case of television sound, the devia-
cent; 1006 and 994 -36 percent; 1008 and tion ratio is 1.67; the maximum modulation
992 -39 percent; 1010 and 990 -26 percent; frequency is 1 5,000 Hz, and the transmitter
1012 and 998 -13 percent. The carrier deviation for full modulation is 25 kHz. The
strength (1000 kHz) will be 18 percent of sound carrier frequency in a standard TV
its modulated value. Changing the amplitude signal is located exactly 4.5 MHz higher
of the modulating signal will change the than the picture carrier frequency. In the
deviation, and thus the modulation index intercarrier TV sound system, which is wide-
will be changed, with the result that the ly used, this constant difference between the
side frequencies, while still located in the picture carrier and the sound carrier is em-
same places, will have different strength ployed within the receiver to obtain an f -m
values from those given above. subcarrier at 4.5 MHz. This 4.S MHz sub -
The deviation ratio is similar to the modu- carrier then is demodulated by the f -m de-
lation index in that it involves the ratio tector to obtain the sound signal which
between a modulating frequency and devia- accompanies the picture.
tion. In this case, however, the deviation in
question is the peak frequency shift obtained Narrow -Band Narrow -band f -m trans-
under full modulation, and the audio fre- F -M Transmission mission has becomestand-
quency to be considered is the maximum ardized for use by the
audio frequency to be transmitted. When mobile services such as police, fire, and
the maximum audio frequency to be trans- taxicab communications, and is also au-
mitted is 5000 Hz, for example, a deviation thorized for amateur work in portions of
ratio of 3 would call for a peak deviation of each of the amateur radiotelephone bands.
3 X 5000, or 15 kHz at full modulation. A maximum deviation of 15 kHz has
The noise -suppression capabilities of frequen- been standardized for the mobile and com-
cy modulation are directly related to the mercial communication services, while a
deviation ratio. As the deviation ratio is in- maximum deviation of 3 kHz is authorized
creased, the noise suppression becomes better for amateur narrow -band f -m communica-
if the signal is somewhat stronger than the tion.
noise. Where the noise approaches the signal
in strength, however, low deviation ratios Bandwidth Required by As the above discus -
allow communication to be maintained in Frequency Modulation sion has indicated,
many cases where high- deviation -ratio fre- many side frequencies
quency modulation and conventional ampli- are set up when a radio- frequency carrier is
tude modulation are incapable of giving frequency modulated; theoretically, in fact,
service. This assumes that a narrow -band an infinite number of side frequencies is
f -m receiver is in use. For each value of r -f formed. Fortunately, however, the ampli-
signal -to -noise ratio at the receiver, there is a tudes of those side frequencies falling out-
maximum deviation ratio which may be used, side the frequency range over which the
beyond which the output audio signal -to- transmitter is swung are so small that most
noise ratio decreases. Up to this critical of them may be ignored. In f -m transmis-
312 Frequency Modulation THE RADIO
Is 9
19s
14-2 Direct F -M Circuits
-6 kHz -3 kHz +3 kHz + 6 kHz
www.americanradiohistory.com
HANDBOOK Direct F -M Circuits 313
AUDIO
IN OSCILLATOR IN
1.75- MN= RANGE
C3..00
AUDIO
- 0.1
Figure 6 Figure 7
6AU6 500 A
Hew TO GRID OR CATHODE
OF V.F.O.
Figure 8
the r -f excitation voltage to the grid of the Adjusting the One of the simplest methods
reactance tube is obtained effectively through Phase Shift of adjusting the phase shift to
reversing the R and C3 of figure 6. In this the correct amount is to place
circuit a small capacitance is used to couple a pair of earphones in series with the oscilla-
r.f. into the grid of the reactance tube, with tor cathode -to- ground circuit and adjust the
a relatively small value of resistance from phase -shift network until minimum sound is
grid to ground. This circuit has the advan- heard in the phones when frequency modula-
tage that the grid of the tube is at relatively tion is taking place. If an electron -coupled
low impedance with respect to r.f. However, or Hartley oscillator is used, this method re-
the circuit normally is not suitable for oper- quires that the cathode circuit of the oscilla-
ation above a few MHz due to the shunting tor be inductively or capacitively coupled to
capacitance within the tube from grid to the grid circuit, rather than tapped on the
ground. grid coil. The phones should be adequately
Either of the reactance -tube circuits may bypassed to r.f. of course.
be used with any of the common types of
oscillators. The reactance modulator of fig- Stabilization Due to the presence of the
ure 6 is shown connected to the high- imped- reactance -tube frequency mod-
ance point of a conventional hot -cathode ulator, the stabilization of an f -m oscillator
Hartley oscillator, while that of figure 7 is in regard to voltage changes is considerably
shown connected across the low- impedance more involved than in the case of a simple
capacitors of a series -tuned Clapp oscillator. self -controlled oscillator for transmitter fre-
There are several possible variations of the quency control. If desired, the oscillator it-
basic reactance -tube modulator circuits self may be made perfectly stable under
shown in figures 6 and 7. The audio input voltage changes, but the presence of the
may be applied to the suppressor grid, rather frequency modulator destroys the beneficial
than the control grid, if desired. Another effect of any such stabilization. It thus be-
modification is to apply the audio to a grid comes desirable to apply the stabilizing ar-
other than the control grid in a mixer or raneement to the modulator as well as the
pentagrid converter tube which is used as oscillator. If the oscillator itself is stable
the modulator. under voltage changes, it is only necessary to
Generally it will be found that the trans - apply voltage- frequency compensation to the
conductance variation per volt of control - modulator.
element voltage variation will be greatest
when the control (audio) voltage is applied Reactance -Tube Two simple reactance -tube
to the control grid. In cases where it is Modulators modulators that may be
desirable to separate completely the audio applied to an existing vfo
and r -f circuits, however, applying audio are illustrated in figures 8 and 9. The circuit
voltage to one of the other elements will of figure 8 is extremely simple, yet effective.
often be found advantageous despite the Only two tubes are used exclusive of the
somewhat lower sensitivity. voltage regulator tubes which perhaps may
HANDBOOK Direct F -M Modulation 315
.01
Figure 9
be already incorporated in the vfo. A 6AU6 the best modulation level consistent with
serves as a high -gain voltage amplifier stage, minimum sideband splatter. Minimum clip-
and a 6CL6 is used as the reactance modula- ping occurs when the Adj. Clip. potentiome-
tor since its high value of transconductance ter is set for maximum voltage on the plates
will permit a large value of lagging current of the 6ALS clipper tube. As with the case
to be drawn under modulation swing. The of all reactance modulators, a voltage -regu-
unit should be mounted in close proximity to lated plate supply is required.
the vfo so that the lead from the 6CL6 to
the grid circuit of the oscillator can be as Linearity Test It
is almost a necessity to run
short as possible. A practical solution is to static test on the reactance -
a
mount the reactance modulator in a small tube frequency modulator to determine its
box on the side of the vfo cabinet. linearity and effectiveness, since small changes
By incorporating speech clipping in the in the values of components, and in stray
reactance modulator unit, a much more ef- capacitances will almost certainly alter the
fective use is made of a given amount of modulator characteristics. A frequency -
deviation. When the f -m signal is received versus- control -voltage curve should be plot-
on an a -m receiver by means of slope detec- ted to ascertain that equal increments in
tion, the use of speech clipping will be control voltage, both in a positive and a
noticed by the greatly increased modulation negative direction, cause equal changes in
level of the f -m signal, and the attenuation frequency. If the curve shows that the mod-
of the center frequency null of no modula- ulator has an appreciable amount of non-
tion. In many cases, it is difficult to tell a linearity, changes in bias, electrode voltages,
speech- clipped f -m signal from the usual r -f excitation, and resistance values may be
a -m signal. made to obtain a straight -line characteristic.
A more complex f -m reactance modulator
incorporating a speech clipper is shown in
TO MODULATOR
figure 9. A 12AX7 double- triode speech am- CONTROL CLEMENT
www.americanradiohistory.com
HANDBOOK Phase Modulation 317
REACTANCE
TUBE
CRYSTAL
OSCILLATOR
TUBE
--yNExr STAGE
R-F INPUT
fo
6U6 PHASE- MODULATED
OUTPUT
LOW- C
TANA
CIRCUIT +B Zoo v.
01
AUDIO IN
RFC
XL ABOUT 1500 (, AT fo
XC ABOUT 750 R AT f0
AUDIO +6
Figure 11 Figure 13
6C L6
250 PF r --Hi-_ .-TO DOUBLER STAGES
Figure 12
REACTANCE MODULATOR FOR 10, 15, AND 20 METER OPERATION
318 Frequency Modulation THE RADIO
capacitors should be adjusted for maximum when the modulation index has a value of
phase -modulation capabilities of the circuit. 2.405-in other words, when the deviation
Resonance of the circuit is established by is 2.405 times the modulation frequency.
the iron slug of coil L1 -L2. A 6CL6 is used For example, if a modulation frequency of
as a doubler to 7 MHz and delivers approxi- 1000 Hz is used, and the modulation is in-
mately 2 watts on this band. Additional creased until the first carrier null is obtained,
doubler stages may be added after the 6CL6 the deviation will then be 2.405 times the
stage to reach the desired band of operation. modulation frequency, or 2.405 kHz. If the
Still another p -m circuit, which is quite modulating frequency happened to be 2000
widely used commercially, is shown in figure Hz, the deviation at the first null would be
13. In this circuit L and C are made reso- 4.810 kHz. Other carrier nulls will be ob-
nant at a frequency which is 0.707 times the tained when the index is 5.52, 8.654, and
operating frequency. Hence at the operating at increasing values separated approximately
frequency the inductive reactance is twice by r. The following is a listing of the modu-
the capacitive reactance. A cathode -follower lation index at successive carrier nulls up to
tube acts as a variable resistance in series the tenth:
with L and C to make up the tank cir-
cuit. The operating point of the cathode fol-
lower should be chosen so that the effective Zero carrier Modulation
resistance in series with the tank circuit point no. index
(made up of the resistance of the cathode - 1 2.405
follower tube in parallel with the cathode 2 5.520
bias resistor of the cathode follower) is equal 3 8.654
to the capacitive reactance of the tank ca- 4 11.792
pacitor at the operating frequency. The 5 14.931
circuit is capable of about plus or minus %2 6 18.071
radian deviation with tolerable distortion. 7 21.212
8 24.353
Measurement When a single- frequency mod - 9 27.494
of Deviation ulating voltage is used with 10 30.635
an f -m transmitter the rela-
tive amplitudes of the various sidebands and
the carrier vary widely as the deviation is The only equipment required for making
varied by increasing or decreasing the amount the measurements is a calibrated audio oscil-
of modulation. Since the relationship be- lator of good wave form, and a communi-
tween the amplitudes of the various side - cation receiver equipped with a beat oscillator
bands and carrier to the audio modulating and crystal filter. The receiver should be
frequency and the deviation is known, a sim- used with its crystal filter set for minimum
ple method of measuring the deviation of bandwidth to exclude sidebands spaced from
a frequency- modulated transmitter is possi- the carrier by the modulation frequency.
ble. In making the measurement, the result The unmodulated carrier is accurately tuned
is given in the form of the modulation index on the receiver with the beat oscillator oper-
for a certain amount of audio input. As pre- ating. Then modulation from the audio oscil-
viously described, the modulation index is lator is applied to the transmitter, and the
the frequency of the audio modulation. modulation is increased until the first car-
The measurement is made by applying a rier null is obtained. This carrier null will
sine-wave audio voltage of known frequency correspond to a modulation index of 2.405,
to the transmitter, and increasing the mod- as previously mentioned. Successive null
ulation until the amplitude of the carrier points will correspond to the indices listed
component of the frequency -modulated wave in the table.
reaches zero. The modulation index for zero A volume indicator in the transmitter
carrier may then be determined from the audio system may be used to measure the
table below. As may be seen from the table, audio level required for different amounts of
the first point of zero carrier is obtained deviation, and the indicator thus calibrated
HANDBOOK F -M Reception 319
MIRCR I -r LIMITER --
rONCY
DCRCTCCUCTOq
AUDIO
AMPLIFIER AMP.
(DISCRIMINATOR)
OSCILLATOR
R C OUCNCY
Figure 14 Figure 15
short. Any frequency variation which ex- depend on the discriminator frequency, the
ceeds the linear portion will cause distortion Q of the tuned circuits, and the value of the
of the recovered audio. It is also obvious by diode load resistors. As the intermediate (and
inspection of figure 15 that an a -m receiver discriminator) frequency is increased, the
used in this manner is vulnerable to signals peaks must be separated further to secure
on the peak of the resonance curve and also good linearity and output. Within limits, as
to signals on the other side of the resonance the diode load resistance or the Q is reduced,
curve. Further, no noise -limiting action is the linearity improves, and the separation
afforded by this type of reception. This sys- between the peaks must be greater.
tem, therefore, is not recommended for f -m
reception, although it may be widely used by Foster -Seeley The most widely used form of
amateurs for occasional nbfm reception. Discriminator discriminator is that shown in
Travis Discriminator Another form of fre- figure 18. This type of dis-
quency detector or dis- criminator yields an output voltage- versus-
criminator, is shown in figure 16. In this frequency characteristic similar to that
arrangement two tuned circuits are used, one shown in figure 19. Here, again, the output
tuned on each side of the i -f amplifier fre- voltage is equal to the algebraic sum of the
quency, and with their resonant frequencies voltages developed across the load resistors of
spaced slightly more than the expected the two diodes, the resistors being connected
transmitter swing. Their outputs are com- in series to ground. However, this Foster -
bined in a differential rectifier so that the Seeley discriminator requires only two tuned
voltage across series load resistors R, and R, circuits instead of the three used in the pre-
is equal to the algebraic sum of the individ- vious discriminator. The operation of the
ual output voltages of each rectifier. When circuit results from the phase relationships
a signal at the i -f midfrequency is received, existing in a transformer having a tuned
the voltages across the load resistors are equal secondary. In effect, as a close examination
and opposite, and the sum voltage is zero. of the circuit will reveal, the primary cir-
As the r -f signal varies from the midfre- cuit is in series for r.f. with each half of
quency, however, these individual voltages the secondary to ground. When the received.
become unequal, and a voltage having the signal is at the resonant frequency of the
polarity of the larger voltage and equal to secondary, the r-f voltage across the second-
the difference between the two voltages ap- ary is 90 degrees out of phase with that
pears across the series resistors, and is applied across the primary. Since each diode is con-
to the audio amplifier. The relationship be- nected across one half of the secondary
tween frequency and discriminator output winding and the primary winding in series,
voltage is shown in figure 17. The separation the resultant r -f voltages applied to each are
of the discriminator peaks and the linearity equal, and the voltages developed across each
of the output voltage- versus - frequency curve diode load resistor are equal and of opposite
www.americanradiohistory.com
HANDBOOK F -M Reception 321
each other. At the center frequency to which supplied with a large amount of signal, so
the discriminator transformer is tuned, the that the amplitude of its output will not
voltage appearing at the top of the 1- change for rather wide variations in ampli-
megohm potentiometer will be one -half the tude of the signal. Noise, which causes little
d -c voltage appearing at the avc output frequency modulation but much amplitude
terminal, since the contribution of each diode modulation of the received signal, is virtual-
will be the same. However, as the input ly wiped out in the limiter.
frequency varies to one side or the other of The voltage across the grid resistor varies
the tuned value (while remaining within the with the amplitude of the received signal.
passband of the i -f amplifier feeding the For this reason, conventional amplitude-
detector) the relative contributions of the modulated signals may be received on the
two diodes will be different. The voltage ap- f -m receiver by connecting the input of the
pearing at the top of the 1- megohm volume audio amplifier to the top of this resistor,
control will increase for frequency deviations rather than to the discriminator output.
in one direction and will decrease for fre- When properly filtered by a simple RC cir-
quency deviations in the other direction cuit, the voltage across the grid resistor may
from the mean or tuned value of the trans- also be used as avc voltage for the receiver.
former. The audio output voltage is equal to When the limiter is operating properly avc is
the ratio of the relative contributions of the neither necessary nor desirable, however, for
two diodes, hence the name ratio detector. f -m reception alone.
The ratio detector offers several advan-
tages over the simple discriminator circuit. Receiver Design One of the most important
The circuit does not require the use of a Considerations factors in the design of an
limiter preceding the detector since the cir- f -m receiver is the frequen-
cuit is inherently insensitive to amplitude cy swing which it is intended to handle. It
modulation on an incoming signal. This will be apparent from figure 17 that if the
factor alone means that the r -f and i -f gain straight portion of the discriminator circuit
ahead of the detector can be much less than covers a wider range of frequencies than
the conventional discriminator for the same those generated by the transmitter, the audio
over-all sensitivity. Further, the circuit pro- output will be reduced from the maximum
vides avc voltage for controlling the gain of value of which the receiver is capable.
the preceding r -f and i -f stages. The ratio In this respect, the term modulation per-
detector is, however, susceptible to variations centage is more applicable to the f -m re-
in the amplitude of the incoming signal as ceiver than it is to the transmitter, since the
in any other detector circuit except the modulation capability of the communication
discriminator with a limiter preceding it, system is limited by the receiver bandwidth
so that avc should be used on the stage pre- and the discriminator characteristic; full
ceding the detector. utilization of the linear portion of the char-
acteristic amounts, in effect, to 100 percent
Limiters The limiter of an f -m receiver modulation. This means that some sort of
using a conventional discriminator standard must be agreed on, for any particu-
serves to remove amplitude modulation and lar type of communication, to make it un-
pass on to the discriminator a frequency - necessary to vary the transmitter swing to
modulated signal of constant amplitude; a accommodate different receivers.
typical circuit is shown in figure 21. The Two considerations influence the receiver
limiter tube is operated as an i -f stage with bandwidth necessary for any particular type
very low plate voltage and with grid -leak of communication. These are the maximum
bias, so that it overloads quite easily. Up to audio frequency which the system will
a certain point the output of the limiter will handle, and the deviation ratio which will
increase with an increase in signal. Above be employed. For voice communication, the
this point, however, the limiter becomes maximum audio frequency is more or less
overloaded, and further large increases in fixed at 3000 to 4000 Hz. In the matter of
signal will not give any increase in output. deviation ratio, however, the amount of
To operate successfully, the limiter must be noise suppression which the f -m system will
www.americanradiohistory.com
HANDBOOK F -M Receiver Design 323
ao K
nal of the discriminator. The receiver is now
a PF tuned back and forth across the frequency
455. MHz 100;i/
K
of the incoming signal, and the movement
bF IN
220K 100 of the voltmeter noted. When the receiver
is exactly tuned on the signal the voltmeter
PF vOLT-
- 0.1 39 N
2w T I-J.W. MILLER O/E-CS
METER
reading should t zero. When the receiver is
s.2wv.
AT 3 MA.
tuned to one side of center, the voltmeter
NOTE: ALL CAPACITORS IN JO' UNLESS OTHERWISE NOTED
reading should increase to a maximum value
ALL RESISTORS O. S WATT UNLESS OTHERWISE NOTED and then decrease gradually to zero as the
signal is tuned out of the passband of the
Figure 23 receiver. When the receiver is tuned to the
NBFM ADAPTER FOR 455 -kHz -F SYSTEM 1
other side of the signal the voltmeter should
increase to the same maximum value but
in the opposite direction or polarity, and
i -f system. The unit may be tuned by placing then fall to zero as the signal is tuned out
a high resistance d -c voltmeter across R, of the passband. It may be necessary to make
and tuning the trimmers of the i -f trans- small adjustments to C, and CL to make the
former for maximum voltage when an un- voltmeter read zero when the signal is tuned
modulated signal is injected into the i -f strip in the center of the passband.
CHAPTER FIFTEEN
Radioteletype Systems
MARK
r --
-,J
LETTER
^ -
*R" and made to print the letter corresponding
to the key pressed. Synchronization of ma-
chines is accomplished by means of start
and stop pulses transmitted with each char-
acter. An electromechanical device driven
by the motor of the teleprinter is released
SPACE
when a key is pressed and transmission of
22MS 22M5,22MS 22MS 22MS 22MS
the complete character is automatic.
163 MS
TIME -4. The receiving apparatus operates in re-
verse sequence, being set in operation by the
Figure 1 first pulse of a character sent by the trans-
mitter mechanism. While each character is
THE TELEPRINTER CODE sent at the speed of 60 w.p.m., actual trans-
Teleprinting is based on a simple binary code mission of a sequence of characters may be
made up of spaces and pulses, each of 22 much slower, depending on the speed of the
milliseconds duration. Normal transmission Is
at the rate of 60 w.p.m. The sequence of mark operator. A simplified diagram of a one -way
and space pulses for the lettir R are shown RTTY circuit is shown in figure 2.
here. Start space provides time for machine
synchronization and stop pulse provides time
for sending and receiving mechanisms to 15 -2 RTTY Reception
position themselves for transmission of the
following character.
The RTTY receiving mechanism must re-
spond to a sequence of pulses and spaces
the equipment, locked to the a -c line fre- transmitted by wire or radio. Frequency -shift
quency. The sequence of mark and space keying may be demodulated by a beat -fre-
pulses for the letter R is shown in figure 1. quency technique, or by means of a discrimi-
The start space provides time for synchro- nator as employed in f -m service. The re-
nization of the receiving machine with the ceived signal is converted into d -c pulses
sending machine. The stop pulse provides which are used to operate the printing mag-
time for the sending mechanism as well as nets in the teleprinter. Conversion of RTTY
the receiving mechanism to properly po- signals into proper pulses is accomplished by
sition themselves for transmission of the a receiving converter (terminal unit, abbrevi-
following character. ated TU) . RTTY converters may be either i -f
The FSK system normally employs the
higher radio frequency as the mark and the
lower frequency as the space. This relation- KEYBOARD
AUDIO DISCR M-
FILTER INATOR
KEYER
PRINTER
Figure 3
discriminator or audio discriminator units. A keyer. In its simplest form, the i -f converter
block diagram of an intermediate -frequency requires that adequate selectivity and inter-
converter is shown in figure 3A. The RTTY ference rejection be achieved by means of
signal in the i -f system of the receiver is the i -f system of the receiver.
considered to be a carrier frequency -modu- The schematic of a typical -f RTTY i
12AU7 Ry
6BH6 DISCRIMINATOR D -C AMPLIFIER (W E-255.1)
LIMITER
IN54
55 kHz
I-F INPUT
I -F RTTY CONVERTER
Typical i -f converter circuit illustrates this technique. Some type of indication that the RTTY
signal is properly tuned is required, particularly on the hf bands. With the i -f terminal
unit, a zero -center microammeter may be connected across discriminator load resistors (A -B).
328 Radioteletype Systems THE RADIO
12A7 RY
D-C AMPLIFIER (WE-ISSA)
12AT7 6C4 SCOPE DISCRIMINATOR
LIMITER AMP INPUT
AUDIO
1N64
INPUT 001
Figure 5
nal required by the teleprinter. The beat space intelligence of the RTTY code. Fre-
oscillator of the receiver is used to provide quency-shift keying (FSK) may be accom-
the beat tones of 212f and 297f Hz re- plished by varying the frequency of the
quired in the usual 850-Hz shift system. transmitter oscillator in a stable manner be-
Either frequency may be used for either tween two chosen frequencies. The amount
mark or space, and the signals may be easily of shift must be held within close tolerances
inverted by tuning the beat oscillator to the as the shift must match the frequency differ-
opposite side of the i -f passband of the re- ence between the selective filters in the
ceiver. The schematic of a simple audio - receiving terminal unit. The degree of fre-
frequency RTTY converter is shown in
figure 5.
Receiving converters of both types usually
include clipping and limiting stages which
hold the signal at constant amplitude and TO
KEYBOARD
KEYBOARD
Figure 7
SPACE
CONTACTS
MARK
CONTACTS Figure 10
AFSK OSCILLATOR
Figure 9
Audio frequency -shift keying Is often used on
TRANSMITTER -DISTRIBUTOR (T -D) UNIT vhf bands to avoid problems of holding close
radio -frequency stability. The L -C, circuit is
T -D unit is electromechanical device which tuned to 2975 Hx (with keyboard open). Clos-
senses perforations in a teleprinter tape and ing the keyboard parallels capacitor CI and
translates this information into the ele of lowers the oscillator frequency to 2125 Hz. The
impulses of the teleprinter code. Information coil L is on 88 mH toroid (with about 44" of
derived from the tape by contact fingers is wire removed). Capacitors CI and C. are high
transmitted in proper time sequence by a quality paper or mylar. Compression mica ca-
commutator- distributor driven by a constant - pacitors may be used as padders to place the
speed motor. oscillator on the co frequencies.
330 Rad ioteletype Systems
punched tape by contact fingers is trans- to use by an audio converter of the type
mitted in the proper time sequence by a shown in figure 4. The beat oscillator in the
commutator -distributor driven at a constant receiver is not used for this form of recep-
speed by a synchronous motor (figure 9). tion. AFSK is permitted only on those ama-
Used in conjunction with the T -D is a teur bands on which A2 emission is author-
tape perforator which punches the teleprinter ized. A simple AFSK oscillator circuit is
code in a paper tape. The perforator operates shown in figure 10.
mechanically from a teleprinter keyboard
Obtaining Sources available to radio ama-
for originating messages. A reperf orator may
Teleprinter teurs include several nonprofit
be connected to receiving equipment to
"tape" an incoming message for storage or Machines RTTY societies, established in
retransmission. various areas of the United
States for the purpose of disposing of tele-
Audio Frequency- Audio frequency -shift key - printer equipment discarded by commercial
Shift Keying ing (AFSK) is often used services. These societies can be contacted
by radio amateurs on the through active RTTY amateurs. The com-
vhf bands in order to avoid the problems of mercial services, including the Bell Tele-
holding close radio- frequency stability. An phone Company, generally cannot dispose of
audio oscillator is employed to generate a used equipment directly to radio amateurs.
2125 -Hz tone (mark) and a 2975 -Hz tone Commercial services should not be contacted
(space) when driven by the keyboard of a regarding used teleprinters. Many radio ama-
teleprinter, or by a tape T -D unit. The teurs, active in RTTY, rebuild machines
audio signal is then applied to the modulator from junked or damaged equipment at nom-
of the vhf transmitter and the resulting am- inal cost. These amateurs are also an excel-
plitude- modulated signal is detected and put lent source of maintenance support.
www.americanradiohistory.com
CHAPTER SIXTEEN
Sideband Transmission
While single -sideband transmission (SSB) The problem solved by the carrier system
has attracted significant interest on amateur was that of translating the 300- to 3000 -Hz
frequencies only in the past few years, the voice band of frequencies to a higher fre-
principles have been recognized and put to quency (for example, 40.3 to 43.0 kHz) for
use in various commercial applications for transmission on the toll wires, and then to
many years. Expansion of single-Sideband for reverse the translation process at the receiv-
both commercial and amateur communica- ing terminal. It was possible in some short -
tion has awaited the development of econom- haul equipment to amplitude modulate a 40-
ical components possessing the required kHz carrier with the voice frequencies, in
characteristics (such as sharp -cutoff filters which case the resulting signal would occupy
and high- stability crystals) demanded by a band of frequencies between 37 and 43
SSB techniques. The availability of such kHz. Since the transmission properties of
components and precision test equipment wires and cable deteriorate rapidly with in-
now makes possible the economical testing, creasing frequency, most systems required
adjustment, and use of SSB equipment on a the bandwidth conservation characteristics
wider scale than before. Many of the seem- of single -sideband transmission. In addition,
ingly insurmountable obstacles of past years the carrier wave was generally suppressed to
no longer prevent the amateur from achiev- reduce the power handling capability of the
ing the advantages of SSB for his class of repeater amplifiers and diode modulators. A
operation. large body of literature on the components
and techniques of SSB has been generated
16 -1 Commercial by the continuing development effort to pro-
Applications of SSB duce economical carrier telephone systems.
The use of SSB for overseas radiotelephony
Before discussion of amateur SSB equip- has been practiced for several years though
ment, it is helpful to review some of the the number of such circuits is small but
commercial applications of SSB in an effort growing. Moreover, the economic value of
to avoid problems that are already solved. such circuits has been great enough to war-
The first and only large scale use of SSB rant elaborate station equipment. It is from
has been for multiplexing additional voice these stations that the impression has been
circuits on long- distance telephone toll wires. obtained that SSB is too complicated for all
Carrier systems came into wide use during but a corps of engineers and technicians to
the 30's, accompanied by the development of handle. Components such as lattice filters
high -Q toroids and copper -oxide ring modu- with 40 or more crystals have suggested
lators of controlled characteristics. astronomical expense.
332 Sideband Transmission THE RADIO
www.americanradiohistory.com
HANDBOOK Derivation 333
www.americanradiohistory.com
HANDBOOK Derivation 335
MOD
SIDE-
BAND
POLTA.E oUTPUT
CAI IE 7 1
vOLTAGE
SIDE -
MOD. BAND
VOLTAGE OUTPUT
7
PUSH-PULL
AUDIO N RING
MODULATOR - DOUBLE-BALANCED
MODULATOR
Figure 8 I
GAPgIEP l
vOLTAGE
SHOWING TWO COMMON TYPES
Figure 9
OF BALANCED MODULATORS
Notice that a balanced modulator changes TWO TYPES OF DIODE BALANCED
the circuit condition from single ended to MODULATOR
push -pull, or vice versa. Choice of circuit de-
pends on external circuit conditions since Such balanced modulator circuits are com-
both the A and S arrangements can give monly used in carrier telephone work and In
satisfactory generation of a double -sideband single -sideband systems where the carrier
suppressed-carrier signal. frequency and modulating frequency are rela-
tively close together. Vacuum diodes, copper -
oxide rectifiers, or crystal diodes may be used
in the circuits.
tion and the usual a -m detector would de-
liver no output.
The phase of the reinserted carrier is of no phase, and with the carrier outputs in paral-
importance in SSB reception and by using a lel. The carrier will be balanced out of the
strong reinserted carrier, exalted- carrier re- output circuit, leaving only the two side -
ception is in effect realized. Selective fading bands. Such a circuit is called a balanced
with one sideband simply changes the ampli- modulator.
tude and the frequency response of the sys- Any nonlinear element will produce mod-
tem and very seldom causes the signal to be- ulation. That is, if two signals are put in,
come unintelligible. Thus the receiving tech- sum and difference frequencies as well as the
niques used with SSB are those which inher- original frequencies appear in the output.
ently greatly minimize distortion due to se- This phenomenon is objectionable in ampli-
lective fading. fiers and desirable in modulators or mixers.
In addition to the sum and difference fre-
16 -3 Carrier -Elimination quencies, other outputs (such as twice one
Circuits frequency plus the other) may appear. All
combinations of all harmonics of each input
Various circuits may be employed to elim- frequency may appear, but in general these
inate the carrier to provide a double -side- are of decreasing amplitude with increasing
band signal. A selective filter may follow the order of harmonic. These outputs are usually
carrier-elimination circuit to produce a rejected by selective circuits following the
single- sideband signal. modulator. All modulators are not alike in
Two modulated amplifiers may be con- the magnitude of these higher -order outputs.
nected with the carrier inputs 180 out of Balanced diode rings operating in the square-
www.americanradiohistory.com
HANDBOOK Carrier Elimination 337
law region are fairly good and pentagrid Care must be taken in adapting this circuit
converters much poorer. Excessive carrier to minimize the carrier power that will be
level in tube mixers will increase the relative lost in these elements. The shunt and series
magnitude of the higher-order outputs. Two quad circuits are usable when the output
types of triode balanced modulators are frequencies are entirely different (i.e.: audio
shown in figure 8, and two types of diode and r.f.). The shunt quad (B) is used with
modulators in figure 9. Balanced modulators high source and load impedances and the
employing vacuum tubes may be made to series quad (C) with low source and load
work very easily to a point. Circuits may be impedances. These two circuits may be
devised wherein both input signals may be adapted to use only two diodes, substituting
applied to a high- impedance grid, simplifying a balanced transformer for one side of the
isolation and loading problems. The most bridge, as shown in figure 12. It should be
important difficulties with these vacuum - noted that these circuits present a half -wave
tube modulator circuits are: (1) Balance load to the carrier source. In applying any
is not independent of signal level. (2) Bal- of these circuits, r -f chokes and capacitors
ance drifts with time and environment. (3) must be employed to control the path of
The carrier level for low high -order output signal and carrier currents. In the shunt pair
is critical. (4) Such circuits have limited shown, a blocking capacitor is used to prevent
dynamic range. the r -f load from shorting the audio input.
A number of typical circuits are shown To a first approximation, the source and
in figure 10. Of the group the most satis- load impedances should be an arithmetical
factory performance is to be had from plate - mean of the forward and back resistances
modulated triodes. of the diodes employed. A workable rule of
thumb is that the source and load imped-
Diode Ring Modulation in telephone car- ances be ten to twenty times the forward
Modulators rier equipment has been very resistance for semiconductor rings. The high -
successfully accomplished with frequency limit of operation in the case of
copper -oxide double balanced ring modula- junction and copper -oxide diodes may be
tors. More recently, germanium diodes have appreciably extended by the use of very low
been applied to similar circuits. The basic source and load impedances.
diode ring circuits are shown in figure 11. Copper-oxide diodes suitable for carrier
The most widely applied is the double bal- work are normally manufactured to order.
anced ring (A). Both carrier and input are They offer no particular advantage to the
balanced with respect to the output, which amateur, though their excellent long -term
is advantageous when the output frequency stability is important in commercial appli-
is not sufficiently different from the inputs cations. Rectifier types intended to be used
to allow ready separation by filters. It should as meter rectifiers are not likely to have
be noted that the carrier must pass through the balance or high- frequency response de-
the balanced input and output transformers. sirable in amateur SSB transmitters.
00
o---I
OUT
e Rd
T
R-F
CAR-
IER
IN
BALANCED MODULATORS
338 Sideband Transmission THE RADIO
HIGH -2 HIGH -Z
MODULAT. SIDESAND MODULATING SIDE SAND LOW -2 LOW -2
VOLTAGE OUTPUT VOLTAGE OUTPUT MODULATING SIDE SAND
VOLTAGE OUTPUT
O
Figure 11
www.americanradiohistory.com
HAN DBOOK Generation of SSB 339
B*
6C4
AJDIO
IN
70
6AL5 - -9 -2 -,
kHz DEVIATION
o I
Figure 15
BANDPASS CHARACTERISTIC OF A
LOW- FREQUENCY (50 kHz) SINGLE -
SIDEBAND FILTER
50 kHz 1750-1950-M4
OSCILLATOR OSCILLATOR
Figure 14
w
C R T'es
Mn .33R FILTER
3 e P---o B 111 ( J
.002 00 R
.05 71C .01
3 R .01
F.
,M
350
O
1350 V
SOMA.
OC3 PUSH -PULL /1-It TO BALANCED
SIC 20W MODULATOR FOR CONVERSION
TO 160 METERS
100 00M
Figure 16
chosen r -f signal, is mixed in a ring modu- Wave Filters The heart of a filter -type SSB
lator, made up of small germanium diodes. exciter is the sideband filter.
Depending on the choice of frequency of the Conventional coils and capacitors may be
r -f oscillator, either the upper or lower side - used to construct a filter based on standard
band may be applied to the input of the me- wave -filter techniques. The Q of the filter
chanical filter. The carrier, to some extent, inductances must be high when compared
has been rejected by the ring modulator. Ad- with the reciprocal of the fractional band-
ditional carrier rejection is afforded by the width. If a bandwidth of 3 kHz is needed at
excellent passband characteristics of the me- a carrier frequency of 50 kHz, the band-
chanical filter. For simplicity, the mixing width expressed in terms of the carrier fre-
and filtering operation usually takes place at quency is 3/50, or 6 percent. This is expressed
a frequency of 455 kHz. The single -sideband in terms of fractional bandwidth as 1/16.
signal appearing at the output of the me- For satisfactory operation, the Q of the
chanical filter may be translated directly to filter inductances should be 10 times the
a higher operating frequency. Suitable tuned reciprocal of this, or 160. Appropriate Q is
circuits must follow the conversion stage to generally obtained from toroidal inductances,
eliminate the signal from the conversion though there is some possibility of using
oscillator. iron -core solenoids between 10 and 20 kHz.
SPEECH
AMPLIFIER
LOW-2
PHASE
INVERTER
SHUNT-QUAD
RING
MODULATOR
---- 455-MMa
MECHANICAL
FILTER
CONVERTER H RIT AMPLIFIER
LIFT -Q
WITH HIGH
CIRCUIT
FOR OUTPUT ON
3953- MHz
450- 4532Nr
3500-kHz
OSCILLATOR
Figure 17
www.americanradiohistory.com
HANDBOOK Generation of SSB 341
CNSO CHSO
r I -F TRANS I -F TRANS
r
I-F TRANS
-1 0
160.5 4605
kHz kHz 10
OUTI I IN
11
J 20
CH50 CNSO
FT -24I CHANNEL 19 CRYSTAL 461. I4Hz
0 30
FT -141 CHANNEL SO CRYSTAL a 462900
z *-CARRIER
Figure 18 o FREQUENCY
(7
A characteristic impedance below 1000 ohms The war -surplus low- frequency crystals
should be selected to prevent distributed may be adapted to this type of filter with
capacitance of the inductances from spoiling some success. Experimental designs usually
over -all performance. Paper capacitors in- synthesize a selectivity curve by grouping
tended for bypass work may not be trusted sharp notches at the side of the passband.
for stability or low loss and should not be Where the width of the passband is greater
used in filter circuits. Care should be taken than twice the spacing of the series and par-
that the levels of both accepted and rejected allel resonance of the crystals, special circuit
signals are low enough so that saturation of techniques must be used. A typical crystal
the filter inductances does not occur. filter using these surplus crystals, and its
approximate passband is shown in figure 18.
Crystal Filters The best known filter re-
Mechanical Filters Filters using mechanical
sponses have been obtained
with crystal filters. Types designed for pro- resonators have been
gram carrier service cut off 80 db in less studied by a number of companies and are
than SO Hz. More than 80 crystals are used offered commercially by the Collins Radio
in this type of filter. The crystals are cut to Co. They are available in a variety of band-
control reactance and resistance as well as widths at center frequencies of 250 and 455
the resonant frequency. The circuits used are kHz. The 250-kHz series is specifically in-
based on full lattices. tended for sideband selection. The selectivity
attained by these filters is intermediate be-
I
tween good LC filters at low center fre-
quencies and engineered quartz -crystal filt-
o--, ers. A passband of two 250 -kHz filters is
LOWER SIDE6AND UIERSIDE6AND
shown in figure 19. In application of the
10
20
CARRIER
-I
FREQUENCY
I mechanical filters some special precautions
are necessary. The driving and pickup coils
should be carefully resonated to the opera-
30-- ting frequency. If circuit capacitances are
1
unknown, trimmer capacitors should be used
40.
across the coils. Maladjustment of these
tuned coils will increase insertion loss and
SO
the peak -to- valley ratio. On high- impedance
AA
filters (ten- to twenty- thousand ohms) sig-
246 247 246 2,011 250 251 252 253 254 255
nals greater than 2 volts at the input should
FREQUENCY I kHz)
be avoided. Direct current should be blocked
Figure 19 out of the end coils. While the filters are
PASSBAND OF LOWER AND UPPER rated for S ma of coil current, they are not
SIDEBAND MECHANICAL FILTER rated for d -c plate voltage.
342 Sideband Transmission THE RADIO
BALANCED
I MODULATOR
N 1
20Q ]000 AUDIO
SPEECH SPEECH PHASE - _ TO POWER AMPLIFIER STAGES
AMPLIFIER FILTER SPLITTING OR DIRECTLY TO ANTENNA SYSTEM
NETWORKS
BALANC ED
ZQ 2 MODULA
N'2
PHASE DIFFERENCE IErWEEN 01 AND e2' RD' 91 62
PRASE DIFFERENCE JETWEEN AND 02 r0
RADIO -FRED
PHASE -
SPLITTING
NETWORA
RADIO FRED.
SIGNAL At
CARRIER FRED.
Figure 20
The Phasing There are a number of points order to obtain sideband at the desired out-
a
System of view from which the op- put frequency, the case with the filter
as in
eration of the phasing system method of sideband generation.
of SSB generation may be described. We Assuming that we feed a speech signal to
may state that we generate two double- the balanced modulators along with the
sideband suppressed -carrier signals, each in 3900 -kHz carrier we will obtain in the out-
its own balanced modulator, that both the put of the balanced modulators a signal
r-f phase and the audio phase of the two which is either the sum of the carrier signal
signals differ by 90 degrees, and that the and the speech band, or the difference be-
outputs of the two balanced modulators are tween the carrier and the speech band. Thus
added with the result that one sideband is if our speech signal covers the band from
increased in amplitude and the other one is 200 to 3000 Hz, we will obtain in the out-
cancelled. This, of course, is a true descrip- put a band of frequencies from 3900.2 to
tion of the action that takes place. But it is 3903 kHz (the sum of the two, or the "up-
much easier to consider the phasing system as per" sideband), or a band from 3897 to
a method simply of adding (or of subtract- 3899.8 kHz (the difference between the two,
ing) the desired modulation frequency and or the "lower" sideband). A further ad-
the nominal carrier frequency. The carrier vantage of the phasing system of sideband
frequency of course is not transmitted, as is generation is the fact that it is a very
the case with all SSB transmissions, but only simple matter to select either the upper side-
the sum or the difference of the modulation band or the lower sideband for transmission.
band from the nominal carrier is transmitted A simple double -pole double -throw revers-
(figure 20). ing switch in two of the four audio leads to
The phasing system has the obvious ad- the balanced modulators is all that is re-
vantage that all the electrical circuits which quired.
give rise to the single sideband can operate in
a practical transmitter at the nominal output High -Level The plate- circuit efficien-
frequency of the transmitter. That is to say Phasing Versus cy of the four tubes
that if we desire to produce a single sideband Low-Level Phasing usually used to make up
whose nominal carrier frequency is 3.9 MHz, the two balanced modu-
the balanced modulators are fed with a 3.9- lators of the phasing system may run as high
MHz signal and with the audio signal from as f0 to 70 percent, depending on the oper-
the phase splitters. It is not necessary to go ating angle of plate current flow. Hence it
through several frequency conversions in is possible to operate the double balanced
HAN DBOOK Generation of SSB 343
O.
1
FOUR -PHASE A F.
Figure 21
formance and is very easy to adjust is shown
in figure 22. The adjustments for carrier
balance are made by adjusting the potentio-
meter for voltage balance and then the small
variable capacitor for exact phase balance of
TWO CIRCUITS FOR SINGLE -
the balanced carrier voltage feeding the diode
SIDEBAND GENERATION BY THE
modulator.
PHASING METHOD.
5+
Tho circuit of A offers the advantages of
simplicity in the single-ended input circuits
plus a push -pull output circuit. Circuit I re- AUDIO
0.2 VOLT
MECHANICAL
FILTER
quires double -ended input circuits but allows
all the plates to bo connected in parallel for
the output circuit. R 0.1 0.1 VOLT
OUTPUT
1
modulator directly into the antenna system 12A7
er
as the output stage of the transmitter. 0R
2.7R 10R
The alternative arrangement is to generate
the SSB signal at a lower level and then to
amplify this signal to the level desired by
means of class -A or class -B r -f power ampli-
R-F CARRIER
2.3 VOLTS Nl r.
fiers. If the SSB signal is generated at a level /2
01
2.7 R
of a few milliwatts it is most common to
make the first stage in the amplifier chain a
Figure 22
class -A amplifier, then to use one or more
class -B linear amplifiers to bring the output BALANCED MODULATOR FOR USE
up to the desired level. WITH MECHANICAL FILTER
www.americanradiohistory.com
344 Sideband Transmission THE RADIO
. ISO V.
3900
0242209
TO SAL.
1.100. I
AUDIO
SIGNAL
7900
3900
TO SAL.
41OD. 2
00575 20 M
TS 3900
-$4121[ A F.
+150 v.
Figure 29
Figure 24
LOW -Q R -F PHASE -SHIFT NETWORK
DOME AUDIO- PHASE-SHIFT NETWORK
The r-f phase -shift system Illustrated above
is convenient le e case where It is desired to arrangement is convenient for ob-
This circuit
make small changes in the operating fre- taining the audio phase shift whew It is
quency of the system witheet the necessity desired to use a minimum of circuit cempo-
of being precise in the edustment of two neets and tube elements.
coupled circuits as used for r-f phase shift in
the circuit of Agar* 21.
The inductance chosen for use at L must
Rodio- Frequency A single -sideband genera - take into account the cancelling effect of the
Phasing for of the phasing type re- input capacitance of the tubes and the cir-
quires that the two bal- cuit capacitance; hence the inductance
anced modulators be fed with r -f signals should be variable and should have a lower
having a 90- degree phase difference. This r -f value of inductance than that value of in-
phase difference may be obtained through the ductance which would have the same react-
use of two loosely coupled resonant circuits, ance as resistor R. Inductor L may be con-
such as illustrated in figure 21A and 21B. sidered as being made up of two values of in-
The r-f signal is coupled directly or induc- ductance in parallel: (1) a value of induct-
tively to one of the tuned circuits, and the ance which will resonate at the operating
coupling between the two circuits is varied frequency with the circuit and tube capaci-
until, at resonance of both circuits, the r -f tances, and (2) the value of inductance
voltages developed across each circuit have which is equal in reactance to resistance R.
the same amplitude and a 90- degree phase In a network such as shown in figure 23,
difference. equal and opposite 45- degree phase shifts are
The 90- degree r-f phase difference also provided by the RL and RC circuits, thus
may be obtained through the use of a low -Q providing a 90- degree phase difference be-
phase- shifting network, such as illustrated in tween the excitation voltages applied to the
figure 23; or it may be obtained through the two balanced modulators.
use of a lumped- constant quarter -wave line.
The low -Q phase- shifting system has proved Audio -Frequency The audio- frequency phase -
quite practical for use in single -sideband Phasing shifting networks used in
systems, particularly on the lower frequen- generating a single -sideband
cies. In such an arrangement the two resist- signal by the phasing method usually are
ances (R) have the same value, usually in based on those described by Dome in an ar-
the range between 100 and a few thousand ticle in the December, 1946, Electronics. A
ohms. Capacitor C, in shunt with the input relatively simple network for accomplishing
capacitances of the tubes and circuit capaci- the 90- degree phase shift over the range
tances, has a reactance at the operating fre- from 160 to 3500 Hz is illustrated in figure
quency equal to the value of resistor R. Also, 24. The values of resistance and capacitance
inductor L has a net inductive reactance must be carefully checked to ensure mini-
equal in value at the operating frequency to mum deviation from a 90- degree phase shift
resistance R. over the 200- to 3000 -Hz range.
www.americanradiohistory.com
HANDBOOK Generation of SSB 345
Figure 26
Figure 25
plexity with the filter method. The selective
A VERSION OF THE DOME
circuits for rejection of unwanted frequen-
AUDIO -PHASE -SHIFT cies operate at a relatively low frequency,
NETWORK are designed for this one frequency and have
a relatively high order of Q. Carrier rejec-
Another version of the Dome network is tion of the order of 50 db or so may be ob-
shown in figure 25. This network employs tained with a relatively simple filter and a
three 12AU7 tubes and provides balanced balanced modulator, and unwanted sideband
output for the two balanced modulators. As rejection in the region of 60 db is econom-
with the previous network, values of the ically possible.
resistances within the network must be held The phasing method of SSB generation ex-
to very close tolerances. It is necessary to changes the problems of high -Q circuits and
restrict the speech range to 300 to 3000 Hz linear amplification for the problems of ac-
with this network. Audio frequencies out- curately controlled phase -shift networks. If
side this range will not have the necessary the phasing method is employed on the act-
phase -shift at the output of the network and ual transmitting frequency, change of fre-
will show up as spurious emissions on the quency must be accompanied by a corres-
sideband signal, and also in the region of the ponding rebalance of the phasing networks.
rejected sideband. A low -pass 3500-Hz In addition, it is difficult to obtain a phase
speech filter, such as the Stancor Electron- balance with ordinary equipment within 2
ics Co. LPF -2 should be used ahead of this percent over a band of audio frequencies.
phase -shift network. This means that carrier suppression is limited
A passive audio phase -shift network that to a maximum of 40 db or so. However,
employs no tubes is shown in figure 26. This when a relatively simple SSB transmitter is
network has the same type of operating re- needed for spot- frequency operation, a phas-
strictions as those described above. Addi- ing unit will perform in a satisfactory man-
tional information concerning phase -shift ner.
networks will be found in The Single Side - Where a high degree of performance in
band Digest published by the American the SSB exciter is desired, the filter method
Radio Relay League. A comprehensive side- and the phasing method may be combined.
band review is contained in the December, Through the use of the phasing method in
1956 issue of Proceedings of the I.E.E.E. the first balanced modulator those undesired
sideband components lying within 1000 Hz
Comparison of Filter Either the filter or the of the carrier may be given a much higher
and Phasing Methods phasing method of degree of rejection than is attainable with
of SSB Generation single-sideband gener- the filter method alone, with any reasonable
ation is theoretically amount of complexity in the sideband filter.
capable of a high degree of performance. Then the sideband filter may be used in its
In general, it may be said that a high normal way to attain very high attenuation
degree of unwanted signal rejection may be of all undesired sideband components lying
attained with less expense and circuit corn- perhaps further than 500 Hz away from the
346 Sideband Transmission THE RADIO
6BE6 2000 kHz as shown in figure 27, the output
2000-kHz
CONVERSION
may be tuned to select either 2250 or 1750
FREQUENCY TUNE TO SELECT
2000 +2S0.22501H7
kHz.
(2.sv) OR
2000- 25007504H, Not only is it necessary to select the de-
sired mixing product in the mixer output
250
SIGNAL
-IH, SSS but also the undesired products must be
(o.25v) highly attenuated to avoid having spurious
Figure 27
output signals from the transmitter. In gen-
eral, all spurious signals that appear within
PENTAGRID MIXER CIRCUIT FOR the assigned frequency channel should be at
SSB FREQUENCY CONVERSION least 60 db below the desired signal, and
those appearing outside of the assigned fre-
carrier, and to restrict the sideband width on quency channel at least 80 db below the
the desired side of the carrier to the specified signal level.
frequency limit. When mixing 250 kHz with 2000 kHz
as in the above example, the desired product
is the 2250 -kHz signal, but the 2000 -kHz
16 -5 Single -Sideband injection frequency will appear in the output
Frequency-Conversion Systems about 20 db stronger than the desired signal.
To reduce it to a level 80 db below the de-
In many instances the band of sideband sired signal means that it must be-attenuated
frequencies generated by a low -level SSB 100 db.
transmitter must be heterodyned up to the The principal advantage of using bal-
desired carrier frequency. In receivers the anced- modulator mixer stages is that the in-
circuits which perform this function are jection frequency theoretically does not ap-
called converters or mixers. In sideband work pear in the output. In practice, when a con-
they are usually termed mixers or modula- siderable frequency range must be tuned by
tors. the balanced modulator and it is not practi-
cal to trim the push -pull circuits and the
Mixer Stages One circuit which can be used tubes into exact amplitude and phase bal-
for this purpose employs a re- ance, about 20 db of injection- frequency
ceiving -type mixer tube, such as the 6BE6. cancellation is all that can be depended on.
The output signal from the SSB generator is With suitable trimming adjustments the can-
fed into the #1 grid and the conversion fre- cellation can be made as high as 40 db, how-
quency into the #3 grid. This is the reverse ever, in fixed -frequency circuits.
of the usual grid connections, but it offers
about 10 db improvement in distortion. The The Twin -Triode MixerThe mixer circuit
plate circuit is tuned to select the desired shown in figure 28
output frequency product. Actually, the has about 10 db lower distortion than the
output of the mixer tube contains all har- conventional 6BE6 converter tube. It has a
monics of the two input signals and all pos- lower voltage gain of about unity and a
sible combinations of the sum and difference
frequencies of all the harmonics. In order to
avoid distortion of the SSB signal, it is fed to
the mixer at a low level, such as 0.1 to 0.2 100
(--- -+ SSS OUTPUT
volts. The conversion frequency is fed in at 12A7
a level about 20 db higher, or about 2 volts. 2.0 VOLT
0.2 VOLT
CONVERSION
By this means, harmonics of the incoming SIGNAL INPUT 470 SIGNAL
SSB signal generated in the mixer tube will
be very low. Usually the desired output fre-
quency is either the sum or the difference of
the SSB generator carrier frequency and the Figure 28
conversion frequency. For example, using an
TWIN-TRIODE MIXER CIRCUIT FOR
SSB generator carrier frequency of 250 kHz
SSB FREQUENCY CONVERSION
and a conversion injection frequency of
HANDBOOK Frequency Conversion 347
6.6.56 '3
IA LANCE
2N1225
0.1 SS,
OUTPUT
0.o1 VOLT
SSI SIG.
INPUT
120
Figure 29
lower output impedance which loads the first attenuation at 4000 kHz (which is only 250
tuned circuit and reduces its selectivity. In kHz away) is a practical example. Adding
some applications the lower gain is of no the requirement that this selective circuit
consequence but the lower distortion level is must tune from 2250 to 4250 kHz further
important enough to warrant its use in high complicates the basic requirement. The best
performance equipment. The signal- to -dis- solution is to cascade a number of tuned cir-
tortion ratio of this mixer is of the order of cuits. Since a large number of such circuits
70 db compared to approximately 60 db for may be required, the most practical solution
a 6BE6 mixer when the level of each of two is to use permeability tuning, with the cir-
tone signals is 0.5 volt. With stronger sig- cuits tracked together. An example of such
nals, the 6BE6 distortion increases very circuitry is found in the Collins 32S side -
rapidly, whereas the 12AU7 distortion is band transmitter.
comparatively much better. If an amplifier tube is placed between each
In practical equipment where the injection tuned circuit, the over-all response will be
frequency is variable and trimming adjust- the sum of one stage multipled by the num-
ments and tube selection cannot be used, it ber of stages (assuming identical tuned cir-
may be easier and more economical to obtain cuits). Figure 30 is a chart which may be
this extra 20 db of attenuation by using an used to determine the number of tuned cir-
extra tuned circuit in the output than by cuits required for a certain degree of at-
using a balanced modulator circuit. Two tenuation at some nearby frequency. The Q
balanced modulator circuits of interest are of the circuits is assumed to be 50, which is
shown in figure 29, providing a minimum of normally realized in small permeability-
20 db of carrier attenuation. tuned coils. The number of tuned circuits
Selective Tuned Circuits
with a Q of 50 required for providing 100
The selectivity re-
quirements of the db of attenuation at 4000 kHz while pass-
ing 4250 kHz may be found as follows:
tuned circuits following a mixer stage often
become quite severe. For example, using an
input signal at 250 kHz and a conversion in-
Of is 4250 - 4000 = 250 kHz
jection frequency of 4000 kHz the desired where,
output may be 4250 kHz. Passing the 4250 - fr is the resonant frequency (4250 kHz),
kHz signal and the associated sidebands
without attenuation and realizing 100 db of and,
348 Sideband Tronsmission THE RADIO
o
M..........M M.MM.......M.......I...........MMOMO... o
NONNI .....NNB......... Mira .......NN..
..i....n .. OEM IMO ..........r,NN.....NIMI.
11111111111111111111111111111111111'/NI111111111111111111
...OM..MM./..........
111111111111111111111111111 ,/ o
NNi.......1.......I......'/.......N..........N.P_
...1.NN1r...NN..INN.i.
1111111111111111111111111111111N1111111111111
111111111111111111111111911111111111111111111111j/I111
............
M..........
l
011rnOOIuNuvOOOipIIIIai
.................W...........r..I
I......
1./I..RRI....d...IN/..N J
111111111111111I 111111111111111r/11N11111111111/111
11111111111111111111111111111111111
' li'' II1''"
I .I ..111.I........r:..M.r!-
......r.........r.........r....
..1111 _.rill
1II
......,.....V/ .........N.r!i...Nr!..
.Id....ViN....ri...mC.yll ..!...rcam..
1111,1111111 P5I11 II1P 111111rd1111'."t_11Or!S i!i
1111/1111111111C.11111i111111!.1M1/.111I1..1P_
111'/111111,11111/.11I11:IIIIMIllealM11!l.:iIll
111111111111!III!!IIII!,!;,!,.,IIIIIIIII
r . .- r"1rr:iir--
,I,liil/1;;;;;:,
o
(I III,,
oN oIn
' o o
IC)
o(1) oN- oc0 o oo o o'
N
90
Figure 30
www.americanradiohistory.com
HAN D B OO K Frequency Conversion 349
Of 250
0.059
fr 4250 f
- SIGNALDISTORTION
TO
The point on the chart where .059 inter- (vol RATION
sects 100 db is between the curves for 6 and
7 tuned circuits, so 7 tuned circuits are re-
quired.
Another point which must be considered
I l
in practice is the tuning and tracking error Sp-4g 4p-3q 3p 24 2p -g P 4 247 4.1:3P 5R4P
34-21.
of the circuits. For example, if the circuits
were actually tuned to 4220 kHz instead of
FROM SSB
GENERATOR
GAIN CONTROL
PREAMPLIFIER
- POWER
AMPLIFIER
STAGE
-.TO ANT.
frequencies are generated. These represent all
of the possible combinations of the sum and
difference frequencies of all harmonics of the
R-F FROM P -A
PLATE CIRCUIT original frequencies. For purposes of test
and analysis, a two-tone test signal (two
CONTROL BIAS R F equal -amplitude tones) is used as the SSB
RECTIFIER,
source. Since the SSB radio -frequency ampli-
I fiers use tank circuits, all distortion products
DELAY BIAS VOLTAGE
FROM POWER SUPPLY are filtered out except those which lie close
to the desired frequencies. These are all odd -
Figure 32 order products; third order, fifth order, etc.
BLOCK DIAGRAM OF AUTOMATIC The third -order products are 2p -q and
LOAD CONTROL (A.L.C.) SYSTEM 2q -p where p and q represent the two SSB
r -f tone frequencies. The fifth order products
range. There are many modulation products
in the output of the mixer and a frequency
- -
are 3p 2q and 3q 2p. These and some
higher order products are shown in figure 31
scheme must be found that will not have un- A, B, and C. It should be noted that the fre-
desired output of appreciable amplitude at quency spacings are always equal to the dif-
or near the desired signal. When tuning ference frequency of the two original tones.
across a frequency range some products may Thus when an SSB amplifier is badly over-
"cross over" the desired frequency. These loaded, these spurious frequencies can extend
undesired crossover frequencies should be at far outside the original channel width and
least 60 db below the desired signal to meet cause an unintelligible "splatter" type of in-
modern standards. The amplitude of the un- terference in adjacent channels. This is usu-
desired products depends on the particular ally of far more importance than the distor-
characteristics of the mixer and the particu- tion of the original tones with regard to
lar order of the product. In general, most intelligibility or fidelity. To avoid interfer-
products of the 7th order and higher will be ence in another channel, these distortion
at least 60 db down. Thus any crossover fre- products should be down at least 30 db below
quency lower than the 7th order must be the adjacent channel signal. Using a two -
avoided since there is no way of attenuating tone test, the distortion is given as the ratio
them if they appear within the desired pass - of the amplitude of one test tone to the
band. The book Single Sideband Principles amplitude of a third -order product. This is
and Circuits by Pappenfus, McGraw Hill called the signal -to- distortion ratio (Sm)
Book Co., Inc., N. Y., covers the subject of and is usually given in decibels. The use of
spurious products and incorporates a "mix feedback r -f amplifiers make S/D ratios of
selector" chart that is useful in determining greater than 40 db possible and practical.
spurious products for various different mix- Automatic Two means may be used to
ing schemes. Load Control keep the amplitude of these dis-
In general, for most applications when the
intelligence -bearing frequency is lower than
tortion products down to ac-
the conversion frequency, it is desirable that ceptable levels. One is to design the ampli-
fier for excellent linearity over its amplitude
the ratio of the two frequencies be between
S to 1 and 10 to 1. This a compromise or power range. The other is to employ a
means of limiting the amplitude of the SSB
between avoiding low -order harmonics of
envelope to the capabilities of the amplifier.
this signal input appearing in the output,
An automatic load control system (ALC)
and minimizing the selectivity requirements
may be used to accomplish this result. It
of the circuits following the mixer stage.
should be noted that the r -f wave shapes of
16-6 Distortion Products the SSB signal are always sine waves because
the tank circuits make them so. It is the
Due to Nonlinearity of change in gain with signal level in an ampli-
R -F Amplifiers fier that distorts the SSB envelope and gen-
When the SSB envelope of a voice or multi - erates unwanted distortion products. An
tone signal is distorted, a great many new ALC system may be used to limit the input
www.americanradiohistory.com
HANDBOOK Distortion Products 351
Figure 34 WU
SIMPLIFIED SCHEMATIC
OF AUTOMATIC LOAD TO IAA
CONTROL AMPLIFIER.
OPERATING POINT OF ALC
CIRCUIT MAY BE SET
BY VARYING BLOCKING
BIAS ON CATHODE OF
6X4 SIGNAL RECTIFIER
SENS.
ALC ZERO
=COMPRESSION ADJ.
INDICATOR
352 Sideband Transmission THE RADIO
to limit the range of the signal fed into the
SSB generator.
Figure 33 shows the effectiveness of the
ALC in limiting the output signal to the
capabilities of the power amplifier. An ad-
justment of the delay bias will place the
threshold of compression at the desired
power output. Figure 34 shows a simplified
schematic of an ALC system. This ALC uses
Figure 35
two variable -gain amplifier stages and the
JR. MODULATOR CIRCUIT
SSB
R-F and A -F sources ara applied in series to maximum over-all gain is about 20 db. A
balanced modulator. meter is incorporated which is calibrated in
db of compression. This is useful in adjust-
passing signals above full power output by ing the gain for the desired amount of load
the amount of compression desired. Since the control. A capacitance voltage divider is
signal level through the SSB generator should used to step down the r -f voltage at the
be maintained within a limited range, it is plate of the amplifier tube to about f 0 volts
unlikely that more than 12 db ALC action for the ALC rectifier. The output of the
will be useful. If the input signal varies more ALC rectifier passes through RC networks
than this, a speech compressor should be used to obtain the desired attack and release times
PHASE -SHIFT NETWORK TAL OR 12AU7
IIn
T2 V O
BLUE,. RED-wNIrE
It]
20 K AMr 13 YEd b 1.001
T M
II 7
,- C2A^ ,y77:11a2.
Z 12AU7 I: Re Ce
12AT7
AUDIO R9 Ce = y001
INPUT I I
kcF , M 250 + 10%
II E C10
ilE y 71 2 I I
30 L--R14-
1.6 K
L1
:2
* 5% *5% 20K L2
N\ 0_0,Q9 i< ir
RFC
0.SMH N, *10b
4
12AU7
12T7
R -F OUTPUT
TWIST RFC
0.5.
MN 6407
-10.5 V. +. B-
C+,13-
6.3 V.
C2A,B,C,D = EACH SECTION 20 UF, 450 V. ELECTROLYTIC a1,2,3,4= 1H52 GERMANIUM D100E OR EQUIVALENT
C7 =2430 PF (.002 UFO MICA *5% WITH 170 -760 PF TRIMMER) L I. L2= 33T. N 21 E. WIRE CLOSE WOUND ON MILLEN N.09046
C6 =4660 PF (0043 UFO MICA *514 WITH 170 -760 PF TRIMMER) IRON -CORE ADJUSTABLE SLUG COIL FORM. LINK OF 6
C9=1215 PF (.001 LIED MICA *S% WITH 50 -360 PF TRIMMER) TURNS OF HOOKUP WIRE WOUND ON OPEN END.
C10 =607.5 PF (500 PF MICA *5% WITH 9 -160 PF TRIMMER) L3 =16 T. NI9 E. WIRE SPACED TO PILL MILLEN N. 99046
C16 =350 PF 600V MICA * 10% (250 PF AND 100 PF PARALLEL) COIL FORM TAP AT 6 TURNS. OAK OF 1 TURN AT CENTER.
R7,RIO= 133.300 OHMS, 1/2 WATT * 11b L4 =SAME AS Li EXCEPT NO LINK USED.
R6,R9= 100.000 OHMS, 1/2 WATT * 1% L5= 26 T. OF N19 E. WIRE. LINK ON END TO MATCH LOAD.
T1= SrANCOR A -53C TRANSFORMER (4 TURN LINK MATCHES 72 OHM LOAD)
T2,T3= UC R -384 TRANSFORMER.
51 =OPOT TOGGLE SWITCH *= MOUNTING ENO OF COILS
Figure 36
www.americanradiohistory.com
HANDBOOK SSB Exciters 353
and through r-f filter capacitors. The 3.3K tuning L2, however, will be degraded side-
resistor and 0.1 -fd capacitors across the band suppression.
rectifier output stabilize the gain around the Power requirements of the SSB, Jr. are 300
ALC loop to prevent "motorboating." volts at 60 ma, and -10.5 volts at 1 ma.
n 00009-o000
.3
z _
Figure 37
LOW-2
MIC. +
2
GE -2 GE -2
SPEECH AMPLIFIER
Figure 38
amplifier provides ample gain for use of a a carrier null indication on the S meter of a
low- impedance dynamic microphone. Audio receiver coupled to the output of the side-
gain is controlled by varying the base bias band filter.
on the second speech amplifier stage. A crystal is now placed in the conversion
The 9 -MHz SSB signal from the filter is oscillator and proper operation is checked by
beat to the desired operating frequency in a monitoring the conversion frequency with
transistor mixer stage. Typical r -f voltages the nearby receiver. The mixer stage is final-
in the mixer stage are indicated in the sche- ly adjusted for maximum output at the de-
matic. The collector of the mixer is tapped sired frequency.
down on the output tank circuit to provide
optimum impedance match. Output of the 16 -8 Reception of
mixer stage is about 0.1 volt.
Selection of the upper or lower sideband is Single -Sideband Signals
accomplished by placing the carrier oscillator
on the proper slope of the sideband filter. Single- sideband signals may be received,
The oscillator should be set at approximately after a certain degree of practice in the tech-
the 20- decibel suppression point of the pass - nique, in a quite adequate and satisfactory
band for best operation. If the oscillator is manner with a good communications re-
closer in frequency to the filter passband ceiver. However, the receiver must have
than this, carrier rejection will suffer. If the quite good frequency stability both in the
oscillator is moved further away in frequen- high- frequency oscillator and in the beat
cy from the passband, the lower voice fre- oscillator. For this reason, receivers which
quencies will be attenuated and the SSB sig- use a crystal -controlled first oscillator are
nal will sound high -pitched and tinny. The likely to offer a greater degree of satisfaction
two carrier- balance controls are adjusted for than the self -controlled oscillator type.
356 Sideband Transmission THE RADIO
Beat- oscillator stability in most receivers creased for good signal level, or until the
is usually quite adequate, but many receivers point is reached where best oscillator injec-
do not have a sufficient amplitude of beat os- tion becomes insufficient and the signal be-
cillator injection to allow reception of strong gins to distort.
SSB signals without distortion. In such re-
ceivers it is necessary either to increase the Single -Sideband Greatly simplified tuning,
amount of beat -oscillator injection into the Receivers and coupled with strong atten-
diode detector, or the manual gain control of Adapters uation of undesired sig-
the receiver must be turned down quite low. nals, can be obtained
The tuning procedure for SSB signals is as through the use of a single -sideband receiv-
follows: The SSB signals may first be located er or receiver adapter. The exalted- carrier
by tuning over the band with receiver set principle usually is employed in such receiv-
for the reception of c -w; that is, with the ers, with a phase- sensitive system sometimes
manual gain at a moderate level and with the included for locking the local oscillator to
beat oscillator operating. By tuning in this the frequency of the carrier of the incoming
manner SSB signals may be located when signal. In order for the locking system to
they are far below the amplitude of conven- operate, some carrier must be transmitted
tional a -m signals on the frequency band. along with the SSB signal. Such receivers and
With the beat oscillator on the wrong side adapters include a means for selecting the
of the sideband, the speech will sound in- upper or lower sideband by the simple oper-
verted; that is to say that low- frequency ation of a switch. For the reception of a
modulation tones will have a high pitch and single-sideband signal the switch obviously
high- frequency modulation tones will have must be placed in the correct position. But
a low pitch -and the speech will be quite for the reception of a conventional a -m or
unintelligible. With the beat oscillator on phase-modulated signal, either sideband may
the correct side of the sideband but too far be selected, allowing the sideband with the
from the correct position, the speech will least interference to be used.
have some intelligibility but the voice will The Product Detector An unusually satis-
sound quite high pitched. Then as the correct factory form of de-
setting for the beat oscillator is approached modulator for SSB service is the product de-
the voice will begin to sound natural but tector, shown in one form in figure 39. This
will have a background growl on each syl- circuit is preferred since it reduces intermod-
lable. At the correct frequency for the beat ulation products and does not require a large
oscillator the speech will clear completely local carrier voltage, as contrasted to the
and the voice will have a clean, crisp, qual- more common diode envelope detector. This
ity. It should also be mentioned that there is product detector operates much in the same
a narrow region of tuning of the beat oscilla- manner as a multigrid mixer tube. The SSB
tor a small distance on the wrong side of the
sideband where the voice will sound quite 6BE6 RFC
,AUDIO
bassy and difficult to understand. SIGNAL
SOO -kN,
With a little experience it will be possible CARRIER
INJECTION
to identify the sound associated with im-
proper settings of the beat -oscillator control
so that corrections in the setting of the con-
trol can be made. Note that the main tuning
control of the receiver is not changed after SSO
SIGNAL }}
FROM
the sideband once is tuned into the passband IF AMP.
www.americanradiohistory.com
HANDBOOK SSB Reception 357
Transmitter Design
www.americanradiohistory.com
Resistors 359
+3 +3
+4 +4
W
+3 +3
z
W
V +2 +2
F
z N
<+I +1
N_ W
v1 0 CC 0
W z
Z W
.J 2 UZ 2
2
a -) IU -3
I
J 4 f! 4
!
_30 40 70 0 10 100
-20 -IO 0 10 20 30 40 SO O 70 !0 90 100 -20 - 0 0 10 20 30 30 60
DEGREES CENTIGRADE DEGREES CENTIGRADE
Figure 1
resistance value. Matched resistors used in sirable amounts of inductance and distrib-
phase- inverter service can be heated out of uted capacitance. These quantities are illus-
tolerance by the act of soldering them into trated in figure 2A, the general equivalent
the circuit. Long leads should be left on the circuit of a resistor. This circuit represents
resistors and long -nose pliers should grip the the actual impedance network of a resistor at
lead between the iron and the body of the any frequency. At a certain specified fre-
resistor to act as a heat block. General tem- quency the impedance of the resistor may be
perature characteristics of typical carbon thought of as a series reactance (X,) as
resistors are shown in figure 1. The behavior shown in figure 211. This reactance may be
of an individual resistor will vary from these either inductive or capacitive depending on
curves depending on the manufacturer, the whether the residual inductance or the dis-
size and wattage of the resistor, etc. tributed capacitance of the resistor is the
dominating factor. As a rule, skin effect
Inductance of Every resistor because of its tends to increase the reactance with fre-
Resistors physical size has in addition to quency, while the capacitance between turns
its desired resistance, less de- of a wirewound resistor, or capacitance be-
Figure 2
C-
RDC
0
L
01 -+-o
p---rn- x , --0
10
FREQUENCY MHz
Figure 3
EQUIVALENT CIRCUIT OF A RESISTOR FREQUENCY EFFECTS ON SAMPLE
AT A PARTICULAR FREQUENCY COMPOSITION RESISTORS
360 Transmitter Design THE RADIO
60
R SHUNT
R=aa000n
`
WOWalli
SO
yp p \--y'y"y"o
C L RSERiEa
DO MEMINIMI Figure 5
11. a io ia 20
of the capacitor and the capacitive reactance
are equal and opposite, and the capacitor is
in itself series resonant at this frequency. As
FREQUENCY UM: the operating frequency of the circuit in
which the capacitor is used is increased above
Figure 4 the series -resonant frequency, the effective-
CURVES OF THE IMPEDANCE OF WIRE - ness of the capacitor as a bypassing element
WOUND RESISTORS AT RADIO deteriorates until the unit is useless.
FREQUENCIES
Bypass The usual forms of bypass ca-
tween the granules of a composition resistor
Capacitors pacitors have dielectrics of paper,
tends to cause the reactance and resistance to mica, or ceramic. For audio
drop with frequency. The behavior of var- work, and low- frequency r -f work up to
ious types of composition resistors over a perhaps 2 MHz or so, the paper capacitors
large frequency range is shown in figure 3. are satisfactory as their relatively high in-
By proper component design, noninductive ternal inductance has little effect on the
resistors having a minimum of residual re- proper operation of the circuit. The actual
actance characteristics may be constructed. amount of internal inductance will vary
Even these have reactive effects that cannot widely with the manufacturing process, and
be ignored at high frequencies. some types of paper capacitors have satis-
Wirewound resistors act as low -Q in- factory characteristics up to a frequency of
ductors at radio frequencies. Figure 4 shows f MHz or so.
typical curves of the high- frequency char- When considering the design of transmit-
acteristics of cylindrical wirewound resistors. ting equipment, it must be remembered that
In addition to resistance variations wire - while the transmitter is operating at some
wound resistors exhibit both capacitive and relatively low frequency (for example, 7
inductive reactance, depending on the type MHz), there will be harmonic currents flow-
of resistor and the operating frequency. In ing through the various bypass capacitors of
fact, such resistors perform in a fashion as the order of 10 to 20 times the operating
low -Q r -f chokes below their parallel self - frequency. A capacitor that behaves properly
resonant frequency. at 7 MHz however, may offer considerable
impedance to the flow of these harmonic
currents. For minimum harmonic generation
17 -2 Capacitors and radiation, it is obviously of greatest im-
portance to employ bypass capacitors having
The inherent residual characteristics of ca- the lowest possible internal inductance.
pacitors include series resistance, series in- Mica -dielectric capacitors have much less
ductance and shunt resistance, as shown in internal inductance than do most paper ca-
figure S. The series resistance and inductance pacitors. Figure 6 lists self -resonant fre-
depend to a large extent on the physical con- quencies of various mica capacitors having
figuration of the capacitor and on the ma- various lead lengths. It can be seen from in-
terial from which it is composed. Of great spection of this table that most mica ca-
interest to the amateur constructor is the pacitors become self -resonant in the 12- to
series inductance of the capacitor. At a cer- 50-MHz region. The inductive reactance
HANDBOOK Capacitors 361
CAPACITOR LEAD LENGTHS RESONANT FREQ. hibits piezoelectric effects, and capacitors
.02 fd MICA NONE 44.5 MHz
employing it for a dielectric will tend to
.002 fd MICA NONE 23.5 MHz "talk- back" when a -c voltages are applied
.01 fd MICA %4n 10 MHz across them. When these capacitors are used
.0009 fd MICA 140 55 MHz as plate bypass units in a modulated trans-
.002 fd CERAMIC s/e" 24 MHz
.001 fd CERAMIC '4" 55 MHz
mitter they will cause acoustical noise.
300 pfd BUTTON NONE 220 MHz Otherwise they are excellent for general r -f
.0005 fd CERAMIC 1'4/n 90 MHz work.
.01 fd CERAMIC % 'f 14.5 MHz A recent addition to the varied line of ca-
pacitors is the coaxial, or Hypass, type of
Figure 6
capacitor. These capacitors exhibit superior
SELF -RESONANT FREQUENCIES OF bypassing qualities at frequencies up to 200
VARIOUS CAPACITORS WITH MHz and the bulkhead type are especially
RANDOM LEAD LENGTH effective when used to filter leads passing
through partition walls between two stages.
they would offer to harmonic currents of 100
Variable Air Even though air is the perfect
MHz, or so, would be of considerable magni-
tude. In certain instances it is possible to Capacitors dielectric, air capacitors exhibit
deliberately series- resonate a mica capacitor losses because of the inherent
to a certain frequency somewhat below its resistance of the metallic parts that make up
normal self -resonant frequency by trimming the capacitor. In addition, the leakage loss
the leads to a critical length. This is some- across the insulating supports may become of
times done for maximum bypassing effect in some consequence at high frequencies. Of
the region of 40 to 60 MHz. greater concern is the inductance of the ca-
The button -mica capacitors shown in fig- pacitor at high frequencies. Since the capaci-
ure 7 are especially designed to have ex- tor must be of finite size, it will have tie rods,
tremely low internal inductance. Certain metallic braces, and end plates; all of which
types of button -mica capacitors of small contribute to the inductance of the unit.
physical size have a self- resonant frequency The actual amount of the inductance will
in the region of 600 MHz. depend on the physical size of the capacitor
Ceramic-dielectric capacitors in general
have the lowest amount of series inductance
per unit of capacitance of these three univer-
sally used types of bypass capacitors. Typical
resonant frequencies of various ceramic units
are listed in figure 6. Ceramic capacitors are
available in various voltage and capacitance
ratings and different physical configurations.
Standoff types such as shown in figure 7 are
useful for bypassing socket and transformer
terminals. Two of these capacitors may be
mounted in close proximity on a chassis and
connected together by an r-f choke to form
a highly effective r -f filter. The inexpensive
disc type of ceramic capacitor is recom-
mended for general bypassing in r -f cir-
cuitry, as it is effective as a bypass unit to
well over 100 MHz. Figure 7
The large TV doorknob capacitors are
useful as by -pass units for high voltage lines. TYPES OF CERAMIC AND MICA CAPACI-
These capacitors have a value of 500 pf, and TORS SUITABLE FOR HIGH -FREQUENCY
are available in voltage ratings up to 40,000 BYPASSING
volts. The dielectric of these capacitors is The Centrale', 8SBS (1000 pi) Is recommended
usually titanium dioxide. This material ex- for screen and plate circuits of tetrads) tubes.
362 Transmitter Design THE RADIO
and the method used to make contact to the wire ten inches long (a not uncommon
stator and rotor plates. This inductance may length for a plate lead in a transmitter) can
be cut to a minimum value by using as small have a self -inductance of 0.15 microhenrys.
a capacitor as is practical, by using insulated This inductance and that of the plate tuning
tie rods to prevent the formation of closed capacitor together with the plate -to- ground
inductive loops in the frame of the unit, and capacity of the vacuum tube can form a
by making connections to the centers of the resonant circuit which may lead to parasitic
plate assemblies rather than to the ends as is
oscillations in the vhf regions. To keep the
commonly done. A large transmitting ca- self -inductance at a minimum, all r -f carry-
pacitor may have an inherent inductance as ing leads should be as short as possible and
large as 0.1 microhenry, making the capaci-
should be made out of as heavy material as
tor susceptible to parasitic resonances in the possible.
50- to 150 -MHz range of frequencies.
At the higher frequencies, solid enameled
The question of optimum C/L ratio and
copper wire is most efficient for r -f leads.
capacitor plate spacing is covered in Chapter Tinned or stranded wire will show greater
Eleven. For all -band operation of a high
losses at these frequencies. Tank -coil and
power stage, it is recommended that a capaci- tank -capacitor leads should be of heavier
tor just large enough for 40 -meter phone wire than other r -f leads.
operation be chosen. (This will have suffi- The best type of flexible lead from the en-
cient capacitance for phone operation on all velope of a tube to a terminal is thin copper
higher- frequency bands.) Then use fixed
strip, cut from thin sheet copper. Heavy,
padding capacitors for operation on 80 met-
rigid leads to these terminals may crack the
ers. Such padding capacitors are available in
envelope glass when a tube heats or cools.
air, ceramic, and vacuum types.
Wires carrying only audio frequencies or
Specially designed variable capacitors are
direct current should be chosen with the volt-
recommended for uhf work; ordinary ca-
age and current in mind. Some of the low -fila-
pacitors often have "loops" in the metal
ment -voltage transmitting tubes draw heavy
frame which may resonate near the operating
frequency. current, and heavy wire must be used to
avoid voltage drop. The voltage is low, and
Variable Vacuum Variable vacuum capacitors hence not much insulation is required. Fila-
ment and heater leads are usually twisted
Capacitors because of their small phy-
together. An initial check should be made on
sical size have less inherent
the filament voltage of all tubes of 25 watts
inductance per unit of capacity than do
or more plate dissipation rating. This voltage
variable air capacitors. Their losses are ex-
should be measured right at the tube sockets.
tremely low, and their dielectric strength is
If it is low, the filament -transformer volt-
high. Because of increased production the
age should be raised. If this is impossible,
cost of such units is now within the reach of
heavier or parallel wires should be used for
the designer of amateur equipment, and their
filament leads, cutting down their length if
use is highly recommended in high -power
possible.
tank circuits.
Coaxial cable may be used for high -volt-
age leads when it is desirable to shield them
17 -3 Wire and Inductors from r -f fields. RG -8/U cable may be used
at d -c potentials up to 8000 volts, and the
Any length of wire, no matter how short, lighter RG -17/U may be used to potentials
has a certain value of inductance. This prop- of 3000 volts. Spark plug -type high- tension
erty is of great help in making coils and in- wire may be used for unshielded leads, and
ductors, but may be of great hindrance when will withstand 10,000 volts.
it is not taken into account in circuit design If this cable is used, the high -voltage leads
and construction. Connecting circuit ele- may be cabled with filament and other low -
ments (themselves having residual induct- voltage leads. For high -voltage leads in low -
ance) together with a conductor possessing power exciters, where the plate voltage is not
additional inductance can often lead to puz- over 450 volts, ordinary radio hookup wire
zling difficulties. A piece of No. 10 copper of good quality will serve the purpose.
www.americanradiohistory.com
HANDBOOK
Rc L Rc
Figure 8
L
C DISTRISUT[D
L
363
No r -f leads should be cabled; in fact it is nated to the operating frequency, will often
better to use enameled or bare copper wire permit satisfactory coupling.
for r -f leads and rely on spacing for insula-
Coil Placement For best Q a coil should be
tion. All r-f joints should be soldered, and
the joint should be a good mechanical junc- in the form of a solenoid with
tion before solder is applied. length from one to two times the diameter.
The efficiency and Q of air coils com- For minimum interstage coupling, coils
monly used in amateur equipment is a factor should be made as small physically as is
of the shape of the coil, the proximity of the practicable. The coils should then be placed
coil to other objects (including the coil so that adjoining coils are oriented for min-
form), and the material from which the coil imum mutual coupling. To determine if this
is made. Dielectric losses in so- called "air - condition exists, apply the following test:
wound " coils are low and the Q of such the axis of one of the two coils must lie
coils runs in the neighborhood of 300 to 500 in the plane formed by the center turn of
at medium frequencies. Unfortunately, most the other coil. If this condition is not met,
of the transmitting -type plug -in coils on the there will be appreciable coupling unless the
market designed for link coupling have far unshielded coils are very small in diameter
too small a pickup link for proper opera- or are spaced a considerable distance from
tion at 3.5 and 7 MHz. The coefficient of each other.
coupling of these coils is about 0.5, and ad- Insulation On frequencies above 7 MHz,
ditional means must be employed to provide ceramic, polystyrene, or Mycalex
satisfactory coupling at these low fre- insulation is to be recommended. Cold flow
quencies. Additional inductance in series must be considered when using polystyrene
with the pickup link, the whole being reso- (Amphenol 912, etc.). Bakelite has low losses
on the lower frequencies but should never be
used in the field of high- frequency tank cir-
24 cuits.
x
0
u
MIL Lucite (or Plexiglas), which is available
in rods, sheets, or tubing, is satisfactory for
Ii1Uiil. . E
:
use at all radio frequencies where the r -f
voltages are not especially high. It is very
easy to work with ordinary tools and is not
lasormon
Z o
EXTERNAL FIELD
TINAN BOTTOM WITH
FLUTED EDGE PRESSED
AGAINST PANEL
RUBBER GROMMET HOLES FOR COAX IAL SOCKET
METER STUDS
METER NUT
00I CERAMIC
LEAD
RFC
Figure 11A
-OPEN- SOX
SIMPLE METER SHIELD
HOLE EXTERNAL
FIELD
2
I 12000
10000
s.
yir-SMALL HOLE IN SHIELD
r
cl
' TO
osc.
fit the depth of the meter. This complete
shield assembly is shown in figure 11A.
3
s
630
00
Iso
11e
iCI
"-
RFC
C2
1172
Careful attention should be paid to leads
entering and leaving the electrically tight
box. Harmonic currents generated inside the
box can easily flow out of the box on power
'`I 2
7
70
140
1C1
CS
3MIt1
I
rFNCI
r* CIRI
C2
or control leads, or even on the outer shields
of coaxially shielded wires. Figure 11B illus-
trates the correct method of bringing
00 shielded cables into a box where it is desired
110 to preserve the continuity of the shielding.
CI RfC RFC
10 50
I
L----
I
I
I2 TRACE
C2_ the lead. Combinations of r -f chokes and low -
inductance bypass capacitors should be used
R -
1000 A CARSON C2 -.005 DISC CERAMIC
I
www.americanradiohistory.com
HANDBOOK Parasitic Oscillations 367
\j
RADIATION
FIELD \
ICKUP
LOOP
ywADIAT
LOOP
HOLE
I
RE- RADIATED
FIELD
WRONG
Chassis Material From a point of view of
electrical properties, alumi-
num is a poor chassis material. It is difficult
to make a soldered joint to it, and all grounds
must rely on a pressure joint. These pressure
^w,PAS3
GPACITOR j TBY
PASS
uMCITOR joints are prone to give trouble at a later
date because of high resistivity caused by
ILLUSTRATION OF HOW A SUPPOSEDLY the formation of oxides from electrolytic
GROUNDED POWER LEAD CAN COUPLE action in the joint. However, the ease of
ENERGY FROM ONE COMPARTMENT working and forming the aluminum material
TO ANOTHER far outweighs the electrical shortcomings,
and aluminum chassis and shielding may be
used with good results provided care is taken
hECTRICALLY TIGHT
UPARTNIENT
ELECTRICALLY TIGHT
COUPARIUE NT
in making all grounding connections. Cad-
mium and zinc plated chassis are preferable
RADIATION
FIELD \ BULKHEAD TYPE from a corrosion standpoint, but are much
(BYPASS CANCITOR
R IGHT more difficult to handle in the home work-
shop.
>P
T 17 -7 Parasitic Oscillation
in R -F Amplifiers
ILLUSTRATION OF LEAD ISOLATION BY
PROPER USE OF BULKHEAD BYPASS Parasitics (as distinguished from self -oscil-
CAPACITOR lation on the normal tuned frequency of the
amplifier) are undesirable oscillations either
spurious effect. In particular, a push -pull 810 of very-high or very-low frequencies which
amplifier using .001 -pfd filament bypass ca- may occur in radio- frequency amplifiers.
pacitors had a filament resonant loop that They may cause spurious signals (which
fell in the 7 -MHz amateur band. When the are often rough in tone) other than normal
amplifier was operated near this frequency harmonics, hash on each side of a modulated
marked instability was noted, and the fila- carrier, key clicks, voltage breakdown or flash-
ments of the 810 tubes increased in brilliance
when plate voltage was applied to the ampli-
fier, indicating the presence of r.f. in the
filament circuit. Changing the filament by-
pass capacitors to .01 fd lowered the fila-
ment resonance frequency to 2.2 MHz and
cured this effect. A ceramic capacitor of
.01 fd used as a filament bypass capacitor
on each filament leg seems to be satisfactory
from both a resonant and a TVI point of
view. Filament bypass capacitors smaller in
WRONG RIGHT
value than .01 fd should be used with
caution.
Various parasitic resonances are also Figure 14
found in plate and grid tank circuits. Push - DOUBLE RESONANCE EFFECTS IN PUSH -
pull tank circuits are prone to double reson- PULL TANK CIRCUIT MAY BE ELIMI-
ances, as shown in figure 14. The parasitic NATED BY THE INSERTION OF AN
resonance circuit is usually several MHz R -F CHOKE IN THE COIL CENTER
L2
RFCi
RFC i: _ .RFCt GM()
fANM
F-3
Figure 15
www.americanradiohistory.com
HANDBOOK Parasitic Oscillations 369
rent. If the inclusion of such a resistor upsets It may be said in general that the presence
the operating bias of the stage, an r -f choke of low -frequency parasitics indicates that
may be used, with a 100 -ohm 2 -watt carbon somewhere in the oscillating circuit there is
resistor in series with the choke to lower the an impedance which is high at a frequency
operating Q of the choke. If this expedient in the upper-audio or low r -f range. This
does not eliminate the condition, and the impedance may include one or more r -f
stage under investigation uses a beam -tetrode chokes of the conventional variety, power
tube, negative resistance can exist in the supply chokes, modulation components, or
screen circuit of such tubes. Try larger and the high- impedance may be presented simply
smaller screen bypass capacitors to determine by an RC circuit such as might be found in
whether or not they have any effect. If the the screen -feed circuit of a beam -tetrode.
condition is coming from the screen circuit
an audio choke with a resistor across it in 17 -8 Elimination of VHF
series with the screen -feed lead will often
eliminate the trouble. Parasitic Oscillations
Low -frequency parasitic oscillations can
often take place in the audio system of an Vhf parasitic oscillations are often diffi-
a -m transmitter, and their presence will not cult to locate and difficult to eliminate since
be known until the transmitter is checked on their frequency often is only moderately
a receiver. It is easy to determine whether or above the desired frequency of operation.
not the oscillations are coming from the But it may be said that vhf parasitics al-
modulator simply by switching off the ways may be eliminated if the operating
modulator tubes. If the oscillations are com- frequency is appreciably below the upper
ing from the modulator, the stage in which frequency limit for the tubes used in the
they are being generated can be determined stage. However, the elimination of a persist-
by removing tubes successively, starting ent parasitic oscillation on a frequency only
with the first speech amplification stage, un- moderately higher than the desired operating
til the oscillation stops. When the stage has frequency will involve a sacrifice in either
been found, remedial steps can be taken on the power output or the power sensitivity of
that stage. the stage, or in both.
If the stage causing the oscillation is a Beam -tetrode stages, particularly those
low -level speech stage it is possible that the using 807 type tubes, will almost invariably
have one or more vhf parasitic oscillations
trouble is coming from r -f or power -supply
unless adequate precautions have been taken
feedback, or it may be coming about as a
in advance. Many of the units described in
result of inductive coupling between two the constructional section of this edition
transformers. If the oscillation is taking had parasitic oscillations when first con-
place in a high -level audio stage, it is pos- structed. But these oscillations were elimi-
sible that inductive or capacitive coupling is nated in each case; hence, the expedients used
taking place back to one of the low -level in these equipments should be studied. Vhf
speech stages. It is also possible, in certain parasitics may be readily identified, as they
cases, that parasitic push -pull oscillation can cause a neon lamp to have a purple glow
take place in a class -B or class-AB modulator close to the electrodes when it is excited by
as a result of the grid -to -plate capacitance the parasitic energy.
within the tubes and in the stage wiring. Parasitic Oscillations Triode stages are less
This condition is more likely to occur if with Triodes subject to parasitic os-
capacitors have been placed across the sec- cillations primarily be-
ondary of the driver transformer and across cause of the much lower power sensitivity of
the primary of the modulation transformer such tubes as compared to beam tetrodes.
to act in the reduction of the amplitude of But such oscillations can and do take place.
the higher audio frequencies. Relocation of Often, however, it is not necessary to in-
wiring or actual neutralization of the audio corporate suppressors as normally is the
stage in the manner used for r -f stages may case with beam tetrodes, unless the triodes
be required. are operated quite near to their upper fre-
370 Transmitter Design THE RADIO
PC
Figure 19
www.americanradiohistory.com
HANDBOOK Parasitic Oscillations 373
www.americanradiohistory.com
Types of TV I 375
SHORT
300-OHM TOINPUT 300 OHM LEADS
LINE FROM TERMINALS TO
LINE FROM ANTENNA
ANTENNA OF TV SET ANTENNA TERM ON
TV SET
TO TV ANTENNA
COAX COAX
FITTING FITTING
TOANTENNA
TERMINALS
OF TV SET
Figure 1 Figure 2
TUNED TRAPS FOR THE TRANSMITTER HIGH -PASS TRANSMISSION LINE FILTERS
FUNDAMENTAL
The arrangement at A will stop the passing of
The arrangement at A has proven to be ef- all signals below about 45 MHz from the an-
fective in eliminating the condition of general tenna transmission line into the TV set. Coils
L, are each 1.2 microhenrys (17 turns No. 24
blocking as caused by a 28 -MHz transmitter in enam. closewound on 14 -inch dia. polystyrene
the vicinity of a TV receiver. The tuned circuits rod) with the center tap grounded. It will be
L C; are resonated separately to the fre- found best to scrape, twist, and solder the
quency of transmission. The adjustment may center tap before winding the coil. The number
be done at the station, or it may be accom- of turns each side of the tap may then be
plished at the TV receiver by tuning for mini- varied until the tap is in the exact center of
mum interference on the TV screen. Shown at the winding. Coil L. is 0.6 microhenry (72 turns
B is an alternative arrangement with a series - No. 24 enam. closewound on 14 -inch dia. poly-
tuned circuit across the antenna terminals of styrene rod). The capacitors should be about
the TV set. The tuned circuit should be reso- 16.5 pf, but either 15- or 20 -pf ceramic capaci-
nated to the operating frequency of the trans- tors will give satisfactory results. A similar
mitter. This arrangement gives less attenuation filter for coaxial antenna transmission line is
shown at B. Both coils should be 0.12 micro -
of the interfering signal than that at A; the henry (7 turns No. 18 enam. spaced to 7'2 inch
circuit has proven effective with interi on 14 -inch dia. polystyrene rod). Capacitors
from transmitters on the 50 -MHz band, and C, should be 75 -pf midget ceramics, while C
with low -power 28 -MHz transmitters. should be a 40-pf ceramic.
will be substantially continuous over the en- reception, work may then be begun on or in
tire frequency coverage of the BC or TV re- the vicinity of the transmitter toward elim-
ceiver. Channels 2 through 13 will be af- inating the other two types of interference.
fected in approximately the same manner.
With the overloading type of interference, Taking Out More or less standard BCI-
the problem is simply to keep the funda- the Fundamental type practice is most com-
mental of the transmitter out of the affected monly used in taking out
receiver. Other types of interference may or fundamental interference. Wavetraps and
may not show up when the fundamental is filters are installed, and the antenna system
taken out of the TV set (they probably will may or may not be modified so as to offer
appear), but at least the fundamental must less response to the signal from the amateur
be eliminated first. transmitter. In regard to a comparison be-
The elimination of the transmitter funda- tween wavetraps and filters, the same con-
mental from the TV set is normally the only siderations apply as have been effective in
operation performed on or in the vicinity of regard to BCI for many years; wavetraps are
the TV receiver. After the fundamental has quite effective when properly installed and
been eliminated as a source of interference to adjusted, but they must be readjusted when-
376 TV and Broadcast Interference THE RADIO
www.americanradiohistory.com
HANDBOOK Harmonic Radiation 377
JW
2ND 3RD 4TH 5TH 6TH 7TH 8TH 9TH 10TH
zO Z
f- u.
Figure 4
VIDEO
54
t
I
!CHANNEL'
I
TV
O
SOI+ND
O
1
60
+
1
'CHANNEL
1
Tv
O
V
6e
I
I
J
I
I TV
'CHANNEL
I
O
F
I
72
I OTHER
LOW BAND
76
t
I
'CHANNEL
I
TV
O
I
e2
O
I
'CHANNEL
1
TV
O
IF
ee toe
VIDEO SOIIIND
N
V e .
N
I
I Tv
1
I
NJ
t
Tv
v
t.
I
N
1
i TV
1
I
1
I TV
I
I
t
I TV
I
I
.
I TV I
-. N
I TV
.
I
'CHANNEL'
1
I
0 I
I
I
I
I
I
I
1
O I
I
1
I
@ I
1
I
1
11 I
I
I
I
I
I
I
I
1.41 I
I
HIGH BAND
Flgur* 5
FREQUENCIES OF THE VHF TV CHANNELS
Showing the frequency ranges of TV channels 2 through 13, with the picture carrier and sound
carrier frequencies also shown.
1. An unmodulated carrier, such as a c -w measures have been taken, operate the trans-
signal with the key down or an a -m mitter on a frequency such that the interfer-
signal without modulation, will give a ing harmonic will fall as far as possible from
crosshatch or herringbone pattern on the picture carrier. The worst possible inter-
the TV screen. This same general type ference to the picture from a continuous car-
of picture also will occur in the case rier will be obtained when the interfering
of a narrow -band f -m signal either signal is very close in frequency to the video
with or without modulation. carrier.
2. A relatively strong a -m or SSB signal Isolating Throughout the testing pro-
will give in addition to the herringbone the Source of cedure it will be necessary to
a very serious succession of light and the Interference have some sort of indicating
dark bands across the TV picture. device as a means of deter-
3. A moderate strength c -w signal with-
mining harmonic field intensities. The best
out transients, in the absence of over- indicator for field intensities some distance
loading of the TV set, will result from the transmitting antenna will probably
merely in the turning on and off of be the TV receiver of some neighbor with
the herringbone on the picture. whom friendly relations are still maintained.
To discuss condition 1 above, the herring- This person will then be able to give a check,
bone is a result of the beat note between the occasionally, on the relative nature of the
TV video carrier and the amateur harmonic. interference. But it will probably be neces-
Hence the higher the beat note the less ob- sary to go periodically and personally check
vious will be the resulting crosshatch. Fur- the results obtained, since the neighbor
ther, it has been shown that a much stronger probably will not be able to give any sort
signal is required to produce a discernible of a quantitative analysis of the progress
herringbone when the interfering harmonic which has been made.
is as far away as possible from the video car- An additional device for checking relative-
rier, without running into the sound carrier. ly high field intensities in the vicinity of the
Thus, as a last resort, or to eliminate the last transmitter will be almost a necessity. A sim-
vestige of interference after all corrective ple crystal -diode wavemeter, shown in fig-
www.americanradiohistory.com
HANDBOOK Low -Pass Filters 379
ure 6, will accomplish this function. Also, it one rated at a kilowatt is only a matter of
will be very helpful to have a receiver, with about 17 db. Yet the degree of harmonic at-
an S meter, capable of covering at least the tenuation required to eliminate interference
50- to 100 -MHz range and preferably the caused by harmonic radiation is from 80 to
range to 216 -MHz. This device may consist 120 db, depending on the TV signal strength
merely of a station receiver and a simple con- in the vicinity. This is not to say that it is
verter using the two halves of a 6J6 as not a simpler job to eliminate harmonic in-
oscillator and mixer. terference from a low -power transmitter
The first check can best be made with the than from a kilowatt equipment. It is simp-
neighbor who is receiving the most serious ler to suppress harmonic radiation from a
or the most general interference. Turn on the low -power transmitter simply because it is a
transmitter and check all channels to deter- much easier problem to shield a low -power
mine the extent of the interference and the unit, and the filters for the leads which enter
number of channels affected. Then discon- the transmitter enclosure may be constructed
nect the antenna and substitute a group of less expensively and smaller for a low -power
100 -watt lamps as a dummy load for the unit.
transmitter. Experience has shown that eight
100 -watt lamps series- connected in two 18 -3 Low -Pass Filters
groups of four in parallel will take the out-
put of a kilowatt transmitter on 28 MHz
if connections are made symmetrically to the After the transmitter has been shielded,
and all power leads have been filtered in such
group of lamps. Then note the interference.
a manner that the transmitter shielding has
Now remove plate voltage from the final
amplifier and determine the extent of inter- not been rendered ineffective, the only re-
ference caused by the exciter stages. maining available exit for harmonic energy
lies in the antenna transmission line. Hence
In the average case, when the final ampli-
fier is a beam -tetrode stage and the exciter is
the main burden of harmonic attenuation
will fall on the low -pass filter installed be-
relatively low powered and adequately shield-
tween the output of the transmitter and the
ed, it will be found that the interference
drops materially when the antenna is re-
antenna system.
moved and a dummy load substituted. It will Experience has shown that the low -pass
also be found in such an average case that
filter can best be installed externally to the
the interference will stop when the exciter main transmitter enclosure, and that the
transmission line from the transmitter to the
only is operating.
lowpass filter should be of the coaxial type.
Transmitter It should be made clear at this Hence the majority of low -pass filters are
Power Level point that the level of power designed for a characteristic impedance of 52
used at the transmitter is not of chms, so that RG -8 /U cable (or RG -58 /U
great significance in the basic harmonic re- for a small transmitter) may be used be-
duction problem. The difference in power tween the output of the transmitter and the
level between a 20 -watt transmitter and antenna transmission line or the antenna
tuner.
Transmitting -type low -pass filters for
10' PICKUP WIRE amateur use usually are designed in such a
manner as to pass frequencies up to about
sT.ISE. 30 MHz without attenuation. The nominal
QS' DIA., O.! LONG COYFAAiE S!-iI0MHz
1 N34 cutoff frequency of the filters is usually be-
I TURN tween 38 and 45 MHz, and m- derived sec-
o-I tions with maximum attenuation in channel
2 usually are included. Well- designed filters
capable of carrying any power level up to
one kilowatt are available commercially from
Figure 6 several manufacturers. Alternatively, filters
Crystal -diode wavemeter suitable for chocking in kit form are available from several manu-
high -intensity harmonics in TV region. facturers at a somewhat lower price. Effec-
380 TV and Broadcast Interference THE RADIO
www.americanradiohistory.com
HANDBOOK Low -Pass Filters 381
Figure 9
PHOTOGRAPH OF THE B FILTER WITH COVER REMOVED
The midsection in this filter adjusted for maximum rejection of channel 4. Note that the
is
main coils of the filter are mounted at on angle of about 45 degrees so that there will be
minimum inductive coupling from one section to the next through the holes in the aluminum
partitions. Mounting the coils in this manner was found to give a measurable improvement
in the attenuation characteristics of the filter.
danger of damage to the capacitors, provided portion of C;, can be tuned until resonance at
the filter is feeding a 52 -ohm resistive load. the correct frequency has been obtained.
It may be practical to use higher levels of Note that there is so little difference between
power with this type of ceramic capacitor in the constants of this intermediate section for
the filter, but at a power level of 200 watts channels 5 and 6 that variation in the setting
on the 28 -MHz band the capacitors run just of C:, will tune to either channel without
perceptibly warm to the touch. As a point materially changing the operation of the
of interest, it is the current rating which is filter.
of significance in the capacitors used in filters The coils in the intermediate sections of
such as illustrated. Since current ratings for the filter (L_, L3, La, and L, in figure 7A,
small capacitors such as these are not readily and L2, L;,, L,, and L,; in figure 7B) may be
available, it is not possible to establish an checked most conveniently outside the filter
accurate power rating for such a unit. The unit with the aid of a small ceramic capaci-
high -power unit illustrated in figure 9, tor of known value and a grid -dip meter.
which uses Centralab type 850S and 854S The ceramic capacitor is paralleled across the
capacitors, has proven quite suitable for small coil with the shortest possible leads.
power levels up to one kilowatt. Then the assembly is placed on a cardboard
Capacitors C,, C_, C1, and C7 can be box and the resonant frequency checked with
standard manufactured units with normal 5 a grid -dip meter. A reactance slide rule may
percent tolerance. The coils for the end sec- be used to ascertain the correct resonant fre-
tions can be wound to the dimensions given quency for the desired LC combination and
L,,, and L,) . Then the resonant fre- the coil altered until the desired resonant fre-
quency of the series -resonant end sections quency is attained. The coil may then be in-
should be checked with a grid -dip meter, stalled in the filter unit, making sure that
after the adjacent input or output terminal it is not squeezed or compressed as it is being
has been shorted with a very short lead. The installed. However, if the coils are wound ex-
coils should be squeezed or spread until actly as given under figure 7, the filter may
resonance occurs at 57 MHz. be assembled with reasonable assurance that
The intermediate m- derived section in the it will operate as designed.
filter of figure 7B may also be checked with
a grid -dip meter for resonance at the cor- Using Low -Pass The low -pass filter con -
rect rejection frequency, after the hot end Filters netted in the output trans-
of L, has been temporarily grounded with a mission line of the transmit-
low- inductance lead. The variable -capacitor ter is capable of affording an enormous de-
382 TV and Broadcast Interference THE RADIO
been discussed in the previous section. Sec-
ondly, the filter must be operated into a load
impedance approximately equal to its design
characteristic impedance. The filter itself
will have very low losses (usually less than
0.5 db) when operated into its nominal
value of resistive load. But if the filter is
Figure 10 not terminated correctly, its losses will be-
come excessive, and it will not present the
SCHEMATIC OF THE SINGLE -SECTION correct value of load impedance to the trans-
HALF -WAVE FILTER mitter.
The constants given below are for a character- If a filter, being fed from a high-power
istic impedance of 52 ohms, for use with transmitter, is operated into an incorrect ter-
RG -8 /U and RG -58 /U cable. Coil L, should be
checked for resonance at the operating fre- mination it may be damaged; the coils may
quncy with C,, and the same with L, and C. be overheated and the capacitors destroyed
This check can be made by soldering a low -
inductance grounding strap to the lead be- as a result of excessive r-f currents. Hence
tween L, and L, where it posses through the it is wise when first installing a low -pass
shield. When the coils have been trimmed to filter, to check the standing -wave ratio of
resonance with a grid -dip meter, the ground-
ing strap should of course be removed. This the load being presented to the output of the
filter type will give an attenuation of about
30 db to the second harmonic, about 48 db to
the third, about 60 db to the fourth, 67 to the FROM TRANS. TO RECEIVER
fifth, etc., increasing at a rate of about 30 db
per octave.
C,, C, C , C,- Silver mica or small ceramic for Rr
low power, transmitting type ceramic for L l! La L4
MI T vvv AN,
r
high power. Capacitance for cliff
given below.
t bands is
5o
C
TLx
100 60 60
I 1 60
Ls
203MH
MHZ
160 meters -1700
80 meters- 850
40 meters- 440
pf
pf
pf
ca c
PF f a E
T60
TT C 6 2114 PF
20 meters- 220 pf
10 meters- 110 pf 30
6 meters- 60 pf m
O 40
L -May made up of sections of 8&W
be z
Miniductor for power levels below 250 watts, o 50
or of No. 12 enam. for power up to one kilo-
watt. Approximate dimensions for the coils are so
g iven below, but the coils should be trimmed z
w
te resonate at the proper freq y with a f- 70
g rid -dip meter as discussed above All coils
except the ones for 160 meters are wound 8 eo
t per inch. loo 150 200 250 300 400
160 meters -4.2 ph; 22 turns No. 16 enam. 1" F (MHZ)
dia. 2" long
80 meters -2.1 ph; 13 t. dia. (No. 3014 Mini -
1 Figure 11
ductor or No. 12)
40 meters -1.1 ph; 8 t. 1" dia. (No. 3014 or No. SIX METER TVI FILTER
12 at 8 t.p.i.)
20 meters -0.55 ph; 7 t. 3/4" dia. (No. 3010 or C, -50 -pf Centralab 8505 -50Z. Resonates with
No. 12 at 8 t.p.I.) L, to 200 MHz.
10 meters-0.3 ph; 6 t. !_" dia. (No. 3002 or C,, C -4-pf piston capacitor. HD type 1/C -40.
No. 12 at 8 t.p.i.) C , C ,C C, -60 pf. Three 20 -pf capacitors
. in
6 meters -0.17 ph; 4 t. '2" dia. (No. 3002 or parallel. Centralab 853A -20Z.
No. 12 at 8 t.p.i.) I.,-Copper strap, I/2" wide, 21/4" long, 1,4"
between mounting holes, approximately 0.01"
thick. Strap is bent in U -shape around capaci-
gree of harmonic attenuation. However, the tor and bolted to capacitor terminals.
L. -11 turns # 18 enam. wire, 1/4" diameter,
filter must be operated in the correct manner 1/4" long, airwound. Resonates to 100 MHz with
or the results obtained will not be up to ex- capacitor C..
L., 1, -3 turns 3/16" tubing, 11/4" i.d., spaced
pectations. to occupy about 21/7 ". Turns are adjusted to
In the first place, all direct radiation from resonate each section at 50 MHz.
L -6 turns #18 enom. wire, 1/4" diameter,
the transmitter and its control and power 5/6" long, oirwound. Resonates to 200 MHz
leads must be suppressed. This subject has with capacitor C..
www.americanradiohistory.com
HAN DBOOK Low -Pass Filters 383
Figure 12
filter with a standing -wave meter of any of ment in series with the input is tuned to
the conventional types. Then the antenna 200 MHz to provide additional protection
termination or the antenna coupling should to television channels 11, 12, and 13.
be adjusted, with low power on the transmit- The filter (figure 12) is built in an alumi-
ter, until the s.w.r. of the load being pre- num box measuring 4" x 4" x 10" and uses
sented to the filter is less than 2.0, and pre- type -N coaxial fittings. The half -wave filter
ferably below 1.5. coils are wound of 3/16 -inch diameter cop-
Half -Wave Filters A half -wave filter is an per tubing and have large copper lugs sol-
effective device for TVI dered to the ends. The 60 -pf capacitors are
suppression and is easily built. It offers the made up of three 20 -pf, Sky ceramic units
advantage of presenting the same value of in parallel. A small sheet of copper is cut in
impedance at the input terminal as appears triangular shape and joins the capacitor
as a load across the output terminal. The terminals and a coil lug is attached to the
filter is a single -band unit, offering high at- center of the triangle with heavy brass bolts.
tenuation to the second- and higher -order The parallel -tuned 200 -MHz series filter
harmonics. Design data for high- frequency element at the input terminal is made of a
half -wave filters is given in figure 10. length of copper strap shunted across a 50-
pf, Sky ceramic capacitor. In this particular
A High -Power The second and higher har- filter, the parallel circuit was affixed to the
Filter for monics of a six -meter trans- output capacitor of the pi- network tank cir-
Six Meters mitter fall directly into the cuit of the transmitter and does not show in
f -m and uhf and vhf tele- the photograph.
vision bands. An effective low -pass filter is The filter is adjusted by removing the
required to adequately suppress unwanted connections from the ends of the half -wave
transmitter emissions falling in these bands. sections and adjusting each section to 50
Described in this section is a six -meter TVI MHz by spreading the turns of the coil with
filter rated at the two -kilowatt level which a screwdriver while monitoring the resonant
provides better than 75 decibels suppression frequency with a grid -dip oscillator. The
of the second harmonic and better than 60 next step is to ground the top end of each
decibels suppression of higher harmonics of series -tuned section (C,, L, and C;, L5) with
a six -meter transmitter (figure 11) . The a heavy strap. The input section is tuned to
unit is composed of a halfwave filter with 100 MHz and the output section to 200
added end sections which are tuned to 100 MHz. When tuning adjustments are com-
MHz and 200 MHz. An auxiliary filter ele- pleted, the straps are removed and the top
384 TV and Broadcast Interference THE RADIO
of the filter box is held in place with sheet - result in much less interference to broadcast
metal screws. reception than either amplitude- modulated
telephony, SSB, or straight keyed c.w. This is
true because, insofar as the broadcast receiver
18 -4 Broadcast is concerned, the amateur f -m transmission
Interference will consist of a plain unmodulated carrier.
There will be no key clicks or voice reception
Interference to the reception of signals in picked up by the broadcast set (unless it
the broadcast band (540 to 1600 kHz) or in happens to be an f -m receiver which might
the f -m broadcast band (88 to 108 MHz) pick up a harmonic of the signal), although
by amateur transmissions is a serious mat- there might be a slight click when the trans-
ter to those amateurs living in densely popu- mitter is put on or taken off the air.
lated areas. Although broadcast interference
Interference Depending on whether it is
has recently been overshadowed by the
Classifications traceable directly to causes
seriousness of television interference, the con-
dition of BCI is still present. within the station or within
In general, signals from a transmitter the receiver, broadcast interference may be
operating properly are not picked up by re- divided into two main classes. For example,
ceivers tuned to other frequencies unless the that type of interference due to transmitter
receiver is of inferior design, or is in poor overmodulation is at once listed as being
condition. Therefore, if the receiver is of caused by improper operation, while an inter-
good design and is in good repair, the burden fering signal that tunes in and out with a
of rectifying the trouble rests with the broadcast station is probably an indication of
cross modulation or image response in the re-
owner of the interfering station. Phone and
c -w stations both are capable of causing
ceiver, and the poorly designed input stage
broadcast interference, key -click annoyance of the receiver is held liable. The various
from the code transmitters being particularly types of interference and recommended cures
objectionable. will be discussed in the following paragraphs.
A knowledge of each of the several types Blanketing This is not a tunable effect, but
of broadcast interference, their cause, and atotal blocking of the receiver.
methods of eliminating them is necessary A more or less complete "washout" covers
for the successful disposition of this trouble. the entire receiver range when the carrier is
An effective method of combating one vari- switched on. This produces either a com-
ety of interference is often of no value plete blotting out of all broadcast stations,
whatever in the correction of another type. or else knocks down their volume several
Broadcast interference seldom can be cured decibels-depending on the severity of the
by "rule -of- thumb" procedure. interference. Voice modulation of the carrier
Broadcast interference, as covered in this causing the blanketing will be highly dis-
section refers primarily to standard (ampli- torted or even unintelligible. Keying of the
tude- modulated, 550 -1600 kHz) broadcast. carrier which produces the blanketing will
Interference with f -m broadcast reception is cause an annoying fluctuation in the volume
much less common, due to the wide separa- of the broadcast signals.
tion in frequency between the f -m broad- Blanketing generally occurs in the imme-
cast band and the more popular amateur diate neighborhood (inductive field) of a
bands, and due also to the limiting action powerful transmitter, the affected area being
which exists in all types of f -m receivers. directly proportional to the power of the
Occasional interference with f -m broadcast transmitter. Also it is more prevalent with
by a harmonic of an amateur transmitter has transmitters which operate in the 160 -meter
been reported; if this condition is encount- and 80 -meter bands, as compared to those on
ered, it may be eliminated by the procedures the higher frequencies.
discussed in the first portion of this chapter The remedies are to (1) shorten the re-
under Television Interference. ceiving antenna and thereby shift its reso-
The use of frequency -modulation trans- nant frequency, (2) remove it to the interior
mission by an amateur station is likely to of the building, (3) change the direction of
www.americanradiohistory.com
HAN DBOOK Broadcast Interference 385
Figure 13
Figure 14
HIGH -ATTENUATION WAVETRAP
COIL AND CAPACITOR TABLE FOR
CIRCUIT
AMATEUR -BAND WAVETRAPS
The two circuits may be tuned to the same fre-
quency for highest attenuation of a strong
signal, or the two traps may be tuned sep- frequency range and must be readjusted for
arately for different bands of operation.
other frequencies.
The wavetrap must be installed as close
either the receiving or transmitting antenna to the receiver antenna terminal as practica-
to minimize their mutual coupling, or, (4) ble, hence it should be as small in size as
keep the interfering signal from entering the possible. The variable capacitor may be a
receiver input circuit by installing a wave - midget air -tuned trimmer type, and the coil
trap tuned to the signal frequency (see fig- may be wound on a 1 -inch dia. form. The
ure 12) or a low -pass filter as shown in fig- table of figure 14 gives winding data for
ure 21. wavetraps built around standard variable
A suitable wavetrap is quite simple in con- capacitors. For best results, both a shunt and
struction, consisting only of a coil and mid- a series trap should be employed as shown.
get variable capacitor. When the trap circuit Figure 11 shows a two -circuit coupled
is tuned to the frequency of the interfering wavetrap which is somewhat sharper in tun-
signal, little of the interfering voltage ing and more efficient. The specifications for
reaches the grid of the first tube. Commer- the secondary coil (L1) may be obtained
cially manufactured wavetraps are available from the table of figure 14. The primary coil
from several concerns, including the J. W.
Miller Co. in Los Angeles. However, the
majority of amateurs prefer to construct the
traps from spare components selected from
the junk box.
The circuit shown in figure 13 is particu-
larly effective because it consists of two
traps. The shunt trap blocks or rejects the
frequency to which it is tuned, while the
series trap across the antenna and ground
terminals of the receiver provides a very -
low- impedance path to ground at the fre-
quency to which it is tuned and bypasses the
signal to ground. In moderate interference Figuro 15
cases, either the shunt or series trap may be
MODIFICATION OF THE FIGURE 13
used alone, while similarly, one trap may be
CIRCUIT
tuned to one of the frequencies of the inter-
fering transmitter and the other trap to a In this circuit arrangement the parallel -tuned
different interfering frequency. In either tank is inductively coupled to the antenna
lead with a 3 to 6 turn link instead of being
case, each trap is effective over but a small placed directly in series with the antenna lead.
386 TV and Broadcast Interference THE RADIO
www.americanradiohistory.com
HAN D BO O K Broadcast Interference 387
Figure 18
CV69
L_ -_ wavetrap coils, while figure 19 illustrates
SHIELD BRAID the method of connecting these wavetraps.
J Observe that these traps are enclosed in
a shield box of iron or aluminum, well
Figure 19
grounded.
Image Interference In addition to those types
RESONANT POWER -LINE
of interference already
WAVE -TRAP CIRCUIT
discussed, there are two more which are com-
The resonant type of power -line Alter is more mon to superhet receivers. The prevalence of
effective than the more c tional "brute
force" type of line Biter, but requires tuning these types is of great concern to the ama-
to the operating frequency of the transmitter. teur, although the responsibility for their
existence more properly rests with the broad-
transmitter, or picked up from the air by cast receiver.
overhead power lines. Underground lines are The mechanism whereby image production
seldom responsible for spreading this inter- takes place may be explained in the following
ference. manner: when the first detector is set to the
To check the path whereby the interfering frequency of an incoming signal, the high -
signals reach the line, it is only necessary to frequency oscillator is operating on another
replace the transmitting antenna with a frequency which differs from the signal by
dummy antenna and adjust the transmitter the number of kHz of the intermediate
for maximum output. If the interference frequency. Now, with the setting of these
then ceases, overhead lines have been picking two stages undisturbed, there is another
up the energy. The trouble can be cleared up signal which will beat with the high -fre-
by installing a wavetrap or a commercial quency oscillator to produce an i -f signal.
line filter in the power lines at the receiver. This other signal is the so- called image,
If the receiver is reasonably close to the which is separated from the desired signal
transmitter, it is very doubtful that chang- by twice the intermediate frequency.
ing the direction of the transmitting an- Thus, in a receiver with a 175 -kHz inter-
tenna to right angles with the overhead lines mediate frequency tuned to 1000 kHz; the
will eliminate the trouble. h -f oscillator is operating on 1175 kHz, and
If, on the contrary, the interference con- a signal on 1350 kHz (1000 kHz plus 2 X
tinues when the transmitter is connected to 175 kHz) will beat with this 1175 kHz
the dummy antenna, radio- frequency energy oscillator frequency to produce the 175 -kHz
is being fed directly into the power line by i -f signal. Similarly, when the same receiver
the transmitter, and the station must be is tuned to 1450 kHz, an amateur signal on
inspected to determine the cause. 1800 kHz can come through.
One of the following reasons for the trou- If the image appears only a few Hz or
ble will usually be found: (1) the r -f stages kHz from a broadcast carrier, heterodyne
are not sufficiently bypassed and /or choked, interference will be present as well. Other-
(2) the antenna coupling system is not per- wise, it will be tuned in and out in the
forming efficiently, (3) the power trans- manner of a station operating in the broad-
formers have no electrostatic shields; or, if cast band. Sharpness of tuning will be com-
shields are present, they are ungrounded, (4) parable to that of broadcast stations produc-
power lines are running too close to an an- ing the same avc voltage at the receiver.
tenna or r -f circuits carrying high currents. The second variety of superhet interfer-
If none of these causes apply, wavetraps ence is the result of harmonics of the receiver
must be installed in the power lines at the high- frequency oscillator beating with ama-
transmitter to remove r-f energy passing teur carriers to produce the intermediate
back into the lighting system. frequency of the receiver. The amateur
390 TV and Broadcast Interference THE RADIO
transmitter will always be found to be on task of removing amateur frequencies from
a frequency equal to some harmonic of the broadcast receivers.
receiver hf oscillator, plus or minus the A good low -pass filter designed for maxi-
intermediate frequency. mum attenuation around 1700 kHz will pass
As an example: when a broadcast superhet all broadcast carriers, but will reject signals
with 465 -kHz intermediate frequency is originating in any amateur band. Naturally
tuned to 1000 kHz, its high- frequency oscil-
lator operates on 1465 kHz. The third har-
monic of this oscillator frequency is 4395
kHz, which will beat with an amateur signal
on 3930 kHz to send a signal through the
i -f amplifier. The 3930 kHz signal would be
tuned in at the 1000 -kHz point on the dial. INPUT OUTPUT
www.americanradiohistory.com
HANDBOOK H i-Fi Interference 391
www.americanradiohistory.com
Power Systems 393
xMTR xMTR
a
OUTLET
STRIP
DESK
USE
XMTR
AND
mTE
DESK
'NMI
MAIN CONDUIT
FUSE BOX 7
AND =__ TO OTHER
SWITCH _= HOUSE CIRCUITS
_IJ
FROM LINE PLAN OA
SHORT COROS FROM
RECEIVER VFO..CLOCK,
FRED. METER, TO
Figure 1 PLAN OUTLET STRIP.
Checking on To make sure that an outlet may be inserted in the outlet strip through-
Outlet with a will stand the full load of the out its length. In many cases it will be de-
Heavy Load entire transmitter, plug in an sirable to reduce the equipment cord lengths
electric heater rated at about so that they will plug neatly into the outlet
50 percent greater wattage than the power strip without an excess to dangle behind the
you expect to draw from the line. If the line desk.
voltage does not drop more than f volts
(assuming a 115 -volt line) under load and Contactors and The use of power- control
the wiring does not overheat, the wiring is Relays contactors and relays often
adequate to supply the transmitter. About will add considerably to the
600 watts total drain is the maximum that operating convenience of the station instal-
should be drawn from a 115 -volt lighting lation. The most practical arrangement usu-
outlet or circuit. For greater power, a sep- ally is to have a main a -c line switch on the
arate pair of heavy conductors should be front of the transmitter to apply power to
run right from the meter box. For a 1 -kw. the filament transformers and to the power-
phone transmitter the total drain is so great control circuits.
that a 230 -volt "split" system ordinarily It also will be found quite convenient to
will be required. Most of the newer homes have a single a -c line switch on the operating
are wired with this system, as are homes desk to energize or cut the power from the
utilizing electricity for cooking and heating. outlet strip on the rear of the operating
With a three -wire system, be sure there is desk. Through the use of such a switch it
no fuse in the neutral wire at the fuse box. is not necessary to remember to switch off
A neutral fuse is not required if both "hot" a large number of separate switches on each
legs are fused, and, should a neutral fuse of the items of equipment on the operating
blow, there is a chance that damage to the desk.
radio transmitter will result. The alternative arrangement, and that
If you have a high -power transmitter and which is approved by the Underwriters, is
do a lot of operating, it is a good idea to to remove the plugs from the wall both for
check on your local power rates if you are the transmitter and for the operating -desk
on a straight lighting rate. In some cities a outlet strip when a period of operation has
lower rate can be obtained (but with a been completed.
higher "minimum ") if electrical equipment While the insertion of plugs or operation
such as an electric heater drawing a specified of switches usually will be found best for
amount of current is permanently wired in. applying the a -c line power to the equip-
It is not required that you use this equip- ment, the changing over between transmit
ment, merely that it be permanently wired and receive can best be accomplished through
into the electrical system. Naturally, how- the use of relays. Such a system usually in-
ever, there would be no saving unless you volves three relays, or three groups of relays.
expect to occupy the same dwelling for a
considerable length of time.
The relays and their functions are: (1)
power- control relay for the transmitter
applies 115 -volt line to the primary of the
-
Outlet Strips The outlet strips which have
been suggested for installation
in the baseboard or for use on the rear of a
high -voltage transformer and turns on the
exciter; (2) control relay for the receiver
makes the receiver inoperative by any one of
-
desk are obtainable from the large electrical - a number of methods when closed, also may
supply houses. If such a house is not in the apply power to the vfo and to a keying or a
vicinity it is probable that a local electrical phone monitor; and (3) the antenna change-
contractor can order a suitable type of strip over relay -connects the antenna to the
from one of the supply -house catalogs. These transmitter when the transmitter is energized
strips are quite convenient in that they are and to the receiver when the transmitter is
available in varying lengths with provision not operating. Several circuits illustrating
for inserting a -c line plugs throughout their the application of relays to such control ar-
length. The a -c plugs from the various items rangement are discussed in the paragraphs
of equipment on the operating desk then to follow in this chapter.
www.americanradiohistory.com
HANDBOOK Power Systems 395
TO EXCITER POWER SUPPLIES acts as the exciter for the higher -powered
stage when full power output is required.
But the majority of the amateurs using high -
s power equipment have some provision for
reducing the plate voltage on the high -level
stages when reduced power output is desired.
IIS V.A.C.
LINE One of the most common arrangements
for obtaining two levels of power output
K
NI -LO
POWER RELAY
involves the use of a plate transformer hav-
ing a double primary for the high -voltage
e POWER - CONTROL RELAY
power supply. The majority of the high -
power plate transformers of standard manu-
TO FILAMENT TRANSFORMERS facture have just such a dual -primary ar-
rangement. The two primaries are designed
for use with either a 115 -volt or 230 -volt
line. When such a transformer is to be
operated from a 115 -volt line, operation of
both primaries in parallel will deliver full
output from the plate supply. Then when
the two primaries are connected in series and
still operated from the 115 -volt line the out-
put voltage from the supply will be reduced
approximately to,one half. In the case of the
230 V.A C
normal class -C amplifier, a reduction in plate
SINGLE PHASE
WITH GROUNDED
NEUTRAL voltage to one half will reduce the power
input to the stage to one quarter.
If the transmitter is to be operated from a
TRANSFORMERS 230 -volt line, the usual procedure is to op-
erate the filaments from one side of the line,
Figure 3
the low -voltage power supplies from the
FULL -VOLTAGE /HALF -VOLTAGE other side, and the primaries of the high -
SYSTEM OF POWER CONTROL voltage transformer across the whole line
circuit at A is for use with a 11S -volt
Tho
for full power output. Then when reduced
lino. Transformer T is of the standard type
a -c power output is required, the primary of
having two IIS-volt primaries; these primaries the high- voltage plate transformer is oper-
are connectod in series for half -voltage output
when the power -control relay K, is energised ated from one side to center tap rather than
but the hi-lo relay (K,) is not operated. When across the whole line. This procedure places
both relays are ised the full output volt- 115 volts across the 230 -volt winding the
age is obtained. At S is a circuit for use with
a standard 230 -volt residence line with same as in the case discussed in the previous
grounded neutral. The two relays control the paragraph. Figure 3 illustrates the two stand-
output of the power supplies the same as at A.
ard methods of power reduction with a plate
transformer having a double primary; A
Controlling Transmitter It is necessary, in shows the connections for use with a 115-
Power Output order to comply with volt line and B shows the arrangement for a
FCC regulations, that 230 -volt a -c power line to the transmitter.
transmitter power output be limited to the The full -voltage /half -voltage methods for
minimum amount necessary to sustain com- controlling the power input to the transmit-
munication. This requirement may be met ter, as just discussed, are subject to the limi-
in several ways. Many amateurs have two tation that only two levels of power input
transmitters; one is capable of relatively high (full power and quarter power) are obtain-
power output for use when calling, or when able. In many cases this will be found to be a
interference is severe, and the other is capa- limitation to flexibility. When tuning the
ble of considerably less power output. In transmitter, the antenna coupling network,
many cases the lower -powered transmitter or the antenna system itself it is desirable to
396 Transmitter Keying and Control THE RADIO
TO H.V.
POWER SUPPLY
INTERLOCKS
IN TRANSMITTER
11 A.C. F
LINE POWER CONTROL
RELAY
TRANSMITTER
GREEN FILAMENT
PLUG FOR PILOT TRANSFORMERS
ABLE TO
ARIAC OR
POWERSTAT
DUMMY PLUG FOR
STRAIGHT OPERATION
115 V. TO EXCITER AND
RED HIGH -VOLTAGE RELAYS.
AND TO RECEIVER CON-
TO FILAMENT PI LOT
TROL AND ANTENNA -
TRANSFORMERS CHANGEOVER RELAYS
L sa J
EXTERNAL VARIAC Figure 5
OR POWERSTAT
be able to reduce the power input to the capable of about 1500 to 1800 watts. The
final stage to a relatively low value, and it maximum power- output capability of these
is further convenient to be able to vary the units is available only at approximately the
power input continuously from this rela- nominal line voltage, and must be reduced to
tively low input up to the full power capa- a maximum current limitation when the out-
bilities of the transmitter. The use of a put voltage is somewhat above or below the
variable -ratio autotransformer in the circuit input line voltage. This, however, is not an
from the line to the primary of the plate important limitation for this type of appli-
transformer will allow a continuous varia- cation since the output voltage seldom will
tion in power input from zero to the full be raised above the line voltage, and when
capability of the transmitter. the output voltage is reduced below the line
Variable -Ratio voltage the input to the transmitter is re-
There are several types of
Autotransformers duced accordingly.
variable -ratio autotrans-
One convenient arrangement for using a
formers available on the
Variac or Powerstat in conjunction with the
market. Of these, the most common are the
high -voltage transformer of a transmitter is
Variac manufactured by the General Radio
illustrated in figure 4. In this circuit a heavy
Company, and the Powerstat manufactured
three -wire cable is run from a plug on the
by the Superior Electric Company. Both these
types of variable -ratio transformers are ex-
transmitter to the Variac or Powerstat. The
cellently constructed and are available in a Variac or Powerstat then is installed so that
wide range of power capabilities. Each is it is accessible from the operating desk so that
capable of controlling the line voltage from the input power to the transmitter may be
zero to about 15 percent above the nominal controlled during operation. If desired, the
line voltage. Each manufacturer makes a cable to the Variac or Powerstat may be un-
single -phase unit capable of handling an out- plugged from the transmitter and a dummy
put power of about 175 watts, one capable plug inserted in its place. With the dummy
of about 750 to 800 watts, and a unit plug in place the transmitter will operate at
www.americanradiohistory.com
HANDBOOK Transmitter Control 397
normal plate voltage. This arrangement al- where it is only necessary to speak into the
lows the transmitter to be wired in such a microphone or touch the key to change both
manner that an external Variac or Powerstat transmitter and receiver over to the transmit
may be used if desired, even though the unit condition. Most amateur stations are inter-
is not available at the time that the trans- mediate between the two extremes in the
mitter is constructed. control provisions and use some relatively
Notes on the Use
simple system for transmitter control.
Plate voltage to the modu- In figure 5 is shown an arrangement which
of the Voriac lators may be controlled at
or Powerstat
protects mercury-vapor rectifiers against pre-
the same time as the plate mature application of plate voltage without
voltage to the final ampli- resorting to a time -delay relay. No matter
fier is varied if the modulator stage uses
which switch is thrown first, the filament
beam -tetrode tubes; variation in the plate will be turned on first and off last. However,
voltage on such tubes used as modulators
double -pole switches are required in place of
causes only a moderate change in the stand-
the usual single -pole switches.
ing plate current.
When assured time delay of the proper
Since the final amplifier plate voltage is
interval and greater operating convenience
being controlled simultaneously with the
are desired, a group of inexpensive a -c relays
modulator plate voltage, the conditions of
may be incorporated into the circuit to give
impedance match will not be seriously up-
a control circuit such as is shown in figure
set. In several high -power transmitters using
6. This arrangement uses a 115-volt thermal
this system, and using beam - tetrode modu-
(or motor -operated) time-delay relay and a
lator tubes, it is possible to vary the plate
input from about 50 watts to one kilowatt dpdt 115 -volt control relay. Note that the
protective interlocks are connected in series
without a change other than a slight in-
with the coil of the relay which applies high
crease in audio distortion at the adjustment
voltage to the transmitter. A tune -up switch
which gives the lowest power output from
has been included so that the transmitter
the transmitter.
may be tuned up as far as the grid circuit of
With triode tubes as modulators it usually
the final stage is concerned before application
will be found necessary to vary the grid bias
at the same time that the plate voltage is
cf high voltage to the final amplifier. Pro-
visions for operating an antenna -changeover
changed. This will allow the tubes to be
relay and for cutting the plate voltage to
cperated at approximately the same relative
the receiver when the transmitter is operat-
point on their operating characteristic when
ing have been included.
the plate voltage is varied. When the mod-
A circuit similar to that of figure 6 but
ulator tubes are operated with zero bias at
full plate voltage, it will usually be possible incorporating push- button control of the
to reduce the modulator voltage along with transmitter is shown in figure 7. The circuit
features a set of START -STOP and TRANS-
the voltage on the modulated stage, with no
MIT- RECEIVE buttons at the transmitter
apparent change in the voice quality. How-
and a separate set at the operating position.
ever, it will be necessary to reduce the audio
The control push buttons operate independ-
gain at the same time that the plate voltage
is reduced.
ently so that either set may be used to con-
trol the transmitter. It is only necessary to
push the START button momentarily to
19 -2 Transmitter Control light the transmitter filaments and start the
Methods time -delay relay in its cycle. When the
standby light comes on it is only necessary to
Almost everyone, when getting a new touch the TRANSMIT button to put the
transmitter on the air, has had the experience transmitter on the air and disable the re-
of having to throw several switches and pull ceiver. Touching the RECEIVE button will
or insert a few plugs when changing from turn off the transmitter and restore the re-
receive to transmit. This is one extreme in ceiver. After a period of operation it is only
the direction of how not to control a trans- necessary to touch the STOP button at
mitter. At the other extreme we find systems either the transmitter or the operating posi-
398 Transmitter Keying and Control THE RADIO
USCS
SAFETY SWITCH
(SEC Fm.* )
RECEIVER POWER
\SI HUSKY TOGGLE SWITCH
ON TRANSMITTER
TRANSFORMER C.T.
THCRMAL PROTECTIVE
TIMO -DEL AY INTERLOCKS
RCLAY O O
-o TRANSMIT -
RECEIVE SWITCH
HIGH-VOLT.
(I15-V.)
011S-V. ANTENNA-
CHANGEOVER
o`o o
TUNO -UP
INDICATOR LIGHTS SWITCH
Figure 6
tion to shut down the transmitter. This type structing a transmitter, the following safety
of control arrangement is called an elec- considerations should be given attention.
trically locking push-to- transmit control Grounds For the utmost in protection, every-
system. Such systems are frequently used in thing of metal on the front panel of
industrial electronic control. a transmitter capable of being touched by
the operator should be at ground potential.
19 -3 Safety Precautions This includes dial set screws, meter zero -
The best way for an operator to avoid adjustment screws, meter cases if of metal,
serious accidents from the high voltage sup- meter jacks, everything of metal protruding
plies of a transmitter is for him to use his through the front panel or capable of being
head, act only with deliberation, and not touched or nearly touched by the operator.
take unnecessary chances. However, no one This applies whether or not the panel itself is
is infallible, and chances of an accident are
of metal. Do not rely on the insulation of
greatly lessened if certain factors are taken meter cases or tuning knobs for protection.
into consideration in the design of a trans- The B negative or chassis of all plate power
mitter, in order to protect the operator in supplies should be connected together, and to
the event of a lapse of caution. If there are an external ground such as a waterpipe.
too many things one must "watch out for" Exposed Wires It
not necessary to resort
is
or keep in mind there is a good chance that and Components to
rack- and -panel con-
sooner or later there will be a mishap; and struction in order to pro-
it only takes one. When designing or con- vide complete enclosure of all components
www.americanradiohistory.com
HANDBOOK Safety Precautions 399
SAP ET Y SWITCH
-o o- .DFlo- Ong -I
(SEE FIG.12)
PROTECTIVE
INTERLOCKS
OVERLOAD"'
RELAY
CONTACTS
O RECEIVER POWER
o TRANSFORMER C.T.
THERMAL
TIME-DELAY
RELAY
HIGH.VOLT.
FILS ON STANDBY (11S-V.)
15-V. ANTENNA -
CHANGEOVER
RELAY
TUNE -UP
INDICATOR LIGHTS SWITCH
900, lOOO,
III
ALL FILAMENT TRANSFORMERS EXCITER H -V HIGH- VOLTAGE
TRANSFORMER TRANSFORMER
Figure 7
and wiring of the transmitter. Even with when changing coils, neutralizing, adjusting
metal -chassis construction it is possible to coupling, or troubleshooting. The latter pro-
arrange things so as to incorporate a protec- cedure can be made safe by making it pos-
tive shielding housing which will not inter- sible for the operator to be absolutely certain
fere with ventilation yet will prevent con- that all voltages have been turned off and
tact with all wires and components carrying that they cannot be turned on either by
high voltage d.c. or a.c., in addition to short circuit or accident. This can be done
offering shielding action. by incorporation of the following system of
If everything on the front panel is at main primary switch and safety signal lights.
ground potential (with respect to external
ground) and all units are effectively housed Combined Safety The common method of
with protective covers, then there is no Signal and Switch using red pilot lights to
danger except when the operator must reach show when a circuit is on
into the interior part of the transmitter, as is useless except from an ornamental stand-
400 Transmitter Keying and Control THE RADIO
!.S V.
TO GREEN PILOT LIGHTS ON behind green bezels or dipped in green
FRONT PANEL AND ON EACH CHASSIS
lacquer. One should be placed on the front
.11.J
panel of the transmitter; others should be
FIL. placed so as to be easily visible when chang-
ing coils or making adjustments requiring
the operator to reach inside the transmitter.
For 100 percent protection, just obey the
MAIN IIS V SUPPLY O- -1(0 POT SWITCH
following rule: never work on the transmit-
ter or reach inside any protective cover ex-
111S -V A C TO ENTIRE TRANSMITTER
cept when the green pilots are glowing. To
avoid confusion, no other green pilots should
be used on the transmitter; if you want an
Figure 8
indicator jewel to show when the filaments
COMBINED MAIN SWITCH AND are lighted, use amber instead of green.
SAFETY SIGNAL
Safety Bleeders Filter capacitors of good
When shutting down the transmitter, throw
the main switch to neutral. If work is to be quality hold their charge for
done on the transmitter, throw the switch all some time, and when the voltage is more
the way to "pilot," thus turning on the green than 100 volts it is just about as dangerous
pilot lights on the panel and on each chassis,
and ensuring that no voltage can exist on the to get across an undischarged 4 -td filter
primary of any transformer, even by virtue capacitor as it is to get across a high -voltage
of a short or accidental ground.
supply that is turned on. Most power sup-
plies incorporate bleeders to improve regula-
point. When the red pilot is not lit it usually tion, but as these are generally wirewound
means that the circuit is turned off, but it resistors, and as wirewound resistors occa-
can mean that the circuit is on but the lamp sionally open up without apparent cause, it is
is burned out or not making contact. desirable to incorporate an auxiliary safety
To enable you to touch the tank coils in bleeder across each heavy -duty bleeder. Car-
your transmitter with absolute assurance bon resistors will not stand much dissipation
that it is impossible for you to obtain a and sometimes change in value slightly with
shock except from possible undischarged age. However, the chance of their opening
filter capacitors (see following topic for up when run well within their dissipation
elimination of this hazard), it is only neces- rating is very small.
sary to incorporate a device similar to that To make sure that all capacitors are bled,
of figure 8. It is placed near the point where it is best to short each one with an insulated
the main 115 -volt leads enter the room (pre- screwdriver. However, this is sometimes awk-
ferably near the door) and in such a position ward and always inconvenient. One can be
as to be inaccessible to small children. Notice virtually sure by connecting auxiliary car-
that this switch breaks both leads; switches bon bleeders across all wirewound bleeders
that open just one lead do not afford com- used on supplies of 1000 volts or more. For
plete protection, as it is sometimes possible to every 500 volts, connect in series a 500,000 -
complete a primary circuit through a short ohm 1 -watt carbon resistor. The drain will
or accidental ground. Breaking just one side be negligible (1 ma) and each resistor will
of the line may be all right for turning the have to dissipate only 0.5 watt. Under these
transmitter on and off, but when you are conditions the resistors will last indefinitely
going to place an arm inside the transmitter, with little chance of opening up. For a
both 115 -volt leads should be broken. 1500 -volt supply, connect three 500,000 -ohm
When you are all through working your resistors in series. If the voltage exceeds an
transmitter for the time being, simply throw integral number of 500 -volt divisions, as-
the main switch to neutral. sume it is the next higher integral value; for
When you find it necessary to work on instance, assume 1800 volts as 2000 volts and
the transmitter or change coils, throw the use four resistors.
switch so that the green pilots light up. Do not attempt to use fewer resistors by
These can be ordinary 6.3 -volt pilot lamps using a higher value for the resistors; not
www.americanradiohistory.com
HANDBOOK Transmitter Keying 401
over 500 volts should appear across any a 150 -volt circuit under certain circum-
single 1 -watt resistor. stances. The 150 -volt limit usually will
In the event that the regular bleeder opens mean that grid -bias packs as well as high -
up, it will take several seconds for the auxil- voltage packs should have their primary cir-
iary bleeder to drain the capacitors down to cuits opened when any interlock is opened.
a safe voltage, because of the very high re-
sistance. Therefore it is best to allow 10 or
15 seconds to elapse after turning off the
19 -4 Transmitter Keying
plate supply before attempting to work on
the transmitter. The carrier from a c -w telegraph trans-
If a 0 -1 d -c milliammeter is at hand, it mitter must be broken into dots and dashes
may be connected in series with the auxiliary for the transmission of code characters. The
bleeder to act as a high- voltage voltmeter. carrier signal is of constant amplitude while
"Hot" Adjustments Some amateurs contend the key is closed, and is entirely removed
that it is almost impossi- when the key is open. When code characters
are being transmitted, the carrier may be
ble to make certain adjustments, such as
considered as being modulated by the keying.
coupling and neutralizing, unless the trans-
mitter is running. The best thing to do is to If the change from the no- output condition
to full -output, or vice versa, occurs too
make all neutralizing and coupling devices
rapidly, the rectangular pulses which form
adjustable from the front panel by means of
flexible control shafts which are broken with
the keying characters contain high -fre-
insulated couplings to permit grounding of quency components which take up a wide
frequency band as sidebands and are heard
the panel bearing.
as clicks.
If your particular transmitter layout is
such that this is impractical and you refuse To be capable of transmitting code char-
to throw the main switch to make an adjust- acters and at the same time not splitting the
ment -throw the main switch -take a read- eardrums of neighboring amateurs, the c -w
ing-throw the main switch-make an ad- transmitter must meet two important speci-
justment -and so on, then protect yourself fications.
by making use of long adjustment rods made 1. It must have no parasitic oscillations
from 1/4-inch dowel sticks which have been either in the stage being keyed or in
wiped with oil when perfectly free from any succeeding stage.
moisture. 2. It must have some device in the keying
Protective Interlocks With the increasing ten- circuit capable of shaping the leading
dency toward construc- and trailing edge of the waveform.
tion of transmitters in enclosed steel cabinets Both these specifications must be met be-
a transmitter becomes a particularly lethal fore the transmitter is capable of c -w opera-
device unless adequate safety provisions have tion. Merely turning a transmitter on and off
been incorporated. Even with a combined by the haphazard insertion of a telegraph
safety signal and switch as shown in figure 8 key in some power lead is an invitation to
it is still conceivable that some person un- trouble.
familiar with the transmitter could come in The two general methods of keying a
contact with high voltage. It is therefore transmitter are those which control the ex-
recommended that the transmitter, when- citation to the keyed amplifier, and those
ever possible, be built into a complete metal which control the plate or screen voltage ap-
housing or cabinet and that all doors or ac- plied to the keyed amplifier.
cess covers be provided with protective inter-
locks (all interlocks must be connected in Key -Click Key -click elimination is accom-
series) to remove the high voltage whenever Elimination plished by preventing a too -rapid
these doors or covers are opened. The term make and break of power to the
"high voltage" should mean any voltage antenna circuit, thus rounding off the keying
above approximately 150 volts, although it characters so as to limit the sidebands to a
is still possible to obtain a serious burn from value which does not cause interference to
402 Transmitter Keying and Control THE RADIO
adjacent channels. Too much lag will pre- waveform. A heavily excited doubler stage or
vent fast keying, but fortunately key clicks class -C stage acts as a peak clipper, tending
can be practically eliminated without limit- to square up a rounded keying impulse, and
ing the speed of manual (hand) keying. the cumulative effect of several such stages
Some circuits which eliminate key clicks in- cascaded is sufficient to square up the keyed
troduce too much time lag and thereby add waveform to the point where bad clicks are
tails to the dots. These tails may cause the reimposed on a clean signal.
signals to be difficult to copy at high speeds. A good rule of thumb is to never key back
farther than one stage removed from the
Location of Considerable thought should be final amplifier stage, and never key closer
Keyed Stage given as to which stage in a than one stage removed from the frequency -
transmitter is the proper one to controlling oscillator of the transmitter.
key. If the transmitter is keyed in a stage Thus there will always be one isolating stage
close to the oscillator, the change in r -f load- between the keyed stage and the oscillator,
ing of the oscillator will cause the oscillator and one isolating stage between the keyed
to shift frequency with keying. This will stage and the antenna. At this point the
cause the signal to have a distinct chirp. waveform of the keyed signal may be most
The chirp will be multiplied as many times easily controlled.
as the frequency of the oscillator is multi-
plied. A chirpy oscillator that would be pass- Keyer Circuit In the first place it may be es-
able on 80 meters would be unusable on 28- Requirements tablished that the majority of
MHz c.w. new design transmitters, and
Keying the oscillator itself is an excellent many of those of older design as well, use a
way to run into keying difficulties. If no medium power beam tetrode tube either as
key -click filter is used in the keying circuit, the output stage or as the exciter for the
the transmitter will have bad key clicks. If a output stage of a high power transmitter.
key -click filter is used, the slow rise and de- Thus the transmitter usually will end up
cay of oscillator voltage induced by the filter with a tube such as type 2E26, 807, 6146,
action will cause a keying chirp. This action 813, 4 -61A, 4E27/5-125B, 4 -125A or simi-
is true of all oscillators, whether electron lar, or one of these tubes will be used as the
coupled or crystal controlled. stage just ahead of the output stage.
The more amplifier or doubler stages that Second, it may be established that it is un-
follow the keyed stage, the more difficult it desirable to key further down in the trans-
is to hold control of the shape of the keyed mitter chain than the stage just ahead of the
final amplifier. If a low -level stage, which is
followed by a series of class -C amplifiers, is
keyed, serious transients will be generated in
the output of the transmitter even though
the keyed stage is being turned on and off
very smoothly. This condition arises as a re-
sult of pulse sharpening, which has been dis-
cussed previously.
Third, the output from the stage should
be completely cut off when the key is up,
and the time constant of the rise and decay
Figure 9 of the keying wave should be easily con-
trollable.
CENTER-TAP KEYING WITH CLICK
Fourth, it should be possible to make the
FILTER
rise period and the decay period of the key-
The constants shown above are suggested as ing wave approximately equal. This type of
starting values; considerable variation in these keying envelope is the only one tolerable for
values can be expected for optimum keying of
amplifiers of different operating conditions. It commercial work, and is equally desirable for
is suggested that a keying relay be substituted
for the key in the circuit above wh obtaining clean -cut and easily readable sig-
practical. nals in amateur work.
www.americanradiohistory.com
HANDBOOK Cathode Keying 403
TO
CATHODE R -F TUBE
SOMA SILICON- lu BREAK) 1M(MAKE)
RECTIFIER
O
471W 47K,2W O 1
- frRF
120V.
10
150 V.
2.SMN O
100 IDO
W KEY M
TO %MTR
GRID.
STANCOR PAl421
2A3
OR6AS7
350 -0350
50 MA.
80
70 K
IW
- 4
teJO v
K Y
RFC
2.5 MN
100 . 1
e-
.00
N105
4
H-v SUPPLY
E-41.
Figure 10
Fifth, it is desirable that the keying circuit the backwave that might otherwise leak
be usable without a keying relay, even when through the keyed stage.
a high -power stage is being keyed. The simplest cathode keying circuit is il-
Last, for the sake of simplicity and safety, lustrated in figure 9, where a key -click filter
it should be possible to ground the frame of is employed, and a hand key is used to break
the key, and yet the circuit should be such the circuit. This simple keying circuit is not
that placing the fingers across the key will recommended for general use, as consider-
not result in an electrical shock. In other able voltage will be developed across the key
words, the keying circuit should be inher- when it is open.
ently safe. An electronic switch can take the place of
All these requirements have been met in the hand key. This will remove the danger
the keying circuits to be described. of shock. At the same time, the opening
and closing characteristics of the electronic
19 -5 Cathode Keying switch may easily be altered to suit the par-
ticular need at hand. Such an electronic
The lead from the cathode or center -tap switch is called a vacuum -tube keyer. Low
connection of the filament of an r-f amplifier internal resistance triode tubes such as the
can be opened and closed for a keying cir- 2A3 or 6AS7 are used in the keyer. These
cuit. Such a keying system opens the plate tubes act as a very high resistance when
voltage circuit and at the same time opens sufficient blocking bias is applied to them,
the grid bias return lead. For this reason, the and as a very low resistance when the bias is
grid circuit is blocked at the same time the removed. The desired amount of lag or
plate circuit is opened. This helps to reduce cushioning effect can be obtained by em-
404 Transmitter Keying and Control THE RADIO
NI -.y_ TRIODE
(811 -A ETC.)
Figure 11 +N.V.
ploying suitable resistance and capacitance as cathode bias on the keyed tube, assuming
values in the grid of the keyer tube(s). Be- the bias return is made to ground, and should
cause very little spark is produced at the key, be taken into consideration when providing
due to the small amount of power in the key bias.
circuit, sparking clicks are easily suppressed. Some typical cathode circuit vacuum -tube
One type 6AS7 tube (both sections) keying units are shown in figure 10.
should be used for every 250 ma of plate cur-
rent. Type 2A3 tubes may also be used; allow
one 2A3 tube for every 80 ma of plate cur- 19 -6 Grid -Circuit Keying
rent.
Because of the series resistance of the keyer Grid- circuit, or blocked -grid keying is an-
tubes, the plate voltage at the keyed tube other effective method of keying a c -w trans-
will be from 30 to 60 volts less than the mitter. A basic blocked -grid keying circuit is
power supply voltage. This voltage appears shown in figure 11. The time constant of the
keying is determined by the RC circuit,
807,6146. ETC. which also forms part of the bias circuit of
the tube. When the key is closed, operating
bias is developed by the flow of grid current
LOW rowER BUFFER through R1. When the key is open, sufficient
(SAG? ETC.) fixed bias is applied to the tube to block it,
preventing the stage from functioning. If
an un- neutralized tetrode is keyed by this
method, there is the possibility of a consider-
able backwave caused by r -f leakage through
RFC the grid -plate capacitance of the tube.
2.5 MN
O KEY Certain high-ji. triode tubes, such as the
00T 1 811 -A and the 3-400Z, automatically block
themselves when the grid- return circuit is
6.3V. TO 6C4
opened. It is merely necessary to insert a key
fool and associated key -click filter in the grid -re-
turn lead of these tubes. No blocking bias
Figure 13 supply is needed. This circuit is shown in
TWO -STAGE BLOCKED -GRID KEYER
figure 12.
A more elaborate blocked-grid keying sys-
A separate filament transformer must be used tem using a 6C4 and VR -150 is shown in
for the tube, as its filament is at a potential
of -400 volts. figure 13. Two stages are keyed, preventing
HANDBOOK Grid -Circuit Keying 405
Figure 15
?N.V.
KEYER UNIT
ILOCA/NIfA/O VOLTASE OurPY/r 7O SCFEENOr
/
IOr
O POINT KEY UP KEY DOWN
VOLTS
I
0
TIME
-\
- VOLTS
TIME
A
B
C
-35
-110
-110
340
O
0
0 375 375
6K6
E -275 -275
KEY 21K
470 K
2W
470K,I W
ION, SOMA.
SY3
5
SAX! /K/ NTR.
2AU
6AX3-GT I
4.7K,2W
350-0-350
50 MA.
IISV.ti
Figure 16
www.americanradiohistory.com
HAN DBOOK Differential Keying 407
D .
o 6AL5 TO CATHODE
BLOCKING DIODES
O
V l
U
7
ti
D
rr
I-
3
CIRCUIT
KEYED STAGE
OF
. U
a
I70K
.
< r
O -250 V VACUUM-
TUBE
t CUTOFF VALVE KEYER
a. AMPLIFIER .oOl - ( F/G. lo )
O
1 KEY
CUTOFF
VALUE
OSC.
Figure 18
BLOCKING DIODES EMPLOYED TO VARY
TIME CONSTANT OF "MAKE" AND "BREAK"
j.--- KEY IS DEPRESSED -.
)
-PI
1
6AL5
ZERO
BEAT
+300 V.
sI 7
1
ttOK 50 K
2w
OUTPUT
CONTROL
RI
100
2w
Figure 19
V, V2 V3
OSCILLATOR BUFFER DRIVER
+300 V.
100 M 100 K
500 V4 0
I!
12A7 MEYER TUBE
6
KEY HI'
.005
R3 005
Figure 20
cutoff bias of about -110 volts is applied vfo between marks of keyed characters, thus
to the screen -grid circuits of the keyed allowing rapid break -in operation.
stages. When the key is closed, the screen -
grid voltage rises to the normal value at a 19 -9 VOX Circuitry
rate determined by the time constant (RC).
On opening the key again, the screen volt-
age returns to cutoff value at the predeter- A form of VOX ( voice- operated trans-
mined rate. mission) is often employed in SSB operation.
The potentiometer (R1) serves as an out- The VOX circuitry makes use of a trans-
put control, varying the minimum internal mitter control relay that is actuated by the
resistance of the 12BH7 keyer tube, and is a operator's voice and is held open by an anti-
vox circuit actuated by the audio system of
useful device to limit power input during
the station receiver. Voice -controlled break -
tuneup periods. Excitation to the final amp-
in operation is thus made possible without
lifier stage may be controlled by the screen
annoying feedback from the receiver speaker.
potentiometer (R,) in the second buffer
stage. An external bias source of approxi-
mately -120 volts at 10 ma is required for PROM SPEECH VIA VON
ro CONTROL
MPLIFIER
operation of the kever, in addition to the ,412AT7
}
.01
6AL5
V3A
300 -volt screen supply. 2 2,6 1_ 100 12AT7
,00 K VF
Blocking voltage may be removed from vol
GAIN
100K
1001
the oscillator for zeroing purposes by closing R
A representative VOX system is illustrated in The antivox signal voltage is derived from
figure 21. The VOX signal voltage is taken the speaker circuit of the receiver, adjusted
from the speech amplifier of the SSB trans- to the proper amplitude by the antivox -gain
mitter and adjusted to the proper amplitude potentiometer and rectified by diode V_B to
by means of VOX -gain potentiometer. The provide a negative voice impulse which
signal is rectified by diode V2A and the posi- biases the vox diode (V_A) to a noncon-
tive voice impulses are applied to the grid of ducting state. The relay is held in a cut -off
the VOX relay tube (V,A) which is normal- position until a positive override signal from
ly biased to cutoff. An RC network in the the VOX circuit defeats the antivox signal
VOX rectifier circuit permits rapid relay ac- taken from the station receiver. The relay
tion yet delays the opening of the relay so tube may also be actuated by the manual
that VOX action is sustained during syl- switch which drops the bias level, causing
lables and between words. Delay periods of the tube to draw a heavy plate current and
up to 0.5 second are common. trip the VOX relay.
CHAPTER TWENTY
Radiation, Propagation,
and Transmission Lines
Radio waves are electromagnetic waves occurs at the open end of a wire. Therefore,
similar in nature to, but much lower in fre- a dipole has a great mismatch at each end,
quency than, light waves or heat waves. Such producing a high degree of reflection. We
waves represent electric energy traveling say that the ends of a dipole are terminated
through space. Radio waves travel in free in an infinite impedance.
space with the velocity of light and can be A returning wave which has been reflected
reflected and refracted much the same as meets the next incident wave, and the volt-
light waves. age and current at any point along the an-
tenna are the vector sum of the two waves.
20 -1 Radiation from an At the ends of the dipole, the voltages add,
Antenna while the currents of the two waves cancel,
thus producing high voltage and low current
Alternating current passing through a con- at the ends of the dipole or half -wave section
ductor creates an alternating electromagnetic of wire. In the same manner, it is found that
field around that conductor. Energy is alter- the currents add while the voltages cancel
nately stored in the field, and then returned at the center of the dipole. Thus, at the
to the conductor. As the frequency is raised, center there is high current but low voltage.
more and more of the energy does not return Inspection of figure 1 will show that the
to the conductor, but instead is radiated off current in a dipole decreases sinusoidally
into space in the form of electromagnetic toward either end, while the voltage similar-
waves, called radio waves. Radiation from a ly increases. The voltages at the two ends of
wire, or wires, is materially increased when- the antenna are 180 out of phase, which
ever there is a sudden change in the electrical means that the polarities are opposite, one
constants of the line. These sudden changes being plus while the other is minus at any
produce reflection, which places standing instant. A curve representing either the volt-
waves on the line. age or current on a dipole represents a stand-
When a wire in space is fed radio -fre- ing wave on the wire.
quency energy having a wavelength of ap-
proximately 2.1 times the length of the wire Radiation From Radiation can and does take
in meters, the wire resonates as a half -wave Sources Other place from sources other
dipole antenna at that wavelength or fre- Than Antennas than antennas. Undesired ra-
quency. The greatest possible change in the diation can take place from
electrical constants of a line is that which open -wire transmission lines, both from sin-
www.americanradiohistory.com
Antenna Characteristics 411
`.. __ '
VOLTAGE electrostatic lines of force will be vertical.
cwlwf't'"
CE-+_NTER
Likewise, a simple horizontal antenna will
waeKr radiate horizontally polarized waves.
.' \. `. Because the orientation of a simple linear
I
`
l
.
radiator is the same as the polarization of
the waves emitted by it, the radiator itself
IULLI-WAVE ANTENNA
is referred to as being either vertically or
horizontally polarized. Thus, we say that a
SHOWING HOW STANDING WAVES
MST ON A HORIZONTAL ANTENNA. horizontal antenna is horizontally polarized.
VOLTAGE
CURRENT IS hIAXIIPLISI AT CENTER.
VOLTAGE IS MAJUWUNI AT DIOS. Figure 2A illustrates the fact that the
polarization of the electric field of the radi-
Figure 1 ation from a vertical dipole is vertical.
Figure 2B, on the other hand, shows that
STANDING WAVES ON A RESONANT the polarization of electric -field radiation
ANTENNA from a vertical slot radiator is horizontal.
This fact has been utilized in certain com-
gle -wire lines and from lines comprised of mercial f -m antennas where it is desired to
more than one wire. In addition, radiation have horizontally polarized radiation but
can be made to take place in a very efficient where it is more convenient to use an array
manner from electromagnetic horns, from of vertically stacked slot arrays. If the me-
plastic lenses or from electromagnetic lenses tallic sheet is bent into a cylinder with the
made up of spaced conducting planes, from slot on one side, substantially omnidirection-
slots cut in a piece of metal, from dielectric al horizontal coverage is obtained with hor-
wires, or from the open end of a waveguide. izontally polarized radiation when the cyl-
inder with the slot in one side is oriented
Directivity of The radiation from any phys- vertically. An arrangement of this type is
Radiation ically practical radiating sys- shown in figure 2C. Several such cylinders
tm is directive to a certain may be stacked vertically to reduce high -
degree. The degree of directivity can be en- angle radiation and to concentrate the radi-
hanced or altered when desirable through the ated energy at the useful low radiation
combination of radiating elements in a pre- angles.
scribed manner, through the use of reflec- In any event the polarization of radiation
ting planes or curved surfaces, or through from a radiating system is parallel to the
the use of such systems as mentioned in the electric field as it is set up inside or in the
preceding paragraph. The construction of vicinity of the radiating system.
directive antenna arrays is covered in detail
in the chapters which follow. 20 -2 General Character-
Polarization Like light waves, radio waves istics of Antennas
can have a definite polarization. All antennas have certain general char-
In fact, while light waves ordinarily have to acteristics to be enumerated. It is the result
be reflected or passed through a polarizing of differences in these general characteristics
medium before they have a definite polari- which makes one type of antenna system
zation, a radio wave leaving a simple radi- most suitable for one type of application
ator will have a definite polarization, the and another type best for a different appli-
polarization being indicated by the orienta- cation. Six of the more important charac-
tion of the electric -field component of the teristics are: (1) polarization, (2) radiation
wave. This, in turn, is determined by the resistance, (3) horizontal directivity, (4)
orientation of the radiator itself, as the vertical directivity, (f) bandwidth, and
magnetic -field component is always at right (6) effective power gain.
angles to a linear radiator, and the electric - The polarization of an antenna or radi-
field component is always in the same plane ating system is the direction of the electric
as the radiator. Thus we see that an antenna field and has been defined in Section 20 -1.
that is vertical with respect to the earth will The radiation resistance of an antenna
transmit a vertically polarized wave, as the system is normally referred to the feed point
412 Radiation, Propagation, and Lines THE RADIO
Figure 2
ANTENNA POLARIZATION
The polarization (electric field) of
the radiation from a resonant di-
pole such as shown at A is par-
allel to the length of the radia-
I ELECTRIC
FIELD
(POLARIZATION)
VERTICAL
tor. In the case of a resonant slot
cut in a sheet of metal and used as
a radiator, the polarization (of the
electric field) is perpendicular to the
2 O
length of the slot. In both cases,
however, the polarization of the
radiated field is parallel to the po-
tential gradient of the radiator; in ELECTRIC FIELD
the case of the dipole the electric (POLARIZATION)
lines of force are from end to end, HORIZONTAL
while in the case of the slot the
field is across the sides of the slot.
The metallic sheet containing the TO POINTS Al
FEEDERS CCOOKCT
Nil DE CYLINDER
slot may be formed into a cylinder
to make up the radiator shown at
C. With this type of radiator the
radiated field will be horizontally
polarized even though the radiator
Is mounted vertically.
www.americanradiohistory.com
HANDBOOK Antenna Characteristics 413
length. Figure 3
A radio wave in space can be compared to CHART SHOWING SHORTENING OF A RESO-
a wave in water. The wave, in either case,
NANT ELEMENT IN TERMS OF RATIO OF
has peaks and troughs. One peak and one LENGTH TO DIAMETER
trough constitute a full wave, or one wave-
length. The use of this chart is based on the basic
formula where radiator length in feet is
Frequency describes the number of wave equal to 468 /frequency in MHz. This formula
cycles or peaks passing a point per second. applies to fr ties below perhaps 30 MHz
when the radiator is made from wire. On
Wavelength describes the distance the wave higher frequencies, or on 14 and 28 MHz when
travels through space during one cycle or the radiator is made of large-diameter tubing,
the radiator is shortened from the value ob-
oscillation of the antenna current; it is the tained with the above formula by an amount
distance in meters between adjacent peaks or determined by the ratio of length to diameter
adjacent troughs of a wave train. of the radiator. The amount of this shortening
is obtainable from the chart shown above.
As a radio wave travels 300,000,000
meters a second (speed of light), a fre-
quency of 1 cycle per second (1 Hz) cor- Length of a half -wave radiator made from
responds to a wavelength of 300,000,000 wire (No. 14 to No. 10):
meters. So, if the frequency is multiplied by 3.5 -MHz to 30 -MHz bands
a million, the wavelength must be divided
by a million, in order to maintain their cor-
rect ratio.
Length in feet - 468
Freq. in MHz
A frequency of 1,000,000 cycles per sec-
ond (1000 kHz) equals a wavelength of
300 meters. Multiplying frequency by 10 50 -MHz band
and dividing wavelength by 10, we find: a
frequency of 10,000 kHz equals a wave- -
Length in feet
460
Freq. in MHz
length of 30 meters. Multiplying and divid-
ing by 10 again, we get: a frequency of
Length in inches -
5600
100,000 kHz equals 3 meters wavelength. Freq. in MHz
Therefore, to change wavelength to fre-
quency (in kilohertz), simply divide 300,-
000 by the wavelength in meters (X). 144 -MHz band
300,000 5500
Fkll> = X
Length in inches
Freq. in MHz
= 300,000 Length -to- Diameter a half -wave radi-
When
FkH. Ratio ator is constructed from
Now that we have a simple conversion tubing or rod whose
formula for converting wavelength to fre- diameter is an appreciable fraction of the
quency and vice versa, we can combine it length of the radiator, the resonant length
with our wavelength -versus-antenna length of a half-wave antenna will be shortened.
formula, and we have the following: The amount of shortening can be deter-
414 Radiation, Propagation, and Lines THE RADIO
mined with the aid of the chart of figure 3.
In this chart the amount of additional short-
- (K -.05) X 492
Freq. in MHz
ening over the values given in the previous
paragraph is plotted against the ratio of the where,
length to the diameter of the half -wave K equals number of %2 waves on antenna,
radiator. L equals length in feet.
The length of a wave in free space is Under conditions of severe current atten-
somewhat longer than the length of an an-
uation, it is possible for some of the nodes,
tenna for the same frequency. The actual or loops, actually to be slightly greater than
free -space half wavelength is given by the
a physical half wavelength apart. Practice
following expressions:
has shown that the most practical method of
Half wavelength - Freq. 492
in MHz
in feet
resonating a harmonically operated antenna
accurately is by cut and try, or by using a
feed system in which both the feedline and
Half wavelength -
5905
in inches antenna are resonated at the station end as
Freq. in MHz an integral system.
Harmonic A wire in space can resonate at A dipole or half -wave antenna is said to
Res more than one frequency. The operate on its fundamental or first harmonic.
lowest frequency at which it A full -wave antenna, 1 wavelength long,
resonates is called its fundamental frequency, operates on its second harmonic. An antenna
and at that frequency it is approximately a with five half wavelengths on it would be
half wavelength long. A wire can have two, operating on its fifth harmonic. Observe that
three, four, five, or more standing waves on the fifth harmonic antenna is 21/2 wave-
it, and thus it resonates at approximately lengths long, not 5 wavelengths.
the integral harmonics of its fundamental
Antenna Most types of antennas operate
frequency. However, the higher harmonics
Resonance most efficiently when tuned, or
are not exactly integral multiples of the
lowest resonant frequency as a result of end resonated, to the frequency of
effects. operation. This consideration of course does
A harmonic -operated antenna is somewhat
not apply to the rhombic antenna and to the
longer than the corresponding integral num- parasitic elements of Yagi arrays. However,
ber of dipoles, and for this reason, the dipole
length formula cannot be used simply by 000
multiplying by the corresponding harmonic. DI .4
Di -,
also as transmission lines. For the latter rea-
son, the resonant length will be dependent Figure 4
to an extent on the method of feed, as there
will be less attenuation of the current along EFFECT OF SERIES INDUCTANCE AND CA-
the antenna if it is fed at or near the center PACITANCE ON THE LENGTH OF A HALF -
than if fed toward or at one end. Thus, the WAVE RADIATOR
antenna would have to be somewhat longer The top antenna has been electrically length-
if fed near one end than if fed near the ened by placing a coil in series with the cen-
ter. In other words, an antenna with a lumped
center. The difference would be small, how- inductance in its center can be made shorter
ever, unless the antenna were many wave- for a given frequency than a plain wire radia-
tor. The bottom antenna has been capacitive-
lengths long. ly shortened electrically. In other words, an
The length of a center -fed harmonically antenna with a capacitor in series with it must
be made longer for a given frequency since its
operated doublet may be found from the effective electrical length as compared to plain
formula: wire is shorter.
www.americanradiohistory.com
HANDBOOK Radiation Resistance 415
10000 +6000
IJDIAMETER= T5650
6000 +5000
DIAMETER-
000 +4000
7000- +7000
r:I
_
h
6000
+2
_
5000. 2 +1
,/am
u
4000
DI
2
u
% DIAMETLR' -,
_
I/m I/
10
3000 1000
O:
1-
2000 z 2000
1000
0 3000
W
4
0
0.15A 05) LOA 1.51 2.0) 2.51
OVER -ALL LENGTH OF RADIATOR
5000
Figure 5
II\v7
I11
.1
\NV\2101
N';;ii'f^
15\U/!i1
11i
=11
11
1111NImrI
IM11PW111
1111
. .2
.
.4
.1 .4
THE RADIO
111
.5
Figure 7
.7
.a
.73
1A
rises uniformly toward each end, where it Several interesting things will be noted
is about 2000 ohms for a dipole remote from with respect to these curves. The reactive
ground, and about twice as high for a verti- component disappears when the over -all
cal Marconi. physical length is slightly less than any num-
If a vertical half -wave antenna is set up ber of half waves long, the differential in-
so that its lower end is at the ground level, creasing with conductor diameter. For over-
the effect of the ground reflection is to in- all lengths in the vicinity of an odd number
crease the radiation resistance to approxi- of half wavelengths, the center feed point
mately 100 ohms. When a horizontal half - looks like a series -resonant lumped circuit to
wave antenna is used, the radiation resist- the generator or transmission line, while
ance (and, of course, the amout of energy for over -all lengths in the vicinity of an
radiated for a given antenna current) de- cien number of half wavelengths, it looks
pends on the height of the antenna above like a parallel- resonant or antiresonant
www.americanradiohistory.com
HANDBOOK Antenna Impedance 417
lumped circuit. Both the feed -point resist- The radiation resistance of an antenna
ance and the feed -point reactance change generally increases with length, although
more slowly with over -all radiator length this increase varies up and down about a
(or with frequency with a fixed length) as constantly increasing average. The peaks and
the conductor diameter is increased, indi- dips are caused by the reactance of the an-
cating that the effective Q is lowered as the tenna, when its length does not allow it to
diameter is increased. However, in view of resonate at the operating frequency.
the fact that the damping resistance is
nearly all radiation resistance rather than Antenna Antennas have a certain loss re-
loss resistance, the lower Q does not repre- Efficiency sistance as well as a radiation re-
sent lower efficiency. Therefore, the lower Q sistance. The loss resistance defines
is desirable, because it permits use of the the power lost in the antenna due to ohm-
radiator over a wider frequency range with- ic resistance of the wire, ground resistance
out resorting to means for eliminating the (in the case of a Marconi), corona dis-
reactive component. Thus, the use of a charge, and insulator losses.
large diameter conductor makes the over -all The approximate effective radiation effi-
system less frequency sensitive. If the diam- ciency (expressed as a decimal) is equal to:
eter is made sufficiently large in terms of
wavelengths, the Q will be low enough to Ra
qualify the radiator as a broadband anten- N` R. + Rr.
na.
The curves of figure 7 indicate the theo- where,
retical center -point radiation resistance of a R;, equals the radiation resistance,
half -wave antenna for various heights above R1, equals loss resistance of antenna.
perfect ground. These values are of im- The loss resistance will be of the order of
portance in matching untuned radio -fre- 0.25 ohm for large- diameter tubing con-
auency feeders to the antenna, in order to ductors such as are most commonly used in
obtain a good impedance match and an ab- multi -element parasitic arrays, and will be of
sence of standing waves on the feeders. the order of 0.5 to 2.0 ohms for arrays of
normal construction using copper wire.
Ground Losses Above average ground, the When the radiation resistance of an an-
actual radiation resistance of tenna or array is very low, the current at a
a dipole will vary from the exact value of voltage node will be quite high for a given
figure 7 since the latter assumes a hypothet- power. Likewise, the voltage at a current
ical, perfect ground having no loss and per- node will be very high. Even with a heavy
fect reflection. Fortunately, the curves for conductor and excellent insulation, the losses
the radiation resistance over most types of due to the high voltage and current will be
earth will correspond rather closely with appreciable if the radiation resistance is suf-
those of the chart, except that the radiation ficiently low.
resistance for a horizontal dipole does not Usually, it is not considered desirable to
fall off as rapidly as is indicated for heights use an antenna or array with a radiation re-
below an eighth wavelength. However, with sistance of less than approximately 5 ohms
the antenna so close to the ground and the unless there is sufficient directivity, com-
soil in a strong field, much of the radiation pactness, or other advantage to offset the
resistance is actually represented by ground losses resulting from the low radiation re-
loss; this means that a good portion of the sistance.
antenna power is being dissipated in the
earth, which, unlike the hypothetical perfect Ground The radiation resistance of a
ground, has resistance. In this case, an ap- Resistance Marconi antenna, especially,
preciable portion of the radiation resistance should be kept as high as pos-
actually is loss resistance. The type of soil sible. This will reduce the antenna current
also has an effect upon the radiation pattern, for a given power, thus minimizing loss re-
especially in the vertical plane, as will be sulting from the series resistance offered by
seen later. the earth connection. The radiation resist-
418 Radiation, Propagation, and Lines THE RADIO
ance can be kept high by making the Mar- A half -wave resonant doublet, because of
coni radiator somewhat longer than a quar- its different current distribution and signifi-
ter wave, and shortening it by series ca- cant length, exhibits slightly more free -space
pacitance to an electrical quarter wave. This power gain as a result of directivity than
reduces the current flowing in the earth con- does the infinitesimal dipole, for reasons
nection. It also should be removed from which will be explained in a later section.
ground as much as possible (vertical being The space- directivity power gain of a half -
ideal). Methods of minimizing the resistance wave resonant doublet is 1.63 (or 2.15 db)
of the earth connection will be found in the referred to an isotropic radiator.
discussion of the Marconi antenna.
Horizontal When choosing and orienting an
20-4 Antenna Directivity Directivity antenna system, the radiation
All practical antennas radiate better in patterns of the various common
some directions than others. This character- types of antennas should be given careful
istic is called directivity. The more directive consideration. The directional characteristics
an antenna is, the more it concentrates the are of still greater importance when a di-
radiation in a certain direction, or direc- rective antenna array is used.
tions. The more the radiation is concen-
trated in a certain direction, the greater will
be the field strength produced in that direc-
tion for a given amount of total radiated
power. Thus the use of a directional an-
tenna or array produces the same result in
the favored direction as an increase in the
power of the transmitter.
The increase in radiated power in a cer-
tain direction with respect to an antenna in
free space as a result of inherent directivity
is called the free -space directivity power gain
II Er
or just space directivity gain of the antenna
(referred to a hypothetical isotropic radiator
II 51111111
which is assumed to radiate equally well in
all directions). Because the fictitious iso-
tropic radiator is a purely academic antenna,
`MEW/MIME M
not physically realizable, it is common prac-
tice to use as a reference antenna the sim-
plest ungrounded resonant radiator, the half - .s
wave dipole, or resonant doublet. As a half - .3
wave doublet has a space directivity gain of .1
www.americanradiohistory.com
HANDBOOK Antenna Directivity 419
Horizontal directivity is always desirable points. This means, of course, that more
on any frequency for point -to -point work. than one angle of radiation can be used. If
However, it is not always attainable with no elevation directivity is being used under
reasonable antenna dimensions on the lower th;s condition of propagation, selective fad-
frequencies. Further, when it is attainable, as ing will take place because of interference
on the frequencies above perhaps 7 MHz, between the waves arriving over the dif-
with reasonable antenna dimensions, oper- ferent paths.
ating convenience is greatly furthered if the On the 28 -MHz band it is by far the
maximum lobe of the horizontal directivity most common condition that only one mode
is controllable. It is for this reason that ro- of propagation will be possible between two
tatable antenna arrays have come into such points at any one time. This explains, of
common usage. course, the reason why rapid fading in gen-
Considerable horizontal directivity can be eral and selective fading in particular are
used to advantage when: (1) only point -to- almost absent from signals heard on the 28-
point work is necessary, (2) several arrays MHz band (except for fading caused by
are available so that directivity may be local effects).
changed by selecting or reversing antennas, Measurements have shown that the angles
(3) a single rotatable array is in use. Signals useful for communication on the 14 -MHz
follow the great- circle path, or within 2 or band are from 3 to about 30 , angles
3 degrees of that path under all normal above about 15 being useful only for local
propagation conditions. However, under work. On the 28 -MHz band, measurements
turbulent ionospheric conditions, or when have shown that the useful angles range
unusual propagation conditions exist, the from about 3 to 180; angles above about
deviation from the great- circle path for 12 being useful only for local (less than
greatest signal intensity may be as great as 3000 miles) work. These figures assume
90. Making the array rotatable overcomes normal propagation by virtue of the F_
these difficulties, but arrays having extremely layer.
high horizontal directivity become too cum-
bersome to be rotated, except perhaps when Angle of Radiation It now becomes of in-
desiened for operation on frequencies above of Typical Antennas terest to determine the
50 MHz. and Arrays amount of radiation
Vertical Vertical directivity is of the available at these useful
Directivity greatest importance in obtain- lower angles of radiation from commonly
used antennas and antenna arrays. Figure 8
ing satisfactory communication
shows relative output voltage plotted against
above 14 MHz whether or not horizontal
directivity is used. This is true simply be- elevation angle (wave angle) in degrees
above the horizontal, for horizontal and
cause only the energy radiated between
certain definite elevation angles is useful for vertical doublets elevated 0.6 wavelength
communication. Energy radiated at other above two types of ground. It is obvious by
elevation angles is lost and performs no use- inspection of the curves that a horizontal
ful function. dipole mounted at this height above ground
(20 feet on the 28 -MHz band) is radiating
Optimum Angle The optimum angle of ra- only a small amount of energy at angles use-
of Radiation diation for propagation of ful for communication on the 28 -MHz
signals between two points band. Most of the energy is being radiated
is dependent on a number of variables. uselessly upward. The vertical antenna above
Among these significant variables are: (1) a good reflecting surface appears much bet-
height of the ionosphere layer which is pro- ter in this respect -and this fact has been
viding the reflection, (2) distance between proven many times by actual installations.
the two stations, (3) number of hops for It might immediately be thought that the
propagation between the two stations. For amount of radiation from a horizontal or
communication on the 14 -MHz band it is vertical dipole could be increased by raising
often possible for different modes of propa- the antenna higher above the ground. This is
gation to provide signals between two true to an extent in the case of the hori-
420 Radiation, Propagation, and Lines THE RADIO
Figure 9
VERTICAL RADIATION
PATTERNS
Showing the vertical radiation
patterns for half -wave antennas
(or collinear half -wave or ex-
tended half -wave antennas) at
dill heights above average
ground and perfect ground. Note
that such antennas one- quarter
wave above ground concentrate
most of the radiation at the
very high angles which are use-
ful for communication only on
the lower- frequency bands. An-
tennas one -half wave above
ground are not shown, but the
POWER OUTPUT elevation pattern shows one
lobe on each side at an angle
of 30 above horizontal.
zontal dipole; the low -angle radiation does radiation to high -angle radiation it is neces-
increase slowly after a height of 0.6 wave- sary to place the antenna high above ground,
length is reached but at the expense of and in addition it is necessary to use addi-
greatly increased high -angle radiation and tional means for suppressing high -angle ra-
the formation of a number of nulls in the diation.
elevation pattern. No signal can be trans-
mitted or received at the elevation angles Suppression of High -angle radiation can be
where these nulls have been formed. Tests High -ongle suppressed, and this radiation
have shown that a center height of 0.6 Radiation can be added to that going
wavelength for a vertical dipole (0.35 wave- out at low angles, only
length to the bottom end) is about optimum through the use of some sort of directive
for this type of array. antenna system. There are three general
Figure 9 shows the effect of placing a types of antenna arrays composed of dipole
horizontal dipole at various heights above elements commonly used which concentrate
ground. It is easily seen by reference to fig- radiation at the lower more effective angles
ure 9 (and figure 10 which shows the radi- for high- frequency communication. These
ation from a dipole at 3/4 wave height) that types are: (1) The close -spaced out -of -phase
a large percentage of the total radiation system as exemplified by the "flat -top"
from the dipole is being radiated at relative- beam, or WBJK array. Such configurations
ly high angles which are useless for commu- are classified as end -fire arrays . (2) The wide -
nication on the 14 -MHz and 28 -MHz bands. spaced in -phase arrays, as exemplified by the
Thus we see that in order to obtain a worth- "Lazy H" antenna. These configurations are
while increase in the ratio of low -angle classified as broadside arrays. (3) The close-
Figure 10
VERTICAL RADIATION
PATTERNS
Showing vertical -plane radiation
patterns of a horizontal single-
section flat-top beam with one-
eighth wave spacing (solid
curves) and a horizontal half -
wave antenna (dashed curves)
when both are 0.S wavelength
(A) and 0.75 wavelength (D)
above ground.
GAIN IN FIELD STRENGTH
www.americanradiohistory.com
HANDBOOK Antenna Bandwidth 421
Figure 11
COMPARATIVE VERTICAL
RADIATION PATTERNS
Showing the vertical radiation
patterns of a horizontal single -
section flat -top beam (A), an ar-
ray of two stacked horizontal
in -phase half -wave elements -
half of a "Lazy H " -(S), and a
horizontal dipole (C). In each
case the top of the antenna sys-
tem is 0.75 wavelength above
ground, as shown to the left of
the curves.
1.0 1. 2.0 z.s
spaced parasitic systems, as exemplified by its bandwidth will be greater than that of a
the three -element rotary beam. similar array constructed of single wires.
A comparison between the radiation from The radiation resistance of antenna arrays
a dipole, a "flat -top beam" and a pair of of the types mentioned in the previous para-
dipoles stacked one above the other (half of graphs may be increased through the use of
a "lazy H "), in each case with the top of wider spacing between elements. With in-
the antenna at a height of 3/4 wavelength is creased radiation resistance in such arrays
shown in figure 11. The improvement in the the radiation efficiency increases since the
amplitude of low -angle radiation at the ex- ohmic losses within the conductors become a
pense of the useless high -angle radiation with smaller percentage of the radiation resist-
these simple arrays as contrasted to the ance, and the bandwidth is increased pro-
dipole is quite marked. portionately.
Figure 12 compares the patterns of a 3-
element beam and a dipole radiator at a 20 -6 Propagation of
height of 0.75 wavelength. It will be noticed Radio Waves
that although there is more energy in the The preceding sections have discussed the
lobe of the beam as compared to the dipole, manner in which an electromagnetic -wave
the axis of the beam is at the same angle
above the horizontal. Thus, although more
Nsw \
radiated energy is provided by the beam at
low angles, the average angle of radiation of A-DIPOLE
the beam is no lower than the average angle B-3- ELEMENT
PARASITIC ARRAY
of radiation of the dipole.
20 -5 Bandwidth
The bandwidth of an antenna or an an-
tenna array is a function primarily of the
radiation resistance and of the shape of the
conductors which make up the antenna
system. For arrays of essentially similar con- .5 1.0 1.5 2.0 2.S SA z.s 4.0
struction the bandwidth (or the deviation GAIN IN FIELD STRENGTH
in frequency which the system can handle Figure 12
without mismatch) is increased with in- VERTICAL RADIATION PATTERNS
creasing radiation resistance, and the band- Showing vertical radiation patterns of a hori-
width is increased with the use of con- zontal dipole (A) and a horizontal 3- element
parasitic array (B) at a height above ground
ductors of large diameter (smaller ratio of of 0.75 wavelength. Note that the axes of the
length to diameter) . This is to say that main radiation lobes are at the same angle
if an array of any type is constructed above the horizontal. Note also the suppres-
sion of high angle radiation by the parasitic
of large diameter tubing or spaced wires, array.
www.americanradiohistory.com
422 Radiation, Propagation, and Lines THE RADIO
www.americanradiohistory.com
HANDBOOK Ground -Wave Communication 423
two waves, which travel over different path refraction and diffraction of the signal
lengths (since one of the waves has been around the spherical earth cause a smaller
reflected from the ground) results in an reduction in field strengththan would oc-
interference pattern. The interference be- cur in the absence of such bending, so that
tween the two waves brings about a cyclic the average radio horizon is somewhat be-
variation in signal strength as the receiving yond the geometrical horizon. The equation
antenna is raised above the ground. This d = 1.4 V H is sometimes used for deter-
effect is illustrated in figure 14. From this mining the radio horizon.
figure it can be seen that best spacewave
reception of a vhf signal often will be ob- Tropospheric Propagation by signal bending
tained with the receiving antenna quite close Propagation in the lower atmosphere, called
to the ground. tropospheric propagation, can
The distance from an elevated point to result in the reception of signals over a
the geometrical horizon is given by the ap- much greater distance than would be the
proximate equation: d = 1.22-V H where case if the lower atmosphere were homo-
distance d is in miles and antenna height H geneous. In a homogeneous or well -mixed
is in feet. This equation must be applied
lower atmosphere, called a normal, or stand-
separately to the transmitting and receiving ard, atmosphere, there is a gradual and uni-
antennas and the results added. However, form decrease in index of refraction with
height. This effect is due to the combined
effects of a decrease in temperature, pressure,
TRANSMITTING and water -vapor content with height.
ANTENNA DI ,DIRECT WAVES
This gradual decrease in refractive index
with height causes waves radiated at very
low angles with respect to the horizontal to
be bent downward slightly in a curved path.
RECEIVING
The result of this effect is that such waves
GROUND - REFLECTED
WAVES
Da Da ANTENNA
AT DIFFERENT
will be propagated beyond the trite, or geo-
HEIGHTS metrical, horizon. In a so- called standard
atmosphere the effect of the curved path is
Figure 14 the same as though the radius of the earth
WAVE INTERFERENCE WITH HEIGHT
were increased by approximately one -third.
This condition extends the horizon by ap-
When the source of a horizontally -polarized proximately 30 percent for normal propa-
space -wave signal is above the horizon, the
received signal at a distant location will go gation, and the extended horizon is known
through a cyclic variation as the antenna as the radio -path horizon, mentioned before.
height is progressively raised. This is due to
the diff in total path length between Conditions Leading to When the tempera -
the direct wave and the ground -reflected Tropospheric
wave, and to the fact that this path length ture, pressure, or wa-
diff changes with antenna height. Stratification ter -vapor content of
When the path length diff is such that
the two waves arrive at the receiving anten-
the atmosphere does
na with a phase diff f 360 or some not change smoothly with rising altitude, the
multiple of 360 , the two waves will appear discontinuity or stratification will result in
to be in phase as far as the antenna is con -
d and maximum signal will be obtained. the reflection or refraction of incident vhf
On the other hand, when the antenna height sienils. Ordinarily this condition is more
is such that the path length difference for the prevalent at night and in the summer. In
two waves causes the waves to arrive with a
phase difference of an odd multiple of 180 certain areas, such as along the west coast
the two waves will substantially cancel, and a of North America, it is frequent enough to
null will be obtained at that antenna height.
The diff between D, and D. plus D, is be considered normal. Signal strength de-
the path-length difference. Note also that creases slowly with distance and, if the
there is an additional 180 phase shift in the favorable condition in the lower atmosphere
ground -reflected wave at the point where it is
reflected from the ground. It is this latter covers sufficient area, the range is limited
phase shift which causes the space -wave field only by the transmitter power, antenna gain,
intensity of a horizontally polarized wave to
be zero with the receiving antenna at ground receiver sensitivity, and signal -to -noise ratio.
level. There is no skip distance. Usually, transmis-
424 Radiation, Propagation, and Lines THE RADIO
sion due to this condition is accompanied by 15). However, such ducts usually are formed
slow fading, although fading can be violent only on an over -water path. The depth of
at a point where direct waves of about the the duct over the water's surface may be
same strength are also received. only 20 to 50 feet, or it may be 1000 feet
Bending in the troposphere, which refers deep or more. Ducts exhibit a low- frequency
to the region from the earth's surface up to cutoff characteristic similar to a waveguide.
about 10 kilometers, is more likely to occur The cutoff frequency is determined by depth
on days when there are stratus clouds than of the duct and by the strength of the dis-
on clear, cool days with a deep blue sky. continuity in refractive index at the upper
The temperature or humidity discontinuities surface of the duct. The lowest frequency
may be broken up by vertical convection that can be propagated by such a duct sel-
currents over land in the daytime but are dom goes below 50 MHz, and usually will
more likely to continue during the day over not be greater than 450 MHz even along
water. This condition is in some degree pre- the Pacific Coast.
dictable from weather information several
days in advance. It does not depend on the Stratospheric Communication by virtue of
sunspot cycle. Like direct communication, Reflection stratospheric reflection can be
best results require similar antenna polariza- brought about during magnetic
tion or orientation at both the transmitting storms, aurora borealis displays, and during
and receiving ends, whereas in transmission meteor showers. DX communication during
via reflection in the ionosphere (that part extensive meteor showers is characterized by
of the atmosphere between about 50 and frequent bursts of great signal strength fol-
500 kilometers high) it makes little differ- lowed by a rapid decline in strength of the
ence whether antennas are similarly polar- received signal. The motion of the meteor
ized. forms an ionized trail of considerable extent
Duct Formation When bending conditions are which can bring about effective reflection of
particularly favorable they signals. However, the ionized region persists
may give rise to the formation of a duct only for a matter of seconds so that a shower
which can propagate waves with very little of meteors is necessary before communica-
attenuation over great distances in a manner tion becomes possible.
similar to the propagation of waves through The type of communication which is pos-
a waveguide. Guided propagation through
sible during visible displays of the aurora
a duct in the atmosphere can give quite re-
borealis and during magnetic storms has
markable transmission conditions (figure been called aurora-type DX. These condi-
tions reach a maximum somewhat after the
sunspot cycle peak, possibly because the spots
on the sun are nearer to its equator (and
more directly in line with the earth) in the
latter part of the cycle. Ionospheric storms
generally accompany magnetic storms. The
INVERS-ION
I
www.americanradiohistory.com
HANDBOOK Ionospheric Propagation 425
being contacted. Distances out to 700 or the earth. Thus we see that high- frequency
800 miles have been covered during magnetic radio waves may travel over short distances
storms, using 30- and S0 -MHz transmitters, in a direct line from the transmitter to the
with little evidence of any silent zone be- receiver, or they can be radiated upward into
tween the stations communicating with each the ionosphere to be bent downward in an
other. Generally, voice- modulated transmis- indirect ray, returning to earth at consider-
sions are difficult or impossible due to the able distance from the transmitter. The
tone or noise modulation on the signal. Most wave reaching a receiver via the ionosphere
of the communication of this type has route is termed a sky wave. The wave reach-
taken place by c.w. or by tone -modulated ing a receiver by traveling in a direct line
waves with a keyed carrier. from the transmitting antenna to the re-
ceiving antenna is commonly called a ground
20 -8 Ionospheric wave.
The amount of bending at the ionosphere
Propagation which the sky wave can undergo depends
on its frequency, and the amount of ioniza-
Propagation of radio waves for communi- tion in the ionosphere, which is in turn de-
cation on frequencies between perhaps 3 and pendent on radiation from the sun. The sun
30 MHz is normally carried out by virtue of increases the density of the ionosphere layers
ionospheric reflection or refraction. Under (figure 16) and lowers their effective height.
conditions of abnormally high ionization in For this reason, the ionosphere acts very dif-
the ionosphere, communication has been ferently at different times of day, and at
known to have taken place by ionospheric different times of the year.
reflection on frequencies higher than SO The higher the frequency of a radio wave,
MHz. the farther it penetrates the ionosphere, and
The ionosphere consists of layers of ion- the less it tends to be bent back toward the
ized gas located above the stratosphere, and earth. The lower the frequency, the more
extending up to possibly 300 miles above easily the waves are bent, and the less they
penetrate the ionosphere. 160 -meter and 80-
200 meter signals will usually be bent back to
F2 earth even when sent straight up, and may
150 be considered as being reflected rather than
F1
MIDDAY refracted. As the frequency is raised beyond
100 about 5000 kHz (dependent on the critical
E frequency of the ionosphere at the moment),
y1 50
J it is found that waves transmitted at angles
D
higher than a certain critical angle never
_ zooo return to earth. Thus, on the higher fre-
F2
_ quencies, it is necessary to confine radiation
V 150
W to low angles, since the high -angle waves
S
MIDNIGHT simply penetrate the ionosphere and are lost.
100
I SUMMER
-
an almost constant height of about 70 miles. z
SUNSPOT
MINIMUM
14
Since the recombination time of the ions at W
D 12
this height is rather short, the E layer dis- a
W IN. ....._+"...-
appears almost completely a short time after cr
local sunset.
The D Layer Below the E layer at a height 2
www.americanradiohistory.com
HANDBOOK 1 1 -Year Sunspot Cycle 427
ticularly the outer portion of the skip zone. reached by the transmitter. After a period
Distortion is much less severe than in the of from 10 to 40 seconds, recombination
case of short scatter, and while the signal is and diffusion have progressed to the point
likewise weak, it sometimes can be utilized where the effect of a single fairly large
for satisfactory communication. meteor is not perceptible. However, there
During a severe ionosphere disturbance in are many small meteors impinging on earth's
the north auroral zone, it sometimes is pos- atmosphere every minute, and the aggregate
sible to maintain communication between effect of their transient ionized trails, in-
the Eastern United States and Northern cluding the small amount of residual ioni-
Europe by the following mechanism: That zation that exists for several minutes after
portion of the energy which is radiated in the original flash but is too weak and dis-
the direction of the great circle path is com- persed to prolong a "burst," is believed to
pletely absorbed on reaching the auroral contribute to the existence of the nighttime -
zone. However, the portion of the wave E layer, and perhaps also to sporadic-E
leaving the United States in a southeasterly patches.
direction is refracted downward from the While there are many of these very small
F.: layer and encounters scattering patches meteors striking the earth's atmosphere every
or media on its downward trip at a distance minute, meteors of normal size (sufficiently
of approximately 2000 miles from the trans- large to produce individual "bursts ") do
mitter. There it is reflected by "long scatter" not strike nearly so frequently except during
in all directions, this scattering region acting some of the comparatively rare meteor
like an isotropic radiator fed with a very "showers." During one of these displays a
small fraction of the original transmitter "quivering" ionized layer is produced which
power. The great circle path from this is intense enough to return signals in the
southerly point to northern Europe does not lower vhf range with good strength, but
encounter unfavorable ionosphere conditions, with a type of "flutter" distortion which is
and the wave is propagated the rest of the characteristic of this type of propagation.
trip as though it had been radiated from the
scattering region. 20 -9 Transmission Lines
Another type of scatter is produced when For many reasons it is desirable to place
a sky wave strikes certain areas of the earth. an antenna or radiating system as high and
On striking a comparatively smooth surface in the clear as is physically possible, utilizing
such as the sea, there is little scattering, the some form of nonradiating transmission line
wave being shot up again by what could be to carry energy with as little loss as possible
considered specular, or mirror, reflection. from the transmitter to the radiating anten-
But on striking a mountain range, for in- na, and conversely from the antenna to the
stance, the reradiation or reflected energy is receiver.
scattered, some of it being directed back There are many different types of trans-
toward the transmitter, thus providing an- mission lines and, generally speaking, practi-
other mechanism for producing a signal cally any type of transmission line or feeder
within the skip zone. system may be used with any type of anten-
na. However, mechanical or electrical con-
Meteors end When a meteor strikes the siderations often make one type of trans-
"Bursts" earth's atmosphere, a cylindrical mission line better adapted for use to feed a
region of free electrons is particular type of antenna than any other
formed at approximately the height of the E type.
layer. This slender ionized column is quite Transmission lines for carrying r -f energy
long, and when first formed is sufficiently are of two general types: nonresonant and
dense to reflect radio waves back to earth resonant. A nonresonant transmission line
most readily, including vhf waves which is one on which a successful effort has been
are not ordinarily returned by the F_ layer. made to eliminate reflections from the ter-
The effect of a single meteor, or normal mination (the antenna in the transmitting
size, shows up as a sudden "burst" of signal case and the receiver for a receiving anten-
of short duration at points not ordinarily na) and hence one on which standing waves
430 Radiation, Propagation, and Lines THE RADIO
do not exist or are relatively small in mag-
nitude. A resonant line, on the other hand,
is a transmission line on which standing
waves of appreciable magnitude do appear,
either through inability to match the char-
acteristic impedance of the line to the termi-
/
/.dIorioi/ ,oraup..
w%
di
nation or through intentional design.
The principal types of transmission line in
_
o
:%=:i%%%11:ililll
use or available at this time include the %I:i1%E'/_E2/111111111
open -wire line (two -wire and four -wire
types), two -wire solid -dielectric line (twin -
I:iMiM- = 111111111
lead and similar ribbon or tubular types),
two -wire polyethylene -filled shielded line,
., , I i 3 f
INCHES, CENTER TO CENTER
7 q a is
www.americanradiohistory.com
HANDBOOK Transmission Lines 431
Open wire line, No. 12 0.15 0 3 0 8 0.96- - Based on 4" spacing below 50 MHz; 2" spacing above
copper. 0.99 50 MHz. Radiation losses included. Clean, low -loss ceram-
ic insulation assumed. Radiation high above 150 MHz.
Ribbon line, rec. type, For clean dry line, wet weather performance rather
300 ohms. poor, best line is slightly convex. Avoid line that has
(7, 28 conductors) 0.86 2.2 5.3 0.82' 6' concave dielectric. Suitable for low -power transmitting
applications. Losses increase as line weathers. Handles
4lxl watts at 30 MHz if SWR is low.
Tubular "twin -lead" Characteristics similar to receiving -type ribbon Iwe
rec. type, 300 ohms, - - -- - - except for much better wet -weather performance.
5. 16" O.D., (Amphenol
type 14 -271)
Ribbon line, trans. Characteristics vary somewhat with manufacturer, but
type, 300 ohms. - - -- - - approximate those of receiving -type ribbon except for
greater power -handling capability and slightly better
wer weather performance.
Tubular "twin- lead" For use where receiving -type tubular "twin -lead" does
trans. type, 7,16 O.D. 0.85 2.3 5.4 0.79 6.1 not have sufficient power -handling capability. Will
l.smphenol 14 -076) handle 1 kw at 30 MHz if SWR is low.
Ribbon line, receive 1.1 2.7 6.0 0.77' 10' Useful for quarter -wave matching sections. No longer
type, 150 ohms. widely used as a line.
Ribbon line, receive Useful mainly in the hf range because of excessive
type, 75 ohms. 2.0 5.0 11 0.68' 19' losses at vnt and uhf. Less atfectea by weather than
300 -ohm ribbon.
Ribbon line, trans. Very satisfactory for transmitting applications below
type, 75 ohms. 1.5 3.9 8.0 0.71 18 ;sI MHz at powers up to 1 kw. Not significantly af-
fected by wer weather.
RG -8 U coax (52 ohms) 1.0 2.1 4.2 0.66 29.5 Will handle 2 kw at 30 MHz if SWR is low. 0.4" O.D.
7, 21 conouctor.
RG -11. U coax (75 ohms) 0.94 1.9 3.8 0.66 20.5 Will handle 1.4 kw at 30 MHz if SWR is low. 0.4"
0.D. 7 26 conductor.
RG -17 U coax (52 ohms) 0.38 0.85 1.8 0.66 29.5 Will handle 7.8 kw at 30 MHz if SWR is low. 0.87"
0.D. 0.19" dia. conductor.
RG -58 U coax (53 ohms) 1.95 4.1 8.0 0.66 28.5 Will handle 430 watts at 30 MHz if SWR is low. 0.2"
U.D. No. 20 conductor.
RG -59 U coax (73 ohms) 1.9 3.8 7.0 0.66 21 Will handle 680 watts at 30 MHz if SWR is low.
0.24" U.D. No. 22 conductor.
TV -59 coax (72 ohms) 2.0 4.0 7.0 0.66 22 "Commercial" version of RG -59 U for less exacting ap-
plications. Less expensive.
RG 22. U shielded 1.7 3.0 5.5 0.66 16 For shielded, balanced -to- ground applications. Very
pair (95 ohms) low noise pickup. 0.4" O.D.
K -111 shielded pair Designed for TV lead -in in noisy locations. Losses higher
(300 ohms) 2.0 3.5 6.1 - 4 than regular 300 -ohm ribbon, but do not increase as
much from weathering.
FIGURE 21
Older type coaxial lines have a useful life of three to six years after which the cable at-
tenuation gradually rises, especially under conditions of heat. Newer cables (designated by
the suffix A: RG -8A /U for example) have useful life up to twelve years or so. The 52 -ohm
series cables have been recently replaced with 50 -ohm cables, RG -8A /U now being designated
RG- 213/U. Long-life versions of the RG -58 family are: RG -58B /U (53.5 -ohm) and RG-58C/U
(50 -ohm).
where,
Since expresses ratio only, the units
S is the exact distance between wire cen- d a
Ill/
this condition is automatically satisfied. If
types of power levels up to one kilowatt in connectors are not used, it is necessary that
IIII
the hf range, are listed with their pertinent some type of moisture -proof sealing corn-
characteristics, in the table of figure 21.
111
Coaxial Line Several types of coaxial cable 204
have come into wide use for ZO 134 LOGp
feeding power to an antenna system. A
g111/I
/.III
170
iiPii`liz
COAXIAL OR
cross -sectional view of a coaxial cable (some- 11111i=ig111IIM111 CONCENTRIC LINC
tIII
%III
times called concentric cable or line) is
shown in figure 22.
1311
MiMiDi1=
=111111111111MI1
As in the parallel -wire line, the power lost 100
in a properly terminated coaxial line is the M1111M1III=1111M1
MIIIIII01111==
lllIMIII
sum of the effective resistance losses along 70
DI.1 NSIDC DIAMETER Of
OUTER CONDUCTOR
the length of the cable and the dielectric 32 IIIiF/I10011I1
MIdl1111I= D .o1.1r310E DIANETEA Or
losses between the two conductors.
Of the two losses, the effective resistance
loss is the greater; since it is largely due to
10
IMlEIII
.M1g111
rNII111=IMINI
2.311.121 5 7 10 15 30
INNER CONOUCTOR
www.americanradiohistory.com
HANDBOOK Tuned Lines 433
pound be applied to the end of the cable even when used in an untuned system.
where it will be exposed to the weather. Strictly speaking, a line is untuned, or non -
Nearby metallic objects cause no loss, and resonant, only when it is perfectly flat, with
coaxial cable may be run up air ducts or ele- a standing -wave ratio of 1 (no standing
vator shafts, inside walls, or through metal waves). However, some mismatch can be
conduit. Insulation troubles can be forgot- tolerated with open -wire untuned lines, so
ten. The coaxial cable may be buried in the long as the reactance is not objectionable,
ground or suspended above ground. or is eliminated by cutting the line to ap-
Standing Waves proximately resonant length.
Standing waves on a trans-
mission line always are the
result of the reflection of energy. The only 20 -11 Tuned or
significant reflection which takes place in Resonant Lines
a normal installation is that at the load end
of the line. But reflection can take place If a transmission line is terminated in its
from discontinuities in the line, such as characteristic surge impedance, there will be
caused by insulators, bends, or metallic ob- no reflection at the end of the line, and the
jects adjacent to an unshielded line. current and voltage distribution will be uni-
When a uniform transmission line is ter- form along the line. If the end of the line is
minated in an impedance equal to its surge either open- circuited or short -circuited, the
impedance, reflection of energy does not oc- reflection at the end of the line will be 100
cur, and no standing waves are present. percent, and standing waves of very great
Thus, for proper operation of an untuned amplitude will appear on the line. There will
line (with standing waves eliminated), some still be practically no radiation from the line
form of impedance- matching arrangement if it is closely spaced, but voltage nodes will
must be used between the transmission line be found every half wavelength, the voltage
and the antenna, so that the radiation resist- loops corresponding to current nodes (fig-
ance of the antenna is reflected back into the ure 23).
line as a nonreactive impedance equal to the If the line is terminated in some value of
line impedance. resistance other than the characteristic surge
The termination at the antenna end is the impedance, there will be some reflection, the
only critical characteristic about the untuned amount being determined by the amount of
line fed by a transmitter. It is the reflection mismatch. With reflection, there will be
from the antenna end which starts waves standing waves (excursions of current and
moving back toward the transmitter end. voltage) along the line, though not to the
When waves moving in both directions along same extent as with an open -circuited or
a conductor meet, standing waves are set up. short -circuited line. The current and voltage
loops will occur at the same points along the
Semiresonant A well -constructed open - line as with the open- or short -circuited line,
Parallel -Wire Lines wire line has acceptably and as the terminating impedance is made to
low losses when its length approach the characteristic impedance of the
than about two wavelengths even when
is less line, the current and voltage along the line
the voltage standing -wave ratio is as high as will become more uniform. The foregoing
10 to 1. A transmission line constructed of assumes, of course, a purely resistive (non -
ribbon or tubular line, however, should have reactive) load. If the load is reactive, stand-
the standing -wave ratio kept down to not ing waves also will be formed. But with a
more than about 3 to 1 both to reduce reactive load the nodes will occur at different
power loss and because the energy dissipation locations from the node locations encount-
on the line will be localized, causing over- ered with improper resistive termination.
heating of the line at the points of maximum A well built 500- to 600 -ohm transmission
current. line may be used as a resonant feeder for
Because moderate standing waves can be lengths up to several hundred feet with very
tolerated on open -wire lines without much low loss, so long as the amplitude of the
loss, a standing -wave ratio of 2/1 or 3/1 is standing waves (ratio of maximum to mini-
considered acceptable with this type of line, mum voltage along the line) is not too great.
434 Radiation, Propagation, and Lines THE RADIO
www.americanradiohistory.com
HANDBOOK Broadband Balun 435
Antennas for the lower- frequency portion it is sometimes justifiable to bring part of
of the high- frequency spectrum (from 1.8 the radiation system directly to the trans-
to 7.0 MHz), and temporary or limited -use mitter, feeding the antenna without benefit
antennas for the upper portion of the high - of a transmission line. This is permissible
frequency range, usually are of a relatively when (1) there is insufficient room to erect
simple type in which directivity is not a a 75- or 80 -meter horizontal dipole and feed
prime consideration. Also, it often is desir- line, (2) when a long wire is also to be
able, in amateur work, that a single anten- operated on one of the higher- frequency
na system be capable of operation at least bands on a harmonic. In either case, it is
on the 3.5- and 7.0 -MHz ranges, and pre- usually possible to get the main portion of
ferably on other frequency ranges. Con- the antenna in the clear because of its
sequently, the first portion of this chapter length. This means that the power lost by
will be devoted to a discussion of such an- bringing the antenna directly to the trans-
tenna systems. The latter portion of the mitter is relatively small.
chapter is devoted to the general problem Even so, it is not best practice to bring
of matching the antenna transmission line the high- voltage end of an antenna into the
to antenna systems of the fixed type. Match- operating room because of the increased
ing the antenna transmission line to the difficulty in eliminating BCI and TVI. For
rotatable directive array is discussed in this reason one should dispense with a feed
Chapter Twenty-four. line in conjunction with a Hertz antenna
only as a last resort.
21 -1 End -Fed Half-Wave
Horizontal Antennas End -Fed The end -fed antenna has no form
Antennas of transmission line to couple it
The half -wave horizontal dipole is the to the transmitter, but brings the
most common and the most practical an- radiating portion of the antenna right down
tenna for the 3.5- and 7 -MHz amateur to the transmitter, where some form of
bands. The form of the dipole, and the coupling system is used to transfer energy
manner in which it is fed are capable of a to the antenna.
large number of variations. Figure 2 shows Figure 1 shows two common methods of
a number of practical forms of the simple feeding the Fuchs antenna,or end-fed Hertz.
dipole antenna along with methods of feed. Some harmonic- attenuating provision (in
Usually a high- frequency doublet is addition to the usual low -pass TVI filter)
mounted as high and as much in the clear must be included in the coupling system,
as possible, for obvious reasons. However, since an end -fed antenna itself offers no dis-
www.americanradiohistory.com
Center -Fed Antennas 437
300-600
II LINE
TUNED DOUBLET
i
LINE -FED 0.77 OF
with f in MHz. For the
folded dipole, length is com-
600 D. LINE 300 A TWINLEAD
puted by: length (feet) =
ANY LENGTH 462/f, with f in MHz. Above
30 MHz, the length of the
dipole is affected to an im-
portant degree by the diam-
eter of the element and the
method of supporting the
1.-0.94 A/2-.1 0.64 )./2 -0.95 1/2 -.{ dipole (see VHF and UHF
Antennas and Radiation, Prop-
2-OR 60 FOR DS LTA agation, and Lines chap-
O
300/1TWINLEAD
LOW
IN
OPENED O FEEDER
SPREADERS
DIMENSIONS
SEE CHAP 19 ters).
TWINLEAD 2 -WIRE DOUBLET DELTA - MATCHED
"FOLDED DIPOLE' OR `FOLDED DIPOLE DOUBLET
I
D =14% OF
O O O TOTAL LENGTH
www.americanradiohistory.com
HANDBOOK Multiwire Doublets 439
tuning capacitor that can optionally be as the transformer between the center of
placed either in series with the antenna coil the dipole and a piece of 300 -ohm twin -
or in parallel with it. A series -tuning capaci- lead. In any case the impedance of the
tor can be placed in series with one feeder quarter -wave transformer will be of the
leg without unbalancing the system. order of U0 to 200 ohms. The use of sec-
The tuned -doublet antenna is shown in tions of transmission line as linear trans-
figure 2D. The antenna is a current -fed formers is discussed in detail in Section 21 -8.
system when the radiating wire is a half
wave long electrically, or when the system Multiwire An alternative method for in-
is operated on its odd harmonics, but be- Doublets creasing the feed -point impedance
comes a voltage -fed radiator when operated of a dipole so that a medium -
on its even harmonics. impedance transmission line may be used is
The antenna has a different radiation shown in figures 2J and 2K. This system
pattern when operated on its harmonics, as utilizes more than one wire in parallel for
would be expected. The arrangement used the radiating element, but only one of the
on the second harmonic is better known as wires is broken for attachment of the feeder.
the Franklin collinear array and is described The most common arrangement uses two
in Chapter Twenty -two. The pattern is wires in the flat top of the antenna so that
similar to a half -wave dipole except that it an impedance multiplication of four is ob-
is sharper in the broadside direction. On tained.
higher harmonics of operation there will be The antenna shown in figure 2J is the
multiple lobes of radiation from the system. so- called twin -lead folded dipole which is a
Figures 2E and 2F show alternative ar- commonly used antenna system on the
rangements for using an untuned transmis- medium- frequency amateur bands. In this
sion line between the transmitter and the arrangement both the antenna and the
tuned -doublet radiator. In figure 2E a half - transmission line to the transmitter are con-
wave shorted line is used to resonate the structed of 300 -ohm twin -lead. The flat
radiating system, while in figure 2F a quar- tep of the antenna is made slightly less than
ter -wave open line is utilized. The adjust- the conventional length (462/Fml1, instead
ment of quarter -wave and half -wave stubs of 468 for a single -wire flat top) and
is discussed in Section 21 -8.
the two ends of the twin -lead are joined
together at each end. The center of one of
Doublets with The average value of feed im- the conductors of the twin -lead flat top is
Quarter-Wave pedance for a center -fed half - broken and the two ends of the twin -lead
Transformers wave doublet is 75 ohms. The feeder are spliced into the flat -top leads. As
actual value varies with a protection against moisture, pieces of flat
height and is shown in Chapter Twenty. polyethylene taken from another piece of
Other methods of matching this rather low 300 -ohm twin -lead may be molded over the
value of impedance to a medium -impedance joint between conductors with the aid of a
transmission line are shown in G, H, and I soldering iron.
of figure 2. Each of these three systems uses Better bandwidth characteristics can be
a quarter -wave transformer to accomplish obtained with a folded dipole made of rib-
the impedance transformation. The only bon line if the two conductors of the ribbon
difference between the three systems lies in line are shorted a distance of 0.82 (the
the type of transmission line used in the velocity factor of ribbon line) of a free -
quarter -wave transformer. G shows the space quarter- wave -length from the center
Q -match system whereby a line made up or feed point. This procedure is illustrated
of %2 -inch dural tubing is used for the low - in figure 3A. An alternative arrangement
impedance linear transformer. A line made for a twin -lead folded dipole is illustrated in
up in this manner is frequently called a set figure 3B. This type of half -wave antenna
of Q bars. Illustration H shows the use of system is convenient for use on the 3.5-
a four -wire line as the linear transformer, MHz band when the 116- to 132 -foot dis-
and I shows the use of a piece of 150 -ohm tance required for a full half -wave is not
twin -lead electrically 1/4-wave in length quite available in a straight line, since the
440 Antennas and Antenna Matching THE RADIO
462 frequent use of the multi -wire radiator as
FMN:
the driven element in a parasitic antenna
array.
Delta- Matched These two types of radiat-
300-OHM 6I1160N Doublet and ing elements are shown in
Standard Doublet figure 2L and figure 2M.
The delta-matched doublet
is described in detail in Section eight of this
chapter. The standard doublet, shown in
404
FMN1
figure 2M, is fed in the center by means of
75 -ohm twin -lead, either the transmitting
or the receiving type, or it may be fed by
means of twisted -pair feeder or by means of
parallel -wire lamp cord. Any of these types
of feed line will give an approximate match
to the center impedance of the dipole, but
the 75 -ohm twin -lead is far to be preferred
over the other types of low- impedance feed-
Figure 3 er due to the much lower losses of the poly-
FOLDED DIPOLE WITH SHORTING ethylene- dielectric transmission line.
STRAPS The coaxial -fed doublet shown in figure
The impedance match and bandwidth char-
2N is a variation on the system shown in
acteristics of a folded dipole may be im- figure 2M. Either 52 -ohm or 75 -ohm coaxial
proved by shorting the two wires of the cable may be used to feed the center of the
ribbon a distance out from the center equal
to the velocity factor of the ribbon times the dipole, although the 52 -ohm type will give
half -length of the dipole as shown at A. An a somewhat better impedance match at
alternative arrangement with bent down ends lower antenna heights. Due to the asym-
for space con atlon Is illustrated at B.
metry of the coaxial feed system, difficulty
may be encountered with waves traveling
single -wire end pieces may be bent away or on the outside of the coaxial cable. For this
downward from the direction of the main reason the use of twin -lead is normally to
section of the antenna. be preferred over the use of coaxial cable
Figure 2K shows the basic type of two - for feeding the center of a half -wave dipole.
wire doublet or folded dipole wherein the
radiating section of the system is made up Off-Center- The system shown in figure
of standard antenna wire spaced by means Fed Doublet 20 is sometimes used to feed
of feeder spreaders. The feeder again is made a half -wave dipole, especially
of 300 -ohm twin -lead since the feed -point
impedance is approximately 300 ohms, the
same as that of the twin -lead folded dipole.
The folded -dipole type of antenna has the
broadest response characteristics (greatest 0.04./r 0.04 A/ O.MA/1
www.americanradiohistory.com
HANDBOOK Vertical Antennas 441
LOADING COIL
AROXI 34 TURNS
I2
AND I
Y
WIRE. 4.5' DIAMETER
FOOT LONG
RADIALS EACH
MHz)
4 RADIALS EACH
MHi)
LOADING
COILS
Figure 8
LOADING THE
MARCONI ANTENNA
The various loading systems
are di d ln the accom-
panying text. 11
gives the least loading but is the most in the form of heat. Improving the ground
practical mechanically. The system shown connection, therefore, provides a definite
at figure 8E gives an intermediate amount means of reducing this power loss, and thus
of loading, while that shown at figure 8F, increasing the radiated power.
utilizing a "hat" just above the loading The best possible ground consists of as
coil, gives the greatest amount of loading. many wires as possible, each at least a quar-
The object of all the top -loading methods ter wave long, buried just below the surface
shown is to produce an increase in the ef- of the earth, and extending out from a com-
fective length of the radiator, and thus to mon point in the form of radials. Copper
raise the point of maximum current in the wire of any size larger than No. 16 is satis-
radiator as far as possible above ground. factory, and the larger sizes will take
Raising the maximum- current point in the longer to disintegrate. In fact, the radials
radiator above ground has two desirable need not even be buried; they may be sup-
results: The percentage of low -angle radia- ported just above the earth, and insulated
tion is increased and the amount of ground from it. This arrangement is called a
current at the base of the radiator is re- counterpoise, and operates by virtue of its
duced, thus reducing the ground losses. high capacitance to ground.
Amateurs primarily interested in the If the antenna is physically shorter than
higher- frequency bands, but liking to a quarter wavelength, the antenna current
work 80 meters occasionally, can usually is higher, due to lower radiation resistance;
manage to resonate one of their antennas consequently, the power lost in resistive
as a Marconi by working the whole system soil is greater. The importance of a good
(feeders and all) against a water pipe ground with short, inductive- loaded Mar-
ground, and resorting to a loading coil if coni radiators is, therefore, quite obvious.
necessary. A high- frequency rotary, zepp. With a good ground system, even very short
doublet, or single- wire -fed antenna will (one -eighth wavelength) antennas can be
make quite a good 80 -meter Marconi if expected to give a high percentage of the
high and in the clear, with a rather long efficiency of a quarter -wave antenna used
feed line to act as a radiator on 80 meters. with the same ground system. This is
Where two -wire feeders are used, the feed- especially true when the short radiator is
ers should be tied together for Marconi top loaded with a high -Q (low-loss) coil.
operation.
Water -Pipe Water pipe, because of its com-
Importance of With a quarter -wave an- Grounds paratively large surface and
Ground Connection tenna and a ground, the cross section, has a relatively
antenna current general- low r -f resistance. If it is possible to attach
ly is measured with a meter placed in the to a junction of several water pipes (where
antenna circuit close to the ground con- they branch in several directions and run
nection. If this current flows through a for some distance under ground), a satis-
resistor, or if the ground itself presents factory ground connection will be obtained.
some resistance, there will be a power loss If one of the pipes attaches to a lawn or
444 Antennas and Antenna Matching THE RADIO
-
446 Antennas and Antenna Matching
110'
higher than the radiation resistance, caus-
ing a loss of 50 percent or more of the
0111:-
priArryi
transmitter power output. If the radiation
resistance of the Marconi antenna is raised,
FOR DETAIL SEE FIG. A THE TWO WIRES MAY SE
SPREAD EITHER HORIZ- the amount of power lost in the ground
BLOCK 2'XI.SX0.3 ON OR VERTICALLY.
RAP CABLES AND BLOCK
ITH SCOTCH ELECTRICAL
resistance is proportionately less. If a Mar-
SPACE BLOCKS
ALONG BALUN
PP /i/ coni antenna is made out of 300 -ohm TV-
type ribbon line, as shown in figure 10, the
radiation resistance of the antenna is raised
from a low value of 10 or 15 ohms to a
more reasonable value of 40 to 60 ohms.
0.S The ground losses are now reduced by a
factor of 4. In addition, the antenna may
be directly fed from a 50 -ohm coaxial line,
or directly from the unbalanced output of
FIGURE A FIGURE B a pi- network transmitter.
CUT OFF SHIELD ANO OUTER REMOVE OUTER JACKET
JACKET AS SHOWN. ALLOW
DIELECTRIC TO EXTEND PART
PROM A SHORT LENGTH OF
CABLE AS SHOWN HERE.
Since a certain amount of power may still
WAY TO OTHER CABLE. COVER UNBRAID THE SHIELD Of
ALL EXPOSED SHIELD ANO COAX C CUTOFF THE DI- be lost in the ground connection, it is still
DIELECTRIC ON BOTH CABLES ELECTRIC AND INNER CON-
WITH A CONTINUOUS WRAP-
PING Of SCOTCH ELECTRICAL
DUCTOR FLUSH WITH TN of greatest importance that a good, low -
61
OUTER JACKET. DO NOT CUT
TAPE TO EXCLUDE MOISTURE. THE SHIELD. WRAP SHIELD
OF COAX C. AROUND SHIELD
resistance ground be used with this antenna.
Of COAX D. SOLDER THE
KEEP BALUN AT ' LEAR CONNECTION. BEING VERY
(ILCpICMADVAGAL A Broadband
co GROUND AND OTHER OBJECTS. THE
HOLD CABLE D STRAIGHT
. Shown in figures 11 and 12
FOR DETAIL SEE FIGURE B WHILE SOLDERING. COVER
THE AREA WITH A CONTI N-
Dipole System are broadband dipoles for the
Sa- OHM RE-S /U, ANY LENGTH UOUS
ELECTRICALT
NG OF SCOTCH
NOCOH 40- and 80 -meter amateur
NTOIMNBIICONDUC-
TORS. bands. These fan -type dipoles have excellent
broadband response, and are designed to be
fed with a 52 -ohm unbalanced coaxial line.
DIMENS IONS SHOWN HERE ARE POR TAWSD-METER SAND. THIS ANT-
ENNA MAr SE SU/LT /UR OTHER SANDS Sr LOINS DIMENSIONS rAUr
ARE MULTIPLES OR SUBMULTIPLES OP THE DIMENSIONS SHOWN.
The antenna system consists of a fan -type
SALON SPACING IS I.!'
ON ALL SANDS. dipole, a balun matching section, and a suit-
Figure 12 able coaxial feedline. The Q of the half -
wave 80 -meter doublet is lowered by de-
BROADBAND ANTENNA WITH QUARTER - creasing the effective length -to- diameter
WAVE UNBALANCED -TO- BALANCED ratio. The frequency range of operation of
TRANSFORMER (BALUN) FEED SYSTEM the doublet is increased considerably by
FOR 80 -METER OPERATION this change. A typical SWR curve for the
80 -meter doublet is shown in figure 13.
The balanced doublet is matched to the
by a shortened antenna system is approxi- unbalanced coaxial line by the quarter -wave
mately in proportion to the amount of
balun. If desired, a shortened balun may be
shortening which has been employed. For used (figure 14). The short balun is ca-
example, the antenna system shown in fig- pacitance loaded at the junction between the
ure 9C may be operated over the range balun and the broadband dipole.
from 3800 to 4000 kHz without serious
standing waves on the feed line. If the an- !
tenna had been made full length it would
be possible to cover about half again as
much frequency range for the same amount
of mismatch at the extremes of the fre-
quency range. 3S 3.11 3.7 3.1 3S 4.0
FREQUENCY (IAN)
The Twin -Lead Much of the power loss in
Marconi Antenna the Marconi antenna is a Figure 13
result of low radiation re-
sistance and high ground resistance. In some SWR CURVE OF 80 -METER BROADBAND
DIPOLE
cases, the ground resistance may even be
www.americanradiohistory.com
HANDBOOK Multiband Antennas 447
ANTENNA
used from the antenna to the transmitter
and since the antenna system is balanced
LU with respect to ground. With operation on
SO METERS
Figure 18
END-FED ZEPP
www.americanradiohistory.com
HANDBOOK Multiband Antennas 449
100 {
a pi- network transmitter coupling circuit.
The use of a low -pass TVI filter is of course
recommended.
ENO 0/ ila
LINE .s APRO/. LE
S.S4EN3 132
PRONE 10 PARALLEL 120o ONUs 30
) 04001 PARALLEL 1200 ONUS
4101. 1200 ONUS
111 OS SO
A-NN3 1200 0001
21-NN3 1200 ONOS
I S41INI
1]1 Iu suas 1> OROS
14 21012 HSlts tf 0003 ES
ar OR
1/21N1
WV LONG 300 -01M OPEN -WIRE
PARALLEL 1200 ONUS TV TYPE LINE
T. INN IoS OS PARALLEL 1200 0005 67
IA.NN3 PAII ALLlL 1200 00341
T-NN3 SEAMS TS ONUS
14.11143 3 1 PARALLEL 200 0001 SS
21-WI 1200 00S
T.NMI PARALLEL 1200 ONUS
1A.NN3
2WN3
100 PARALLEL 1200 0003 a
1200 O01
ANTENNA TUNER
OR
*MATCH BOA AA IAL
LINE
Figure 21
Figure 20
MULTIBAND ANTENNA USING FAN -
DIMENSIONS FOR CENTER -FED MULTI - DIPOLE TO LIMIT IMPEDANCE EXCUR-
BAND ANTENNA SIONS ON HARMONIC FREQUENCIES
450 Antennas and Antenna Matching THE RADIO
If the feed line between the transmitter Folded Flat Top As has been mentioned
and the antenna is made to have a character- Dual -Band Antenna earlier, there is an in-
istic impedance of approximately 300 ohms creasing tendency
the excursions in end -of- feeder impedance among amateur operators to utilize rotary
are greatly reduced. or fixed arrays for the 14 -MHz band and
There are several practical types of trans- those higher in frequency. In order to af-
mission line which can give an impedance ford complete coverage of the amateur
of approximately 300 ohms. The first is, bands it is then desirable to have an addi-
obviously, 300 -ohm twin -lead. Twin -lead tional system which will operate with equal
of the receiving type may be used as a reso- effectiveness an the 3.5- and 7 -MHz bands,
but this low- frequency antenna system
will not be required to operate on any
bands higher in frequency than the 7 -MHz
band. The antenna system shown in figure
22 has been developed to fill this need.
This system consists essentially of an
open -wire folded dipole for the 7 -MHz band
with a special feed system which allows the
antenna to be fed with minimum standing
waves on the feed line on both the 7 -MHz
and 3.5-MHz bands. The feed -point imped-
ance of a folded dipole on its fundamental
frequency is approximately 300 ohms.
Hence the 300 -ohm twin lead shown in
figure 22 can be connected directly into the
Figure 22 center of the system for operation only on
the 7 -MHz band and standing waves on
FOLDED -TOP DUAL -BAND ANTENNA the feeder will be very small. However, it
is possible to insert an electrical half wave
nant feed line in this case, but its use is not of transmission line of any characteristic
recommended with power levels greater impedance into a feeder system such as this
than perhaps 150 watts, and it should not and the impedance at the far end of the line
be used when lowest loss in the transmis- will be exactly the same value of impedance
sion line is desired. which the half -wave line "sees" at its termi-
For power levels up to 250 watts or so, nation. Hence this has been done in the
the transmitting type tubular 300 -ohm antenna system shown in figure 22; an elec-
line may be used, or the open -wire 300 -ohm trical half wave of line has been inserted
TV line may be employed. For power levels between the feed point of the antenna and
higher than this, a 4 -wire transmission line, the 300 -ohm transmission line to the trans-
or a line built of one -quarter inch tubing mitter.
should be used. The characteristic impedance of this ad-
Even when a 300 -ohm transmission line ditional half -wave section of transmission
is used, the end -of- feeder impedance may
line has been made about 715 ohms (No. 20
reach a high value, particularly on the sec- wire spaced 6 inches), but since it is an
ond harmonic of the antenna. To limit the electrical half wave long at 7 MHz and
impedance excursions, a two -wire flat top operates into a load of 300 ohms at the
may be employed for the radiator, as shown antenna the 300 -ohm twin lead at the bot-
in figure 21. The use of such a radiator will tom of the half -wave section still "sees" an
limit the impedance excursions on the har- impedance of 300 ohms. The additional half -
monic frequencies of the antenna and make wave section of transmission line introduces
the operation of the antenna matching unit a negligible amount of loss since the current
much less critical. The use of a two -wire flowing in the section of line is the same
radiator is highly recommended for any which would flow in a 300 -ohm line at
center -fed multiband antenna. each end of the half -wave section, and at
www.americanradiohistory.com
HANDBOOK The Multee Antenna 451
a< 3.5
W
; 3.0
as
ii
iti 2.0
CUT -OFF FREQUENCY
15
D:
3
VI
22 26 30 34 36 42 4e 54 56
I
e 10 14 16 SO
FREQUENCY (Poll)
Figure 25
SWR CURVE FOR A 13.2-MHz DISCONE
ANTENNA. SWR IS BELOW 1.5 TO 1 FROM
13.0 MHz TO 58 MHz
VOLTAGE CURVES
AS Guy
POST
win. POR W000
wOOD
Figure 26
meter. The number of active turns in the of the vertical antenna is adjusted in length
coil and the feedline tap are varied a turn for 10 -meter resonance at about 28.7 MHz,
at a time until proper transmitter loading is followed by adjustment of the center section
achieved with a reasonably low value of for resonance at 21.2 MHz. The last ad-
SWR on the transmission line (below 1.5 /1 justment is to the top section for resonance
or so at the center frequency in each band). at about 14.2 MHz.
The trap technique described in the Di- It must be remembered that trap, or other
rective Antennas chapter can be used for a multifrequency antennas are capable of ra-
three -band vertical antenna as shown in diating harmonics of the transmitter that
figure 29. This antenna is designed for may be coupled to them via the transmission
operation on 10, 15, and 20 meters and uses line. It is well to check for harmonic radi-
a separate radial system for each band. No ation with a nearby radio amateur. If such
adjustments need be made to the antenna harmonics are noted, an antenna tuner sim-
when changing frequency from one band to ilar to the one described later in this chapter
another. Substitution of a ground connection should be added to the installation to reduce
for the radials is not recommended because unwanted harmonics to a minimum.
of the high ground loss normally encoun-
tered at these frequencies. Typical trap con- 21 -8 Matching
struction is discussed in the reference chap- Nonresonant Lines
ter, and the vertical radiator is built of to the Antenna
sections of aluminum tubing, as described
earlier. Present practice in regard to the use of
Each trap is built and grid- dipped to the transmission lines for feeding antenna sys-
proper frequency before it is placed in the tems on the amateur bands is about equally
radiator assembly. The 10 -meter trap is divided between three types of transmission
line: (1) Ribbon or tubular molded 300 -
self- resonant at about 27.9 MHz and the
15 -meter trap is self- resonant at about 20.8
ohm line is widely used up to moderate
MHz. Once resonated, the traps need no power levels (the "transmitting" type is
further adjustment and do not enter into usable up to the kilowatt level). (2) Open -
later adjustments made to the antenna. The wire 400- to 600 -ohm line is most common-
ly used when the antenna is some distance
complete antenna is resonated to each ama-
teur band by placing a single -turn coil from the transmitter, because of the low
between the base of the vertical radiator and attenuation of this type of line. (3) Coaxial
line (usually RG -8 /U with a 52 -ohm char-
the radial connection and coupling the grid -
acteristic impedance) is widely used in vhf
dip oscillator to the coil. The coaxial line
work and also on the lower frequencies
is removed for this test. The lower section
where the feed line must run underground
L
or through the walls of a building. Coaxial
line also is of assistance in TVI reduction
since the r -f field is entirely enclosed within
MATCHING SECTION
the line. Molded 75 -ohm line is sometimes
used to feed a doublet antenna, but the
doublet has been largely superseded by the
folded -dipole antenna fed by 300 -ohm rib-
bon or tubular line when an antenna for a
single band is required.
Standing Waves As was discussed earlier,
standing waves on the an-
tenna transmission line, in the transmitting
NON RESONANT
LINE case, are a result of reflection from the point
Figure 30 where the feed line joins the antenna system.
THE DELTA -MATCHED DIPOLE ANTENNA
The magnitude of the standing waves is
for the portions of the an-
The dimensions determined by the degree of mismatch be-
tenna are given in the text. tween the characteristic impedance of the
456 Antennas and Antenna Matching THE RADIO
www.americanradiohistory.com
HANDBOOK Matching Systems 457
Figure 33
Stub Length Current -Fed Voltage -Fed
Electrical) Radiator Radiator
MATCHING STUB APPLICATIONS
1/4-3/4-11/4-etc. Open Shorted
wavelengths Stub Stub An end-fed half -wave antenna with a quar-
ter -wave shorted stub is shown at A. 8 shows
1/2 -1- 11/2.2-etc. Shorted Open the use of a half -wave shorted stub to feed
wavelengths Stub Stub o relatively low impedance point such as the
center of the driven element of a parasitic
array, or the center of a half -wave dipole.
Any number of half waves can be added The use of an open -ended quarter -wave stub
to either a quarter -wave stub or a half -wave to feed a low impedance is illustrated at
stub without disturbing the operation, C. D shows the c tional use of a shorted
quarter -wave stub to voltage -feed two half -
though losses and frequency sensitivity will wave antennas with a 180 phase duff
HANDBOOK Matching Systems 459
be lowest if the shortest usable stub is em- Quarter -Wave The impedance inverting
ployed (see chart). Matching characteristic of a quarter -
Transformers wave section of transmission
Linear R -F A resonant quarter -wave line line is widely used by mak-
Transformers has the unusual property of ing such a section of line act as a quarter -
acting much as a transform- wave transformer. The quarter -wave trans-
er. Let us take, for example, a section con- former may be used in a wide number of
sisting of No. 12 wire spaced 6 inches, applications wherever a transformer is re-
which happens to have a surge impedance of quired to match two impedances whose
600 ohms. Let the far end be terminated geometric mean is somewhere between per-
with a pure resistance, and let the near end haps 25 and 750 ohms when transmission -
be fed with radio- frequency energy at the line sections can be used. Paralleled coaxial
frequency for which the line is a quarter lines may be used to obtain the lowest im-
wavelength long. If an impedance measuring pedance mentioned, and open -wire lines
set is used to measure the impedance at the composed of small conductors spaced a
near end while the impedance at the far end moderate distance may be used to obtain the
is varied, an interesting relationship between higher impedance. A short list of imped-
the 600 -ohm characteristic surge impedance ances, which may be matched by quarter -
of this particular quarter -wave matching wave sections of transmission line having
line, and the impedance at the ends will be specified impedances, follows.
discovered.
When the impedance at the far end of
the line is the same as the characteristic Load or Ant.
Impedance 300 480 600 Impedance
surge impedance of the line itself (600 .
L FEET
anced input. For some time the Collins
MHz
Radio Co. has been using a balun and ta-
zl pered -line system for matching a coaxial -
output transmitter to an open -wire balanced
L
_
234
F(NHz)
f- TUBING
Zo= z
Q- HATCHING SECTION transmission line. Illustratd in figure 35
is one type of matching system which is
proving satisfactory over a 4:1 frequency
range. Z, is the transmitter end of the sys-
tem and may be any length of 52 -ohm
coaxial cable. Z2 is one -quarter wavelength
2
long at the midfrequency of the range to be
covered and is made of 75 -ohm coaxial
cable. ZA is a .quarter- wavelength shorted
section of cable at the midfrequency. Z
(ZA and Z2) forms a 200 -ohm quarter -
UNTUNED LINE
wave section. The ZA section is formed of a
ANY LENGTH conductor of the same diameter as Z2. The
Figure 34 difference in length between ZA and. Z2 is
HALF-WAVE RADIATOR FED accounted for by the fact that Z2 is a co-
BY "Q BARS" axial conductor with a solid dielectric,
The Q matching section is simply a quarter-
whereas the dielectric for Zo is air. Z, is
wave transformer whose impedance is equal one -quarter wavelength long at the mid -
to the geometric mean between the imped- frequency and has an impedance of 123
ance at the center of the antenna and the
impedance of the transmission line to be ohms. Z, is one -quarter wavelength long at
used to feed the bottom of the transformer. the midfrequency and has an impedance of
The transformer may be made up of parallel
tubing, "ribbon" line, or any other type of 224 ohms. Z., is the balanced line to be
transmission line which has the correct value matched (in this case 300 ohms) and may
of impedance. be any length.
Other system parameters for different
A Q- matched system can be adjusted output and input impedances may be cal-
precisely, if desired, by constructing a culated from the following:
matching section to the calculated dimen-
sions with provision for varying the spacing N Zoo,
of the Q- section conductors slightly, after r Zin
the untuned line has been checked for stand-
ing waves. where,
N is the number of sections. In the above
Center to Impedance Impedance case:
Center in Ohms in Ohms
Spacing for 1/2" for 1/4" 3 Z,
in Inches Diameters Diameters r = Z,
1.0 170 250
1.25 188 277
1.5 207 298
Zo Z3 Z4
1.75 225 318
ZS
2.0 248 335 Z2
ZS=30011
Z3-12311 Z4'22411 (ANY LENGTH)
PARALLEL TUBING SURGE IMPEDANCE
ZA
FOR MATCHING SECTIONS L INNER & OUTER CONDUCTORS
SHORTED AT EACH END
www.americanradiohistory.com
HANDBOOK Antenna Construction 461
Impedance between sections, as Z2_ 3, is r practical. They can be made self- support-
times the preceding section. Z2_3 =
r X ing, but a few guys will enable one to use a
Z1, and Z:1_, = r X Z2_3 smaller cross section without danger from
Midfrequency (in): high winds. The torque exerted on the base
+ of a high self -supporting mast is terrific
- Fi
2
F2
during a strong wind.
The "A- Frome" Figures 36A and 36B show
+
For 40 -20 -10 meters -7 30
- 18.5 MHz Most the standard method of
construction of the A-
and one -quarter wavelength = 12 feet. f rame type of mast. This type of mast is
+54 quite frequently used since there is only a
For 20 -10 -6 meters =
14
2
- 34 MHz moderate amount of work involved in the
construction of the assembly and since the
and one -quarter wavelength = 5.5 feet. material cost is relatively small. The three
pieces of selected 2 by 2 are first set up on
The impedance of the sections are:
three sawhorses or boxes and the holes
z_ =V' Z,XZ2 -3 drilled for the three 1/4-inch bolts through
Z3 = Zp_3 X Z3_1 the center of the assembly. Then the base
legs are spread out to about 6 feet and the
Z,= bottom braces installed. Finally the upper
Z =s%XZ, braces and the cross pieces are installed and
Generally, the larger the number of taper the assembly given several coats of good -
sections the greater will be the bandwidth quality paint as a protection against weath-
of the system. ering.
Figure 36C shows another common type
21 -9 Antenna of mast which is made up of sections of 2
Supports by 4 placed end -to -end with stiffening sec-
tions of 1 by 6 bolted to the edge of the 2
The foregoing portion of this chapter has by 4 section. Both types of mast will require
been concerned primarily with the electrical a set of top guys and another set of guys
characteristics and considerations of anten- about one -third of the way down from the
nas. Some of the physical aspects and me- top. Two guys spaced about 90 to 100 de-
chanical problems incident to the actual grees and pulling against the load of the
erection of antennas and arrays will be dis- antenna will normally be adequate for the
cussed in the following section. top guys. Three guys are usually used at
Up to 30 feet, there is little point in using the lower level, with one directly behind
mast -type antenna supports unless guy wires the load of the antenna and two more
either must be eliminated or kept to a mini- spaced 120 degrees from the rear guy.
mum. While a little more difficult to erect, Raising the mast is made much easier if a
because of their floppy nature, fabricated gin pole about 20 feet high is installed about
wood poles of the type to be described will 30 or 40 feet to the rear of the direction
be just as satisfactory as more rigid types, in which the antenna is to be raised. A line
provided many guy wires are used. from a pulley on the top of the gin pole is
Rather expensive when purchased through then run to the top of the pole to be raised.
the regular channels, 40- and 50 -foot tele- The gin pole comes into play when the
phone poles sometimes can be obtained quite center of the mast has been raised 10 to 20
reasonably. In the latter case, they are hard feet above the ground and an additional ele-
to beat, inasmuch as they require no guying vated pull is required to keep the top of the
if set in the ground six feet (standard mast coming up as the center is raised fur -
depth), and the resultant pull in any lateral ther above ground.
direction is not in excess of a hundred
pounds or so. Using TV Masts Steel tubing masts of the
For heights of 80 to 100 feet, either three - telescoping variety are
or four -sided lattice -type masts are most widely available at a moderate price for use
462 Antennas and Antenna Matching THE RADIO
le
2%
Figure 36 BOLTS
p_.x1r
cowcRere:We
in supporting television antenna arrays. If "dead .men" are used for the guy wire
These masts usually consist of several 10- terminations, the wire or rod reaching from
foot lengths of electrical metal tubing the dead men to the surface should be of
(EMT) of sizes such that the sections will nonrusting material, such as brass, or given
telescope. The 30- and 40 -foot lengths are a heavy coating of asphalt or other protec-
well suited as masts for supporting antennas tive substance to prevent destructive action
and arrays of the type used on the amateur by the damp soil. Galvanized iron wire will
bands. The masts are constructed in such a last only a short time when buried in moist
manner that the bottom 10 -foot length may soil.
be guyed permanently before the other sec- Only strain -type (compression) insulators
tions are raised. Then the upper sections should be used for guy wires. Regular ones
may be extended, beginning with the top- might be sufficiently strong for the job, but
mast section, until the mast is at full length it is not worth taking chances, and egg -type
(provided a strong wind is not blowing) strain halyard insulators are no more expen-
following which all the guys may be an- sive.
chored. It is important that there be no load Only a brass or bronze pulley should be
on the top of the mast when the "vertical" used for the halyard, as a high pole with a
raising method is to be employed. rusted pulley is truly a sad affair. The bear-
ing of the pulley should be given a few
Guy Wires Guy wires should never be pulled drops of heavy machine oil before the pole
taut; a small amount of slack is or tower is raised. The halyard itself should
be of good material, preferably waterproofed.
desirable. Galvanized wire, somewhat heavier
than seems sufficient for the job, should be Hemp rope of good quality is better than
window sash cord from several standpoints,
used. The heavier wire is a little harder to
and is less expensive. Soaking it thoroughly
handle, but costs only a little more and in engine oil of medium viscosity, and then
takes longer to rust through. Care should be wiping it off with a rag, will not only ex-
taken to make sure that no kinks exist when tend its life but minimize shrinkage in wet
the pole or tower is ready for erection, as weather. Because of the difficulty of re-
the wire will be greatly weakened at such placing a broken halyard it is a good idea
points if a kink is pulled tight, even if it to replace it periodically, without waiting
is later straightened. for it to show excessive deterioration.
www.americanradiohistory.com
HANDBOOK Antenna Construction 463
It is an excellent idea to tie both ends of only the best from a preservative standpoint,
the halyard line together in the manner of but looks very well. This type of paint,
a flag -pole line. Then the antenna is tied when purchased in quantities, is consider-
onto the place where the two ends of the ably cheaper than might be gathered from
halyard are joined. This procedure of mak- the price asked for quarter -pint cans.
ing the halyard into a loop prevents losing Portions of posts or poles below the sur-
the top end of the halyard should the an- face of the soil can be protected from ter-
tenna break near the end, and it also pre- mites and moisture by painting with cresote.
vents losing the halyard completely should While not so strong initially, redwood will
the end of the halyard carelessly be allowed deteriorate much more slowly when buried
to go free and be pulled through the pulley than will the white woods, such as pine.
at the top of the mast by the antenna load. Antenna Wire
A somewhat longer piece of line is required The antenna or array itself
but the insurance is well worth the cost of presents no special problem.
the additional length of rope. A few considerations should be borne in
mind, however, For instance, soft -drawn
Trees as Often a tall tree can be used to copper should not be used, as even a short
Supports support one end of an antenna, span will stretch several percent after whip-
but one should not attempt to at- ping around in the wind a few weeks, thus
tach anything to the top, as the swaying of affecting the resonant frequency. Enameled
the top of the tree during a heavy wind will copper wire, as ordinarily available at radio
complicate matters. stores, is usually soft -drawn, but by tying
If a tree is utilized for support, provision one end to some object such as a telephone
should be made for keeping the antenna taut pole and the other to the frame of an auto, a
without submitting it to the possibility of few husky tugs can be given and the wire,
being severed during a heavy wind. This can after stretching a bit, is equivalent to hard -
be done by the simple expedient of using a drawn.
pulley and halyard, with weights attached to Where a long span of wire is required,
the lower end of the halyard to keep the or where heavy insulators in the center of
antenna taut. Only enough weight to avoid the span result in considerable tension, cop-
excessive sag in the antenna should be tied per-clad steel wire is somewhat better than
to the halyard, as the continual swaying of hard -drawn copper. It is a bit more expen-
the tree submits the pulley and halyard to sive, though the cost is far from prohibitive.
considerable wear. The use of such wire, in conjunction with
Galvanized iron pipe, or steel -tube con- strain insulators is advisable where the
duit, is often used as a vertical radiator, and antenna would endanger persons or property
is quite satisfactory for the purpose. How- should it break.
ever, when used for supporting antennas, it For transmission lines and tuning stubs
should be remembered that the grounded steel -core or hard -drawn wire will prove
supporting poles will distort the field pat- awkward to handle, and soft -drawn copper
tern of a vertically polarized antenna unless should, therefore, be used. If the line is long,
spaced some distance from the radiating the strain can be eased by supporting it at
portion. several points.
More important from an electrical stand-
Pointing The life of a wood mast or pole point than the actual size of wire used is
can be increased several hundred the soldering of joints, especially at current
percent by protecting it from the elements loops in an antenna of low radiation resist-
with a coat or two of paint. And, of course, ance. In fact, it is good practice to solder
the appearance is greatly enhanced. The all joints, thus ensuring quiet operation when
wood should first be given a primer coat of the antenna is used for receiving.
flit white outside house paint, which can be
thinned down a bit to advantage with sec- Insulation A question that often arises is
ond -grade linseed oil. For the second coat, that of insulation. It depends, of
which should not be applied until the first course, on the r -f voltage at the point at
is thoroughly dry, aluminum paint is not which the insulator is placed. The r-f volt-
464 Antennas and Antenna Matching THE RADIO
age, in turn, depends on the distance from a virtual necessity for eliminating interference
current node, and the radiation resistance to TV reception. However, the TVI- elimi-
of the antenna. Radiators having low radia- nation portion of the problem may always
tion resistance have very high voltage at the be accomplished by adequate shielding of the
voltage loops; consequently, better than transmitter, by filtering of the control and
usual insulation is advisable at those points. power leads which enter the transmitter
Open -wire lines operated as nonresonant enclosure, and by the inclusion of a har-
lines have little voltage across them; hence monic- attenuating filter between the output
the most inexpensive ceramic types are suffi- of the transmitter and the antenna system.
ciently good electrically. With tuned lines, Although TVI may be eliminated through
the voltage depends on the amplitude of the inclusion of a filter between the output of a
standing waves. If they are very great, the shielded transmitter and the antenna system,
voltage will reach high values at the voltage the fact that such a filter should be included
loops, and the best spacers available are in the link between transmitter and antenna
none too good. At the current loops the makes it necessary that the transmitter -
voltage is quite low, and almost anything loading problem be re-evaluated in terms of
will suffice. the necessity for inclusion of such a filter.
When insulators are subject to very high Harmonic- attenuating filters must be op-
r -f voltages, they should be cleaned occa- erated at an impedance level which is close
sionally if in the vicinity of sea water or to their design value; therefore they must
smoke. Salt scum and soot are not readily operate into a resistive termination substan-
dislodged by rain, and when the coating be- tially equal to the characteristic impedance
comes heavy enough, the efficiency of the of the filter. If such filters are operated into
insulators is greatly impaired. an impedance which is not resistive and ap-
If a very pretentious installation is to be proximately equal to their characteristic
made, it is wise to check up on both Under- impedance: (1) the capacitors used in the
writer's rules and local ordinances which filter sections will be subjected to high peak
might be applicable. If you live anywhere voltages and may be damaged, (2) the har-
near an airport, and are contemplating a monic- attenuating properties of the filter
tall pole, it is best to investigate possible will be decreased, and (3) the impedance at
regulations and ordinances pertaining to the input end of the filter will be different
towers in the district, before starting con- from that seen by the filter at the load end
struction. (except in the case of the half -wave type
of filter) . It is therefore important that
21 -10 Coupling to the the filter be included in the transmitter -to-
antenna circuit at a point where the im-
Antenna System pedance is close to the nominal value of the
filter, and at a point where this impedance is
When coupling an antenna feed system to likely to remain fairly constant with varia-
a transmitter the most important considera- tions in frequency.
tions are as follows: (1) means should be
Block Diagrams of There are two basic
provided for varying the load on the ampli-
Transmitter -to- Antenna arrangements which
fier; (2) the load presented to the final
Coupling Systems include all the provi-
amplifier should be resistive (nonreactive)
in character; and (3) means should be pro- sions required in the
vided to reduce harmonic coupling between transmitter -to- antenna coupling system, and
the final amplifier plate tank circuit and which permit the harmonic -attenuating fil-
the antenna or antenna transmission line to ter to be placed at a position in the coupling
an extremely low value. system where it can be operated at an im-
pedance level close to its nominal value.
The Transmitter- The problem of coupling These arrangements are illustrated in block
Loading Problem the power output of a high - diagram form in figures 37 and 38.
frequency or vhf trans- The arrangement of figure 37 is recom-
mitter to the radiating portion of the an- mended for use with a single -band antenna
tenna system has been complicated by the system, such as a dipole or a rotatable array,
HANDBOOK Antenna Coupling Systems 465
EXCITER
PORTION - FILIFINAL
AMPER
COUPLING
AOJVS1/T
- IIARMDNIC
TTENWTIN - - - -
SYSTEM TRANSMISSION
LINE
IMPEDANCE
MATCHING
\T ANTENNA SYSTEM
Figure 37
wherein an impedance matching system is the harmonic filter. Hence the arrangement
included within or adjacent to the antenna. of figure 38 is more general than the figure
The feed line coming down from the anten- 37 system, since the inclusion of the anten-
na system should have a characteristic im- na coupler allows the system to feed an
pedance equal to the nominal impedance of antenna transmission line of any reasonable
the harmonic filter, and the impedance impedance value, and also without regard to
matching at the antenna should be such that the standing -wave ratio which might exist
th; standing -wave ratio on the antenna on the antenna transmission line. Antenna
feed line is less than 2 to 1 over the range of couplers are discussed in a following section.
frequency to be fed to the antenna. Such
an arrangement may be used with open -wire Output Coupling It will be noticed by refer -
line, ribbon or tubular line, or with coaxial Adjustment erence to both figure 37
cable. The use of coaxial cable is to be and figure 38 that a box
recommended, but in any event the imped- labeled Coupling Adjustment is included in
ance of the antenna transmission line should the block diagram. Such an element is nec-
be the same as the nominal impedance of the essary in the complete system to afford an
harmonic filter. The arrangement of figure adjustment in the value of load impedance
37 is more or less standard for commercial- presented to the tubes in the final amplifier
ly manufactured equipment for amateur stage of the transmitter. The impedance at
and commercial use in the high -frequency the input terminal of the harmonic filter is
and vhf range. established by the antenna, through its
The arrangement of figure 38 merely adds matching system and the antenna coupler,
an antenna coupler between the output of
if used. In any event the impedance at the
the harmonic attenuating filter and the an-
tenna transmission line. The antenna coupler input terminal of the harmonic filter should
will have some harmonic -attenuating action, be very close to the nominal impedance of
but its main function is to transform the the filter. Then the Coupling Adjustment
impedance at the station end of the antenna provides means for transforming this im-
transmission line to the nominal value of pedance value to the correct operating value
AT TRANSMITTER AT ANTENNA
_sitisua
i
HARMONIC IMPEDANCE RADIATING
EACITER FINAL COUPLING ATTLNUATINI
ANTENNA MATCHING
PORTION AMPLIFIER 11WI /STMENT COUPLER SYSTEM
SYSTEM TRANSMISSION T ANTENNA
LINE
L J
Figure 38
TO ANTENNA
- ATTENUATING
HARMONIC
FILTER
COAL ANT.
CHANGEOVER
RELAY
FEEOLINE OR
TO ANTENNA
COUPLER
Figure 39
of load impedance which should be presented mitter, due to the low Q and low impedance
to the final amplifier stage. of the coupling circuit.
There are two common ways for accom-
plishing the antenna coupling adjustment, Pi- Network The pi- network coupling sys-
as illustrated in figures 39 and 40. Figure 39
Coupling tem offers two advantages: (1)
shows the variable link arrangement often a mechanical coupling variation
used in home- construction equipment, while is not required to vary the loading of the
the pi- network coupling arrangement is il- final amplifier, and (2) the pi- network (if
lustrated in figure 40. Either method may be used with an operating Q of about 10) of-
used, and each has its advantages. fers within itself a harmonic attenuation of
30 db or more, in addition to the harmonic
Variable Link The variable link method il- attenuation provided by the additional har-
Coupling lustrated in figure 39 pro- monic attenuating filter. Some commercial
vides good rejection to sub - equipment incorporates an L- network in ad-
harmonics. For greatest bandwidth of oper- dition to the pi- network, for accomplishing
ation of the coupling circuit, the reactance the impedance transformation in two steps
of link coil L and the reactance of link to provide additional harmonic attenuation.
tuning capacitor C should both be between
3 and 4 times the nominal load impedance Tuning the Tuning a pi- network
of the harmonic filter. This is to say that Pi- Section Coupler coupling circuit such as
the inductive reactance of coupling link L illustrated in figure 40 is
should be tuned out or resonated by capaci- accomplished in the following manner: First
tor C, and the operating Q of the LC link place a dummy load on the output terminal
circuit should be between 3 and 4. If the of the transmitter. Tune C2 to a capacitance
link coil is not variable with respect to the which is large for the band in use, adding
tank coil of the final amplifier, capacitor C suitable additional capacitance by switch S
may be used as a loading control; however, if operation is to be on one of the lower-
this system is not recommended since its use frequency bands. Apply reduced plate volt-
will require adjustment of C whenever a age to the stage and dip to resonance with
frequency change is made at the transmitter. C,. It may be necessary to vary the induct-
If L and C are made resonant at the center ance in coil L, but in any event, resonance
of a band, with a link circuit Q of 3 to 4, should be reached with a setting of C, which
and coupling adjustment is made by physical is approximately correct for the desired
adjustment of L with respect to the final value of operating Q of the pi- network.
amplifier tank coil, it usually will be possi- Next, couple the load to the amplifier
ble to operate over an entire amateur band (through the harmonic filter) , apply re-
without change in the coupling system. duced plate voltage again and dip to reso-
Capacitor C normally may have a low volt- nance with C,. If the plate- current dip with
age rating, even with a high -power trans- load is too low (taking into consideration
HANDBOOK Antenna Couplers 467
COATI TO
RECEIVER
To FEEDLINE
OR ANTENNA
COUPLER
Figure 40
the reduced plate voltage), decrease the ca- antenna, or where a long wire without a
pacitance of C_ and again dip to resonance, separate feed line is to be fed from the out-
repeating the procedure until the correct put of the harmonic filter. In such cases an
value of plate current is obtained with full antenna coupler is required.
plate voltage on the stage. There should be In certain cases when a pi- network is
a relatively small change required in the being used at the output of the transmitter,
setting of C, (from the original setting the addition of an antenna coupler will pro-
of C, without load) if the operating Q vide sufficient harmonic attenuation. But in
of the network is correct and if a large all normal cases it is prudent to include a
value of impedance transformation is being harmonic filter between the output of the
employed -as would be the case when trans- transmitter and the antenna coupler.
forming from the plate impedance of a
single -ended output stage down to the 50- Function of on The function of the an-
ohm impedance of the usual harmonic filter Antenna Coupler tenna coupler is, basically,
and its subsequent load. to transform the imped-
In a pi- network of this type the har- ance of the antenna system being used to
monic attenuation of the section will be the correct value of resistive impedance for
adequate when the correct value of C, and the harmonic filter, and hence for the trans-
L are being used and when the resonant dip mitter. Thus the antenna coupler may be
in C, is sharp. If the dip in C, is broad, or if used to resonate the feeders or the radiating
the plate current persists in being too high portion of the antenna system, in addition
with Co, at maximum setting, it means that to its function of impedance transforma-
a greater value of capacitance is required at
tion.
C2, assuming that the values of C, and I.
It is important to remember that there is
are correct.
nothing that can be done at the antenna
21 -11 Antenna Couplers coupler which will eliminate standing waves
on the antenna transmission line. Standing
As stated in the previous section, an an- waves are the result of reflection from the
tenna coupler is not required when the im-
antenna, and the coupler can do nothing
pedance of the antenna transmission line is
about this condition. However, the antenna
the same as the nominal impedance of the
harmonic filter, and the antenna feed line is coupler can resonate the feed line (by intro-
being operated with a low standing -wave ducing a conjugate impedance) in addition
ratio. However, there are many cases where to providing an impedance transformation.
it is desirable to feed a multiband antenna Thus, a resistive impedance of the correct
from the output of the harmonic filter, value can be presented to the harmonic fil-
where a tuned line is being used to feed the ter, as in figure 38, regardless of any reason-
468 Antennas and Antenna Matching THE RADIO
COAX. TO
RECEIVER
SINGLE -WIRE
HERTZ ANTENNA
OR FEEDER
Figure 41
ALTERNATIVE ANTENNA COUPLER CIRCUITS
Plug -in coils, one or two variable capacitors of the split- stator variety, and a system of
switches or plugs and jacks may be used in the antenna coupler to accomplish the feeding
of different types of antennas and antenna transmission lines from the coaxial input line from
the transmitter or from the antenna changeover relay. Link L should be resonated with
capacitor C at the operating frequency of the transmitter so that the harmonic filter will
operate into a resistive load impedance of the correct nominal value.
52 R INPUT
FROM XMTR
RI IA
5
250 SINGLE -
WIRE
Ca S1
L2 ANT
25 M
KV
MICA TUNE
CI
-_
SENSITIVITY 154
NV.
0 -I
Li-35 TURNS e 1e,
3.5^ LONG (A /P
2-DUE)
DIA., C1- JOHNSON 350E20
TAP AT IST., 27T., C2- CENTRALAATYPE 822
FROM POINT A J I -TYPE 50-239 RECEPTACLE
Figure 44 La-JOHNSON 229 -207 VAR IABL E RI -TEN 1O-OHM I -WATT CAR-
INDUCTOR (10 LN) BON RESISTORS IN PARA-
LLEL. /RC TYPE ATA
ANTENNA TUNER IS HOUSED IN
METAL CABINET 7 INCHES X 8
INCHES IN SIZE
Figure 45
Inductance switch S, and sensitivity control
are at left with counter dial for L, at center.
Output tuning capacitor C, is at right. SWR SCHEMATIC OF A SINGLE -WIRE
meter is mounted above S,. ANTENNA TUNER
Final adjustments are made with the sensi- Connection to the antenna is made by
tivity control at its maximum (clockwise) means of a large feedthrough insulator
position. The bridge is balanced when the mounted on the back of the tuner cabinet.
input impedance of the tuner is 52 ohms This insulator is not visible in the photo-
resistive. This is the condition for maximum graphs.
energy transfer between transmission line
and antenna. The meter is graduated in arbi- Bridge The SWR bridge must be cali-
trary units, since actual SWR value is not Calibration brated for 52 -ohm service. This
required. can be done by temporarily dis-
Tuner
connecting the lead between the bridge and
Major parts placement in the the antenna tuner and connecting a 2 -watt,
Construction tuner is shown in figures 44 52 ohm carbon resistor to the junction of
and 46. Tapped coil L1 is R, and the negative terminal of the 1N56
mounted on Vi -inch ceramic insulators, and diode. The opposite lead of the carbon re-
all major components are mounted above
sistor is grounded to the chassis of the bridge.
deck with the exception of the SWR bridge A small amount of r -f energy is fed to the
(figure 47). The components of the bridge input of the bridge until a reading is ob-
are placed below deck, adjacent to the co- tained on the r -f voltmeter. The 25 -pf
axial input plug mounted on the rear apron bridge -balancing capacitor Co (see figure
of the chassis. The ten 10 -ohm resistors are
47) is then adjusted with a fiber -blade
soldered to two 1 -inch rings made of copper
wire as shown in the photograph. The bridge
screwdriver until a zero reading is obtained
capacitors are attached to this assembly with on the meter. The sensitivity control is ad-
extremely short leads. The 1N56 crystal vanced as the meter null grows, in order
mounts at right angles to the resistors to to obtain the exact point of bridge balance.
ensure minimum amount of capacitive cou- When this point is found, the carbon resistor
pling between the resistors and the detector. should be removed and the bridge attached
The output lead from the bridge passes to the antenna tuner. The bridge capacitor
through a ceramic feedthrough insulator to is sealed with a drop of nail polish to pre-
the top side of the chassis. vent misadjustment.
HANDBOOK Tuner for Center -Fed Systems 471
Figure 46
REAR VIEW OF
TUNER SHOWING
PLACEMENT OF
MAJOR
COMPONENTS
Rotary inductor is
driven by Johnson
116-208-4 counter dial.
Coaxial input recep-
tacle J, is mounted
directly below rotary
inductor.
Tuner All tuning adjustments are the meter will read zero. All loading adjust-
Adjustments made to obtain proper trans- ments may then be made with the transmit-
mitter loading with a bal- ter controls. The tuner should be readjusted
anced (zero- meter -reading) bridge condi- whenever the frequency of the transmitter
tion. The tuner is connected to the trans- is varied by an appreciable amount.
mitter through a random length of S2 -ohm
coaxial line, and the single-wire antenna is 21 -13 A Tuner for
attached to the output terminal of the
tuner. Transmitter loading controls are set Center -Fed Antenna
to approximate a S2 -ohm termination. The Systems
transmitter is turned on (preferably at
reduced input) and resonance is established Center -fed antennas require a balanced
in the amplifier tank circuit. The sensitivity antenna tuner to allow them to be used with
control of the tuner is adjusted to provide transmitters having unbalanced coaxial an-
near full -scale deflection on the bridge tenna terminations. Shown in this section is
meter. Various settings of S1, L_, and C, a simple and inexpensive antenna tuner
should be tried to obtain a reduction of (figure 48) which, when used in conjunc-
bridge reading. As tuner resonance is ap- tion with an SWR meter, will permit center -
proached, the meter reading will decrease fed antennas of practically any configura-
and the sensitivity control should be ad- tion to be used with modern coaxial- output
vanced. When the system is in resonance, transmitters.
472 Antennas and Antenna Matching THE RADIO
Tuner To conserve space, yet allow
Construction maximum circuit Q to be
achieved, the tuner is con-
structed in a wooden box measuring 13"
wide, 10" high, and 12" deep. A piece of
masonite is used for the panel. The two vari-
able capacitors are mounted to the panel,
as are the selector switch and the airwound
inductor. The inductor is spaced away from
the panel by two 2" long ceramic insulators.
The four- winding inductor is made from a
single section of coil stock, as shown in the
drawing. Adjustable taps are made at the
chosen coil turns by means of small phos-
phor- bronze clips attached to flexible in-
sulated leads. The various terminals are
mounted on a small aluminum plate which
is mounted in a cutout area in the rear of
the wood cabinet.
Figure 47
is
tions other than the desired one. Such cap-
radiation in certain desired directions and to
abilities involve the use of directive antenna
neutralize radiation in other directions, a
arrays.
directive antenna array is formed.
Few simple antennas, except the single
vertical element, radiate energy equally well The function of a directive antenna when
in all azimuth (horizontal or compass) used for transmitting is to give an increase
in signal strength in some direction at the
directions. All horizontal antennas, except
those specifically designed to give an omni- expense of radiation in other directions. For
directional azimuth radiation pattern such reception, one might find useful an antenna
as the turnstile, have some directive proper-
giving little or no gain in the direction from
which it is desired to receive signals if the
ties. These properties depend on the length
of the antenna in wavelengths, the height antenna is able to discriminate against inter-
above ground, and the slope of the radiator.
fering signals and static arriving from other
The various forms of the half -wave hor- directions. A good directive transmitting
izontal antenna produce maximum radiation antenna, however, can also be used to good
at right angles to the wire, but the direc- advantage for reception.
If radiation can be confined to a narrow
tional effect is not great. Nearby objects
beam, the signal intensity can be increased
also minimize the directivity of a dipole
a great many times in the desired direction
radiator, so that it hardly seems worth while of transmission. This is equivalent to in-
to go to the trouble to rotate a simple half - creasing the power output of the transmit-
wave dipole in an attempt to improve trans- ter. On the higher frequencies, it is more
mission and reception in any direction. economical to use a directive antenna than
The half -wave doublet, folded- dipole, to increase transmitter power, if more than
zepp, single- wire -fed, matched-impedance, a few watts of power is being used.
and Q- section antennas all have practically Directive antennas for the high- frequency
the same radiation pattern when properly range have been designed and used commer-
built and adjusted. They all are dipoles, and cially with gains as high as 23 db over a
the feeder system, if it does not radiate in simple dipole radiator. Gains as high as 3 f
Directive Antennas 475
.V-
IIii
DOUeIC HOP
- -_,
3o-
.
40.
30. 12
20' --
SINGLE HOP
W.
no
10
I
30
Mil
30 SOO 300 300 1000 3000 10000
GREAT CIRCLE DISTANCE IN MILES
Figure 1
Figure 3
DIRECTIVE GAIN OF
LONG -WIRE ANTENNAS
0
O a 1 4 S S 7 10
DB POWER RATIO OF MAIN LOBE TO A DIPOLE
Types of There is an enormous vari- broadside lobes get smaller and smaller in
Directive Arrays ety of directive antenna amplitude, even though numerous (fig-
arrays that can give a sub- ure 2).
stantial power gain in the desired direction The horizontal radiation pattern of such
of transmission or reception. However, some antennas depends on the vertical angle of
are more effective than others which require radiation being considered. If the wire is
the same space. In general it may be stated more than 4 wavelengths long, the maxi-
that long -wire antennas of various types, mum radiation at vertical angles of 15 to
such as the single long wire, the V beam, 20 (useful for DX) is in line with the
and the rhombic, are less effective for a wire, being slightly greater a few degrees
given space than arrays composed of reso- either side of the wire than directly off the
nant elements, but the long-wire arrays have ends. The directivity of the main lobes of
the significant advantage that they may be radiation is not particularly sharp, and the
used over a relatively large frequency range minor lobes fill in between the main lobes
while resonant arrays are usable only over a to permit working stations in nearly all
quite narrow frequency band. directions, though the power radiated broad-
side to the radiator will not be great if the
22 -2 Long -Wire Radiators radiator is more than a few wavelengths
long. The directive gain of long -wire an-
Harmonically operated long wires radiate tennas, in terms of the wire length in wave-
better in certain directions than others, but lengths is given in figure 3.
cannot be considered as having appreciable To maintain the out -of -phase condition
directivity unless several wavelengths long. in adjoining half -wave elements throughout
The current in adjoining half -wave elements the length of the radiator, it is necessary
flows in opposite directions at any instant, that a harmonic antenna be fed either at
and, thus, the radiation from the various one end or at a current loop. If fed at a
elements adds in certain directions and can- voltage loop, the adjacent sections will be
cels in others. fed in phase, and a different radiation pat-
A half -wave doublet in free space has a tern will result.
"doughnut" of radiation surrounding it. A The directivity of a long wire does not
full wave has 2 lobes, 3 half waves 3, etc. increase very much as the length is in-
When the radiator is made more than 4 half creased beyond about 15 wavelengths. This
wavelengths long, the end lobes (cones of is due to the fact that all long -wire anten-
radiation) begin to show noticeable power nas are adversely affected by the r -f resist-
gain over a half -wave doublet, while the ance of the wire, and because the current
www.americanradiohistory.com
HANDBOOK Long -Wire Radiators 477
Frequency
In MHz 1X 11/2a 2X 21/4X 3X 31/,X 4X 41/sX
amplitude begins to become unequal at dif- of the antenna. A ground is sometimes con-
ferent current loops as a result of attenua- nected to the center of the tuned coil.
tion along the wire caused by radiation and If desired, the antenna can be opened and
losses. As the length is increased, the tuning current -fed at a point of maximum current
of the antenna becomes quite broad. In by means of low -impedance ribbon line, or
fact, a long wire about I S waves long is by a quarter -wave matching section and
practically aperiodic, and works almost open line.
equally well over a wide range of frequen-
cies. 22 -3 The V Antenna
One of the most practical methods of
feeding a long -wire antenna is to bring one If two long -wire antennas are built in
end of it into the radio room for direct the form of a V, it is possible to make two
connection to a tuned antenna circuit which of the maximum lobes of one leg shoot in
is link- coupled through a harmonic- attenu- the same direction as two of the maximum
ating filter to the transmitter. The antenna lobes of the other leg of the V. The result-
can be tuned effectively to resonance for ing antenna is bidirectional (two opposite
operating on any harmonic by means of the directions) for the main lobes of radiation.
tuned circuit which is connected to the end Each side of the V can be made any odd or
1O
140
Figure 4
120
o
o a 4 so sa
LENGTH IN "L' WAVELENGTHS
www.americanradiohistory.com
478 High- Frequency Directive Antennas
Figure 5
TYPICAL V BEAM ANTENNA
IL)
m
ao
ew
4
\
GAIN IN DECIBELS OVER HALF WAVE
soiV!
fi:9111
111=111
Figure 6
DIRECTIVE GAIN OF A V BEAM
THE RADIO
_i
\1
A1
L.
&
-
1111111
the angle 8 are listed in the V- Antenna De- a somewhat smaller angle than that deter-
sign Table for various frequencies in the mined by the lobes. Optimum directivity
10 -, 15 -, 20- and 40 -meter amateur bands. for a one -wave V is obtained when the angle
Apex angles for all side lengths are given in is 90 rather than 180 , as determined by
figure 4. The gain of a V beam in terms of the the ground pattern alone.
Frequency L = X L = 2X L = 4X L = EX
In kHz 5= 90 5= 70 5= 52 5 = 39
www.americanradiohistory.com
HANDBOOK The Rhombic Antenna 479
above ground.
MMMMM 0M00000
22-4 The Rhombic Antenna
Termination nated with a resistance of a for the "maximum output" design; the shorter
lengths (I') are for the "alignment" method
value between 700 and 800 which gives approximately 1.5 db less gain
ohms the rear lobes are eliminated, the for- with a considerable reduction in the space
required for the antenna. The values of side
ward gain is increased, and the antenna can length, tilt angle, and height for a given
be used on several bands without changes. wave angle are obtained by drawing a ver-
The terminating resistance should be cap- tical line upward from the desired wave
angle.
able of dissipating one -third the power out-
put of the transmitter, and should have very
little reactance. For medium- or low -power consisting of No. 25 nichrome wire, spaced
transmitters, the noninductive plaque resis- 6 inches and terminated with 800 ohms,
tors will serve as a satisfactory termination. will serve satisfactorily. Because of the at-
Several manufacturers offer special resistors tenuation of the line, the lumped resistance
suitable for terminating a rhombic antenna. at the end of the line need dissipate but a
The terminating device should, for tech- few watts even when high power is used.
nical reasons, present a small amount of in- A half dozen 5000 -ohm 2 -watt carbon re-
ductive reactance at the point of termina- sistors in parallel will serve for all except
tion. very high power. The attenuating line may
A compromise terminating device com- be folded back on itself to take up less
monly used consists of a terminated 250 - room.
foot or longer length of line made of The determination of the best value of
resistance wire which does not hale too terminating resistor may be made while re-
much resistance per unit length. If the latter ceiving, if the input impedance of the
qualification is not met, the reactance of receiver is approximately 800 ohms. The
the line will be excessive. A 250 -foot line value of resistor which gives the best di-
www.americanradiohistory.com
480 High- Frequency Directive Antennas THE RADIO
31
Figure 8
TYPICAL RHOMBIC
ANTENNA DESIGN
The antenna system illus-
trated above may be used
over the frequency range from LINE TO TX
N 14 SPACED I-
7 to 29 MHz without change.
The directivity of the system SPACING BETWEEN SIDES S.21 FEET TERMINATING LINE
may be reversed by the sys- TOTAL LENGTH SO2 FEET OP 230' OF N 211
rectivity on reception will not give the above ground is increased in the same man-
most gain when transmitting, but there ner as with a dipole antenna. The rhombic
will be little difference between the two. should not be tilted in any plane. In other
The input resistance of the rhombic words, the poles should all be of the same
which is reflected into the transmission line height and the plane of the antenna should
that feeds it is always somewhat less than be parallel to the ground.
the terminating resistance, and is around A considerable amount of directivity is
700 to 750 ohms when the terminating re- lost when the terminating resistor is left off
sistor is 800 ohms. the end and the system is operated as a reso-
The antenna should be fed with a non - nant antenna. If it is desired to reverse the
resonant line having a characteristic imped- direction of the antenna it is much better
ance of 650 to 700 ohms. The four corners practice to run transmission lines to both
of the rhombic should be at least one -half ends of the antenna and then run the
wavelength above ground for the lowest terminating line to the operating position.
frequency of operation. For three -band op- Then with the aid of two dpdt switches it
eration the proper tilt angle (0) for the will be possible to connect either feeder to
center band should be observed. the antenna changeover switch and the
The rhombic antenna transmits a hori- other feeder to the terminating line, thus
zontally polarized wave at a relatively low reversing the direction of the array and
angle above the horizon. The angle of radia- maintaining the same termination for either
tion (wave angle) decreases as the height direction of operation.
Figure 9
z 3
A
10 II 12 IS 14 IS 111 17 IS M 20
www.americanradiohistory.com
HANDBOOK Stacked -Dipole Arrays 481
Lt 2 L2 Lt
le S -1
L3 La
f
/ PLANE OF WIRES w
I
END VIEW
QUARTER -WAVE STUBS
NON RESONANT
FEED LIN
GAIN APPRO% 4 s DB
O Figure 11
I
I THE FRANKLIN OR COLLINEAR
ANTENNA ARRAY
s,+S.
I IN PHASE
An antenna of this type, regardless of the
IRO. OUT OF PHASE
number of elements, attains all of its direc-
FLAT -TOP BEAN, ETC.) _- / (LAZY H, STERRA CURTAIN)
tivity through sharpening of the horizontal
or azimuth radiation pattern; no vertical di-
Figure 10 rectivity is provided. Hence a long antenna
of this type has an extremely sharp azimuth
RADIATION PATTERNS OF A PAIR OF pattern, but no vertical directivity.
DIPOLES OPERATING WITH IN -PHASE
EXCITATION, AND WITH EXCITATION
180 OUT OF PHASE 90 , and 135 for instance), the pattern is
unsymmetrical, the radiation being greater
If the dipoles are oriented horizontally most
of the directivity will be in the vertical to one direction than in the opposite direc-
plane; if they are oriented vertically most tion.
of the directivity will be in the horizontal With spacings of more than 0.8 wave-
plane.
length, more than two main lobes appear
for all phasing combinations; hence, such
Figure 7 gives curves for optimum-design spacings are seldom used.
rhombic attennas for both the maximum -
output method and the alignment method. In -Phase With the dipoles driven so as to
The alignment method is about 1.5 db down Spacing be in phase,the most effective
from the maximum output method but spacing is between 0.5 and 0.7
requires only about 0.74 as much leg length. wavelength. The latter provides greater
The height and tilt angle are the same in gain, but minor lobes are present which do
either case. Figure 8 gives construction data not appear at 0.5- wavelength spacing. The
for a recommended rhombic antenna for the radiation is broadside to the plane of the
7.0- through 29.7 -MHz bands. This antenna wires, and the gain is slightly greater than
will give about 11 db gain in the 14.0 -MHz can be obtained from two dipoles out of
band. The approximate gain of a rhombic phase. The gain falls off rapidly for spac-
antenna over a dipole (both above normal ings less than 0.375 wavelength, and there is
soil) is given in figure 9. little point in using spacing of 0.25 wave-
length or less with in -phase dipoles, except
where it is desirable to increase the radiation
22 -5 Stacked -Dipole Arrays resistance. (See Multiwire Doublet.)
The characteristics of a half -wave dipole Out-of-Phase When the dipoles are fed 180
already have been described. When another Spacing out of phase, the directivity is
dipole is placed in the vicinity and excited through the plane of the wires,
either directly or parasitically, the resultant and is greatest with close spacing, though
radiation pattern will depend on the spac- there is but little difference in the pattern
ing and phase differential, as well as the after the spacing is made less than 0.125
relative magnitude of the currents. With wavelength. The radiation resistance be-
spacings less than 0.65 wavelength, the radi- comes so low for spacings of less than 0.1
ation is mainly broadside to the two wires wavelength that such spacings are not prac-
(bidirectional) when there is no phase dif- tical.
ference, and through the wires (end fire) In the three foregoing examples, most of
when the wires are 180 out of phase. With the directivity provided is in a plane at a
phase difference between 0 and 180 (45 , right angle to the wires, though when out
www.americanradiohistory.com
482 High- Frequency Directive Antennas THE RADIO
COLLINEAR ANTENNA DESIGN CHART
600
FIMHZ) F1
Frequency
in MHz L1 L2 L3 FIIMHz)
28.5 16'8" 17' 8'6" A B
21.2 228 23,3" 11'6"
A- B ISOR FEED POINT GAINAFPROA. 3 O
14.2 33'8" 34,7 17'3"
7.15 67' 68'8" 34'4" Figure 12
4.0 120' 123' 61'6" DOUBLE EXTENDED ZEPP ANTENNA
3.6 133' 136'5" 68'2" For hest results, antenna should he tuned to
operating frequency by means of grid -dip
oscillator.
of phase, the directivity is in a line through
the wires, and when in phase, the directivity dipole. This consideration applies whether
is broadside to them. Thus, if the wires arc the elements are of normal length or are
oriented vertically, mostly horizontal di- extended.
rectivity will be provided. If the wires are The collinear antenna consists of two or
oriented horizontally, most of the directiv- more radiating sections from 0.5 to 0.65
ity obtained will be vertical directivity. wavelengths long, with the current in phase
To increase the sharpness of the directivity in each section. The necessary phase reversal
in all planes that include one of the wires, between sections is obtained through the
additional identical elements are added in use of resonant tuning stubs as illustrated in
the line of the wires, and fed so as to be in figure 11. The gain of a collinear array us-
phase. The familiar lazy -H array is one array ing half -wave elements (in decibels) is ap-
utilizing both types of directivity in the proximately equal to the number of ele-
manner prescribed. The two -section 8JK ments in the array. The exact figures are as
flat -top beam is another. follows:
These two antennas in their various forms
are directional in a horizontal plane, in addi-
Number of Elements 2 3 4 5 6
Gain in Decibels 1.8 3.3 4.5 5.3 6.2
tion to being low-angle radiators, and are
perhaps the most practical of the bidirec- As additional in -phase collinear elements
tional stacked- dipole arrays for amateur use. are added to a doublet, the radiation resist-
More phased elements can be used to pro- ance goes up much faster than when addi-
vide greater directivity in planes including tional half waves are added out of phase
one of the radiating elements. The H then (harmonic operated antenna).
becomes a Sterba- curtain array. For a collinear array of from 2 to 6 ele-
For unidirectional work the most prac- ments, the terminal radiation resistance in
tical stacked -dipole arrays for amateur - ohms at any current loop is approximately
band use are parasitically excited systems 100 times the number of elements.
using relatively close spacing between the It should be borne in mind that the gain
reflectors and the directors. Antennas of from a collinear antenna depends on the
this type are described in detail in a later
chapter. The next most practical unidirec-
tional array is an H or a Sterba curtain with
l'- iH,i + ---i-- l'MN-, -'I
a similar system placed approximately one -
quarter wave behind. The use of a reflector
system in conjunction with any type of
stacked -dipole broadside array will increase
the gain by 3 db.
Collinear The simple collinear antenna ar-
PHASE- REVERSING SWITCH
Arrays ray is a very effective radiating FOR CLOVERLEAF PATTERN
www.americanradiohistory.com
HANDBOOK Broadside Arrays 483
a,
--ee e -. -
eee
-eS,e'-
c
'- a'e--4
sr eos
a phase reversing switch, the two elements
may be operated out of phase, producing a
cloverleaf pattern with slightly less maxi-
3zw 32v- mum gain.
1 A three- element precut array for 40-
MANE STUBS a:4 4 e E. ---30011 RIBBON TO
meter operation is shown in figure 14. It is
TRANSMITTER, ANY LENGTH
W ERE, SPACED TO
fed directly with 300 -ohm ribbon line, and
GAIN APPROX. 3 De may be matched to a 52-ohm coaxial output
transmitter by means of a balun.
Figu e 14
www.americanradiohistory.com
HANDBOOK Broadside Arrays 485
NONRESONANT FEEDER
Ls L2 L3
Figure 16
L -+ L -+ f L -I L -
!0011 LINE GAIN APPROX. S De . NTS
I Or MAXI MUM
CURRENT
DIMENSION L
IOM. ISM. 120 /I014. ANTENNA f0
TRANSMITTER
s9- IS 17'S- TUNER
Figure 18
'
IOM. ISM. M
9 db.
.
L IsY O SECTION
12.3, I7- M14E. WIRE
SPACED
of the array.
The feed impedance of the Double Bruce I SOIL LINE TO TRANSMITTER
is about 750 ohms. The array may be fed
with a quarter -wave stub made of 300 -ohm
ribbon line and a feedline made of 150 -ohm
ribbon line. Alternatively, the array may be Figure 19
fed directly with a wide -spaced 600 -ohm
transmission line (figure 18) The feedline .
THE BI- SQUARE BROADSIDE ARRAY
should be brought away from the Double This bidirectional array is related to the
Lazy N, and in spite of the oblique elements,
Bruce for a short distance before it drops is horizontally polarized. It has slightly less
downward, to prevent interaction between gain and directivity than the Lazy N, the
the feedline and the lower part of the center free -space directivity gain being approxi-
mately 4 db. Its chief advantage is the fact
phasing section of the array. For best re- that only a single pole is required for sup-
sults, the bottom sections of the array should port, and two such arrays may be supported
from a single pole without interaction if the
be one -half wavelength above ground. planes of the elements are at right angles.
Arrays such as the X -array and the A 600-ohm line may be substituted for the
Double Bruce are essentially high -impedance twin lead, and either operated as a resonant
line, or made sonant by the incorpora-
devices, and exhibit relatively broadband tion of a matching stub.
characteristics. They are less critical of ad-
justment than a parasitic array, and they
work well over a wide frequency range such lessthan that of a Lazy H, but is still
as is encountered on the 28- to 29.7 -MHz
worthwhile, being approximately 4 db over
a half -wave horizontal dipole at the same
band.
average elevation.
The Bi-Square Illustrated in figure 19 is a When two Bi- Square arrays are suspended
Broadside Array simple method of feeding a at right angles to each other (for general
small broadside array. As coverage) from a single pole, the 0-sections
two arrays of this type can be supported at should be well separated or else symmetrical-
right angles from a single pole without in- ly arranged in the form of a square (the
teraction, it offers a solution to the problem diagonal conductors forming one Q- section)
of suspending two arrays in a restricted in order to minimize coupling between
space with a minimum of erection work. them. The same applies to the line if open
The free space directivity gain is slightly construction is used instead of twin lead, but
www.americanradiohistory.com
HANDBOOK Cubical Quad 487
L
may be placed 0.15 wavelength behind this
The Bobtail Another application of
antenna to provide an over -all gain of 8.5
Bidirectional vertical orientation of the
db. The reflector may be tuned by means of
Broadside Curtain radiating elements of an
a quarter -wave stub which has a movable
shorting bar at the bottom end. array in order to obtain
oa 3 D2
flat -top portion. This is not sufficient to
affect the power gain appreciably, but does
degrade the discrimination somewhat.
END LINK COIL TO A moderate amount of sag can be tol-
TUNE FREQUENCY
C=IOOpr
erated at the center of the flat top, where it
DIMENSIONS connects to the driven vertical element. The
40M.
524 6O M.
poles and antenna tank should be so located
COAXIAL LINE DI 46'4- 126
D2 55' 60' with respect to each other that the driven
D3 50.7050' 54.7066 vertical element drops approximately straight
down from the flat fop.
Figure 22
Normally the antenna tank will be lo-
BOBTAIL BIDIRECTIONAL BROADSIDE cated in the same room as the transmitter,
CURTAIN FOR THE 7 -MHz OR THE to facilitate adjustment when changing
4.0 -MHz AMATEUR BANDS frequency. In this case it is recommended
that the link- coupled tank be located across
This simple vertically polarized array pro-
vides low angle radiation and response with the room from the transmitter if much
comparatively low pole heights, and is very power is used, in order to minimize r-f
effective for DX work on the 7 -MHz band or feedback difficulties which might occur as
the 4.0 -MHz phone band. Because of the
phase relationships, radiation from the hori- a result of the asymmetrical high- impedance
zontal portion of the antenna is effectively feed. If tuning of the antenna tank from
suppressed. Very little current flows in the
ground lead to the coupling tank; so an elab- the transmitter position is desired, flexible
orate ground system is not required, and the shafting can be run from the antenna tank
length of the ground lead is not critical so capacitor to a control knob at the trans-
long as it uses heavy wire and is reasonably
short. mitter.
The lower end of the driven element is
quite "hot" if much power is used, and
low -angle radiation at the lower end of the the lead -in insulator should be chosen with
high- frequency range with low pole heights this in mind. The ground connection need
is illustrated in figure 22. When precut to not have very low resistance, as the current
the specified dimensions this single -pattern flowing in the ground connection is com-
array will perform well over the 7 -MHz paratively small. A stake or pipe driven a
amateur band or the 4 -MHz amateur phone few feet in the ground will suffice. How-
band. For the 4 -MHz band, the required ever, the ground lead should be of heavy
two poles need be only 70 feet high, and wire and preferably the length should not
the array will provide a practical signal gain exceed about 10 feet at 7 MHz or about 20
averaging from 7 to 10 db over a horizontal feet at 4 MHz in order to minimize reactive
half -wave dipole utilizing the same pole effects due to its inductance. If it is impos-
height when the path length exceeds 2500 sible to obtain this short a ground lead, a
miles. piece of screen or metal sheet about four
The horizontal directivity is only moder- feet square may be placed parallel to the
ate, the beam width at the half-power points earth in a convenient location and used as
being slightly greater than that obtained an artificial ground. A fairly high C/L ratio
from three cophased vertical radiators fed ordinarily will be required in the antenna
with equal currents. This is explained by the tank in order to obtain adequate coupling
fact that the current in each of the two and loading.
outer radiators of this array carries only
about half as much current as the center - 22 -7 End -Fire Directivity
driven element. While this binomial current
distribution suppresses the end -fire lobe that By spacing two half -wave dipoles, or col-
occurs when an odd number of parallel ra- linear arrays, at a distance of from 0.1 to
diators with half-wave spacing are fed equal 0.25 wavelength and driving the two 180
www.americanradiohistory.com
HANDBOOK End -Fire Directivity 489
out of phase, directivity is obtained through feeders. Proper stub adjustment is covered
the two wires at right angles to them. in Chapter Twenty -four. Suitable radiator
Hence, this type of bidirectional array is lengths and approximate stub dimensions are
called end -fire. A better idea of end -fire di- given in the accompanying design table.
rectivity can be obtained by referring to Figure 23 shows top views of eight types
figure 10. of flat -top beam antennas. The dimensions
Remember that end-fire refers to the rad- for using these antennas on different bands
iation with respect to the two wires in the are given in the design table. The 7- and
array rather than with respect to the array 28 -MHz bands are divided into two parts,
as a whole. but the dimensions for either the low- or
The vertical directivity of an end -fire bi- high- frequency ends of these bands will be
directional array which is oriented horizon- satisfactory for use over the entire band.
tally can be increased by placing a similar In any case, the antennas are tuned to the
end -fire array a half wave below it, and ex- frequency used, by adjusting the shorting
cited in the same phase. Such an array is a wire on the stub, or tuning the feeders, if
combination broadside and end -fire affair. no stub is used. The data in the table may
be extended to other bands or frequencies
8JK Flat -Top A very effective bidirectional by applying the proper factor. Thus, for
Beam end -fire array is the 8JK 50- to 52 -MHz operation, the values for 28
Flat-Top Beam. Essentially, to 29 Mc. are divided by 1.8.
this antenna consists of two close- spaced All of the antennas have a bidirectional
dipoles or collinear arrays. Because of the horizontal pattern on their fundamental
close spacing, it is possible to obtain the frequency. The maximum signal is broad-
proper phase relationships in multisection side to the flat top. The single -section type
flat tops by crossing the wires at the voltage has this pattern on both its fundamental
loops, rather than by resorting to phasing frequency and second harmonic. The other
stubs. This greatly simplifies the array. (See types have four main lobes of radiation on
figure 23.) Any number of sections may be the second and higher harmonics. The nom-
used, though the one- and two -section ar- inal gains of the different types over a half -
rangements are the most popular. Little wave comparison antenna are as follows:
extra gain is obtained by using more than single- section, 4 db; two -section, 6 db; four -
four sections, and trouble from phase shift section, 8 db.
may appear. The maximum spacings given make the
A center -fed single-section flat -top beam beams less critical in their adjustments. Up
cut according to the table, can be used quite to one -quarter wave spacing may be used on
successfully on its second harmonic, the pat- the fundamental for the one -section types
tern being similar except that it is a little and also the two -section center -fed, but it is
sharper. The single- section array can also not desirable to use more than 0.15 wave-
be used on its fourth harmonic with some length spacing for the other types.
success, though there then will be four Although the center -fed type of flat -top
cloverleaf lobes, much the same as with a generally is to be preferred because of its
full -wave antenna. symmetry, the end -fed type often is con-
If a flat -top beam is to be used on more venient or desirable. For example, when a
than one band, tuned feeders are necessary. flat -top beam is used vertically, feeding
The radiation resistance of a flat -top beam from the lower end is in most cases more
is rather low, especially when only one sec- convenient.
tion is used. This means that the voltage If a multisection flat -top array is end -fed
will be high at the voltage loops. For this instead of center -fed, and tuned feeders are
reason, especially good insulators should be used, stations off the ends of the array can
used for best results in wet weather. be worked by tying the feeders together and
The exact lengths for the radiating ele- working the whole affair, feeders and all, as
ments are not especially critical, because a long -wire harmonic antenna. A single -pole
slight deviations from the correct lengths double -throw switch can be used for chang-
can be compensated in the stub or tuned ing the feeders and directivity.
490 High- Frequency Directive Antennas THE RADIO
CENTER FED TO CENTER
Or rLAT TOP
1-SECTION I - SECTION
2- SECTION
3- SECTION
4- SECTION
1
L3-A-ht- -La -tp1 L3 -.401 V--- L D
Spec-
,... x
a
...
- x
X
o
x
O
x
O
Frequency ing
Q a g Q
S L1 Li 13 14 M D Al N
7.2-7.2 MHz X/8 17'4" 34' 60' 52'8" 44' 8'10" 4' 26' 60' 96' 4'
7.2-7.3 X/8 17'0" 33'6" 59' 51'8" 43'1" 8'8" 4' 26' 59' 94' 4'
14.0-14.35 X/8 8'8" 17' 30' 26'4" 22' 4'5" 2' 13' 30' 48' 2'
14.0-14.35 .15X 10'5" 17' 30' 25'3" 20' 5'4" 2' 12' 29' 47' 2'
14.0-14.4 .20X 13'11" 17' 30' 22'10" 7'2" 2' 10' 27' 45' 3'
14.0-14.4 X/4 17'4" 7' 30' 20'8" 8'10" 2' 8' 25' 43' 4'
28.0-29.0 .15X 5'2" 8'6" 15' 12'7" 10' 2'8" 1'6" 7' 15' 24' 1'
28.0-29.0 X/4 8'8" 8'6" 15' 10'4" 4'5" 1'6" 5' 13' 22' 2'
28.7- 29.7 .15a 5'0" 8'3" 14'6" 12'2" 9'8" 2'7" 1'6" 7' 15' 23' 1'
28.7- 29.7 X/4 8'4" 8'3" 14'6" 10'0" 4'4" 1'6" 5' 13' 21' 2'
Dimension chart for flat-top beam antennas. The meanings of the symbols are as fol ows:
L , the lengths of the sides of the flap -top sections as shown. L, is length of the
sides of single- section center -fed, L, single- section end-fed and 2- section center fed, L 4-
section center -fed and end-sections of 4- section end -fed, and L middle sections of 4- section
end -fed.
S, the spacing between the Rat -top wires.
M, the wire length from the outside to the center of each cross -over.
D, the spacing lengthwise between sections.
A (I: j), the approximate length for a quarter -wave stub.
A (I/z), the approximate length for a half -wave stub.
A (3/4), the approximate length for a three -quarter wave stub.
X, the approximate distance above the shorting wire of the stub for the connection of a
600 -ohm line. This distance, as given in the table, is approximately correct only for 2- section
flat -tops.
For single- section types it will be smaller and for 3- and 4- section types it will be larger.
The lengths given for a half -wave stub are applicable only to single -section center -fed
flat -tops. To be certain of sufficient stub length, it is advisable to make the stub a foot or
so longer than shown in the table, especially with the end-fed types. The lengths, A, are
measured from the point where the stub connects to the flat -top.
Both the center and end -fed types may be used horizontally. However, where a vertical
antenna is desired, the flat -tops can be turned on end. In this case, the end-fed types may
be more convenient, feeding from the lower end.
www.americanradiohistory.com
HANDBOOK Combination Arrays 491
i
Figure 24
DOM--
-\
THE TRIPLEX FLAT -TOP BEAM
ANTENNA FOR 10, 15, AND 20
METERS
MAX. RADIATION
4.S DE
1I
MAXIMUM
RADIATION
4.5 DS
1000 LIN! TO
TRANSMITTER
ANY LENCTH
DIMENSIONS
IOM. 15M. 20M MATERIAL
L 1!'r 71'S' 7!'7' WR[S,SAC[b7
S S'Y T'S II'
D T'! ID'S' 14'4- 300-OHM RIE/ON
The Triplex The Triplex beam is a modified while maintaining a bidirectional pattern
Beam version of the flat -top antenna the individual arrays must be spaced suffi-
which uses folded dipoles for the ciently to reduce the mutual impedances to
half wave elements of the array. The use of a negligible value.
folded dipoles results in higher radiation When two flat -top beams, for instance,
resistance of the array, and a high over -all are placed one above the other or end to
system performance. Three wire dipoles are end, a center spacing on the order of one
used for the elements, and 300 -ohm twin - wavelength is required in order to achieve
lead is used for the two phasing sections. A a worthwhile increase in gain, or approxi-
recommended assembly for Triplex beams mately 3 db.
for 28, 21, and 14 MHz is shown in figure Thus it is seen that, while maximum gain
24. The gain of a Triplex beam is about 4.5 occurs with two stacked dipoles at a spacing
db over a dipole. of about 0.7 wavelength and the space direc-
tivity gain is approximately db over one S
siderable spacing is required in order to under a half wavelength, and while the
realize a gain approaching 3 db (the re- proper phase relationship is 180 degrees re-
quired spacing being a function of the size gardless of the spacing for all spacings not
of the curtains) . While it is true that a exceeding one half wavelength, the situation
space- directivity gain of approximately 4 db is different in the case of two curtains
can be obtained by placing one half -wave placed in end -fire relationship to give a bi-
dipole an eighth wavelength ahead of an- directional pattern. For maximum gain at
other and feeding them 180 degrees out of zero wave angle, the curtains should be
phase, a gain of less than 1 db is obtained spaced an odd multiple of one -half wave-
when the same procedure is applied to two length and driven so as to be 180 degrees
large broadside curtains. To obtain a gain of out of phase, or spaced an even multiple of
approximately 3 db and retain a bidirectional one half wavelength and driven in the same
pattern, a spacing of many wavelengths is phase. The optimum spacing and phase rela-
required between two large curtains placed tionship will depend on the directivity
one ahead of the other. pattern of the individual curtains used
A different situation exists, however, alone, and as previously noted the optimum
when one driven curtain is placed ahead of spacing increases with the size and directiv-
an identical one and the two are phased so ity of the component arrays.
as to give a unidirectional pattern. Whin a A concrete example of a combination
unidirectional pattern is obtained, the gain broadside and end -fire array is two Lazy H
over one curtain will be approximately 3 arrays spaced along the direction of maxi-
db regardless of the spacing. For instance, mum radiation by a distance of four wave-
two large curtains one placed a quarter lengths and fed in phase. The space- directiv-
wavelength ahead of the other may have a ity gain of such an arrangement is slightly
space- directivity gain of only 0.5 db over less than 9 db. However, approximately the
one curtain when the two are driven 180 same gain can be obtained by juxtaposing
degrees out of phase to give a bidirectional the two arrays side by side or one over the
pattern (the type of pattern obtained with other in the same plane, so that the two
a single curtain). However, if they are combine to produce, in effect, one broad-
driven in phase quadrature (and with equal side curtain of twice the area. It is obvious
currents) the gain is approximately 3 db. that in most cases it will be more expedient
The directivity gain of a composite array to increase the area of a broadside array
also can be explained on the basis of the than to resort to a combination of end -fire
directivity patterns of the component arrays and broadside directivity.
alone, but it entails a rather complicated One exception, of course, is where two
picture. It is sufficient for the purpose of curtains are fed in phase quadrature to ob-
this discussion to generalize and simplify by tain a unidirectional pattern and space -
saying that the greater the directivity of directivity gain of approximately 3 db with
an end -fire array, the farther an identical ar- a spacing between curtains as small as one
ray must be spaced from it in broadside re- quarter wavelength. Another exception is
lationship to obtain optimum performance; where very low angle radiation is desired
and the greater the directivity of a broad- and the maximum pole height is strictly
side array, the farther an identical array limited. The two aforementioned Lazy H
must be spaced from it in end -fire relation- arrays when placed in endfire relationship
ship to obtain optimum performance and re- will have a considerably lower radiation
tain the bidirectional characteristic. angle than when placed side by side if the
It is important to note that while a bi- array elevation is low, and therefore may
directional end -fire pattern is obtained with under some conditions exhibit appreciably
two driven dipoles when spaced anything practical signal gain.
www.americanradiohistory.com
CHAPTER TWENTY -THREE
The very- high -frequency or vhf fre- 16 db will make a 25 -watt transmitter
quency range is defined as that range falling sound like a kilowatt at the other station.
between 30 and 300 MHz. The ultrahigh- Even a much simpler and smaller three- or
frequency or uhf range is defined as falling four- element parasitic array having a gain
between 300 and 3000 MHz. This chapter of 7 to 10 db will produce a marked im-
will be devoted to the design and construc- provement in the received signal at the other
tion of antenna systems for operation on station.
the amateur 50 -, 144 -, 235 -, and 420 -MHz However, as all vhf and uhf workers
bands. Although the basic principles of an- know, the most important contribution of a
tenna operation are the same for all fre- high -gain antenna array is in reception. If a
quencies, the shorter physical length of a remote station cannot be heard it obviously
wave in this frequency range and the differ- is impossible to make contact. The limiting
ing modes of signal propagation make it factor in vhf and uhf reception is in almost
possible and expedient to use antenna sys- every case the noise generated within the re-
tems different in design from those used in ceiver itself. Atmospheric noise is almost
the range from 3 to 30 MHz. nonexistent and ignition interference can
almost invariably be reduced to a satisfac-
tory level through the use of an effective
23 -1 Antenna noise limiter. Even with a grounded -grid or
Requirements neutralized triode first stage in the receiver,
the noise contribution of the first tuned
Any type of antenna system usable on the circuit in the receiver will be relatively
lower frequencies may be used in the vhf large. Hence it is desirable to use an antenna
and uhf bands. In fact, simple nondirective system which will deliver the greatest signal
half -wave or quarter -wave vertical antennas voltage to the first tuned circuit for a given
are very popular for general transmission field strength at the receiving location.
and reception from all directions, especially Since the field intensity being produced at
for short -range work. But for serious vhf the receiving location by a remote transmit-
or uhf work the use of some sort of direc- ting station may be assumed to be constant,
tional antenna array is a necessity. In the the receiving antenna which intercepts the
first place, when the transmitter power is greatest amount of wave front (assuming
concentrated into a narrow beam the appar- that the polarization and directivity of the
ent transmitter power at the receiving sta- receiving antenna is proper, will be the an-
tion is increased many times. A "billboard" tenna which gives the best received signal -
array or a Sterba curtain having a gain of to -noise ratio. An antenna which has two
494 VHF and UHF Antennas THE RADIO
square wavelengths of effective area will pick is wise to use the very minimum length of
up twice as much signal power as one which transmission line possible since transmission -
has one square wavelength area, assuming line losses at frequencies above about 100
the same general type of antenna and that MHz mount very rapidly.
both are directed at the station being re- Open lines should preferably be spaced
ceived. Many instances have been reported closer than is common for longer wave-
where a frequency band sounded completely lengths, since a few inches are an appreciable
dead with a simple dipole receiving antenna fraction of a wavelength at 2 meters. Radia-
but when the receiver was switched to a tion from the line will be greatly reduced if
three -element or larger array a considerable 1 -inch or 11/2-inch spacing is used, rather
amount of activity from 80 to 160 miles than the wider spacing used in the uhf
distant was heard. region.
Ordinary TV -type 300 -ohm ribbon or the
Angle of The useful portion of the signal new coaxial foamflex line may be used on
Radiation in the vhf and uhf range for the 2 -meter band for feeder lengths of about
short- or medium- distance com- 50 feet or less. For longer runs, either the
munication is that which is radiated at a uhf or vhf TV open -wire lines may be used
very low angle with respect to the surface with good over -all efficiency. The vhf line
of the earth; essentially it is that signal is satisfactory for use on the amateur 420 -
which is radiated parallel to the surface of MHz band.
the earth. A vertical antenna transmits a
Antenna It is recommended that the
portion of its radiation at a very low angle
Changeover same antenna be used for trans-
and is effective for this reason; its radiation
is not necessarily effective simply because it mitting and receiving in the vhf
is vertically polarized. A simple horizontal and uhf range. An ever-present problem in
dipole radiates very litle low -angle energy this connection, however, is the antenna
and hence is not a satisfactory vhf or uhf changeover relay. Reflections at the antenna
radiator. Directive arrays which concentrate changeover relay become of increasing im-
a major portion of the radiated signal at a portance as the frequency of transmission is
low radiation angle will prove to be effective increased. When coaxial cable is used as the
radiators whether their signal is horizontal- antenna transmission line, satisfactory co-
ly or vertically polarized. axial antenna changeover relays with low
In all cases, the radiating system for vhf reflection can be used.
and uhf work should be as high and as much On the 235- and 420 -MHz amateur
in the clear as possible. Increasing the height bands, the size of the antenna array becomes
of the antenna system will produce a very quite small, and it is practical to mount two
marked improvement in the number and identical antennas side by side. One of these
strength of the signals heard, regardless of antennas is used for the transmitter, and the
the actual type of antenna used. other antenna for the receiver. Separate
transmission lines are used, and the antenna
Transmission
relay may be eliminated.
Transmission lines to vhf and
Lines uhf antenna systems may be Effect of Feed A vertical radiator for
either of the parallel- conduc- System on Radiation general- coverage uhf use
tor or coaxial- conductor type. Coaxial line Angle should be made either
is recommended for short runs and closely %- or 1/2-wavelength
spaced open wire line for longer runs. long. Longer vertical antennas do not have
Waveguides may be used under certain their maximum radiation at right angles to
conditions for frequencies greater than the line of the radiator (unless co- phased),
perhaps 1500 MHz but their dimensions and, therefore, are not practical for use
become excessively great for frequencies where greatest possible radiation parallel to
much below this value. Nonresonant trans- the earth is desired.
mission lines will be found to be con- Unfortunately, a feed system which is not
siderably more efficient on these frequen- perfectly balanced and does some radiating,
cies than those of the resonant type. It not only robs the antenna itself of that
www.americanradiohistory.com
HANDBOOK Antenna Requirements 495
VECTOR SUM OF
Z PATTERNS
Figure 1
144 -MHz bands almost invariably use hor- angles are not recommended for vhf and
izontal polarization. uhf work. It is for this reason that the
Experience has shown that there is a great horizontal dipole and horizontally disposed
attenuation in signal strength when using collinear arrays are generally unsuitable for
crossed polarization (transmitting antenna work on these frequencies. Arrays using
with one polarization and receiving antenna broadside or end -fire elements do concen-
with the other) for all normal ground -wave trate radiation at low elevation angles and
contacts on these bands. When contacts are are recommended for vhf work. Arrays such
being made through sporadic -E reflection, as the lazy H, Sterba curtain, flat -top beam,
however, the use of crossed polarization and arrays with parasitically excited elements
seems to make no discernible difference in are recommended for this work. Dimensions
signal strength. So the operator of a station for the first three types of arrays may be
doing vhf work (particularly on the 50- determined from the data given in the pre-
MHz band) is faced with a problem: If vious chapter, and reference may be made
contacts are to be made with all stations to the Table of Wavelengths given in this
doing work on the same band, provision chapter.
must be made for operation on both hori- Arrays using vertically stacked horizontal
zontal and vertical polarization. This prob- dipoles, such as are used by commercial
lem has been solved in many cases through television and f-m stations, are capable of
the construction of an antenna array that giving high gain without a sharp horizontal
may be revolved in the plane of polarization radiation pattern. If sets of crossed dipoles,
in addition to being capable of rotation in as shown in figure IA, are fed 90 out of
the azimuth plane. phase the resulting system is called a turn-
An alternate solution to the problem stile antenna. The 90 phase difference be-
which involves less mechanical construction tween sets of dipoles may be obtained by
is simply to install a good ground -plane feeding one set of dipoles with a feed line
vertical antenna for all vertically -polarized which is one -quarter wave longer than the
work, and then to use a multielement hori- feed line to the other set of dipoles. The
zontally polarized array for DX work. field strength broadside to one of the dipoles
is equal to the field from that dipole alone.
23 -2 Simple Horizontally The field strength at a point at any other
Polarized Antennas angle is equal to the vector sum of the fields
from the two dipoles at that angle. A near-
Antenna systems which do not concen- ly circular horizontal pattern is produced
trate radiation at the very low elevation by this antenna.
www.americanradiohistory.com
HANDBOOK Antenna Polarization 497
www.americanradiohistory.com
HANDBOOK Discone Antenna 499
0.7 D
Figure 4
gain at zero angular elevation, but the re- THE DISCONE BROADBAND
duction is very slight. RADIATOR
Below the frequency at which the slant
height of the conical skirt is equal to a f ree- This antenna system radiates a vertically
polarized wave over a very wide frequency
space quarter wavelength the standing -wave range. The "disc" may be made of solid
ratio starts to climb, and below a frequency metal sheet, a group of radials, or wire
screen; the "cone" may best be constructed
approximately 20 percent lower than this by forming a sheet of thin aluminum. A single
the standing -wave ratio climbs very rapidly. antenna may be used for operation on the
This is termed the cutoff frequency of the 50 -, 144 -, and 220 -MHz amateur bands. The
dimension D is determined by the lowest fre-
antenna. By making the slant height approx- quency to be employed, and is given in
imately equal to a free -space quarter wave- Chart 1.
TH RADIO
_''l
500 VHF and UHF Antennas E
--_-=I
When using circularly polarized antennas
a
DMO
180
140
120
-_'MII MIN
MM Ela
MIN11111=M
IMMW1MIMIIItI
MO
.1
at both ends of the circuit, however, both
must be left handed or both must be right
handed. This offers some interesting possibil-
ities with regard to reduction of interference.
At the time of writing, there has been no
standardization of the "twist" for general
eo-__ I.11.MI\
;
amateur work.
0 11 Perhaps the simplest antenna configura-
70
On tion for a directional beam antenna having
.1
0S 10 13 2 2.5 3
II
4 3
circular polarization is the helical beam
which consists simply of a helix working
D IN FEET against a ground plane and fed with coaxial
line. In the uhf and the upper vhf range
CHART 1
the physical dimensions are sufficiently small
DESIGN CHART FOR THE DISCONE to permit construction of a rotatable struc-
ANTENNA ture without much difficulty.
When the dimensions are optimized, the
mast and to the outer conductor of the characteristics of the helical beam antenna
coaxial line. The line is run down through are such as to qualify it as a broadband an-
the supporting mast. An alternative arrange- tenna. An optimized helical beam shows
ment, one suitable for certain mobile appli- little variation in the pattern of the main
cations, is to fasten the base of the skirt lobe and a fairly uniform feed -point im-
directly to an effective ground plane such pedance averaging approximately 125 ohms
as the top of an automobile. over a frequency range of as much as 1.7
www.americanradiohistory.com
HANDBOOK Corner -Reflector Antenna 503
450-ONM TV LINE
OA UHF HORN ANTENNA
ANGLE BETWEEN
SIDES OF HORN -60
Figure 9
OB VHF HORIZONTALLY POLARIZED HORN
THE 60 HORN ANTENNA FOR USE ON
Figure 8
FREQUENCIES ABOVE 144 MHz
TWO TYPES OF HORN ANTENNAS
The "two -sided horn" of Figure 8e may be compared to a half -wave dipole at frequen-
fed by means of an open -wire transmission cies higher than cutoff is:
line.
I-
0
O s0 0 = TILT ANGLE R,, R2 = 390 OHMS EACH
NON INDUCTIVE
2A 45 65 eA toA
SIDE LENGTH, S h L -.l
Figure 10 Figure 11
www.americanradiohistory.com
HANDBOOK VHF Beam Antennas 505
-
A- DRIVEN ELEMENT
suitable termination consists of a "lossy" 'LONG
SEND RADIUS
line consisting of stainless -steel wire (cor-
responding to No. 24 or 26 gauge) spaced Is' FEED LIN
THRU NOL
2 inches, which in turn is terminated by two D- DIRECTOR DETAIL
390 -ohm 2 -watt carbon resistors. The dissi- 3s' LONG
pative line should be at least 6 wavelengths I X2 WELL -SEASONED
WOOD 34' LONG
long. GAIN 7.506 (20' LONG FOR I} METERS)
FLATTEN ENDS OF
TUBING AND DRILL
23 -8 Multielement VHF FOR 6/32 SCREWS
Beam Antennas 75 OM
TV LUE
REFLECTOR
40
RING BOLT
- 48" BOOM
60-
RG -59
EACH
BOOM
/U CABLES
40 LONG
MAINBOOMS
ELEMENTS -..
- <I
RG -8/U CABLE
SHAPE ENDS OF SHORT PIECES '
TO T. COAXIAL
TO FIT CONTOUR FITTING
AS SHOWN, ANTENNA IS
HORIZONTALLY POLARIZED-
CONTROL CORDS
RADIAL BEARING
Figure 13
CONSTRUCTION DRAWING OF AN EIGHT -ELEMENT TILTABLE 144 -MHz ARRAY
line. The complete antenna is light enough An 8- Element Figures 13 and 14 illus-
to be turned by a TV rotator. "Tiltable" Array trate an 8- element rotary
A simple Yagi antenna of this type will for 144 MHz array for use on the 144 -
provide a gain of 7 db over the entire 2- MHz amateur band. This
meter or 11/4-meter band, and is highly rec- array can be tilted to obtain either hori-
ommended as an "easy -to- build" beam in- zontal or vertical polarization. It is necessary
volving very little expense for the novice or that the transmitting and receiving station
beginner. use the same polarization for the ground-
www.americanradiohistory.com
HANDBOOK VHF Beam Antennas 507
Figure 14
wave signal propagation which is character- is sharper when the antenna system is ori-
istic of this frequency range. Although ented for horizontal polarization.
polarization has been loosely standardized in The changeover from one polarization to
various areas of the country, exceptions are the other is accomplished simply by pulling
frequent enough so that it is desirable that on the appropriate cord. Hence, the operation
the polarization of antenna radiation be is based on the offset head sketched in
easily changeable from horizontal to verti- figure 13. Although a wood mast has been
cal. used, the same system may be used with a
The antenna illustrated has shown a signal pipe mast.
gain of about 11 db, representing a power The 40 -inch lengths of RG -59 /U cable
gain of about 13. Although the signal electrically ys- wavelength) running from
gain of the antenna is the same whether it is the center of each folded dipole driven ele-
oriented for vertical or horizontal polariza- ment to the coaxial T- junction allow enough
tion, the horizontal beam width is smaller slack to permit free movement of the main
when the antenna is oriented for vertical boom when changing polarity. Type RG-
polarization. Conversely, the vertical pattern 8 U cable is run from the T- junction to
508 VHF and UHF antennas THE RADIO
!
POLYTHENE
' DIPOLE
Construction of
the Array
Most of the constructional
aspects of the antenna array
are self- evident from figure
13. However, the pointers given in the fol-
S'4
----
MI
lowing paragraphs will be of assistance to
those wishing to reproduce the array. TAW
The drilling of holes for the small ele- 3' 4f
ments should be done carefully on accurate- 14.91!
ly marked centers. A small angular error in
the drilling of these holes will result in a
considerable misalignment of the elements 3'4}. CROSS
WIVES
after the array is assembled. The same con-
sideration is true of the filing out of the
rounded notches in the ends of the main
boom for the fitting of the two-antenna
booms. 3'4i DIA.
Short lengths of wood dowel are used
freely in the construction of the array. The
ends of the small elements are plugged with
an inch or so of dowel, and the ends of the
y s'4
ALL JOINTS WELDED
antenna booms are similarly treated with
larger discs pressed into place.
Figure 15
The ends of the folded dipoles are made
in the following manner: Drive a length of DETAIL OF LAYOUT AND DIMENSIONS
dowel into the short connecting lengths of OF SCREEN -BEAM
aluminum tubing. Then drill down the cen-
ter of the dowel with a clearance hole for the
connecting screw. Then shape the ends of The main booms were made from 3/4-inch
the connecting pieces to fit the sides of the aluminum electrical conduit. Any size of
element ends. After assembly the junctions small tubing will serve for making the
may be dressed with a file and sandpaper elements. Note that the main boom is
until a smooth fit is obtained. mounted at the balance center and not nec-
The mast used for supporting the array is essarily at the physical center. The pivot
a 30 -foot spliced 2 by 2. A large discarded bolt in the offset head should be tightened
ball bearing is used as the radial load bearing sufficiently so that there will be adequate
and guy -wire termination. Enough of the friction to hold the array in position. Then
upper -mast corners were removed with a an additional nut should be placed on the
drawknife to permit sliding the ball bearing pivot bolt as a lock.
down about 9 feet from the top of the In connecting the phasing sections be-
mast. The bearing then was encircled by an tween the T-junction and the centers of the
assembly of three pieces of durai ribbon to folded dipoles, it is important that the cen-
form a clamp, with ears for tightening ter conductors of the phasing sections be
screws and attachment of the guy wires. connected to the same side of the driven
The bearing then was greased and covered elements of the antennas. In other words,
with a piece of auto inner tube to serve as when the antenna is oriented for horizontal
protection from the weather. Another junk - polarization and the center of the coaxial
box bearing was used at the bottom of the section goes to the left side of the top an-
mast as a thrust bearing. tenna, the center conductor of the other
www.americanradiohistory.com
HANDBOOK VHF Beam Antennas 509
TRANSFORMER - B
SNORTING BAR -C
I
C, & C2 = 5ORF
WATERT GAT
I
COMPARTMENT
APPROX.
BALUN - D
F SNORTING BAR -
COPPER TUBING
72 COAX CABLE
Figure 18
Figure 17
HORIZONTAL RADIATION PATTERN OF
THE SCREEN -BEAM ARRAY. THE FRONT -
THE MATCHING UNIT IN DETAIL FOR TO -BACK RATIO IS ABOUT 28 db IN
THE SCREEN -BEAM DESIGN, WHICH AMPLITUDE, AND THE FORWARD
ALLOWS THE USE OF 72 -OHM COAX GAIN APPROXIMATELY 15 db
510 VHF and UHF Antennas THE RADIO
long. These tubes are welded onto the center justable matching stub (C) and two vari-
tube of each group of three horizontal brac- able capacitors (C, and Co) are employed
ing tubes, and are so located to support the for impedance matching. The two variable
horizontal dipole at its exact center. The capacitors are mounted in a watertight box,
dipoles are attached to the supporting rods with the balun and matching stubs entering
by means of small phenolic insulating the bottom and top of the box, respectively.
blocks, as shown in figure 16. The radiators The matching procedure is carried out by
are therefore insulated from the screen the use of a standing -wave meter (SWR
reflector. The inner tips of the radiators are bridge). A few watts of power are fed to
held by small polystyrene blocks for rigidity, the array through the SWR meter, and the
and are cross connected to each other by a setting of the shorting stub on C and the
transposed length of TV -type 400 -ohm open setting of the two variable capacitors are
wire line. The entire array is fed at the adjusted for lowest SWR at the chosen
point A -A, illustrated in figure 15. operating frequency. The capacitance set-
The matching system for the beam is tings of the two variable capacitors should
mounted behind the reflector screen, and is be equal. The final adjustment is to set the
shown in figure 17. A quarter -wave trans- shorting stub of the balun (D) to remove
former (B) drops the relatively high imped- any residual reactance that might appear
ance of the antenna array to a suitable value on the transmission line. With proper ad-
for the low- impedance balun (D). An ad- justment, the SWR of the array may be
Dio D11
Dv
De
D7
Ds
LENGTH SPACING
ELEMENT FROM
(D /A M. WS -) 144 Yi, 143 MS 1461111. 147 MB. DIPOLE
Figure 19
www.americanradiohistory.com
HANDBOOK VHF Beam Antennas 511
Rotary Beams
The rotatable antenna array has become directional parasitic end -fire array, and is
almost standard equipment for operation on actually a type of yagi array. The flat -top
the 28- and 50 -MHz bands and is common- beam is a type of bidirectional end -fire ar-
ly used on the 14- and 21 -MHz bands and ray. The broadside type of array is also quite
on those frequencies above 144 MHz. The effective in obtaining low -angle radiation,
rotatable array offers many advantages for and, although widely used in f -m and TV
both military and amateur use. The direc- broadcasting, has seen little use by amateurs
tivity of the antenna types commonly em- in rotatable arrays because of its size.
ployed (particularly the unidirectional ar-
rays) offers a worthwhile reduction in inter- 24 -1 Unidirectional
ference from undesired directions. Also,
the increase in the ratio of low -angle radia- Parasitic End -Fire Arrays
tion plus the theoretical gain of such ar- (Yogi Type)
rays results in a relatively large increase in
both the transmitted signal and the signal If a single parasitic element is placed on
intensity from a station being received. one side of a driven dipole at a distance of
A significant advantage of a rotatable from 0.1 to 0.25 wavelength the parasitic
antenna array in the case of the normal element can be tuned to make the array sub-
station is that a relatively small amount of stantially unidirectional.
space is required for erection of the antenna This simple array is termed a two -element
system. In fact, one of the best types of parasitic beam.
installation uses a single telephone pole with
the rotating structure holding the antenna The Two -Element The two- element parasitic
mounted atop the pole. To obtain results in Beam beam provides the greatest
all azimuth directions from fixed arrays amount of gain per unit
comparable to the gain and directivity of a size of any array commonly used by radio
single rotatable three -element parasitic beam amateurs. Such an antenna is capable of a
would require several acres of surface. signal gain of 5 db over a dipole, with a
There are two normal configurations of front -to-back ratio of 7 to 15 db, depend-
radiating elements which, when horizontally ing on the adjustment of the parasitic ele-
polarized, will contribute to obtaining a low ment. The parasitic element may be used
angle of radiation. These configurations are either as a director or as a reflector.
the end -fire array and the broadside array. The optimum spacing for a reflector in a
The conventional three- or four -element ro- two -element array is approximately 0.13
tary beam may properly be called a uni- wavelength and with optimum adjustment
www.americanradiohistory.com
Unidirectional Arrays 513
50
45
40
35 t
m5
o
50
23
Z2 2 20
D
.....----"-
10
e
O1 0.15 02 0.25 0 01 0.15 02 0.25
ELEMENT SPACING (X) ELEMENT SPACING (X)
Figure 1 Figure 2
GAIN VERSUS ELEMENT SPACING FOR A RADIATION RESISTANCE AS A FUNCTION
TWO-ELEMENT CLOSE-SPACED PARASITIC OF THE ELEMENT SPACING FOR A
BEAM ANTENNA WITH PARASITIC ELE- TWO- ELEMENT PARASITIC ARRAY
MENT OPERATING AS A DIRECTOR OR
REFLECTOR
of the length of the reflector a gain of tion resistance of about 25 ohms, it may be
approximately 5 db will be obtained, with of advantage in some instances to choose the
a feed -point resistance of about 25 ohms. antenna with the higher radiation resistance,
If the parasitic element is to be used as a assuming other factors to be equal.
director, the optimum spacing between it Figure 3 shows the front -to -hack ratio
and the driven element is 0.11 wavelength. for the two -element parasitic array for both
The general characteristics of a two -ele- the reflector and director cases. To produce
these curves, the elements were tuned for
ment parasitic array may be seen in figures
maximum gain of the array. Better front -
1, 2 and 3. The gain characteristics of a
to -back ratios may be obtained at the ex-
two -element array when the parasitic ele- pense of array gain, if desired, but the gen-
ment is used as a director or as a reflector eral shape of the curves remains the same.
are shown. It can be seen that the director It can be readily observed that operation
provides a maximum of 5.3 db gain at a of the parasitic element as a reflector pro-
spacing of slightly greater than 0.1 wave- duces relatively poor front -to -back ratios
length from the antenna. In the interests of except when the element spacing is greater
greatest power gain and size conservation, than 0.15 wavelength. However, at this ele-
therefore, the choice of a parasitic director ment spacing, the gain of the array begins
would be wiser than the choice of a parasitic to suffer.
reflector, although the gain difference be- Since a radiation resistance of 17 ohms is
tween the two is small. not unduly hard to match, it can be argued
Figure 2 shows the relationship between that the best all- around performance may
the element spacing and the radiation resist- be obtained from a two -element parasitic
ance for the two -element parasitic array for beam employing 0.11 element spacing, with
both the reflector and the director case. the parasitic element tuned to operate as a
Since the optimum antenna -director spacing director. This antenna will provide a for-
for maximum gain results in an antenna ward gain of 5.3 db, with a front -to -back
radiation resistance of about 17 ohms, and ratio of 10 db, or slightly greater. Closer
the optimum antenna -reflector spacing for spacing than 0.11 wavelength may be em-
maximum gain results in an antenna radia- ployed for greater front -to -back ratios, but
514 Rotary Beams THE RADIO
_40
the radiation resistance of the array becomes m
quite low, the bandwidth of the array be-
comes very narrow, and the tuning becomes 3,
I-
quite critical. Thus the Q of the antenna tr
system will be increased as the spacing be- 20
tween the elements is decreased, and smaller CO
www.americanradiohistory.com
HANDBOOK Three -Element Array 515
a dipole in free space. In actual practice, the the order of 0.25 wavelength between ele-
array will often show 7 to 8 db apparent ments) will result in greater bandwidth for
gain over a horizontal dipole placed the a specified maximum standing -wave ratio on
same height above ground (at 28 and 14 the antenna transmission line. Smaller spac-
MHz) . ings may be used when boom length is an
The use of more than three elements is important consideration, but for a specified
desirable when the length of the supporting standing -wave ratio and forward gain the
structure is such that spacings of approxi- frequency coverage will be smaller. Thus the
mately 0.15 wavelength between elements O of the antenna system will be increased
becomes possible. Four -element arrays are as the spacing between the elements is de-
quite common on the 28- and 50 -MHz creased, resulting in smaller frequency cov-
bands, and five elements are sometimes used erage, and at the same time the feed -point
for increased gain and discrimination. As impedance of the driven element will be
the number of elements is increased the gain decreased.
and front -to -back ratio increase but the For broad -band coverage, such as the
radiation resistance decreases and the band- range from 28.0 to 29.7 MHz or from 50
width or frequency range over which the to 54 MHz, 0.2 wavelength spacing from
antenna will operate without reduction in the driven element to each of the parasitic
effectiveness is decreased. elements is recommended. For narrower
bandwidth, such as would be adequate for
Material for the 14.0- to 14.4 -MHz band or the 144- to
While the elements may consist
Elements 148 -MHz band, the radiator -to- parasitic
of wire supported on a wood
framework, self-supporting ele- element spacing may be reduced to 0.12
ments of tubing are much to be preferred.
wavelength, while still maintaining adequate
The latter type array is easier to construct, array bandwidth for the amateur band in
looks better, is no more expensive, and question.
avoids the problem of getting sufficiently
good insulation at the ends of the elements. Length of the Experience has shown that
The voltages reach such high values toward Parasitic Elements it is practical to cut the
the ends of the elements that losses will be parasitic elements of a
excessive, unless the insulation is excellent. three -element parasitic array to a predeter-
The elements may be fabricated of thin - mined length before the installation of such
walled steel conduit, or hard -drawn thin - an antenna. A pretuned antenna such as this
walled copper tubing, but (bird tubing is will give good signal gain, adequate front -
much better. Dural tubing may be obtained to -back ratio, and good bandwidth factor.
in telescoping sizes from large metal -supply By carefully tuning the array after it is in
houses in many cities. Various manufacturers. position the gain may be increased by a
moreover, supply beam antenna kits of all fraction of a db, and the front -to -back ratio
by several db. However the slight improve-
types and prices. The majority of these
ment in performance is usually not worth
beams employ dural elements because of the
the effort expended in tuning time.
good weather -capability of this material.
The closer the lengths of the parasitic
elements are to the resonant length of the
Element Spacing The optimum spacing for driven element, the lower will be the feed -
a two -element array is, as point resistance of the driven element, and
has been mentioned before, approximately the smaller will be the bandwidth of the
0.11 wavelength for a director and 0.13 array. Hence, for wide frequency coverage
wavelength for a reflector. However, when the director should be considerably shorter,
both a director and a reflector are combined and the reflector considerably longer than
with the driven element to make up a three - the driven element. For example, the direc-
element array the optimum spacing is estab- ter should still be less than a resonant half -
lished by the bandwidth which the antenna wavelength at the upper frequency limit of
will be required to cover. Wide spacing (of the range wherein the antenna is to be oper-
516 Rotary Beams THE RADIO
TYPE
3-ELEMENT
3-CLEMENT
PRIVEN ELEMENT
LENGTH
473
F Dow
-
.)
REFLECTOR
LENGTH
501
F pool
F(MH,)
1STDIRECTOR
LENGTH
45
F DPW
F(MMS)
- -
2ND DIRECTOR
LENGTH
3RD DIRECTOR SPACING SET-
WEDI
.1S -.IS
.2S -.2S
S
APPROR.GAIN
DS
7.5
3.5
APPROx RADIATION
RE i3TAHCE(1L)
20
as
-1
Figure 5
aced, and the reflector should still be long more. In the vhf range, where the additional
enough to act as a reflector at the lower fre- elements may be added without much diffi-
quency limit. Another way of stating the culty, and where required bandwidths are
same thing is to say, in the case of an array small, the use of more than two parasitic
to cover a wide frequency range such as the elements is quite practical.
amateur range from 28 to 29.7 MHz that the
director should be cut for the upper end Stocking of Parasitic arrays (yagis) may be
of the band and the reflector for the lower Yogi Arrays stacked to provide additional
end of the band. In the case of the 28- to gain in the same manner that
29.7 -MHz range this means that the director dipoles may be stacked. Thus if an array of
should be about 8 percent shorter than the six dipoles would give a gain of 10 db, the
driven element and the reflector should be substitution of yagi arrays for each of the
about 8 percent longer. Such an antenna dipoles would add the gain of one yagi ar-
will show a relatively constant gain of about ray to the gain obtained with the dipoles.
6 db over its range of coverage, and the pat- However, the yagi arrays must be more
tern will not reverse at any point in the widely spaced than the dipoles to obtain this
range. theoretical improvement. As an example, if
Where the frequency range to be covered six 5- element yagi arrays having a gain of
is somewhat less, such as the 14.0- to 14.4 - about 10 db were substituted for the di-
MHz amateur band, or the lower half of poles, with appropriate increase in the spac-
the amateur 28 -MHz phone band, the re- ing between the arrays, the gain of the
flector should be about 5 percent longer than whole system would approach the sum of
the driven element, and the director about f the two gains, or 20 db. A group of arrays
percent shorter. Such an antenna will per- of yagi antennas, with recommended spac-
form well over its rated frequency band, ing and approximate gains, is illustrated
will not reverse its pattern over this band, in figure 6.
and will show a signal gain of 7 to 8
db. See figure 5 for design figures for 24 -3 Feed Systems for
3- element arrays. Parasitic (Yagi) Arrays
More Thon A small amount of additional
Three Elements gain may be obtained through
The table of figure f gives, in addition to
other information, the approximate radia-
use of more than two parasitic tion resistance referred to the center of the
elements, at the expense of reduced feed - driven element of multielement parasitic
point impedance and lessened bandwidth. arrays. It is obvious, from these low values
One additional director will add about 1 db, of radiation resistance, that special care
and a second additional director (making a must be taken in materials used and in the
total of five elements including the driven construction of the elements of the array
element) will add slightly less than 1 db to ensure that ohmic losses in the conductors
www.americanradiohistory.com
HANDBOOK Feed Systems 517
0.2a-+-112A-.0.2a-4+r0.2A.
45 a30
F RH, F Nw
OGAIN ABOUT 12 DB
WITH 2 SECTIONS
FEEDER LINE
Figure 6
will not be an appreciable percentage of the low radiation resistance of these antenna ar-
radiation resistance. It is also obvious that rays to the normal range of characteristic
some method of impedance transformation impedance used for antenna transmission
must be used in many cases to match the lines.
518 Rotary Beams THE RADIO
4
R RADIATION
i FOR
FOR
DI
Dt t-
Da RFEED
RpA
O 2 D2o.
5. 1.s
RY
RAD.
x8.9
FOLDED- ELEMENT
MATCH
FEED
FOR DIt
D2.11'10.5
5I.5
FORDtt-
D2'.25
1 - 18
5. RAD.
D
1
T
FORD
*12
FoR
S
WIRE
D t-
3t1
1.
RA D.
sa-
12 WIRE
RA-14
D-1.
FOR
12 1.s -RA 18 D.
I
WIRE
FOR D
e S
WIRE
1- VIAL 24
FOR D 1-
S. 1- RZU12. 32
12 WIRE RAO.
Figure 7
DATA FOR
FOLDED -ELEMENT
R FEED 2 9 MATCHING SYSTEMS
3 -WIRE MATCH
In all normal applications of
the data given the main ele-
ment as shown is the driven
element of a multi element
parasitic array. Directors and
reflectors have not been shown
for the sake of clarity.
R FEED =APPROa 25
5 -WIRE MATCH R RAD
www.americanradiohistory.com
HANDBOOK Feed Systems 519
L
L-.1*-44x L
Figure 8 L
taxL-+-ta%L
AVERAGE DIMENSIONS
FOR THE DELTA AND T-o
"T" MATCH
Dt3D2
OB 'T' MATCH
20011 ON 300 a
TWIN LINE
ment proper. The wire may be spaced below acteristic impedance of the transmission line
the self- supporting element by means of and the radiation resistance of the driven
several small strips of polystyrene which element of the antenna array. Inspection of
have been drilled for both the main element the ratios given in figure 7 shows that the
and the small wire and threaded on the fourth set of dimensions given under figure
main element. 7B will give a 24 -to -1 step -up, which is
sufficiently close. So it is merely necessary to
The Folded- Element The calculation of the use a 1 -inch diameter driven element with
Match Calculations operating conditions of a No. 8 wire spaced on 1 -inch centers ( %-
the folded -element inch below the outside wall of the 1 -inch
matching systems and the yoke match, as tubing). The No. 8 wire is broken and a
shown in figures 7A and 7B is relatively
2 -inch insulator placed in the center. The
simple. A selected group of operating con-
feed line then carries from this insulator
ditions has been shown on the drawing of
figure 7. In applying the system it is only down to the transmitter. The center insula-
necessary to multiply the ratio of feed to tor should be supported rigidly from the f-
radiation resistance (given in the figures to inch tube so that the spacing between the
the right of the suggested operating dimen- piece of tubing and the No. 8 wire will be
sions in figure 7) by the radiation resistance accurately maintained.
of the antenna system to obtain the imped- In many cases it will be desired to use the
ance of the cable to be used in feeding the folded -element or yoke matching system
array. Approximate values of radiation re- with different sizes of conductors or differ-
sistance for a number of commonly used ent spacings than those shown in figure 7.
parasitic- element arrays are given in figure S. Note, then, that the impedance transforma-
As an example, suppose a 3- element array tion ratio of these types of matching systems
with 0.15D-0.15R spacing between ele- is dependent Goth on the ratio of conductor
ments is to be fed by means of a 465 -ohm diameters and on their spacing. The follow-
line constructed of No. 12 wire spaced 2 ing equation has been given by Roberts
inches. The approximate radiation resistance (RCA Review, June, 1947) for the deter-
of such an antenna array will be 20 ohms. mination of the impedance transformation
Hence we need a ratio of impedance step -up when using different diameters in the two
of 23 to obtain a match between the char- sections of a folded element:
520 Rotary Beams THE RADIO
52 O. COAXIAL CABLE
QUARTER -WAVE
TRANSFORMER FEED
Figure 9
OD ROTARY -LINK
COUPLING
COIL SPACED COILS 10
APPROX. 0.5" DIAMETER
C
1 TURN LINKS ARE PARALLEL
C 13 200 PF VARIABLE
S0 -BOO IL LINE
www.americanradiohistory.com
HANDBOOK Feed Systems 521
TO
STUB
" Rotary Beams
TRANSMITTER
FLAT LINE
SWR =1.0
Figure 11
MATCHING STUB
\
less expensive than equivalent filters for two -
required by reference to a chart such as
wire line. figure 12 and using them as a starter. Usual-
The gamma -match is illustrated in figure
\:
10, and may be considered as one -half of a MIMMIN IMMI1111I11I
IlkINNIMINIMINIMIMMI11II11111
10
1102
T- match. One resonating capacitor is used,
placed in series with the gamma rod. The
capacitor should have a capacity of 7 pf
1=M iIIII
IILIMIMIMMINIIMMIN11
NIMNINIINI11NI11
1111111111
i/MIMI11111111111
1
_....eesl 70
eo 0
W
3 4
o"o\s
5
ductance of the gamma rod. By adjustment SWR
of the length of the gamma rod, and the Figure 12
setting of the gamma capacitor, the SWR
on the coaxial line may be brought to a very SHORTED -STUB LENGTH AND POSITION
low value at the chosen operating frequency. CHART
The use of an Antennascohe, described in From the standing -wave ratio and current or
the Test Equipment chapter is recommended voltage null position it is possible to deter-
mine the theoretically correct length and
for precise adjustment of the gamma match. position of a shorted stub. In actual prac-
tice a slight discrepancy usually will be
The Matching Stub If an open -wire line is found between the theoretical and the ex-
low- imped-
used to feed a perimentally optimised dimensions; therefore
it may be necessary to "touch up" the di-
ance radiator, a section of the transmission mensions after using the above data as a
line may be employed as a matching stub as starting point.
www.americanradiohistory.com
HANDBOOK Unidirectional Driven Arrays 523
T
\`
ly only a slight "touching up" will produce DIRECTIONAL
a perfect match and flat line. { 4
In order to utilize figure 12, it is first nec-
essary to locate accurately a voltage node or
current node on the line in the vicinity that
has been decided on for the stub, and also to
determine the SWR.
Stub adjustment becomes more critical as
the SWR increases, and under conditions of OGAIN ABOUT B DB FEED LINE
24 -4 Unidirectional
Driven Arrays
Three types of unidirectional driven ar-
rays are illustrated in figure 13. The array
shown in figure 13A is an end -fire system
which may be used in place of a parasitic
array of similar dimensions when greater
frequency coverage than is available with
the yagi type is desired. Figure 13B is a
combination end -fire and collinear system FEED LINE
which will give approximately the same gain
as the system of figure 13A, but which re-
quires less boom length and greater total Figure 13
element length. Figure 13C illustrates the UNIDIRECTIONAL ALL -DRIVEN ARRAYS
familiar lazy -H with driven reflectors (or
A unidirectional all -driven end-lire array Is
directors, depending on the point of view) shown at A. I shows an array with two half
in a combination which will show wide waves in phase with driven reflectors. A
lazy -H array with driven reflectors is shown
bandwidth with a considerable amount of at C. Note that the directivity is through the
forward gain and good front -to -back ratio elements with the greatest total feed -line
over the entire frequency coverage. length in arrays such as shown at B and C.
A simple driven array is the so- called
ZL Special, which is one -half the array of Unidirectional Stacked Three practical types
figure 13B. The ZL Special is fed at the Broadside Arrays of unidirectional
center point of the half -wave elements and stacked broadside ar-
provides a cardioid pattern with a gain of rays are shown in figure 14. The first type,
about 3 decibels. shown at figure 14A, is the simple lazy -H
524 Rotary Beams THE RADIO
GAIN APPROX. R DB
-.4
BROADSIDE HALF -WAVES
WITH REFLECTORS
Figure 14
300 n. LINE
BROADSIDE ARRAYS
4 WITH PARASITIC
REFLECTORS
"TWO OVER TWO OVER TWC
WITH REFLECTORS The apparent gain of the ar-
rays illustrated will be
GAIN APPROX 11 S DB greater than the values
given due to concentration
of the radiated signal at the
lower elevation angles.
8SLI LINE
type of antenna with parasitic reflectors for unidirectional array when both elements are
each element. Figure 14B shows a simpler the same length. Using this procedure will
antenna array with a pair of folded dipoles give a gain of 3 db with the reflectors over
spaced one -half wave vertically, operating the gain without the reflectors, with only a
with reflectors. In figure 14C is shown a moderate decrease in the radiation resistance
more complex array with six half waves of the driven element. Actually, the radia-
and six reflectors which will give a very tion resistance of a half -wave dipole goes
worthwhile amount of gain. down from 73 ohms to 60 ohms when an
In all three of the antenna arrays shown identical half -wave element is placed one -
the spacing between the driven elements quarter wave behind it.
and the reflectors has been shown as one- A very slight increase in gain for the en-
quarter wavelength. This has been done to tire array (about 1 db) may be obtained at
eliminate the requirement for tuning of the the expense of lowered radiation resistance,
reflector, as a result of the fact that a half - the necessity for tuning the reflectors, and
wave element spaced exactly one- quarter decreased bandwidth by placing the re-
wave from a driven element will make a flectors 0.15 wavelength behind the driven
www.americanradiohistory.com
HANDBOOK Construction of Rotatable Arrays 525
elements and making them somewhat longer supporting elements are used in the flat -top
than the driven elements. The radiation re- beam the problem of losses due to insulators
sistance of each element will drop approxi- at the ends of the elements is somewhat re-
mately to one -half the value obtained with duced. With a single- section flat -top beam a
untuned half -wave reflectors spaced one - gain of approximately 4 db can be expected,
quarter wave behind the driven elements. and with two sections a gain of approxi-
Antenna arrays of the type shown in fig- mately 6 db can be obtained.
ure 14 require the use of some sort of lat- Another type of bidirectional array which
tice work for the supporting structure since has seen less use than it deserves is shown
the arrays occupy appreciable distance in in figure 15B. This type of antenna system
space in all three planes. has a relatively broad azimuth or horizontal
beam, being capable of receiving signals with
Feed Methods The requirements for the feed little diminution in strength over approxi-
systems for antenna arrays of mately 40 degrees, but it has a quite sharp
the type shown in figure 14 are less critical elevation pattern since substantially all radi-
than those for the close- spaced parasitic ar- ation is concentrated at the lower angles of
rays shown in the previous section. This is radiation if more than a total of four ele-
a natural result of the fact that a larger
ments is used in the antenna system. Figure
number of the radiating elements are direct- 15B gives the approximate gain over a half -
ly fed with energy, and of the fact that the wave dipole at the height of the center of
effective radiation resistance of each of the the array which can be expected. Also shown
driven elements of the array is much higher in this figure is a type of "rotating- mast"
than the feed -point resistance of a parasitic structure which is well suited to rotation of
array. As a consequence of this fact, arrays this type of array.
of the type shown in figure 14 can be ex- If six or more elements are used in the
pected to cover a somewhat greater fre- type of array shown in figure 15B, no match-
quency band for a specified value of stand- ing section will be required between the
ing -wave ratio than the parasitic type of antenna transmission line and the feed point
array. of the antenna. When only four elements
In most cases a simple open -wire line may are used, the antenna is the familiar lazy H
be coupled to the feed point of the array and a quarter -wave stub should be used for
without any matching system. The stand- matching the antenna transmission line to
ing -wave ratio with such a system of feed the feed point of the antenna system.
will often be less than 2 -to -1. However, if If desired, and if mechanical considera-
a more accurate match between the antenna
tions permit, the gain of the arrays shown
transmission line and the array is desired a in figure 15B may be increased by 3 db by
conventional quarter -wave stub, or a quar- placing a half -wave reflector behind each of
ter -wave matching transformer of appro- the elements at a spacing of one -quarter
priate impedance, may be used to obtain a wave. The array then becomes essentially
low standing -wave ratio. the same as that shown in figure 14C and
the same considerations in regard to reflec-
24 -5 Bidirectional tor spacing and tuning will apply. However,
Rotatable Arrays the factor that a bidirectional array need be
The bidirectional type of array is sometimes rotated through an angle of less than 180
used on the 28- and S0 -MHz bands where should be considered in this connection.
signals are likely to be coming from only one
general direction at a time. Hence the sacri-
24 -6 Construction of
fice of discrimination against signals arriv- Rotatable Arrays
ing from the opposite direction is likely to A considerable amount of ingenuity may
be of little disadvantage. Figure 15 shows be exercised in the construction of the sup-
two general types of bidirectional arrays. porting structure for a rotatable array.
The flat -top beam, which has been described Every person has his own ideas as to the
in detail earlier, is well adapted to installa- best method of construction. Often the
tion atop a rotating structure. When self- most practical method of construction will
526 Rotary Beams THE RADIO
O
STUB FLAT -TOP BEAM FOR
ROTATABLE ARRAY
GAIN TO 6 De
7.8 DB
OF BIDIRECTIONAL
9.0 DB
ARRAYS
10.0 De 10
Average gain figures are
given for both the flat -top
beam type of array and for n
the broadside -collinear array
with different numbers of RADIAL LOAD 465 0. FEEDERS
elements. BEARING N. 12 WIRE SPACED 2
GUY WIRES
ROPES TO
PCRATING POSITION
rTHRUST
REARING
be dictated by the availability of certain durai tubing for the self -supporting ele-
types of construction materials, but in any ments. Other materials may be used but an
event be sure that sound mechanical en- alloy known as 2024 has proven over a
gineering principles are used in the design of period of time to be quite satisfactory. Cop-
the supporting structure. There are few per tubing is too heavy for a given strength,
things quite as discouraging as the picking and steel tubing, unless copper plated, is
up of pieces, repairing of the roof, etc., likely to add an undesirably large loss re-
when a newly constructed rotary comes sistance to the array. Also, steel tubing,
down in the first strong wind. If the prin- even when plated, is not likely to withstand
ciples of mechanical engineering are under- salt atmosphere (such as is encountered along
stood it is wise to calculate the loads and the seashore) for a satisfactory period of
torques which will exist in the various mem- time. Do not use a soft aluminum alloy for
bers of the structure with the highest wind the elements unless they will be quite short;
velocity which may be expected in the 2024 is a hard alloy and is noncorrosive.
locality of the installation. If this is not Alloy 2017 and 6061 are also satisfactory,
possible it will usually be worth the time cheaper, and easier to obtain. Do not use al-
and effort to look up a friend who under- loys 5052, 2014, or 3003 (EMT), as these
stands these principles.
signify alloys which have not been heat
Radiating One thing more or less standard treated for strength and rigidity. However,
Elements about the construction of rotat- these softer alloys, and aluminum electrical
able antenna arrays is the use of conduit, may be used for short radiating
HANDBOOK Beam Construction 527
break at the mounting holes when the array that a constant amount of power will
issubject to heavy winds. be fed to the array under test. Also, if
you use this system, use very low power ( 5
24-7 Tuning the Array or 10 watts of power is usually sufficient)
and make sure that the antenna transmission
line is effectively grounded as far as d -c
Although satisfactory results may be ob-
tained by precutting the antenna array to plate voltage is concerned. The use of the
dimensions given earlier in this chapter, the method described in the previous paragraph
occasion might arise when it is desired to of course eliminates these problems.
make a check on the operation of the anten-
One satisfactory method of tuning the
na before calling the job complete.
array proper, assuming that it is a system
with several parasitic elements, is to set the
The process of tuning an array may sat-
directors to the dimensions given in figure 5
isfactorily be divided into two more or less
and then to adjust the reflector for maxi-
distinct steps: the actual tuning of the ar- mum forward signal. Then the first director
ray for best front -to -back ratio or for maxi-
should be varied in length until maximum
mum forward gain, and the adjustment to
forward signal is obtained, and so on if ad-
obtain the best possible impedance match
ditional directors are used. Then the array
between the antenna transmission line and may be reversed in direction and the reflector
the feed point of the array.
adjusted for best front -to-back ratio. Sub-
Tuning the The actual tuning of the array sequent small adjustments may then be
Array for best front -to-back ratio or made in both the directors and the reflector
maximum forward gain may for best forward signal with a reasonable
best be accomplished with the aid of a low - ratio of front -to -back signal. The adjust-
power transmitter feeding a dipole antenna ments in the directors and the reflector will
(polarized the same as the array being be found to be interdependent to a certain
tuned) at least four or five wavelengths degree, but if small adjustments are made
away from the antenna being tuned and after the preliminary tuning process a sat-
located at the same elevation as that of the isfactory set of adjustments for maximum
antenna under test. A calibrated field - performance will be obtained. It is usually
strength meter of the remote -indicating type best to make the end sections of the ele-
is then coupled to the feed point of the ments smaller in diameter so that they will
antenna array being tuned. The transmis- slip inside the larger tubing sections. The
sions from the portable transmitter should smaller sliding sections may be clamped in-
be made as short as possible and the call side the larger main sections.
sign of the station making the test should In making the adjustments described, it
be transmitted at least every ten minutes. is best to have the rectifying element of the
It is, of course, possible to tune an array remote -indicating field- strength meter di-
with the receiver connected to it and with a rectly at the feed point of the array, with
station a mile or two away making trans- a resistor at the feed point of the estimated
missions on your request. But this method value of feed -point impedance for the array.
is more cumbersome and is not likely to
give complete satisfaction. It is also possible Matching to the The problem of matching
to carry out the tuning process with the Antenna Trans- the impedance of the an-
transmitter connected to the array and with mission Line tenna transmission line to
the field- strength meter connected to the the array is much simpli-
remote dipole antenna. In this event the fied if the process of tuning the array is
indicating instrument of the remote -indi- made a substantially separate process as just
cating field- strength meter should be visible described. After the tuning operation is
from the position where the elements are complete, the resonant frequency of the
being tuned. However, when the array is driven element of the antenna should be
being tuned with the transmitter connected checked, directly at the center of the driven
to it there is always the problem of making element if practical, with a grid -dip meter.
continual adjustments to the transmitter so It is important that the resonant frequency
www.americanradiohistory.com
HANDBOOK Tuning the Array 529
S. P O.T.
RELAY
SOCKET PLUG DOCKET I
ANTENNA ROTATOR
PLUG
TO PROP MOTOR
REVERSE LEADS
o0U
,p
PILOT
LIGHT
01!
ITO ttS-V A.C. Oi s0
a
TOGGLE
SWITCH
SOCKET J PLUG SOCKET L PLUG
J
DIRECTION INDICATOR
Figure 19
sired reading, the length of the match- termined near the ground is marked so that
ing rod should be decreased. After the adjustments will not be lost, the array
each change in length of the matching may be raised to rotatable height and the
rod, the series capacitor in the match- fastening clamps left loose enough so that
ing system should be re- resonated for the elements may be slid in by means of a
best null on the meter of the Anten- long bamboo pole. After a series of trials a
nascope. satisfactory set of adjustments can be ob-
tained.
Raising and A practical problem always pres- The matching process does not require ro-
Lowering ent when tuning up and match - tation, but it does require that the antenna
the Array ing an array is the physical lo- proper be located at as nearly its normal
cation of the structure. If the operating position as possible. However, on
array is atop the mast it is inaccessible for a particular installation the standing -wave
adjustment, and if it is located on step- ratio on the transmission line near the trans-
ladders where it can be adjusted easily it mitter may be checked with the array in the
cannot be rotated. One encouraging factor air, and then the array may be lowered to
in this situation is the fact that experience ascertain whether or not the SWR has
has shown that if the array is placed 8 or changed. If it has not, and in most cases if
I o feet above ground on some stepladders the feeder line is strung out back and forth
for the preliminary tuning process, the rais- well above the ground as the antenna is
ing of the system to its full height will not lowered they will not change, the last ad-
produce a serious change in the adjustments. justment may be determined, the standing -
So it is usually possible to make preliminary wave ratio again checked, and the antenna
adjustments with the system located slightly re- installed in its final location.
greater than head height above ground, and
then to raise the antenna to a position where 24 -8 Indication of Direction
it may be rotated for final adjustments. If The most satisfactory method for indi-
the position of the matching device as de- cating the direction of transmission of a ro-
www.americanradiohistory.com
HANDBOOK Three -Band Beams 531
tatable array is that which uses Selsyns or shown in figure 20. At the lowest operating
Synchros for the transmission of the data frequency, the tuned traps exert a minimum
from the rotating structure to the indicating influence on the element which resonates at
pointer at the operating position. A num- a frequency determined by the electrical
ber of Synchros and Selsyns of various types length of the configuration, plus a slight de-
are available on the surplus market. Some gree of loading contributed by the traps.
of them are designed for operation on 115 At some higher frequency (generally about
volts at 60 Hertz, some are designed for 1.5 times the lowest operating frequency)
operation on 60 Hertz but at a lowered volt- the outer set of traps is in a parallel reso-
age, and some are designed for operation nant condition, placing a high impedance
from 400 -Hertz or 800 -Hertz energy. This between the element and the tips beyond
latter type of high - frequency Selsyn is the the traps. Thus, the element resonates at a
most generally available type, and the high - frequency 1.5 times higher than that deter-
frequency units are smaller and lighter than mined by the overall length of the element.
the 60 -Hertz units. Since the indicating As the frequency of operation is raised to
Selsyn must deliver an almost negligible approximately 2.0 times the lowest operating
amount of power to the pointer which it frequency, the inner set of traps becomes
drives, the high- frequency types will operate resonant, effectively disconnecting a larger
quite satisfactorily from 60 -Hertz power if portion of the element from the driven sec-
the voltage on them is reduced to somewhere tion. The length of the center section is
between 6.3 and 20 volts. In the case of resonant at the highest frequency of opera-
many of the units available, a connection tion. The center section, plus the two ad-
sheet is provided along with a recommenda- jacent inner sections are resonant at the
tion in regard to the operating voltage when intermediate frequency of operation, and the
they are run on 60 Hertz. In any event the complete element is resonant at the lowest
operating vltage should be held as low as it frequency of operation.
may be and still give satisfactory transmis- The efficiency of such a system is deter-
sion of data from the antenna to the operat- mined by the accuracy of tuning of both
ing position. Certainly it should not be the element sections and the isolating traps.
necessary to run such a voltage on the units In addition the combined dielectric losses of
that they become overheated. the traps affect the overall antenna effi-
A suitable Selsyn indicating system is ciency. As with all multipurpose devices,
shown in figure 19. some compromise between operating con-
Systems using a potentiometer capable of venience and efficiency must be made with
continuous rotation and a milliammeter, antennas designed to operate over more than
along with a battery or other source of cne narrow band of frequencies. Taking into
direct current, may also be used for the in- account the theoretical difficulties that must
dication of direction. be overcome it is a tribute to the designers
of the better multiband beams that they
24 -9 Three -Band Beams perform as well as they do.
A popular form of beam antenna intro- The Isolating The parallel -tuned circuit
duced during the past few years is the so- Trap which serves as an isolating
called three -band beam. An array of this trap for a multiband anten-
type is designed to operate on three adjacent na should combine high circuit Q with good
amateur bands, such as the 10 -, 15 -, and environmental protection. A highly satis-
20 -meter group. The principle of operation factory trap configuration based on the
of this form of antenna is to employ original design of W3DZZ is shown in fig-
parallel -tuned circuits placed at critical ure 21. The trap capacitor, which has a
positions in the elements of the beam which value of about 25 pf, is made of two sec-
serve to electrically connect and disconnect tions of aluminum tubing which form a
the outer sections of the elements as the portion of the antenna element. The capaci-
frequency of excitation of the antenna is tor dielectric is moulded lucite, or similar
changed. A typical three -band element is plastic material, given a coat of epoxy to
532 Rotary Beams
"N
IIf RESONANT
3
AT HIGHEST FREQUENCY
h. AT
RESONANT -
INTERMEDIATE WREQUENCY
www.americanradiohistory.com
CHAPTER TWENTY -FIVE
Mobile Equipment
Design and Installation
Mobile operation is permitted on all ama- SSB equipment, as contrasted to the heavy
teur bands. Tremendous impetus to this power drain of conventional a -m gear has
phase of the hobby was given by the suit- encouraged the use of relatively high-power
able design of compact mobile equipment. sideband equipment in many mobile instal-
Complete mobile installations may be pur- lations. The rigid frequency stability re-
chased as packaged units, or the whole mo- quirement for satisfactory SSB reception,
bile station may be home built, according to however, has obsoleted the once -popular
the whim of the operator. tuned -converter and auto - receiver combina-
The problems involved in achieving a sat- tion formerly used for a -m reception.
isfactory two -way installation vary some- Transistor, crystal -controlled converters
what with the band, but many of the prob- have attained some measure of popularity
lems are common to all bands. For instance, when combined with transistor auto radios
ignition noise is more troublesome on 10 for casual mobile reception of amateur sig-
meters than on 75 meters, but on the other nals. If the converter includes a demodu-
hand an efficient antenna system is much lating bfo, it may be used for satisfactory
more easily accomplished on 10 meters than SSB reception.
on 75 meters. Also, obtaining a worthwhile
amount of transmitter output without ex- A Transistorized
cessive battery drain is a problem on all
25 -1
bands. Mobile Converter
Specialized mobile equipment is available
for operation on the 2- and 6 -meter bands This inexpensive three- transistor mobile
and a small amount of mobile use is made converter may be used in conjunction with
of the 432 -MHz band. The availability of a transistor auto radio for a -m, c -w or
surplus equipment, moreover, has stimulated SSB reception on the 80- or 40 -meter ama-
f -m mobile activity, especially on 2 meters, teur bands. The converter is self-powered
where the use of fixed f -m repeaters placed from a 9 -volt miniature battery and pro-
on elevated locations has done much to en- vides satisfactory reception when used in
hance vhf mobile operation. conjunction with a tuned mobile whip an-
The recent trend has been toward the tenna.
use of SSB mobile transceivers for high - The schematic of the converter is shown
frequency operation. The low duty -cycle of in figure 1. The unit uses inexpensive RCA
534 Mobile Equipment Design THE RADIO
RCA
ANT. L, SK3006 L3 S50-I050MHz(8OY) Band selection is accomplished by the
J, 1550-850 MHZ(10M) choice of proper coils and crystal. The cost
470
Jz
TO
AUTO RADIO
i of the converter is so moderate that it is
better to construct separate converters for
each amateur band than to try to make a
SM band -change system for a single converter.
The converter may be built on a section
RCA
Y,_ SK3006 4
of copper-plated phenolic circuit board
and placed in a miniature aluminum utility
box, much in the manner shown for the
construction of the vhf converters in the
next chapter of this Handbook. It is sug-
gested that transistor sockets be used to
prevent soldering heat from damaging the
BF0
5M 3006 transistors.
9V.
The converter should be tested in stages.
TA Operation of the mixing oscillator is
E
checked by monitoring the crystal fre-
Figure 1
quency in a nearby receiver as the slug of
SCHEMATIC OF TRANSISTORIZED coil L_ is adjusted. A test signal in the
MOBILE CONVERTER chosen amateur band should then be in-
L,-80 meters; Approx. 40 ,H. 56 turns #24, 1" jected into the antenna receptacle (J1) and
diam., approx. 2" long. (Air -Dux 8327). the converter temporarily connected to the
Link windings A and B are each 10
turns #22 d.c.c. at ground end of L, station receiver, or a broadcast set tuned to
40 meters; Approx. 17 ,H. 24 turns #24, the lower end of the broadcast band. Tuned
1" diam., approx. 3/4" long. (Air -Dux
8327). Link windings A and B are each
circuits L, and L, are peaked for maximum
7 turns 422 d.c.c. at ground end of L, signal strength near the center of the ama-
L,-80 meters; Approx. 35 H. (J.W. Miller teur band. The last adjustment is to set the
21A335R81). Link winding B is 4 turns
#22 d.c.c. at end of winding A. Capaci- slug of bfo coil L, for best SSB reception.
tor C, = SO pf If oscillator injection is too weak for good
40 meters; Approx. 20 H. (J. W. Miller
21A225R81). Link winding same as for SSB reception, it is suggested that the lead
80 meters. Capacitor C, = 30 pf from the 10 -pf coupling capacitor to coil
L,-190 -330 H. (J. W. Miller 4513). Peak to L,B be disconnected and run into the auto
portion of band in use. Link winding
8 is 30 turns 422 d.c.c, on bottom end radio and placed near one of the i -f stages
of
L - L,
3.3-4.1 mH (J. W.
for proper SSB
Miller 27A333181). Peak
reception
of the receiver.
7,-80 meters, 2.950 MHz. 40 meters, 6.450 Two-Meter Reception For reception on the
MHz
144 -MHz amateur
band, and those higher in frequency, the
"universal replacement" transistors in an simple converter /auto -set combination has
easily constructed circuit. Transistor Q, not proven very satisfactory. The primary
serves as a mixer stage, with the incoming reason for this is the fact that the relatively
80- or 40 -meter signal impressed on the sharp i -f channel of the auto set imposes too
base circuit and the local mixing signal link severe a limitation on the stability of the
coupled into the emitter circuit. A crystal - high- frequency oscillator in the converter.
controlled transistor oscillator stage (Q) And if a crystal -controlled beating oscillator
provides the proper mixing frequency. A is used in the converter, only a portion of
separate beat oscillator (Q) is used for the band may be covered by tuning the
SSB and c -w reception, the bfo being tuned auto set.
to the intermediate frequency of the auto The most satisfactory arrangement has
radio which, in most cases, is 262 kHz. The been found to consist of a separately
bfo is coupled to the i -f circuitry through mounted i -f, audio, and power -supply sys-
the stray capacitance of the input circuit tem, with the converter mounted near the
of the auto radio. steering column. The i -f system should have
www.americanradiohistory.com
HANDBOOK Mobile Power Sources 535
ALTERNATOR 30
12.5 V. PER LEG
AUTOMOTIVE
3 0 12 V.D.C. ELECTRICAL
RECTIFIER SYSTEM
3 0, 115 V.
PER LEG TO
HIGH VOLTAGE
POWER
SUPPLY
30 DISTRIBUTION TRANSFORMER
Figure 2
a bandwidth of 30 to 100 kHz and may receiver together with a common audio
have a center frequency of 10.7 MHz if system and power supply. The complete vhf
standard i -f transformers are to be used. station may be packaged in a single cabinet.
The control head may include the 144 -MHz
r -f, mixer, and oscillator sections, and 25 -2 Mobile Power Sources
sometimes the first i -f stage. Alternatively,
the control head may include only the high - As in the case of transmitters for fixed -
frequency oscillator, with a broadband r -f station operation, there are many schools of
unit included within the main receiver as- thought as to the type of transmitter which
sembly along with the i -f and audio sys- is most suitable for mobile operation. One
tems. school states that the mobile transmitter
An alternative arrangement is to build a should have very low power drain, so that
converter, 10.7 -MHz i -f channel, and sec- no modification of the electrical system of
ond detector unit, and then to operate this the automobile will be required, and so that
unit in conjunction with the auto -set pow- the equipment may be operated without
er supply, audio system, and speaker. Such a serious regard to discharging the battery
system makes economical use of space and when the car is stopped, or overloading the
power drain, and can be switched to provide generator when the car is in motion. A
normal broadcast -band auto reception or re- total transmitter power drain of about 80
ception through a converter for the high - watts from the car battery (6 volts at 13
frequency amateur bands. amperes, or 12 volts at 7 amperes) is about
A recent development has been the vhf the maximum that can be allowed under
transceiver, typified by the Gonset Com- these conditions. For maximum power effi-
municator. Such a unit combines a crystal - ciency it is recommended that a transistor
controlled transmitter and a tunable vhf type of supply be used as opposed to a dyna-
536 Mobile Equipment Design THE RADIO
motor supply, since the conversion efficiency addition, the 60 -Hz rating of the transform-
of the transistor unit is high compared to er may be considerably exceeded at the
that of the dynamotor. higher supply frequency, particularly in the
A second school of thought states that case of low duty -cycle SSB equipment.
the mobile transmitter should be of rela- Shown in figure 3 is a Leece- Neville 3-
tively high power to overcome the poor phase alternator mounted atop the engine
efficiency of the usual mobile whip antenna. block, and driven with a fan belt. The
In this case, the mobile power should be voltage regulator and silicon rectifier for
drawn from a system that is independent charging the car battery from the a -c system
from the electrical system of the automo- replace the usual d -c generator. These new
bile. A belt- driven high- voltage generator is items are mounted in the front of the car
often coupled to the automobile engine in radiator. The alternator provides a balanced
this type of installation. delta output circuit wherein the line voltage
Three -Phase is equal to the coil voltage, but the line cur-
With many SSB mobile radio
Power Systems installations now requiring rent is Vi times the coil current. The coil
400 watts peak power or voltage is a nominal 12- volts, rms and
more from the automotive electrical system,
it usually is necessary to run the car engine
when the equipment is operated for more
than a few minutes at a time to avoid dis-
charging the battery. Many commercial ve-
hicles faced with this problem have 3-phase
alternators installed to provide extra power
for two -way radio equipment. A block dia-
gram of such an installation is shown in
figure 2. Voltage regulation of the alternator
system is very good, although the frequency
varies with engine speed, ranging from 100
Hz or so with the engine idling to nearly
1000 Hz at top speed. Modern power trans- Figure 3
formers, however, even though rated for 60- LEECE- NEVILLE 3 -PHASE
Hz operation, are capable of operating effi- ALTERNATOR IS ENGINE DRIVEN
ciently over this range of frequencies. In BY AUXILIARY FAN BELT
4 10 K 4 70 K 470K
3 0, 115 V.
PER LEG
Figure 4
THREE -PHASE -MOBILE KILOWATT SUPPLY
Three -phase power from a system such as shown in figure 3 may be used to provide high
voltage for mobile transmitting equipment. For 1800 volts, transformers T,, T,, and T, are 11S-
volt primary, 830-volt secondary (Stanco, PC- 8301). For 2400 volts, T,, Tr and T are 11S -volt
primary, 1030 -volt secondary (Stance, PC- 8302). Three type 1N1697 or 1N4005 diodes ore
used in each stock.
www.americanradiohistory.com
HANDBOOK Mobile Antennas 537
three 12 -volt, 20- ampere filament trans- This auxiliary trimmer should have a range
formers may be connected in delta on the of about 3 to 50 pf, and may be of the
primary and secondary windings to step the inexpensive compression mica type.
12 volts up to 3 -phase 115 volts. If desired, With the trimmer cut out and the con-
a special 115 -volt, 3 -phase step -up trans- verter turned off (bypassed by the "in -out"
former may be wound which will occupy switch), peak the regular antenna trimmer
less space than the three filament trans- on the auto set at about 1400 kHz. Then
formers. Since the ripple frequency of a turn on the converter, with the receiver
3 -phase d -c power supply will be quite high, tuned to 1500 kHz, switch in the auxiliary
a single 4 -td filter capacitor will suffice. trimmer, and peak this trimmer for maxi-
The schematic of an 1800 -volt, 3 -phase mum background noise. The auxiliary trim-
kilowatt power supply for SSB service is mer then can be left switched in at all times
shown in figure 4. except when receiving very weak broadcast -
band signals.
Some auto sets, particularly certain Gen-
25 -3 Antennas for Mobile eral Motors custom receivers, employ a high-
Work
Auxiliary Antenna One modification of the
Trimmer auto receiver which may
or may not be desirable
depending on the circumstances is the ad-
dition of an auxiliary antenna trimmer
capacitor. If the converter uses an untuned
output circuit and the antenna trimmer on
the auto set is peaked with the converter
cut in, then it is quite likely that the trim-
mer adjustment will not be optimum for
broadcast -band reception when the convert-
ole
er is cut out. For reception of strong broad-
cast -band signals this usually will not be
serious, but where reception of weak broad-
cast signals is desired the loss in gain often
cannot be tolerated, especially in view of
the fact that the additional length of an-
tenna cable required for the converter in-
stallation tends to reduce the strength of
broadcast -band signals.
If the converter has considerable reserve
gain, it may be practical to peak the an-
tenna trimmer on the auto set for broadcast -
band reception rather than resonating it to
the converter output circuit. But oftentimes
this results in insufficient converter gain,
excessive image troubles from loud local
amateur stations, or both. Figure 5
The difficulty can be circumvented by in- A CENTER-LOADED 80 -METER WHIP
corporation of an auxiliary antenna trimmer USING AIR -WOUND COIL MAY BE USED
connected from the "hot" antenna lead on WITH HIGH- POWERED TRANSMITTERS
the auto receiver to ground, with a switch
An anti- corona loop is placed at the top of
in series for cutting it in or out. This ca- the whip to reduce loss of power and burning
pacitor and switch can be connected across of tip of antenna. Number of turns in coil is
critical and adjustable, high -Q coil is recom-
the converter end or the set end of the mended. Whip may be used over frequency
cable between the converter and receiver. range of about 15 kHz without retuning.
538 Mobile Equipment Design THE RADIO
www.americanradiohistory.com
HANDBOOK Mobile Antennas 539
\o ,
Use only a stiff, heavy chassis unless the
chassis is quite small.
Use lock washers or lock nuts when mount-
HOT i2H
ing components by means of screws.
Use stranded hookup wire except where ALTERNATE
CONTROL
MAIN POWER
RELAY
RECEIVER
MUTING
ANTENNA
CHANGEOVER
ANY
OTHER
r -f considerations make it inadvisable (such SWITCH RELAY RELAY RELAYS
www.americanradiohistory.com
HANDBOOK Noise Suppression 541
-41. RING
TIP OF MIRE
25 -5 Vehicular Noise
SHELL
PLUG Suppression
(GROUND)
ryI
.RESS-TC >.
SWITCH Satisfactory reception on frequencies
Figure 9 above the broadcast band usually requires
STANDARD CONNECTIONS FOR THE
greater attention to noise- suppression meas-
PUSH -TO -TALK SWITCH ON A HAND-
ures. The required measures vary with the
HELD SINGLE -BUTTON CARBON
particular vehicle and the frequency range
MICROPHONE
involved.
Most of the various types of noise that
are present in a vehicle may be broken down
not a fixed value, but depends on (1) the into the following main categories:
voltage, (2) whether it is a.c. or d.c., and
(3) whether the circuit is purely resistive or (1) Ignition noise.
is inductive. If in doubt, refer to the man- (2) Wheel static (tire static, brake static,
ufacturer's recommendations. Also keep in and intermittent ground via front
mind that a dynamotor (if you use one) wheel bearings).
presents almost a dead short until the arma- (3) "Hash" from voltage regulator con-
ture starts turning, and the starting relay tacts.
should be rated at considerably more than (4) "Whine" from generator commuta-
the normal dynamotor current. tor segment make and break.
(f) Static from scraping connections be-
Microphones The standardized connections tween various parts of the car.
and Circuits for a majority of hand -held
microphones provided with There is no need to suppress ignition noise
push -to -talk switch are shown in figure 9. completely, because at the higher frequencies
Practically all hand -held military -type sin- ignition noise from passing vehicles makes
gle- button microphones on the surplus mar- the use of a noise limiter mandatory any-
ket use these connections. way. However, the limiter should not be
There is an increasing tendency among given too much work to do, because at high
mobile operators toward the use of micro- engine speeds a noisy ignition system will
phones having better frequency and dis- tend to mask weak signals, even though
tortion characteristics than the standard with the limiter working, ignition "pops"
single- button type. The high -impedance dy- may appear to be completely eliminated.
namic type is probably the most popular, Another reason for good ignition suppres-
with the ceramic- crystal type next in popu- sion at the source is that strong ignition
larity. The conventional crystal type is not pulses contain enough energy, when inte-
suitable for mobile use since the crystal unit grated, to block the avc circuit of the re-
will be destroyed by the high temperatures ceiver, causing the gain to drop whenever
which can be reached in a closed car parked the engine is speeded up. Since the avc cir-
in the sun in the summer time. cuits of the receiver obtain no benefit from
The use of low -level microphones in mo- a noise clipper, it is important that ignition
bile service requires careful attention to the noise be suppressed enough at the source that
elimination of common -ground circuits in the avc circuits will not be affected even
the microphone lead. The ground connection when the engine is running at high speed.
for the shielded cable which runs from the
transmitter to the microphone should be Ignition Noise The following procedure
made at only one point, preferably directly
should be found adequate
adjacent to the input of the first tube or
transistor in the speech amplifier. The use of for reducing the ignition noise of practically
a low -level microphone usually will require any passenger car to a level which the
the addition of two speech stages, but these clipper can handle satisfactorily at any en-
stages will take only a milliampere or two gine speed at any frequency from SOO kHz
of current. to 148 MHz. Some of the measures may al-
542 Mobile Equipment Design THE RADIO
ready have been taken when the auto re- Voltage Regulator Certain voltage regula-
ceiver was installed. Hash " tors generate an objection-
First either install a spark -plug suppressor able amount of "hash"
on each plug, or else substitute resistor plugs. at the higher frequencies, particularly in the
The latter are more effective than suppres- vhf range. A large bypass capacitor will
sors and on some cars ignition noise is re- affect the operation of the regulator and pos-
duced to a satisfactory level simply by in- sibly damage the points. A small bypass can
stalling them. However, they may not do be used, however, without causing trouble.
an adequate job alone after they have been A 0.001 -fd mica capacitor placed from
in use for a while, and it is a good idea to the field terminal of the regulator to ground
take the following additional measures. with the shortest possible leads often will
Check all high- tension connections for produce sufficient improvement. If not, a
gaps, particularly the "pinch -fit" terminal choke consisting of about 60 turns of No.
connectors widely used. Replace old high - 18 d.c.c. wound on a 3/4 -inch form can be
tension wiring that may have become leaky. added. This should be placed at the regulator
Check to see if any of the high- tension terminal, and the 0.001 -fd bypass placed
wiring is cabled with low -tension wiring, from the generator side of the choke to
or run in the same conduit. If so, reroute ground.
the low tension wiring to provide as much
separation as practical. Generator Whine'. Generator "whine" often
Bypass to ground the 12 -volt wire from can be satisfactorily sup-
the ignition switch at each end with a 0.1- pressed from 550 kHz to 148 MHz simply
d molded -case paper capacitor in parallel by bypassing the armature terminal to
with a .001 -fd mica or ceramic, using the ground with a special "auto radio" capacitor
shortest possible leads. of 0.25 or 0.5 fd in parallel with a 0.001-
Check to see that the hood makes good pfd mica or ceramic capacitor. The former
ground contact to the car body at several usually is placed on the generator when an
points. Special grounding contactors are auto radio is installed, but must be aug-
available for attachment to the hood lacings mented by a mica or ceramic capacitor with
short leads in order to be effective at the
on cars that otherwise would present a
higher frequencies as well as on the broadcast
grounding problem. band.
If the high - tension coil is mounted on When more drastic measures are required,
the dash, it may be necessary to shield the special filters can be obtained which are de-
high - tension wire as far as the bulkhead, signed for the purpose. These are recom-
unless it already is shielded with armored mended for stubborn cases when a wide
conduit. frequency range is involved. For reception
over only a comparatively narrow band of
Wheel Static Wheel static is either static frequencies, such as the 10 -meter amateur
electricity generated by rota- band, a highly effective filter can be im-
tion of the tires and brake drums, or is noise provised by connecting a resonant choke be-
generated by poor contact between the front tween the previously described parallel by-
wheels and the axles (due to the grease in pass capacitors and the generator armature
the bearings). The latter type of noise sel- terminal. This may consist of No. 10
dom is caused by the rear wheels, but tire enameled wire wound on a suitable form
and shunted with an adjustable trimmer ca-
static may of course be generated by all four
pacitor to permit resonating the combina-
tires. tion to the center of the frequency band
Wheel static can be eliminated by inser- involved. For the 10 -meter band 11 turns
tion of grounding springs under the front close wound on a one -inch form and shunted
hub caps, and by inserting "tire powder" in by a 3 -30 -pf compression -type mica trim-
all inner tubes. Both items are available at mer is suitable. The trimmer should be ad-
radio parts stores and from most auto radio justed experimentally at the center frequen-
dealers. cy of the band in use.
www.americanradiohistory.com
HANDBOOK Noise Suppression 543
Vt V2A V3 V4 VS
R -F AMP. MIXER MIXER 2e2-kHz I.F. DETr13F0-AVC
ivy;
RY2A AUDIO GAIN
R2
I = Xi
2775 KC. Ve
R-F GAIN AUDIO
RI
eve REC. TUNE MIC.
05C1 L.
V2B
TRANS. TUNE
V?
AUDIO
SPEAKER
MODULATOR
Figure 11 Ve,V9
gine mount in a similar manner. If a rear - Wheel static, when present, will persist
mounted whip is employed, the exhaust tail when the car clutch is disengaged and the
pipe also should be bonded to the frame if ignition switch turned off (or to the "left"
supported by rubber mounts. position), with the car coasting.
Body noise will be noticeably worse on
Locating Determining the source of a bumpy road than on a smooth road, par-
Noise Sources certain types of noise is made ticularly at low speeds.
difficult when several things
are contributing to the noise, because elimi- 25 -6 A Six -Band Trans -
nation of one source often will make little Receiver for Portable
or no apparent difference in the total noise.
The following procedure will help to isolate
or Emergency Work
and identify various types of noise.
Ignition noise will be present only when While the trend to SSB operation is
strong, there exists a need for amplitude-
the ignition is on, even though the engine
modulated equipment, particularly for emer-
is turning over.
gency operation. Mobile participation in
Generator noise will be present when the emergency service is a cornerstone of ama-
motor is turning over, regardless of whether teur radio; and a compact, all -band trans-
the ignition switch is on. Slipping the drive receiver, suitable for portable and mobile
belt off will kill it. use in emergency work is a convenient unit
Gauge noise usually will be present only for any amateur to have at hand. A small
when the ignition switch is on or in the unit of this type is described in this section
"left" position provided on some cars. and consists of a complete six -band, SO -watt
www.americanradiohistory.com
HANDBOOK Six -Band Trans- Receiver 545
a -m station. For portable work in an emer- simple, stable, and rugged as possible, using
gency, a trans -receiver is much simpler and a single inductor wound on a ceramic form
more convenient to use than the usual com- and a double -bearing capacitor in the fre-
bination of separate receiver and transmit- quency- determining circuit. The vfo oper-
ter. In addition, the trans -receiver can easily ating range is 3.5 to 4.0 MHz. The har-
receive on a different frequency (or band) monics of this range are used for the higher -
than that used for transmitting; a feature frequency bands, except for 6 meters, where
not common to all transceivers. crystal control is used.
Two 6BQ6 TV -type sweep tubes are
Trans -Receiver The trans -receiver consists of triode connected as a class -B modulator
Circuitry a separate receiver and trans-
mitter making use of a com-
stage (V,_ ) in a simplified circuit with
the screens driven, and with the grids at
mon audio system. A block diagram of the cathode (ground) potential.
complete unit is shown in figure 11 and the
schematic is given in figure 12. Trans- Receiver The compact assembly is
The Receiver Section-The receiver is a Controls achieved by utilizing small
dual- conversion circuit which reduces image components, dual -purpose
response by virtue of a high first interme- tubes, and multiple use of some tubes for
diate frequency, and gains selectivity by em- both receiving and transmitting modes.
ploying a second lower- frequency i -f sec- The receiver and transmitter, however,
tion. A total of thirteen tubes is used in operate independently of each other, each
the trans -receiver, including a voltage regu- having its own bandswitch and tuning as-
lator. In the receiver portion, a semiremote- semblies. The receiver has separate r -f and
cutoff 6BZ6 is used in the r-f amplifier (V1), audio gain controls incorporated in a con-
followed by a 6U8A mixer -oscillator (V2A- centric potentiometer along with the power
B), gang -tuned by a dual capacitor (CA- switch (R1, R_, Ss). The automatic noise
B). The first intermediate frequency of limiter and beat oscillator are controlled by
2515 kHz is converted to the second inter- slide switches (S., S3) on the panel, and a
mediate frequency (262 kHz) by a crystal - frequency spotting switch (S ) turns on
controlled 6BE6 second mixer-oscillator the transmitter vfo or crystal oscillator for
(V3). A 6B6 i -f amplifier (V1) is fol- spotting the transmitting frequency in the
lowed by a multipurpose 6T8 (V.,) which receiver. The bfo has no panel pitch con-
combines the functions of second detector, trol, but it may be set as desired by adjust-
noise limiter, and bfo. A second 6U8A is ing the slug of the bfo coil.
used as a pentode audio amplifier (V,;) with The receiver dial tunes the dual capacitor
the triode section used as a microphone am- of the oscillator and mixer stages (C2A-B).
plifier for the transmitter. The 6AQ5 audio A panel trimmer capacitor (CO resonates
stage (V;) performs the dual function of the input circuit of the r-f stage.
receiver audio stage and cathode -follower Few controls are required for the trans-
driver for the 6BQ6 modulator tubes of the mitter section. The tuning dial (C3) sets
transmitter section, by switching a capaci- the transmitter frequency while grid drive
tor between cathode and plate of the on all bands is preadjusted during original
6AQ5 by relay RY,C, thus obtaining the alignment procedure and is broadbanded.
dual circuitry desired. A voltage regulator The tuning and loading capacitors of the
(V,:,) stabilizes the voltage on both receiver final amplifier (C4, C5) provide a variable
and transmitter variable oscillators. match to a portable or mobile antenna sys-
The Transmitter Section -The transmit- tem. Switch S4A-B selects vfo or crystal-
ter portion of the trans -receiver consists of controlled operation. An external carbon
a 6AU6 crystal oscillator or vfo stage microphone plugs into jack J, on the front
(V, ), a 12BY7A tuned buffer -multiplier panel and key jack J:, is placed on the rear
stage and a 6146B final amplifier (V12). apron of the chassis, along with the phone,/
The output circuit of the final amplifier is c -w switch (S;), speaker jack J., coaxial
a pi- network on all bands. The vfo is the antenna receptacle J,, and the heavy -duty
heart of the transmitter and is made as eight- contact power plug. Power switch S.
546 Mobile Equipment Design THE RADIO
Figure 12
www.americanradiohistory.com
HANDBOOK Six -Band Trans -Receiver 547
applies all power to the trans -receiver, plac- in diameter. Small angle brackets affixed to
ing the receiver in operation. Pressing the the chassis at both ends of the pan ensure
microphone switch actuates the control re- rigidity. The completed dials are covered
lays (RY, and RY2), disconnecting voltage with a protective plexiglas sheet.
from portions of the receiver and transfer- A small chassis 4" X 1/2" X I" is used to
1
ring it to the transmitter. The two audio mount the 2515 -kHz i -f transformer (T1),
tubes (V,;, V;) have voltages applied at all the 6BE6 mixer tube (V:1), and the con-
times as their operation is common to both version crystal, plus the first 262 -kHz i -f
modes. transformer (T_). The unit is wired with
connecting leads left long enough to pass
Trans- Receiver The trans- receiver is built on
through a hole in the chassis for connection
Construction an aluminum chassis measur- to the under-chassis receiver -section wiring.
ing 8 %" X 11" which is sup- A separate low- capacitance shielded lead
plied with the perforated cabinet. Layout of (RG- 178/U) connects to the plate terminal
the trans -receiver is conventional, the only of the 2515-kHz i -f transformer and is fed
extra metal work required is fabrication of through a separate chassis hole to make a
a few small brackets and shield plates bent short connection to the plate pin of the
of light sheet aluminum. The dials are made 6U8 mixer tube socket. Note that holes
from a single piece of aluminum bent to must be drilled in the main chassis below the
form a shallow pan 11" long and 4 1/2" subchassis unit in line with the bottom
high, having 3/R" lips on all sides. This sub - slugs of the two i -f transformers so that
panel is bolted to the top of the chassis they may be adjusted from under the chas-
placing the lips flush with the front apron sis when the unit is assembled. The 2515-
of the chassis, leaving the 3/K" depth as a kHz i -f transformer is a 1500 -kHz unit
protected enclosure for the dial mechanisms modified by removing about five feet of
and pointers, and the miniature meter (fig- wire from each winding. The 262 -kHz
ure 13). transformers are standard units, and the
The inside area of this pan is sprayed with bfo transformer (L_ :,) may be made from a
white lacquer (aerosol) and the calibration 262 -kHz input i -f transformer by removing
marks and lettering are done directly on the the mica padding capacitor and using one
lacquered surface with India ink. The ver- winding for the bfo inductance. Silver mica
nier drive mechanisms are removed from padding capacitors are connected externally
their base and are mounted directly on the across the winding.
pan in an %g" cutout. A clear plastic
1
Receiver R -F Section-The r -f portion of
pointer is shaped to fit over the metal dial the receiver is contained in the under -chas-
face and is attached to it by three small
sis area below the receiver tuning capacitor.
screws holding the dial to the vernier as-
sembly. This unit conserves space and makes Two shield partitions separate the r-f coils.
a large, easy -to -read dial nearly five inches Both partitions measure 5" X 15/R" with
548 Mobile Equipment Design THE RADIO
Figure 13
1/4" lipsbent on all sides. They are separated on the switch mechanism. A panel bearing
1 The trimmer capacitors for the various
%err. is placed in the shaft hole of the oscillator
coils mount on the top lip of these parti- partition (rear) to ensure proper alignment
tions, and the bandswitch segments (S,C- of the phenolic switch shaft. The r -f coil
D-E) are mounted on the sides of the parti- partition is 13/4" from the front panel and
tions. Bandswitch segment S,A -B is mounted crosses the center of the 6BZ6 r -f tube
www.americanradiohistory.com
HANDBOOK Six -Band Trans -Receiver 549
nol
Figure 14
socket. A cutout in the partition clears the post of the socket. It is soldered to the
socket, but a small shield of copper "flash- center post and then bolted to the aluminum
ing" material is cut to fit over the center partition to completely shield the input from
550 Mobile Equipment Design THE RADIO
the plate circuit of the r -f stage. Proper RECEIVER COIL DATA
shielding at this point is necessary to pre- Band Coil Range 1MHzI Turns
vent regeneration or oscillation in the r -f Ll 3.5-4.0 loopstick
80 12 (Antenna) 121. No. 32E.
stage. L13 3.5-4.0 Loopstick
Feedthrough capacitors are mounted on 119 6.015-6.515 46T. No. 30E.
the oscillator and mixer side partitions to 13 7.0.7.3 35T. No. 32E.
bring high voltage into the two coil areas. 40 1.4 (Antenna) 71. No. 30E.
The side plates fasten to the partitions to 114 7.0.7.3 321. No. 32E.
9.515-9.815 17T. No. 24E.
complete the shielding and to provide L2e
15 14.0-14.4 201. No. 28E.
rigidity to the assembly. Leads from the L6 (Antenna) ST. No. 30E.
tuning capacitor are brought down to the 20
115
14.0-14.4 20T. No. 20E.
various bandswitch segments through holes Center-Tap.
16.515.16.915
in th; chassis. The tuning capacitor corn - L21 81. No. 20E.
www.americanradiohistory.com
HANDBOOK Six -Band Trans -Receiver 551
to the 6146B. The space below the loading ceiver alignment tests. The i -f section is
capacitor is used for placement of the 25K, aligned first by applying a 2515 -kHz test
10 -watt screen- dropping resistor and the signal to the input grid of the 6U8A mixer
5 -ohm, 10 -watt filament- balancing resistor. tube (pin 2, V2A) and adjusting the slugs
The amplifier tank coil (L:,;-L:10 is made of the three i -f transformers for maximum
in two sections as shown in the coil table signal. This may be judged by ear or by a
(figure 16) with the low- frequency section v.t.v.m. placed on the avc line. If the con-
mounted horizontally above the chassis on version oscillator is working, the complete
ceramic insulators. The 10- and 6 -meter i -f section may be aligned with the 2515 -
coil is placed in a vertical position below a kHz test signal.
large rectangular cutout in the chassis Using the given coil data, all bands are
which permits short connections between spread out over nearly 180 degrees of the
coil and bandswitch. receiver dial. Except for the 6 -meter band,
Wiring the coils in the transmitter section all oscillator circuits tune to the high -fre-
presents no problems. The short under -chas- quency side of the incoming signal. Align-
sis shield partition near the panel supports ment procedure is the same for all bands.
switch segments S:,A and S;,B (one on each The frequency range of each mixer oscil-
side of the panel) and the two oscillator lator circuit is monitored with a calibrated
coils L,; and L_.. All coils for the 12BY7A receiver or BC -221 (LM) frequency meter
buffer stage are placed between the two and is adjusted with the individual trim-
partitions with their alignment screws pro- mer capacitor and by varying the spacing
jecting out of the top of the chassis. The of the turns on the oscillator coil, if neces-
air -wound 6 -meter buffer coil (L,,,) is sary. Alternatively, alignment can be done
soldered between a terminal of the band - by checking the frequency range of the
switch (S:,B) and a tiepoint on the chassis. oscillator with a grid -dip meter. The r-f
www.americanradiohistory.com
HANDBOOK Transistor Power Supplies 553
Figure 18
switch on the push -to -talk circuit of the The switch is actuated by a magnetic coil
microphone. and breaker circuit requiring appreciable
power which must be supplied by the pri-
mary source.
25 -7 Transistor One of the principal applications of the
Power Supplies transistor is in switching circuits. The tran-
sistor may be switched from an "off" con-
The vibrator -type of mobile supply dition to an "on" condition with but the
achieves an over-all efficiency in the neigh- application of a minute exciting signal.
borhood of 707(. The vibrator may be When the transistor is nonconductive it
thought of as a mechanical switch reversing may be considered to be an open circuit.
the polarity of the primary source at a When it is in a conductive state, the in-
repetition rate of 120 transfers per second. ternal resistance is very low. Two transistors
properly connected, therefore, can replace
W
the single -pole, double -throw mechanical
u switch representing the vibrator. The tran-
RISE TIME = !O//S
0 2.5 VOLTS nt- sistor switching action is many times faster
.200 -2000 -
> 35-. US than that of the mechanical vibrator and
p 30 the transistor can switch an appreciable
,(:j) 25 amount of power. Efficiencies in the neigh-
20 --
O
borhood of 95 percent can be obtained with
28 -volt primary -type transistor power sup-
10.-
W
N
a
CO
5-
0_t
0.5 vOLT
r-- 200-
us
plies, permitting great savings in primary
power over conventional vibrators and dyna-
motors.
TIME - Transistor The transistor operation resembles
Figure 19 Operation a magnetically coupled multivi-
brator, or an audio- frequency
BASE -COLLECTOR WAVEFORM OF
push -pull square wave oscillator (figure
SWITCHING CIRCUIT, 18C). A special feedback winding on the
FOR 12 -VOLT CIRCUIT
power transformer provides 180 degree phase-
Square waveshape produces almost ideal shift voltage necessary to maintain oscilla-
switching action. Small "spike" on leading
edge of pulses may be reduced by proper tion. In this application the transistors are
transformer design. operated as on -off switches; i.e., they are
554 Mobile Equipment Design THE RADIO
2N27
+ 250V. AT SO MA =12 WATTS
2N278 oR2N301A
T2
Figure 20
either completing the circuit or opening it. that an increase in load produces very little
The oscillator output voltage is a square extra heating of the transistors. This feature
wave having a frequency that is dependent means that it is impossible to burn out the
on the driving voltage, the primary in- transistors in the event of a shorted load
ductance of the power transformer, and since the switching action merely stops.
the peak collector current drawn by the
conducting transistor. Changes in trans- Transistor The power capability of the
former turns, core area, core material, and Power Rating transistor is limited by the
feedback turns ratio have an effect on the amount of heat created by the
frequency of oscillation. Frequencies in com- current flow through the internal resistance
mon use are in the range of 120 Hz to of the transistor. When the transistor is con-
3500 Hz. ducting, the internal resistance is extremely
The power consumed by the transistors low and little heat is generated by current
is relatively independent of load. Loading flow. Conversely, when the transistor is in
the oscillator causes an increase in input a cut -off condition the internal resistance is
current that is sufficient to supply the re- very high and the current flow is extremely
quired power to the load and the additional small. Thus, in both the "on" and "off"
losses in the transformer windings. Thus, conditions the transistor dissipates a min-
the over -all efficiency actually increases imum of power. The important portion of
with load and is greatest at the heaviest load the operating cycle is that portion when
the oscillator will supply. A result of this is the actual switching from one transistor to
www.americanradiohistory.com
HANDBOOK Transistor Power Supplies 555
Figure 22
SCHEMATIC, TRANSISTOR
Di Da
POWER SUPPLY FOR
12 VOLT AUTOMOTIVE
SYSTEM
T -Transistor power transformer.
12 -volt primary, to provide 275v.
275 volts at 125 ma. Chicago A7125 MA
Stancor DCT -1
D -D -Sarkes- Tarzian silicon recti-
fier, type M -500, or equiva-
lent
- 12 V. BATTERY +
er transformer design can reduce this "spike" The complete power package is built on
to a minimum value. An excessively large an aluminum chassis -box measuring 5 %s"
"spike" can puncture the transistor junc- X 3" X 2". Paint is removed from the
tion and ruin the unit. center portion of the box to form a simple
heat sink for the transistors. The box there-
25 -8 Two fore conducts heat away from the collector
Transistorized elements of the transistors. The collector of
Mobile Supplies the transistor is the metal case terminal
and in this circuit is returned to the nega-
The new Stancor Electronics Corp. series tive terminal of the primary supply. If
of power transformers designed to work the negative of the automobile battery is
in transistor -type power supplies permits grounded to the frame of the car the case of
the amateur and experimenter to construct the transistor may be directly grounded to the
efficient mobile power supplies at a fraction
of their former price. Described in this sec-
tion are two power supplies designed around
these efficient transformers. The smaller sup-
ply delivers 35 watts (275 volts at 125 mil-
liamperes) and the larger supply delivers 85
watts (500 volts at 125 milliamperes and
250 volts at 50 milliamperes, simultaneous-
ly). Both power units operate from a 12-
volt primary source.
The 35 -Watt The 35 -watt power unit uses
Supply two inexpensive RCA 2N-
301A PNP -type power tran-
sistors for the switching elements and four
silicon diodes for the high- voltage rectifiers.
The complete schematic is shown in figure
22. Because of the relatively high switching
Figure 23
frequency only a single 20 -pfd filter capaci-
tor is required to provide pure direct cur- UNDER -CHASSIS LAYOUT OF PARTS
rent. Two 10-Ad capacitors are connected in parallel
Regulation of the supply is remarkably for 20 -fd
output filter capacitor. Silicon recti-
fiers are mounted in dual fuse clips at end of
good. No -load voltage is 310 volts, dropping chassis. Transistor should be insulated from the
to 275 volts at maximum current drain of chassis with thin mica sheets and fibre washers
if supply is used with positive -grounded pri-
125 milliamperes. mary system.
www.americanradiohistory.com
HANDBOOK Trorsistor Power Supplies 557
2N27B Figure 24
SCHEMATIC,
85 -WATT
TRANSISTOR
POWER SUPPLY
FOR 12 -VOLT
AUTOMOTIVE
SYSTEM
T- Transistor power trans-
former. 12 -volt pri-
mary to provide 275
volts at 125 ma. Chi-
cago 5tancor DCT -2.
D -D -Sarkes -Tarzian sili-
con rectifier, type M-
500
unpainted area of the chassis. If the positive insulators. These may be used in place of
terminal of the car battery is grounded it is the mica washers, if desired. The under -
necessary to electrically insulate the transis- chassis wiring may be seen in figure 23.
tor from the aluminum chassis, yet at the
The 85-Watt Figure 24 shows the schematic
same time permit a low thermal barrier to
exist between the transistor case and the
Supply of a dual -voltage transistor
power supply chassis. A simple method of mobile power supply. A
accomplishing this is to insert a thin mica bridge rectifier permits the choice of either
sheet between the transistor and the chassis. 250 volts or 500 volts, or a combination of
Two -mil (0.002") mica washers for tran- both at a total current drain that limits the
sistors are available at many large radio secondary power to 85 watts. Thus, 500
supply houses. The mica is placed between volts at 170 milliamperes may be drawn,
with correspondingly less current as addi-
the transistor and the chassis deck, and fibre
tional power is drawn from the 250 -volt
washers are placed under the retaining nuts
tap.
holding the transistor in place. When the The supply is built on an aluminum box
transistors are mounted in place, measure the chassis measuring 7" X 5" X 3", the layout
collector to ground resistance with an ohm- closely following that of the 35 -watt supply.
meter. It should be 100 megohms or higher Delro 2N278 PNP -type transistors are used
in dry air. After the mounting is completed, as the switching elements and eight silicon
spray the transistor and the bare chassis diodes form the high -voltage bridge rectifier.
section with plastic Krylon to retard oxida- The transistors are affixed to the chassis in
tion. Several manufacturers produce anodized the same manner as the 2N301A mounting
aluminum washers that serve as mounting described previously.
CHAPTER TWENTY -SIX
Receiver construction has just about be- he never uses, and the phone man must pay
come a lost art. Excellent general coverage for the c -w man's narrow -band filter. For
receivers are available on the market in many one amateur, the receiver has too much
price ranges. However, even the most mod- bandspread; for the next, too little. For
est of these receivers is relatively expensive, economy's sake and for ease of alignment,
and most of the receivers are designed as a low -Q coils are often found in the r -f cir-
compromise-they must suit the majority of cuits of commercial receivers, making the
users, and they must be designed with an eye set a victim of crosstalk and overloading
to the price. from strong local signals. Rarely does the
It is a tribute to the receiver manufac- purchaser of a commercial receiver realize
turers that they have done as well as they that he could achieve the results he desires in
have. Even so, the c -w man must often pay a homebuilt receiver if he left off the frills
for a high- fidelity audio system and S -meter and trivia which he does not need but which
Figure 1
COMPONENT NOMENCLATURE
RESISTORS
CAPACITORS: I- RESISTANCE VALUES ARE STATED IN OHMS, THOUSANDS
Of OHMS (K), AND MEGONMS (M ).
I- VALUES BELOW 999 PF ARE INDICATED IN UNITS.
EXAMPLE, 270 OHMS 270
EXAMPLE: ISO PF DESIGNATED AS ISO.
700 OHMS 4 7 K
2- VALUES ABOVE 999 Pf ARE INDICATED IN DECIMALS. 33000 OHMS
100,000 OHMS' 100K
' S3 K
OR 0.1 M
EXAMPLE. OOSLFD DESIGNATED AS .005. 33,000,000 OHMS' 33 M
3- OTHER CAPACITOR VALUES ARE AS STATED. 2- ALL RESISTORS ARE 1 -WATT COMPOSITION TYPE UNLESS
EXAMPLE: I0,UFD, 0.3PF, Esc.
- TYPE OF CAPACITOR IS INDICATED BENEATH THE VALUE
DESIGNATION.
OTHERWISE NOTED. WATTAGE NOTATION IS THEN INDICATED
BELOW RESISTANCE VALUE.
4711
EXAMPLE: rci-
SM= SILVER MICA
C' CERAMIC
M. MICA
P. PAPER INDUCTORS:
200, .01 .M1 MICROHENRIES =LH
EXAMPLE:
MILLIHENRIES NH
HENRIES= H
S- VOLTAGE RATING OF ELECTROLYTIC OR ^FILTER
CAPACITOR IS INDICATED BELOW CAPACITY DESIGNATION.
EXAMPLE,
0
Io
.
a.* tII
SCHEMATIC SYMBOLS1
0,9
+
1
CONDUCTORS JOINED
I- THE CURVED LINE IN CAPACITOR SYMBOL REPRESENTS
'
THE OUTSIDE FOIL GROUND OF PAPER CAPACITORS,
THE NEGATIVE ELECTRODE Of ELECTROLYTIC CAPACITORS,
II I
www.americanradiohistory.com
The Transceiver 559
www.americanradiohistory.com
560 Receivers and Transceivers THE RADIO
PARALLEL
6CW4 WIRES
2 6U6A to ,+116
R -F AMP. O.5PF 470 MIXER V.
2
(SOMHr)1 51MH,)
6 'Hp (/I) (17MH!)
1
L2 ! L3 4 M LS
J2
20 14 -15 MNI I -F OUTPUT TO
M RECEIVER
10 V.001
J1 T T
50-54 WM .001 -
= L1
470 (SOMHI) .001
C
P1
105V
+90 V. HEATERS
6.3 V.
1L6 (36M/U
NOTES, ALL RESISTORS Ixt -WArr AURA
UNLESS OTHERWISE SPECIFIED. 3e-hot.. 5
C. CERAMIC CAPACITOR RTAL A
Figure 3
or spots placed on the body of the compo- built on a small piece of etched circuit,
nent. Several color codes have been used copper -clad, paper -base phenolic board in-
in the past, and are being used in modified stead of the usual aluminum or plated -steel
form at present to indicate component chassis.
values. The most important of these color
codes for resistors, capacitors, power trans- Circuit Description The schematic of the 50-
formers, chokes, i -f transformers, etc. can MHz converter is shown
be found in the appendix at the end of this in figure 3. A grounded -grid 6CW4 nuvis-
Handbook. tor triode is used as an r -f amplifier. The
r -f input circuit (L0) is made of a single
tapped length of miniductor coil material.
26 -1 A Nuvistor Converter This circuit is broadly resonated by the in-
for Six Meters put capacitance of the 6CW4. The nuvistor
plate circuit is lightly coupled to a 6U8A
The nuvistor series of miniature tubes (pentode section) mixer by means of a
brings low- noise -level vhf reception within small capacitor made of two lengths of par-
the economic capability of the average radio allel wire having a capacitance of about
amateur. Described in this section is a sim- 0.5 pf.
ple and reliable 50-MHz converter (figure The triode section of the 6U8A is the
2) that makes use of the 6CW4 nuvistor mixing oscillator employing a third -over-
vhf triode. The inherent noise figure of this tone 36 -MHz crystal. This choice of mixing
converter is about 3 decibels which compares frequency produces an intermediate- frequen-
favorably with units employing special and cy range of 14 to 18 MHz for signals rang-
costly low -noise tubes. The converter is ing in the 50- to H -MHz band. Coupling
www.americanradiohistory.com
HANDBOOK A Nuvistor Converter 561
Figure 4
UNDER -CHASSIS VIEW OF CONVERTER
Layout of the principal components are shown in this view. The shield of copper -clad phenolic
board is seen placed in a vertical position between L. and L The shield is soldered to the
.
chassis board. Two parallel insulated wires extending through a hole in the board make up
the interstage coupling capacitor. The antenna input receptacle (J,) and r -f coil are at the
left, and crystal oscillator plate coil L, is near the bottom of the photograph, just right of
center. The i -f coils are at the extreme right, with the output receptacle (J ) immediately
above coil L .
between the oscillator and mixer stage is copper laminate board simplifies the job of
both capacitive and inductive by virtue of obtaining good r -f grounds as components
the spacing between coils L3 and L6. may be soldered directly to the board. The
The pentode mixer plate coil (L, and L:;) use of this material also reduces construc-
are stagger -tuned over the 14- to 18 -MHz tion time as soldering lugs do not have to be
i -f range for relatively flat converter gain. mounted for the various ground connec-
tions. Soldering is easily done with a 25 -watt
Converter Construction The 50 -MHz con- "pencil" iron. Increased electrical stability
verter is built on a is the result of this construction technique
4" X 6" sheet of copper laminate (two which more than offsets the slight additional
sides) phenolic circuit board (figure 4). A care required in drilling and cutting the
4" X 6" X 2" inverted aluminum chassis board to avoid flaking the thin layers of
serves as a shield box. The use of inexpensive bonded copper.
562 Receivers and Transceivers THE RADIO
' .+ 1. w 9 ++ --F s- 7
Ls _ i TOP VIEW
: I
xrAL
TD
---1
1
4. --
1.-CHASSIS
BONDIR
I +
aa
1-. ---- N D LIJ
TT
}
BOTTOM EDGE
2
SHIELD
Figure 5
www.americanradiohistory.com
HANDBOOK An "Antioverload" Converter for 50 MHz 563
Figure 6
l
7
.-i
I
470 TO RECEIVER
lo +1.0V.
J C
2.2 K
c
470
6e = in G
1W.
HEATERS O5C.
f
1,5
6BA7 6C 15K
3-MHZ i lw.
+105v.
41
I
3613326 5 3 4 %TAL 3 MA.
PIN 2
PI 1 T1
470
C
476L .00E. 470
Figure 7
The 50 -MHz converter described in this mixing oscillator. Coupling between the r -f
section (figure 6) is a popular design on the amplifier plate coil (L1) and the 6BA7
West Coast, particularly in areas of high mixer grid coil (L,;) may be varied to suit
6 -meter activity and areas in proximity to the strong -signal situation by adjustment of
high -power Channel -2 television transmit- link coils L, and L.
ters. The converter has good sensitivity, an The 6C4 crystal oscillator is capacitively
acceptable noise figure of about 5 decibels, coupled to the 6BA7 mixer, and the former
and excellent high -signal overload char- employs a 43 -MHz third- overtone crystal
acteristics. It is the "city ham's" 50 -MHz to produce a 7- to 11 -MHz intermediate
converter par- excellence. frequency for signals in the range of 50 to
54 MHz. Plate voltage for the 6C4 oscilla-
Circuit Description The schematic of the
tor is fed separately from the rest of the
antiorerload converter is
converter so the oscillator may be turned
shown in figure 7. A 6BZ6 semiremote -cut- off during periods of transmission.
off pentode is used as an r -f amplifier. The
6BZ6 has a cathode gain control (R1) to Converter Construction The 50 -MHz convert-
permit adjustment of stage gain when strong er is built on a 4" X
local signals are encountered. A 6BA7 6" copper -laminate (two sides) phenolic
pentagrid mixer having an exceptionally circuit board. A 4" X 6" X 2" aluminum
large dynamic signal range is used, in con- chassis box serves as a base and a shield for
junction with a 6C4 fundamental -frequency the wiring and components. All components
www.americanradiohistory.com
HANDBOOK An "Antioverlood" Converter for 50 MHz 565
L.
a-
FRONT EDGE J
2
3
TOP
}
POTENTIOMETER
BEND UP
G0
Figure 8
except the manual gain control (R1) are Placement of the major components may
mounted on the chassis plate. The gain be seen in the underchassis view (Figure 9)
control mounts on the chassis box as shown and the drilling layout (Figure 8). The
in Figure 8. A short length of flexible hook- three 50 -MHz coils are made of sections of
up wire connects the potentiometer terminal niiniductor coil stock. A heated razor blade
to a phenolic tie -point mounted on the held with pliers is a good tool to sever the
plate. The ground terminal of the gain con- plastic ribs when cutting the coil stock to
trol is grounded to the plate by a separate length. The coils are mounted in place by
lead so that an electrical connection is made their leads.
if the converter is operated outside of the A small shield is placed across the 6BZ6
box. This permits the chassis plate to be socket to prevent oscillation of the r -f stage.
swung up and out while the converter is The shield is cut from thin "flashing" cop-
aligned and tested. per and measures 2" long by 3/44' high. It
The usable area of the chassis plate is 3" is soldered to the chassis plate at either side
X 5" since a one -half inch border around the of the socket and to the socket center post.
plate must be left to clear the chassis base The shield should be placed in position be-
lip. All parts (except the gain control) fore wiring is started.
must be located inside the border. Self - Most of the small components are sup-
tapping screws secure the chassis -plate to ported by their leads, between socket pins,
the box. or from socket pin to nearby phenolic tie-
566 Receivers and Transceivers THE RADIO
Figure 9
point terminals. All leads are short and the r -f and mixer stages. The 6C4 may be
direct. The drilling layout shows the lo- operated from the 250-volt supply without
cation of the major components. If other regulation if an additional series dropping
components are substituted for the ones resistor of 50,000 ohms, 2 watts is placed be-
used, it would be wise to check the layout tween pins 3 and 4 of plug P,. Filament
before drilling any holes as space is rather power for the converter is 6.3 volts at 0.75
limited in some areas.
ampere.
Adjustment of A regulated supply voltage All wiring should be checked before
the Converter of 105 volts at 3 milliamperes power is applied to the converter. The 6C4
is required for the oscillator oscillator is checked in the manner described
and 250 volts at about 25 milliamperes for in the previous section of this chapter. The
www.americanradiohistory.com
HANDBOOK A Nuvistor Two -Meter Converter 567
6BA7 and 6BZ6 tubes are inserted in the be stagger- tuned. Capacitor C2 is the
proper sockets, the converter is attached to sharpest tuning of the three circuits, and
a receiver tuned to 7 MHz and a low -level should be peaked at the center of the oper-
50 -MHz signal is injected into antenna re- ating range to be covered. Coil Ls should be
ceptacle (J,) of the converter. The various adjusted to provide maximum gain at the
tuning capacitors are adjusted for maximum center of the operating range. As a starter,
signal. capacitor C, should be peaked at 50.5 MHz,
Final adjustment of coupling coils L, and capacitor C2 at 51 MHz, and capacitor C:,
L; should be done after the user has had at 52 MHz. If the converter is to be used
experience with the converter in the pres- only in the lower one megahertz of the
ence of strong local signals. With all cir- band, all circuits may be peaked at 50.5
cuits peaked for maximum signal, the link - MHz. Adjustment of link coil L, and tun-
coupling coils should be adjusted for mini- ing of input capacitor C1 have an effect on
mum signal consistent with good reception the noise figure of the converter. These ad-
and the prevailing state of nearby strong justments may be made on a weak signal or
signals. Too close coupling will limit the with the aid of a noise meter.
ability of the converter to withstand strong
local signals and too loose coupling will re- 26 -3 A Nuvistor Two -Meter
sult in an excessive loss of gain. Adjust-
ment of the link coils coupled with experi-
Converter
mentation with the gain control will achieve
the ultimate in usable sensitivity and excel- The wide acceptance of the nuvistor
lent overload capability. tube as an r -f amplifier for vhf reception
For flattest response across the whole 50- has shown its superiority over conventional
MHz band, capacitors C1, Cand C, should triodes for weak -signal reception. When
used with a nuvistor tetrode mixer stage,
moreover, the performance of the 6CW4
as a low -noise r -f amplifier is considerably
enhanced.
Described in this section is a three- nuvis-
tor converter for 144 MHz that exhibits a
noise figure close to 3 decibels (figure 10) .
The circuit may be modified for "antiover-
load" characteristics for protection from
strong local signals while still retaining its
excellent noise figure.
Circuit Description The schematic of the
nuvistor two -meter con-
verter is given in figure 11. A 6CW4 is
used as a low -noise r -f amplifier in a neutral-
ized, grid- driven configuration. The mixer
stage uses a 7587 nuvistor tetrode which
Figure 10 combines low lead inductance with ex-
tremely high transconductance and re-
OBLIQUE VIEW OF TWO METER
duced input loading. A minimum of local -
NUVISTOR CONVERTER
oscillator injection is required and as the
This all-nuvistor converter provides superior tube has high conversion gain, it provides
performance on the two -meter band. Across
the near edge (left to right) are the slug - good i -f output signal voltage.
tuned output coil (L ), the 7857 nuvistor, the A second 6CW4 is used as a "hot -cath-
interstage coils (L; and L,), followed by the
6CW4 nuvistor and r -f tuning capacitor C,. ode" conversion oscillator employing a
The 6CW4 crystal oscillator stage is at the 39.33 -MHz overtone crystal. The third
back (right) with coaxial receptacle .1 near harmonic of the crystal frequency appears
the crystal. The power plug is at the rear
(left). The complete converter is built on a in the plate circuit of the 6CW4 stage (118
copper -plated, phenolic -base circuit board. MHz) providing an intermediate- frequency
568 Receivers and Transceivers THE RADIO
JI
LI
o
Le
0005y1
-e- .
6CW4+65
5O
u
(745MHZ)
V.
*L1
Lz
6.66
C3
I
(7L;MHI)
_
7587
+103
21150 v
/SOOT
M
7V.
V.
IeH
NO '470
C430
(E4MHr)
L4
) L5 J2
26-30
26- -MHz 1- FOUTPUT TO RCCCI V CR
6 T50
M J M PLUG
PI
+105V. HEATERS 6.3 V.
O 0
Cs (SEE rEXT) 001
.001-T-
111
4 2
6CW4
L7 2 XTAL p
(779MN7) (3 r.33MNt) BLANK'
NOTES: ALL RESISTORS 7/2 -WATT UNLESS
OTHERWISE SPECIfIEO.
6////CW4 6CW4 7567
G. CERAMIC CAPACITOR
M. SILVER MICA CAPACITOR
10/12
500
-e-
Figure 11
SCHEMATIC OF TWO -METER NUVISTOR CONVERTER
C,-0.5- to 5 -pf tubular trimmer (Erie 532A or equiv.)
Cr C,- 3.3 -pf tubular ceramic (Centralab TCZ -3R3 or equiv.)
C,-2.2 -pf tubular ceramic (Centralab TCZ -2R2 or equiv.)
C5,C. -30 -pf ceramic (Centralab DD or equiv.)
J BNC receptacle type UG -625/U
L, -5 turns #16 bare wire, (/4" diameter, spaced wire diameter. Tap 2 turns up and adjust for
best noise figure
L,, L;4 turns #26 enamelled wire, y4" diameter, close wound on slug -tuned form (CTC -PLST
or equiv.)
L5-11 turns #26 enamelled wire, 3/e" diameter, closewound on slug -tuned form (CTC -LS3 or
equiv.)
LS 3 turns insulated wire closewound around 8-plus end of L,
1,-5 turns #26 enamelled wire, 3 /j'
diameter, closewound on slug -tuned form (CTC-LS3 or
equiv.)
L-7 turns #26 enamelled wire, 1/4" diameter, closewound on slug -tuned form (CTC -PLST or
equiv.)
6,-25 turns #30 enamelled wire, wound on 1- megohm, %2 -watt resistor, approximately 5/16"
long; adjust for neutralisation (see text)
Note: All S00 -pf ceramic capacitors are disc type (Centralab DD-S01 or equiv.).
All 500 -pf silver mica capacitors are silver button type (Erle 370-C8-501K or equiv.).
Nuvistor sockets are Cinch 133 -65 -10-0.011 or equiv.
range of 26 to 30 MHz. For lower i -f of copper -laminate (two sides) phenolic cir-
ranges, only the crystal and the i -f output cuit board. A similar size inverted aluminum
coil (L,) need be changed. If an i -f range chassis 1 %z" deep is used as a shield box.
of 14 to 18 MHz is desired, a conversion Layout of major components, the nuvistor
crystal frequency of 43.3 MHz should be socket shield, and the nuvistor mounting
used. The plate circuit of the oscillator is
hole are shown in figure 12. Because of their
now tuned to 130 MHz, however no
small size, nuvistor sockets are clamped
changes are necessary in the tuned circuit.
Output coil L, requires 22 turns for cover- (rather than bolted) to the chassis by bend-
ing two lugs on the socket. After the chassis
age of the 14- to 18 -MHz range.
hole is drilled, two notches are filed to en-
The oscillator is coupled to the grid of
sure a tight fit of socket to chassis. For
the mixer stage by the stray capacitance of
a wire run from the grid end of mixer coil
grounding, both socket lugs are soldered
L3 to an unused lug on the plate end of
to the chassis. All ground connections for
the nuvistor socket should be made to the
oscillator coil L7. No further coupling is
socket lugs, except in the case of the r -f
required.
amplifier, which uses the base shield as the
Converter Construction The converter is built common ground return. This shield is cut
on a 3" X 4" sheet from a thin piece of copper and is soldered
www.americanradiohistory.com
HANDBOOK A Nuvistor Two-Meter Converter 569
DIA.
T SHIELD PLATE NUVISTOR SOCKET HOLE if
SEND 90
a.
(f
4 2
POWER PLUG
Le BNC COAX
NTENNAI
32 2
+
SOCKET
5NC COAX
(OUTPUT)
"28 > r DIA.
L7
STANDOFF +f20 3"
HOLE POR
TOP CAP
LEAD y +! DIA.
INSULATOR
+
C9 C7 Ce
+*38 + 538 C1
+ 16 DIP
*20 +
L4 -LS L3 L2
{ DIA. + +
*20
+
20
Ce
+ 38 SEE FIG FO
CHASSIS LAYOUT
Figure 12
DRILLING LAYOUT FOR CONVERTER BOARD (G), BASE SHIELD LAYOUT (E),
AND NUVISTOR SOCKET HOLE (F).
to socket pins 8 and 10 and to the chassis. tuning eliminates the need for squeezing
As in all vhf construction, good grounds and adjusting the coils for correct tuning.
are essential and all ground- return leads Use of the layout of figure 12 ensures that
should be short. Connection to the top cap the coils are mounted in the proper posi-
of the 7587 is best made with a short piece tion with unwanted coupling and spurious
of steel (piano) wire looped into a tight -fit- feedback paths eliminated.
ting, one -turn coil. Note that all components are held clear
All coils except the input coil have been of the chassis edge so that they do not inter-
wound on slug -tuned forms to provide neat fere with the lip of the chassis box (figure
construction and ease of alignment. Slug- 13).
570 Receivers and Transceivers THE RADIO
PPP-
_
" 0
40, A 1 *;
r` r
.031
Q
Figure 13
www.americanradiohistory.com
HANDBOOK A Nuvistor Two -Meter Converter 571
receiver should fall off. A slight adjustment r -f stage socket. Neutralizing coil L. is ad-
of oscillator cathode coil L, may be neces- justed by starting with a few extra turns
sary to start oscillation. Oscillator plate coil and removing one turn at a time to find the
L7 should be peaked for maximum oscilla- point of minimum signal feedthrough when
tor output. the other tubes are operating. This adjust-
Tune in a signal about 145 MHz and ad- ment is not critical.
just r -f amplifier plate coil L for maximum
indicated signal. Repeat at 147 MHz and "Antioverlood" Strong local two -meter sig-
adjust mixer grid coil L. Find a signal near Modification nais may, under certain cir-
146 MHz and adjust the r -f input circuit cumstance~, cause overloading of any two -
(L1, CO for maximum signal. meter converter designed for low noise figure
The r -f amplifier should be properly and weak- signal reception. Cross modula-
neutralized for best noise figure. The fila- tion can be reduced in this converter by
ment lead should be opened at the 6CW4 the use of automatic gain control (agc or
Figure 15
FRONT VIEW OF TRANSCEIVER
The transceiver panel measures 1218" wide by 658" high. The two large controls at center
are for final amplifier tank and vfo tuning. On the left area of the panel are the modulator
balance control (top), r -f gain adjustment, receiver volume, and microphone gain control
(next to the microphone jack). The lower switch is the main power control (S ) and the meter
switch is at the top, right. Below the plate tuning control are the grid tuning adjustment and
the function switch, S On the right of the panel are the carrier level control, R , and the
.
antenna loading capacitor, C The cabinet is a wrap- around style made from two pieces of
.
perforated aluminum sheet bent into a U -shape and riveted together at the sides. Panel and
cabinet are primed and painted with aerosol (spray) paint.
572 Receivers and Transceivers THE RADIO
avc) and the substitution of a 6DS4 semi - The source of the agc voltage in the com-
remote -cutoff nuvistor for the sharp -cutoff munications receiver may be found by
6CW4. The agc control voltage is taken studying the schematic and locating the agc
from the communications receiver used as line in the chassis wiring. The agc voltage
the i -f strip. Because the agc voltage in a (as measured with a vacuum -tube volt-
typical communications receiver is not de- meter) should vary from zero volts at no
veloped until a reasonably strong signal is signal to about -10 volts at maximum sig-
received, the converter still retains maxi- nal level. The receiver must have zero volts
mum sensitivity for weak -signal reception. on the agc line in the absence of signals or
Modification of the converter is simple. this system will not work.
The 6DS4 is substituted for the 6CW4 and
one resistor and two capacitors are added to
the circuit as shown in figure 14. The agc
26 -4 A Single -Band
control voltage is obtained from the re- SSBTransceiver
ceiver. The original grid resistor (47K) is
lifted from ground and rewired through
new resistor R, to the spare contact on the
power plug. Capacitors C. and C are added
Probably the most popular item of equip-
ment for SSB operation is the transceiver
a complete station in one compact package.
-
close to R1. Since many of the tubes and components
V, Vz Ti V3 Va VS
MIC PI-
NETWORK
CI Ca
TRANSMITTER
LCoTo'r N J SECTION
Ve V7 e TuNE
YI
z FREQUENCY CONTROL
SECTION
SEE COIL DATA FOR TUN/NG RANGE
RECEIVER SECTION
TO PI-NET.
Ve V,o V,1 Ta V,z V,S VI
TUNED
9MHz I.F.
CIRCUIT
EXTERNAL
SPEAKER
Figure 16
www.americanradiohistory.com
HANDBOOK A Single-Band SSB Transceiver 573
are common to both the transmitting and A 12AU6 (operated at slightly reduced fila-
receiving functions, the transceiver can be ment voltage) serves as the oscillator tube
built compactly and rather inexpensively, (Vs).
and it is well suited for both fixed -station The 9 -MHz i -f signal passes through the
and mobile operation. selective crystal lattice filter (ACF -4) and
The most economical and least complex is amplified in a common i -f stage (V,)
transceiver to build is one designed for use which is transformer coupled to a second
on a single amateur band. Multiple mixing (receiving) i -f stage (V ,,) and then fed
schemes and complex coil catacombs are to a product detector (V12). At this point
thus eliminated, and the "birdie" problem in the circuit, carrier is injected in the
is greatly simplified. Shown in this section detector from the 6U8A common crystal
is a 200 -watt PEP, single -band transceiver oscillator (VG) and the resulting audio
(figure 15) which may be used on any one product is amplified in one -half of the
amateur band from 160 to 20 meters. It 12AX7 dual triode (V) and the 6AQSA
is relatively simple in design and is an ideal output tube (V ,,) . A portion of the audio
"first" project for those amateurs interested signal returns to the 6ALS automatic gain
in building their own sideband gear. While control rectifier (V,:,) to provide an audio -
a commercial 9 -MHz crystal filter is used, derived agc voltage for the receiver section.
substitution of a homemade crystal filter is A fixed positive voltage taken from the
practical, further reducing the cost of the cathode of the 6AQSA stage provides delay
transceiver. voltage for the agc circuit to allow maxi-
mum receiver sensitivity to be realized with
The Transceiver The transceiver circuit is weak signals. Receiver volume is controlled
Circuit a proven one that has been in the grid of the 6AQSA stage instead of
employed in many com- the low -level audio circuit so that agc action
mercial units and is a version of the original is independent of the audio volume level.
W6QKI (Swan) circuit. Fifteen tubes are Transmission-In the transmitting mode,
used, including a voltage regulator and the the circuit takes the form of a single -con-
unit is designed to be operated from either version, crystal -filter SSB exciter, featuring
a 115/230-volt a -c primary supply or a a 7360 balanced modulator and a 6DQS
12 -volt transistor power pack (external). linear amplifier. Switching the circuitry
Operation of the single -band SSB trans- from receive to transmit is accomplished by
ceiver and the dual function of some of the a single relay (RY) which applies blocking
tubes and tuned circuits may be seen from bias ( -100 volts) to inactivate tubes used
an inspection of the block diagram of fig- only in the receiving mode. The relay also
ure 16. applies screen voltage to the 6DQS r -f am-
Reception -In the receiving mode, the plifier (V,,) and grounds the cathode of
circuit takes the form of a single- conversion the common 6BA6 i -f amplifier stage to
superheterodyne featuring product detection. nullify the receiving r -f gain control dur-
The received SSB signal is resonated in the ing transmission. The receiver r -f amplifier
antenna input circuit which, in this case, is stage remains connected to the plate circuit
the pi- network of the transmitter portion of the linear amplifier of the transmitter
of the unit. The network is capacitively section, but the 6BA6 amplifier is protected
coupled to a 6BA6- remote cutoff r -f ampli- from strong- signal damage by virtue of the
fier (V). The plate circuit (L, -C,) of the high negative bias applied to it in the trans-
6BA6 is common to both receiver and trans- mission mode.
mitter circuits. A 12BE6 (V,,,) serves as a When transmitting, the sideband carrier
receiver mixer, the input signal being mixed isgenerated by the common crystal oscillator
with the local vfo signal to produce a 9- and buffer stage (V). The carrier is coupled
MHz intermediate frequency. The vfo stage into #1 grid of the 7360 balanced modula-
is common to both transmit and receive cir- tor (V ;) and the audio signal from the
cuits and tunes approximately 200 kHz 12AX7 speech amplifier is applied to one
in the region of S to 8 MHz, the exact deflection plate of the 7360. The resulting
tuning range depending on the band in use. double-sideband signal passes into the crystal
574 Receivers and Transceivers THE RADIO
Figure 17
www.americanradiohistory.com
HANDBOOK A Single -Band SSB Transceiver 575
C,, Ct 50 -pf each; two Hammarlund HF -S0 RFC,, -I mH choke; Miller 4652
,
ganged RFC, -Use Miller RFC -14 for 80 -40-20 meters;
C -20 -pf variable mica trimmer Miller RFC-3.S for :60 meters
Use
C,- 15-p, type APC. RY -4PDT, 12 -volt coil; Potter-Brumfield KHP-
C -235 -pf. Gap .024 "; Bud 1859 17-D 11
C,- 365 -pf per section; J. W. Miller 2113 S,- Centralab PA -2007
C.- 50 -pf. Centralab 827 T,, T5- 10.7 -MHz i -f transformer; capacitor X is
C,- SO -pf, type APC internal part of unit; Miller 1457
C,-? 04 -pf precision capacitor; Miller 2101 T -5000 ohms to 4 ohms; Stancor A -3877
CR, -IN34 Y, -Int local Crystal Co types CY6 -9L0
J,- Amphenol 80 -PC2F (9001.5 kHz) or CY6 -9Hl (8998.S kHz) as re-
J5- Coaxial receptacle. SO -239 quired
J,-Chassis receptacle; Cinch -Jones P -308A8 ACF- 4- International Crystal Co 9 MHz SSB Al-
MA- Calrad, 0 -1 ma d -c, 13/4" meter ter
PC -4 turns #I8 around 100 -ohm, 2 -watt resis- 1- Chassis, 10" x 12" x 3 ", Bud AC -413
tor -Box, 4" x S" x 3", Bud AU -1028
1
R -Meter shunt for300 ma. Use #30 enamelled, 1-Box, 4" x 4" x 2"; Bud AU -1083
wire wound on 47 -ohm, 1/2-watt resistor 2- Insulated shaft couplers; Johnson 104 -264
RFC, thru RFC,-2.5 mH subminiature choke; 1-Dial drive; Eddystone 892
Miller 70E- 253 -Al
filter which suppresses the undesired side - transceiver for 160 -, 80 -, 40 -, or 20 -meter
band and the carrier, which is already some- operation using standard components. The
what attenuated by the balanced modulator layout has been planned to allow short r -f
stage. The desired sideband is amplified in leads where necessary, and to permit proper
the common 6BA6 i -f stage and passed to circuit isolation. In most cases, resistors and
the 12BE6 transmitting mixer (V,) where bypass capacitors are mounted directly at
it is mixed with the vfo signal to produce the tube -socket pins with liberal use of tie -
an SSB signal on the same frequency as the point terminals to achieve solid construction.
signal being received. The SSB signal is The resistor network for balancing the volt-
further amplified in the 12BY7A driver age on the deflection plates of the 7360
stage (V,) and the 6DQS linear amplifier modulator tube is mounted on a separate
V,). When the pi- network plate circuit terminal board fastened to the side of the
of the 6DQ5 has been properly tuned for chassis, and a second terminal board is used
transmission, it is also tuned for optimum for mounting the r -f choke in the vfo
reception and requires no further adjustment cathode circuit and the associated capaci-
unless a large frequency excursion is made. tors (figure 21) The power plug, relay ter-
.
The same is true of the 12BY7A tuned minal strip, final amplifier bias potentiometer,
circuit (marked grid lune). and speaker jack are placed on the rear
Transceiver Layout
apron of the chassis.
The transceiver measures Final amplifier components are placed in-
and Assembly 121/4" wide by 61/4"
side the utility box bolted to the top rear
high by 101/4" deep. A
corner of the chassis. The chassis area be-
10" X 12" X 3" aluminum chassis is used
neath the 6DQ5 tube is cut out and covered
for the assembly, with the vfo components
with a perforated aluminum sheet, as are
mounted in two 4" X 4" X 2" aluminum
the top and rear of the box, to achieve prop-
utility boxes, one atop and one beneath the
er circulation of air around the tube.
chassis. The final amplifier plate circuit com-
ponents are inclosed in a third utility box
The vfo (figure 22) is placed at the
measuring 4" X S" X 3" in size. Layout of front -center of the chassis and is constructed
the major components may be seen in the on a 1/4-inch thick plate of aluminum meas-
drawings and photographs (figures 18, 19, uring 4" X 41" in size. The vfo tuning
and 20). The cabinet is a homemade wrap- capacitor is fastened to this sturdy base by
around type made of two pieces of perfo- mounting bolts from the underside of the
rated aluminum sheet bent into a U- shaped plate. A precision, silver- plated tuning ca-
inclosure and riveted together at the sides. pacitor having ball bearings and closely
Data is given in the tables for coils, crys- controlled torque is used in conjunction with
tals and frequencies to be used to build a a 10 -to -1 ratio epicyclic driving head to
576 Receivers and Transceivers THE RADIO
Figure 18
achieve a smooth, backlash -free tuning sys- serve as a shield compartment for the vfo
tem. coil and circuit components. The vfo coil
One aluminum utility box is bolted to is made from airwound inductor stock
this mounting plate from the bottom side to (Ininidrrftor) securely affixed to a 1/4-inch
www.americanradiohistory.com
HANDBOOK A Single -Band SSB Transceiver 577
GROMMET
VFO
)
GROMMET
XTAL L4 SHIELD
C7 BOX
O O
o Ci
o O
V
V13 L5 Ve 4/
o o
o 0
0
0
(/F USED)
I
If_
(- C2
I
o I I
o
V12
V1 C
V Vio V9
GROMMET
B B
O O C
GROMMET
Ta
0o I
RY
I
1
I
1
I
1
3K
25 Ww
V14 I I I
I I I I
T3 9MHzFILTER Vis
I
ol
B B
v1;
Ti
V2
B 8
O
J3 J4 J2
Figure 19
thick block of pleAiglas or other insulating of the utility box and is bolted to the box
material which, in turn, is bolted to the in line with the capacitor shaft and affixed
chassis with similar insulating blocks spac- to it with a flexible coupler. A 4!/," diame-
ing it away from the metal. ter circular piece of sheet plastic is at-
Operating voltages are brought into the tached to the drive head to form the tuning
under- chassis shield box via feedthrough ca- dial. It is spray -painted white and calibra-
pacitors and the vfo output leads are con- tion marks are lettered on it with India
nected to feedthrough bushings on the sides ink after final calibration is completed. Suf-
of the box nearest the transmitting and re- ficient clearance is left between the dial and
ceiving mixer tubes. A second utility box the chassis so the plastic does not rub on
is bolted to the top of the vfo plate, spaced the metal.
about % inch back from the front apron The front panel is spaced away from the
of the chassis to permit clearance for the chassis by virtue of the large nuts holding
dial and drive mechanism. The drive head the various controls on the front apron of
is passed through a 3/4-inch hole in the front the chassis and is affixed in place with a
578 Receivers and Transceivers THE RADIO
Figure 20
UNDER -CHASSIS VIEW OF THE TRANSCEIVER
The bottom plate has been removed from the vfo compartment to show internal layout. The
three -gong antenna loading capacitor, C, is bolted to the side apron of the chassis right) as
is the audio output transformer (left). Small components are soldered directly to tube socket
terminals and adjacent fie -point strips, leaving the sockets clear for voltage measurements.
See Figure 19 for placement of major components.
second set of nuts on the control bushings. Transceiver Wiring It is suggested that the
The Vs-inch space thus created provides receiver portion of the
room for the dial to rotate freely. A cutout transceiver be wired and tested first. The
is made in the panel in front of the dial sideband filter comes as a wired package
to match the appearance of the meter. The with matching transformers and requires
opening is covered with a section of ple.vr- only a slight modification. The mounting
glas or Incite inscribed with a hairline plate is cut down to a width of 13/4" to
indicator. A pilot light behind the dial pro- conserve space and new mounting holes are
vides proper illumination. The hole in the drilled along the edges of the plate. The
panel for the tuning shaft should be made filter assembly is then attached to the trans-
sufficiently large so the shaft does not touch ceiver chassis over a slot cut just behind
the panel, making the tuning mechanism the vfo assembly. The output connection of
independent of any panel movement. the filter assembly goes to the grid of the
www.americanradiohistory.com
HANDBOOK A Single-Band SSB Transceiver 579
R4 Figure 23
160 V. R4 CT 180 V.
COIL DATA
L1, La
3/6' DIA. SLUG TUNED COILS
160 MCTCRS- 220H MILLER #21A224118/
80 METERS- 22M MILLER R 27225R8/
40 METERS - 151.111 MILLER 0211S3 Rel
20 METERS- 3.3.1.11.1 MILLER 2/336R81
L3
160 METERS - 53 TURNS 20 ENAMEL WIRE CLOSE WOUND.
1/4 DIA., I 3/4' LONG.
SPIN 2 PINS PINS/ 1
6BA6 i -f amplifier tube (V2). The grounded 40 8100 -6150 kHz SAME DATA AS ABOVE, EXCEPT
PHONE TAP 2ND TURN FROM GND END
side of the input transformer secondary 20 5000 -3500 kHz SAME DATA AS FOR 80 METERS.
is lifted from ground, bypassed and con- 20 5200 -5550 kHz SAME DATA AS FOR 75 PHONE
PHONE ADJUST TRIMMER CS FOR DE-
nected to the 1000 -ohm decoupling resistor SIRED RANGE.
R- 0 0 0 6 175 75 .6
tuned -circuit data for the various amateur V2 6BA6 T- o o o 6 175 70 .5
ceiver, it is suggested that a wiring check NOTE: MEASUREMENTS MADE WITH A 20,000 ONM-PER-VOLT
be made and a voltage check be done with a METER. NO SIGNAL INPUT, R-F CAIN ADVANCED
AUDIO GAIN OFF. FILAMENTS A.C.
TO MAX/MUM,
www.americanradiohistory.com
HANDBOOK A Single-Band SSB Transceiver 581
point. With the 80 -meter unit as an exam- meter indication. The pi- network circuit,
ple, the vfo must tune from 5.5 to 5.0 -MHz of course, is in resonance for this operation,
for proper coverage of 3.5 to 4.0 MHz. as determined by a grid -dip oscillator.
The carrier crystal is at 9001.5 kHz to Up to this point, all tuning has been done
properly place the carrier on the slope of with carrier injection. For proper sideband
the filter for lower sideband output. Coil operation, the carrier must be removed and
L, of the 6BE6 transmit mixer is tuned to the unit excited by an SSB signal. The tech-
3.5 MHz with the aid of a grid -dip oscil- nique is to position the carrier crystal fre-
lator, the slug being adjusted with capaci- quency properly on the filter "slope" and
tor C, set near maximum capacitance. The then to balance out the carrier in the 7360
entire 80 -meter band can then be covered modulator stage. Capacitor C; varies the
by peaking the pi- network and grid- circuit frequency of the crystal oscillator a suffi-
tuning controls. cient amount to find the proper point for
Alignment of the transmitting circuits is the carrier on the passband slope of the filter.
best done with the v.t.v.m. using an r -f The adjustment of this point can best be
probe for signal indication. The function made by ear, when receiving a sideband
switch is placed in the tune position and the signal. Adjust capacitor C, until the re-
carrier -level control (R,;) advanced to- ceived audio of an SSB signal sounds natural
ward maximum. R -f voltage at the plate and pleasing. The crystal should be about
of the 6U8A oscillator should measure about 1500 Hz away from the 9 -MHz filter center
3 or 4 volts, and about the same value frequency. The frequency displacement, of
should be observed at the plate of the buffer course, will remain the same while trans-
section of this tube. Inasmuch as the filter mitting.
transformers and transformer T, have been Carrier null is accomplished by adjustment
adjusted previously, no further adjustment of the balance control (R,) on the panel.
of these circuits is required. The r -f probe The r-f probe is placed at the grid of the
can now be placed at the grid of the 6DQ5 6DQ5 stage and the function switch turned
amplifier tube socket and the slug in coil to transmit. No audio signal is desired. The
I._ adjusted for maximum r -f voltage read- balance potentiometer is adjusted for min-
ing. This peaks grid tuning so that coil L, imum indicated reading on the v.t.v.m.,
will track with the previous alignment of which should be 1 volt or less. Operation of
coil I.,. the audio system and balanced modulator
may now be checked by noting the voltage
Final Adjustment The 12BY7A stage should swing while talking into the microphone. A
and Neutralization now be neutralized. To sustained audio tone will swing the meter to
accomplish this, all pow- 30 or 40 volts peak reading. It is helpful to
er is turned off and the screen lead tempor- monitor the signal in a nearby receiver while
arily removed from the 12BY7A socket. these adjustments are being made.
With power again turned on, circuits reso-
nated, and the function switch in the tune Transmit Operation The screen -voltage lead
position, neutralization capacitor C, is ad- may now be reconnected
justed with a nonmetallic screwdriver for to the 6DQ5 tube socket and high voltage
minimum feedthrough of r -f voltage as provided for the plate circuit. Potentials be-
measured with the v.t.v.m. probe placed at tween 400 and 800 volts may be used for
the #1 grid terminal of the 6DQ5 socket. the 6DQ5, with proportionately higher out-
The screen lead to the 12BY7A socket is put at the higher plate voltages. An antenna
replaced when this operation is concluded. or dummy load must be connected to the
The same technique is employed with the transceiver to complete the final checkout
6DQ5 stage as was used with the driver and bias adjustment. The meter switch is
stage. With screen (and plate) voltage re- set for plate current and the function switch
moved from the 6DQ5, but with drive ap- for transmit. The bias potentiometer on the
plied, the v.t.v.m. is placed on the antenna rear apron is adjusted for a 6DQ5 resting
terminal of the transceiver and neutralizing plate current of 25 milliamperes. Antenna
capacitor C, adjusted for minimum volt- loading is done with the function switch in
582 Receivers and Transceivers THE RADIO
the lune position. As the carrier control is ma. As the audio level is raised, speech will
advanced, the final -amplifier plate current kick the indicated current up to values in
will rise in a linear fashion. The amplifier the vicinity of 125 to 170 milliamperes
plate circuit is brought into resonance and depending on the individual voice. Too high
the grid circuit adjusted for peak plate cur- values of peak current will result in dis-
rent reading. Loading control C,, is adjusted tortion and splatter.
for further increase, reestablishing resonance The meter may be switched to read rela-
with the tuning control until the indicated tive power output which, in some cases,
cathode current reaches a value of 275 to will simplify loading the amplifier, especially
300 milliamperes. Full load current should during mobile operation, as tuning may be
not be run for more than 20 seconds at a done for maximum output reading under
time to achieve maximum amplifier tube a controlled level of excitation.
life. When the function switch is advanced The 80 -meter version of the SSB trans-
to transmit, amplifier plate current will ceiver is shown in the photographs. The
drop back to the original idling value of 25 only difference in a unit designed for a
Figure 25
200 WATT PEP SIDEBAND TRANSCEIVER FOR 80, 40, AND 20 METERS
Less thon a cubic foot in volume, this inexpensive transceiver will fit into today's "compact"
automobile. Unit may also be used with auxiliary 11S -volt a -c supply for the home station.
The major controls on the panel are (I. to r.): sideband switch (S,), SSB /a -m selector switch
(S ), audio volume (R ), microphone gain (R,), carrier injection (S 1, band- selector switch
(S ,), microphone jack (J), r -f gain (R), meter -selector switch (S.,), antenna loading capacitor
(C, ), and final amplifier tuning (C,;). The main frequency -control dial (C ) is at top center.
Wrap -around, perforated cabinet provides ventilation and acts as TV! shield.
Once adjusted for a particular band, the only tuning required is done with the vfo control.
Bandpass coupling allows large excursions in fr cy. The vfo tuning mechanism with
100:1 ratio makes sideband tuning a pleasure.
www.americanradiohistory.com
HANDBOOK A 200 -Watt 3 -Band Sideband Transceiver 583
V, V T3 VS T4 Ve
PI-
ETWORK
TRANSMITTER SECTION To RY 2
5.0-5.3MHz OR 16.0-16.3MMr
I PDI. S RHz 2 rLt V3 Ve
Y,
Y3
0E6
PREMIX 2/ 5 MHz
3. 0 -3.33 MC.
FREQUENCY CONTROL SECTION
Figure 26
stage to a second 6BA6 receiver i -f ampli- over switch is routed through the main
fier (V15) whose output circuitry is capaci- changeover relay (RYIB) so voltage is
tively coupled to a 6BE6 product detector applied to the carrier oscillator when trans-
(V12). Oscillator injection for SSB reception mitting, regardless of the setting of the
is from either of the two sideband crystals SSB a -m switch (S;,).
in the grid circuit of the 7360 carrier Mobile operation requires a receiver hav-
oscillator -balanced modulator (V_) which is ing a reserve of audio power and the audio
common to receive and transmit sections. section is designed to meet this requirement.
Collector plate voltage is removed from the Two 6EB8 triode -pentode tubes (V,,, V1,)
7360 during reception by relay RY,C but are employed, with the pentode sections
the oscillator section always functions since used as a push -pull audio stage. One triode
deflector and screen voltage is applied in section of the first 6EB8 is used as an audio
either mode. phase inverter and the second triode is used
The 6BE6 product detector (V,2) may as the driving amplifier for the phase in-
be switched to function as a plate detector verter. The two dual -purpose tubes take up
for reception of a -m signals (S,ABC). This no more space than the usual two -tube am-
changeover requires disabling the 7360 car- plifier stages but produce nearly 5 watts of
rier oscillator, but since this oscillator is re- high -quality audio. The speaker is not in-
quired for transmitting, the a -m change- corporated in the transceiver, since use of
www.americanradiohistory.com
HANDBOOK A 200-Watt 3-Band Sideband Transceiver 585
the speaker in the auto radio is contem- tor removed from the amplifier stage of a
plated. For home use, an auxiliary speaker surplus SCR -274N /ARC -5 transmitter. Only
is incorporated in the 115 -volt a -c power the worm gear and frame assembly are used
unit. and the original capacitor plates are re-
The Transmitter Portion -The transmit- moved (figure 28). A double bearing 140 -
ter portion of the unit starts with a 12AX7 pf receiving -type variable capacitor is in-
two -stage speech amplifier (V1) driving a stalled in the frame in place of the original
deflection plate of the 7360 carrier oscilla- capacitor assembly, slipping the spring -
tor-balanced modulator (V2). When trans- loaded drive gear over the shaft of the new
mitting, voltage is applied to the collector capacitor so that it engages the worm gear
plates of the 7360 via relay RY,C and the as did the rotor of the original capacitor.
carrier is generated by the triode section The free space inside the framework is used
of the tube functioning as a crystal oscil- to mount the various components of the
lator. Choice of upper or lower sideband is vfo as shown in the photograph. An alumi-
made by proper crystal selection by means num plate is bolted to the back frame to
of sideband -selector switch S,. The balanced - support the tub: socket (V7) and an L-
modulator plate circuit of the 7360 is link - shaped shield is bolted over the top and
coupled to the 9 -MHz filter for rejection end of the frame to inclose the assembly.
of the unwanted sideband and passage of the A circular dial cut from 1/16 -inch plastic
desired sideband to the common 6BA6 i -f or plexiglas is placed on the large gear in
amplifier (V:,). The sideband signal is then lieu of the original metal dial. The new
transformer- coupled to the 6BE6 transmit- dial is spray -painted white on the front and
ter mixer (V'). This mixer stage receives calibration marks are lettered with India
its mixing voltage from the vfo and buffer ink. The complete vfo is bolted to a base
premixer stages (V7, V5) in the same man- plate of %8" thick aluminum, slightly larger
ner as the receiver. Output of the 6BE6 in area than the capacitor framework. The
transmitter mixer is at either 80, 40, or 20 completed assembly is then bolted to the
meters and is bandpass -coupled on the desired transceiver chassis with the center of the
band to a 12BY7 amplifier- driver (V,). dial at the center line of the chassis. The
This stage, in turn, is bandpass- coupled to plastic dial will extend below the front
a neutralized 6DQ5 (V) serving as a apron of the chassis, requiring a slight
class -AB, linear amplifier. The final tank amount of clearance so that it does not rub.
circuit of the amplifier is a pi- network The panel is spaced away from the chassis
configuration providing good harmonic at- apron by the lock washers and nuts that
tenuation and ease of adjustment. fasten the various controls, allowing clear-
Transceiver
ance for the dial. The panel is secured in
Transceiver construction is place with a second nut on each control.
Construction straightforward and should The upper edge of the panel and the rear
be no problem for the ad- lip of the chassis are bolted to the wrap-
vanced amateur. The vfo is built as a sep - around cabinet to provide a rigid structure
arate unit and may be tested and aligned immune to vibration.
before it is installed in the transceiver. The Component Layout-Most of the major
receiver portion of the unit should be wired
components are mounted atop the chassis as
and tested before the various transmitter
shown in figures 29 and 30. The antenna
stages are completed. The transceiver is
constructed on a 10" X 12" X 3" steel receptacle (J:,), power plug (P,) and jack
chassis. Layout of the major components and for the external speaker (J_) are placed on
shield partitions are observed in the photo- the rear apron of the chassis and all other
graphs and drawings. The 6DQS amplifier major controls are mounted on the front
tube socket is recessed so that panel height panel with the exception of the phase -bal-
is only 6 %2 ". Standard parts are used ance capacitor (CO and voltage- balance po-
throughout with the exception of the vfo tentiometer (R._) which are placed on the
tuning capacitor. The vfo is built as a unit chassis to the rear of the 7360 tube socket.
on the frame of a worm -gear driven capaci- These controls need be adjusted only in the
586 Receivers and Transceivers THE RADIO
Figure 27
SCHEMATIC OF TRANSCEIVER
www.americanradiohistory.com
HANDBOOK A 200-Watt 3 -Band Sideband Transceiver 587
C, -20 -pf differential capacitor (E. F. Johnson #32 copper wire wound on 47 -ohm, t/3-
160-311) watt resistor placed at cathode terminal
C C -12 -pf ceramic variable (Centralab CRL - of 60Q5
827) RFC- 2.S -mH, 300-ma. (National R -3000)
C, -50 -9f (Hammarlund MAPC) RY 3PDT relay, 12 -volt d -c coil (Potter-Brum-
C-- 140 -pf (Hammarlund MC -140M) field KM -14D or equiv.)
C, -25 -pf ceramic variable (Centralab CRL -827) RY DPDT relay, 12 -volt d-c coil (Potter -Brum-
C thru C,t 50 -pf ceramic variable ( Centralab field KM -11D or equiv.)
CRL -827) S A, B, C -3 -pole, 2- position wafer switch
-1S -pf (Hammarlund MAPC). (Centralab CRL PA1007)
-
C
C, 235 -pf (Bud 1859) S A, B; SA, B; S,A, B; S.A, B;S, -2 -pole ceramic
C 1200 -pf, 3 -gang broadcast -type capacitor wafer sections (Centralab PA -2 each,
(J. W. Miller 2113) ganged on Centralab PA -301 index as-
CR, thru CR -Diode, ,434 or equivalent
1 sembly)
FL,-9-MHz crystal sideband filter (McCoy 558- T,-Transformer, 10.7 -MHz TV 1 -f type, (J. W.
9, McCoy Electronics, Mt. Holly Springs, Miller 1463). (x indicates internal compon-
Pa.) ent)
M -0 -1 d -cmilliammeter, 13/4 square (Cal -Rad, T , T , T,-Transformer, 4.5 -MHz TV interstage
or equiv.). type .(J. W. Miller 6270). (c indicates inter-
PC- Parasitic choke. 7 turns #18e. wire on 100 - nal component)
ohm, -watt composition resistor
1 1,- Universal output transformer, 10K plate -to-
P,- 8-contact chassis -mounting plug (Cinch- plate (Stancor A -3823)
Jones P- 308AB) Y,-8898.S -kHz crystal (furnished with FL,)
R,- 1- megohm potentiometer with switch S, Vi- 9001.5 -kHz crystal (furnished with FL)
attached Y,-21.50 -MHz crystal (International Crystal
R -Meter shunt for 300 -ma range. Approx. 10 Co. FA -S)
initial alignment and ordinarily require no hand. These are ready -made 4.5 -MHz TV
further attention. replacement interstage transformers (T2, T:,,
The main bandswitch runs down the and T,). They are used without alteration
center line of the under -chassis area with and provide the desired bandpass effect by
wafer sections S,_- (inclusive) bolted indi- virtue of stagger- tuning between 3.8 and
vidually to the small partitions that act as 4.0 MHz.
interstage shields. Switch wafer S, for the A great deal of the wiring may be done
6DQ5 amplifier plate tank coil is mounted before the shield partitions or switch as-
in the amplifier compartment on the rear semblies are put in place. The switch wafers
apron of the chassis below tank coil are installed one at a time, beginning with
with the connecting wires from the coil the receiver segment at the rear of the
brought below deck through an oblong hole chassis. The side and front shield plates are
in the chassis. The shaft of this switch is made of thin aluminum and are installed
ganged to the main bandswitch shaft by last, being bolted to each other, the switch
means cf a link -and -arm arrangement shown partitions, and the chassis to make a rigid
in figure 31. Two small lever arms are made assembly (figure 33).
by taking apart a flexible shaft coupler. Terminal boards are used for the small
One arm is slipped over the main band - components of the balanced modulator and
switch shaft at the point where it enters audio systems. Other small components are
the under- chassis shield plate behind the mounted to tube- socket terminals and tie -
main panel, and the second arm is attached point terminal strips.
to the fiber extension shaft driving the am-
plifier switch wafer (S,) mounted on the Testing and The transceiver will operate
rear apron of the chassis. The two lever Alignment with any power supply capable
arms are interconnected by a narrow strip of of delivering between 500 and
aluminum having a hole at each end for 800 volts at an intermittent load of 250
small bolts to secure it to the two lever milliamperes for the final amplifier, and
arms. Panel bushings in the shield plate act 250 volts at 125 milliamperes for the re-
as bearings for the switch shafts. ceiver and exciter sections. Bias requirement
The bandpass coils are constructed as in- is -50 volts at 5 milliamperes (adjustable).
dicated in the coil table figure 32) with For fixed -station use and bench align-
the exception of the coils for the 80 meter ment, a voltage- doubler power supply using
588 Receivers and Transceivers THE RADIO
www.americanradiohistory.com
HANDBOOK A 200 -Watt 3 -Bond Sideband Transceiver 589
i't alletfll(Mt
- o o a) ,......'1.
t 1? f
Figure 29
remove them from the active circuit and a coils are then resoldered to the bandswitch
grid -dip oscillator is used to set the fre- terminals and the 6B6 r -f amplifier tube
quency of the primary circuits (L__, L_,) by )
(V is removed from its socket. This raises
adjustment of the slugs. The 40 -meter plate the resonant frequency of the primary wind-
coil is adjusted to 7.3 MHz and the 20- ings so they will not affect the adjustment
meter plate coil to 14.35 MHz. The grid of the grid circuitry. The grid coils are
590 Receivers and Transceivers THE RADIO
SPEAKER
J2fPLUG POWER
PLUG, II
n
11
P1
ANTENNA
CONNECTOR -J3
Ls
L6
F L,
9 MHz
FILTER
o Tz o
L22 L24
L3
L4 Ti
-
o T3 o
L13 Lis
-- -
V7
L7 BAUE
L17 L19
Li
Ti o
Lz 10-o jjF
450 V. L UNDER -CHASSIS SHIELD
o PHENOLIC
SHAFT
C10 R20
OL11
V2 VFO ASSEMBLY
7360
C2 C3
Yi \- A A_A i Yz
METER "1*--PLEXIGLASS
DIAL
Figure 30
CHASSIS DECK
SHIELD PARTITI TO PANEL Transmitter Alignment Alignment of the
ALUMINUM
STRIP transmitter section is
.f(N".t. ARM FROM done with the high voltage disconnected and
\3-1.I
1 COUPLER
SHAFT TO FINAL AMPLIFIER SWITCH INDEX SHAFT
with screen voltage removed from the 6DQ5
SWITCH SEGMENT.
Se S4 -ST
amplifier. If the 0A2 screen -regulator tube
is wired so that the dropping resistor goes
to pin #1 and the screen lead to pin #5, the
screen voltage will be disconnected by re-
Figure 31 moving the 0A2 from its socket, since the
BANDSWITCH ARM DETAIL 0A2 has an internal jumper between these
pins.
Much of the transmitter alignment is
then dipped to 7.0 and 14.0 MHz. With the completed once the receiver section has been
r -f tube back in its socket, the transceiver adjusted. The 7360 balanced -modulator
can be turned on and checked for receiver plate coil (L1) is tuned first, placing the
operation on each band. r -f probe of the v.t.v.m. at the grid (pin
www.americanradiohistory.com
HANDBOOK A 200 -Watt 3 -Band Sideband Transceiver 591
PLATE
#7) of the 6BE6 transmitting mixer (V,)
LS -3 COIL FORM
to obtain an r -f voltage reading. The trans-
mitter circuitry is energized by pressing the
push -to -talk switch on the microphone
(with the microphone gain control R,
BANDPASS COIL turned down). The carrier control (R3) is
turned on and advanced to provide carrier
injection until a reading is obtained on the
x X x v.t.v.m. The slug of coil L, is adjusted for
N
4.1 8 S maximum r -f indication. The phase -balance
AI
capacitor (CO should be set for equal ca-
COMPONENTS FOR BALANCED COMPONENTS POR AUDIO AND PRO -
pacitance and the voltage -balance potenti-
MODULATOR CIRCUIT. DUCT DETECTOR CIRCUIT.
ometer (R2) set near the center of rotation.
TERMINAL BOARD LAYOUT
When the carrier control is turned off, the
indicated r -f voltage will drop and balance
Figure 32
potentiometer R;, should be adjusted for a
COIL TABLE FOR TRANSCEIVER minimum r -f reading. This is the adjust-
ment for carrier suppression and at this
L -12 bifllar turns ,24 in all) "24 enamel wire,
closewound on slug -tuned form, 12" diam. time the phase -balance capacitor should be
(National XR -50). Tune to 9 MHz adjusted slightly to achieve lowest possible
L,-4 turns 1,24 hookup wire around center of
L,
r -f reading. Both controls affect carrier sup-
L -4 turns a24 hookup wire on "cold" end of pression and are slightly interlocking and
L3
L.-30 turns a 30 enamel closewound on
should be adjusted in sequence for lowest
L , L ,
5/16" diameter form. Tune to 9 MHz reading on the v.t.v.m. The whole process
L -4 turns $24 hookup wire on "cold" end of may be monitored with a receiver used as
L
L -12 turns a24 enamel closewound on 1/4' an r -f probe with the antenna lead placed
diam. slug -tuned form /CTC -L53 or equiv.). near the socket of the 6BE6 transmitter
Tune to 16 MHz
L -8 turns a24 enamel wire closewound on
mixer tube (V1).
3e" length of 2" diam. polystyrene tubing
7 Carrier Oscillator Adjustment- Capaci-
slipped over top end of coil to make
L
tors C, and C_ across the upper- and lower -
premixer transformer. Tune to 16 MHz
1.,,,-Ferrite rod loop -antenna coil "loopstick" ,
sideband crystals are used to trim the crystal
with turns removed to resonate to 5MHz frequencies for proper positioning of the
(J. W. Miller 6300)
LIB 15 turns #24 enamel wire closewound on carrier on the slope of the sideband filter.
Vs" diam. slug -tuned form (CTC -L53). Tune To realize the rated sideband rejection of 40
L;to3011).
21.5 MHz
turns "20, 3," diam.,
77/3 3 )4" long (B & W decibels, the carrier oscillator should be
Tunes 5.0 to 5.35 MHz placed 1500 Hz above or below the 9 -MHz
L , L , L enamel wire close-
--30 turns (130
center frequency of the filter. Carrier sup-
wound on 3/3" diam. slug -tuned form (CTC -
L53). Tune to 7 MHz pression is also affected by proper position-
L L, , L -25 turns "30 enamel wire close- ing of the carrier frequency on the filter
wound on 3/4" length of 2" diam. poly-
styrene tubing cemented to top of L L , , slope. When making the frequency adjust-
L to make bandpass transformer see
ments, carrier suppression should be checked
sketch). Tune to 7 MHz
L, L, L -14 turns r28 enamel wire close- on both upper -and lower-sideband positions.
wound on 3'8" diam. slug -tuned form (CTC - The minimum voltage reading with carrier
L53). Tune to 14 MHz
L, , L , L -12 turns .028 enamel closewound on turned off should be very nearly the same
38" length of 1.2" diam. polystyrene tubing with either crystal. Final adjustment may
cemented to top of L , L L to make ,
bandpass transformer. Tune to 14 MHz be made with voice modulation, striving for
L -Final amplifier tank coil. 32 turns "16 good audio quality on either sideband as
wire, with 16 turns spaced twice wire
diameter; 16 turns spaced wire diameter monitored in a nearby receiver.
592 Receivers and Transceivers THE RADIO
IN FL III se
Figure 33
is bolted to the side apron of the chassis near antenna receptacle and tank switch wafer S .
The opposite side apron is used to mount the audio output transformer (T ) and two terminal
boards that support most of the resistors and capacitors in the audio and balanced modulator
circuits. Terminal strips and tie points are used to mount small components securely to resist
vibration encountered in mobile work. The antenna relay (RY ) is mounted on the rear apron
above the 68E6 (V, ) socket. The voltage changeover relay RY. is mounted in the center of
the chassis area between the i -f amplifier tubes and the audio tubes.
Bandpass Adjustment-The bandpass cir- tuned for uniform 6DQ5 drive -voltage read-
cuits in the linear amplifier stages of the ing over the 200 -kHz tuning range with the
transmitter are aligned in the same manner r-f probe placed at the grid of the 6DQS. A
as the receiver circuits using carrier injec- maximum of 15 to 20 volts rms can be
tion from either sideband crystal. The 40- obtained with full carrier injection. Under
and 20 -meter coils are checked with a grid - final operating conditions, the 40- and 20-
dip oscillator as before, but the 80 -meter meter coils may require some slight adjust-
transformers (T,, T4) as well as the second- ment for uniform drive across these bands.
ary of T are adjusted with voltage applied Amplifier Neutralization-The last step
to the transmitter and the transformer slugs is to neutralize the final amplifier stage.
www.americanradiohistory.com
HANDBOOK A 200-Watt 3-Band Sideband Transceiver 593
CHART 1
TUBE 1 2 3 4 5 6 7 8 9
Low voltage -250 volts at 115 ma receive High voltage -600 to 800 volts at 300 ma, xmit only
80 ma xmit Filaments -12.6 volts a.c. or d.c. at 4 A
--50
Bias volts d.c. 5 ma Relay -12 V.D.C. 80 ma, xmit only
With plate and screen voltage removed and up to 200 watts peak input may be run to
grid drive applied to the 6DQ5, neutraliza- the 6DQS without overheating the tube.
tion is accomplished by placing the r -f probe Carrier injection and tune -up conditions, on
at the antenna receptacle and adjusting neu- the other hand, impose maximum dissipa-
tralizing capacitor C,:, for minimum r -f tion conditions on the tube and tune -up
indication when the 6DQS tank circuit is operation at full input should be limited to
tuned to resonance. periods of 20 seconds or less in one minute
Final Anmhlificr Adjust men /-
Amplifier as tube dissipation runs near 65 watts or so
under these conditions. With the average
bias is adjusted to provide 50 ma of resting
current. The transceiver should be coupled voice, peak plate- current indication on the
to a dummy load and loading and grid drive meter will run below 50 percent of the full
(carrier insertion) adjusted to provide the carrier injection plate current, even taking
desired input level.
into account the alc action of this circuit.
Thus, under intermittent carrier tune -up
Antenna loading requires that a fixed
at 800 volts plate potential, maximum plate
ratio of grid drive to plate -load impedance
current may run as high as 275 to 300 mil-
be achieved. Maximum drive level is fixed liamperes, with indicated voice peaks run-
and loading is accomplished at this level ning about 125 to 175 milliamperes meter
and may be increased until flat- topping is reading. Excessive peak plate current read-
first observed on a monitor oscilloscope. ings under voice conditions indicate flat -
Advantage is taken of the high peak-to -av- topping and consequent distortion of the
erage -power ratio in the human voice, and signal.
594 Receivers and Transceivers THE RADIO
432
26 -6 A 432 MHz Low ANTENNA
MN,
Ill MO 29 MHr
Noise Converter System
www.americanradiohistory.com
HANDBOOK A 432 MHz Low Noise Converter System 595
Lt C2 La 2N3478 L3 FO uP L
+ S0- + ig
ANTENNA SHIELD I,
1
I
I
l
rII
iw 7' Z T
T Ti
-Due
it-1 ,y
2.1R 9.9K
NOTE: ALL RESISTORS 0.25 WATT CHASSIS LAYOUT SHIELD
UNLESS OTHERWISE VIEWED FROM
SPECIFIED.
-O 9V. +
INPUT SIDE.
1
7 STRIPS OF 1 /6I INCH THICKNESS FLASHING
COPPER.
3. ADJACENT EDGES OF L2 LS ARE 13/16
1 +1 INCH APART.
4. SN /ELO MADE OF 1 /9I INCH THICKNESS
CONNECT BASE 112 FLASHING COPPER.
LEAD HERE CONNECTS
HERE FOR 5 WE ARE BUTTON CAPACITORS (ERIE 29221
!
C
O
Figure 36
SCHEMATIC AND LAYOUT FOR 432 -MHz R -F AMPLIFIER
C , C , C 0.5-
to 8 -pf piston trimmer capacitor. JF0 PC3S -H080 or equiv.
C -3 -pf ceramic capacitor. (see text:
RFC -7 turns W22 e., %9 -inch long (Ohmite Z -460)
J, J -DNC connector, chassis mounting. UG -657/U or equiv.
Sox -4" x n x IS /e" (Bud 3002A)
596 Receivers and Transceivers THE RADIO
Figure 37
L3 -C4, iscoupled to the output receptacle the input end of the chassis and capacitor
through a blocking capacitor. Additional C, is mounted with the lead pointing to-
feedthrough capacitors are used to decouple ward the transistor shield. The lead of ca-
power leads passing through the wall of the pacitor C, points toward receptacle J_. Cut
inclosure. the inductors from thin copper (see illustra-
tion), shape them, and solder in place, start-
R -F AmplifierThe amplifier is built in an ing with inductor L,. An insulated standoff
Construction aluminum box measuring 4" terminal is used to support the r -f choke and
X 2/8" X 15/8" (Bud 3002 - the two adjacent resistors. The intrastage
A). A piece of copper-clad shield is drilled for the transistor and mount-
(one side) phenolic board (see figure 37) is ing holes and mounted with lips facing the
mounted on the inside of the box, copper input end of the chassis. The emitter bypass
side towards the viewer, to provide a good capacitor is soldered on the input side of the
r -f ground. All grounds are made to the shield near the transistor mounting hole. The
copper foil. Small 4 -40 screws in the corners mounting lugs of the capacitor serve as
of the board affix the board to the box and feet. The 2N3478 is finally mounted by
ground it to the box. carefully soldering the ground lead (shortest
The first step is to drill the board and of the four) to the shield, with the transis-
mount the coaxial connectors, followed by tor suspended upside down and centered in
the feedthrough capacitors. The leads on the hole. The shield lead is tinned at the
piston capacitors C,, C,, and C, are trimmed point it solders to the metal plate. A heat
to a length of 5/16 inch. Capacitor C, sink must be used between the solder point
is mounted with the lead pointing toward and the transistor. Long -nose pliers are ef-
www.americanradiohistory.com
HANDBOOK A 432 MHz Low Noise Converter System 597
e
i
e
-
CI 'STANDOFF MTG
CI FOR
LS
J3 C2 CR1MTa
t + --
le
J3. I -F OUTPUT le
-- . ++- --__.
CHASSIS LAYOUT
32 MC.
MIXER UNIT
H
MTG. HOLE
I, I 11, -
Figure 39
403 MHz is used to produce a 29 -MHz of copper-clad (one side) phenolic board
intermediate frequency. If the mixer is used (figure 40) is mounted on the inside of the
without the benefit of the transistor r -f box, and a piece of the same material is
stage ahead of it, a 1N2IF diode is suggested fastened to the inside end of the box where
Figure 40
www.americanradiohistory.com
HANDBOOK A 432 MHz Low Norse Converter System 599
eceptacle J_ is placed. Small 4 -40 screws When fastening the two halves of the
are placed near the corners of the boards to aluminum box together, two additional #6
fasten them in place, and the boards are sheet -metal screws should be used on each
soldered together where the two edges meet. side of the box to ensure good r -f contact
The boards are removed from the box after between the halves.
this operation and the remaining holes
Adjusting the Connect the 432 -MHz mixer
drilled. A single 6 -32 screw secures the
Mixer Unit unit to the oscillator /multi-
ground end of inductors L, and L. The
center pin of receptacle J, is cut short to plier unit, the i -f system,
allow clearance between inductor L1 and and a suitable antenna. Set the i -f system
the pin, as shown in the photograph. to 29 MHz and turn on a nearby, stable
Spacing between L, and L:, is about I/8- 432 -MHz test signal. Set capacitor Co at
inch. A 1/4-inch high insulator with a midrange. With the oscillator /multiplier cir-
fuse clip attached is used to support the base cuits peaked, a diode current of about 0.5
end of the mixer diode. A pin receptacle ma should be achieved. Tune in the i -f
from an octal socket makes connection to signal and peak capacitors C, and C_ for
the small end, and is soldered to a short, maximum observed signal.
1/4 -inch wide copper strap fastened to in- Oscillator /Multiplier This local oscillator
ductor L:,. This permits the diode to be Unit for 432 MHz (LO) unit employs two
easily changed with minimum effort. The tubes and a 3rd -over-
local oscillator injection loop (L,), should tone crystal in the 40- to 50-MHz range to
be as small as practical and still permit 0.5- provide injection voltage for a 432 -MHz
ma diode current. The ground end of in- converter unit. By choosing the proper crys-
ductor L2 is soldered to the copper board tal frequency, an i -f output of 7, 29, or
adjacent to the nearby BNC receptacle 50 MHz may be obtained for a 432 -MHz
mounting nut. For maximum transfer of input signal. An optional feature is remote
power from the local oscillator to the mixer vernier tuning of the local crystal oscillator
diode the sum of the length of inductor L, (and thereby the received signal). This
the coaxial cable, and the output link of the feature is extremely useful when tuning for
injection unit should be a half wavelength weak DX signals and SSB stations. Filament
(electrical) or multiple thereof at the in- power is 6.3 volts at 0.48 ampere and 150
jection frequency. volts d.c. (regulated) at 15 ma. The unit
will deliver more than enough power to a
diode mixer to develop better than 0.5
milliampere of diode current.
The LO Circuitry A 12AT7 double triode is
used in a cathode oscil-
lator /tripler circuit. A 44.777 -MHz 3rd -
overtone crystal is used when a 29 -MHz
i-f signal is desired for 432 MHz. The plate
circuit of the first triode section is resonated
at the overtone frequency (f). The plate
circuit of the second triode section is tuned
Figure 41 to the third harmonic (f:,), 134.331 MHz.
THE OSCILLATOR -MULTIPLIER UNIT
Inductive coupling is used between this cir-
cuit and the following stage to minimize
The LO features o varactor- controlled remote the transfer of undesired harmonics. Coils
tuning circuit. Components are mounted on an
inexpensive aluminum box. Atop the chassis L, and L, are made from a single length of
(I. to r.) are: f X I output receptacle J , 12A17 inductor stock. The spacing between the
oscillator (with crystal behind it), tip jacks for
alignment, and 6AKS multiplier tube. Output adjacent ends of the coils is %8 inch. The
receptacle J: is at far right. Tuning capacitors 6AK5 plate circuit is tuned to the ninth
C,, C., and C, are in the foreground. Output
tuning capacitor C, is to the right of the 6AKS harmonic (f,,, 403 -MHz). Inductor L; is
tube. a hairpin loop of 3 '16 inch wide "flashing"
600 Receivers and Transceivers THE RADIO
JI
(fx1) nL5. VIEWED ON EDGE.
OUT FLASHING COPPER,
3 /1e' WIDE.
O- m I
Le, SAME SIZE.
L2 MADE OF R20 WIRE.
1 )KLI
OOIa t
Cd .
20 ci
d (fxl) f6
GNO -
C 2A17
e
7
1 c
(fx3)
L3 L
6AK5
5
L5
le 1
.I
J2
L 20
REMOTE TUNING CNT. Xi C2 (f59)
(f ) OUT
270 70 KIOK
C IK
12A1 GARS
9
I. n5 3
e.nv.
11
.001 CHASSIS LAYOUT
p
PLI
1! wi l ... s
GND. FIL. e+
REMOTE VERNIER TUNING
(OPTIONAL)
Figure 42
SCHEMATIC OF LOCAL OSCILLATOR MULTIPLIER
C,, C. -20 -pf (Johnson 20M11 miniature) (B 6 W 3006)
C,-9 -pf (Johnson 9M11 miniature) L -2 turns as per L,
C,-5-pl (Johnson SMI? miniature) PL,- 3-circuit chassis mounting (Cinch -Jones P-
C- Variable capacitance diode (5.2- to 31 -pf, 303A0)
Pacific Semiconductor V -12 Voricop) RFC -50 -MHz choke (Ohmite Z -50 or equiv.)
J.-Coaxial receptacle, UHF or BNC type
.1,, X,- Third -overtone crystal. Use 44.777 MHz for
L,-9 turns #20 e., closewound, 3/e in. diam. 29 -MHz i -f; 42.444 MHz for 50 -MHz i -f; 42.500
L.-One turn loop at 11-plus end of L, MHz for double c sien to 7 MHz (see text)
L -3 turns spaced I/ inch, I/2 inch diam. Chassis -Bud CU -2106A
copper strap. The coupling inductor, L,;, is 28 MHz using an auxiliary i -f amplifier,
approximately the same size as L and is described later.
mounted parallel to it, about 3/16 inch The second optional circuit is remote
away. vernier tuning of the crystal oscillator by
Two optional circuits are available to the means of a capacitance diode (l'aractor)
user. The first option provides LO injection across the crystal -oscillator plate circuit.
at both the ninth and fundamental overtone Excellent vernier tuning is provided by this
frequencies. By using both the ninth and simple circuit. A blocking capacitor serves
the fundamental frequency of a 42.5 -MHz to remove the d -c plate voltage from the
3rd -overtone crystal, a converter with first diode, and an r -f choke isolates the tank cir-
and second mixer stages could be built that cuit from the external control circuit. At the
would convert 432 MHz to 49.5 MHz (for remote- control point, a ten -turn potentiom-
good image rejection) and then convert eter, limiting resistor, and battery are
again to 7 MHz where many receivers have used to adjust the diode voltage, thereby
better performance than at 28 MHz. This tuning the overtone crystal over a narrow
option was not exercised in this design, as range. The values indicated will provide a
good receiver performance was achieved at tuning range of about 3 kHz at 432 MHz.
www.americanradiohistory.com
HANDBOOK A 432 MHz Low Noise Converter System 601
Figure 43
A counter dial reading from 0 to 1000 for the 6AKS stage should be short and direct.
10 turns of the potentiometer will provide The socket for the 12AT7 is oriented so that
good bandspread when tuning a few kHz pin 6 faces the 6AKS socket, which has pin
for a weak moonbounce signal. A shielded 1 facing the 12AT7 socket. Point -to -point
cable connects between the remote control wiring is used with many components
unit and the LO unit. The plus terminal of mounted directly to socket pins or to adja-
the battery connects to the cathode of the cent ground lugs or tie points.
tuning diode. If the builder elects to omit
this feature, the portion of the circuit in Adjustment of Tune -up procedure of the
the dotted inclosure may be eliminated. the LO Unit LO unit is simple and
straightforward. Power is ap-
Construction of The LO unit is built in an plied, including bias to the tuning diode (if
the LO Unit aluminum box measuring used). The f, and f, circuits are adjusted
S' " X 3" X 2'/R" (Bud for maximum negative voltage read by a
CU 2106A). Layout of major components
is shown in the photographs (figure 43).
high- resistance meter at test point J.
This
is the grid bias voltage of the 6AKS and
Note that a piece of thin copper sheet is should approximate three volts. Connect a
fastened to the inside of the box where the low -range milliammeter across test points
6AKS socket is mounted. This provides a J, and J, and watch for a faint change in
good r -f ground for this stage. All leads in 6AK5 plate current while adjusting the
602 Receivers and Transceivers THE RADIO
+
LI
SRIELD
I
-F
INPUT 6CW4
L3 j, FOLD
_-I-.
IFIL.1
Ja F OLD I
4
I-F
OUTPUT - - -- etwA-
_
J3
J2+._ L1+ La
FOLDI
470 FOLD
05 2
.9U
DIODE
MIXER
CURRENT
(- .00,
1
"ST 1
FT
3 4 T
C.045515 LAYOUT SHIELD
GND V 'I0V.,5MA
Figure 45
L -19 turns 028 enam. on 35 -inch form (powered -iron slug). 3.10- to 6.8 -. hy. IJ. W. Miller
4405)
L -3 turns insulated wire at 8 -plus end of L
Box -4 x 218" x 1, 8" (Bud 3002A)
www.americanradiohistory.com
HANDBOOK A 432 MHz Low Noise Converter System 603
- ..,,_ ..
Figure 46
is provided by a small capacitor. Diode in order to achieve a snug fit. The shield is
current of the preceding mixer stage is soldered to the copper -clad board after all
monitored by inserting a low resistance 0- parts are mounted. The side flanges of the
to-1 d -c milliameter in jack J,. The cath- shield should make contact with the sides
ode of the 6CW4 is tapped down on the of the box and it may be fastened in posi-
input circuitry to achieve a good impedance tion with sheet -metal screws after final as-
match. The three circuits may be stagger - sembly.
tuned if fairly flat response is desired over 1-F Amplifier The amplifier is connected to
the 28- to 30 -MHz range. Power leads are Adjustment the mixer stage and to the
decoupled by feedthrough capacitors to communications receiver by
reduce fundamental- frequency pickup. coaxial lines. A 29 -MHz or 432 -MHz test
Amplifier Construction The i -f amplifier is signal may be used, depending on whether
built in an aluminum or not the mixer stage is available. The
box measuring 4" X 2 %g" X 15/8" (Bud slugs of coils L, and L. are peaked for max-
3002A). A piece of copper -clad (one side) imum signal, followed by adjustment of
phenolic board is mounted on the inside of coil L:,. A check across the 10 -meter band
the box with a 4 -40 screw in each corner. A will show if the amplifier circuits need to
shield of thin "flashing" copper straddles the be stagger -tuned to flatten the response.
nuvistor socket, isolating the input circuit Optimizing The As is, two of these simple
from the plate circuit. The grid terminal 432 -MHz R -F r -f amplifier units will out-
of the socket is bent down to the shield Amplifier perform most vacuum -tube
and soldered to it. The notch in the shield amplifiers by several decibels
to clear the socket should be filed to shape of noise figure at 432 MHz. Little adjust-
604
1111111
Receivers and Transceivers
11U111111111111111MVUIL.
UMIINES11S11111`
M- NM/
=.1 I. IS
wtttt tt
%1111111
11.1..=1111
ittjit111i
tttttttSW=
..
1N
MIIMINIM.
tlPI/
7R 1111
Irra
..
r
I
r
-i=
'0 ..4i
-
II
THE RADIO
oommom
raM7111r
/ IPIt.'
Mara
-15 MIN -I
Figure 47
ment need be made when using the 2N3478 optimum value of resistance is not critical
transistors, or other transistors of this gen- (plus or minus a few hundred ohms) and
eral family. No doubt better and cheaper may easily be determined by listening tests.
transistors will be developed over the years
as the trend to transistor television receivers 26 -7 The Deluxe
grows. With suitable modifications in the HBR Receiver
bias circuitry, this circuit will work well
with other transistor types. One of the most popular receiver designs
The gain and noise figure of this circuit in recent years has been the HBR circuit,
may be set by adjustment of the base - the creation of Ted Crosby, W6TC, and
ground resistor (2.7K). The optimum value others. Described in this section is a mod-
of resistance may vary slightly between ernized version of this popular receiver in-
transistors, even of the same type. To op- corporating many improvements over the
timize the circuitry, therefore, it is only earlier HBR models. The Deluxe HBR am-
necessary to replace the fixed resistor with ateur -band receiver is expressly designed
a SK potentiometer and to adjust it for for high -quality performance on SSB and
optimum gain and noise figure on a weak c.w. and has a high order of selectivity and
signal. It will be found that the lower the stability. It has good dynamic signal range
value of the base -ground resistance, the to help protect it from excessive cross mod-
lower will be the current drain and circuit ulation caused by strong signals and fea-
gain. When the resistance is too high, the tures high-Q r-f circuits for "up front"
stage will break into oscillation. Before the selectivity. Best of all, the receiver may
point of oscillation is reached, the signal -to- be built for a modest price and without the
noise ratio of the stage will deteriorate. The use of special metal- handling tools.
www.americanradiohistory.com
HANDBOOK The Deluxe HBR Receiver 605
Figure 48
THE DELUXE HBR RECEIVER
The 19 -tube HBR receiver is a double -conversion superheterodyne covering the amateur
bands. Employing plug -in coils, the receiver combines simplicity of design with good r -f
selectivity. Delayed automatic gain control and an efficient product detector make the re-
ceiver well suited for SSB reception. Layout of the panel controls may be seen in this photo-
graph. To the left of the main tuning control are the antenna trimmer (C and the r -f gain
control (R ), with the Q- multiplier tuning (C, centered below. To the right of the main tuning
control is the first i -f gain potentiometer (R with the noise limiter switch (S ! and the
1
100-kHz calibration oscillator switch to the right. Across the lower edge of the panel are
I. to r.): Q- multiplier control (S ), bandwidth- adjust potentiometer (R ), mixer -gain potentio-
meter (R ), second i -f gain potentiometer (R ), .node selector switch (5 ), audio -gain potentio-
meter (R ) and agc time- constant switch (S). Below, to the right are the phone jack and the
main power switch.
The Deluxe HBR receiver (figure 48) is receiver in such a way as to accommodate
a double conversion superheterodyne employ- his particular operating conditions, and this
ing nineteen tubes. A high -C electron - flexibility has proven to be one of the out-
coupled oscillator is used in the first con- standing features of this receiver. A delayed
version stage to combine a good order of agc system provides ample control for local
stability with circuit simplicity. Receiver signals, yet allows full sensitivity for weak
coverage is restricted to the high- frequency signals.
amateur bands only( 80 through 10 meters), Auxiliary circuitry includes a 100 -kHz
and inexpensive plug -in coils are used to sim- transistor crystal calibrator, i -f noise limiter,
plify receiver construction and to achieve S- meter, and Q- multiplier. Construction of
high -Q circuitry. Separate gain controls are the Deluxe HBR receiver is done on two
provided for the r -f stage, the first i -f stage, chassis, with the r -f circuitry placed on a
the second mixer, and the low- frequency separate small chassis that may be assembled
i -f system. These adjustments permit the and tested as a separate unit, if desired. Re-
operator to establish the over -all gain of the ceiver r -f alignment is easily accomplished
606 Receivers and Transceivers THE RADIO
VI4 VIS
DET -AUDIO AUDIO
1r
VI V2 VS V9 V9 Vio I 2AU7
R IST MIX. TI I.F. T2 2ND MIX. T3 T4 I.F. TS
95kN BSkH,
GAIN GAIN
V12
PROD. DET.
Vil
CD NOISE
LIMITER
100 kHz
CALIB. V13
OSCILLATOR BFO
BAND RANGE (MN,)
so 5.1- 5.6 VIA V17 V4 V19
40 S.4-5.7 AVC AMP. NYC DIODES RECTIFIER REGULATOR
20 9.2-9.375
15 9.7 -9.95
10 13.2-14.05
Figure 49
by separate bandset and bandspread adjust- injection is used on the 20- ,15- , and 10-
ments using an auxiliary signal source. meter bands.
Electrical and mechanical bandspread
The Receiver A block diagram of the Deluxe tuning are both employed in this receiver.
Circuit HBR receiver is shown in fig- A high -ratio tuning dial (110:1) is used,
ure 49.
The R -F Section
the amateur bands between 80 and 10
-
The receiver covers
permitting easy tuning of SSB signals. In
addition, a tapped -coil bandspread technique
(see Radio Receiver Fundamentals chapter)
meters with sufficient overlap at the band is employed (figure 51). The tuning rate of
edges for auxiliary activities such as MARS. the high- frequency oscillator (expressed as
Plug -in coils are employed in the r-f tuned a percentage of frequency) may be matched
circuits. The r-f stage employs a 6BZ6 to the rate of the r -f and detector stages by
semiremote -cutoff pentode (V1) to provide
maximum weak- signal performance while proper adjustment of the padding capacitors
in the bandspread circuit.
still allowing freedom from crosstalk and
front -end overload. The r -f stage gain con- The I -F Section -Two intermediate fre-
trol is normally run open and is only backed quencies are used in the Deluxe HBR re-
off in the presence of strong local signals. A ceiver. The first intermediate frequency is
1615 kHz which provides good image re-
6BC5 serves as a high -gain, low -noise mixer
jection in the high- frequency range. The
(V_), with conversion oscillator injection on
second intermediate frequency is 85 kHz
the control grid. The injection level is ad- which provides excellent adjacent -channel
justable to provide optimum signal -to -noise selectivity. Separate gain controls are pro-
level consistent with good overload capabil- vided for the two i -f sections and also for
ity. The first conversion oscillator (V.1) is the second mixer stage. Normally, the i -f
a 6BH6 in a plate- feedback circuit having and mixer gains are retarded as the re-
good frequency stability. Fundamental-fre - ceiver has more than sufficient gain for all
quency injection is employed on the 80- modes of operation. Ample selectivity is
and 40 -meter bands and second -harmonic available at 1615 kHz to prevent broadcast
www.americanradiohistory.com
HANDBOOK The Deluxe HBR Reciever 607
and 1650 -kHz navigational -aid signals from (V,;) and a 12AU7 double triode (V16)
causing interference to desired signals. A comprise the agc system. The double -diode
6BJ6 (V5) is used in the high- frequency circuit (V,;A and B) and the 470K/0.01-
i -f stage, followed by a 6BC5 (V6) second pH RC combination determine the "on"
mixer. The second conversion oscillator is time of the attack network, permitting the
a 6C4 (V;) in a crystal -controlled Pierce 0.1 -fd agc capacitor to charge up in a
circuit. The conversion frequency is 1.7 relatively quick time. The capacitor remains
MHz. charged, as the 12AU7 (V16A) triode is
Small, high -Q i-f transformers designed cut off by this action, and there remains no
for 100 -kHz operation are padded down to discharge path to ground in the agc circuit,
85 kHz to provide excellent over -all selec- even when the voltage across the RC net-
tivity. The nose of the selectivity curve is work is removed. The time constant of the
under 2 kHz in width, with the over -all release network is considerably longer, and
passband measuring about 4 kHz wide at 60 is adjusted by the "slow- fast -off" agc switch
decibels below the reference signal level. A (S,). After a predetermined period, the
12AX7 Q- multiplier (V8) provides addi- voltage across this network decays sufficient-
tional i -f selectivity for c -w reception or ly to permit the triode release gate (V,6A)
may be used to place a rejection notch at to conduct and discharge the agc line ca-
any point in the i -f passband to attenuate pacitor. A slight degree of delayed agc
interference. An i -f noise limiter employing action is provided by applying fixed bias to
a 6AL5 diode clipper (V11) is employed in the attack diode to prevent the circuit
the plate circuit of the last 8f -kHz i -f from being tripped by background noise or
The Detector, AGC and Audio Section
Dual detectors are provided in the Deluxe
-
amplifier stage. Clipping level is adjustable. weak signals.
A single 12AU7 triode section followed
by a 6AQ5 provides sufficient audio level
HBR receiver. A 7360 is used as a product for earphone reception, or to drive a speaker
detector for SSB and c -w reception (V,_) to good room volume. Feedback is incor-
with local-oscillator injection on the control porated in the 6AQ5 stage to provide
grid. The i -f signal is injected at one deflec- smooth audio response.
tion plate, with the resulting output con- The S -Meter and Power Supply-The S-
taining signal components produced by the meter consists of a simple v.t.v.m. that
product of the input signals. The desired compares agc voltage against a fixed volt-
audio component is selected and filtered in age reference. The 6CB6 bridge plate cir-
the plate circuit of the 7360. A 6BH6 beat - cuit (V) is balanced for a meter null with
frequency oscillator (V12) serves for c -w no signal input to the receiver, and agc
and SSB reception. A 12AU7 infinite -im- voltage in the presence of a signal unbal-
pedance detector (V14) is used for a -m re- ances the bridge, causes a reading on the
ception. meter proportional to signal strength. The
The agc system makes use of an audio - meter may be used for all modes of recep-
controlled hang circuit especially designed tion, providing useable readings on c -w
for SSB and c -w modes. It features a very signals as well as SSB or amplitude modu-
rapid response that prevents receiver over- lation.
load on a syllabic burst of SSB, instantly re- The power supply utilizes a choke input
ducing the receiver gain. The gain reduc- circuit, with critical voltages regulated by
tion remains in effect as long as the signal is an 0A2. Standby is accomplished by relay
in evidence, then hangs on for about 0.5
RY which breaks the r-f stage cathode and
second after the removal of the signal. This
speaker circuits of the receiver when actu-
sequence reduces to a minimum the usual
"thump" that occurs at the start of a ated by the VOX circuit of the transmitter.
syllable and removes the "rush" of back- A separate filament transformer is used for
ground noise at the end of a syllable, both the oscillator tubes, permitting them to run
of which occur with less sophisticated agc continuously, thus drastically reducing the
circuits. A choice of fast or slow release warmup drift of the receiver, especially in
action may be made. A triple diode 6BC7 a humid location.
608 Receivers and Transceivers THE RADIO
Figure 50
www.americanradiohistory.com
HANDBOOK The Deluxe HBR Receiver 609
C A, B, C- 5- 23 -pf, 3- section. Miller 2102, or S -Two- section, 2 -poles per section, 4- position.
Polar C28 -143 -6 015 Centralab PA -2010
(Note: one rotor plate must be removed
from each section of the Polar capacitor to
T,, T - 1800 -kHz transformer. Pad to 1615 kHz
with 62 -pf silver micas. Miller 1730
achieve proper bandspread. The Miller unit T,, T,, T- 100 -kHz high -Q transformer, 2.5 -kHz
needs no modification) bandwidth. Pad with 130 -pf silver micas to
C -15 -pf Hammarlund MAPC -15B 85 kHz. Miller 1709. For 3 -kHz bandwidth,
C -2 -pf Hammarlund MAPC -15 with all but one use Miller 1710 transformers
rotor and one stator plate removed. T,- 132 -kHz bhp transformer padded to 85 kHz
C , C , C -See coil table, figure 57. (Hammar- with 17S -pf silver mica. Remove compres-
lund type MAPC) sion trimmer and use threaded bushing of
C -10 -pf Centralab 822EZ trimmer as mount for lugs to hold wires
C -740-pf Hammarlund APC -140B from bfo coil. Miller 012 -MS
C -7 -pf Centralab 822EZ T. -SK to 4 ohms. Stancor A -3877
Centralab 822AZ T, -3:1 audio interstage. Triad A -31X
Le L., L,-See coil fable, figure 57 T,-550-volt c.t., 110-ma. S volt 2 amp., 6.3
L 3.5 -mH Miller 9003 volt S amp. Triad R -12A
L,, L, -6- henry, 110 -ma, Triad C -11X T, -6.3-volt 1 amp. Stancor P -6134
L.-0.7 -mH, tapped. Miller 9012 S- meter -0 -I d-c milliammeter
RFC,100-mH Miller 960 Chassis -(1) 10" x 14" x 3" (2) S" x 7" x 2"
RFC, -50-mH Miller 958 BFO box -2s/%" x 2" x 4"
RT -dpdt relay with coil to match transmitter Cabinet -IS" x 11" x 9 ". Wyco CR -7725
control circuit Dial- Eddystone 898
S -Single -section, 3 -pole, 3-position. Centralab
PA -2006
Receiver A receiver such as this is a 6 -32 spade bolts mounted in the corners of
Constrrcfion complex device and its con- the smaller unit. Placement of the subchassis
struction should only be under- is shown in figure 54.
taken by a person familiar with receiving Alignment of the dial on the panel is
equipment and who has built and aligned determined by the positioning of the main
equipment approaching this complexity. The tuning capacitor (C,A -B -C). The capaci-
first step is to lay out the chassis, panel, tor used is a high -quality unit having full
tuning dial, and larger components in a ball -race bearings front and back and a
"mockup" assembly to ensure that the re- controlled torque. This unit provides mini-
ceiver will go together without a conflict mum drag on the geared dial. The tuning
between the components. The receiver is capacitor is mounted above the chassis on
built on an aluminum chassis measuring two bolts as shown in the various illustra-
10" X 14" X 3". A chassis having welded dons. Note that the capacitor is insulated
seams with triangular braces in each corner from the top of the chassis and the dial, the
is recommended for maximum rigidity. The rotor frame being grounded by a separate
complete receiver fits in a steel cabinet ground strap running from the capacitor
measuring 11" X 15" X 9". A series of 1/8- frame to the under -chassis area of the r -f
inch holes are drilled around the upper edges unit. This grounding technique avoids
of the sides and back of the chassis for spurious ground loops in the r -f assembly
ventilation and another series of 3/B- inch that may give rise to regeneration and in-
holes is drilled across the rear top and bot- stability. The tuning capacitor is driven via
tom edges of the cabinet. The bottom of a rigid insulated bushing and the supporting
the cabinet, in addition, is "honeycombed" bolts are adjusted and locked after align-
with 3/8 inch holes. Additional holes are also ment of the front panel, thus assuring mini-
required in the rear of the cabinet for mum dial drag.
various cables and terminations to plugs The receiver chassis rests on the floor of
and receptacles mounted on the rear apron the cabinet and the chassis is mounted nearly
of the chassis. flush with the bottom edge of the panel.
The r -f circuits of the receiver are built The dial assembly is positioned as shown in
on a steel subchassis measuring 5" X 7" X the layout drawing (figure 55). It is sug-
2" which is mounted above the main chassis gested that the r -f chassis be temporarily
as shown in the photographs. The subchassis placed on the larger chassis after mounting
is affixed to the main chassis by means of the tuning capacitor and dial placement
610 Receivers and Transceivers THE RADIO
The chassis, subchassis, dial, and panel mixer (6BC5, V,;) are located on the main
should be assembled and studied before the chassis directly behind the r -f subchassis.
chassis holes are drilled. Placement of the The 12AX7 Q- multiplier is placed in the
remaining components may be done from a rear left corner of the chassis, with the 6C4
www.americanradiohistory.com
HANDBOOK The Deluxe HBR Receiver 61 1
Figure 52
REAR OBLIQUE VIEW OF RECEIVER
R -f components are mounted on a subchassis placed atop receiver chassis (right). The chassis
is held in position by spade bolts in the corners. The main tuning capacitor is insulated from
the chassis and grounded to it by a separate strap. It is held in position by long bolts
insulated from the chassis by fibre washers.
Along the rear apron of the chassis ore (1. to r.): power cord, line fuse, speaker jack,
accessory socket, notch -depth potentiometer (R,), and antenna receptacle. Ventilation holes
may be seen around the edge of the chassis. A plug -in silicon rectifier is in use in place of
the 5V4.
crystal oscillator and 1.7 -MHz crystal at The noise limiter and detector tubes are
the right end of the i -f strip, as seen in the near the front of the chassis, adjacent to
photographs. The 85 -kHz i -f system is posi- i -f transformer T. The product detector
tioned from front to back along the center and audio system are to the right -front of
of the chassis. The transformers are oriented the chassis.
so that plate and grid leads to adjacent
Beneath the chassis (figure 56), the bfo
sockets are short and do not cross over each
components are housed in an aluminum box
other. Sockets, too, are oriented to provide
measuring 2" X 23/4" X 4", centered under
short grid and plate leads.
The AGC, Audio, and Power Systems
The power supply occupies the right rear
- the bfo tube socket. Various chokes and
transformers are mounted to the side wall
corner of the chassis, with the `S -meter ad- of the chassis, with the Q- multiplier con-
justment' and "S -meter zero' potentiometers trols mounted on a subpanel placed near the
placed in front of the power transformer. rear of the receiver. Extension shafts couple
612 Receivers and Transceivers THE RADIO
ma AIN Mk .111
Figure 53
TOP VIEW OF RECEIVER
Major above- chassis components may be seen in this photograph. Directly behind the 5 -meter
(upper left) are the 6AQ5 and 7360 tubes and nearer the power transformer are the 6BC7
and the 12AU7 agc tubes. Adjacent to the power transformer are the two S -meter potentiom-
eters. To the right of the power transformer are the 5V4 rectifier and the dual 40 -40 ;:fd
filter capacitor. Next to the capacitor (towards the panel) are the 0A2 regulator and the
6CB6 5 -meter tubes and the agc test -point jack. Closer to the panel is the bfo transformer
(T) with the bfo tube between it and the panel. Down the center of the chassis )rear to
front) are the 1.7 -MHz crystal, 6C4 oscillator, 85 -kHz transformer )T ), i -f amplifier 6BJ6 (V
transformer T , i -f amplifier 6BJ6 (V, ), and transformer T (near panel). To the left of trans-
former T are the 12AU7 (V ) and the 6AL5 (V I. The a n I control (R ) is to the side of
transformer T
Along the rear of the chassis (at right) are the slug of coil L (in the corner), 12AX7 (V
6BJ6 (V 1, transformer T, and 100 -kHz test point. At the front of the chassis (between the
subchassis and the panel) are the 100 -kHz crystal and associated components. Note that the
tuning capacitor is coupled to the dial with an inflexible (rigid) coupling and short shaft
extension. The flat ground strap on the main tuning capacitor may be seen passing through
a slot to the under -chassis area where it is grounded.
these controls to the panel dials. Panel separate leads running from the filament
bushings are placed on all extension shafts. transformer to each branch. To reduce r -f
ground- current intercoupling, all grounds
Receiver Wiring The receiver should be for a single stage should be returned to that
wired in an orderly man- stage, preferably to a common ground point
ner, a stage at a time. The power supply near the tube socket. The cathode, agc -by-
and filaments should be wired first. The pass, and screen capacitors, for example, can
builder should avoid overloading the fila- all return to a ground connection near the
ment wiring by wiring the sockets in sev- cathode pin of the socket in question. Com-
eral branches of four to five tubes, with ponents should be grouped about a socket
www.americanradiohistory.com
HANDBOOK The Deluxe HBR Receiver 613
6 +-
COMPONENTS CENTERED
ALONG TN/S LINE.
5
POWER TRANSFORMER
-F CRASSI 7
R
APPROX 3f 3f
o
SRI[LD ASS'Y.
*
17
COMPONENTS CENTERED
io W ALONE STS LINE
R
2W
Q
W,2
ON
2+ I.D.
Z.;
42
u
COMPONENTS CENTERED
ALONG TANS LINE.
\"CT\OUT\ ,+
#*
31.1*- ... -2+ F = DIMENSION TO BE CHECKED IN PRELIMINARY ASSEMBLY.
Figure 54
where possible, and not "stacked" above the should run in shielded braid, with the
socket so the latter can be reached for volt- braid grounded at both ends of the leads.
age measurements. The i -f system and crystal conversion
Before bfo transformer T6 is mounted, oscillator may be checked by injecting a
the compression -type mica trimmer capaci- 1615-kHz test signal at the grid of the
tor is removed from the case and the trans- first i -f amplifier tube (V;) . The complete
former reassembled. A 25 -pf ceramic trim- receiver, less the r -f assembly may be com-
mer (C11) in parallel with a 175 -pf silver - pleted and checked, stage by stage as work
mica capacitor are mounted under the chas- progresses.
sis in the bfo inclosure. This substitution
removes a slight frequency instability The R -F Circuits Plug -in coil data is given
noted on SSB signals due to the flexing of in the accompanying table.
the spring on the compression capacitor. The coils are easy to wind and the receiver
After the power supply has been wired is simple to align. Five -pin, 11/4-inch diam-
and tested, the audio system may be checked eter polystyrene coil forms, available from
out by applying an audio signal to the top Allied Radio Co., Chicago, Ill. (catalog
end of the audio gain control (R;) . The number 46 -Z-696) are used. Figure 57
bfo may be checked for proper operation summarizes the windings and shows them in
with a v.t.v.m., if handy, by measuring the relation to the coil pins. All coils are tightly
rectified r-f voltage at the plate of the wound in the same direction on the form,
6BH6 oscillator tube, which should be about and an air padding capacitor is mounted
10 volts. Power leads from the bfo inclosure within the form. The MAPC -style capaci-
614 Receivers and Transceivers THE RADIO
-4-4 A
e
Zi , -4-
2 h -
REF. POINT
FOR DIAL CUTTING AND DRILLING TEMPLATE
PROVIDED WITH DIAL I
- -fr,
/ 1
/ -- -\ 11I
SJ
1
\ .-TL
1
i S-MITER
-.--,
CUT OUT TO MATCH
9
l 1
I
,
\
4-
OIDIAL
1
_ ll
I-
-EID
4i-
-
H
4
I ,-I
I
4E0- Tti
15
NOTE' ALL PANEL HOLES FOR CONTROL SNAPTS ARE VO/A. HOLES FOR TOGGLE SWITCHES ARE I- O /A.
Figure 55
tor used as a trimmer should have brass The amount of wire for the winding is
(not aluminum) plates for best frequency estimated and one end of the length is
stability. The Hammarlund units are recom- cleaned, passed through the predrilled hole
mended. The trimmer capacitor is held in above pin 1, and into the pin. It is brought
place by Duco cement, plus the wire leads out the end of the pin and quickly soldered
running to it from the proper base pins. A with a hot iron. Hold the pin with a long -
silver -mica padding capacitor is used in the nose pliers acting as a heat sink so that the
oscillator coil assembly and is mounted in coil form will not be deformed by excessive
position as shown in the drawing. The rotor heat. The free end of the length of wire is
terminal of the MAPC padding capacitor is now attached to a vise or stationary object.
connected to the r -f ground end of the coil The wire is straightened by a gentle pull,
winding and the stator to the topmost turn and wound on the coil form under tension,
(grid) of the coil. rotating the top of the form towards you,
The first step is to wind the primary coil
keeping the wire taut at all times. When the
(B winding). Make two small holes in the
sidewall of the form for the connections to proper number of turns are on the form,
pin 1 and pin 5. The holes are about % -inch grasp the winding to prevent it from un-
apart with the pin -5 hole close to the point raveling and cut the wire a few inches
where the bottom turn of the secondary longer than the length needed to go through
winding will be placed. This allows suffi- the predrilled hole above pin f and to pro-
cient space to slide the primary winding up trude through the pin. Clean the end of the
or down the coil form to provide the proper wire, thread it through the form and out
degree of coupling between the windings. the pin. Pull it taut, solder the wire to the
www.americanradiohistory.com
HANDBOOK The Deluxe HBR Receiver 615
lik swam It
Figure 56
pin tip and trim off the excess length. approximate the turn spacing as the coil is
Scrape the pin free of rosin and solder. wound, and the spacing may be adjusted
The same technique is used for the larger after the winding is finished, if necessary.
secondary winding. An extra hole is needed The winding can be temporarily wound on
in the coil form for the tap connection to the form in order to determine the position
the winding. The tap hole is somewhat of the tap hole. Once the winding is posi-
larger than the others (make it about Y4" tioned, the enamel on the wire is scraped
diameter) to permit the joint to be soldered away at the tap point and a small length of
without the iron damaging the low- melting wire soldered at this spot, passed through the
polystyrene form. The ends of the secondary tap hole and out through pin 3.
winding are not soldered until the additional The last step is to fix the MAPC padding
wires of the MAPC padding capacitor are capacitor in place and to mount the auxiliary
also inserted in the same pins. It is easy to silver -mica padding capacitor used in some
616 Receivers and Transceivers THE RADIO
COIL TABLE
L'A,L2A 29 TURNS No 26, L18: 57/6 TURNS No 26. L IA. L2A. 111/2 TURNS No. LIB: 3-7/6 TURNS No. 26,
CLOSE -WOUND. THEN 31/2 SPACED 3/6 IN. FROM LIA. 22. LENGTH 15/16 INCH. SPACED 5/16 IN. FROM LIA.
TURNS SPACED I/4 INCH, L2 B 9 -7/6 TURNS No. 26 TAPPED AT
THEN 4 TURNS CLOSE-W.0
1/4 TURNS. L2B: 3.7/6 TURNS No. 26.
SPACED 5/16 IN FROM L24 LEA: 6.1/2 TURNS No. 22. SPACED 7/16 IN. FROM L2A.
TAPPED AT 31-1/4 TURNS; L3B, 11.7/6 TURNS No. 26, 14 MHz LENGTH 1/2 IN. TAP'D AT
(TOTAL 36-1/2 TURNS).
L3B' 11-7/6 TURNS No 26,
SPACED 3/16 IN. FROM L3A. 6-1/4 TURNS. SPACED I/O IN. FROM L3A.
3.5MHz L3A: 15 TURNS No 22 CLOSE - C4,C5: 25 -PF AIR PADDER
WOUND, THEN 312 TURNS Ce. 50 -PF AIR PADDER,
SPACE -WOUND OVER 5/16 +200 -PF SILVER MICA.
IN., TAPPED AT 16 -1/4
TURNS (TOTAL 161/2 T'S).
C4, CS:S0 -PF AIR PADDE R. LIA,L2A: p 1/2 TURNS No 22 LIB' 3.7/6 TURNS No. 26,
C6:75 -PF AIR PADDER. LENGTH 7/6 IN; TAP'D AT SPACED SA6 IN. FROM L1 A.
2-1/4 TURNS. L2e: 3 -7/6 TURNS No. 26,
21MHz L3A: 5 -1/2 TURNS No. 22 SPACED SA6 IN. FROM L2A.
LENGTH 3/61N, TAP'D AT L3B 6.7A! TURNS No. 26,
LIA. L24. 6-1/2 TURNS No. 22, Ll B. 2 -7/6 TURNS No. 26. '1/4 TURNS. SPACED 5/32 IN. FROM L3A.
CLOSE -WOUND, THEN 16 SPACED 7/16 IN. FROM LIA. C4.CS: 25 -PF AIR PADDER.
TURNS SPACE-W.13 TO AN L2B: 3 -7/6 TURNS No. 26, Ca- SO -PF AIR PADDER
OVER -ALL LENGTH OF 11N; SPACED 3/6 IN. FROM L2A. +140 -PF SILVER MICA.
TAPPED AT 9-3/6 TURNS. L3B: 10 -7/6 TURNS No. 26,
(TOTAL 22 -1/2 TURNS ) SPACED 3/32 IN FROM L3A.
7 MHz L3: 6 -1/2 TURNS No.22, LIA,L2A, 5-1/2
CLOSE -W'D THEN 7 T'S
TURNS No 22. LIB: 3-7/6 TURNS No. 26,
LENGTH IS /16IN; TAPPED SPACED 361N. FROM LIA
SPACE -WOUND TO AN 0/ER- AT 2-3/6 TURNS.
ALL LENGTH OF 9/16 INCH; L2B: 3 -7/6 TURNS N. 26,
L3A S 1/2 TURNS No. 22,
1
SPACED 3/6 IN. FROM L24
TAPPED AT 13 -1/4 TURNS, 28MHz LENGTH 1/2 IN; TAPPED L3B: 6 -7/6 TURNS No 26.
TOTAL 13-1/2 TURNS ) AT 5 -1/4 TURNS. SPACED 5/32 IN FROM L3A
C4, :SO -PF AIR PADDER C4,C5: 25 -PF AIR PADDER
Ce.50 -PF AIR PADDER, C6: 50 -PF AIR PADDER,
+66 -PF SILVER MICA. +47 -PF SILVER MICA.
NOTES;
ALL COILS WOUND WITH ENAMELED WIRE ON 1f
- INCH DIAMETER POLYSTYRENE 5 -PIN PLUG -IN FORMS,
(ALLIED RADIO 46-Z-495) TAPS ARE COUNTED FROM BOTTOM END OF COIL.
ON 'A' COILS, TURNS SHOULD BE EVENLY SPACED TO LENGTH SPECIFIED; 'B"
COILS ARE CLOSE -WOUND.
"A' AND 'B "COILS ARE WOUND IN SAME DIRECTION.
TAP 2
PIN 3
S
2 4 2 1 5 4
(A) (B)
Figur. 57
COIL TABLE FOR DELUXE HBR RECEIVER
oscillator coils. Lengths of bare wire are toothpick. The fixed padding capacitor of
soldered to the rotor and stator terminals of the oscillator coil is mounted between one
the variable capacitor. The wires pass into stator post of the MAPC capacitor and the
pins 2 and 3, with the rotor going to pin 2 wire dropping down from the MAPC rotor
(r -f ground) and the stator going to pin 3. to pin 2. The assembly of MAPC capacitor
Once the wires of the MAPC capacitor are and fixed padding capacitor is soldered to-
in place, along with the ends of the second-
gether before the unit is slipped into the coil
ary winding, the pins may be soldered. Be- form.
fore soldering, check that no wires interfere When one set of coils is finished, it may
with the rotation of the capacitor and that be placed in the receiver and the circuits ad-
the capacitor may be turned easily without
justed to the approximate frequencies by
the rotor binding on the side of the coil means of a grid -dip oscillator.
form. When soldering is completed, the
MAPC capacitor may be wedged in place be- Receiver Alignment The i -f system of the
fore cement is applied using a length of Deluxe HBR receiver is
www.americanradiohistory.com
HANDBOOK The Deluxe HBR Receiver 617
aligned first, followed by the r -f section. test signal loosely coupled to the plate of
While an experienced builder can align the the 6BC5 first mixer (V2). Transformers
receiver "by ear ", it is recommended that a T, and T2 are aligned for maximum re-
BC -221 (or LM) frequency meter be used sponse at this frequency. Gain controls are
for alignment along with a grid -dip oscilla- retarded to prevent overload as receiver
tor and a general -coverage receiver. Align- gain rises. Instability may be noted when
ment is done by injecting signals of various controls are advanced to a position of maxi-
frequencies into the receiver and peaking mum gain. The receiver has a large reserve
the adjustable capacitors of the tuned cir- of gain, and with correct alignment, no sign
cuits for maximum response. If the test of oscillation will be apparent at normal
signal is modulated with an audio tone, the operational gain levels.
receiver response may be noted on a high R -f Alignment -R -f alignment is ac-
impedance a -c voltmeter placed across the complished by adjustment of the various
speaker output terminals. If the test signal padding capacitors in the plug -in coils, and
is unmodulated, the receiver S -meter may by varying the inductance of the coils if
be employed, or a vacuum -tube voltmeter need be. Two adjustments are necessary-
placed across the agc line may be used. tracking and bandspread. Both of these ad-
I -f Alignment -The receiver bfo is used justments are performed on the high-fre-
for alignment of the 85-kHz i -f system. quency oscillator circuit and are then re-
With the "function switch" in the bfo posi- peated for the mixer and r -f circuits. The
tion, the second harmonic of the bfo is ad- procedure is best carried out first on a low -
justed to 170.0 kHz using the low- frequency frequency band, and the 40 -meter adjust-
range of the BC-221 frequency meter. A ments are chosen as an example. The align-
small amount of signal from the bfo is ment chart shows that the proper 40 -meter
coupled into the plate circuit of the second alignment is achieved when the high -fre-
mixer tube (V,;) and transformers T, T,, quency oscillator stage tunes the 5.4- to 5.7-
and T; adjusted for maximum response. To MHz range while the r -f and mixer stages
couple the bfo to the second mixer, run a tune the 7.0- to 7.3 MHz range. The first
length of wire from the top of the 7360 step is to adjust the oscillator range for
socket (remove the tube and probe pin 3) proper bandspread, which is accomplished
to the mixer. Wrap the wire around the with the aid of the BC-221 frequency meter.
bulb of the mixer. Switch S, is set in the The BC-221 is set to 5.4 MHz and the
SSB position for this procedure. Once the main tuning dial of the receiver is adjusted
i -f system is aligned to 85 kHz, the bfo so that the tuning capacitor is about 90
may be adjusted for proper c -w and side - percent meshed. Oscillator trimming ca-
band selection. This is done with the BC- pacitor CA is set at mid -capacitance. Oscil-
221 frequency meter, adjusting the second lator padding capacitor Cc (in the coil form)
harmonic of the bfo to 168.4 kHz for SBI is adjusted until the frequency of the oscil-
reception and to 171.6 kHz for SB2 re- lator is 5.4 MHz, as measured on the fre-
ception. The bfo is adjusted to 169.2 kHz quency meter. The dial reading of the re-
for c -w reception. Note that SRI and SB2 ceiver is noted and the BC-221 is then set
alternate between upper and lower sideband. to 5.7 MHz. The main tuning dial of the
On 80 meters where the high- frequency receiver is adjusted to tune the high -fre-
tuning oscillator is higher than the received quency oscillator to the same frequency,
signal, upper sideband is reversed from the which should occur with the tuning ca-
20 -meter situation, where the tuning oscil- pacitor about 10 percent meshed. If it does
lator is lower than the received signal. not, the oscillator padding capacitor (CO
Once the low- frequency i -f system is should be readjusted to properly place the
aligned, the 1.7 -MHz conversion crystal and 5.7 -MHz checkpoint on the receiver dial.
6C4 oscillator tube are plugged in and If the padder must be increased in capaci-
crystal operation is checked by tuning the tance to align the circuit to 5.7 MHz, it
general coverage receiver to 1.7 MHz and indicates the tap on the oscillator coil is too
noting stable oscillator operation. Next, high and that the portion of the winding
the BC -221 is tuned to 1615 kHz and the between the tap and ground must be spread
618 Receivers and Transceivers THE RADIO
CHART 1.
VOLTAGE CHART FOR RECEIVER
Measurements made with VTVM. All controls 1/2 advanced except r -f gain which is fully
advanced. Q- multiplier on boost, Function switch on 582, avc on fast,Noise limiter on.
TUBE 1 2 3 4 5 6 7 8 9
V, -4.1 0 6.3 0 50 88 0 - -
V, - 250 - 280 - 280 - 250 -
V,. -0.3 31 6.3 0 235 215 0 - -
V 0 31 6.3 0 240 220 31 - -
V: 110 - 6.3 0 110 -15 0 - -
V, 230 0 2.8 0 0 -0.7 -0.5 0 6.3
slightly apart to decrease the inductance. tuned to the 5.4 -MHz oscillator point. The
If the padder must be decreased in capaci- test signal should be heard in the receiver,
tance, it indicates that the tap on the coil and the detector padding capacitor adjusted
is too low, and therefore the portion of the for maximum response. The signal generator
winding between tap and ground must be is now set to 7.3 MHz and the receiver dial
bunched together to raise the inductance. tuned to the 5.7 -MHz oscillator point. The
By slight adjustment of the lower portion detector padding capacitor is readjusted for
of the oscillator coil winding, the 5.4 -MHz maximum response, noting whether the ca-
and 5.7 -MHz points may be placed near the pacitance is increased or decreased. The
ends of the tuning dial, and the proper winding below the tap of the mixer coil
coverage is positioned on the dial without (L_) and ground is now adjusted in the
necessitating readjustment of the padding fashion described for the oscillator coil until
capacitor in the oscillator coil. If more or the setting of the padding capacitor remains
less bandspread is desired, the tap on the
the same at both ends of the tuning range.
coil may be moved a fraction of a turn,
changing the inductance of the winding be- This adjustment is repeated with the r -f
low the tap. stage, with the r -f trimming capacitor (C._)
Once the oscillator circuit tracks across set at mid -scale. With due care, the whole
the appropriate range, the mixer tube may alignment operation should take less than an
be placed in its socket, along with mixer hour for the first set of coils, and with ex-
coil, L. The BC -221 frequency meter is perience the adjustments to the remaining
adjusted to 7.0 MHz and the receiver dial coils may be done in less time.
www.americanradiohistory.com
HANDBOOK The Deluxe HBR Receiver 619
www.americanradiohistory.com
CHAPTER TWENTY -SEVEN
www.americanradiohistory.com
6- and 2 -Meter Exciters 621
6U8A 6CL6
05C. (6.25-90MN) rims. -QUAD (250 -270MM) DOUBLER (46 -54 MHz)
68
X, J3
"--o) 50 MHz
6
if
,00 n = 00,
68
220
+250 v
+250 v
Jt to J,. J2
+ 250 v
6116* .00,
ALL RESISTORS I WATT UNLESS OTHERWISE
NOTED
ALL .001 OF ARE DISC CAPACITORS.
Hi.
6CL6
00, Jt
Figure 2 1 2 3 4 5 6 7
Output Xtal and Mull. Doubler Tripler BASIC TWO -METER EXCITER
(MHz) L1 -C1 L2-C2, L4-C4,
L6-C6
ITC, Push -pull 6CL6 (or 6360)tripler stage is added
L, -C, to exciter of figure 2 to reach 2 meters.
6.25- Crystal switch and multiple crystal sockets
6.75 MHz 50.0- are added. Panel controls (I. to r.) are: crys-
30
8.334-
25.0-
27.0 MHz 54.0 MHz
tal switch, coils L, and L,, coils L and l, and
plate- tuning capacitor C,. Separate oscillator
9.0 MHz coils are used for each crystal and are
6.0- mounted behind the two octal sockets used os
144 6.166 MHz 24.0-
crystal mounts.
48.0- 144.0-
8.0- 24.66 MHz 49.33 MHz 148 MHz
8.222 MHz
6CL6
6CL6 TOI PL ER
DOUBLER (48.0 -4933) e
tripler stage is tuned to 144 MHz and out- L4 L5 7
put taken from a pickup link inserted at
is
the center of the plate coil.
Proper operation is monitored by grid -
current metering in the various multiplier
stages, accomplished with a 0 to 1 -ma d -c
meter. The metering circuit provides a full
scale reading of ma at test point J, and
1
ALL .001 tIF
S ma at 1.. and J:,, The screen voltage for AND 220 PF
ARE DISC
the multiplier stages has been brought out CAPACITORS
www.americanradiohistory.com
HANDBOOK 6- and 2 -Meter Exciters 623
Sockets and other components are posi- tuned for maximum meter indication (about
tioned to permit very short interconnections. 0.3 ma). Adjust the slug so that the oscil-
The coils in the bandpass coupler in the lator starts immediately each time plate volt-
grid circuit of the tripler stage (L, -L5) are age is applied.
placed on a small angle bracket mounted be- Screen voltage is next applied to the
tween the two 6CL6 tripler sockets and the multiplier section of the 6U8A and the
driver tube socket. The coils are spaced 5/e" 6CL6 is placed in its socket. The 8.45 -MHz
center -to- center and are mounted 3/4" below crystal is substituted for the 8.375 -MHz
the chassis. A second small angle bracket unit and the slug in coil L2 is adjusted for
supports the 144 -MHz tripler plate- tuning maximum 6CL6 grid current (about 1.3
capacitor. ma). This step is followed by adjustment of
Extension shafts for the tripler tuning the slug of coil L. for maximum grid cur-
adjustments brought out through the front rent using the 8.375 -MHz crystal. Switch-
of the chassis are made from !!" lengths
1 ing crystals back and forth and adjusting
of 1/4" diameter brass rod. A tapped hole for
a 6 -32 screw is placed in one end of two of
the rods and a screwdriver slot is cut in the
opposite end. After the coils have been
mounted to the bracket, a 6 -32 nut is run
on the slug shafts and a brass extension
threaded on each coil shaft and locked in
position by the nut. The shaft is supported
in position by a panel bushing.
The tripler plate- circuit tuning capacitor
(CO has a 3 /16" diameter shaft and the
extension for this control is attached to it
by a special coupling made of a short length
of ?/g" diameter brass rod, drilled at one end
for the capacitor shaft and at the other end
for the panel shaft. The coupling is soldered
ro the panel shaft and is drilled and tapped
for a set screw for the smaller shaft. Figure 7
All disc ceramic capacitors are placed in
UNDER -CHASSIS VIEW OF
position with the shortest possible lead length
TWO -METER EXCITER
and those units which bypass the screen -grid
terminals in the various stages should be Crystal switch, multiple crystal sockets, and
oscillator plate coils are at the left. Switch is
connected between the screen and cathode a 2 -pole, 4-position ceramic deck; one pole
(ground) pins of the respective tube socket. switching the crystals and the other pole
switching the coils. Broadband circuit Li-L, is
Most resistors are soldered between socket mounted on bracket near the center of the
pins and tie -point terminals. Power leads are chassis, with 6CL6 tripler sockets at right.
run close to the chassis to reduce r -f pickup. Air -wound tank coil 1, and butterfly tuning
capacitor are at right.
Exciter The 50 -MHz Exciter -The ex-
Adjustment citer is adjusted for broadband the slugs of coils L_ and L, will permit
operation between 50 and 51 substantially constant 6CL6 grid current
MHz. Two crystals at approximately 8.375 to be measured at test point J2 across the
and 8.450 MHz (output frequencies of appropriate frequency range.
50.25 and 50.7 MHz respectively) are used The next step is to connect a suitable
for initial adjustment. After the wiring has dummy load, such as four #47 pilot lamps
been checked, power is applied to the 6U8A (6.3 -volt, brown bead) in parallel to output
tube. The negative terminal of a 0 to 1 -ma receptacle J:, and apply all screen and plate
d -c meter is connected to test point J, and voltages to the exciter. The slug of plate
the positive terminal of the meter is coil L, is adjusted for maximum bulb bril-
grounded. The slug of oscillator coil L, is liance. Using various crystals in the proper
624 Low -Power Transmitters THE RADIO
range, the exciter coils may now be repeaked The power output of either exciter is
to provide nearly constant power output ample to drive any of the popular twin
from the exciter across the lower 1 -MHz beam -power tubes designed for operation in
portion of the six -meter band. the vhf spectrum (832A, 829B, 815, 5894,
When the 50-MHz exciter is coupled to etc.). Sufficient drive also exists for a single
the grid circuit of a succeeding amplifier 4X1S0A, 4CX2S0B, or 4CX300A tetrode
through a short length of coaxial line, plate at a power level up to several hundred watts.
coil L, should be readjusted so that maxi-
The 6360 A single 6360 double
mum drive to the amplifier occurs at about
Frequency Multiplier tetrode may be substi-
50.3 MHz. If the amplifier grid tank is then
resonated for maximum drive at about 50.6 tuted for the push -pull
MHz, little variation in drive will be noticed 6CL6 multiplier stage (figure 8). Higher
over the 1 -MHz operating range. plate voltage-up to 350 volts -may be
The 144 -MHz Exciter -The procedure applied to the 6360, provided that this
outlined for the 50 -MHz exciter is followed voltage is applied only to this stage and not
for the 144 -MHz exciter, with the plate to the rest of the exciter. Power output of
circuit of the 6U8A tripler tuned to 24.15 the 6360 will be about 10 watts at the
MHz and the grid circuit of the 6CL6 higher plate potential.
doubler tuned to 24.45 MHz. The plate cir-
cuit of the 6CL6 is tuned to 48.3 MHz, 27 -2 A 175 -Watt
(crystal frequency of 8.05 MHz). The grid SSB Exciter
coil of the push -pull tripler stage is peaked
for maximum grid current (about 2.5 ma) Building a single -sideband exciter or
at 48.9 MHz (crystal frequency of 8.1 S transmitter is simpler and less expensive
MHz). This staggered- tuning adjustment than construction of a -m equipment of
should result in little variation of grid cur- equivalent power rating. Physically, the
rent in either the doubler or tripler stages SSB exciter can be made more compact
over the range of 144 to 148 MHz. and lighter in weight for an equivalent de-
The tripler plate tuning capacitor (CG) gree of "talk power" as compared to the
is resonated to the operating frequency and a-m equipment. The wide acceptance of SSB
should be retuned each time a shift in fre- has produced suitable inexpensive parts and
quency greater than 200 kHz is made. components so that it is now no longer
difficult or expensive to build and align a
high- quality SSB exciter. Only the simplest
6360
TRIPLER of test equipment is required and the use of
6CL6
DOVBLER 9-33 MM) a commercial crystal sideband filter in the
L4 L5 3 Le
exciter eliminates critical circuit adjust-
144 MIR ment and tinkering. In addition, the side -
band suppression may be easily optimized
and spurious responses attenuated to a great
degree.
The exciter described in this section (fig-
ure 9) is of a proven design and is recom-
ALL .001.1.1F
AND 220 Pi
ARE DISC
mended to those experimenters wishing to
CAPACITORS build their first piece of sideband trans-
mitting equipment.
The Exciter This filter -type exciter incorpo-
+250V. +250V. +250 -350V. Circuit rates all the desirable features of
more expensive exciters, covering
the amateur bands between 10 and 80 meters
Figure 8 with a minimum of controls and adjust-
ments. The output stage utilizes a pair of
ALTERNATIVE 6360 TRIPLER CIRCUIT highly linear 65 50 tetrode tubes run at a
www.americanradiohistory.com
HANDBOOK 175 -Watt SSB Exciter 625
Figure 9
PEP input of about 175 watts. A block product of these two signals is used as the
diagram of the exciter is shown in figure 10. basic exciter frequency range of 3.5 to 4.0
The sideband generator is designed around MHz, which appears in the plate circuit of
a 9 -MHz crystal lattice filter and consists the mixer stage. All the higher amateur
of a 7360 oscillator /balanced modulator bands are derived by mixing this SSB signal
(V7) with a 12AX7 speech amplifier (V1) with an auxiliary crystal oscillator. The out-
modulating one deflection plate of the 7360. put from the 6BA7 mixer is bandpass cou-
The sideband filter has low- impedance in- pled to a 12BY7A (V3) which operates as
put and output terminations and is link - an amplifier on the 80 -meter band and as a
coupled to the output circuit of the bal- second mixer for all the higher- frequency
anced modulator. The filter drives a 6B6 bands. The mixing oscillator is a 6C4 (V10)
i -f amplifier stage (V7) to bring the signal whose output is always higher in frequency
up to the proper mixing level. Two carrier than the desired mixer product. A second
crystals in the grid circuit of the 7360 os- 12BY7A (V4) serves as the driver stage for
cillator section permit sideband selection. the two parallel connected 6550 tetrode
The 9 -MHz SSB signal is coupled to the amplifier tubes. The 6550 tubes are chosen
#2 grid of a 6BA7 mixer (V,), and here it to provide the required power output
is combined with the 5.5- to 5.0 -MHz out- with the minimum degree of intermod-
put of a very stable vfo. The difference ulation distortion. The measured third- and
www.americanradiohistory.com
626 Low -Power Transmitters THE RADIO
MIC. VI-AUDIO
VS
.1 AMP. ANT
V2 Ve V3 V4
V7 -MOD.. 9MN[ FILTER I.F. MIX. MIX. -AMP. AMP.
FL -1
-CO - 12eV7 12BY7
PI-
NETWORK
o v9
VFO (D V,O
OSC. Ve
L
TUNE AMP.
Y, g
USB LSB
Y2
B+ REG. +e0OV. +LV
4 4 S3A
POWER
SUPPLY
REG. Y3 - - -Y7
V r1 X TALS
Figure 10
fifth -order distortion products of these tubes ond position (c -w) of the function switch
under the given operating parameters run provides c -w operation (with carrier in-
better than 30 decibels below one tone of jection) and the third position (PTT) places
a two -tone test signal. This degree of dis- the exciter in readiness for push -to -talk
tortion is f to 10 decibels less than other voice operation.
small tetrodes of the same power capability. The power supply makes use of a low -
Spurious products are reduced by incor- cost TV -type replacement transformer in
poration of a low -pass filter (L.,, L5, and a bridge circuit utilizing inexpensive silicon
associated capacitors) in the output circuit diode rectifiers. The center tap of the trans-
of the vfo stage to suppress the second and former is utilized for the necessary lower
higher order harmonics of the oscillator. In d -c voltages. Regulated voltage for the var-
addition, careful design of the various tuned ious oscillators is provided by an 0A2 regu-
circuits ensures that unwanted "birdies" are lator tube and a separate zener diode regu-
held to the very minimum. lated supply provides operating bias for the
The exciter is activated by two relays 6550 amplifier tubes. The complete sche-
which are energized by a push -to -talk matic of the exciter is shown in figure 11.
switch on the microphone. All tuning ad-
justments are accomplished with a single Exciter The general layout of the ex-
meter that measures the cathode current of Layout citer may be seen in the var-
the final amplifier, plus an auxiliary grid - ious photographs (figures 12 and 15). An
current meter. A three-position function 11" X 17" X 3" steel chassis is used for the
switch (S7A-B) enables the operator to foundation. The final amplifier assembly
zero -in an a chosen frequency without above the chassis is inclosed in a three -sided
placing an interfering signal on the air or inclosure measuring about 71/2" long by
without disabling his receiver. The zero 41/2" deep by 6" high. The sides are made
position of switch S; also disables the micro- from perforated aluminum sheet. A two -
phone circuit so that the exciter cannot be sided L- shaped aluminum dust cover com-
accidentally turned on in this mode. A sec- pletes the inclosure.
www.americanradiohistory.com
HANDBOOK 175 -Watt SSB Exciter 627
Under the chassis, the vfo components are from the amplifier section of a "surplus"
inclosed in a U- shaped shield made of light AN /ARC -S (SCR -274N) transmitter (any
aluminum sheet measuring about 4" square model). It has wide plate spacing, glass
and 2 3/4" high. The tuned circuits and bead insulation, and a smooth worm gear
6AU6 socket are mounted on a heavy %8" drive that lends itself very nicely to the
thick aluminum plate measuring 4" X 5" assembly of a simple home -made slide rule
mounted atop the chassis above the alumi- dial. The vfo coil is a section of miniduclor
num shield. The shield has 1/4" lips bent on stock securely cemented to a 1/4" thick
all sides to fasten it to the chassis and to square of plexiglas which is solidly
the side apron of the chassis. The balanced mounted on two ceramic pillars inside the
modulator and speech amplifier tubes are vfo shield compartment. The coil is placed
at the opposite end of the chassis, and their to one side of the chassis away from sources
under -chassis components are contained of heat.
within an L- shaped aluminum inclosure at
the panel. The design of the bandswitch as- The Main The assembly details of
sembly, placed at the center of the chassis, is Bandswitch Assembly a typical bandswitch
shown in figure 13A. and coil section are
The final amplifier tube sockets are shown in figure 13A. The coils and asso-
mounted on a sheet of perforated aluminum ciated padding capacitors are preassembled
which is bolted above a cutout in the chassis to the shield plate and wired before the
to permit good circulation of air past the plate is mounted to the chassis. Although
sockets and envelopes of the tubes. The leads the bandswitch shaft is positioned along
from the tap on the amplifier plate coil the centerline of the chassis, the switch
(L20) pass down through a slot cut in the wafers are placed slightly off center on the
chassis to the ceramic bandswitch segment partition (see illustration) to allow space
(S5A -B) which is driven by the main band- for the ganged APC capacitors (CAA -B) .
switch assembly. The switch segment is These capacitors are insulated from the par-
fastened to the back apron of the chassis tition (ground) and from each other, and
and is connected to the bandswitch by a are ganged with insulated flexible couplings
phenolic shaft extension. to the panel control. As each set of coils is
A small U- shaped aluminum shield is identical for any one band and is wired to
placed across the center of the 6550 sockets the switch section in an identical manner
to isolate the plate parasitic chokes and on each partition, a satisfactory degree of
leads from the nearby grid wiring. The tracking is achieved by the use of parallel
shield forms a compartment about an inch padding capacitors. These capacitors are
wide over the sockets, as can be seen in the mounted near the top lip of each shield par-
under -chassis photograph (figure 15). To tition. Isolating the two APC capacitors
minimize heat under the chassis the 40K, avoids a possible source of undesired inter -
20 -watt high -voltage bleeder resistors are stage coupling caused by circulating ground
mounted in a vertical position above the currents. The crystal oscillator coils are
chassis by means of a long bolt placed ver- mounted on the front side of the partition
tically in the rear corner of the final am- nearest the panel, with leads from the
plifier inclosure. The 6S0 -ohm, 25 -watt re- switch wafer left long enough to be at-
sistor in the B -plus voltage dropping net- tached to the tube and crystal socket be-
work is mounted in the same manner to the neath the bandswitch catacomb.
outside of the amplifier inclosure near the
hieh- voltage filter choke. The Dial The slide -rule dial is patterned
The most critical assembly of any good Assembly after a "short -wave style" dial
SSB exciter is the vfo which must have and is made from a flat plate of
rigid construction and use the best available aluminum mounted to triangular brackets
parts for the job. Silver mica padding ca- that fasten it to the chassis. The center of
pacitors, a ceramic tube socket and a pre- the plate is cut out leaving a rectangular
cision tuning capacitor ensure the stability hole, and a sheet of clear plerigla.c (spray -
of this unit. The tuning capacitor is taken painted white on the front) is fastened to
628 Low -Power Transmitters THE RADIO
Figure 11
www.americanradiohistory.com
HANDBOOK 175 -Watt SSB Exciter 629
C,-15-pf differential capacitor. Johnson 160- PC- Parasitic suppressor. 100 -ohm, 1 -watt re-
308 sistor, wound with 4 turns # 18 enam.
C,, Cr C,, C -12 -pf ceramic capacitor. Centra- RFC, -2.5 mH, 300 ma. National R -3000
lab 822 RFC., RFC, -2.5 mH, 100 ma. National R -100
C,, C -35 pf Hammarlund MAPC RFC , RFC., RFC, -750 H. National R -33
C,, C,.-50 pf Hammarlund APC RFC -I Miller 4652
mH, 100 ma.
C -150 -p1 tuning capacitor front ARC-S trans- S,, S,, S,,S,- Bandswitch. Four Centralab PA -2
mitter ceramic decks, each two -pole six-position,
C,A- B -35 -pf Hammarlund APC ganged with in- with Index Assembly PA -301
sulated coupling T,-10.7-MHz I-f transformer (tune to 9 MHz).
C -50 -pf ceramic capacitor. Centralab 827 Miller 1463
C,,,-30-pf ceramic capacitor. Centralab 822 T.- Bandpass transformer (3.5 -4.0 MHz). See
C -50 -pf APC padder for 3.5 MHz figure 138
C-
C,,-250-pf, 0.024 spacing. Bud 1859
365-pf per section. Miller 2112
CR,-1N34A
T.-6.3 volts, ampere. Wire in reverse
Stancor
1
the rear of the plate with 4 -40 bolts or the weight of the exciter (less the trans-
"pop" rivets. A slider moves along the former) is quite small.
smooth top edge of the plate and carries the After the exciter is assembled and wired,
pointer over the dial face. A dial cord is at- a voltage check should be made. The sche-
tached to a drum mounted on the large matic shows various voltages derived from
gear of the vfo tuning capacitor and drives the B -plus divider network. The no -load
the pointer via small dial pulleys taken from high voltage is approximately 800 volts,
an obsolete slide rule dial. As the gear moves dropping to about 750 volts with a 200 -ma
almost 360 degrees for 180 degree rotation load. The 12BY7A driver tube operates at
of the capacitor, a 21/2" diameter drum will 300 volts, as do the screens of the 6550
provide almost 7" of pointer travel. The dial tubes. The vfo and crystal oscillator receive
cord passes around the bottom of the dial regulated voltage from the 0A2, and all
face on idler pulleys placed at the corners, other stages are supplied from the 210 volt
then back to the vfo cord drum. tap of the supply network. Bias for the final
Wiring and Testing It prudent to wire the
is
amplifier is regulated at - 36 volts by
the Exciter
means of the zener diodes.
exciter in sections and to Oscillator Alignment -The vfo may be
get one section at a time adjusted to the 5.5- to 5.0 -MHz range by
in working order before proceeding to the
loosely coupling it to the station receiver and
next section. It is suggested that the speech
checking the tuning range against the re-
amplifier, 7360 stage, and 9 -MHz amplifier ceiver. Better yet, a BC -221 (LM) fre-
be wired and tested first. The next step
quency meter may be used. Alternatively,
would be to wire and test the vfo, first
mixer, and I 2BY7A amplifier. An auxiliary alignment may be checked by placing the
power supply can be utilized for these tests. bandswitch in the 80 -meter position and
Next, the crystal oscillator, driver, and final listening for the mixed signal in the 80-
amplifier stages are wired and tested, fol- meter range. The values of inductance and
lowed by completion of the power- supply capacitance given for the vfo tuned circuit
and control wiring. If the power transform- allow slightly more than 500 -kHz coverage
er is not mounted until the last step, it will and the optimum frequency placement on
be quite simple to move the unit about as the main tuning dial is done by adjusting
630 Low -Power Transmitters THE RADIO
Figure 12
the padding capacitor (C6) in the oscillator sideband is selected by sideband switch S,
circuit. and the trimmer capacitor across the crystal
The conversion oscillator is adjusted by in use is adjusted to place the oscillator
proper tuning of the plate circuit, checking frequency at the correct point on the slope
oscillation with a low -range voltmeter placed of the sideband filter. This is done by moni-
at the test point (TP,) in the grid circuit toring the signal from the filter and adjust-
of the 6C4 oscillator. The voltmeter reading ing the proper trimmer for natural sound
indicates the degree of crystal activity. Once of voice modulation.
adjusted to frequency, this circuit requires Modulator Alignment -The alignment is
no further attention. The upper or lower accomplished with the aid of a vacuum -tube
www.americanradiohistory.com
HANDBOOK 175 -Watt SSB Exciter 631
FIVE CERAMIC
TRIMMERS BOLTED on and advanced, a small reading will be
TO TOP LIP OF
evident on the v.t.v.m. The slug of the
PARTITION
balanced modulator coil (L1) is adjusted for
maximum reading. The probe is next moved
to the #2 grid (pin 7) of the 6BA7 mixer
(V,). Function switch S7 is placed in the
zero position to close the cathode circuit of
the 6BA6 amplifier stage and the slugs of
transformer T, are tuned to achieve a maxi-
mum indicated signal level of 6 to 10 volts.
To adjust carrier suppression, the carrier
TYPICAL BANDSWITCH AND COIL ASS'Y. switch (S2) is turned off (the v.t.v.m.
reading should drop considerably) and the
carrier null potentiometer (R2) is adjusted
for minimum meter reading. Differential
capacitor C, will affect the suppression, and
3 -4MHz
0
BANDPASS TR ANSrORMER
these two controls should be adjusted alter-
nately for the minimum possible meter read-
Figure 13 ing.
Bandhass Transformer Alignment -Band-
A- Sandswitch and coil assembly. Two assem-
blies are required. The partition is made of pass transformer T_ is aligned with the band -
light sheet aluminum with 14" lips bent on all switch placed in the 80 -meter position and
sides to mount it to the chassis and side sup- with the r -f probe placed at the plate (pin
ports. The rotor of capacitor C. is insulated
from ground and joined to the similar capaci- 9) of the 6BA7 mixer tube (Vs). The
tor on the opposite partition with an insulated exciter vfo is adjusted for a carrier fre-
coupler. The rotor is grounded to a common
ground (ground lug at the bottom of the quency of 3.S MHz (vfo frequency of Si
partition, below the bandswitch). The switch MHz) and carrier is inserted to obtain a
segment is mounted with 4 -40 bolts and metal
spacers to the partition with a similar size
meter reading. The primary capacitor (C4)
section of copper -clad phenolic circuit board
spaced 1'2" behind it. The ground connections
of the various coils are returned to this 2
board, which is grounded to the common
ground lug by a short flexible lead.
C- Bandpass transformer assembly. The two
MAPC 35 -pf capacitors are mounted on a
phenolic board 13/4" it 11/2 ". The double coil Is
wound on two 1/2" diameter lengths of poly-
styrene tubing slipped over a 1/4" diameter
polystyrene rod. Each section of tubing is O O
1/3" long, and the spacing betwn the coils is
3/e ". The primary winding (plate) is 55 turns,
k30 enam., bank -wound over 1" length. The
secondary winding (grid) is 45 turns, #30
nam., wound in similar fashion. A piece of
phenolic board the same size as the top
mounting board supports the coils by means
NV e+ NV
of 4 -40 bolts and spacers. Two long 6 -32
bolts join the sections and secure them to the PHENOLIC OR FIBERGLAS SHEET - CENTER CUT OUT.
shield can. The construction is similar to that RESISTORS MOUNTED ON BACK OF BOARD.
of an air -tuned I -f transformer. Adjustment of
the bandwidth is made by varying the spacing Figure 14
between the two windings.
DIODE RECTIFIER ASSEMBLY
voltmeter having an r -f probe. The probe is The eight silicon diodes are mounted on the
phenolic board by means of brass eyelets.
placed at the input terminal of the sideband The 220K equalizing resistors are mounted on
filter, and the differential capacitor (C1) the reverse side of the board in parallel with
each diode. The assembly is mounted below
in the 7360 circuit is set for balanced ca- the power transformer on the side apron of
pacitance. The balance potentiometer (R2) the chassis by means of 6 -32 bolts and spac-
is set near the center of its range. With the
rs. A cutout is made in the apron of the chas-
sis in line with the assembly for proper cool-
carrier -injection potentiometer (R:,) turned ing and ventilation.
632 Low -Power Transmitters THE RADIO
Figure 15
of the transformer is adjusted for maximum reading. When the dial is tuned across the
meter reading. The probe is now moved to 80 -meter band the voltmeter reading should
the grid (pin 2) of the 12BY7A amplifier / remain relatively constant, indicating proper
mixer (V) and the transmitter vfo is adjusted alignment of the bandpass circuit.
for a carrier frequency of 4.0 MHz (vfo Amplifier Alignment-The plate circuit
frequency of 5.0 MHz). With carrier in- of the 12BY7A amplifier /mixer is untuned
jection, the secondary capacitor (C,) of the for 80 -meter operation and the remaining
transformer is tuned for maximum meter alignment on this band is accomplished with
www.americanradiohistory.com
HANDBOOK 175 -Watt SSB Exciter 633
Li, L2 Balanced modulator coil. 12 bifilar turns tion and the vfo adjusted for carrier output
(24 in all) No. 24 e. 1/2" dia., 114" long at 7.5 MHz, the ganged capacitors are set at
on National XR -50 form. near -minimum capacitance. The 40 -meter
Link: 4 turns No. 24 e around tenter of L1.
_-
_- -.
Figure 17
THE HBT -200 SIDEBAND TRANSMITTER- EXCITER
The HBT -200 is a high -performance SSB exciter covering the 80- through 70 -meter
amateur
oands. The main tuning dial is at the upper left, calibrated every 100 kHz, with readout to
one kHz on the vernier window. Main panel controls are (I. to .): Upper row -calibrate level
R ), pilot lamp, main tuning control, mixer- driver tuning (C. , C. ), final
amplifier tuning (C ),
final amplifier bandswitch (S I. Center row -sideband- selector switch (S,), Carrier balance
potentiometer (Rd, function switch (S ), audio gain control (R,), main bandswitch (S ), drive
level potentiometer (R,), meter switch (S ), final amplifier loading (C,.). Bottom row -key
jack, VOX hold (R ), antitrip (R:), microphone ack, p.a. bias control (R,), r -f sensitivity (R )
proper position for the band in use, and serted carrier is removed, the plate current
carrier- injection switch S2 is turned off. will drop back to the original idling level.
When the push -to -talk switch on the micro- The final step is to determine the ratio of
phone is closed, the plate meter will indicate grid drive to plate current loading. If this
an idling current of about 70 ma. The car- ratio is improperly set, the exciter will "flat-
rier control is turned on and advanced top" before full output level is reached
slightly, and grid tuning is peaked for a (excessive -drive, light- loading condition),
rise in plate current. The plate tank capaci- or transmitter output will be low (insuffi-
tor is tuned for current dip and loading cient- drive, heavy- loading condition). Ex-
adjustments are made using regular pi -net- citation is set by means of the r-f level
work procedure. Maximum loaded plate cur- control (R1) which may be calibrated for
rent with full carrier insertion is 200 to 240 each band. To do this, it is necessary to place
ma and this value is reached by advancing a O to 1 d -c milliammeter between test
the carrier control, together with an in- points 2 and 3 in the grid circuit of the
crease in amplifier loading. When the in- final amplifier stage. No grid current will be
www.americanradiohistory.com
HANDBOOK HBT -200 SSB Transmitter 635
V2 VT ve VT Vv VII
P BAL. FIRST SECOND THIRD
CARRIER OSC MOD 9MHz FILTER MIXER MIXER MIXER 134X5-8 I15
en- lse
EL I 11.0-11 65
210-2165
28 0-26 65
USB .2865.2P50
LSe 51 R2
Cw CARRIER SAL
VI
MIC. SPEECH
RTg PI -
AL' R An NETWORR
Figure 18
BLOCK DIAGRAM OF HBT -200 TRANSMITTER- EXCITER
drawn until the peak of the r -f driving The tune -up procedure for c.w. is the
signal exceeds the bias level (a nominal same as above except that carrier is inserted
-36 volts). Maximum power output will and the audio gain control turned down.
be obtained in class -AB, Service when the A -m operation is possible by inserting suffi-
amplifier tubes are driven just to the point cient carrier for a plate current of about
of grid current and with plate loading then 100 ma and advancing the gain control
adjusted for maximum power output at a while monitoring the ratio of grid drive to
plate current of 200 to 240 ma. Using car- antenna loading with an oscilloscope to
rier insertion, then, the point of grid current achieve maximum modulation level without
is monitored on the temporary test meter and distortion.
plate loading is adjusted for the proper
plate current. The setting of the r -f level
27 -3 The Deluxe HBT -200
control is logged and carrier is removed
and the transmitter modulated by voice. SSB Transmitter- Exciter
The audio gain control is advanced until,
with the r -f level control untouched, the The Deluxe HBT -200 SSB transmitter -
grid- circuit test meter just indicates a exciter is a companion unit to the HBR
flicker of grid current: one scale division or receiver described in an earlier chapter. De-
less. The setting of the audio gain control is signed for high quality SSB and c -w per-
then logged. This calibration procedure formance, the HBT -200 is capable of 200
should be run on each amateur band and the watts PEP input on amateur bands between
settings of the controls noted for future use. 80 and 10 meters (upper or lower sideband)
If desired, a small 1 -inch diameter milliam- and features automatic load control (alc) ;
meter may be panel mounted on the unit VOX or push -to -talk operation; all -band
for a continual check of the amplifier peak coverage and a nuvistor -type, high -C vfo.
signal driving point. It should be noted The transmitter has a high order of fre-
that under maximum peak voice conditions, quency stability and a very minimum of
the plate meter will swing to about 100 spurious "birdies" and image signals.
ma. Operation may be monitored with an The HBT -200 transmitter (figure 17) is
oscilloscope to check "flat -topping." designed around easily obtainable compo-
636 Low -Power Transmitters THE RADIO
MIXER TUNING
7
,oeo VII TeA
o 7360 TeB eO
TOC-0 3N
lo
TOE 5aE
ro 001. 53F TOE T60-0
.
S 3C ^ y I 1 2 N
r V12
oo, 2.711
39
I La GPO ION 1211
Op
To Sao
9
JJ C11
CIO C13
C 2
V3 Ti Ve
.-1-1II DIl c 7360 68A6 TOT]
a O8N
3 1
7 Ya -L2N II Ce
6CW4 3 ?9
SI
C1
= .01 1211
.1..-
C IW 080 N
1 IM
+75 V. REG. (Va,V6, Vto,Via
+
10
z5
1.300 V.
FIGURE 19A
www.americanradiohistory.com
HANDBOOK HBT -200 SSB Transmitter 637
V13 Vt4-6146B V15-6146B
6CL6
-
ANT
J2
Pct
Pcz
EPc4 E Pc
So3i, RFC 1
7 2.!-
I4,eI,,.
,0 SM
MN=
5
1L]E
,;_
C1e
is
PC3 }{ +eoo v Re
R-F SENSITIVITY
1e134
01
4-300 V.
Re
-
2011
DRIVE LEVEL
I=
RFC 5
450
SK
RF
100
SSA S5
1
I
4
W
2.51N 170
leo IIH Sa
.01 G
-C 1/C
ALC GAIN o011_
RFC 1011
ALC 2. SW,
T
060 ii
11 N647 Y1 C I9 P.A. BIAS Se OPERATE ..TUNE
1 1-,sK
100
68 K/111/
Re
as K
o.I_ ISK ]K
1w - 2w zw
R7 - -150V.
33K -
ALC BIAS TO VOX
RELAY
+300 V. REG.
V? V9
T3 7360 T4 7380
01 /C .001 001
C
68K
=1
1 T24
211
47011 3
dl-ICS d
1
11 /rSI
3 e
6 se
710
Sac. 0
J
001 J
C
I01 I
?;M
01
+160v. REG
i AI
TO VOX
1 DI
T01
+ 180 REG.
OTg
TD.II.Qj 1w
`2. 7
11
K
rrra
10 IS
Y5e7e9 0 L3B pCoe :' 1L3E
2K LIC, D
+75v REG.
I LIA
.01
C
12K
r
01
+75v. REG.
4-300v.
Figure 19B
type. J. W. Miller 2113 PC, -4 turns # 14 enam., Ye" diam, around 47-
C -1S -pf Hammarlund HF -15X ohm, 2 -watt composition resistor
M -0 -1 d-c milliammeter. 47 ohms resistance. Simp- Cabinet-Wyco CR -7725.
son 1212A
638 Low -Power Transmitters THE RADIO
2R C-W KEYER
w VOX RELAY
TO V90 o
`
RYIA Ve
I RY1
PL-1 1
iRVI
PTr
Sze
TUNE
T
cw I
1
--
B.SV.
TOAILOD 1eR
KEYD
TO PA BIAS
CONTROL R
(VON) IM
FILS 21 .Dt RrID
= R 1.6 KV
.111
0
O O O O O b d O I
I
GNO
2 3 3
B.Sv. -150y. 300V. 300V. 600V.
C A 20MA ISO MA REG. 250 MA.
"---v-'
7
EXT. CONTROL
10
ANTI-
VOA
I11CVR-SPKR.1
TERMINAL CONNECTIONS
Figure 19C
nents, including the prefabricated exciter hand key, bug, or electronic keyer. A sep-
tuned circuits. Nuvistor tubes are used in arate power supply is used with the trans-
the local -oscillator stages and the filament mitter to reduce the weight of the unit
voltage of the vfo is regulated to achieve and to keep heat -producing components out
maximum frequency stability. A high qual- of the transmitter enclosure. Transmitter
ity 9 -MHz sideband filter having superior alignment is easily accomplished without the
shape factor ensures a clean, crisp SSB sig- need of complicated test equipment.
nal.
The power output of the transmitter is Transmitter block diagram of the HBT-
A
better than 100 watts PEP on all bands and Circuitry 200 transmitter is shown in
is sufficient to drive most of the popular figure 18 and the schematic is
"grounded- grid" linear amplifier tubes, al- given in figures 19A, B, and C. The filter
through the HBT -200 will give a good ac- system of sideband generation is used, and
count of itself on the air when operated as -is. the transmitter employs bandswitching of
The efficient alc circuit allows a degree of the various r -f stages in 650 kHz segments
speech compression that imparts a "punch" between 80 and 10 meters. Sufficient over-
to the signal and ensures that voice modula- lap may exist at the band ends for auxiliary
tion is held at a high level at all times. activities such as MARS.
In addition to the SSB features, the HBT- The R -F Section -A
7199 is used as a
200 incorporates semiautomatic break-in key- two -stage speech amplifier (V,) to drive one
ing for c -w operation, utilizing either a deflection plate of a 7360 balanced modu-
www.americanradiohistory.com
HANDBOOK HBT -200 SSB Transmitter 639
Figure 20
REAR VIEW OF HBT -200 TRANSMITTER CHASSIS
Symmetrical arrangement of components prevents crowded appearance of transmitter chassis.
The nuvistor vfo is in the cast aluminum box at the center of the chassis, with its output
coupled via a coaxial line to the third mixer stage. Aluminum shield separates 6146B am-
plifier plate tank circuit from low -level stages. On rear apron of chassis are (I. to r.): An-
tenna receptacle (J 1, rear shaft of bandswitch S surrounded by hf oscillator coils .L A -E .
Counterclockwise, coils are: 80- 40- 20 -1S -10 meters. To the right of the coil assembly are the
power receptacles and speaker jack. At the extreme right are the second mixer balance
control (R and the first mixer balance control (R ).
Atop the chassis, in the foreground are the 1S -MHz crystal Y with the 6CW4 nuvistor
tube behind it; the 7360 first mixer (V ) to the left, along with transformer T Behind the
.
mixer stage are the 6BA6 i -f tube and the crystal filter. To the right, rear near the panel:
are the three carrier -frequency crystals, the 7360 balanced modulator, the 6CW4 carrier oscil-
lator tube and the slug adjustment of coil L,.
Immediately in front of the vfo box are the 7199 audio tube, the 12AT7 relay tube, and
the 12AT7 d -c amplifier. In front of these tubes are the 7360 third mixer (V J and transformers
(T A, B, C). Towards the near edge of the chassis are the 7360 second mixer (V ), the 6CW4
mixing oscillator (V ), and the hf crystals (Y -Y. ). To the right are the two voltage -regulator
tubes.
Next to the shield partition at the left are four r -f transformers (T A, B, C, D,), and im-
mediately behind them transformers T D and E. Buffer plate coils L A -E are chassis- mounted
adjacent to the 6CL6 socket. Magnetic shields are placed on the 7360 tubes (Millen 80801 -D3'.
lator (V ). The carrier signal is generated at of upper or lower sideband, or c.w. The d -c
9 MHz by a 6CW4 nuvistor oscillator (Ve). voltage applied to the second deflection
Three crystals are provided to permit choice plate of the 7360 balanced modulator may
640 Low -Power Transmitters THE RADIO
be adjusted by carrier- balance potentiometer circuit layout does not require it; however
R2 to null the unwanted sideband and car- the stage should be checked for stability as
rier in the plate circuit. Carrier injection for changes in wiring or layout may require that
c.w. or for calibration is accomplished by neutralization be added. If so, the neutral-
unbalancing this circuit by means of switch izing technique employed in the final am-
S2A and level -control potentiometer R,_. plifier may be applied to the driver stage.
A 9 -MHz crystal lattice filter (FL,) hav- The final linear amplifier uses a pair of
ing excellent skirt selectivity is transformer - 6146B tetrodes connected in parallel and
coupled between the balanced modulator and operated in class -AB, mode. The plate cir-
the 6BA6 intermediate- frequency amplifier cuit is a conventional pi- network using an
(V6). Alc voltage is applied to the grid of r-f voltmeter to measure relative r -f power
the 6BA6 to provide proper control and to output. Extra capacitance is added to the
reduce flat- topping of the SSB signal under output section of the pi- network on the 80-
overdrive conditions. meter band to achieve proper loading into
A second 7360 is used as the first mixer a 50- to 70 -ohm antenna system.
(V7) and a 6CW4 nuvistor serves as a Grid, screen, and cathode currents of the
15.3 -MHz crystal oscillator (V8), convert- final amplifier are monitored by multimeter
ing the SSB signal to 6.3 MHz for further M, and a tune -operate switch (S6) reduces
mixing into the various amateur bands. (If screen voltage to protect the final amplifier
a 6.3 -MHz crystal filter is available, the tubes during loading and tuning adjust-
V7 and V8 stages may be omitted and the ments.
i -f amplifier stage modified for 6.3 MHz A fraction of the r -f plate voltage is
operation). This particular intermediate fre- selected by a capacitance bridge (C,., plus a
quency is chosen since the creation of spur- 5 -pf ceramic series capacitor) and is recti-
ious mixer products that fall in the amateur fied and filtered to obtain an alc voltage,
bands is less severe than with a 9 -MHz which is fed via filter RFC, to the grid of
intermediate frequency. the 6BA6 i -f amplifier.
The basic 6.3 -MHz SSB signal is fed to The VOX and Control Section-Voice-
a second mixer stage utilizing a 7360 (Va). control (VOX) voltage is derived from the
The 6CW4 nuvistor mixing oscillator (V,,,) speech amplifier and further amplified in a
makes use of appropriate crystals to provide separate VOX amplifier (V4A). The VOX
six channels of SSB which may be mixed and antiVOX signal (from the receiver) are
into the amateur bands with a minimum of rectified, filtered, and passed through a d -c
"birdies" and beats. The six intermediate amplifier (V5A) to the VOX relay circuit
SSB channels are combined in a third 7360 (V,B) . The VOX relay is a 4pdt device,
mixer stage (V11) with the signal from two circuits of which are used to energize
the master variable oscillator which employs the transmit pilot light (PL,) and to actuate
a 7587 nuvistor (V12) and tunes the 6.0- to the auxiliary circuits (antenna relay, etc.).
6.65 -MHz region. This provides an SSB tun- A third relay circuit provides standby block-
ing range of 650 kHz on each position of ing bias for the second mixer stage (V )
the bandswitch. Six tunable bands are thus and the fourth circuit actuates the c -w
generated, covering the various amateur keyer (V5B) and removes blocking bias
bands with generous overlaps up to 10 from the final amplifier stage.
meters. On this band a total of 1.3 MHz is The sequence of c -w operation is as fol-
covered in two ranges (either 28.0 to 29.3 lows: Closing the transmitter key grounds
MHz or 28.4 to 29.7 MHz may be utilized the grid of keyer tube (V.,B) which actu-
by proper choice of conversion crystals Y ates the control relay tube when the emis-
and Y10). sion switch S: is placed in the c -w position.
Satisfactory image suppression is achieved The relay is held closed by virtue of an
by the use of double- tuned, ganged circuits adjustable RC network (VOX -hold) in the
between the last mixer and driver stage. A bias leg (R11) for normal c -w speeds. The
6CL6 is used as the driver (V,3) with the maximum time constant is about 1.5 second.
drive level controlled by screen potentiome- Longer delays may be achieved by increasing
ter R,;. The 6CL6 is not neutralized since the the value of the 0.22-pfd capacitor in the
www.americanradiohistory.com
HANDBOOK HBT -200 SSB Transmitter 641
ait
Figure 22
UNDER- CHASSIS VIEW OF HBT -200 TRANSMITTER- EXCITER
The transmitter is wired and tested in stages for simplicity and reliability of operation. Inter -
stage shields are employed between circuits of different operating frequencies. Shields are
slotted so that power leads may pass through the shields, yet permit the shields to be taken
out for wiring and easy access to coils and components. Bandswitch shields are also slotted
so that they may be installed after bandswitch is wired to the second mixer. Crystal oscil-
lator coils L A -E are mounted to the rear apron of the chassis and are placed in position after
bandswitch has been mounted and wired.
Power amplifier loading capacitor (C, is at upper left, with its shield and below it are
(
the 61468 sockets. A common ground point is used for each socket. The shaft of p.a. neutral-
izing capacitor C projects through the chassis between the sockets. The plate coils of the
6CL6 stage (L A -E) are grouped around the front section of the main bandswitch, with the
shielded microphone jack mounted over the switch section. The third mixer balance potenti-
ometer (R) is mounted to the rear bandswitch partition.
At the right of the chassis is the balanced modulator and carrier -oscillator compartment,
with the crystal padding capacitors in view, as well as the differential capacitor (C ). Con-
trol relay RY, is mounted to the center shield partition to the left.
to drill the holes on the front chassis apron. 2 2" X 7 %2" X 1/4" Fiberglas board with
!
The control nuts arc used to space the panel I -inch spacers. The board, in turn, is
/2
www.americanradiohistory.com
HANDBOOK HBT -200 SSB Transmitter 643
Figure 23
COIL AND T RANSFORMER DATA
L 9 MHz oscillator coil. Approx. 24H. J. W. and 20 -meter portion: S turns of 3/16"
Miller 42A-226 -CB1. diam. tubing, 1- 11/16" inside diameter,
L. -15.3 MHz oscillator coil. Approx. 1.14 H. 11/4" long. Three turns from plate end to
J. W. Miller 42A -106 -C111. 1S -meter tap. Third section is 40- and 80-
Li-Plate of 6CW4 hf oscillator (V, ). meter portion: 10 turns #I2 wire, 2" In-
Band Ou. Freq. J. W. Miller C (pf) side diameter, six turns per inch. Four
(MHz) Coil turns for 40 -meter portion. (Illumitronix
80 16.45 42A- 106 -C81 75 "Pi -Dux" 1608 -D6 cut to size).
40 19.85 42A- 106 -CBI 62 T, -9 -MHz 558 transformer, bifilar wound. J. W.
20 26.95 42A- 476 -CBI Miller 1739.
15 33.95 42A- 336 -C81 T -9 -MHz 558 transformer. J. W. Miller 1741.
10 41.45 42A- 226 -CBI T -9 -MHz transformer (10.7 MHz i -f trans-
10 41.95 42A- 226 -C81 former. Pad windings with 24 -pf silver
adjusted to approximately the following
(L, mica capacitors). J. W. Miller 1451.
inductance: 80 meters, 1.1 H; 40 meters, T -6.3 -MHz transformer. J. W. Miller 1800-1.
0.98 H; 20 meters, 5H; 15 meters, 3.1 T- Second Mixer transformer for V, stage.
H; 10 meters, 2.1 H). Band Frequency (MHz) J. W. Miller No.
Note: L A (80- meter) and LB (40- meter) coils
80 10.15 1800 -6
are supplied 8 -plus through a 12K resistor.
Others are supplied 8 -plus directly. This is 40 13.55 1800 -5
20 20.65 1800 -4
done to equalize the difle in output
15 27.65 1800 -3
between the fundamental and overtone
10 35.15 1800 -2
mode crystals.
L -Oscillator coil. 103/a turns #22 enam. wound
10 35.65 1800 -2
to a length of 15/16 inch on 1/4-inch diam. 7,-Third Mixer transformer for V stage.
ceramic form. National XR -71 with slug re- Band Frequency (MHz) J. W. Miller No.
moved. Approx 1.5 H with Q of ISO. 80 3.5 -4.0 1800 -11
L -Plate coil of 6CL6 stage (V ). 40 7.0 -7.3 1800 -10
Band Freq. (MHz) J. W. Miller Coil uN 20 14.0 -14.35 1800 -9
15 21.0 -21.45 1800 -8
80 3.5 -4.0 42A -155 -CB1 15.0
10 28.4-29.05 1800 -7
40 7.0 -7.3 42A- 686 -C81 6.0 29.05 -29.7 1800 -7
10
20 14.0 -14.35 42A -226 -C81 2.2
15 21.0 -21.45 42A -106 -C8l 1.2 Crystals-Y,, Y,-supplied with filter.
10 28.4 -29.05 42A -106 -C81 0.7 7,-9.000 MHz. McCoy Type M -1.
10 29.05 -29.7 42A -106 -C81 0.7 Y -15.30 MHz.
(remove 1 turn) Y;- 16.415 MHz.
L -Plate coil of 61468 amplifier stage. Y,- 19.681 MHz.
Plate coil is made in three sections. First Y.-26.950 MHz.
section is 10 -meter portion: 6 turns of V, -33.95 MHz.
3/16" diam. copper tubing, %" inside di- V,-41.45 MHz.
ameter, IN" long. Second section is 15- Y -41.95 or 49.2 MHz.
coupled to the dial by means of a metal wired, tested, and the shield placed around
coupler. The vfo assembly may be wired these stages. The B+ distribution circuits,
and tested as a unit before it is placed on control and VOX assemblies and audio
the transmitter chassis. A regulated 7.S -volt stages are wired next, tested, and the audio
filament supply is used with the vfo to shield positioned. The mixer and driver
provide 6.3 volts at the tube socket through stage are wired next, and the vfo installed.
the d -c resistance of the filament chokes The last step is to wire the final amplifier.
(RFC,, RFC_). Placement of under -chassis components is
Transmitter
"tight," especially in the vicinity of the
A unit of this complexity
bandswitch and use of miniature components
Wiring should be wired and tested in and terminal boards mounted to the inter -
stages so as to simplify as- stage shields is recommended.
sembly. It is suggested that all filament Buffer tuning capacitor (C,:,) and mixer
wiring be completed and the carrier oscil-
lator and balanced modulator be wired first,
tuning capacitor (C)
are mounted to an
tested, and the shield placed around this insulating plate as the rotor of capacitor C,-.
is at B+ potential and is coupled to the
assembly. The shield (and others like it)
should be slotted so that it may be installed ganged capacitor and to the dial with in-
and replaced without displacing the power sulated couplings. Capacitor section C,-,C
leads. Next, the i -f amplifier, first mixer has half the rotor plates removed to provide
and 15-MHz high- frequency oscillator are proper tracking, as outlined in the parts list.
644 Low -Power Transmitters THE RADIO
Power wiring is done with #18 shielded Carrier Oscillator- Modulator Adjustment
wire (filament leads) and #22 shielded wire -The first step is to adjust the carrier
for other interconnections. oscillator frequency with the r -f probe of
Transmitter Bandswitch Assembly-The the v.t.v.m. placed on pin 3 of balanced
main bandswitch is made up of seven cer- modulator V:,. The plate coil (L1) of the
amic switch decks and may be seen in the carrier oscillator is tuned to the high -fre-
under -chassis photograph. The deck nearest quency side of oscillation, using the 8998.f-
the panel is the plate circuit switch for the kHz sideband crystal. With the BC -221
driver stage (S3I) and is mounted to the frequency meter coupled to the modulator,
panel index assembly. The remaining sec- the crystal trimmer capacitors (C1, C2, and
tions are placed towards the rear of the C:,) are set to provide the proper injection
chassis and are driven from the front section frequencies of 8998.3, 9000.0, and 9001.E
via a metal shaft coupler. The section nearest kHz. If necessary, the oscillator inductor is
the index assembly is S:,H (secondary circuit readjusted for reliable operation with each
of transformer TO and immediately behind carrier crystal to provide a peak -to -peak r -f
this section is an integral mounted shield measurement of about f to 8 volts. (Some
plate. Behind the plate are sections S,F and v.t.v.m. scales are calibrated in rms, and
S,G (one deck for each section). A chassis may be converted to a peak reading by
shield cuts across the switch behind these multiplying the reading by 1.41) .
sections, and to the rear of the shield are The r -f probe is now moved to the sec-
sections S3C, D, and E (made up of two ondary of transformer T1 (pin C) and the
decks). Another integral mounted shield c -w carrier crystal (9000 kHz) selected.
plate separates these sections from the rear Carrier balance control R2 is turned fully
switch segment (S:,A and B). Leads should clockwise and the slugs of transformer T,
be attached to the three rear switch seg- are adjusted for maximum voltmeter indi-
ments before the switch is placed in posi- cation. Check the null action of control R_
tion, and the oscillator coils (L,A, B, C, and at this time. Adjust R_ and C, to obtain
D) mounted to the rear apron after the best carrier null. The microphone is now
switch is placed on the chassis. The leads connected, the carrier unbalanced and the
from the switch are trimmed and wired r -f signal at the secondary of transformer
to the proper coils and terminals. Wiring is T, monitored for proper a -m voice modu-
done with tinned, bare wire, cut to length lation as heard in a nearby receiver. Leav-
and covered with insulated tubing. ing the carrier unbalanced, the r -f probe
is moved to the plate circuit of the 6BA6
Transmitter The HBT -200 transmitter may i -f amplifier and the slugs of transformer
Alignment be aligned with the aid of a T_ are adjusted for maximum voltmeter
BC -221 (LM) frequency me- indication. Now connect the probe to pin
ter, a vacuum -tube voltmeter with r -f 9 of the first mixer tube (V7) and adjust
probe, a dummy load, and a general cover- transformer T3 for maximum indication.
age receiver. An audio oscillator, grid -dip When the carrier is now nulled out by the
meter, and oscilloscope are convenient test carrier balance control, a clean 9 -MHz SSB
items but not indispensable. signal will be observed when voice modula-
After checking transmitter wiring and tion is used. (Adjustment of transformers
assembly, filament voltage and all d -c volt- T, and T, for optimum filter passband will
ages (with the exception of 6146B screen be done later) .
and plate voltages) are applied to the trans- Mixer and 1 -F Circuit Alignment -The r-f
mitter. Drive level control R,; is set for probe is placed on pin 3 of the first mixer
zero 6CL6 screen voltage, and ale bias con- (V7). Using the receiver as a monitor, plate
trol R7 is set for maximum negative alc coil L_ of the 15.3-MHz crystal oscillator
bias. Function switch S., is placed in the (V.) is adjusted to the high- frequency side
tune position. In this position, VOX relay of resonance for a peak indication of about
RY, will be energized and normal operating 8 to 10 volts on the v.t.v.m. The probe is
bias will be applied to the second mixer and now moved to pin 9 of the second balanced
amplifier stages. mixer (V.,) and the slugs of interstage
www.americanradiohistory.com
HANDBOOK HBT -200 SSB Transmitter 645
transformer T, adjusted for maximum sig- 90;; meshed to about 60 %f meshed, and on
nal with capacitor C, set at mid -capacitance. 40 meters the capacitor range goes from
Carrier injection is used, and balancing out 50'; to 60% meshed. On the higher bands,
the carrier with control R_ will provide a the capacitor only makes a small movement,
6.3 -MHz SSB signal. but still tunes rather sharply.
The next step is to adjust the 6CW4 The meter switch (Sa) is now set to
high- frequency conversion oscillator stage measure final amplifier grid current and
(V,,,). The slug of the plate coil for each the slugs of coils L,A through L,E are ad-
bandswitch position (L:,A through L,E) is justed for maximum indicated grid drive on
adjusted to the high- frequency side of each band. Drive level is controlled by the
resonance for a peak measurement of 5 to 6CL6 screen potentiometer (R). Use the
10 volts at pin 3 of mixer tube V,,. The p.a. bias potentiometer along with the drive
mixer (V )
probe is now moved to pin 9 of the third
and the primary and second-
ary slugs of transformers LA through LE
level control to adjust the sensitivity of the
meter. Drive is generous and will easily pin
the meter, even at the near -zero setting of
are adjusted for maximum indicated signal the drive control.
on each band, using carrier injection. Mixer Balance Adjustment-The com-
It may not be possible to completely reso- munication receiver is now tuned to H.3
nate the secondary windings of the 10 -, 15 -, MHz and loosely coupled to the secondary
or 20 -meter transformers because of the of interstage transformer T, with a shielded
added probe capacitance, but final alignment lead. The first mixer balance controls (R3
can be accomplished later. All preceding and C,) are adjusted for minimum S-meter
adjustments should now be repeaked for reading. Following the same technique, the
maximum signal. receiver is tuned to 26.95 MHz and loosely
The BC -221 frequency meter is now cou- coupled to pin 9 of the second mixer stage
pled loosely to the vfo and the main tuning (V). The second mixer balance control
capacitor (CO set near zero capacitance. (R,) is adjusted for minimum indicated
Padding capacitor C is adjusted to place signal, with the bandswitch placed in the
the vfo at 6.65 MHz. The tuning range 20 -meter position. In like manner, the re-
should now run from 6.65 MHz to 6.0 ceiver is loosely coupled to pin 2 of the
MHz with some overlap. After temperature 6CL6 amplifier, the exciter adjusted to 40
stability has been achieved, the differential meters, and the third mixer balance control
compensating capacitor C,,, is adjusted for (R,) is adjusted for minimum carrier as
minimum long -term drift. Capacitor C monitored at the vfo frequency (6.0 to
can be readjusted to compensate for adjust-
ments to C,,, as far as frequency is con-
cerned.
6.65 MHz).
Alignment of the Crystal Filter Circuit
The secondary of transformer T, and the
-
The r -f probe is now placed at pin 3 of primary of transformer T_ should now be
the third mixer (V) and the output level adjusted to provide the proper impedance
of the vfo stage is observed. It should be match to the crystal filter consistent with
about 5 to 8 volts, peak, as the vfo is tuned good passband characteristic. If this circuit
across its range. The probe is now moved is improperly adjusted, the passband will be
to the input grid (pin 2) of the 6CL6 unsymmetrical, or uneven, with an un-
driver stage (V13) and the slugs of inter - natural emphasis given to certain audio fre-
stage transformers LA through T,;E are quencies. A simple adjustment technique is
adjusted for maximum indicated signal on to strive for the best sounding signal on
each band, with ganged tuning capacitor both upper- and lower -sideband positions.
C -C,:, set approximately as follows: 80 This may involve shifting the USB and
meters (3.75 MHz), 3/4 meshed; 40 meters,
LSB carrier frequencies slightly. This tech-
'/2 meshed; 20 meters, 1/4 meshed; 10 and
15 meters, %8 meshed. The initial setting
nique will suffice unless the transformers
of the capacitor gang and the slug adjust- are badly out of alignment (they are factory
ments are somewhat self -compensating. On tuned). If it is desired to check adjustment,
80 meters, the capacitor range goes from a more formal and complex approach makes
646 Low -Power Transmitters THE RADIO
Figure 24
use of the BC -221 frequency meter and the mate amplitude, and corresponding to the
V. t. v.m. peaks shown in the filter passband curve
Remove one crystal from the carrier oscil- published in the data sheet supplied with
lator and inject the output of the BC -221 the filter. These peaks occur at approximate-
into the grid circuit of the carrier oscillator ly 8999.2 kHz, 9000.0 kHz, and 9000.8
(V2). The BC-221 will then drive the ex- kHz for the filter specified. The dips in the
citer in the place of the crystal. Connect the filter passband should be no more than one
v.t.v.m. probe to the 6CL6 plate circuit. A decibel below the peaks, corresponding to a
definite peak, or series of peaks, will be difference of 0.89 on the v.t.v.m. For ex-
found in the reading of the v.t.v.m. as the ample, if the peaks are set at 10 volts on
BC -221 is tuned through the passband of the meter (by adjustment of the drive con-
the filter. The goal is to adjust the second- trol), the dips, or valley should be approxi-
ary of transformer T, and the primary of
transformer T2 to obtain three peaks in the
filter passband, all having the same approxi-
mately 8.9 volts.
Functional Test and Neutralization
Once the operation of the low level r -f and
-
www.americanradiohistory.com
HANDBOOK HBT -200 SSB Transmitter 647
Figure 25
HBT -200 TRANSMITTER AND HBR RECEIVER MAKE A MATCHED PAIR AT K6OPZ
audio stages is satisfactory, operation of the pacitance. The r-f probe is connected to the
function switch should be checked for prop- plate cap of one 6146B and neutralizing
er PTT, VOX, CAL, TUNE, and CW capacitor C,, is adjusted for minimum volt-
operation. Remember that in normal stand- meter indication.
by, the second mixer (V9) is biased to cutoff. Amplifier Stage Adjustment -Plate and
In the CAL position the second mixer is screen voltage are applied to the amplifier,
operative but the 6146B stage is biased to and the operate -tune switch (S,;) is set to the
-150 volts. PTT, VOX, TUNE, and CW tune position (low screen voltage). Reso-
are identical as far as transmitter operation nance is established and drive removed
is concerned except that alc is grounded from the stage. Switch S,; is set to the oper-
out in the CW position. ate position and bias potentiometer R9 is
The exciter stages are now adjusted for adjusted for 40 ma static plate current. Car-
10 -meter operation and grid drive to the rier is now inserted and the drive level ad-
final amplifier is adjusted to provide half - vanced to provide an indication of in-
scale meter reading (screen and plate volt- creased amplifier plate current. Resonance
age removed). Resonate the amplifier plate - is established, with the transmitter operating
tuning capacitor for a dip in grid current, into a dummy load, and the transmitter is
with the loading capacitor set at half ca- tuned and loaded in the usual manner to
648 Low -Power Transmitters THE RADIO
www.americanradiohistory.com
HANDBOOK Transistor Keyer 649
3.9K SIK
I.
TO TRANSMITTER
200
Q1-2N404S -- Q2 Q4 Qe 1 N 2858
.01
39K 66 2N404 S0 2N404 2N4045
I68K
Q3
2N 1302 1N34A 1N34
OR GATE
2N1302 Q6
10K 10K
Q10 2N404,s Q11
TONE OSCILLATOR
1N2858
1N2858 1/w N2861
10K
0
123,40
z
53
AUTO. F
V16130-KEYER
1SK
R SEMI A VTO
KEY
2000
15 +
g+
1YHAND +8 V. 117 V..60 CPS
Figure 27
Q7 CURRENT
sistor- capacitor network in the multivibra-
tor base-collector circuit. The more negative RELAY CURRENT
the voltage on the movable arm of the
DASH FORMATION
weight potentiometer, the faster the timing (PADDLE CONNECTING DASH CONTACT OF V /DRO -NEVER ro END.)
capacitors will charge to the conducting Q7 CURRENT
potential of the multivibrator transistor
(not conducting at that instant) . The max- QE CURRENT
www.americanradiohistory.com
HANDBOOK Transistor Keyer 651
Figure 30
Closeup of circuit board of transistorized keyer
Figure 29 showing dot multivibrator circuit (left) and
dash flip -flop circuit (right). In the center are
SIDE VIEW OF KEYER the three transistors comprising the OR gate
and relay control. The board immediately in
Side oblique view of transistorized keyer back includes circuitry for the side -tone oscil-
showing location of speaker, transformers, lator and biasing network.
and filter capacitors. Six- terminal strip con-
nects to contacts of double -pole, double -throw
relay (RY). Phenolic terminal boards are sup-
ported from front panel by long bolts and flop and its clamp transistor and the OR
spacers. Electrolytic capacitors are mounted gate. The front board contains the tone os-
to tie point strips bolted to surface of
inclosure. Towards far side of the phenolic cillator and voltage bridge for the dot and
board are mounted the 1.0 -pf capacitors of the dash clamp circuit.
dot multivibrator. Transistors Q, Q , and Q. The power supply, keying relay, speaker,
are positioned between the capacitors. Tran-
sistors Q Q , and Q are near center of the
, output transformer, and potentiometers are
board, and transistors Q , Q , and Q are in all mounted to the case. The speaker cone is
the foreground.
covered with a small square of perforated
aluminum sheet to protect it from damage.
The keyer may be operated as a semi- All relay contacts are brought out to the
automatic key (bug) by placing switch S, rear of the keyer to the six terminal strip.
in the semiauto position. Although the dots Two circuits may be keyed simultaneously,
are produced automatically, the automatic and the relay also provides normally closed
keying circuits are bypassed when the pad- or normally open circuits. The second set
dle of the key is moved to the dash position, of contacts may be used to mute the sta-
and the dashes must be produced manually. tion receiver during the key down condition.
Note: Some relays do not have nonmetallic
Keyer The complete keyer is housed strikers on either the pole pieces or the
Construction in a miniature aluminum case armature. Consequently the relay action is
4" high, Srr wide and 6" deep. somewhat sluggish. This condition can be
The major circuitry is mounted on two corrected by drilling and tapping the arma-
phenolic boards stacked one behind the other ture and installing a 2 -56 brass screw. The
and mounted to the front panel of the case screw is adjusted so that 2 mm. to 6 mm.
by bolts and spacers. The rear board (seen protrude from the armature. A lock nut is
in figure 30) contains the multivibrator used to prevent a shift in the position of the
circuit and its clamp transistor, the flip- screw.
CHAPTER TWENTY -EIGHT
The trend in design of transmitters for a -m excitation for high -level amplifiers run-
operation on the high- frequency bands is to- ning up to the 1000 -watt power limit. The
ward the use of a single high -level stage. amplifiers shown in this chapter may be
The most common and most flexible ar- easily driven by such exciters.
rangement includes a compact bandswitching
exciter unit, with 15 to 100 watts output 28 -1 Power Amplifier
on all the high- frequency bands, followed Design
by a simple power -amplifier stage. In many
cases the exciter unit is placed on the oper- Choice of Either tetrode or triode tubes may
ating table, with a coaxial cable feeding the Tubes be used in high- frequency power
drive to the power amplifier, although some amplifiers. The choice is usually
operators prefer to have the exciter unit in- dependent on the amount of driving power
cluded in the main transmitter housing. that is available for the power amplifier. If
This trend is a natural outgrowth of the a transmitter -exciter of 100 -watt power ca-
increasing importance of vfo operation on pability is at hand (sideband or a -m) it
the amateur bands. It is not practical to would be wise to employ a power amplifier
make a quick change in the operating fre- whose grid- driving requirements fall in the
quency of a transmitter when a whole suc- same range as the output power of the ex-
cession of stages must be retuned to reso- citer. Triode tubes running 1- kilowatt in-
nance following the frequency change. An- put (plate- modulated) generally require
other significant factor in implementing the some 50 to 80 watts of grid- driving power.
trend has been the wide acceptance of com- Such a requirement is easily met by the out-
mercially produced 100- to 250-watt trans- put level of the 100 -watt transmitter which
mitter- exciters. These supply SSB, c -w and should be employed as the exciter. Tetrode
www.americanradiohistory.com
Power Amplifier Design 653
TOANTENNA TO ANTENNA
CIRCUIT CIRCUIT
Figure 1 Figure 2
www.americanradiohistory.com
HANDBOOK Push -Pull Triode Amplifiers 655
minimum capacitance, and it is possible to All r -f leads should be made as short and
obtain a unit that will be satisfactory on direct as possible. The leads from the tube
all bands from 10 to 80 meters without the grids or plates should be connected directly
need for auxiliary padding capacitors. The to their respective tank capacitors, and the
rotor of the grid capacitor is grounded, leads between the tank capacitors and coils
simplifying mounting of the capacitor and should be as heavy as the wire that is used
providing circuit balance and electrical sym- in the coils themselves. Plate and grid leads
metry. Grounding the rotor also helps to to the tubes may be made of flexible tinned
retard vhf parasitics by bypassing them to braid or flat copper strip. Neutralizing leads
ground in the grid circuit. The LC ratio in should run directly to the tube grids and
the grid circuit should be fairly low, and plates and should be separate from the grid
care should be taken that circuit resonance and plate leads to the tank circuits. Having
is not reached with the grid capacitor at a portion of the plate or grid connections to
minimum capacitance. That is a direct in- their tank circuits serve as part of a neu-
vitation for instability and parasitic oscilla- tralizing lead can often result in amplifier
tions in the stage. The grid coil may be instability at certain operating frequencies.
wound of No. 14 wire for driving powers Excitation In general it may be stated
of up to 100 watts. To restrict the field and Requirements
thus aid in neutralizing, the grid coil should
that the over-all power re-
quirement for grid- circuit ex-
be physically no larger than absolutely neces-
sary. citation to a push -pull triode amplifier is ap-
proximately 10 percent of the amount of
Circuit Layout The most important consid-
the power output of the stage. Tetrodes re-
eration in constructing a quire about 1 percent to 3 percent excita-
push -pull amplifier is to maintain electrical tion, referred to the power output of the
symmetry on both sides of the balanced cir- stage. Excessive excitation to pentodes or
cuit. Of utmost importance in maintaining tetrodes will often result in reduced power
electrical balance is the control of stray output and efficiency.
capacitance between each side of the circuit
and ground. Push -Pull Symmetry is the secret of suc -
Large masses of metal placed near one side Amplifier cessful amplifier design. Shown
of the grid or plate circuits can cause serious Construction in figures 3 and 4 are views of
unbalance, especially at the higher frequen- a 1- kilowatt amplifier designed
cies, where the tank capacitance between for operation in amateur bands between 80
one side of the tuned circuit and ground is and 10 meters. The circuit corresponds to
often quite small in itself. Capacitive un- that shown in figure 2, with the addition of
balance most often occurs when a plate or parasitic suppressors in the grid leads to
grid coil is located with one of its ends the triode tubes.
close to a metal panel. The solution to this Larger triode tubes such as the 810 and
difficulty is to mount the coil parallel to the 230TH make excellent r -f amplifiers at the
panel to make the capacitance to ground kilowatt level, but care must be taken in
equal from each end of the coil, or to place amplifier layout as the interelectrode capac-
a grounded piece of metal opposite the itance of these tubes is of importance. One
"free" end of the coil to accomplish a ca- tube and one neutralizing capacitor is placed
pacity balance. on each side of the tank circuit to permit
Whenever possible, the grid and plate coils very short interconnecting leads. The rela-
should be mounted at right angles to each tive position of the tubes and capacitors is
other, and should be separated far enough transposed on each side of the chassis, as
apart to reduce coupling between them to shown in the illustrations. The plate tank
a minimum. Coupling between the grid and coil is mounted parallel to the front panel
plate coils will tend to make neutralization of the amplifier on a phenolic plate support-
frequency sensitive, and it will be necessary ed by the tuning capacitor which sits atop a
to readjust the neutralizing capacitors of small chassis -type box. The grid -circuit tun-
the stage when changing bands. ing capacitor is located in this box, as seen
656 High- Frequency Power Amplifiers THE RADIO
Figure 3 Figure 4
UNIQUE CHASSIS LAYOUT PERMITS LEFT-HAND VIEW OF KILOWATT
SHORT LEADS IN KILOWATT AMPLIFIER OF FIGURE 3
AMPLIFIER Above shielded meter box is the protective
Large size components required for high -level "microswitch" which opens the primary power
amplifier often complicate amplifier layout. circuit when the panel door is not closed. Tube
In this design, the plate tank capacitor sits sockets are recessed in the chassis so that
astride small chassis running lengthwise on top of tube socket shells are about 1'2-inch
main chassis. Inductor is mounted to phenolic above chassis level. On right side of amplifier
plate atop capacitor. Variable link is panel (facing it from the rear) the tube socket is
driven through right -angle gear drive. Plate nearest the panel, with the neutralizing ca-
circuit is grounded by safety arm when panel pacitor behind it. On the opposite side, the
door is opened. Note that plate capacitor is capacitor is nearest the panel with the tube
mounted on four TV -type capacitors which directly behind it. This layout transposition
serve to bypass unit, and also act as supports. produces very short neutralizing leads, since
A small parasitic choke is visible next to the
connections may be made through the stator
grid terminal of the 810 tube. of plate tuning capacitor.
www.americanradiohistory.com
HANDBOOK Push -Pull Tetrode Amplifiers 657
Figure 6
www.americanradiohistory.com
HANDBOOK Tetrode Pi- Network Amplifiers 659
the plate lead) shunted by the suppressor The pi circuit is usually shunt -fed to re-
resistor. move the d -c plate voltage from the coils
Because of the compact size of many tet- and capacitors. The components are held at
rodes, it is necessary to cool the filament and ground potential by completing the circuit to
plate seals with a blast of air. A small blower ground through the choke (RFC,). Great
may be mounted on the pressurized chassis stress is placed on the plate -circuit choke
to direct a stream of cooling air at the fila- (RFC,). This component must be specially
ment seals of the tube, and through the base designed for this mode of operation, having
to the glass envelope. Many transmitting low interturn capacitance and no spurious
tubes require special air sockets and chim- internal resonances throughout the operat-
neys to provide adequate cooling. Cooling ing range of the amplifier.
data for most popular transmitting tetrodes Parasitic suppression is accomplished by
may be obtained on request from the manu- means of chokes PC, and PC_ in the screen,
facturer. grid, or plate leads of the tetrode. Suitable
values for these chokes are given in the parts
list of figure 8. Effective parasitic suppres-
28 -4 Tetrode Pi- Network sion is dependent to a large degree on the
Amplifiers choice of screen bypass capacitor C1. This
component must have extremely low induct-
The most popular amplifier today for both ance throughout the operating range of the
commercial and amateur use is the pi -net- amplifier and well up into the vhf parasitic
work configuration shown in figure 8. This range. The capacitor must have a voltage
circuit is especially suited to tetrode tubes, rating equal to at least twice the screen po-
although triode tubes may be used under tential (four times the screen potential for
certain circumstances. plate modulation). There are practically no
A common form of pi- network amplifier capacitors available that will perform this
is shown in figure 8A. The pi circuit forms difficult task. One satisfactory solution is
the matching system between the plate of to allow the amplifier chassis to form one
the amplifier tube and the low- impedance, plate of the screen capacitor. A "sandwich"
unbalanced, antenna circuit. The coil and is built on the chassis with a sheet of insu-
input capacitor of the pi may be varied to lating material of high dielectric constant
tune the circuit over a 10 to 1 frequency and a matching metal sheet which forms the
range (usually 3.0 to 30 MHz). Operation screen side of the capacitance. A capacitor
over the 20- to 30 -MHz range takes place of this type has very low internal induct-
when the variable slider on coil L_ is ad- ance but is very bulky and takes up valuable
justed to short this coil out of the circuit. space beneath the chassis. One suitable ca-
Coil L, therefore comprises the tank in- pacitor for this position is the Centralab
ductance for the highest portion of the type 858S -1000, rated at 1000 pf at 5000
operating range. This coil has no taps or volts. This compact ceramic capacitor has
sliders and is constructed for the highest relatively low internal inductance and may
possible Q at the high- frequency end of the be mounted to the chassis by a 6 -32 bolt.
range. The adjustable coil (because of the Further screen isolation may be provided by
variable tap and physical construction) a shielded power lead, isolated from the
usually has a lower Q than that of the fixed
screen by a .001 -fd ceramic capacitor and
coil.
a 100 -ohm carbon resistor.
The degree of loading is controlled by ca-
Various forms of the basic pi- network
pacitors C, and C2. The amount of circuit
capacitance required at this point is in- amplifier are shown in figure 8. The A con-
versely proportional to the operating fre- figuration employs the so- called "all- band"
quency and to the impedance of the antenna grid -tank circuit and a rotary pi- network
circuit. A loading capacitor range of 100 coil in the plate circuit. The B circuit uses
to 2500 pf is normally ample to cover the coil switching in the grid circuit, bridge
3.5- to 30 -MHz range. neutralization, and a tapped pi- network
660 High- Frequency Power Amplifiers THE RADIO
C2 L1 2
LO W-Z
1 11 1
OUTPUT
PC, PC2 ll' 3
ao 000
RFC2 TC3 S2 C< R FC 1
0'1
h TST _L
EACITATIOti 01 00 6
Hl, 1. NC y100
1r0
o
q
Hr Hl 001
E11. SCR
O
BfHV
100
M
001
C
BIAS 5 V 1
LOW -Z
OUTPur
EXCITATION
Figure 8
www.americanradiohistory.com
HANDBOOK Tetrode Pi- Network Amplifiers 661
r
SSIRE SUSUPY
SCSE EN SUPPLY ic suppression may be omitted. The price
PLATE SUPPLY
that must be paid is the additional excitation
that is required to develop operating voltage
across grid resistor R,.
The pi- network circuit of figure 8C is in-
teresting in that the rotary coil (L2) and
the plate tuning capacitor (C:,) are ganged
ruNE together by a gear train, enabling the cir-
"COMMON cuit to be tuned to resonance with one panel
MINUS'
LEAD control instead of the two required by the
circuit of figure 8A. Careful design of the
rotary inductor will permit the elimination
Figure 9
of the auxiliary high- frequency coil (L,) ,
GROUNDED -SCREEN -GRID thus reducing the cost and complexity of
CONFIGURATION PROVIDES HIGH the circuit.
ORDER OF ISOLATION IN TETRODE
AMPLIFIER STAGE The Grounded -Screen For maximum shielding,
Configuration it necessary to operate
is
A- Typical amplifier circuit has cathode return the tetrode tube with
at ground potential. All circuits return to
cathode. the screen at r -f ground potential. As the
S -All
circuits return to cathode, but ground
screen has a d -c potential applied to it (in
point has been shifted to screen terminal
of tube. Operation of the circuit remains grid- driven circuits), it must be bypassed
the same, as potential diff between to ground to provide the necessary r -f re-
elements of the tube are the same as in
circuit A. turn. The bypass capacitor employed must
C- Practical grounded -screen circuit. "Common perform efficiently over a vast frequency
minus" lead returns to negative of plate
supply, which cannot be grounded. Switch spectrum that includes the operating range
Ss removes screen voltage for tune -up plus the region of possible vhf parasitic oscil-
purposes. lations. This is a large order, and the usual
bypass capacitors possess sufficient inductance
coil with a vacuum tuning capacitor. Fig- to introduce regeneration into the screen cir-
ure 8C shows an interesting circuit that is cuit, degrading the grid -plate shielding to
becoming more popular for class -AB, a marked degree. Nonlinearity and self -
linear operation. A tetrode tube operating oscillation can be the result of this loss of
under class -AB, conditions draws no grid circuit isolation. A solution to this problem
current and requires no grid -driving power. is to eliminate the screen bypass capacitor,
Only r -f voltage is required for proper by grounding the screen terminals of the
operation. It is possible therefore to dispense tube by means of a low- inductance strap.
662 High- Frequency Power Amplifiers THE RADIO
Screen voltage is then applied to the tube by
grounding the positive terminal of the screen
supply, and "floating" the negative of the
screen and bias supplies below ground po-
tential as shown in figure 9. Meters are
placed in the separate- circuit cathode return
leads, and each meter reads only the cur-
rent flowing in that particular circuit. Oper-
Figure 10
ation of this grounded- screen circuit is nor-
mal in all respects, and it may be applied to INDUCTIVE TUNING ELIMINATES
any form of grid- driven tetrode amplifier INPUT TUNING CAPACITOR
with good results.
www.americanradiohistory.com
HANDBOOK Cathode- Driven Amplifiers 663
R -F OUT R -F OUT
EMC
EXC
O
Figure 11
exciters have power output capabilities of from the large amount of feedback inherent
the order of 70 to 100 watts and are thus in the grounded -grid circuit.
well suited to drive high -power grounded - A simplified version of the grounded -grid
grid linear amplifier stages whose total ex- amplifier is shown in figure 11B. This con-
citation requirements fall within this range. figuration utilizes an untuned input circuit,
and is very popular as an inexpensive and
Distortion Laboratory measurements made simplified form of the more sophisticated
Products on various tubes in the circuit circuit of figure 11A. It has inherent limita-
of figure 11A show that a dis- tions, however, that should be recognized.
tortion reduction of the order of S to 10 In general, slightly less power output and
decibels in odd -order products can be ob- efficiency is observed with the untuned -cath-
tained by operating the tube in cathode - ode circuit, odd -order distortion products
driven service as opposed to grid- driven run 4 to 6 decibels higher, and the circuit
service. The improvement in distortion varies is harder to drive and match to the exciter
from tube type to tube type, but some than is the tuned -cathode circuit of figure
order of improvement is noted for all tube 11A. Best results are obtained when the
types tested. Most amateur -type transmit- coaxial line of the driver stage is very short
ting tubes provide signal -to- distortion ratios -a few feet or so. Optimum linearity re-
of -20 to -30 decibels at full output in quires cathode circuit Q that can only be
class -AB, grid- driven operation. The ratio supplied by a high -C tank circuit.
increases to approximately -25 to -40 Since the single -ended class -B grounded -
decibels for class -B grounded -grid operation. grid linear amplifier draws grid current on
Distortion improvement is substantial, but only one -half (or less) of the operating
not as great as might otherwise be assumed cycle, the sideband exciter "sees" a low-
664 High- Frequency Power Amplifiers THE RADIO
R -F OUT R -F OUT.
Figure 12
TETRODE TUBES MAY BE USED IN CATHODE -DRIVEN AMPLIFIERS
A- Tetrode tube may be used in cathode-driven configuration, with bias and screen voltages
applied to elements which are at r -f ground potential. I -Grid current of grounded -grid tube
is easily monitored by RC network which lifts grid above ground sufficiently to
permit
millivoltmeter to indicate voltage drop across -ohm resistor. Meter is a 0 -1 d -c milliammetera
1
impedance load during this time, and a very grounded -grid stage. The g -g linear ampli-
high -impedance load over the balance of the fier may have either configuration, although
cycle. Linearity of the exciter is thereby the majority of the g -g stages are single
affected and the distortion products of the ended, as push -pull offers no distinct ad-
exciter are enhanced. Thus, the driving vantages and adds greatly to circuit com-
signal is degraded in the cathode circuit of plexity.
the grounded -grid stage unless the unbal- The cathode circuit of the amplifier is
anced input impedance can be modified in resonated to the operating frequency by
some fashion. A high -C tuned circuit, stores means of a high -C tank (figure 11A). Reso-
enough energy over the operating r -f cycle so nance is indicated by maximum grid current
that the exciter "sees" a relatively constant of the stage. A low value of SWR on the
load at all times. In addition, the tuned cir- driver coaxial line may be achieved by ad-
cuit may be tapped or otherwise adjusted so justing the tap on the tuned circuit, or by
that the SWR on the coaxial line coupling varying the capacitors of the pi- network
the exciter to the amplifier is relatively low. (figure 11C). Correct adjustments will pro-
This is a great advantage, particularly in duce minimum SWR and maximum ampli-
the case of those exciters having fixed -ratio fier grid current at the same settings. The
pi- network output circuits designed expressly cathode tank should have a Q of 2 or more.
for a 50 -ohm termination. The cathode circuit should be completely
Finally, it must be noted that removal of shielded from the plate circuit. It is common
the tuned cathode circuit breaks the ampli- practice to mount the cathode components
fier plate- circuit return to the cathode, and in an "r -f tight" box below the chassis of
r -f plate -current pulses must return to the the amplifier, and to place the plate circuit
cathode via the outer shield of the driver components in a screened box above the
coaxial line and back via the center con- chassis.
ductor! Extreme fluctuations in exciter load- The grid (or screen) circuit of the tube is
ing, intermodulation distortion, and TVI operated at r -f ground potential, or may
can be noticed by changing the length of have d -c voltage applied to it to determine
the cable between the exciter and the the operating parameters of the stage (figure
grounded -grid amplifier when an untuned- 11A). In either case, the r -f path to ground
cathode input circuit and a long intercon- must be short, and have extremely low in-
necting coaxial line are used. ductance, otherwise the screening action of
Cathode -Driven Design features of the sin - the element will be impaired. The grid (and
Amplifier gle -ended and push -pull am- screen) therefore, must be bypassed to
Construetion plifiers discussed previously ground over a frequency range that includes
apply equally well to the the operating spectrum as well as the region
www.americanradiohistory.com
HANDBOOK Cathode -Driven Amplifiers 665
of possible vhf parasitic oscillations. This is near full capacitance and the plate tank
quite a large order. The inherent inductance capacitor is adjusted for resonance (mini-
of the usual bypass capacitor plus the length mum plate current) . Drive is advanced until
of element lead within the tube is often suf- grid current is noted and the plate circuit
ficient to introduce enough regeneration into is loaded by decreasing the capacitance of
the circuit to degrade the linearity of the the plate loading capacitor. The drive is
amplifier at high signal levels even though increased until about one -half normal grid
the instability is not great enough to cause current flows, and loading is continued (re-
parasitic oscillation. In addition, it is often resonating the plate tank capacitor as re-
desired to "unground" the grounded screen quired) until loading is near normal. Finally,
or grid sufficiently to permit a metering cir- grid drive and loading are adjusted until
cuit to be inserted. PEP - condition plate and grid currents are
One practical solution to these problems normal. The values of plate and grid current
is to shunt the tube element to ground by should be logged for future reference. At
means of a 1 -ohm composition resistor, by- this point, the amplifier is loaded to the
passed with a .01 -fd ceramic disc capacitor. maximum PEP input condition. In most
The voltage drop caused by the flow of grid cases, the amplifier and power supply are
(or screen) current through the resistor can capable of operation at this power level for
easily be measured by a millivoltmeter whose only a short period of time, and it is not
scale is calibrated in terms of element cur- recommended that this condition be per-
rent (figure 12B). mitted for more than a minute or two.
The plate circuit of the grounded -grid The exciter is now switched to the SSB
amplifier is conventional, and either pi -net- mode and, with speech excitation, the grid
work or inductive coupling to the load may and plate currents of the cathode- driven
be used. There is some evidence to support stage should rise to approximately 40 to SO
the belief that intermodulation distortion percent of the previously logged PEP read-
products are reduced by employing plate cir- ings. The exact amount of meter movement
cuit Q's somewhat higher than normally with speech is variable and depends on meter
used in class -C amplifier design. A circuit damping and the peak to average ratio of
Q of 10 or greater is thus recommended the particular voice. Under no circum-
for ground -grid amplifier plate circuits. stances, however, should the voice meter
readings exceed SO percent of the PEP ad-
Tuning the Since the input and output justment readings unless some form of
Grounded -Grid circuits of the grounded - speech compression is in use.
Amplifier grid amplifier are in series, a To properly load a linear amplifier for
certain proportion of driv- the so- called "two- kilowatt PEP" condition,
ing power appears in the output circuit. If it is necessary for the amplifier to be tuned
full excitation is applied to the stage and the and loaded at the two -kilowatt level, albeit
output circuit is opened, or the plate voltage briefly. It is necessary to use a dummy load
removed from the tube, practically all of to comply with the FCC regulations, or
the driving power will be dissipated by the else a special test signal must be used. To
grid of the tube. Overheating of this ele- achieve a ratio of 2:1 between the tune -up
ment will quickly occur under these circum- condition and the PEP condition an audio
stances, followed by damage to the tube. pulser and single -tone driving signal may
Full excitation should therefore never be be used. Shown in figure 13 is a pulser
applied to a grounded -grid stage unless plate having a duty cycle of about 0.44. For a
voltage is applied beforehand, and the stage d -c meter reading of 880 watts input using
is loaded to the antenna. the pulser and a single audio tone, the PEP
Tuneup for sideband operation consists input level and corresponding amplifier
of applying full plate voltage and sufficient loading adjustments will satisfy the two -
excitation (carrier injection) so that a kilowatt PEP conditions. An oscilloscope and
small rise in resting plate (cathode) current audio oscillator are necessary to conduct this
is noted. The plate loading capacitor is set exercise, but these instruments are required
www.americanradiohistory.com
666 High- Frequency Power Amplifiers THE RADIO
6J5 tion but are not recommended because of
.0015
poor signal -to- distortion ratios.
AUDIO
INPUT Certain types of tetrodes, exemplified by
100 R ' 5 PULSCO AUDIO
the 4 -65A, 4X150A, 4CX300A, and 4CX-
OUTPUT
I000A should not be used as grounded -grid
amplifiers unless grid bias and screen voltage
are applied to the elements of the tube (fig-
ure 12A). The internal structure of these
MERITS S040
OR tubes permits unusually high values of grid
EQUIVALENT
current to flow when true grounded -grid
circuitry is used, and the tube may be easily
,ISV.ti damaged by this mode of operation.
The efficiency of a typical cathode -driven
Figure 13 amplifier runs between 55- and 6f- percent,
indicating that the tube employed should
AUDIO PULSER FOR HIGH -POWER have plenty of plate dissipation. In general,
TUNE -UP OF AMPLIFIER the PEP input in watts to a tube operating
This simple audio pulser modifies the audio in grounded -grid configuration can safely
signal to the sideband exciter so that it has a be about 2.5 to 3 times the rated plate
high peak -to- average power ratio. Amplifier
may be thus tuned for two kilowatt PEP dissipation. Because of the relatively low
input without violating the one -kilowatt average -to -peak power of the human voice
maximum steady-state condition. it is tempting to push this ratio to a higher
figure in order to obtain more output from
items for any well -equipped sideband station. a given tube. This action is unwise in that
For best linearity, the output circuit of the odd -order distortion products rise rapid-
the grounded -grid stage should be over - ly when the tube is overloaded, and because
coupled so that power output drops about no safety margin is left for tuning errors or
2- percent from maximum value. A simple circuit adjustments.
output r -f voltmeter is indispensable for
proper circuit adjustment. Excessive grid Neutralization At some high frequency
of the G -G Stage
the shielding action of the
current is a sign of antenna undercoupling,
and overcoupling is indicated by a rapid drop grid of the g -g amplifier
in output power. Proper grounded -grid stage deteriorates. Neutralization may be neces-
operation can be determined by finding the sary at higher frequencies either because of
optimum ratio between grid and plate cur- the presence of inductance between the ac-
rent and by adjusting the drive level and tive grid element and the common returns
loading to maintain this ratio. Many manu- of the input and output circuit, or because
facturers now provide grounded -grid oper- of excessive plate- cathode capacitance.
ation data for their tubes, and the ratio of Neutralization, where required, may be
grid to plate current can be determined from accomplished by feeding out -of -phase energy
the data for each particular tube. from the plate circuit to the filament circuit
(figure 14A) or by inserting a reactance in
Choice of Tubes Not all tubes are suitable series with the grid (figure 14B). For values
for G -G Service for grounded -grid service. of plate- cathode capacitance normally en-
In addition, the signal -to- countered in tubes usable in g -g service, the
distortion ratio of the suitable tubes varies residual inductance in the grid -ground path
over a wide range. Some of the best g -g provides sufficient reactance, and in some
performers are the 811A, 813, 7094, 4 -125A, cases even series capacitance will be required.
4 -250A, 4 -400A, and 4- 1000A. In addition, Typical tube electrode capacitances are shown
the 3 -400Z and 3 -1000Z triodes are specifi- in figure 15A. These can be represented by
cally designed for low distortion, grounded - an equivalent star connection of three ca-
grid amplifier service. The older types 837 pacitors (figure 15B). If an inductance (L)
and 803 are used extensively for g -g opera- is placed in series with Ce so that a resonant
www.americanradiohistory.com
HANDBOOK Cathode- Driven Amplifiers 667
R -F OUT. R -F our.
Figure 14
Neutralization of the g -g stage may be necessary at the higher frequencies. Energy fed back
in proper phase from plate to cathode is used to neutralize the unwanted energy fed through
the tube (A). Reactance placed in series with the grid return lead (B) will accomplish the
same result. The inductance (L) usually consists of the internal grid lead of the tube, and
capacitor C may be the grid bypass capacitor. A series -resonant circuit at the operating
frequency is thus formed.
circuit is formed (figure 15C), point O will by proper choice of the bias bypass capacitor
be at ground potential (15D). This prevents (figure 14B). Below a certain frequency
the transfer of energy from point P to point determined by the physical geometry of the
K, since there now exists no common cou- tube, neutralization may be accomplished
pling impedance. The determination of by adding inductance to the grid- return
value C, and L are shown in figure 15. lead; above this frequency it may be neces-
It is apparent that when the plate- cathode sary to series tune the circuit for minimum
capacitance of the tube is small as compared energy feedthrough from cathode to plate.
to the plate -grid and the grid -cathode ca- Most tubes are sufficiently well screened so
pacitance, C, is a large value and the re- that series inductive neutralization at the
quired value of inductance L is small. In lower frequencies is unnecessary, but series
practical cases the value of L is supplied by capacitance tuning of the grid- return lead
the tube and lead inductance, and the grid - may be required to prevent oscillation at
to- ground impedance can be closely adjusted some parasitic frequency in the vhf range.
Cn-K
Ccx(CP-cxCr-x)t(CR-xxCs:-x)t(Cs:-x xC-a)
CP-R
1
(2/7r)2 x Cc
Figure 15
Figure 16
www.americanradiohistory.com
HANDBOOK Six -Meter Kilowatt Linear 669
RECEIVER
3-400 Z 001 ANT.
PC PLATE STRAP 5 KV 2 L3 RELAY ANTENNA
50
4 2W
507I 190 ttt--- RFC 2 A
11 11
C2 'IC3 2-50 B
001
.001
KV
3 J1 TOEXCIT. CALIBRATE
140
.001
4
.0 1
TO R Y 0-1 MA
50K GRI0
10W 0- 200 MA
PLATE 01
0 -500 MA C 500
50 RFC 3
IOW Z -50
500
50 0 500 = C
_
G
C 6V
115 V ti GND. -H V +HV 115 V
JS J3 J4
Figure 17
circuit (L, -C,) is used to preserve the wave- circuit is shown for using the linear ampli -
form of the driving signal and to reduce fier with a transceiver (figure 18).
harmonic distortion that may cause TVI.
The cathode coil is made of copper tubing TO
OR
EXCITER
TRANSCEIVER
and filament voltage is fed to the 3 -400Z JI
via the coil and an insulated wire passed
through the tubing. Excitation is tapped on
the coil at a point which provides a nomi-
nal 50-ohm load to the exciter.
The plate circuit of the amplifier utilizes
a pi -L network to achieve a high order of
001
3RVT
r
To L' Figure 18
harmonic suppression and a simple diode
voltmeter is used to monitor the r -f output SUGGESTED ANTENNA -RELAY CIRCUIT
voltage. An antenna relay (RY) is incor- FOR USING AMPLIFIER WITH
porated in the amplifier, and an alternative TRANSCEIVER
www.americanradiohistory.com
670 High- Frequency Power Amplifiers THE RADIO
Metering and It is necessary to measure and is made by merely shunting a portion
Suppression Circuits both grid and plate cur- of the plate strap with a composition resistor.
rent in a cathode- driven
amplifier to establish the proper ratio of grid Amplifier The amplifier is inclosed in an
to plate current. At the higher frequencies Construction "r -f tight" cabinet measuring
it is desirable to directly ground the grid of 13" X 83/" X 10 ". A stand-
the amplifier tube and not to rely on ques- ard 12" X 10" X 3" aluminum chassis is
tionable bypass capacitors to insure that the used, along with an 83/4" X 13" panel (cut
grid remains at ground potential. Grid cur- from a standard aluminum relay rack pan-
rent, therefore, is measured in the cathode - el). The cabinet is made by bending a sheet
return circuit of the amplifier by meter M,. of light aluminum (31" X I1 ") to fit
Plate current is measured in the B -minus lead around the panel. It is riveted to 13" X 11"
to the power supply by meter _. A simpli- bottom plate. The rear of the cabinet is a
fied metering circuit is shown in figure 19. sheet of perforated aluminum fastened to
the cabinet with 1/4-inch aluminum angle
Rs stock. Additional angle stock is cut to
3 -4002 'to- length and fastened to the front edge of
the cabinet to secure the panel. A 4 -inch
hole is cut in the cabinet directly above
the 3 -400Z and is covered with a small
sheet of perforated aluminum. This shielded
vent permits the heated air from the tube to
escape from the inclosure.
After the amplifier is completed and slid
within the cabinet, it is fixed in place by
means of sheet -metal screws placed at the
bottom edges of the chassis and around the
panel, making the inclosure as "r -f tight"
+B as possible.
POWER SUPPLY A meter shield is used to protect the
panel meters from the r -f field of the plate
circuit and to suppress r -f leakage from
Figure 19 the cabinet via the meter faces. The box-
METERING CIRCUITS FOR KILOWATT
like shield is attached to the panel by means
AMPLIFIER of aluminum angle stock which is held to
the panel by the meter mounting bolts. All
A -D -c meter circuit showing grid and plate
meters placed in low- potential return
paint is removed from the rear of the panel
leads. The B -minus of the power supply to provide a good ground connection to the
"Roots" above ground by virtue of the 50- meter shield and to the chassis and cabinet.
ohm resistor, which may be placed in the
power supply, if desired. The 3 -400Z tube requires forced -air cool-
8- Peak -responding voltmeter circuit useful for ing during operation and a blower (B) is
adjusting linsar for a -m service. mounted on the chassis and activated with
application of filament voltage. An F.imac
This amplifier was checked for parasitics SK -410 air -system socket and SK -416 air
and it was found that the usual plate para- chimney are used to achieve proper air flow
sitic choke was not required for stable oper- around the filament and plate seals of the
ation. A variation in circuit layout, how- tube. The air enters the under -chassis area,
ever, or changes in ground- return currents passes through the socket and is directed
may allow weak parasitic oscillation to take over the envelope and plate seal of the 3 -4002
place. If this condition is found, placement by the glass chimney placed over the tube.
of a parasitic choke in the plate lead will The air then passes out through the vent
suppress the unwanted parasitic. A practical placed in the cabinet directly above the
parasitic choke is shown in the schematic 3 -400Z.
www.americanradiohistory.com
HANDBOOK Six -Meter Kilowatt Linear 671
Layout of the major components may be terminal. The two filament leads are covered
seen in the photographs. The air- system with lengths of cambric or plastic tubing,
socket is mounted on the underside of the over which is slipped a length of shield braid,
chassis in a 31/4-inch diameter cutout. The grounded to the chassis at both ends.
spring clips that hold the chimney in place The excitation tap on the coil is placed
fasten with the same bolts used to mount about 3/4 -turn from the bottom (bypassed)
the socket, which is oriented so that filament end. The exact tap point is not critical and
pins 1 and f are facing the front of the may be easily checked by adjusting it for
chassis. The cathode tuning capacitor (C1) minimum SWR on the coaxial line from the
is mounted on the front apron of the chassis exciter to the amplifier.
with insulated washers as the rotor is above The three grid pins of the 3 -400Z socket
ground by the amount of the filament volt- are grounded by passing a 1/4-inch wide
age. The cathode coil is a dual winding, made copper strap through the slot in the socket
of copper tubing having an insulated center adjacent to each grid pin and soldering the
conductor. A section of 1/4-inch soft - strap directly to the flat tab on the pin.
drawn copper tubing about a foot long is The straps are then bolted to the chassis
needed to make the coil. Before the coil is just clear of the socket. The lead from the
wound, the ends of the tubing are smoothed coaxial input receptacle (J1) to the coupling
with a file and a length of #12 cotton - capacitor adjacent to the cathode circuit is
covered (or formvar- insulated) wire is made of a length of RG -58/U coaxial line,
passed through the tubing. The coil is then with the outer shield of the line grounded
wound about a 3/4-inch diameter wood dowel at the input receptacle and also at the
rod used as a temporary form, spacing the mounting stud of the bypass capacitor for
three turns to a length of two inches. The the coil.
tubing is trimmed, and the inner wire is
left projecting about ten inches from each The Plate - Layout of the components
end. The coil is mounted close to the tube Circuit Assembly above the chassis are shown
socket (figure 21) with one end supported in figure 20. The plate
by the filament pins of the tube socket. The tuning and loading capacitors (Co and CO
inner conductor is trimmed to length and are mounted on 1z -inch ceramic insulators.
soldered to one filament pin, and the tubing The tuning capacitor is rotated 90 degrees
is connected to the other filament pin by on its side and held in position with small
means of a short length of copper strap about aluminum brackets. A common ground con-
-inch wide, cut from copper "flashing" nection made of a length of 1/a -inch wide
material. The end of the coil is equidistant copper strap connects the rear rotor terminals
from the filament pins. The strap encircles of the capacitors. In addition, the capacitor
one end of the tubing and is soldered in rotor wipers are connected to the common
place, with the other end soldered to the ground strap.
pin. The filament bypass capacitor is sol- A second strap grounds the rotors to a
dered directly between the filament pins of common ground point on the chassis under
the socket. A second short length of copper the stud of the high -voltage bypass capaci-
strap jumpers the first strap to the stator tor at the lower end of the plate r -f choke.
of the cathode tuning capacitor. The shafts of the variable capacitors are
The opposite end of the cathode coil is driven with insulated couplers to prevent
bypassed to ground by a ceramic capacitor ground -loop currents from flowing through
which also supports the coil. The inner con- the shafts into the panel.
ductor is bypassed to the outside tubing at The pi- section of the plate tank circuit
this point, and a length of copper strap (L2) is made of a length of 3/16 -inch di-
makes a connection to the rotor of the ameter copper tubing, the five turns being
tuning capacitor. The inner conductor con- spaced three inches long, with an inside di-
tinues over to the filament transformer and ameter of 11/8". The ends of the coil are flat-
a second length of #12 wire is run from the tened and drilled to be bolted to the stator
copper tubing to the second transformer lugs of the capacitor with 4 -40 hardware.
www.americanradiohistory.com
672 High- Frequency Power Amplifiers THE RADIO
Figure 20
www.americanradiohistory.com
HANDBOOK Six -Meter Kilowatt Linear 673
,Z.
The L- section of the tank coil (Ls) is A diode r -f voltmeter is mounted beneath
made of a length of %8 -inch diameter tubing the chassis in a small aluminum box posi-
and has four turns, 3/4 inch in diameter and tioned over the r -f feedthrough insulator
two inches long. The coil is placed at right which supports the end of the L-network
angles to the main tank coil and is bolted above the chassis. The lead from the volt-
between the stator of C3 and a ceramic feed - meter circuit to the calibrate potentiometer
through insulator. Large lugs may be placed on the panel is run in shield braid, as are
on the coil to facilitate bolting it in position. the leads from the center tap of the filament
The plate r -f choke is homemade, and is transformer. Tight rubber grommets are
wound on a 1/2-inch diameter ceramic in- used in all chassis holes to restrict air leaks.
sulator. A commercial choke may be used, if
desired. The base of the choke screws on Amplifier When the amplifier has been
the bolt of the high -voltage feedthrough Adjustment wired and inspected, it is ready
insulator on the chassis, and is bypassed at for initial checks. Air is di-
this point with a ceramic capacitor. The r -f rected into the tube socket by means of a
choke is positioned close to the 3 -400Z chim- temporary bottom plate (cardboard) taped
ney to permit a reasonably short plate lead, to the chassis. Filament voltage is applied
and the plate blocking capacitor is mounted and the blower motor should start. A strong
to the top of the choke with a length of blast of air out of the tube chimney should
strap which extends downwards toward the be noted. Tube filament voltage should be
plate tank capacitor. adjusted to 5.0 volts at the socket with an
The coaxial antenna relay is mounted on accurate meter. Filament voltage is now re-
the top of the chassis positioned so the moved and the input and output coaxial re-
output lead from the L- section of the tank ceptacles are temporarily terminated in 50-
circuit can be connected directly to the ohm, 1 -watt composition resistors, which
input receptacle. The connection is made by may be soldered across the receptacles for
trimming down a coaxial connector and this test. A grid -dip meter is tuned to 50
soldering a short length of #10 wire to the MHz and brought near the cathode coil
center terminal to make the connection to (the 3 -400Z being in the socket). The meter
the coil. The antenna receptacle of the relay should show resonance with the cathode
extends beyond the rear apron of the chassis tuning capacitor about two -thirds meshed.
and through the rear of the cabinet. The The plate tank circuit is now tested, with
receive receptacle is fed with a length of the tuning capacitor about one -half meshed
RG -58 %U coaxial cable which terminates and the loading capacitor about two -thirds
at the coaxial receptacle on the rear apron of meshed. Grid -dip resonance at these settings
the chassis. An auxiliary set of contacts on for 50 MHz may be achieved by slight al-
the relay are used to short out the 50K terations in the spacing of the pi- network
self -bias resistor in the cathode circuit of coil. The L- section should also show a dip
the 3 -400Z when transmitting. The resistor around 50 MHz.
serves to bias the tube to near cutoff during Once resonance of the tank circuits has
periods of reception to prevent noise being been verified, the 50 -ohm resistors are re-
generated which may interfere with recep- moved and the amplifier attached to the
tion of weak signals and also to reduce the exciter and coaxial antenna lead. A separate
standby drain on the power supply. The ground lead is run from the amplifier to the
relay is actuated by the control or VOX
power supply. A plate potential of 2500 volts
circuit of the exciter, and the relay coil
is recommended as a maximum (key -down)
should be chosen to match the voltage de-
livered from the exciter control circuit. value, and good operation can be obtained
Plate current metering is accomplished in down to 2000 volts. At the higher potential,
the negative lead to the power supply, and the resting plate current will be about 80
the negative of the supply is lifted above ma. Random variations in resting plate cur-
ground as described in the Power Supplies rent, or a show of grid current when the
chapter of this Handbook. controls are tuned (with no grid drive) is
www.americanradiohistory.com
674 High- Frequency Power Ampl ifiers THE RADIO
Figure 21
www.americanradiohistory.com
HANDBOOK 432 -MHz 500 -Watt Linear 675
ditions are as stated above, and under these remove the distortion on the peaks of the
conditions, the anode of the 3 -400Z will be signal. Under voice modulation, plate and
a cherry red in color. With carrier removed grid current will flicker a small amount
and SSB voice modulation applied, drive is upward.
advanced until voice peaks reach about 200 The combination of a peak- responding
ma plate current and about 70 ma grid cur- voltmeter, an oscilloscope, and an audio
rent. For c -w operation, the full 400 ma oscillator used with tune -up under 100 per-
plate current value may be run. cent single -tone modulation of the exciter
affords a relatively easy and accurate method
A -M Linear The amplifier may be used for of achieving proper a -m linear amplifier
Operation a -m linear service when prop- service.
erly adjusted. The amplifier ef- As with any cathode- driven amplifier,
ficiency at the peak of the modulation cycle drive should never be applied to the ampli-
is about 66 percent and efficiency under fier in the absence of plate voltage, as dam-
carrier conditions (no modulation) is about age to the grid of the tube may result. The
33 percent. As maximum plate dissipation proper sequence is to always apply plate
is 400 watts, the total a -m carrier input to voltage before drive, increasing the drive
the 3 -400Z is limited to about 600 watts level slowly from a minimum value as tun-
(2500 volts at 240 ma). In order to prop- ing adjustments are made.
erly load the amplifier to this condition for
a -m linear service, an oscilloscope and peak -
28 -7 A 500 -Watt
responding voltmeter are necessary. The r -f Amplifier for 432 -MHz
output voltmeter in the amplifier may be Linear or Class-C- Service
converted to a peak- responding instrument
as shown in figure 19B. In addition, a simple This amplifier is designed for SSB or c -w
1000 -Hz audio oscillator is used for the service at the half- kilowatt level in the
following adjustments. 432 -MHz amateur band. Making use of
Fcr preliminary tuneup, the a -m driver
is modulated 100 percent with the 1000 -Hz
tone. A driver capable of about 15 watts
carrier is required. The 3 -400Z amplifier is
loaded and drive level adjusted to 600 watts
input under this condition. Amplifier output
is monitored with the peak- responding volt-
meter, which is adjusted to full -scale reading
at the 600 -watt input level. Grid current
will run about %4 the plate- current value, or
approximately 60 ma. Once this condition
is reached, the modulation of the driver is
removed, leaving only carrier excitation. If
the linear amplifier is properly adjusted, the Figure 22
indication of the peak- responding voltmeter 500 -WATT STRIP -LINE AMPLIFIER FOR
should drop to one -half scale, corresponding 432 MHz
to an output drop to one -quarter power. The 8122 is used in a simple strip -line circuit
If the peak- voltage drop when modulation for high -power operation in the 432 -MHz band.
is removed is less than one -half, the plate Two aluminum chassis are placed back to
back to form the cavities and the strip -lines
circuit loading and grid -drive level of the are cut from aluminum plate. The amplifier is
linear amplifier must be adjusted to provide forced -air cooled by a blower mounted on the
rear of the assembly which forces air through
the correct ratio. This is an indication that the tube socket, past the anode of the tube,
antenna loading is too light for the given and out the vents cut in the top of the plate
grid drive. If this process is monitored with cavity. The r -f output cable is at the top of
the cavity and the grid meter and grid tuning
an oscilloscope, the point of flat -topping can capacitor (C are mounted on the front wall
I
JI
www.americanradiohistory.com
HANDBOOK 432 -MHz 500 -Watt Linear 677
3z 34
..r...
L6
o
-
PLATE LINE (L3)
RADIATOR
,r-INSUL}ATORS
CERAMIC
fl i 1 fIINllllllll//NIIIIIIIIIIII/14
* -20 THREADED ROD
IILL
B II
walli=g7.4
CHIMNEY
E.F JOHNSON 124 -III -1
-E BY SSSON24SCR3EN-GRID
GRID LINE IL2) NoJ FLASHING COPPER
PRIS 72
SURROUNDS TUBE SOCKET;
o CUTOUT ALL SOLDERING MADE
TO THIS PLATE
GRIM
ERENT
0 -50MA
o 0 BLOWER
C4 4 -6-32 SCREWS
- SHIELD PARTITION 0
HOLD SCREEN -GRID
BYPASS AND COPPER
SOLDERING PLATE
Figure 25
a glass piston -type capacitor. The grid line fingers that must be insulated is determined
is made from a piece of copper plate %s -inch by various circuit parameters necessary to
thick which reduces r -f losses and helps obtain complete neutralization, and these
maintain the grid temperature at a safe vary from amplifier to amplifier. In some
level. A combination of fixed and resistor cases, none of the fingers need be insulated.
bias is used in the grid circuit, the fixed In any case, this extra neutralization tech-
bias making certain that the 8122 will be nique is not necessary unless the amplifier
operated within a safe range of anode dis- shows instability after neutralization in the
sipation should the driving power fail. The normal manner.
input coupling link (L,) is series -tuned to The Plate Line-The plate line must main-
reduce reactance and provide optimum cou- tain a low -loss connection with the plate of
pling between the exciter and the amplifier the 8122 to ensure satisfactory performance,
grid circuit. in view of the heavy circulating currents
Cathode pins 1 and 9 of the 8122 socket flowing in this portion of the amplifier. The
are grounded to the chassis. Pin 4 is series - plate line is a half -wavelength strip -line
tuned to ground by a small trimmer capaci- tuned at the open end by a specially con-
tor to provide the required neutralizing structed capacitor. This capacitor (CO is
adjustment. shown in figure 25, and the details of its
The screen (grid 2) of the 8122 is by- construction are outlined in the following
passed to ground at the operating frequency section. Power is transferred to the antenna
by a screen -ring capacitor. An alternative load via a link coupling arrangement shown
method of neutralizing the amplifier is ob- in the illustrations.
tained by eliminating one or more of the
screen -ring contact fingers. This is easily Amplifier The amplifier is mounted in
accomplished by lifting the contact finger Construction two aluminum chassis measur-
from the screen -ring terminal of the tube ing 13" X S" X 3 ". The chassis
and slipping a piece of teflon or polyethylene are fastened back to back to provide sepa-
sheet between the contacts. The number of rate compartments for the grid and plate
678 High- Frequency Power Amplifiers THE RADIO
/- I4 CENTER
LOCATE SERIES
OF D. HOLES
FOR COOLING,
EQUIDISTANT
SPACING
- I16D. HOLE IN
PLATE LINE
CHASSIS
I0 HOLE DRILLED
IN GRID -LINE
CHASSIS FOR
SOCKET
-
DRILL 16 HOLES No. 33 DRILL
CLEARANCE HOLE FOR No. 4 SHEETMETAL SCREW
13
Figure 26
A- Detail of vent and socket holes in plate chassis
I- Detail of top plate for plate -line chassis
(Note: all dimensions in inches)
lines. Thus attached, all mounting holes (C.) . The remaining end is soldered to the
with the exception of the tube socket holes three grid -1 socket pins, as shown in figures
(whose diameters for the two chassis differ) 25 and 27. An aluminum bracket holds the
may be drilled to size. A %S -inch pilot hole grid tuning capacitor (C2) in position. The
is drilled through both chassis to correctly center of C2 and the bushing holding the
center the socket holes for accurate punch- tuning shaft must line up if smooth tuning
ing to final size. After all mounting holes is to be obtained. A shield (figures 25 and
have been drilled, the chassis are separated 30) is placed in the grid compartment to
and holes for the tube socket punched in isolate the grid meter from the r -f field.
each chassis. Figure 26A indicates the sizes Before the tube socket is assembled, all
and locations of the socket holes for both screen contact tabs should be removed from
chassis. the socket -that is, from pins 2, 7, and 10.
All electrical ground connections are The d -c connection to the screen is made in
soldered to a piece of "flashing" copper the plate compartment. Figure 31A shows
which surrounds the base of the tube socket. the method used for making this connec-
The copper is held to the base of the grid - tion.
line compartment by the four 6 -32 screws Details for the construction of the plate
that hold the screen bypass ring, which is line are shown in figures 25 and 28A. The
located in the plate compartment. bracket assembly that guides the 1/4-20
The grid line is held in place by soldering threaded plate capacitor tuning shaft should
one end to the tab of the piston capacitor be constructed close to the dimensions given
www.americanradiohistory.com
HANDBOOK 432 -MHz 500 -Watt Linear 679
iN SHAFT-lOaf
GUSHING
4 4 - z - + 9
i f-
LI
,, LLEAD
CU SCREWAND
DER N PLACE
Tr_ i -
I _
27
I
Figure 27
in figure 29. An improperly constructed frequency for the strip -line with the top
bracket will result in an erratic ground cover in place.
for the plate tuning assembly. The dis- The B -plus r -f choke is connected to the
tance from the plate line to the ground plate line by one of the screws which hold
reference is I1 inches. This distance pro- the plate assembly together. The choke is
vides the correct impedance and resonant connected at the low -voltage point of the
r6 32 CLEARANCE MOLES
A'
}
'
1
j
TAPERED WOOD PLUG
l. FINGER STOCK
SOLDER TO PART A. C0
FLAT-HEAD SwyiO;6-3Y
2Z
.. MATERIAL OPFER
Figure 28
A- Detail of plate line (L.) and assembly
6- Recommended procedure for soldering finger stock to part A' of plate -line assembly.
(Note: 3/i -inch wide beryllium finger stock available from Instrument Specialties Co., Little
Falls, N. J. Stock No. 97 -136).
680 H igh- Frequency Power Amplifiers THE RADIO
DRILL 6 -32
CLEARANCE
AT.9 COPPER
3
SOLDER FINGER STOCK
TO COPPER PLATE
(SEE NOTE) MAT. MICA 0.006 THICK
2
7
CLEARANCE HOLE
-DRILL HOLE TO FIT 10 -32 SCREW
SHAFT LOCK
4
GUIDE BRACKET
f
24
MAT. ALUMINUM
16
Figure 29 L
A- Details of construction of plate tuning capacitor C,.
(Note: Beryllium finger stock available from Instrument Specialties G 24
Co., Little Falls, N. J., Stock No. 97 -115)
S- Details of high -voltage capacitor C. Two sheets of .002 Mylar
may be substituted for the mica dielectric. MAT.
2 BRASS
line. The output -coupling probe is located up for one minute or so before other po-
in this area. The specially constructed high - tentials are applied. Plate voltage should al-
voltage bypass capacitor is made of a 1/32 - ways be applied before screen voltage: never
inch brass plate insulated from the chassis after it.
by a 0.006" thick piece of mica or teflon Neutralizing the Amplifier -The grid and
sheet. plate compartment covers are placed in
The 1/4-inch shaft -lock bushings are used position for the following test. After heater
to guide the shaft for the plate tuning ca- warmup time has been allowed, a small
pacitor and can be adjusted to provide the amount of drive power is applied to the
amount of drag required to maintain a good amplifier without plate or screen potentials
contact. supplied to the tube. Grid current at reso-
Soldering of the finger stock to the plate nance should be about 30 ma for S watts of
assembly can be simplified by use of a tap- drive power. If grid current is low, the fol-
ered wooden plug as shown in figure 28B. lowing adjustments should be made: (1)
The plug will hold the finger stock in place The cathode neutralizing capacitor (Ca)
and prevent excessive heat absorption during should be adjusted for maximum grid cur-
the soldering operation. rent. About half -capacity setting will be
Tuning and Heater power for the 8122 is nearly correct, although further adjustment
Operation 12.5 volts at 1.3 amperes. The may be necessary on application of plate
blower should come on with the and screen voltages. (2) Position the input
heater supply and the tube should be warmed coupling link (L1) for maximum grid cur-
www.americanradiohistory.com
HANDBOOK 432 -MHz 500 -Watt Linear 681
FILE ORAL,
SLOT TO CLEAR
INSULATED LEAD
FOR SCREEN -GRID
CHOKE
SOLDER ,
CHOKE TO
FINGER
- CONTACT FINGERS
Figure 30 RG N.
SOLDER SHIELD
inclosure for the grid line and the right com- VIEW
Figure 32
tion is accomplished by insulating one or
more fingers of the bypass ring from the Cover for plate -line compartment showing
screen -ring terminal of the 8122. Generally, vent holes and location of hole for coupling
loop. Cover is mounted to large top plate and
no more than two fingers have to be insu- attachment holes are slotted so that the cover
lated to obtain complete neutralization. In- and output loop assembly may be moved in
sulation of more than two fingers can result relation to plate line to establish proper de-
gree of coupling. When coupling is established,
in self -oscillation, which will be indicated cover is locked in position with sheet -metal
by excessive and erratic grid current. screws.
682 High- Frequency Power Amplifiers THE RADIO
volts and screen voltage to 200. Plate and utility amplifier provides the experimenter
screen voltage should be applied simulta- with an inexpensive unit which can be
neously, or screen voltage may be applied quickly and easily modified to suit his taste
after plate voltage. The plate capacitor is and pocketbook.
then tuned to resonance and the probe cou- The design and layout of the U -2 ampli-
pling adjusted for maximum power output. fier are such that it may be built in the
Maximum power output does not neces- home workshop with a minimum of tools and
sarily occur at minimum plate current; little complicated metal work. Best of all,
therefore, some form of output power the basic design and assembly remains un-
measuring device is required. changed when various tube types are used.
At resonance with 700 volts on the plate Thus, the builder may start the project at the
and 200 volts on the screen, the power out- 1000 watt PEP level with four inexpensive
put should be about 100 watts. Plate cur- 811A's and later advance to the full legal
rent should be between 260 and 300 ma. PEP level with larger tubes, still making
Grid current should be about 25 to 30 ma. use of the basic U -2 amplifier package.
When the plate voltage is raised to 1500 Of particular interest is the use of the
volts, the power output will rise to about 4X150A family of external -anode tetrodes
235 watts. For a plate potential of 2000 connected as low -p. triodes. Normally un-
volts, power output will be 250 watts for suited for class -B grounded -grid operation,
500 watts input. these compact tubes are connected in a
The amplifier is designed for 50 -ohm out- unique cathode -driven configuration to
put termination and low -loss coaxial line function as semitriodes with the screen ele-
(Foam Heliax) should be used, as common ment of the tube deriving voltage from the
50 -ohm coaxial line (RG -8/U for example) exciting signal.
has prohibitively high losses at this operating
frequency. The Basic The basic circuit of the
Amplifier Circuit U -2 Linear Amplifier is
28 -8 The Utility 2- shown in figure 34A, with
a chart of various tube combinations and
Kilowatt PEP (U-2) their operating parameters given in figure
Linear Amplifier 35. The modifications necessary to adapt
the basic circuit for use with a particular
Described in this section is a general pur- tube type are shown in figure 34B and are
pose utility cathode -driven ("grounded - indicated within the dashed lines of the basic
grid") amplifier. Rated at a maximum input schematic. The utility circuit uses a single
of 2 kilowatts PEP, the U -2 Linear Ampli- tube, such as the zero-bias 3 -1000Z or -
fier can use the following tube combina- 400Z. Substitution of the tetrode 4 -1000A
tions at the discretion of the builder: or 4 -400A ( operated as high - triodes)
Two kilowatt PEP level-One 3- 1000Z, is indicated by the dashed screen grid in the
one 4- 1000A, two 3 -400Z, two 4 -400A, or schematic, which is operated at ground po-
two 4 -250A. 1250 -watt PEP level-Two tential. Socket pins 2, 3, and 4 are grounded
813. 1000 -watt PEP level-One 3 -400Z, in all cases, so tube substitution (4 -1000A
one 4 -400A, one 4 -250A, four 811A, or two for 3 -1000Z and 4 -400A for 3 -400Z) may
4X150A, 4CX250B or 4CX300A connected be accomplished without changing socket
as low -is triodes. connections.
By proper choice of components and cir- Use of two tubes in parallel is shown in
cuitry, the basic configuration of figure 33 figure 34B, the tetrode connection indicated
is used for any of the above listed tube by the dashed screen grids. Low-kt triode
combinations and only sockets, chimneys, connection of the external anode tetrodes is
filament transformer and associated minor shown in circuit 2, and connection for four
components need be changed to adapt the 811A's in parallel is given in circuit 3.
U -2 amplifier to any one of these various The Input Circuit -A tuned -cathode in-
tube combinations. Construction of this put circuit is used to reduce the intermod-
www.americanradiohistory.com
HANDBOOK Utility 2 -KW Linear Amplifier 683
TUNE i
BAND
UTILITY
LINEAR
AMPLIFIEp
yr
Figure 33
VI
SEE ALTERNATE 10
RFC S2
DRAWING 1,283 - 2 C6 T
250 RFC4 00
CIO
.001=
6 N
TO Cg11-3-
EXCITER .001
C6 SMV
B+
L2 L4 L.DOg .01
= C C
01
C
'1 1
Tog
-- LINEAR
ON -OFF
IOW
o o TS
2 3
ON VOX GNO B-
Figure 34
ing high voltage from being applied to the of the high- voltage power supply is thus
plate meter, and removing the grid meter above ground by virtue of the plate meter.
from the r -f grid circuitry. The negative A SO -ohm safety resistor is placed across the
www.americanradiohistory.com
HANDBOOK Utility 2 -KW Linear Amplifier 685
Circuit 2 -Two 4X150A "family" tetrodes is low -.. circuit. PC , PC.-3 turns .16 enam. around
50 -ohm, 2 -watt composition resistor
Circuit 3 -Four 811A. PC thru PC, same as for circuit 2 except coils are 4 turns
Sockets and chimneys -For 3 -1000Z use SK -510 socket and SK -516 chimney. For 4 -1000A use
SK -510 socket and SK -506 chimney. For 3 -4002 use SK -410 socket and SK -416 chimney. For
4 -400A use SK -410 socket and 5K -406 chimney. For 4X150A or 4CX2508 use 5K -640 socket
and SK -606 chimney. For 4CX300A use SK -770 socket (with integral chimney).
meter circuit to protect the operator in the Amplifier the U -2 linear amplifier is
unlikely case a meter coil is defective. Construction built on a 10" X 17" X 4"
The Plate Circuit -A
conventional pi- aluminum chassis and has a
network plate circuit is employed, making 10" high aluminum panel. The panel is cut
use of readily available components. The Q to fit the available cabinet. Construction of
of the plate tank circuit varies under load a suitable TVI -proof inclosure is discussed
from approximately 10 at 80 meters to over in the chapter "Workshop Practice." A bot-
20 at 10 meters, depending on the combina- tom plate is bolted to the chassis to pres-
tion of tubes in use. Plate- circuit efficiency, surize it and to provide proper shielding. A
however, remains good in all cases. The com- forced -air cooling system is required for all
plete circuit is designed to match the ampli- tube combinations except the 811's and
fier to a 50-ohm antenna system having a the 813's. The squirrel -cage blower is
SWR of 2:1 or less. An RL parasitic sup- mounted on the rear apron of the chassis
pressor is placed in the plate lead of each and forces the air into the chassis inclosure,
tube to suppress any tendency toward vhf through the air socket(s) and past the en-
oscillation. Neutralization is not required, velope and plate seals of the tube(s).
and the U-2 linear amplifier remains stable If changes in tube types are contem-
over the operating range of 3.5 to 29.7 plated, it is wise to make a removable plate
MHz. for the tube sockets as is done in this unit.
686 High- Frequency Power Amplifiers THE RADIO
Figure 35
TUBE CHART FOR U -2 LINEAR AMPLIFIER
Typical operating parameters are given for twelve tube combinations. Zero-signal plate cur-
rent varies with no -load plate voltage, rising as plate voltage rises. The indicated values are
measured with no cathode bias (relay RY, closed).
The replaceable socket -mounting plate filament transformer and tube sockets. The
measures 10" X 7" and is fixed in place two current meters are mounted in match-
with 6 -32 bolts. However, if one particular ing holes drilled through panel and chassis,
tube or combination of tubes making use and the plate- circuit loading capacitor (C7)
of a common socket is to be used, it is is placed at the opposite end of the chassis
easier and quicker to drill the main chassis from the meters, beneath the main band-
for the required sockets and to eliminate switch assembly. This capacitor is mounted
the special plate. on aluminum brackets placed at the front
The plate -circuit tuning capacitor (CO and rear of the frame. Immediately behind
is centered on the chassis and mounted on the loading capacitor is the filament trans-
2 -inch high aluminum brackets which posi- former, mounted on its side by small alum-
tion the capacitor shaft at the same height inum brackets affixed to the core bolts. The
as the shaft of the main bandswitch as- transformer is positioned after the cathode
sembly. The bandswitch assembly is placed tank assembly is wired and the inclosure is
to one side of the chassis and drives the bolted in place.
cathode bandswitch (S,) by means of an The grid terminals of the tube socket(s)
insulated, flexible coupling and a right - arc grounded to the chassis by means of
angle gear unit. The cathode tank circuit short lengths of copper strap. In the case
is mounted in an inclosure beneath the chas- of the SK -410 and SK-510 sockets, the
sis, as is shown in figure 37. straps pass through slots in the socket wall
Placement of the under -chassis compo- and are soldered directly to the socket pins.
nents is conventional, and is shown in the Common ceramic sockets are not recom-
photographs. (figure 38A and B). The main mended for the 3 -400Z or the 3 -1000Z as
components remain the same for all tube the flow of cooling air is impeded and ex-
combinations, with the exception of the cessive lateral pressure is exerted on the
www.americanradiohistory.com
HANDBOOK Utility 2 -KW Linear Amplifier 687
Figure 36
www.americanradiohistory.com
HANDBOOK Utility 2 -KW Linear Amplifier 689
`*'
Figure 38
not be operated in the common class -B the output power of the amplifier. This
grounded -grid circuitry since the abnormally drive level is well within the capability of
high levels of grid current which occur may most modern SSB exciters or transceivers.
be destructive to the tube. It is possible and Because the screen of the tetrode is grounded
practical, however, to achieve good perform- in this unique circuit, the inexpensive air-
ance and low values of grid current by con- system sockets having no screen bypass
necting these tubes as a form of low - t capacitor may be used. The use of a ceramic
triode with the control grid strapped to the !octal (receiving type) socket for the
cathode (figure 40). The control grid still
has some effect on the electron stream, since
4X 1 f 0A or 4CX2 f 0B is not recommended
zero -signal resting plate current of this mode since the base temperature of the tube can-
of operation is low. Feedthrough power is not be controlled by application of cooling
quite high and, as a result, the drive level air. A high -speed (6000 r.p.m.) blower
of this mode of operation is about 80 to 90 should be used to supply the required volume
watts per tube, most of which appears in of air through the socket at the back -pres-
690 High- Frequency Power Amplifiers THE RADIO
Figure 39
www.americanradiohistory.com
HANDBOOK Utility 2 -KW Linear Amplifier 691
www.americanradiohistory.com
HANDBOOK Tribander Amplifier 693
cuit as is done in the common grounded - The Tribander linear amplifier construc-
grid configuration. Grid and screen ele- tion is novel in that no regular chassis deck
ments are at r-f ground, while normal is employed. The amplifier is built within an
Class -AB, grid -bias and screen potentials are inclosure made up of two aluminum chassis,
applied to the tubes. Under these condi- each measuring 10" X 14" X 3". One
tions, the power gain of the 4CX300A is chassis is inverted and serves as a pan with-
quite high; approximately 30 watts PEP in which the components are mounted. The
drive being required for full output. second chassis is placed atop the first and
The amplifier plate circuit is a simple serves as a top shield cover. This chassis as-
three -band pi- network, designed for a cir- sembly is hinged along the rear edge, and
cuit Q of 15. Since the low- frequency bands opens up much in the manner of a suitcase.
are not included, only two small self -sup- A single -piece front panel made of alumi-
porting air -wound coils are required. In ad- num is fixed to the lower chassis. The front
dition, the size of the pi- network loading apron of the top section is cut away to pro-
capacitance is considerably smaller than a vide clearance for the meters, switches and
capacitor necessary for all -band operation. capacitors. When the top section is closed,
The amplifier is controlled by a two -deck the cabinet is sealed by a strip of finger
progressively-shorting switch (S,) that re- stock that runs around the inside edges of
motely controls the auxiliary equipment and the lower chassis box. A length of "piano -
provides the operator with a choice of type" hinge fastens the rear edges of the
"tune" or "operate" modes. All control and two chassis together, and the inclosure
low- voltage power leads are suitably filtered halves are held in place by five panel bolts
by LC networks to suppress radiation of which screw into nut plates riveted to the
TVI- producing harmonics. lip of the lid, or top section.
Figure 42
TRIBANDER LINEAR
AMPLIFIER FOR
10 -15 -20 -METER
SIDEBAND
This one kilowatt PEP
linear amplifier is de-
signed for those ama-
teurs interested in the
higher - frequency DX
bands. Using two 4CX-
300A tubes, this com-
pact bandswitching unit
is ideally suited for ex-
citers having a PEP out-
put of about 30 watts.
Panel controls are 1. to
r.,: Screen meter, plate
meter, plate tuning,
plate loading. On the
left is the mode switch
(S ) and on the right
is the band switch (S.). o
Amplifier is mounted on
four rubber "feet" so
that cooling air may be
drawn from under the
cabinet. Geared tuning
dials, switch knobs, and
plate bandswitch are
salvaged from surplus
"TU" tuning drawers
from BC- 191/375 trans -
mitter.
694 High- Frequency Power Amplifiers THE RADIO
EXC. R.P.
INPUT RY1 OUTPUT
RFC
Bt2 XV. 0 SOO
IDD
MU/ =ID RV RFC2
Su 2 ] S _ - r
EE RFC]
its V. 'U 1 0--
T
1TtT`
RFC,
T IH
C+ sa C2
+lH
ANT. RELAY
T T
3 RFC,
B+ SCR.
T
4 RFC,
slAS
R FC3
1,5V.1,e2S0 `r AT EACH
SOCKET
SIB
os+
AUX^ s RFC,
CONTROL o- T
SWITCH , NOTE: I. CAPACITORS C* ARE .007NF.. 100 V..,
TUNE 0/SC CERAMIC.
I. OPP
CONTROL
T T NS - I Z. AUX. 2. ONE CAPACITOR C
ON EACH CR /O
GN0.0--ill
Figure 43
An aluminum partition divides the inte- function switch, and power lead filters. The
rior of the inclosure into two compartments larger compartment contains the two
(figure 44). The smaller compartment con- 4CX300A tubes, the plate circuit pi -net-
tains the blower motor, filament transform- work components and the antenna relay.
er, panel meters, auxiliary control relay, The partition is shaped to fit around the
www.americanradiohistory.com
HANDBOOK Tribonder Amplifier 695
1111r
Figure 44
Ilr
INTERIOR VIEW OF LINEAR AMPLIFIER
Antenna relay Rye is placed in small aluminum box mounted to rear wall of cabinet directly
behind antenna loading capacitor. The two 4CX300A tube sockets are mounted on top of
aluminum shield can taken from oscillator coil section of surplus "command" transmitter.
Micro -switch on partition removes high voltage when cover is opened. Midget relay adjacent
to switch is added for auxiliary control circuits and is not required. At extreme left rear are
feedthrough capacitors mounted on aluminum plate, with r -f chokes beneath them. Filament
transformer is in corner of compartment, in back of mode selector switch. Pi- network
components are at right, with three plate blocking capacitors mounted to aluminum strip
supported by plate tank capacitor.
housing holding the tetrode tube sockets. directly to the bottom chassis "pan." The
As the standard air -system socket with pi- network capacitors and bandswitch are
built -in screen bypass capacitor is both ex- panel mounted, and the remaining com-
pensive and bulky, the smaller phenolic partment area is taken up by the plate coils,
socket having no screen capacitor was used r -f choke, and the plate blocking capacitors.
as an inexpensive substitute. Two of these Antenna relay RY, is mounted in a small
sockets will mount atop an oscillator shield aluminum shield box placed at the back of
can taken from a defunct surplus "Com- the compartment.
mand" transmitter. The can makes an in- Transmitter wiring is simple and straight-
expensive and r -f tight shield for the grid forward. All connections in the meter com-
and cathode components, and is mounted partment are made with unshielded wire.
696 High- Frequency Power Amplifiers THE RADIO
T
1 +2XV.
41 CHI RY2A
et SCR. (390V.)
1113)
20
SOO
VR105
IIF--o GNO.
(119)
eIAs
2 RY CR /0 (14)
IE TER
RESISTOR
1T
T 150
0115 V.111
p
Lo(
RYas (02)
ANT. RELAY
15 V.12
(O
TO RI. CNT.0 -AUX CONTROL
VOX- MANUAL RCCCIVCR
SWITCH (99)
TUNO CONTROL
3 7 3 7 3 7 (117)
TO VOX
CONTROL RELAY
[RATE' CONTROL
CONTACTS
Figure 45
www.americanradiohistory.com
HANDBOOK Tribander Amplifier 697
cuit is completed which can turn on the the high -voltage and bias supply energized.
station receiver or sideband exciter. The Using a voltmeter, the potentiometer should
third position turns on the amplifier tube
filaments and energizes the blower motor
be set to provide about - 60 volts on the
arm. The voice relay is energized dropping
to cool the tubes. Cutoff bias is applied to the cutoff bias out and the potentiometer
the tubes to eliminate diode noise often is carefully reset to provide a static plate
noticed in standby operation. The fourth current of 200 ma as read on the meter. In-
position applies full plate voltage and re- dicated screen current (bleeder current)
duced screen voltage to the amplifier for should be about 22 ma. When the voice
tuning operations, and the fifth switch posi- relay drops out, the plate current should
tion applies full screen voltage. Cutoff bias fall to zero.
is removed by the voice- actuated relay in The amplifier is now fed a small exciting
the power supply. Screen and plate currents signal (single tone) and tuned and loaded
are continually monitored by the two panel for a maximum plate current of 500 mil-
meters. The screen meter is recalibrated to liamperes. Screen current should now be
have an elevated zero point and reads -20 approximately 30 ma. (This is a total of
to + 30 milliamperes. Under certain con- screen and bleeder current.) The output
ditions, negative screen current can flow coupling is now increased slightly so that
and it is important to monitor this sensitive r -f output (as read on an r -f ammeter, or
indicator of amplifier operation. output voltmeter) drops about 2 percent.
The power -supply schematic is shown in Maximum linearity is obtained when the
figure 45. The high voltage supply uses amplifier is slightly overcoupled. Under
3B28 "hash- free" gas rectifier tubes and voice conditions, plate- current peaks should
provides 2000 volts d.c. at 500 ma and reach approximately 250 ma, as read on the
regulated 360 volts at 30 milliamperes. meter. No grid current should be read on
"Jumpers" in the base of the regulator tubes a 0 -1 d -c milliammeter placed across the
are wired in series with the primary relay grid- current terminals in the power supply.
circuit so that the supply cannot be ener- Any flicker of grid current indicates the
gized unless the tubes are in their sockets. amplifier is being overdriven, with a con-
A smaller half -wave semiconductor supply sequent severe increase in distortion. Under
provides operating and cutoff bias for the voice conditions, indicated screen current
amplifier. The bias relay may be actuated by will be relatively constant, since actual cur-
the voice circuit of the exciter to drop the rent drawn by the screen of the tubes will
bias to the correct amount during the time be less than + or - 10 ma., and this
small value is swamped out by the bleeder
the voice circuit is energized.
current, which is constant at 22 ma. Low
values of screen meter current (indicating
Transmitter The only initial adjustment is that the tubes are drawing negative cur-
Adjustment to set the operating bias level rent) indicates excessive loading; high values
and Tuning by means of the potentiometer. of screen current indicate insufficient plate
Initially, the arm should be set circuit loading.
at the high -potential end of the potentiom- Never apply excitation to this (or any
eter to apply full bias to the tubes. The other) grounded -grid amplifier without all
filaments and blower are turned on, and operating potentials applied to the tubes.
CHAPTER TWENTY-NINE
Speech and
Amplitude -Modulation Equipment
Amplitude modulation of the output of a percent of the d -c input to the final stage.
transmitter for radiotelephony may be ac- A detailed discussion of the relative advan-
complished either at the plate circuit of the tages of the different methods for accom-
final amplifier, commonly called high-level plishing amplitude modulation of the output
amplitude modulation or simply plate modu- of a transmitter is given in an earlier chap-
lation of the final stage, or it may be accom- ter.
plished at a lower level. Low -level modula- Two trends may be noted in the design of
tion is accompanied by a plate- circuit systems for obtaining high -level amplitude
efficiency in the final stage of 30 to 45 per- modulation of the final stage of amateur
cent, while the efficiency obtainable with transmitters. The first is toward the use of
high -level amplitude modulation is about tetrodes in the output stage of the high -
twice as great, running from 60 to 80 per- power audio amplifier which is used as the
cent. Intermediate values of efficiency may modulator for a transmitter. The second
be obtained by a combination of low -level trend is toward the use of a high -level splat-
and high -level modulation; cathode modula- ter suppressor in the high -voltage circuit
tion of the final stage is a common way of between the secondary of the modulation
obtaining combined low -level and high -level transformer and the plate circuit of the
modulation. modulated stage.
High -level amplitude modulation is char-
acterized by a requirement for an amount of 29 -1 Modulation
audio power approximately equal to one -
half the d -c input to the plate circuit of the Tetrodo In regard to the use of tetrodes,
final stage. Low -level modulation, as for Modulators the advantages of these tubes
example grid -bias modulation of the final have long been noted for use in
stage, requires only a few watts of audio modulators having from 10 to 100 watts
power for a medium -power transmitter and output. The 6V6, 6L6, and 807 tubes have
10 to 15 watts for modulation of a stage served well in providing audio power out-
with one kilowatt input. Cathode modula- puts in this range. Recently the higher -
tion of a stage normally is accomplished power tetrodes such as the 4 -65A, 813,
with an audio power capability of about 20 4 -125A, and 4 -250A have come into more
www.americanradiohistory.com
Modulation 699
general use as high -level audio amplifiers. the radiated signal is concerned, at least so
The beam tetrodes offer the advantages of long as the linear modulation capability of
low driving power (even down to zero the final amplifier is not exceeded. So the
driving power for many applications) as problem becomes mainly one of constructing
compared to the moderate driving power a modulator /final- amplifier combination so
requirements of the usual triode tubes hav- that negative -peak clipping (modulation in
ing equivalent power-output capabilities. excess of 100 percent in a negative direc-
On the other hand, beam -tetrode tubes tion) cannot normally take place regardless
require both a screen -voltage power supply of any reasonable speech input level.
and a grid -bias source. So it still is expedient
Asyymetricol The speech waveform of the
in many cases to use zero -bias triodes or
even low -ic triodes such as the 304TL in Speech normal male voice is charac-
many modulators for the medium -power terized, as was stated before,
and high -power range. A list of suggested by high-amplitude peaks of short duration.
modulator combinations for a range of pow- But it is also a significant characteristic of
er output capabilities is given in conjunction this wave that these high -amplitude peaks
with several of the modulators to be de- are polarized in one direction with respect to
scribed. the average amplitude of the wave. This is
the "lopsided" or assymetrical speech which
Increasing the It has long been known has been discussed and illustrated in an
Effective Modu- that the effective modu- earlier chapter.
lotion Percentage lation percentage of a The simplest method of attaining a high
transmitter carrying un- average level of modulation without nega-
altered speech waves was necessarily limited tive -peak clipping may be had merely by
to a rather low value by the frequent high - ensuring that these high- amplitude peaks
amplitude peaks which occur in a speech always are polarized in a positive direction at
waveform. Many methods for increasing the the secondary of the modulation transform-
effective modulation percentage in terms of er. This adjustment may be achieved in the
the peak modulation percentage have been following manner: Couple a cathode -ray
suggested in various publications and subse- oscilloscope to the output of the transmitter
quently tried. in the field by the amateur in such a manner that the carrier and its
fraternity. Two of the first methods sug- modulation envelope may be viewed on the
gested were automatic modulation control scope. Speak into the microphone and note
and volume com pression. Both these meth- whether the sharp peaks of modulation are
ods were given extensive trials by operating polarized upward or whether these peaks
amateurs; the systems do give a degree of tend to cut the baseline with the "bright
improvement as evidenced by the fact that spot" in the center of the trace which de-
such arrangements still are used in many notes negative -peak clipping. If it is not
amateur stations. But these systems fall far obvious whether or not the existing polarity
short of the optimum, because there is no is correct, reverse the polarity of the mod-
essential modification of the speech wave- ulating signal and again look at the envel-
form. Some method of actually modifying ope. Since a push -pull modulator almost in-
the speech waveform to improve the ratio variably is used, the easiest way of reversing
of peak amplitude to average amplitude must signal polarity is to reverse either the leads
be used before significant improvement is which go to the grids or the leads to the
obtained. plates of the modulator tubes.
It has been proved that the most serious When the correct adjustment of signal
effect on the radiated signal accompanying polarity is obtained through the above pro-
overmodulation is the strong spurious -side- cedure, it is necessarily correct only for the
band radiation which accompanies negative - specific microphone which was used while
peak clipping. Modulation in excess of 100 making the tests. The substitution of an-
percent in the positive direction is accom- other microphone may make it necessary
panied by no undesirable effects as far as to reverse the polarity, since the new
700 Speech and A -M Equipment THE RADIO
R-F FINAL
microphone may be connected internally in MODULATOR
Figure 2
www.americanradiohistory.com
HANDBOOK Design 701
called high -level splatter suppressor. As fig- vacuum rectifier and utilizes a large -size
ure 1 shows, it is only necessary to add a heater -type dual cathode requiring a warm -
high- vacuum rectifier tube socket, a fila- up time of at least 40 seconds before cur-
ment transformer, and a simple low -pass rent should be passed. The tube is rated at
filter to an existing modulator /final -ampli- an average current of 250 ma. For greater
fier combination to provide high -level sup- current drain by the final amplifier, two or
pression. more 836 tubes may be placed in parallel.
The tube (V1) serves to act as a switch The filament transformer for the cathode
to cut off the circuit from the high -voltage of the splatter- suppressor tube must be in-
power supply to the plate circuit of the sulated for somewhat more than twice the
final amplifier as soon as the peak a -c volt- operating d -c voltage on the plate -modu-
age across the secondary of the modulation lated stage, to allow for a factor of safety
transformer has become equal and opposite on modulation peaks. A filament transform-
to the d -c voltage being applied to the er of the type normally used with high -
plate of the final amplifier stage. A single - voltage rectifier tubes will be suitable for
section low -pass filter serves to filter out the such an application.
high- frequency components resulting from
the clipping action.
Tube V, may be a receiver rectifier with 29 -2 Design of Speech
a 5 -volt filament for any but the highest Amplifiers and Modulators
power transmitters. The 5Y3 -GT is good
for 125 ma plate current to the final stage, A number of representative designs for
the 5R4 -GY and the 5U4 -G are satisfactory speech amplifiers and modulators are given in
for up to 250 ma. For high -power high - this chapter. Still other designs are included
voltage transmitters the best tube is the in the descriptions of other items of equip-
high- vacuum transmitting tube type 836. ment in other chapters. However, those per-
This tube is equivalent in shape, filament sons who wish to design a speech amplifier
requirements, and average- current capabil- or modulator to meet their particular needs
ities to the 866A. However, it is a high- are referred to Chapter Six, Vacuum -Tube
CRYSTAL
MIC.
J, 4.714
^ dd
4 \
103/F 1 m
4.714 tOUF
450
B
- ,:ROUND
-6 3v
r 6 3 V
when an audio power out- 6L6 AB1 360 270 -23 6,600 85-135 27
put of 10 to 120 watts is required of a
modulator. A pair of 6L6 tubes operating 616 AB2 360 270 -23 3.800 85 -205 47
in such a modulator will deliver good plate - 807 AB1 600 300 -34 10,000 35-140 56
circuit efficiency, require only a very small 807 AB, 750 300 -35 12,000 30-140 75
amount of driving power, and impose no
serious grid -bias problems. 807 AB2 750 300 -35 7,300 30.240 120
www.americanradiohistory.com
HANDBOOK General Purpose Modulator 703
V2
4.7K tOUF
450
-T 2 + C 1"--BIS
9 1000-1250
SUPPLY IF USED,
OTNERW/SE SNORT
V.
TERMINALS.
CHI
i VVb
A
5Y3 -GT +JOv
S Y Y
* = MATCNED PAIR
RESISTORS, 1 Rb
TOUF
1S0
TOUF 430
50LF T3 T4
150
1000
115 1,
Figure 5
84-400
6,700
8,200
300
370
should be adjusted to show -
75 volts from
each 810 (if 810's are used) grid terminal
810 2000 -50 60-420 12,000 450
to ground as measured with a high -resistance
810 2500 -75 50-420 17,500 500 voltmeter. If an oscilloscope is available, it
8005 1500 -67 40.330 9,800 330 should be coupled to point "A" on the load
704 Speech and A -M Equipment THE RADIO
www.americanradiohistory.com
HANDBOOK Clipper -Amplifier 705
47 100K
1W 1
fe.3v.
330
I
TO
CLASS -B
MODULATOR
UNLESS OTHERWISE INDICATED;
1
GRID CIRCUIT
EA.
8450V .
MAX.
ALL RESISTORS 0.3 WATT
ALL CAPACITORS IN 1./F
RESISTORS MARKED WITH ASTER/SK W
ADJUST ARE BALANCED PAIRS.
CLIPPING
-ee v
5U4-G
FILS
T1- 430 -0 -450 VOLTS AT 105 MA.
CHICAGO TRANSFORMER PSR -103 22K,2
12- CLASSE DRIVER TRANSFORMER 22K,2W
CHICAGO TRANSFORMER CDS -I EA.
230
40
V.
3
T3 -125 V. AT 15 MA.
STA PICOR AS-04/3 so
t W.
LPF -2 -LOW PASS FILTER UNIT
CHICAGO TRANS. LPF -2 SR
+
L_,- 12 N, 130 M A
.
GN0. 0
Figure 8
volts peak audio signal under maximum clip- that the heat of the soldering iron does not
ping conditions. A double -triode 12AU7 cause the resistors to shift value. The re-
cathode- follower phase inverter follows the sistors should be held firmly by the lead to
clipper stage and delivers a 100 -volt rms be soldered with a long -nose pliers, which
signal to the push -pull grids of the 2A3 will act as a heat sink between the soldered
audio driver tubes. The 2A3 tubes operate joint and the body of the resistor. If this
at a plate potential of 330 volts and have a precaution is taken the two phase- inverter
-68 volt bias voltage developed by a small outputs will be in close balance.
diode rectifier supply applied to their grid
circuit. An audio output of 15 watts is de- Adjustment of When
the wiring of the
veloped across the secondary terminals of the Speech amplifier has been
speech
the class -B driver transformer with less than Amplifier completed and checked, the
5 percent distortion under conditions of no unit is ready to be tested.
clipping. A 5U4 -G and a choke -input filter Before the tubes are plugged in the amplifier,
network provide unusually good voltage reg- the bias supply should be energized and the
ulation of the high- voltage plate supply. voltage across the 600 -ohm bleeder resistor
The resistors in the 12AU7 phase -in- should be measured. It should be -68 volts.
verter plate circuit and the grid circuit of If it is not, slight changes in the value of
the 2A3 tubes should be matched to achieve the series resistor (R:1) should be made until
best phase- inverter balance. The exact value the correct voltage appears across the bleeder
resistor. The tubes may now be inserted in
of the paired resistors is not important, but
the amplifier and the positive and cathode
care should be taken that the values are
voltages checked in accordance with the
equal. Random resistors may be matched on
measurements given in figure 8. After the
an ohmmeter to find two units that are alike unit has been tested and is connected to the
in value. When these matched resistors are modulator, R_ should be set so that it is
soldered in the circuit, care should be taken impossible to overmodulate the transmitter
706 Speech and A -M Equipment
803's
Ti 2
6 -WATT
Zs'o 25K
To To
DRIVER SPEECH MODULATOR
STAGE AMP. LOAD
www.americanradiohistory.com
CHAPTER THIRTY
In view of the high cost of iron -core com- many cases compromises will be involved,
ponents such as go to make up the bulk of a particularly when the power supply is for an
power supply, it is well to consider carefully amateur station and a number of compo-
the design of a new or rebuilt transmitter in nents already on hand must be fitted into
terms of the minimum power supply require- the plan. As much thought and planning
ments which will permit the desired per- should be devoted to the power -supply com-
formance to be obtained from the transmit- plement of an amateur station as usually is
ter. Careful evaluation of the power supply allocated to the r -f and a -f components of
requirements of alternative transmitter ar- the station.
rangements will permit the selection of that The arrival at the design for the power
transmitter arrangement which requires the supply for use in a particular application
minimum of power supply components, and may best be accomplished through the use
which makes most efficient use of such pow- of a series of steps, with reference to the
er supplies as are required. data in this chapter by determining the
values of components to be used. The first
step is to establish the operating require-
30 -1 Power-Supply ments of the power supply. In general these
Requirements are:
A power supply for a transmitter or for a 1. Output voltage required under full
unit of station equipment should be designed load.
in such a manner that it is capable of de- 2. Minimum, normal, and peak output
livering the required current at a specified current.
voltage, that it has a degree of regulation 3. Voltage regulation required over the
consistent with the requirements of the ap- current range.
plication, that its ripple level at full current 4. Ripple voltage limit.
is sufficiently low for the load which will be
S. Rectifier circuit to be used.
fed, that its internal impedance is sufficient- The output voltage required of the power
ly low for the job, and that none of the supply is more or less established by the
components shall be overloaded with the operating conditions of the tubes which it
type of operation contemplated. will supply. The current rating of the sup-
The meeting of all the requirements of ply, however, is not necessarily tied down by
the previous paragraph is not always a a particular tube combination. It is always
straightforward and simple problem. In best to design a power supply in such a man-
708 Power Supplies THE RADIO
Figure 1
i
25,00
A well- designed supply con-
trol panel has separate pri-
mary switches and indicator
lamps for the filament and
ON
4000 -- - 3000
ON
ner that it will have the greatest degree of of inductance which the input choke must
flexibility; this procedure will in many cases have to keep the voltage from soaring when
allow an existing power supply to be used the external load is removed.
without change as a portion of a new trans- The normal current rating of a power
mitter or other item of station equipment. supply usually is a round -number value
So the current rating of a new power supply chosen on the basis of the transformers and
should be established by taking into consider- chokes on hand or available from the catalog
ation not only the requirements of the tubes of a reliable manufacturer. The current
which it immediately will feed, but also with rating of a supply to feed a steady load such
full consideration of the best matching of as a receiver, a speech amplifier, or a con-
power-supply components in the most eco- tinuously operating r -f stage should be at
nomical current range which still will meet least equal to the steady drain of the load.
the requirements. It is often long -run econo- However, other considerations come into
my, however, to allow for any likely addi- play in choosing the current rating for a
tional equipment to be added in the near keyed amplifier, an amplifier of SSB signals,
future. or a class -B modulator. In the case of a
supply which will feed an intermittent load
Current- Rating The minimum current drain such as these, the current ratings of the
Considerations which will be taken from a transformers and chokes may be less than
power supply will be, in the maximum current which will be taken;
most cases, merely the bleeder current. There but the current ratings of the rectifier tubes
are many cases where a particular power to be used should be at least equal to the
supply will always be used with a moderate maximum current which will be taken. That
or heavy load on it, but when the supply is is to say that 300 -ma transformers and
a portion of a transmitter it is best to con- chokes may be used in the supply for a mod-
sider the minimum drain as that of the ulator whose resting current is 100 ma but
bleeder. The minimum current drain from whose maximum current at peak signal will
a power supply is of importance since it, rise to 500 ma. However, the rectifier tubes
in conjunction with the nominal voltage of should be capable of handling the full 500
the supply, determines the minimum value ma.
www.americanradiohistory.com
HANDBOOK Power -Supply Requirements 709
The iron -core components of a power changes in a -c line voltage must be held to
supply which feeds an intermittent load may a few volts at the output terminals.
be chosen on the basis of the current as These three problems are solved in the
averaged over a period of several minutes, normal type of installation in quite different
since it is heating effect of the current which manners. The high -power case where output
is of greatest importance in establishing the voltage must be held to within 10 to 15
ratings of such components. Since iron -core percent is normally solved by using the
components have a relatively large amount proper value of inductance for the input
of thermal inertia, the effect of an intermit- choke and proper value of bleeder at the
tent heavy current is offset to an extent by output of the power supply. The calculations
a key -up period or a period of low modula- are simple: the inductance of the power -
tion in the case of a modulator. However, supply input choke at minimum current
the current rating of a rectifier tube is drain from the supply should be equal in
established by the magnitude of the emission henrys to the load resistance on the supply
available from the filament of the tube; the (at minimum load current) divided by
maximum emission must not be exceeded 1000. This value of inductance is called the
even for a short period or the rectifier tube critical inductance and it is the minimum
will be damaged. The above considerations value of inductance which will keep the
are predicated, however, on the assumption output voltage from soaring in a choke -
that none of the iron -core components will input power supply with minimum load on
become saturated due to the high intermit- the output. The minimum load current may
tent current drain. If good quality com- be that due to the bleeder resistor alone, or
ponents of generous weight are chosen, it may be due to the bleeder plus the mini-
saturation will not be encountered. mum drain of the modulator or amplifier to
which the supply is connected.
Voltage The general subject of voltage The low- voltage low-current supply, such
Regulation regulation can really be divided as would be used for a vfo or the high -fre-
into two sub -problems, which quency oscillator in a receiver, usually is
differ greatly in degree. The first, and more regulated with the aid of glow- discharge
common, problem is the case of the normal gaseous -regulator tubes. These regulators are
power supply for a transmitter modulator, usually called VR. tubes. Their use in various
where the current drain from the supply types of power supplies is discussed in Sec-
may vary over a ratio of four or five to one. tion 30 -12. The electronically regulated
In this case we desire to keep the voltage power supply, such as is used in the 10- to
change under this varying load to a matter 100 -watt power output range, also is dis-
of 10 or 15 percent of the operating voltage cussed in this chapter.
under full load. This is a quite different
problem from the design of a power supply Ripple The ripple- voltage limitation
to deliver some voltage in the vicinity of Considerations imposed on a power supply
250 volts to an oscillator which requires is determined by the load
two or three milliamperes of plate current; which will be fed by the supply. The toler-
but in this latter case the voltage delivered able ripple voltage from a supply may vary
to the oscillator must be constant within a from perhaps 5 percent for a class -B or class -
few volts with small variations in oscillator C amplifier which is to be used for a c -w
current and with large variations in the a -c stage or amplifier of an f -m signal down to
line voltage which feeds the oscillator power a few hundredths of one percent for the
supply. An additional voltage -regulation plate -voltage supply to a low -level voltage
problem, intermediate in degree between amplifier in a speech amplifier. The usual
the other two, is the case where a load must value of ripple voltage which may be toler-
be fed with 10 to 100 watts of power at ated in the supply for the majority of stages
a voltage below 500 volts, and still the volt- of a phone transmitter is between 0.1 and
age variation with changes in load and 2.0 percent.
710 Power Supplies THE RADI O
FULL. WAVE RIPPLE IN TERMS OF C AT FULL LOAD
In general it may be stated that, with TO
RECTIFIER
CAPACITANCE, C PERCENT RIPPLE
60 -Hz line voltage and a single -phase recti- 2 L1 13.1
resistance- capacitance filters is given in a less self -evident that the better the regula-
following paragraph. tion of such a supply, the better will be its
www.americanradiohistory.com
HANDBOOK Power -Supply Requirements 711
8 1 V 12 Mv
tion is determined by the above formula.
TOFULL-WAVE
RECTIFIER Then this percentage is multiplied by the
filter reduction factor of the following sec-
tion of filter. This reduction factor is deter-
mined through the use of the following
Figure 3 formula:
SAMPLE FILTER FOR
CALCULATION OF RIPPLE
Filter reduction factor - LC-1
1.76
(2)
www.americanradiohistory.com
HANDBOOK Rectification Circuits 713
O HALF-AND FULL -VOLTAGE BRIDGE POWER SUPPLY O TWO- VOLTAGE BRIDGE POWER SUPPLY
4-
Eoo.
al
Ti
o1Ef-E
EO TWO -VOLTAGE POWER SUPPLY pF SPECIAL FILTER CIRCUIT FOR BRIDGE RECTIFIER
Figure 6
monly used in a -c power supplies are of the peak -inverse -voltage rating may be used for
low-pass type. a given voltage output.
Note that the center of the high -voltage
Bridge The bridge rectifier (figure winding of the bridge transformer (figure
Rectification 4C) is a type of full -wave 4C) is not a ground potential. Many trans-
circuit in which four rectifier formers having a center -tapped winding are
elements or tubes are operated from a single not designed for bridge service as the insula-
high- voltage winding on the power trans- tion between the center -tap point and
former. ground is inadequate. Lack of insulation at
While twice as much output voltage can this point does no harm in a full -wave cir-
be obtained from a bridge rectifier as from cuit, but may cause breakdown when the
a center -tapped circuit, the permissible out- transformer is used in bridge configuration.
put current is only one -half as great for a
given power transformer. In the bridge cir-
cuit, four rectifiers and three filament -heat- 30 -3 Standard Power -
ing transformer windings are needed, as Supply Circuits
against two rectifiers and one filament wind-
ing in the center -tapped full -wave circuit. Choke input is shown for all three of the
In a bridge rectifier circuit, the inverse -peak standard circuits of figure 4, since choke
voltage impressed on any one rectifier tube is input gives the best utilization of rectifier -
halved, which means that tubes of lower tube and power- transformer capability, and
www.americanradiohistory.com
HANDBOOK Standard Circuits 715
PRIMARY
Eo+.17Es
Is0.5nIDC.
RIPPLE FREQUENCY 3F
RIPPLE PERCENT 19.3
PEAR INVERSE 2.09 EO
TUSE VOLTAGE 2.44 Es
OA 3 -PHASE STAR
1DC. +Eo
Figure 7
COMMON
POLYPHASE -
RECTIFICATION
PRIMARY Eo 1.55 Es
Is 0.409 I O.C. CIRCUITS
RIPPLE FREQUENCY 9 F
RIPPLE PERCENT 4.2 These circuits are used
PEAR INVERSE 2.09 Eo when polyphase power Is
2.113 Es
TUBE VOLTAGE available for the plate
supply of a high -power
transmitter. The circuit
at B is also called a
OB 6 -PHASE STAR t hree -phase full -wave
+Eo rectification system. The
circuits are described in
the accompanying text.
PRIMARY Eo2.34Es
Is 0.1110 t D.C.
RIPPLE FREQUENCY SF
RIPPLE PERCENT 4.2
KAR INVERSE 1.05 EO
TUBE VOLTAGE - 2.44 Es
6 -PHASE BRIDGE
in addition gives much better regulation. rectifier of figure 4A is commonly used with
Where greater output voltage is a require- capacitor -input and resistance- capacitance
ment, where the load is relatively constant filter as a high -voltage supply for a cathode -
so that regulation is not of great significance, ray-tube. In this case the current drain is
and where the rectifier tubes will be operated very small so that the peak-current rating
well within their peak- current ratings, the of the rectifier tube seldom will be exceeded.
capacitor -input type of filter may be used. The circuit of figure 4B is most common-
The capacitor -input filter gives a no -load ly used in medium- voltage power supplies
output voltage equal approximately to the since this circuit is the most economical of
peak voltage being applied to the rectifier filament transformers, rectifier tubes, sock-
tubes. At full -load, the d -c output voltage ets, and space. But the circuit of figure 4C,
is usually slightly above one -half the second- commonly called the bridge rectifier, gives
ary a -c voltage of the transformer, with better transformer utilization so that the
the normal values of capacitance at the in- circuit is most commonly used in higher -
put to the filter. With large values of input powered supplies. The circuit has the advan-
capacitance, the output voltage will run tage that the entire secondary of the trans-
somewhat higher than the rms secondary former is in use at all times, instead of each
voltage applied to the tubes, but the peak side being used alternately as in the case of
current flowing through the rectifier tubes the full -wave rectifier. As a point of inter-
will be many times as great as the d -c output est, the current flow through the secondary
current of the power supply. The half -wave of the plate transformer is a substantially
716 Power Suppl ies THE RADIO
www.americanradiohistory.com
HANDBOOK Standard Circuits 717
only half the current passes through each transformers may be used. A tap at half -
choke. Also, the two chokes (L1) act as voltage may be taken at the junction of the
input chokes so that an additional swinging star transformers, but there will be d -c flow
choke is not required for such a power sup- in the transformer secondaries with the
ply. power -supply center tap in use. The circuit
A conventional two -voltage power supply of figure 7A has the disadvantage that there
with grounded transformer center tap is is an average d -c flow in each of the wind-
shown in figure 6E. The output voltages ings.
from this circuit are separate and not addi- Rectifiers Rectifying elements in high -volt-
tive as in the circuit of figure 6B. Figure age plate supplies are almost in-
6F is of advantage when it is desired to variably electron tubes of either the high -
operate class -B modulators from the half - vacuum or mercury -vapor type, although
voltage output of a bridge power supply and selenium or silicon rectifier stacks contain-
the final amplifier from the full -voltage ing a large number of elements are often
output. Both L, and L should be swinging used. Low -voltage high- current supplies may
chokes but the total drain from the power use argon gas rectifiers (Tungar tubes),
supply passes through L, while only the drain selenium rectifiers, or other types of dry-
of the final amplifier passes through L.. Ca- disc rectification elements. The xenon recti-
pacitors C, and C_ need be rated only half fier tubes offer some advantage over mer-
the maximum output voltage of the power cury -vapor rectifiers for high -voltage ap-
supply, plus the usual safety factor. This ar- plications where extreme temperature ranges
rangement is also of advantage in holding are likely to be encountered. However, such
down the "key -up" voltage of a c -w trans- rectifiers (3B25 for example) are consider-
mitter since both L, and L_ are in series, and ably more expensive than their mercury -
their inductances are additive, insofar as the vapor counterparts.
"critical inductance" of a choke -input filter
Peak Inverse Plate In an a -c circuit, the
is concerned. If 4 -pfd capacitors are used at
Voltage and Peak maximum peak voltage
both C, and C_ adequate filter will be ob-
Plate Current or current is 2, or 1.41
tained on both plate supplies for hum -free
radiophone operation. times that indicated by
the a -c meters in the circuit. The meters
Polyphase It is usual practice in com- read the root mean square (rms) values,
Rectification mercial equipment installa- which are the peak values divided by 1.41
Circuits tions when the power drain for a sine wave.
from a plate supply is to be If a potential of 1000 rms volts is ob-
greater than about one kilowatt to use a tained from a high -voltage secondary wind-
polyphase rectification system. Such power ing of a transformer, there will be 1.410 -
supplies offer better transformer utilization, volts peak potential from the rectifier plate
less ripple output and better power factor to ground. In a single-phase supply the
in the load placed on the a -c line. However, rectifier tube has this voltage impressed on
such systems require a source of three -phase it, either positively when the current flows
(or two -phase with Scott connection) ener- or "inverse" when the current is blocked on
gy. Several of the more common polyphase the other half -cycle. The inverse peak volt-
rectification circuits with their significant age which the tube will stand safely is used
characteristics are shown in figure 7. The as a rating for rectifier tubes. At higher
increase in ripple frequency and decrease voltages the tube is liable to arc back,
in percentage of ripple is apparent from the thereby destroying or damaging it. The
figures given in figure 7. The circuit of fig- relations between peak inverse voltage, total
ure 7C gives the best transformer utilization transformer voltage, and filter output volt-
as does the bridge circuit in the single -phase age depend on the characteristics of the fil-
connection. The circuit has the further ad- ter and rectifier circuits (whether full- or
vantage that there is no average d -c flow half wave, bridge, single-phase or polyphase,
in the transformer, so that three single -phase etc.).
718 Power Supplies THE RADIO
livers twice as much a -c voltage as would
_LINE VOLTS-HEATER VOLTS
HEATER AMPERES be the case with a half -wave rectifier or
bridge rectifier.
+ Q
LINE RECTIFIER
Mercury-Vapor The inexpensive mercury -
Rectifier Tubes vapor type of rectifier tube
+ is almost universally used in
+ C + SILICON the high -voltage plate supplies of amateur
LINE RECTIFIER and commercial transmitters. Most amateurs
T- zT o are quite familiar with the use of these tubes
but it should be pointed out that when new
0
VOLTAGE DOUBLER
or long- unused mercury-vapor tubes are
first placed in service, the filaments should
FULL-WAVE
be operated at normal temperature for ap-
proximately twenty minutes before plate
voltage is applied, in order to remove all
traces of mercury from the cathode and to
clear any mercury deposits from the top of
the envelope. After this preliminary warm -
+ VOLTAGE DOUBLER
_ HALF -WAVE up with a new tube, plate voltage may be
applied within 20 to 30 seconds after the
time the filaments are turned on, each time
the power supply is used. If plate voltage
0 should be applied before the filament is
brought to full temperature, active material
SILICON
RECTIFIER may be knocked from the oxide- coated fila-
VOLTAGE
QUADRUPLER ment and the life of the tube will be greatly
shortened.
Small r -f chokes must sometimes be con-
nected in series with the plate leads of mer-
Figure 8
cury -vapor rectifier tubes in order to pre-
vent the generation of radio- frequency hash.
TRANSFORMERLESS POWER -SUPPLY These r -f chokes must be wound with suf-
CIRCUITS ficiently heavy wire to carry the load cur-
Circuits such as shown above are also fre- rent and must have enough inductance to
quently called line -rectifier circuits. Silicon attenuate the r -f parasitic noise current to
rectifiers, vacuum diodes, or gas diodes may
be used as the rectifying elements in these
prevent it from flowing in the filter supply
circuits. leads and then being radiated into nearby
receivers. Manufactured mercury -vapor
rectifier hash chokes are available in various
Rectifier tubes are also rated in terms of
current ratings from various manufacturers.
peak plate current. The actual direct load
When mercury -vapor rectifier tubes are
current which can be drawn from a given
operated in parallel in a power supply, small
rectifier tube or tubes depends on the type
resistors or small iron -core choke coils should
of filter circuit. A full -wave rectifier with
he connected in series with the plate lead of
capacitor input passes a peak current several
times the direct load current. each tube. These resistors or inductors tend
In a filter with choke input, the peak cur- to create an equal division of plate current
rent is not much greater than the load cur- between parallel tubes and prevent one
rent if the inductance of the choke is fairly tube from carrying the major portion of
high (assuming full -wave rectification). the current. When high -vacuum rectifiers
A full -wave rectifier with two rectifier are operated in parallel, these chokes or
elements requires a transformer which de- resistors are not required.
www.americanradiohistory.com
HANDBOOK Standard Circuits 719
70 j 1 PHASE
degree of filtering and capacitance, should
60 be between 15 to 60 td. In the circuit of
50 figure 8D, capacitor C, should be rated at
40
150 volts and capacitor Co should be rated
30
zo
at 300 volts. In the circuit of figure
,0 8E, capacitors C, and C_ should be rated at
n 150 volts and C3 and C, should be rated at
0 so 100 ,SO zoo 250 300
RELATIVE LOAD CURRENT, 300 volts.
PERCENT OP FULL LOAD.
The d -c output voltage of the line recti-
Figure 9 fier may be stabilized by means of a VR
THE SELENIUM RECTIFIER tube. However, due to the unusually low
A -The selenium rectifier is a semiconductor internal resistance of the silicon rectifier,
stack built up of nickel-plated aluminum transformerless power supplies using this
discs coated on ono side with selenium
alloy. type of rectifying element can normally be
B- Rectifier efficiency is high, reaching 70% expected to give very good regulation.
for single -phase service, dropping slightly
at high current densities. Voltage -Doubler Figures 8C and 8D illus-
Circuits trate two simple voltage -
Transformerless Figure 8 shows a group of doubler circuits which will
Power Supplies five different types of trans - deliver a d -c output voltage equal approxi-
formerless power supplies mately to twice the rms value of the power
which are operated directly from the a -c line voltage. The no -load d -c output volt-
line. Circuits of the general type are normal- age is equal to 2.82 times the rms line volt-
ly found in a -c /d -c receivers but may be age value. At high current levels, the output
used in low -powered exciters and in test voltage will be slightly under twice the line
instruments. When circuits such as shown voltage. The circuit of figure 8C is of ad-
in A and B are operated directly from the vantage when the lowest level of ripple is
a -c line, the rectifier element simply recti-
required from the power supply, since its
fies the line voltage and delivers the alter-
ripple frequency is equal to twice the line
nate half cycles of energy to the filter net- frequency. The circuit of figure 8D is of
work. With the normal type of rectifier advantage when it is desired to use the
tube, load currents up to approximately 75 grounded side of the a -c line in a permanent
ma may be employed. The d -c voltage out- installation as the return circuit for the
put of the filter will be slightly less than the power supply. However, with the circuit of
rms line voltage, depending on the particu- figure 8D the ripple frequency is the same
lar type of rectifier tube employed. With as the a -c line frequency.
the introduction of the miniature silicon
rectifier, the transformerless power supply Voltage The circuit of figure 8E illus-
has become a very convenient source of Quadrupler trates a voltage -quadrupler cir-
moderate voltage at currents up to perhaps cuit for miniature silicon recti-
500 ma. A number of advantages are of- fiers. In effect this circuit is equivalent to
720 Power Supplies T H E R A D I O
'
tions. The threshold voltage will vary with
30 -4 Selenium and
Silicon Rectifiers POSITIVE TERMINAL CONTACT NEGATIVE TERMINAL
www.americanradiohistory.com
HANDBOOK Selenium and Silicon Rectifiers 721
too
U
b
75
W
I-
QF
CC z So
LLW
OK
FK , 25
WU
00
a
WO
aJ
25 SO 75 100 12S 150 1.2 2.0
PERCENT RATED PEAK INVERSE VOLTAGE FORWARD VOLTAGE DROP
VOL rs, o-c
Figure 13
www.americanradiohistory.com
HANDBOOK Selenium and Silicon Rectifiers 723
4
Series diode operation is commonly used
D, Da 03
when the peak- inverse voltage of the source
I I I
E' (5)
+265V
150 MA.
265 V.
150 MA
OA DUAL- VOLTAGE DOUBLER
WITH COMMON B -MINUS
t0 W
300 i! F
D2 D
Tl' D3
111.1 450 MA.
+325V. +
117 V. 15011F
50 MA. D2
ti 230V
l0LF
2aV.V. 02
Figure 15
www.americanradiohistory.com
HANDBOOK Silicon Supplies for SSB 725
45
speaking, the average power capability of a
40 power supply designed for intermittent voice
service (IVS) can be as low as 25 percent
35
of the PEP level. C -w requirements run
30 somewhat higher than this, the average c -w
25
power level running close to 50 percent of
the peak level for short transmissions. Rel-
20 atively small power transformers of modest
15 capability may be used for intermittent voice
and c -w service at a worthwhile saving in
10
weight and cost. The power capability of a
5 transformer may be judged by its weight,
as shown in the graph of figure 16. It must
o
.2 .4 .6 .6 1 0 1 2 1 4 1 6 1 5 2.0 be remembered that the use of alc or voice
POWER CAPACITY KW
compression in SSB service raises the duty,
Figure 16 thus reducing the advantage of the IVS
power rating. The IVS rating is difficult
INTERMITTENT VOICE SERVICE IN SSB
to apply to very small power transformers,
PERMITS LARGE PEAK POWER TO BE
since the d -c resistance of the transformer
DRAWN FROM POWER TRANSFORMER. windings tends to degrade the voltage regu-
PEAK -TO- AVERAGE RATIO OF NEARLY
lation to a point where the IVS rating is
FOUR TO ONE MAY BE ACHIEVED WITH
meaningless. Intelligent use of the IVS rating
MAXIMUM IVS RATING. POWER CA- in choosing a power transformer, stacked
PACITY OF TRANSFORMER MAY BE
silicon rectifiers, and "computer" type elec-
DETERMINED FROM WEIGHT
trolytic capacitors can permit the design
and construction of inexpensive, lightweight
A may be grounded and point B will provide high -voltage power supplies suitable for SSB
half -voltage. Circuit B is a half -wave tripler and c -w service.
that delivers 440 volts at 0.S ampere. In this The Design of IVS The low duty of SSB
circuit, one side of the power line is common Power Supplies and c -w modes can be
to the negative side of the output. Circuit C used to advantage in the
is a 900 -watt, 0.5 ampere supply composed
design of high -voltage power supplies for
of two voltage doublers supplied from a
"distribution" transformer having dual 115
230 -volt windings.
these services.
The Power Transformer - Relatively
low- voltage transformers may be used in
Power Supply Rating The duty cycle (ratio voltage -doubler service to provide a kilo-
for SSB Service of duration of maxi- watt or two of peak power at poten-
mum power output to tials ranging from one to three thousand
total "on" time) of a power supply in volts. Most suitable power transformers are
SSB and c -w service is much smaller than rated for commercial service and the IVS
that of a supply used for a -m equip- rating must be determined by experiment.
ment. While the power supply must be Figure 16 shows a relationship between var-
capable of supplying peak power equal ious services as determined by extensive
to the PEP input of the SSB equipment tests performed on typical transformers.
for a short duration, the average power de- The data illustrates the relationship between
manded by SSB voice gear over a period of transformer weight and power capability.
time usually runs about one -half or less of Transformer weight excludes weight of the
the PEP requirement. Then, too, the inter- case and mounting fixtures. Thus, a plate
vals between words in SSB operation provide transformer weighing about 17 pounds that
periods of low duty, just as the spaces in is rated for 400 watts commercial or in-
c -w transmission allow the power supply to dustrial service should have an 800 -watt
"rest" during a transmission. Generally peak capacity for c -w service and a 950-
726 Power Suppl ies THE RADIO
watt peak capacity for intermittent SSB fiers are available at a cost less than that of
service. A transformer having a so- called building a complete RC stack of diodes. The
"two- kilowatt PEP" rating for sideband may silicon rectifier, if properly used, is rarely
weigh as little as 22 pounds, according to the limiting factor in the design of steady -
this graph. state IVS power supplies, provided proper
Not shown in the graph is the effect of transient protection is incorporated in the
amplifier idling (standby) current taken supply.
from the supply, or the effect of bleeder The Filter Capacitor-Recently developed
current. Both currents impose an extra, con- "computer" -type aluminum -foil electrolytic
tinuous drain on the power transformer and capacitors combine high capacitance per
quickly degrade the IVS rating of the trans- unit of volume with moderate working volt-
former. Accordingly, the IVS curves of fig- age at a low price. Capacitors of this type
ure 16 are limited to the bleeder current can withstand short -interval voltage surges
required by the equalizing resistors for a of 15 percent over their d -c working volt-
series capacitor filter and assume that the age. In a stack, the capacitors should be
idling plate current of the amplifier is cut protected by voltage- equalizing resistors, as
to only a few milliamperes by the use of a shown in the power supplies in this section.
VOX -controlled cathode bias system. If the The capacitors are sheathed in a Mylar
idling plate current of the amplifier assumes jacket and may be mounted on the chassis
an appreciable fraction of the peak plate or adjacent to each other without addi-
current, the power capability of the supply tional insulation between the units. The
decreases to that given for c -w service. stack may be taped and mounted to a metal
Most small power transformers work re- chassis with a metal clamp, as is done in
liably with the center tap of the secondary
winding above ground potential. Some of the
larger transformers, however, are designed
Inrush Current Protection -
some of the units described here.
When the
power supply is first turned on, the filter
to have the center tap grounded and lack capacitors are discharged and present a near
sufficient insulation at this point to permit short circuit to the power transformer and
their use in either a bridge or voltage dou- rectifier stack. The charging current of a
bling configuration. The only way of de- high- capacitance stack may exceed the
termining if the center -tap insulation is maximum peak- recurrent current rating
sufficient is to use the transformer and see of the rectifiers for several cycles, thus dam-
if the insulation breaks down at this point! aging the diodes. Charging current is
It is wise to ground the frame of the trans- limited only by the series impedance of
former so that if breakdown occurs, the the power- supply circuit which consists
frame of the tranformer does not assume mainly of the d -c circuit resistance (pri-
the potential of the secondary winding and marily the resistance of the secondary wind-
thus present a shock hazard to the operator. ing of the power transformer) plus the
-A
The Silicon Rectifier bewildering vari- leakage reactance of the transformer. Trans-
ety of "TV- type" silicon rectifiers exists and formers having high secondary resistance
new types are being added daily. Generally and sufficient leakage reactance usually limit
speaking, 600 -volt PTV rectifiers, having an the inrush current so that additional in-
average rectified current rating of 1 ampere rush protection is unnecessary. This is not
at an ambient temperature of 75C with a the case with larger transformers hiving
maximum single -cycle surge- current rating low secondary resistance and low leakage
of if amperes or better are suitable for use reactance. To be on the safe side, in any
in the power supplies described in this sec- case, it is good practice to limit inrush cur-
tion. Typical rectifiers are packaged in the rent to well within the capability of the
top-hat configuration as well as the epoxy - diode stack. A current -limiting circuit is
encapsulated assembly and either type costs shown in figure 17 which can be added at
less than a dollar per unit. In addition, potted little expense to any power supply. The cur-
stacks utilizing controlled -avalanche recti- rent- limiting resistor (R) is initially in the
www.americanradiohistory.com
HAN DBOOK Silicon Supplies for SSB 727
1I11111111I11I11111111/11111111111111I II II
IIIIIIII1111
where,
R' is determined from figure 18,
/ is the full load current in amperes,
IIIIIIIIII IIIIIIIIIIIIIIIII11111111III II
111111111 II II IIIII1111I III1711111111 I I II
11111111111111111
1111111111 111111
IIIIIIIIIIIIIIIII
1111111111E1111111
IIIII11111
II II
111111111111111111
IIIII
IIIIIIIIIIIII II 11111111111 III II II 111111111111111
I
...
...
1==
The peak rectified voltage across the
complete filter- capacitor stack is equal to M\\./
111010191..N
www.americanradiohistory.com
HANDBOOK 2- Kilowatt PEP Supply 729
Figure 19
former having less secondary resistance and one pays for an intermittent -duty design,
slightly less secondary voltage would provide and care should be exercised in the use of
improved voltage regulation. The 840 -volt this equipment
transformer having an 8-ohm secondary To reduce the standby current and power
winding discussed earlier would be ideal in consumption, it is recommended that cath-
this application. ode bias be applied to the linear amplifier
The power supply is constructed on a stage shown in various designs in this Hand-
steel amplifier foundation chassis and dust book. During transmission, the cathode re-
cover. The diode stack is mounted on a per- sistor may be shorted out by contacts of
forated phenolic board under the chassis. the VOX relay, restoring the stage to proper
The electrolytic capacitors are taped to- operation.
gether and held in position atop the chassis Using the alternative 1100 -volt trans-
by a clamp cut from an aluminum sheet. former, the supply delivers 2600 volts at a
The interior of the clamp is lined with a c -w rating of 380 ma. Peak IVS voice rating
piece of plastic material salvaged from a is 500 ma (1.25 KW, PEP) . No -load volt-
package of frozen vegetables. The voltage - age is about 3100, and eight electrolytic ca-
equalizing resistors are wired across the pacitors are required in the stack instead of
terminals of the capacitors. Normally, it six.
takes 10 seconds or so to fully discharge the
filter capacitors when no external load is
connected to the supply. It is recommended
30 -8 A 2- Kilowatt PEP
that the supply be discharged with a 1000 - Supply for SSB
ohm, 100 -watt resistor before any work is
done on the unit. Power -supply components The power supply described in this section
and all terminals should be well protected is designed for the maximum power rating
against accidental contact. The voltage de- for amateur service. It is capable of 1.2 kilo-
livered by this supply is lethal and the filter watts power for c -w (50 percent duty
capacitors hold a considerable charge for a cycle) and 2 kilowatts IVS for SSB service.
surprising length of time. This is the price The supply is ideally suited for a grounded-
730 Power Supplies T H E R A D I O
3
pfd effective capacitance at 3600 volts
==1111=1========
2
2
DOo
oo
_
iM!^M
iIIIIMIMEMEIMMEICErE3>=M11
bill_-_ RATING 000 VZ)
MAX I VS
NG
working voltage. The voltage across the
"bottom" capacitor in the stack is moni-
tored by a 0 -to -1 d -c milliammeter recali-
brated 0 to 4 KV and which is used with a
Mi llO
S)
2
l
1111=MIM1==1111110O= series multiplier to provide a 0 to 5000-volt
2 000
1===1N1=MMIOO=
11===N==1100= full -scale indication. A 0 -to -1 d -c ammeter
MIMMEN111==00=1
11111=MIO=MIMo= is placed in series with the negative lead to
==IIIM=11===1=
=111Ma=
==l1NO0=11111===o=
00IN
00=1
MINIIMI0MMIN
I=MIMNMIIIII
==l1M111=M11
the high -voltage terminal strip.
The supply is built on a steel amplifier
foundation chassis in the same style as the
OM 1KW supply described previously. All safety
.2 .3 .4 .3 .e
LOAD CURRENT (AMPERES) precautions outlined earlier should be ob-
served with this supply.
Figure 20
www.americanradiohistory.com
HANDBOOK IVS Bridge- Rectifier Supplies 731
Supplies
zeoo .==p003==r:--SSB
N (2 )
Ieoo iirooiiiiiGTiOiiiiN
NNNNNNNNNNIiiiii:Z.NONMI
The bridge- rectifier circuit is somewhat
z400 NiNINNINNiNINNO
NNNNNNNNNNNNNNN>
more efficient than the full -wave circuit in
NNNNNNNNNNNNNNNI that the former provides more direct current
2200
NiNINNNNNNINNNO
NNNNNNNNiNNNNNNMII per unit of rms transformer current for a
NiNNNNNNNiNNINO
NNNiNNNNNNNNNNO given load than does the full -wave circuit.
NNNNNNNNNNNNNNNill
NNNNNNNNNNNNNNNill
Since there are two rectifiers in opposite
NNNNNNNNNNNNNNNIN arms of the bridge in the conducting mode
0
NNNNNNNNNNNNNNNIN
.2
.1 .3 .4 .5 .E .
when the a -c voltage is at its peak value,
LOAD CURRENT (AMPERES) the remaining two rectifiers are back -biased
to the peak value of the a -c voltage. Thus
Figure 22 th: bridge- rectifier circuit requires only
half the PIV rating for the rectifiers as
REGULATION CURVE OF
compared to a center -tap full -wave recti-
2- KILOWATT IVS SUPPLY
fier. The latter circuit applies the sum of
The power supply uses the circuit of figure the peak a -c voltage plus the stored capaci-
17. Surge components are as given in figure
20, except that the surge capacitor (C ) has a tor voltage to one rectifier arm in the maxi-
rating of 5 KV. Twenty type- 1N2071 (600 -volt mum inverse- voltage condition.
PIV) diodes are used in an assembly similar
to that shown in figures 25 and 26. Eight 240 A 500 -Watt IVS Shown in figure 23
:,fd, 450- working -volt (500 -volt peak) ca-
pacitors are used to provide 30 aft/ effective Bridge Power Supply is 500 -watt bridge
a
capacitance. Two 100K, 2 -watt resistors are power supply designed
shunted across each capacitor. Time -delay
circuit components are as suggested in fig- around an inexpensive "TV- replacement"
ure 20. The transformer used has a 115/230 - type power transformer. The secondary
volt primary and an 1100 -volt secondary, with winding is 1200 volts center -tapped at a
an (CAS rating of 1.2 KW. (Berkshire Trans-
former Corp., Kent, Conn. #BTC- 4905B). current rating of 200 ma. The weight of
Figure 23
U 1400
o
n 1200
I-
M- A-X. IVS R-
(480
ATING J'
J W.)
O
0 .1 .2 .3 .4 .5
LOAD CURRENT (AMPERES)
Figure 24
VOLTAGE -REGULATION CURVE OF
500 -WATT BRIDGE POWER SUPPLY
www.americanradiohistory.com
HANDBOOK Supply for SSB Transceiver 733
T1 p,
SA.
115 V. 1, FOO., TB
.01
.1cV 0 qF-pe
y
RO
or p-
IHII1
.01 O B+750 V.
T1eKv 02
Tt. 1cV ?
240
450--
I+ 50
sw
K
240 I+ S0 K
_9i 450 T SW 2
o ONO.
3N-300 MA. CHI CH2 IN-300MA. -1-
II
B+ 250 V.
BIAS (ADJ.)
e
0 12.6 V. 1,
1
?LINE SW.
Figure 28
metal foil separated by several layers of must be correctly connected in the circuit
special paper. Some types of paper capaci- so that the anode always is at a positive po-
tors are wax -impregnated, but the better tential with respect to the electrolyte, the
ones, especially the high -voltage types, are latter actually serving as the other electrode
oil- impregnated and oil -filled. Some capaci- (plate) of the capacitor. A reversal of the
tors are rated both for flash test and normal
polarity for any length of time will ruin
operating voltages; the latter is the impor-
the capacitor.
tant rating and is the maximum voltage
which the capacitor should be required to The dry type of electrolytic capacitor uses
withstand in service. an electrolyte in the form of paste. The di-
The capacitor across the rectifier circuit electric in electrolytic capacitors is not per-
in a capacitor -input filter should have a fect; the capacitors have a much higher
working -voltage rating equal at least to 1.41 direct -current leakage than the paper type.
times the rms voltage output of the rectifier. The high capacitance of electrolytic ca-
The remaining capacitors may be rated pacitors results from the thinness of the
more nearly in accordance with the d -c film which is formed on the plates. The max-
voltage. imum voltage that can be safely impressed
The electrolytic capacitor consists of two across the average electrolytic filter capacitor
aluminum electrodes in contact with a con- is between 450 and 600 volts; the working
ducting paste or liquid which acts as an voltage is usually rated at 450. When elec-
electrolyte. A very thin film of oxide is trolytic capacitors are used in filter circuits
formed on the surface of one electrode, of high -voltage supplies, the capacitors
called the anode. This film of oxide acts should be connected in series. The positive
as the dielectric. The electrolytic capacitor terminal of one capacitor must connect to
www.americanradiohistory.com
HANDBOOK Power -Supply Components 735
the negative terminal of the other, in the Transformers Power transformers and fila-
same manner as dry batteries are connected ment transformers normally
in series. will give no trouble over a period of many
Electrolytic capacitors can be greatly re- years if purchased from a reputable manu-
duced in size by the use of etched aluminum facturer, and if given a reasonable amount
foil for the anode. This greatly increases the of care. Transformers must be kept dry;
surface area, and the dielectric film cover- even a small amount of moisture in a high -
ing it, but raises the power factor slightly. voltage unit will cause quick failure. A
For this reason, ultramidget electrolytic ca- transformer which is operated continuously,
pacitors ordinarily should not be used at full within its ratings, seldom will give trouble
rated J -c voltage when a high a -c compo- from moisture, since an economically de-
nent is present as would be the case for signed transformer operates at a moderate
the input capacitor in capacitor -input filter. temperature rise above the temperature of
the surrounding air. But an unsealed trans-
Bleeder A heavy -duty resistor should be former which is inactive for an appreciable
Resistors connected across the output of a period of time in a highly humid location
filter in order to draw some load can absorb enough moisture to cause early
current at all time. This resistor avoids failure.
soaring of the voltage at no load when
swinging -choke input is used, and also pro- Filter Choke Filter inductors consist of a
vides a means for discharging the filter ca- Coils coil of wire wound on a lami-
pacitors when no external vacuum -tube cir- nated iron core. The size of
cuit load is connected to the filter. This wire is determined by the amount of direct
bleeder resistor should normally draw ap- current which is to flow through the choke
proximately 10 percent of the full load coil. This direct current magnetizes the
current. core and reduces the inductance of the
The power dissipated in the bleeder resistor choke coil; therefore, filter choke coils of
can be calculated by dividing the square of the smoothing type are built with an air
the d -c voltage by the resistance. This power gap of a small fraction of an inch in the
is dissipated in the form of heat, and, if the iron core, for the purpose of preventing
resistor is not in a well -ventilated position, saturation when maximum current flows
the wattage rating should be higher than through the coil winding. The "air gap" is
the actual wattage being dissipated. High - usually in the form of a piece of fiber in-
voltage, high- capacitance filter capacitors serted between the ends of the laminations.
can hold a dangerous charge if not bled off, The air gap reduces the initial inductance of
and wirewound resistors occasionally open the choke coil, but keeps it at a higher value
up without warning. Hence it is wise to under maximum load conditions. The coil
place carbon resistors in series across the must have a great many more turns for the
regular wirewound bleeder. same initial inductance when an air gap is
When purchasing a bleeder resistor, be used.
sure that the resistor will stand not only The d -c resistance of any filter choke
the required wattage, but also the toitage. should be as low as practical for a specified
Some resistors have a voltage limitation value of inductance. Smaller filter chokes,
which makes it impossible to force sufficient such as those used in radio receivers, usually
current through them to result in rated have an inductance of from 6 to 15 henrys,
wattage dissipation. This type of resistor and a d -c resistance of from 200 to 400
usually is provided with slider taps, and is ohms. A high d -c resistance will reduce the
designed for voltage divider service. An un-
tapped, nonadjustable resistor is preferable as
output voltage, due to the voltage drop
across each choke coil. Large filter choke
a high -voltage bleeder, and is less expensive.
coils for radio transmitters and class -B
Several small resistors may be connected in
series, if desired, to obtain the required watt- amplifiers usually have less than 100 ohms
age and voltage rating. d -c resistance.
736 Power Supplies TH E R A D I O
www.americanradiohistory.com
HANDBOOK Special Power Supplies 737
I 115
W_
V. PUTE 115 V FILS. 325
O
TO
450 V.
0
GN0.1
J
regulated bias voltage. Associated with the
triode is a d -c amplifier and a voltage -
L
regulator tube which serves to vary the grid
Figure 30 voltage of the triode regulator tube so that
a constant voltage is maintained across it.
SCHEMATIC OF VOLTAGE - The SK variable potentiometer is adjusted
REGULATED POWER SUPPLY to produce about 20 ma current through
3,-615 or 520 volts each side of c.t., 300 ma. the first regulator tube.
Stancor P -8041
T.-5 volts at 3 amp., 6.3 volts at 6 amp. A Shunt -Regulated Series - regulated power
Stancor P -5009
CH -4 -henry at 250 ma. St C -1412
Bias Supply supplies are usually not
(100 to 600 V) suited for bias units since
the direction of load -cur-
in the 6AS7G is equal to the current through rent flow is opposite from that of a regular
it (output current plus the current passing supply. In the supply shown in figure 32,
through the two bleeder strings) multiplied the regulator tube (6CK4) acts as a vari-
by the drop through the tube (voltage able bleeder resistor which automatically
across the filter capacitor minus the out- adjusts its resistance to a value such that
put voltage of the supply). the grid current flowing through it will de-
j1.14
!0 MA.
5Y3-GT VR-150 65J7
Irtn
O
E.3V.TO
X
www.americanradiohistory.com
HANDBOOK Power -Supply Design 739
VI
AMMOXIMAT2
COMPONENTS OUTPUT VOLTAGt
MAX. 6.3V.
NO FULL CUMtCNTI1LAMtMT
TiI VI CHI CH C1 Ca R1 LOAD LOAD
10 M. 10 M.
10UF, 450 V. 2010,50 V.
330-0-350 3r3-GT
CORNELL- CORNELL-
33K,101
S7AMCOAIC4101 STANCOR STANCOR DOL/LIER OUI/LIER 310 240 SOMA. 3A.
C-10o1 C-1007 LR-1045 LR-2013
600-0-600 554-GT
3-23 M. 20 M. 131/ 600 V. I UF.600 V.
UTC S-I/ UTC S-3t UTE S-3 / SPRAWL SPRAGUE 33 K,25 W 540 410 200MA. A.
CR-II CR-11
100-0-500 5-25 M. 411F.KV I 6 UF,/ K V.
UTC S-I! 35-Cr UTC 5-32
20 M.
UM S-al SPRAGUE
CR-41
SPRAGUE
CR-I1
50K,25W 130 650 175MA. -
Figure 33
VI
1 1 S V.
e0-6O '1,
O
V.
0
COMPONENTS APPROXIMATE
OUTPUT VOLTAGE MAX. 6.3V.
Ti Vi CN 1 C 1 C 2 R
NO PULL CURRENT FILAMENT
1
LOAD LOAD
10 M. 201", 430 V. 201F, 450 V.
260-0-260 5V3-GT STANCOR CORNELL- CORNELL-
STANCORPC-4404 C-1007 OUSILIER OUE/LIER 351,10W 340 240 60MA. 3A.
ER-1045 ER-2043
500-0-600
Inv S-45 5R4-GY
20M.
UTC S-31
AUF, 1.5KV.
SPRAGUE
CR-475
61/F, 1.511V.
SPRAGUE 751 25W
CR-610
1200 610 170 MA. -
Figure 34
DESIGN CHART FOR CAPACITOR -INPUT POWER SUPPLIES
used to hold the minimum current drain of or modulator requires two different supply
the supply to a value greater than 50 milli- voltages, differing by a ratio of about 2:1.
amperes or so. Under these conditions the A simple bridge supply such as shown in
peak plate current of the 872A mercury- figure 36 will provide both of these voltages
vapor tubes will not be exceeded if the input from a simple broadcast "replacement- type"
filter capacitor is 4 fd or less. power transformer. The first supply of fig-
Choke -input filter systems are character- ure 36 is ample to power a transmitter of
ized by lower peak -load currents (1.1 to the 6CL6 -807 type to an input of 60 watts.
1.3 times the average load current) than the The second supply will run a transmitter
capacitor -input filter, and by better voltage running up to 120 watts, such as one em-
regulation. Design charts for capacitor and ploying a pair of 6146 tetrodes in the pow-
choke input filter supplies for various volt- er- amplifier stage. It is to be noted that
ages and load currents are shown in figures separate filament transformers are used for
33, 34, and 35. rectifier tubes V, and V., and that one leg
The construction of power supplies for of each filament is connected to the cathode
transmitters, receivers, and accessory equip- of the respective tube, which is at a high
ment is a relatively simple matter electrically potential with respect to ground. The choke
since lead lengths and placement of parts are CH, in the negative lead of the supply serves
of minor importance and since the circuits as a common filter choke for both output
themselves are quite simple. voltages. Each portion of the supply may be
considered as having a choke -input filter
Bridge Supplies Some practical variations of system. Filaments of V, and V, are energized
the common bridge- rectifier before the primary voltage is applied to T,.
circuit of figure6 are illustrated in figures Bridge supplies may also be used to ad-
36 and 37. In many instances a transmitter vantage to obtain relatively high plate volt-
www.americanradiohistory.com
HANDBOOK Power -Supply Design 741
T2
MAX.
COMPONENTS OUTPUT VO~TAGE CURRENT
CI C2 OCAS)
Ti T2 VI-V2 CHI CH2 R1 LOAD LOAD
Figure 35
ages for high -powered transmitting equip- will be present. However, with bridge cir-
ment. Type 866A and 872A rectifier tubes cuits, the same tubes may deliver up to as
can only serve in a supply delivering under much as 7000 volts d.c. without exceeding
3500 volts in a full -wave circuit. Above the peak- inverse-voltage rating.
this voltage, the peak- inverse -voltage rating The bridge circuit also permits the use
of the rectifier tube will be exceeded, and of the so- called "pole transformer" in high -
danger of flashback within the rectifier tube voltage power supplies. Two KVA trans-
CH2
T1 t M V.111 z
Figure 36
DUAL -VOLTAGE
+ X.V.61
INTERMITTENT -SERVICE
BRIDGE POWER SUPPLIES
o
63V.
O
T3 100 R
00-0-400 MX56T 5U4-G6 161F. 7610. 10 X. N.
20R 106t 1W
625 260 130 MA. 50 MA.
T2,T3 6.3 V., A. 1
STA NCORPC-6472 450 V. 450V. 22SMA. 75MA.
57-ANCOR P-6134
1lV.
50-601.
742 Power Supplies THE RADIO
Ti VI
Figure 37
HIGH -VOLTAGE
POWER SUPPLY
T2
`POLE TwANYM1ICR'
t 71 9-A 2!7 2500 v. 200 w
1900 500 MA.
ur0
formers of this type having a 110 /220 -volt versed, and either 110 or 220 volts applied
secondary winding and a split 2200 -volt to the "primary" winding, approximately
primary winding may often be picked up in 2200 volts rms will be developed across the
salvage yards for a dollar or two. If re- new "secondary" winding. If used in a
bridge circuit as shown in figure 37, a d -c
872A,s
supply voltage of about 1900 volts at a
RYI current of 500 milliamperes may be drawn
from such a transformer. Do not attempt
to use a smaller transformer than the 2KVA
rating, as the voltage regulation of the unit
will be too poor for practical purposes.
For higher voltages, a pole transformer
with a 4400 -volt primary and a 110/220 -
volt secondary may be reversed to provide a
d -c plate supply of about 3800 volts.
Commercial plate transformers intended
TO EXCITER
CONTROL CIRCUIT
for full wave rectifier service may also be
used in bridge service provided that the
Figure 38 insulation at the center -tap point of the
HIGH- VOLTAGE POWER SUPPLY high- voltage winding is sufficient to with-
T,--2900-0-2900 volts at 600 ma., Stanco, stand one -half of the rms voltage of the
P -8035 secondary winding. Many high -voltage trans-
T,-.5 volts, 20 amp.
CH,6 henrys, 700 ma. formers are specifically designed for opera-
C; 0.15 id, 5000 -volt tion with the center tap of the secondary
C_ -Three 4 -td 3000-volt
R,- 100,000 ohms, 200 -watt winding at ground potential; consequently
R:-Eleven 0.5- megohm 2 -watt resistors in the insulation of the winding at this point
series is not designed to withstand high voltage.
RT DPST relay, 110 v. coil, 20 amp contacts,
Potter & Brumfield PRIA It is best to check with the manufacturer
RFC,, -Nash filter, J. W. Miller Co. No. 7868 of the transformer and find out if the in-
www.americanradiohistory.com
HANDBOOK Kilowatt CCS Supply 743
sulation will withstand the increased voltage work, the supply may be used to power a
before a full -wave type transformer is util- 2000 watt PEP linear amplifier.
ized in bridge -rectifier service. Because the total weight of the compo-
nents is over 150 pounds, the supply should
be built directly on the bottom of a relay
30 -14 A Kilowatt CCS rack instead of on a steel chassis.
Power Supply The r -f hash- suppression chokes (RFC,
and RFC_) are fastened directly to the
high -voltage terminals of the plate trans-
Shown in figure 38 is the schematic of a
former. The two 872A rectifier tubes are
power supply capable of delivering 2500
so located that the leads from the r -f chokes
volts at a continuous current drain of 500 to the plate caps are only about three inches
milliamperes, or 1000 milliamperes with an long.
intermittent load. The supply is designed A 0.15 -pfd 5000 -volt paper capacitor is
to power a kilowatt amplifier operating at used to resonate the filter choke to approxi-
2500 volts and 400 ma, in conjunction with mately 120 Hz at a bleeder current of 25
a 500 -watt modulator operating at 2500 milliamperes. When full load current is
volts at a varying current drain of 50 -300 drawn, the inductance of the filter choke
ma. Specifically, the supply is employed drops, detuning the parallel- resonant circuit.
with a transmitter having a pair of 4 -250A Improved voltage regulation is gained by
tetrode tubes in the class -C stage, and a this action; the no-load voltage increases
pair of 810 modulator tubes. For sideband only 200 volts over the full -load voltage.
CHAPTER THIRTY -ON E
Electronic
Test Equipment
All amateur stations are required by law or f -m transmission, and are a necessity if
to have certain items of test equipment single-sideband operation is contemplated.
available within the station. A c -w station A calibrated signal generator is almost a
is required to have a frequency meter or necessity if much receiver work is con-
other means, in addition to the transmitter templated, although a frequency meter of
frequency control, for ensuring that the LM or BC -221 type, particularly if it in-
transmitted signal is on a frequency within cludes internal modulation, will serve in
one of the frequency bands assigned for place of the signal generator. Extensive an-
such use. A radiophone station is required tenna work invariably requires the use of
in addition to have a means of determining some type of field- strength meter, and a
that the transmitter is not being modulated standing -wave meter of some type is very
in excess of its modulation capability, and helpful. Lastly, if much vhf work is to be
in any event not more than 100 percent. done, a simple grid -dip meter will be found
Further, any station operating with a power to be one of the most used items of test
input greater than 900 watts is required to equipment in the station.
have a means of determining the exact input
to the final stage of the transmitter, so as
to ensure that the power input to the plate 31 -1 Voltage and
circuit of the output stage does not exceed Current
1000 watts.
The additional test and measurement The measurement of voltage and current
equipment required by a station will be de- in radio circuits is very important in proper
termined by the type of operation contem- maintenance of equipment. Vacuum tubes
plated. It is desirable that all stations have and transistors of the types used in commu-
an accurately calibrated voltohmmeter for nications work must be operated within
routine transmitter and receiver checking rather narrow limits in regard to filament
and as an assistance in getting new pieces of or collector voltage, and they must be
equipment into operation. An oscilloscope operated within certain maximum limits in
and an audio oscillator make a very desir- regard to the voltage and current on other
able adjunct to a phone station using a -m electrodes.
www.americanradiohistory.com
Voltage and Current 745
+10 100 +250 +1000 ment. The calculation of shunts for extend-
0 0 0
ing the range of d -c milliammeters and am-
1011 100 K* 25011 tM[G meters is discussed in Chapter Two.
A direct current voltmeter is merely a d -c
milliammeter with a multiplier resistor in
+10 +100 +250
series with it. If it is desired to use a low -
+1000
0 0 0 o range milliammeter as a voltmeter the value
of the multiplier resistor for any voltage
range may be determined from the following
1011 9011 150 K 75011 formula:
1000 F,
R
I
1000
10
where,
0-1 o.c 100 250 R equals multiplier resistor in ohms,
10K 150 K 75011
9011
E equals desired full -scale voltage,
I equals full -scale current of meter in ma.
bring the needle to zero ohms when the ter- The ohmmeter diagramed in figure 3 was
minals are shorted; this adjustment should especially designed for the reasonably accu-
always be made before a resistance measure- rate reading of resistances down to 1 ohm.
ment is taken. Higher voltages than 500 can Two scales are provided, one going in one
be read if a higher value of multiplier re- direction and the other scale going in the
sistor is added to an additional tap on the other direction because of the different man-
switch. The proper value for a given full - ner in which the milliammeter is used in
scale reading can be determined from each case. The low scale covers from 1 to
Ohm's Law. 100 ohms and the high scale from 100 to
Resistances higher than 100.000 ohms 10,000 ohms. The high scale is in reality a
cannot be measured accurately with the cir- medium -range scale. For accurate reading
cuit constants shown; however, by increas- of resistances over 10,000 ohms, an ohm-
ing the ohmmeter battery to 45 volts and meter of the type previously described
multiplying the 4000 -ohm resistor and 1000 - should be used.
ohm potentiometer by 10, the ohms scale also The 1- to 100 -ohm scale is useful for
checking transformers, chokes, r -f coils, etc.,
will be multiplied by 10. This would per-
which often have a resistance of only a few
mit accurate measurements up to 1 meg- ohms.
ohm. The calibration scale will depend on the
0 -1 d -c milliammeters are available with internal resistance of the particular make of
special voltohmmeter scales which make in- 1.5 -ma meter used. The instrument can be
dividual calibration unnecessary. Or, special calibrated by means of a Wheatstone bridge
scales can be purchased separately and sub- or a few resistors of known accuracy. The
stituted for the original scale on the milliam- latter can be series- connected and parallel -
meter. connected to give sufficient calibration
www.americanradiohistory.com
HANDBOOK Voltage and Current 747
ovar 1G4-G
SWITCH
Figure 4
Figure 3
SLIDE -BACK V -T VOLTMETER
SCHEMATIC OF A LOW -RANGE By connecting a variable source of voltage in
OHMMETER series with the input to a conventional v -t
voltmeter, or in series with the simple triode
A description of the operation of this circuit voltmeter shown above, a slide -back a-c volt-
is given in the text. With the switch in the meter for peak voltage measurement can be
left position the half-scale reading of the constructed. Resistor R should be about 1000
meter will occur with an external resistance ohms per volt used at battery B. This type of
of 1000 ohms. With the switch in the right v -t voltmeter has the advantage that it can
position, half -scale deflection will be obtained give a reading of the actual peak voltage of
with an external resistance equal to the d-c the wave being measured, without any cur-
resistance of the milliammeter (20 to SO ohms rent drain from the source of voltage.
depending on the make of instrument).
points. A hand -drawn hand -calibrated scale the iron -tane instrument is commonly used.
can be cemented over the regular meter For audio frequency alternating current
scale to give a direct reading in ohms. (50- 20,000 Hz) a d'Arsonval instrument
Before calibrating the instrument or us- having an integral copper- oxide, selenium, or
ing it, the test prods should always be silicon rectifier is usually used. Radio -fre-
touched together and the zero adjuster set quency voltage measurements are usually
accurately. made with some type of vacuum -tube volt-
meter, while r -f current measurements are
Measurement of The measurement of al- almost invariably made with an instrument
Alternating Current ternating current and containing a thermocouple to convert the
and Voltage voltage is complicated radio- frequency current into direct current
by two factors; first, for the meter movement.
the frequency range covered in ordinary Since an alternating- current wave can
communication channels is so great that have an almost infinite variety of shapes,
calibration of an instrument becomes ex- it can easily be seen that the ratios between
tremely difficult; second, there is no single the three fundamental quantities of the
type of instrument which is suitable for all wave (peak, rms,effective, and average after
a -c measurements -as the d'Arsonval type rectification) can also vary widely. So it
of movement is suitable for d -c. The becomes necessary to know beforehand just
d'Arsonval movement will not operate on which quality of the wave under measure-
alternating current since it indicates the ment our instrument is going to indicate.
average value of current flow, and the For the purpose of simplicity we can list
average value of an a -c wave is zero. the usual types of alternating- current meters
As a result of the inability of the reliable along with the characteristic of an alternat-
d'Arsonval type of movement to record an ing- current wave which they will indicate:
alternating current, either this current must
be rectified and then fed to the movement. iron -vane, thermocouple -rms.
or a special type of movement which will Rectifier type (copper- oxide, selenium,
operate from the effective value of the cur- etc.)- average after rectification.
rent can be used. V.t.v.m. -rms, average, or peak, depend-
For the usual measurements of power - ing on design and calibration of the
frequency alternating current (25 -60 Hz), meter.
748 Electronic Test Equipment THE RADIO
www.americanradiohistory.com
HANDBOOK Vacuum -Tube Voltmeter 749
and R, should be able to withstand the same spection of the trace on the cathode -ray
amount of voltage. The easiest and least tube screen. The peak positive excursion of
expensive way of obtaining such resistors the wave will be slightly flattened by the
is to use several low- voltage resistors in action of the v.t.v.m. Usually this flatten-
series, as shown in figure S. Other voltage ing will be so small as to be negligible.
ranges can be obtained by changing the An alternative arrangement, shown in
value of these resistors, but for voltages less figure 6, is quite convenient for the meas-
than several hundred volts a more linear urement of high a -c voltages such as are en-
calibration can be obtained by using a re- countered in the adjustment and testing of
ceiver -type diode. A calibration curve high -power audio amplifiers and modula-
should be run to eliminate the appreciable tors. The arrangement consists simply of a
error due to the high internal resistance of 2X2 rectifier tube and a filter capacitor of
the diode, preventing the capacitor from perhaps 0.25 -fd capacitance, but with a
charging to the full peak value of the volt- voltage rating high enough that it is not
age being measured. likely to be punctured as a result of any
A direct -reading diode peak voltmeter of tests made. Cathode -ray oscilloscope capaci-
the type shown in figure S will load the tors, and those for electrostatic- deflection
source of voltage by approximately one - TV tubes often have ratings as high as 0.25
half the value of the load resistance in the fd at 7500 to 10,000 volts. The indicating
circuit (R1, or R, plus R_, in this case) . instrument is a conventional multiscale d -c
Also, the peak voltage reading on the meter voltmeter of the high- sensitivity type, pref-
will be slightly less than the actual peak erably with a sensitivity of 20,000 or
voltage being measured. The amount of 50,000 ohms per volt. The higher the sen-
lowering of the reading is determined by the sitivity of the d -c voltmeter used with the
ratio of the reactance of the storage capac- rectifier, the smaller will be the amount of
itance to the load resistance. If a cathode - flattening of the a -c wave as a result of the
ray oscilloscope is placed across the terminals rectifier action.
of the v.t.v.m. when a voltage is being Basic D -C Vacuum- A simple v.t.v.m. is
measured, the actual amount of the lower- Tube Voltmeter shown in figure 7 . The
ing in voltage may be determined by in- plate load may be a me-
chanical device, such as a relay or a meter,
2x2/979 or the output voltage may be developed
across a resistor and used for various control
A-c
l IISV.A.0
purposes. The tube is biased by E, and a
VOLTAGE
FROM
fixed value of plate current flows, causing a
SOURCE
WITH
fixed voltage drop across plate -load resistor
O -c
RETURN
CONVENTIONAL
HIGH- SENSITIVITY
R1,. When a positive d -c voltage is applied to
RATH VOLTMETER
the input terminals it cancels part of the
negative grid bias, making the grid more
Figure 6 positive with respect to the cathode. This
grid- voltage change permits a greater
PEAK -VOLTAGE MEASUREMENT
CIRCUIT
Through use of the arrangement shown above
it is possible to make accurate measurements
of peak a -c voltages, such as across the sec-
ondary of a modulation transformer, with a
c tional d-c multivoltmeter. Capacitor C
and transformer T should, of course, be in-
sulated for the highest peak voltage likely to
be encountered. A capacitance of 0.25 -ufd at
C has been found to be adequate. The higher Figure 7
the sensitivity of the indicating d-c voltmeter,
the smaller will be the error between the SIMPLE VACUUM-TUBE
indication on the meter and the actual peak
voltage being measured. VOLTMETER
750 Electronic Test Equipment THE RADIO
amount of plate current to flow, and de-
velops a greater voltage drop across the
plate -load resistor. A negative input voltage
would decrease the plate current and de-
crease the voltage drop across Rp. The vary-
ing voltage drop across R, may be employed
as a control voltage for relays or other de-
vices. When it is desired to measure various
voltages, a voltage range switch (figure 8)
may precede the v.t.v.m. The voltage to be
measured is applied to voltage divider (R1, Figure 9
R2, R3) by means of the voltage range BRIDGE -TYPE VACUUM -TUBE
switch. Resistor R4 is used to protect the VOLTMETER
meter from excessive input voltage to the
v.t.v.m. In the plate circuit of the tube
a battery and a variable resistor (zero ad- rent flows through the meter. If one voltage
justment) are used to balance out the changes, the bridge becomes unbalanced and
meter reading of the normal plate cur- indication of this will be noted by a reading
rent of the tube. The zero -adjustment po- of the meter.
tentiometer can be so adjusted that the A Modern V.T.V.M. For the purpose of an-
meter (M) reads zero current with no input alysis, the operation of a
voltage to the v.t.v.m. When a d -c input modern v.t.v.m. will be described. The
voltage is applied to the circuit, current Heathkit 1M -13 is a fit instrument for such
flows through the meter, and the meter read- a description, since it is able to measure posi-
ing is proportional to the applied d -c volt- tive or negative d -c potentials, a -c rms
age. values, peak -to -peak values, and resistance.
The Bridge -type Another important use of a The circuit of this unit is shown in figure
V.T.V.M. d -c amplifier is to show the 10. A sensitive 0 to 200 d -c microammeter
exact point of balance be- is placed in the cathode circuit of a 12AU7
tween two d -c voltages. This is done by twin triode. The zero- adjust control sets up
means of a bridge circuit with two d -c am- a balance between the two sections of the
plifiers serving as two legs of the bridge triode such that with zero input voltage ap-
(figure 9). With no input signal, and with plied to the first grid, the voltage drop
matched triodes, no current will be read on across each portion of the zero -adjust con-
meter M, since the IR drops across R, and trol is the same. Under this condition of bal-
RZ are identical. When a signal is applied to ance the meter will read zero. When a volt-
one tube, the IR drops in the plate circuits age is applied to the first grid, the balance in
become unbalanced, and meter M indicates the cathode circuits is upset and the meter
the unbalance. In the same way, two d -c indicates the degree of unbalance. The re-
voltages may be compared if they are applied lationship between the applied voltage on
to the two input circuits. When the voltages the first grid and the meter current is linear
are equal, the bridge is balanced and no cur- and therefore the meter can be calibrated
with a linear scale. Since the tube is limited
in the amount of current it can draw, the
meter movement is electronically protected.
The maximum test voltage applied to the
12AU7 tube is about 3 volts. Higher applied
ZERO -ADJUST
voltages are reduced by a voltage divider
which has a total resistance of about 10
megohms. An additional resistance of 1-
megohm is located in the d -c test probe,
thereby permitting measurements to be made
Figure 8 in high -impedance circuits with minimum
D -C VACUUM -TUBE VOLTMETER disturbance.
www.americanradiohistory.com
HANDBOOK Vacuum -Tube Voltmeter 751
Figure 10
Figure 11
Figure 12
FULL-WAVE RECTIFIER
FOR V.T.V.M. R -F PROBE SUITABLE FOR USE
IN 1 kHz -100 MHz RANGE
www.americanradiohistory.com
HANDBOOK Measurement of Power 753
load resistors are available in various resist- 120 -volt a -c line. A rheostat in series with
ances in both 100- and 250-watt ratings. the lead from the a -c line to the lamp is
These are virtually noninductive, and may used to vary its light intensity to the same
be considered as a pure resistance up to 30 value (as indicated by the exposure meter)
MHz. The resistance of these units is sub- as it was putting out as a dummy load. The
stantially constant for all values of current a -c voltmeter in parallel with the lamp and
up to the maximum dissipation rating, but ammeter in series with it is then used to
where extreme accuracy is required, a cor- determine lamp power input by: P = EI.
rection chart of the dissipation coefficient This method of power determination is sat-
of resistance (supplied by the manufac- isfactory for audio and low- frequency r.f.
turer) may be employed. This chart shows but is not satisfactory for vhf work because
the exact resistance for different values of of variations in lamp efficiency due to un-
current through the resistor. even heating of the filament.
Sine -wave power measurements (r-f or Dummy Loads Lamp bulbs make poor dum-
single-frequency audio) may also be made my loads for r -f work, in
through the use of a v.t.v.m. and a resistor general, as they have considerable reactance
of known value. In fact a v.t.v.m of the above 2 MHz, and the resistance of the lamp
type shown in figure 10 is particularly varies with the amount of current passing
suited to this work. The formula, P = E' /R through it.
is used in this case. However, it must be
A suitable r-f load for powers up to a
remembered that a v.t.v.m. of the type few watts may be made by paralleling 2-
shown in figure 10 indicates the peak value
of the a -c wave. This reading must be con-
verted to the rms or heating value of the
wave by multiplying it by 0.707 before sub-
stituting the voltage value in the formula.
The same result can be obtained by using
the formula P = Es /2R. (Note: Some
v.t.v.m.'s are peak reading but are calibrated
rms on the meter scale).
Thus all three methods of determining
power-ammeter -resistor, voltmeter -resistor,
and voltmeter- ammeter-give an excellent
crosscheck on the accuracy of the deter-
mination and upon the accuracy of the
standards.
Power may also be measured through the
use of a calorimeter, by actually measuring
the amount of heat being dissipated.
Through the use of a water-cooled dummy - Figure 13
load resistor this method of power output
2- KILOWATT DUMMY LOAD
determination is being used by some of the FOR 3 -30 MHz
most modern broadcast stations. But the
method is too cumbersome for ordinary Load is built in case measuring 22" deep, 71"
wide and 5" high. Meter is calibrated in watts
power determinations. against microampere scale as follows: (1),
Power may also be determined photometri- 22.3 pa. (5) 50 ba. (101, 70.5 a. (15), 86.5
ua. (20), 100 ria. Scale may be marked off as
cally through the use of a voltmeter, shown in photograph. Calibration technique is
ammeter, incandescent lamp used as a load discussed in text. Alternatively, a standing -
resistor, and a photographic exposure wave bridge (calibrated in watts) such as
"Micromatch" may be used to determine
meter. With this method the exposure meter power input to bridge.
is used to determine the relative visual out- Vents in top of case, and inch holes in
chassis permit circulation of air about re-
put of the lamp running as a dummy -load sistors. Unit should be fan- cooled for con-
resistor and of the lamp running from the tinuous dissipation.
754 Electronic Test Equipment THE RADIO
watt composition resistors of suitable value VENTILATED SHIELD COVER
Figure 15
DUMMY -LOAD
ASSEMBLY
Twelve Globar resistors
(surplus) are mounted
to aluminum "Tee"
stock, six to a side, in
fuse clips. Right end is
supported by ceramic
pillars from front panel.
Probe, meter, and po-
tentiometers are at
right.
www.americanradiohistory.com
HANDBOOK Measurement of Circuit Components 755
(C, + C)
measured, it is connected in parallel with
the standard capacitor and the combination The determination of an unknown capaci-
tuned accurately to some known frequency. tance is somewhat less complicated than the
This tuning may be accomplished either by above. A tuned circuit including a coil, the
using the tuned circuit as a wavemeter and unknown capacitor and the standard ca-
coupling it to the tuned circuit of a refer- pacitor, all in parallel, is resonated to some
ence oscillator, or by using the tuned circuit convenient frequency. The capacitance of
in the controlling position of a two termi- the standard capacitor is noted. Then the
nal oscillator such as a dynatron or transi- unknown capacitor is removed and the cir-
tron. The capacitance required to tune this cuit re- resonated by means of the standard
first frequency is then noted as C,. The cir- capacitor. The difference between the two
cuit or the oscillator is then tuned to the readings of the standard capacitor is then
second harmonic of this first frequency and equal to the capacitance of the unknown
the amount of capacitance again noted, this capacitor.
time as C. Then the distributed capaci-
tance across the coil (including all stray 31 -5 Measurements
capacitances) is equal to: C,, = (C, - with a Bridge
4C2) /3.
This value of distributed capacitance is Experience has shown that one of the
then substituted in the following formula most satisfactory methods for measuring
along with the value of the standard ca- circuit constants (resistance, capacitance,
756 Electronic Test Equipment THE RADIO
and inductance) at audio frequencies is by
means of the a -c bridge. The Wheatstone
(d -c) bridge is also one of the most ac-
curate methods for the measurement of d -c
resistance. With a simple bridge of the type
shown at figure 16A it is entirely practical
to obtain d -c resistance determinations ac-
curate to four significant figures. With an
a -c bridge operating within its normal rat-
ing as to frequency and range of measure- zx= Zs xx= fixs
xs Rx= Rs
ment it is possible to obtain results ac-
ZA= IMPEDANCE BEING MEASURED, RS = RESISTANCE Cri.PDNEM OF ZS
curate to three significant figures. Z5= IMPEDANCE OF STANDARD, XA =REACTANCE COMPONENT OF ZA
Both the a -c and the d -c bridges consist RA= RESISTANCE COMPONENT OF ZA, XS =REACTANCE COMPONENT OF Z5
of a source of energy, a standard or refer-
ence of measurement, a means of balancing
this standard against the unknown, and a
means of indicating when this balance has
been reached. The source of energy in the
d -c bridge is a battery; the indicator is a
sensitive galvanometer. In the a -c bridge
the source of energy is an audio oscillator
(usually in the vicinity of 1000 Hz), and
the indicator is usually a pair of headphones.
The standard for the d -c bridge is a resist-
ance, usually in the form of a decade box.
Standards for the a -c bridge can be re-
sistance, capacitance, and inductance in TWO A -C BRIDGE CIRCUITS
varying forms. The operation of these bridges is essentially
Figure 16 shows two general types of the the same as those of figure 16 except that
a.c. is fed into the bridge instead of d.c. and
Wheatstone or d -c bridge. In A the so- called a pair of phones is used as the indicator in-
"ratio arms" (RA and RB) are fixed (usual- stead of the galvanometer. The bridge shown
ly in a ratio of 1-to-1, 1- to -10, 1 -to -100, or at A can be used for the measurement of re-
sistance, but it is usually used for the meas-
1 -to-1000) and the standard resistor (Rs) is urement of the impedance and reactance of
varied until the bridge is in balance. In coils and capacitors at frequencies from 200
to 1000 Hz. The bridge shown at B is used
commercially manufactured bridges there for the measurement of small values of ca-
are usually two or more buttons on the gal- pacitance by the substitution method. Full
vanometer for progressively increasing its description of the operation of both bridges
is given in the accompanying text.
sensitivity as balance is approached. Figure
16B is the slide -wire type of bridge in which
fixed standards are used and the ratio arm is used for the measurement of resistance. It
continuously variable. The slide wire may is necessary with this type of bridge to use
actually consist of a moving contact along a standard which presents the same type of
a length of wire of uniform cross section in impedance as the unknown being measured:
which case the ratio of RA to R may be resistance standard for a resistance measure-
read off directly in centimeters or inches, or ment, capacitance standard for capacitance,
in degrees of rotation if the slide wire is and inductance standard for inductance
bent around a circular former. Alternative- determination.
ly, the slide wire may consist of a linear - For measurement of capacitances from a
wound potentiometer with its dial calibrated few picofarads to about 0.001 lid, a Wag-
in degrees or in resistance from each end. ner-grounded substitution capacitance bridge
Figure 17A shows a simple type of a -c of the type shown in figure 17B will be
bridge for the measurement of capacitance found satisfactory. The ratio arms RA and
and inductance. It can also, if desired, be R11 should be of the same value within 1
www.americanradiohistory.com
HANDBOOK Transistorized Capacitance Meter 757
31 -6 A Transistorized 2w
Capacitance Meter CORRECTOR
h Puss ro
p TEST
www.americanradiohistory.com
758 Electronic Test Equipment THE RADIO
www.americanradiohistory.com
HANDBOOK Frequency Measurement 759
Figure 21
munications receiver and a SO kHz, 100 - noting whether or not the harmonics of the
kHz, or 200 -kHz band -edge spotter. The oscillator in the amateur bands fall on the
low- frequency oscillator may be self -excited approximate calibration marks of the re-
if desired, but low- frequency- standard crys- ceiver.
tals have become so relatively inexpensive A simple frequency spotter is diagramed
that a reference crystal may be purchased in figure 21.
for very little more than the cost of the
components for a self- excited oscillator. 31 -8 Antenna and
The crystal has the additional advantage Transmission -
that it may be once set so that its har-
monics are at zero beat with WWV and
Line Measurements
then left with only an occasional check to
see that the frequency has not drifted more The degree of adjustment of any amateur
than a few Hz. The self- excited oscillator, antenna can be judged by the study of the
on the other hand, must be monitored very standing -wave ratio on the transmission line
frequently to ensure that it is on frequency. feeding the antenna. Various types of instru-
ments have been designed to measure the
Using a To use a frequency spotter it is SWR present on the transmission line, or to
Frequency only necessary to couple the out- measure the actual radiation resistance of
Spotter put of the oscillator unit to the the antenna in question. The most important
antenna terminal of the receiver of these instruments are the slotted line, the
through a very small capacitance such as bridge -type SWR meter, and the antenna-
might be made by twisting two pieces of scope.
insulated hookup wire together. Station
WWV is then tuned in on one of its har- The Slotted Line It is obviously impractical
monics, 15 MHz will usually be best in to measure the standing -
the daytime and S or 10 MHz at night, wave ratio in a length of coaxial line since
and the trimmer adjustment on the oscil- the voltages and currents inside the line are
lator is varied until zero beat is obtained be- completely shielded by the outer conductor
tween the harmonic of the oscillator and of the cable. Hence it is necessary to insert
WWV. With a crystal reference oscillator some type of instrument into a section of
no difficulty will be had with using the the line in order to be able to ascertain the
wrong harmonic of the oscillator to obtain conditions which are taking place inside the
the beat, but with a self- excited oscillator shielded line. Where measurements of a
it will be wise to ensure that the reference high degree of accuracy are required, the
oscillator is operating exactly on 50, 100, slotted line is the instrument most frequent-
or 200 kHz (whichever frequency has been ly used. Such an instrument, diagramed in
chosen) by making sure that zero beat is figure 22, is an item of test equipment which
obtained simultaneously on all the frequen- could be constructed in a home workshop
cies of WWV that can be heard, and by which includes a lathe and other metal-
760 Electronic Test Equipment THE RADIO
COAX
flrrlNG TAPER PROBE INNER DOCTOR TAPER COAX
FITTING 420 MHz and above where a full wave-
length is 28 inches or less. But for the lower
frequencies such an instrument is mechani-
CARRIER SLOT IN
FOR PROBE SLIDER
CARRYING
CONDUCTOR cally impractical.
PROBE
Bridge -Type The bridge type of stand -
Standing -Wave ing -wave indicator is used
Indicators quite generally for making
measurements on commer-
Figure 22 cial coaxial transmission lines. A simplified
version is available from M. C. Jones Elec-
DIAGRAMATIC REPRESENTATION
tronics Co., Bristol, Conn. ( "Micro- Match ") .
OF A SLOTTED LINE
One type of bridge standing -wave indi-
The conductor ratios in the slotted line, in- cator is diagramed in figure 23. This type
cluding the tapered end sections should be
such that the characteristic impedance of the of instrument compares the electrical im-
equipment is the same as that of the trans- pedance of the transmission line with that of
mission line with which the equipment is to
be used. The indicating instrument may be the resistor R. which is included within the
operated by the d-c output of the rectifier unit. Experience with such units has shown
coupled to the probe, or it may be operated that the resistor R;, should be a good grade
by the a -c components of the rectified signal
if the signal g for or transmitter is am- of noninductive carbon type. The Ohmifc
plitude- modulated at a constant percentage. "Little Devil" type resistor in the 2 -watt
rating has given good performance. The
working tools. Commercially built slotted resistance at R. should be equal to the
lines are very expensive since they are con- characteristic impedance of the antenna
structed with a high degree of accuracy for transmission line. In other words, this resis-
precise laboratory work. The slotted line tor should have a value of 52 ohms for lines
consists essentially of a section of air- dielec- having this characteristic impedance such
tric line having the same characteristic im-
pedance as the transmission line into which
it is inserted. Tapered fittings for the trans- OUTPUT
TO
ANTENNA
mission line connectors at each end of the FEED LINE
www.americanradiohistory.com
HANDBOOK Frequency Measurements 761
lu
the same as before. If the test resistor is
15
13 matched to R:1, and stray capacitances have
!o
ie been held to low values, the indication on
15 the milliammeter will be very small. The
12
test plug with its resistor is removed and
10 the plug for the antenna transmission line
9
e
is inserted. The meter indication now will
7 read the reflection coefficient which exists on
e
the antenna transmission line at the point
s
where the indicator has been inserted. From
4
this reading of reflection coefficient the
3S
3
actual standing -wave ratio on the transmis-
sion line may be determined by reference to
-
23
2 A _
ZR - ZO the chart of figure 24.
ZR + ZO
Measurements of this type are quite help-
+ IAI
- ful in determining whether or not the anten-
1
1 IAI
tS
na is presenting a good impedance match to
0 01 02 03 04 05 06 07 0.6 09 10 the transmission line being used to feed it.
READING ON 0-1 INSTRUMENT, OR FACTOR TIMES FULL SCALE
However, a test instrument of the type
(MAGNITUDE OF REFLECTION COEFFICIENT, A) shown in figure 23 must be inserted into
the line for a measurement, and then re-
Figure 24
COEFFICIENT
IMP INNER CONDUCTOR ou1
This chart may be used to convert reflection- JI J2
coefficient Indications such as are obtained D1
with a bridge -type standing-wave indicator or 270 270 ,i
an indicating twin lamp into values of stand- ` 141
R2 IOM
as RG-8 /U and RG -58 /U. For use with
lines having a nominal characteristic im-
pedance of 70 ohms, a selected "68 -ohm"
resistor having an actual resistance of 70
ohms may be used. SI
FDRWARD REFLECTED!
Balance within the equipment is checked POWER POWER
receptacle of the instrument and a power INNER CONDUCTOR= 1/2 " DIAMETER x 5 /B0, TAPER ENDS
1
ci.r
sm wow
Figure 26
moved from the line when the equipment the degree of match or mismatch and all
is to be operated. Also, the power input to antenna and transmission -line adjustments
the line feeding the input terminal of the should be conducted so as to make this
standing -wave indicator must not exceed 4 reading as low as possible, regardless of its
watts. The power level which the unit can absolute value.
accept is determined by the dissipation lim- The actual meter readings obtained from
itation of resistors R, plus R_. the device are a function of the operating
It is also important, for satisfactory op- frequency, the sensitivity of the instrument
eration of the test unit, that resistors R, being a function of transmitter power, in-
and R_ be exactly equal in value. The actual creasing rapidly as the frequency of opera-
resistance of these two is not critically im- tion is increased. However, the reflectometer
portant, and deviations up to 10 percent is invaluable in that it may be left perma-
from the value given in figure 23 will be nently in the transmission line, regardless
satisfactory. But the two resistors must have of the power -output level of the transmitter.
the same value, whether they are both 21 It will indicate the degree of reflected power
ohms or 24 ohms, or some value in between. in the antenna system, and at the same time
provide a visual indication of the power
31 -9 A Simple Coaxial output of the transmitter.
Reflectometer
The reflectometer is a short section of co- Reflectometer The circuit and assembly in-
axial transmission line containing two r -f Circuit formation for the reflectome-
voltmeters. One voltmeter reads the incident ter are given in figure 25.
component of voltage in the line, and the Two diode voltmeters are coupled back -to-
other reads the reflected component. The back to a short length of transmission line.
magnitude of standing -wave ratio on the The combined inductive and capacitive pick-
transmission line is the ratio of the incident up between each voltmeter and the line is
component to the reflected component, as such that the incident component of the
shown in figure 24. In actual use, calibration line voltage is balanced out in one case and
of the reflectometer is not required since the the inductive component is balanced out in
relative reading of reflected power indicates the other case. Each voltmeter, therefore,
www.americanradiohistory.com
HANDBOOK Coaxial Reflectometer 763
reads only one wave -component. Careful serve that the resistor leads are of equal
attention to physical symmetry of the length and that each half of the assembly is
assembly ensures accurate and complete a mirror -image of the other half. The body
separation of the voltage components by the of the resistor is spaced about '/8-inch away
two voltmeters. The outputs of the two from the center conductor.
voltmeters may be selected and read on an
Testing the The instrument can be ad-
external meter connected to the terminal
Reflectometer justed on the 28 MHz band.
posts of the reflectometer.
Each r -f voltmeter is composed of a load An r -f source of a few watts
resistor and a pickup loop. The pickup loop and nonreactive load are required. The con-
is positioned parallel to a section of trans- struction of the reflectometer is such that
mission line permitting both inductive and it will work well with either 52- or 72 -ohm
capacitive coupling to exist between the coaxial transmission lines. A suitable dummy
center conductor of the line and the loop. load for the 52 -ohm line can be made of
The dimensions of the center conductor and four 220 -ohm, 2 -watt composition resistors
the outer shield of the reflectometer are (Ohmite "Little Devil ") connected in par-
chosen so that the instrument impedance allel. Clip the leads of the resistors short and
closely matches that of the transmission mount them on a coaxial plug. This assembly
line. provides an eight watt, 55-ohm load, suit-
able for use at 30 MHz. If an accurate ohm-
Reflectometer A view of the interior of the meter is at hand, the resistors may be hand
Construction reflectometer is shown in fig- picked to obtain four 208 -ohm units, thus
ure 26. The coaxial input and making the dummy -load resistors exactly
output connectors of the instrument are 52 ohms. For all practical purposes, the SS-
mounted on machined brass discs that are ohm load is satisfactory. A 75 -ohm, eight -
held in place by brass alignment rods, tap- watt load resistor may be made of four
ped at each end. The center conductor is 300 -ohm, 2 -watt composition resistors con-
machined from a short section of brass rod, nected in parallel.
tapered and drilled at each end to fit over R -f power is coupled to the reflectometer
the center pin of each coaxial receptacle. and the dummy load is placed in the "out-
The end discs, the rods, and the center con- put" receptacle. The indicator meter is
ductor should be silver plated before as- switched to the "reflected- power" position.
sembly. When the center conductor is placed The meter reading should be almost zero.
in position, it is soldered at each end to the It may be brought to zero by removing the
center pin of the coaxial receptacles. case of the instrument and adjusting the
One of the alignment rods is drilled position of the load resistor. The actual
and tapped for a 6 -32 bolt at the midpoint, length of wire in the resistor lead and its
and the end discs are drilled to hold "A-inch positioning determine the meter null. Re-
ceramic insulators and binding posts, as place the case before power is applied to the
shown in the photograph. The load resistors, reflectometer. The reflectometer is now re-
crystal diodes, and bypass capacitors are fi- versed and power is applied to the "output"
nally mounted in the assembly as the last receptacle, with a dummy load attached to
step. the "input" receptacle. The second volt-
The two load resistors should be measured meter (forward power) is adjusted for a
on an ohmmeter to ensure that the resistance null reading of the meter in the same
values are equal. The exact value of resist- manner.
ance is unimportant as long as the two re- If a reflected reading of zero is not ob-
sistors are equal. The diodes should also be tainable, the harmonic content of the r -f
checked on an ohmmeter to make sure that source might be causing a slight residual
the front resistances and back resistances meter reading. Coupling the reflectometer
are balanced between the units. Care should to the r -f source through a tuned circuit
be taken during soldering to ensure the ( "antenna tuner ") will remove the offend-
diodes and resistors are not overheated. Ob- ing harmonic and permit an accurate null
764 Electronic Test Equipment THE RADIO
www.americanradiohistory.com
HANDBOOK The Antennascope 765
relay to the antenna system. When there true value of standing -wave ratio on the
are no standing waves on the antenna trans- line. If these conditions are not met, the
mission line the lamp toward the transmitter reading is relative, giving an indication of
will light while the one toward the antenna the degree of mismatch on the line. This
will not light. With 300 -ohm twin -lead handicap is not important since the relative,
running from the antenna changeover relay not the absolute, degree of mismatch is suffi-
to the antenna, and with about 200 watts cient for transmission -line adjustments to
input on the 28 -MHz band, the dial lamp be made.
toward the transmitter will light nearly to The bridge may be calibrated with a
full brilliance. With a standing -wave ratio grid -dip oscillator, and with various values
of about 1.5 to 1 on the transmission line of carbon resistors used as a load.
to the antenna the lamp toward the anten-
na will just begin to light. With a high 31 -11 The Antennascope
standing -wave ratio on the antenna feed line
both lamps will light nearly to full brilli- The Antennascope is a modified SWR
ance. Hence the instrument gives an indi- bridge in which one leg of the bridge is
cation of relatively low standing waves, but composed of a noninductive variable resistor.
when the standing -wave ratio is high the This resistor is calibrated in ohms, and when
its setting is equal to the radiation resistance
twin -lamp merely indicates that they are
high without giving any idea of the actual of the antenna under test, the bridge is in a
magnitude. balanced state. If a sensitive voltmeter is
connected across the bridge, it will indicate
A Balanced Two standing-wave indicators a voltage null at bridge balance. The radia-
SWR Bridge may be placed back -to -back to tion resistance of the antenna may then be
form an SWR bridge capable read directly from the calibrated resistor of
of being used on two -wire balanced trans- the instrument.
mission lines (figure 29). When the dual When the antenna under test is in a non -
bridge is balanced the meter reading is zero. resonant or reactive state, the null indica-
This state is reached only when the line tion on the meter of the Antennascope will
currents are equal and 180 degrees out of be incomplete. The frequency of the exciting
phase and the SWR is unity. When the line signal must then be moved to the resonant
currents are balanced and 180 degrees out frequency of the antenna to obtain accurate
of phase, the balanced bridge will read the readings of radiation resistance from the
dial of the instrument.
Ri
250
250 6uT
Ri
Figure 29
Figure 30
SCHEMATIC OF BRIDGE FOR
BALANCED LINES SCHEMATIC, ANTENNASCOPE
M-0 -200 d-c microammeter R,-1000 -ohm composition potentiometer Ohm-
R, -Hots: Six 250 -ohm resistors are composi- ite type AB or Allen Bradley type J, linear
tion, noninductive units. 1RC type ST, or taper
Ohmite "Little Devil" 1 -watt resistors may R. -SO -ohm, 1 -watt composition resistor, !RC
be used. (see text) type IT, or Ohmite "Little Devil" (see text)
S Dpdt rotary switch. Centralab type 1464 M -0 -200 d-c microammeter
766 Electronic Test Equipment THE RADIO
www.americanradiohistory.com
HANDBOOK The Antennascope 767
Figure 33
www.americanradiohistory.com
HANDBOOK Monitor Scope 769
+30V
audio volume control is adjusted until the IIH
CENTERING
output meter advances to one -quarter scale. CONTROLS
This reading is the basic receiver noise. The 220K 2API 001 R-F
Figure 38
UNDER -CHASSIS
VIEW OF
OSCILLOSCOPE
Filament transformer is
mounted directly behind
scope tube so os to not
distort electron beam.
Centering controls are
mounted on phenolic
board on chassis edge.
Controls are covered af-
ter adjustment to elimi-
nate shock hazard.
is designed to sit atop the receiver. R -f con- computed as shown in figure 40. This data
nection to the transmitter may be made by may be compared with the information
inserting a coaxial "Tee" in the transmission listed in the transistor data sheet to deter-
line and running a short length of similar mine the condition of the transistor under
line from the "Tee" to the scope. Operation test.
of the scope and its uses are covered in
chapter 8, "The Oscilloscope."
31 -14 An Inexpensive
Transistor Checker
This inexpensive and compact transistor
checker will measure the d -c parameters of
most common transistors. Either NPN or
PNP transistors may be checked. A six -
position test switch permits the following
parameters to be measured: (1) I 'o -D -c
collector current when collector junction
is reverse -biased and emitter is open cir-
cuited; (2) I,.0_2,,-collector current when
base current is 20 microamperes; (3) I, -o -Inn
-collector current when base current is Figure 39
100 microamperes; (4) I,.EO- collector cur- TRANSISTOR CHECKER
rent when collector junction is reverse - An expanded-scale meter provides accurate
biased and base is open circuited; (5) Lis measurement of transistor parameters in this
-collector current when collector junction easily built instrument. Six d -c parameters
may be measured and with the data derived
is reverse -biased and base is shorted to emit- from these tests, the a -c forward -current trans-
ter; (6) IEO -emitter current when emitter fer ratios may be computed. Two transistor
sockets are mounted at the left of the tester,
junction is reverse- biased and collector is with the three selector switches to the right.
open circuited. Six -position test switch is mounted to bot-
Using the data derived from these tests, tom side of box. Tip jacks are placed in par-
allel with transistor socket terminals to per-
the static and a -c forward-current transfer mit test of transistors having unorthodox
ratios (hFE and hlr, respectively) may be bases.
www.americanradiohistory.com
HANDBOOK Transistor Checker 771
PUSH TO READ
1.2K J=L
METER ADJUST
TO ADJUST RESULT
TEST WHEN St
ICO VCB =6 V. I READ METER DIRECT
I C 1ST 2OLA 2 Figure 41
Ic IB =tooLA 3 "
ICED VCE =6 V. 4 ' INTERIOR VIEW OF
ICES VCE =B V. 9
Ieo VEO =eV. e '
CALCULATE.
TRANSISTOR CHECKER
hFE 1B= 2O LA 2 hFE = Components of meter diode circuit are
IIC LEADING
13
mounted to phenolic board attached to meter
CALCULATE
hFE 1B =100 LA -
terminals. Other small resistors may be wired
3
hFE_ IC METER READING
directly to switch lugs. The four II 2-volt bat-
le 100 LA
CALCULATE: WHERE: teries are held in o small clamp at the rear
ht. I Br 20SIA 2 _ ICI-IC2 ICI. METER of the case. Chassis is cut out for lever -action
hh .17-3-74r READING switches and opening is covered with three -
CALCULATE IC2 =METER position switch plate.
ICI -Ica READING WITH
-
ht. le'IDDLA 3 hN -20 :Oar S4 CLOSED
WITH 13011 RESISTOR CON - equal to about 6K, the meter scale is com-
eV NECTED TO C -E OF TEST
4 SOCKET, FULL-SCALE METER
BATTERY
DEFLECTION WILL RESULT pressed only one microampere at 20 micro-
WHEN Ss IS PRESSED
amperes. Meter adjust potentiometer R_ is
set to give 10 milliamperes full -scale meter
Figure 40 deflection. The scale may then be calibrated
by comparison with a conventional meter.
SCHEMATIC OF TRANSISTOR CHECKER If the NPN -PNP switch (S_) is in the
S,A, B, C- Three -pole, 6- position. Centralab wrong position, the collector and emitter
1021
S, S , S -Centralab type 1400 nonshorting junctions will be forward biased during the
lever switch 'co and IF:o tests (switch positions 1 and 6).
M -0 -200 d -c microammeter. General Electric
or Simpson (41/2")
The high resulting current may be used as
a check for open or intermittent connections
within the transistor.
The transistor parameters are read on a The transistor checker also measures hrE
0 -100 d -c microammeter placed in a diode with 20 microamperes and 100 microam-
network which provides a nearly linear scale peres base current. Depressing the fife switch
to 20 microamperes, a highly compressed (S,) decreases the base drive about 20 per-
scale from 20 microamperes to one milliam- cent, permitting h1,. to be estimated from
pere, and a nearly linear scale to full scale the corresponding change in collector cur-
at 10 milliamperes. Transistor parameters rent (formulas 1 and 2). All tests are con-
may be read to within 10 percent on all ducted with a 330 -ohm resistor limiting the
transistor types from mesas to power alloys collector current to about 12 milliamperes
without switching meter ranges and with- and the maximum transistor dissipation to
out damage to the meter movement or about 20 milliwatts. The checker therefore
transistor. cannot harm a transistor regardless of how
By making the sum of the internal resist- it is plugged in or how the test switches are
ance of the meter plus series resistor R, set.
772 Electronic Test Equipment
The battery test provides full -scale meter top of the box. Three insulated tip jacks
deflection of 10 milliamperes when the bat- are wired to the leads of one transistor test
tery potential is 6 volts. This is achieved by socket so that transistors having unorthodox
connecting a 150-ohm resistor from collector bases or leads may be clipped to the tester
to emitter of a test socket. by means of short test leads. Four 1 1/2-volt
flashlight cells are mounted to the rear of
Test Set The transistor checker is built the case by an aluminum clamp. Potenti-
Construction in an aluminum box measur- ometer R_, the meter diode, and associated
ing 3" X f" X 7 ", as shown components are fastened to a phenolic
in the photographs. Test switch S, is board attached to the meter terminals.
mounted on the end of the box; and the Switch S, has an indicator scale made of
transistor sockets, microammeter, and the heavy white cardboard, lettered with India
various other switches are placed on the ink and a lettering pen.
www.americanradiohistory.com
CHAPTER THIRTY -TWO
www.americanradiohistory.com
HANDBOOK Tools and Materials 775
Figure 1
SOFT ALUMINUM
SHEET MAY BE CUT
WITH HEAVY
KITCHEN SHEARS
grinding head, etc. If power equipment is pur- tube sockets mount in a 5/8 -inch hole, while
chased, obviously some of the hand tools and 9 -piu sockets mount in a 3/4-inch hole. Green-
accessories listed will be superfluous. A drill lee socket punches can be obtained in these
press greatly facilitates construction work, sizes, or a smaller hole may be reamed to the
and it is unfortunate that a good one costs as proper size. Needless to say, the punch is
much as a small transmitter. much the more satisfactory solution. Mounting
Not listed in the table are several special - screws for miniature sockets are usually of
purpose radio tools which are somewhat of a the 4 -40 size.
luxury, but are nevertheless quite handy, such
as various around - the -corner screwdrivers and Metal Chassis Though quite a few more tools
wrenches, special soldering iron tips, etc. and considerably more time
These can be found in the larger radio parts will be required for metal chassis construction,
stores and are usually listed in their mail or- much neater and more satisfactory equipment
der catalogs. can be built by mounting the parts on sheet
If it is contemplated to use the newer and metal chassis instead of breadboards. This type
very popular miniature series of tubes (6AK5, of construction is necessary when shielding of
6C4, 6B6, etc.) in the construction of equip- the appartus is required. A front panel and a
ment certain additional tools will be required back shield minimize the danger of shock and
to mount the smaller components. Miniature complete the shielding of the enclosure.
Figure 2
CONVENTIONAL
WOOD EXPANSION
BIT IS EFFECTIVE IN
DRILLING SOCKET
HOLES IN REYNOLDS
DO -IT- YOURSELF
ALUMINUM
776 Workshop Practice THE RADIO
Figure 3
SOFT ALUMINUM
TUBING MAY BE
BENT AROUND
WOODEN FORM
BLOCKS. TO PREVENT
THE TUBE FROM
COLLAPSING ON
SHARP BENDS, IT IS
PACKED WITH
WET SAND.
Figure 4
A WOODWORKING
PLANE MAY BE USED
TO SMOOTH OR
TRIM THE EDGES OF
REYNOLDS
DO -IT- YOURSELF
ALUMINUM STOCK.
www.americanradiohistory.com
HANDBOOK Enclosures 777
COMPONENT PARTS
1. Legs and slnngers-
aluminum angle %"N %"
DESK 2. Top -flush door
3. Shelves- W plywood
6'O-
OI%"
Figure 6
TVI ENCLOSURE MADE FROM
i-
SINGLE SHEET OF
I-1'r PERFORATED ALUMINUM
76 Reynolds Metal Co. "Do -it- yourself" aluminum
sheet may be cut and folded to form TVI -
proof enclosure. One-half inch lip on edges Is
bolted to center section with 6 -32 machine
screws.
J 1'6'
f-
O6-
32 -4 Enclosure
Openings
Openings into shielded enclosures may be
made simply by covering them by a piece of
shielding held in place by sheet metal screws.
Openings through vertical panels, however,
usually require a bit more attention to pre-
vent leakage of harmonic energy through the
crack of the door which is supposed to seal
the opening. A simple way to provide a panel
opening is to employ the Bud ventilated door
rack panel model PS -814 or 815. The grille
opening in this panel has holes small enough
in area to prevent serious harmonic leakage.
The actual door opening, however, does not Figure 7
seal tightly enough to be called TVI- proof. TVI -PROOF ENCLOSURE BUILT OF
PERFORATED ALUMINUM SHEET
In areas of high TV signal strength where AND ANGLE STOCK
a minimum of operation on 28 Mc. is con-
templated, the door is satisfactory as is. To
accomplish more complete harmonic suppres-
sion, the edges of the opening should be lined 32-5 Summation
with preformed contact finger stock manufac- of the Problem
tured by Eitel- McCullough, Inc., of San Bruno,
Calif. Eimac finger stock is an excellent means The creation of r -f energy is accompanied
of providing good contact continuity when by harmonic generation and leakage of funda-
using components with adjustable or moving mental and harmonic energy from the generator
contact surfaces, or in acting as electrical source. For practical purposes, radio frequen-
"weatherstrip" around access doors in enclo- cy power may be considered as a form of both
sures. Harmonic leakage through such a sealed electrical and r -f energy. As electrical energy,
opening is reduced to a negligible level. The it will travel along any convenient conductor.
mating surface to the finger stock should be As r -f energy, it will radiate directly from the
free of paint, and should provide a good elec- source or from any conductor connected to the
trical connection to the stock. source. In view of this "dual personality" of
A second method of re- establishing elec- r -f emanations, there is no panacea for all
trical continuity across an access port is to forms of r -f energy leakage. The cure involves
employ Metex shielding around the mating both filtering and shielding: one to block the
edges of the opening. Metex is a flexible knit- paths of conducted energy, the other to pre-
ted wire mesh which may be obtained in vent the leakage of radiated energy. The proper
various sizes and shapes. This r -f gasket ma- combination of filtering and shielding can re-
terial is produced by Metal Textile Corp., duce the radiation of harmonic energy from a
Roselle, N.J. Metex is both flexible and resili- signal source some 80 decibels. In most cases,
ent and conforms to irregularities in mating this is sufficient to eliminate interference caused
surfaces with a minimum of closing pressure. by the generation of undesirable harmonics.
www.americanradiohistory.com
HANDBOOK Construction Prcctice 779
32 -6 Construction
Practice
Chassis The chassis first should be covered
Layout with a layer of wrapping paper,
which is drawn tightly down on FILL
DRILL HOLES SLIGHTLY BREAK OUT
all sides and fastened with scotch tape. This INSIDE DASHED OUTLINE
OF DESIRED HOLE.
PIECE INSIDE
DRILL HOLES.
SMOOTH
removing these. Perhaps the best is to take a pieces of apparatus to chassis or breadboards.
chisel (be sure it is one for use on metal) and The first, using nuts and machine screws, is
set it so that its bottom face is parallel to the slow, and the commercial manufacturing prac-
piece. Then gently tap it with a hammer. This tice of using self- tapping screws is gaining
usually will make a clean job with a little favor. For the mounting of small parts such
practice. If one has access to a counterbore, as resistors and capacitors, "tie points" are
this will also do a nice job. A countersink will very useful to gain rigidity. They also con-
work, although it bevels the edges. A drill of tribute materially to the appearance of finished
several sizes larger is a much used arrange- apparatus.
ment. The third method is by filing off the Rubber grommets of the proper size, placed
burr, which does a good job but scratches the in all chassis holes through which wires are
adjacent metal surfaces badly. to be passed, will give a neater appearing job
and also will reduce the possibility of short
Mounting There are two methods in gen- circuits.
Components eral use for the fastening of
transformers, chokes, and similar Soldering Making a strong, low- resistance
solder joint does not mean just
dropping a blob of solder on the two parts to
NUMBERED DRILL SIZES be joined and then hoping that they'll stick.
Correct for There are several definite rules that must be
--
DI- rapping
DRILL meter Clears Steel or observed.
NUMBER (In.) Screw
-
Brout
-
All parts to be soldered must be absolutely
--
1 .226 clean. To clean a wire, lug, or whatever it may
2 .221 12 -24
3
4
S
.213
.209
.205
12 -20 -
14 -24
-
be, take your pocket knife and scrape it thor-
oughly, until fresh metal is laid bare. It is not
6 .204 enough to make a few streaks; scrape until the
7 .201
.199 part to be soldered is bright.
--
6
Make a good mechanical joint before apply-
--
9 .196
10 .193 10 -32
11 .191 10-24 ing any solder. Solder is intended primarily
12' to make a good electrical connection; mechani-
--
.169
--
13 .165
14 .162 cal rigidity should be obtained by bending the
15 .160 wire into a small hook at the end and nipping
16 .177 12 -24
17 .173 it firmly around the other part, so that it will
--
111. .169 6-32 hold well even before the solder is applied.
19 .166 12 -20
20 .161 Keep your iron properly tinned. It is im-
21. .159 10-32 possible to get the work hot enough to take
--
22 .157
23
24
25.
.154
.152
.149
-- 10 -24
the solder properly if the iron is dirty. To tin
your iron, file it, while hot, on one side until
a full surface of clean metal is exposed. Im-
--
26 .147
27 .. .144
211 .140 6-32 mediately apply rosin core solder until a thin
29 8-32 layer flows completely over the exposed sur-
30 in
.136
--
31 .120 face. Repeat for the other faces. Then take a
32 .116 clean rag and wipe off all excess solder and
33 .113 4 -36 4 -40
rosin. The iron should also be wiped frequently
--
34 .111
35. .110 6-32 while the actual construction is going on; it
36 .106
37 .104 helps prevent pitting the tip.
Apply the solder to the work, not to the
- -
38 .102
39. .100 3.48
40 .096 iron. The iron should be held against the parts
41 .096 to be joined until they are thoroughly heated.
42 .093 4 -36 4-40
43 .089 2 -56 The solder should then be applied against the
44 .066 parts, and the iron should be held in place
45 .062 3-46
until the solder flows smoothly and envelopes
tUw neat size larger drill for tapping
Bakelite and similar composition materials the work. If it acts like water on a greasy
(plastics, etc.). plate, and forms a ball, the work is not suf-
*Sizes most commonly used In radio con- ficiently clean.
The completed joint must be held perfectly
still until the solder has had time to solidify.
Figure 9 If the work is moved before the solder has be-
www.americanradiohistory.com
HANDBOOK Construction Practice 781
HOLD TIGHTLY
Figure 10 Figure 11
come completely solid, a "cold" joint will re- There are also several brands of dull gloss
sult. This can be identified immediately, be- black enamels on the market which adhere
cause the solder will have a dull "white" ap- well to metals and make a nice appearance.
pearance rather than one of shiny "silver." Airdrying wrinkle finishes are sometimes suc-
Such joints tend to be of high resistance and cessful, but a bake job is usually far better.
will very likely have a bad effect upon a cir- Wrinkle finishes, properly applied, are very
cuit. The cure is simple, merely reheat the durable and are pleasing to the eye. If you
joint and do the job correctly. live in a large community, there is probably
Wipe away all surplus flux when the joint an enameling concern which can wrinkle your
has cooled if you are using a paste type flux. work for you at a reasonable cost. A very at-
Be sure it is non- corrosive, and use it with tractive finish, for panels especially, is to
plain (not rosin core) solder. spray a wrinkle finish with aluminum paint.
In any painting operation (or plating, either,
Finishes If the apparatus is constructed on for that matter) , the work should be very
a painted chassis (commonly avail- thoroughly cleaned of all greases and oils.
able in black wrinkle and gray wrinkle), there To protect brass from tarnish, thoroughly
is no need for application of a protective coat- cleanse and remove the last trace of grease by
ing when the equipment is finished, assuming the use of potash and water. The brass must
that you are careful not to scratch or mar the be carefully rinsed with water and dried; but
finish while drilling holes and mounting parts. in doing it, care must be taken not to handle
However, many amateurs prefer to use un- any portion with the bare hands or anything
painted (zinc or cadmium plated) chassis, be- else that is greasy. Then lacquer.
cause it is much simpler to make a chassis
ground connection with this type of chassis. Winding Coils Coils are of two general types,
A thin coat of clear "linoleum" lacquer may those using a form and "air -
be applied to the whole chassis after the wir- wound" types. Neither type offers any particu-
ing is completed to retard rusting. In localities lar constructional difficulties. Figure 10 il-
near the sea coast it is a good idea to lacquer lustrates the procedure used in form winding
the various chassis cutouts even on a painted a coil. If the winding is to be spaced, the
chassis, as rust will get a good start at these spacing can be done either by eye or a string
points unless the metal is protected where the or another piece of wire may be wound simul-
drill or saw has exposed it. If too thick a coat taneously with the coil wire and removed after
is applied, the lacquer will tend to peel. It the winding is in place. The usual procedure
may be thinned with lacquer thinner to permit is to clamp one end of the wire in a vise, at-
application of a light coat. A thin coat will taching the other end to the coil form and with
adhere to any clean metal surface that is not the coil form in hand, walk slowly towards the
too shiny. vise winding the wire but at the same time
An attractive dull gloss finish, almost vel- keeping a strong tension on the wire as the
vety can be put on aluminum by sand -blasting form is rotated. After the coil is wound, if
it with a very weak blast and fine particles there is any possibility of the turns slipping,
and then lacquering it. Soaking the aluminum the completed coil is either entirely coated
in a solution of lye produces somewhat the with a coil or Duco cement or cemented in
same effect as a fine grain sand blast. those spots where slippage might occur.
782 Workshop Practice THE RADIO
Figure 12
GOOD SHOP
LAYOUT AIDS
CAREFUL
WORKMANSHIP
Built in a corner of a
garage, this shop has all
features necessary for
electronic work. Test in-
struments are arranged
on shelves above bench.
Numerous outlets reduce
"haywire" produced by
tangled line cords. Not
shown in picture are drill
press and sander at end
of left bench
V -h-f and u -h -f coils are commonly wound framework made of 2 "x4" lumber. The top
of heavy enameled wire on a form and then of the workbench is covered with hard -sur-
removed from the form as in figure 11. If face Masonite. The edge of the surface is pro-
the coil is long or has a tendency to buckle, tected with aluminum "counter edging" strip,
strips of polystyrene or a similar material may obtainable at large hardware stores. Two wood-
be cemented longitudinally inside the coil. Due en shelves 12" wide are placed above the
allowance must be made for the coil springing bench to hold the various items of test equip-
out when removed from the form, when select- ment. The shelves are bolted to the wall studs
ing the diameter of the form. with large angle brackets and have wooden
On air wound coils of this type, spacing be- end pieces. Along the edge of the lower shelf
tween turns is accomplished after removal a metal "outlet strip" is placed that has an
from the form, by running a pencil, the shank 115 -volt outlet every six inches along its
of a screwdriver or other round object spirally length. A similar strip is run along the back
between the turns from one end of the coil to of the lower shelf. The front strip is used
the other, again making due allowance for for equipment that is being bench -tested, and
spring. the rear strip powers the various items of test
Air -wound coils, approaching the appearance equipment placed on the shelves.
of commercially manufactured ones, can be At the left of the bench is a storage bin
constructed by using a round wooden form for small components. A file cabinet can be
which has been sawed diagonally from end seen at the right of the photograph. This nec-
to end. Strips of insulating material are tem- essary item holds schematics, transformer data
porarily attached to this mandrel, the wire sheets, and other papers that normally are
then being wound over these strips with the lost in the usual clutter and confusion.
desired separation between turns and cemented The area below the workbench has two
to the strips. When dry, the split mandrel may
storage shelves which are concealed by sliding
be removed by unwedging it. doors made of V4 -inch Masonite. Heavier tools,
and large components are stored in this area.
32 -7 Shop Layout On the floor and not shown in the photo-
The size of your workshop is relatively un- graph is a very necessary item of shop equip-
important since the shop layout will deter- ment: a large trash receptacle.
mine its efficiency and the ease with which A compact and efficient shop built in one-
you may complete your work. half of a wardrobe closet is shown in figure
Shown in figure 12 is a workshop built 13. The workbench length is four feet. The
into a 10'x10' area in the corner of a garage. top is made of 4 "x6" lumber sheathed with
The workbench is 32" wide, made up of four hard surface Masonite and trimmed with
strips of 2 "x8" lumber supported on a solid "counter edging" strip. The supporting struc-
www.americanradiohistory.com
HANDBOOK Shop Layout 783
Figure 13
COMPLETE WORKSHOP IN A CLOSET!
Careful layout permits complete electronic
workshop to be placed in one-half of a ward-
robe closet. Work bench is built atop an un-
painted three -shelf bookcase.
Radio Mathematics
and Calculations
Radiomen often have occasion to cal- four tens, plus three units, which could be
culate sizes and values of required parts. This written as follows:
requires some knowledge of mathematics. The
following pages contain a review of those parts 8 thousands (10 x 10 x 10)
of mathematics necessary to understand and 1 hundreds (10 x 10 )
784
www.americanradiohistory.com
Decimal Fractions 785
o o.s I 2 1 4
1 I t I I I L I I I
s-0.7-- b
Figure 1.
AN ILLUSTRATION OF LINEAR FRACTIONS.
varying factors in accordance with complicat- being the clearer. Every time a digit is placed
ed rules, The English system of measurements one space further to the right it represents a
is such an inconsistent and inferior system. ten times smaller part. This is illustrated in
Figure 1, where each large division represents
Decimal Fractions Since we can extend a a unit; each unit may be divided into ten parts
number indefinitely to although in the drawing we have only so di-
the left to make it bigger, it is a logical step to vided the first part. The length ab is equal to
extend it towards the right to make it smaller. seven of these tenth parts and is written as 0.7.
Numbers smaller than unity are fractions and The next smaller divisions, which should be
if a displacement one position to the right di- written in the second column to the right of
vides its value by ten, then the number is re- the decimal point, are each one -tenth of the
ferred to as a decimal fraction. Thus a digit small division, or one one- hundredth each.
to the right of the units column indicates the They are so small that we can only show them
number of tenths, the second digit to the right by imagining a magnifying glass to look at
represents the number of hundredths, the third, them, as in Figure 1. Six of these divisions is
the number of thousandths, etc. Some distin- to be written as 0.06 (six hundredths). We
guishing mark must be used to divide unit from need a microscope to see the next smaller divi-
tenths so that one may properly evaluate each sion, that is those in the third place, which will
symbol. This mark is the decimal point. be a tenth of one one -hundredth, or a thou-
A decimal fraction like four- tenths may be sandth; four such divisions would be written
written .4 or 0.4 as desired, the latter probably as 0.004 (four thousandths).
A e
G F
o
0 o
D N E C
Figure 2.
IN THIS ILLUSTRATION FRACTIONAL PORTIONS ARE REPRESENTED IN THE
FORM OF RECTANGLES RATHER THAN LINEARLY.
ABCD = 1.0; GEED = 0.1; KJEH = 0.01; each small section within KJEH equals 0.001
786 Radio Mathematics and Calculations THE RADIO
It should not be thought that such numbers The result of the operation is called the
are merely of academic interest for very small product.
quantities arc common in radio work. From the examples to follow it will be obvi-
Possibly the conception of fractions may be ous that there are as many partial products
clearer to some students by representing it in as there are digits in the multiplier. In the fol-
the form of rectangles rather than linearly lowing examples note that the righthandmost
(see Figure 2). digit of each partial product is placed one
space farther to the left than the previous one.
Addition When two or more numbers are
to be added we sometimes write 834 834
them horizontally with the plus sign between X 26 X 206
them. + is the sign or operator indicating ad-
dition. Thus if 7 and 12 are to be added to- 5004 5004
gether we may write 7+12=19. 1668 000
1668
But if lager or more numbers are to be added 21684
together they are almost invariably written one 171804
under another in such a position that the deci-
mal points fall in a vertical line. If a number In the second example above it will be seen
has no decimal point, it is still considered as that the inclusion of the second partial prod-
being just to the right of the units figure; such uct was unnecessary; whenever the multiplier
a number is a whole number or integer. Ex- contains a cipher (zero) the next partial prod-
amples: uct should be moved an additional space to
the left.
654 0.654 654
32 3.2 32 Numbers containing decimal fractions may
53041 53.041 5304.1 first be multiplied exactly as if the decimal
point did not occur in the numbers at all; the
53727 56.895 5990.1 position of the decimal point in the product is
determined after all operations have been com-
The result obtained by adding numbers is pleted. It must be so positioned in the product
called the rum. that the number of digits to its right is equal
to the number of decimal places in the multi-
Subtraction Subtraction is the reverse of
addition. Its operator is (the
minus sign) . The number to be subtracted is
- plicand plus the number of decimal places in
the multiplier.
This rule should be well understood since
called the subtrahend, the number from which many radio calculations contain quantities
it is subtracted is the minuend, and the result which involve very small decimal fractions. In
is called the remainder.
the examples which follow the explanatory
minuend notations "2 places," etc., are not actually
written down since it is comparatively easy to
-subtrahend determine the decimal point's proper location
mentally.
remainder
Examples: 5.43 2 places
X 0.72 2 places
65.4 65.4
-32 -32.21 1086
3 801
33.4 33.19
3.9096 2 +2 =4 places
Multiplication When numbers are to be mul-
tiplied together we use the x 0.04 2 places
which isknown as the multiplication or the X 0.003 3 places
times sign. The number to be multiplied is
known as the multiplicand and that by which 0.00012 2+ 3=5 places
it is to be multiplied is the multiplier, which Division Division is the reverse of multi-
may be written in words as follows: plication. Its operator is the =,
which is called the division sign. It is also com-
multiplicand mon to indicate division by the use of the frac-
X multiplier tion bar (/) or by writing one number over
partial product the other. The number which is to be divided
partial product is called the dividend and is written before
the division sign or fraction bar or over the
product horizontal line indicating a fraction. The num-
www.americanradiohistory.com
HANDBOOK Division 787
ber by which the dividend is to be divided is Another example: Divide 0.000325 by 0.017.
called the divisor and follows the division Here we must move the decimal point three
sign or fraction bar or comes under the hori- places to the right in both dividend and di-
zontal line of the fraction. The answer or visor.
result is called the quotient. 0.019
17 0.325
quotient 17
divisor dividend
155
or 153
dividend -divisor= quotient
2
or
dividend
= quotient In a case where the dividend has fewer deci-
divisor mals than the divisor the same rules still may
be applied by adding ciphers. For example to
Examples: divide 0.49 by 0.006 we must move the
126 49 decimal point three places to the right. The
105084 49 TiUg 0.49 now becomes 490 and we write:
834
834 196
81
2168 476 6ri9l0
1668 441 48
5004 35 remainder 10
5004 6
4
Note that one number often fails to divide
into another evenly. Hence there is often a
quantity left over called the remainder. When the division shows a remainder it is
The rules for placing the decimal point sometimes necessary to continue the work so
are the reverse of those for multiplication. as to obtain more figures. In that case ciphers
The number of decimal places in the quotient may be annexed to the dividend, brought down
is equal to the difference between the number to the remainder, and the division continued
of decimal places in the dividend and that in as long as may be necessary; be sure to place
the divisor. It is often simpler and clearer a decimal point in the dividend before the
to remove the decimal point entirely from the ciphers are annexed if the dividend does not
divisor by multiplying both dividend and di- already contain a decimal point. For ex-
visor by the necessary factor; that is we move ample:
the decimal point in the divisor as many places 8033
to the right as is necessary to make it a whole 6 482.00
number and then we move the decimal point 48
in the dividend exactly the same number of
places to the right regardless of whether this 20
18
makes the dividend a whole number or not.
When this has been done the decimal point 20
in the quotient will automatically come di- 18
rectly above that in the dividend as shown in
the following example. 2
Example: Divide 10.5084 by 8.34. Move the
decimal point of both dividend and divisor This operation is not very often required
two places to the right. in radio work since the accuracy of the mea-
surements from which our problems start
1.26 seldom justifies the use of more than three
834 753 0.84 significant figures. This point will be cov-
834 ered further later in this chapter.
2168 Fractions Quantities of less than one
1668 (unity) are called fractions. They
5004 may be expressed by decimal notation as we
5004 have seen, or they may be expressed as vulgar
fractions. Examples of vulgar fractions:
788 Radio Mathematics and Calculations THE RADIO
numerator 3 6 1 the numerator is added. That is in the above
denominator 4 7 5 example we multiply 2 by 7 and then add 3
to obtain 17 for the numerator. The denomi-
The upper position of a vulgar fraction is nator is the same as is the denominator of
called the numerator and the lower position the original fraction. In the following ex-
the denominator. When the numerator is the ample we have added two mixed numbers.
smaller of the two, the fraction is called a
proper fraction; the examples of vulgar frac- 3 3 17 1s r17x4 15x7]
tions given above are proper vulgar fractions. 2 7+ 3 4= 7+ 4 -L 7x4 + 4x7
When the numerator is the larger, the ex-
pression is an improper fraction, which can
be reduced to an integer or whole number
- 68
21 +
105
21 =
173
2$ =
S
6 2S
with a proper fraction, the whole being called Multiplying All vulgar fractions are multi -
a mixed number. In the following examples Fractions plied by multiplying the nu-
improper fractions have been reduced to merators together and the de-
their corresponding mixed numbers. nominators together, as shown in the follow-
ing example:
= 3-
1 1
Adding or Subtracting
4
3-
ly be found much easier to add and subtract be performed mentally. As with addition and
fractions in the form of decimals. This rule subtraction any mixed numbers should be
likewise applies for practically all other oper- first reduced to improper fractions as shown
ations with fractions. However, it is occa- in the following example:
sionally necessary to perform various opera-
tions with vulgar fractions and the rules
should be understood.
3
23 x 4 3
1
- 3
23 x 13
3
39
69 -23
13
When adding or subtracting such fractions Division of Fractions may be most easily
the denominators must be made equal. This Fractions divided by inverting the di-
may be done by multiplying both numerator visor and then multiplying.
and denominator of the first fraction by the Example:
denominator of the other fraction, after
which we multiply the numerator and de-
nominator of the second fraction by the de-
2
S
3
4 - 2
5 x 3-
4
1S
nominator of the first fraction. This sounds In the above example it will be seen that to
more complicated than it usually proves in divide by 3/4 is exactly the same thing as to
practice, as the following examples will show. multiply by 4/3. Actual division of fractions
is a rather rare operation and if necessary is
1x2 usually postponed until the final answer is se-
2+3_
1 1 1 x 3
22Y3+3x2]-
3
6+ 6-
2 S
6 cured when it is often desired to reduce the
resulting vulgar fraction to a decimal frac-
3x3 2x4 t
3
4 - 2
S = 4 x 5 -5x4]= 20 15
- 20 =
7
20
tion by division. It is more common and
usually results in least overall work to re-
Except in problems involving large numbers duce vulgar fractions to decimals at the be-
the step shown in brackets above is usually ginning of a problem. Examples:
done in the head and is not written down.
Although in the examples shown above we = 0.375 32
= 0.15625
have used proper fractions, it is obvious that
the same procedure applies with improper 0.15625
fractions. In the case of problems involving 32 5.00000
mixed numbers it is necessary first to convert 32
them into improper fractions. Example: 1 80
1 60
7 = 2x7y +3-_
3 17 200
2 192
80
The numerator of the improper fraction is 64
equal to the whole number multiplied by the 160
denominator of the original fraction, to which 160
www.americanradiohistory.com
HANDBOOK Division of Fractions 789
It will be obvious that many vulgar fractions In general terms, when a number is to be
cannot be reduced to exact decimal equiva- multipled by itself we speak of raising to a
lents. This fact need not worry us, however, power or involution; the number of times
since the degree of equivalence can always be which the number is to be multiplied by it-
as much as the data warrants. For instance, self is called the order of the power. The
if we know that one -third of an ampere is standard notation requires that the order of
flowing in a given circuit, this can be written the power be indicated by a small number
as 0.333 amperes. This is not the exact written after the number and above the line,
equivalent of 1/3 but is close enough since it called the exponent. Examples:
shows the value to the nearest thousandth of
an ampere and it is probable that the meter 2' = 2 X 2, or 2 squared, or the second
rom which we secured our original data was power of 2
not accurate to the nearest thousandth of an 2' = 2 X 2 X 2, or 2 cubed, or the third
ampere. power of 2
Thus in converting vulgar fractions to a 2' = 2 X 2 X 2 X 2, or the fourth pow-
decimal we unhesitatingly stop when we have er of 2
reached the number of significant figures war-
ranted by our original data, which is very Sometimes it is necessary to perform the
seldom more than three places (see section reverse of this operation, that is, it may be
Significant Figures later in this chapter). necessary, for instance, to find that number
When the denominator of a vulgar fraction which multiplied by itself will give a product
contains only the factors 2 or 5, division can of nine. The answer is of course 3. This
be brought to a finish and there will be no process is known as e.tvracting the root or
remainder, as shown in the examples above. evolution. The particular example which is
When the denominator has other factors
such as 3, 7, 11, etc., the division will seldom
come out even no matter how long it is con-
tinued but, as previously stated, this is of
no consequence in practical work since it may
cited would be written:
= 3
3
- .... =073
= 0.3333 obvious that it must lie between 2 and 3. In
general the square root of such a number can-
= 0.142857142857 .... = 0.14285/ not be exactly expressed either by a vulgar
fraction or a decimal fraction. However, the
The foregoing examples contained only re- square root can be carried out decimally as
peating digits. In the following example a far as may be necessary for sufficient accur-
non -repeating digit precedes the repeating acy. In general such a decimal fraction will
digit: contain a never -ending series of digits with-
out repeating groups. Such a number is an
30 = 0.2333.... =0.21 irrational number, such as
While repeating decimal fractions can be = 2.2361 . . . .
converted into their vulgar fraction equiva-
lents, this is seldom necessary in practical The extraction of roots is usually done by
work and the rules will be omitted here. tables or logarithms the use of which will
be described later. There are longhand meth-
Powers and When a number is to be mul- ods of extracting various roots, but we shall
Roots tiplied by itself we say that give only that for extracting the square root
it is to be squared or to be since the others become so tedious as to make
raised to the second power. When it is to be other methods almost invariably preferable.
multipled by itself once again, we say that Even the longhand method for extracting the
it is cubed or wired to the third power. square root will usually be used only if loga-
790 Radio Mathematics and Calculations THE RADIO
rithm tables, slide rule, or table of roots are found and annex the cipher (2 x 75 or 150
not handy. plus the cipher, which will give 1500). 1500
will go into 5391 3 times. Replace the last
Extracting the First divide the number the cipher with a three and multiply 1503 by 3 to
Square Root root of which is to be ex- give 4509. Place 3 above the third group.
tracted into groups of two Subtract to find the remainder of 882. The
digits starting at the decimal point and going quotient 75.3 which has been found so far is
in both directions. If the lefthandmost group not the exact square root which was desired;
proves to have only one digit instead of two, in most cases it will be sufficiently accurate.
no harm will be done. The righthandmost However, if greater accuracy is desired groups
group may be made to have two digits by of two ciphers can be brought down and the
annexing a zero if necessary. For example, process carried on as long as necessary.
let it be required to find the square root of
5678.91. This is to be divided off as follows: 7 5. 3
J56' 78.91
J56' 78.91 49
www.americanradiohistory.com
HANDBOOK Order of Operations 791
2+ 3 X 4'
In the foregoing example it is obvious that the
we must first square the 4 to get 16; then we 3 in the numerator goes into the 6 in the de-
multiply 16 by 3, making 48, and to the nominator twice. We may thus cross out
product we add 2, giving a result of 50. the three and replace the 6 by a 2. The 2
if a different order of operations were fol- which we have just placed in the denominator
lowed, a different result would be obtained. cancels the 2 in the numerator. Next the 5
For instance, if we add 2 to 3 we would ob- in the denominator will go into the 25 in the
tain 5, and then multiplying this by the square numerator leaving a result of 5. Now we
of 4 or 16, we would obtain a result of 80, have left only a 5 in the numerator and a 7
which is incorrect. in the denominator, so our final result is 5/7.
In more complicated forms such as frac- If we had multiplied 2 X 3 x 25 to obtain
tions whose numerators and denominators may 150 and then had divided this by 6 X 5 X 7
both be in complicated forms, the numerator or 210, we would have obtained the same re-
and denominator are first found separately sult but, with considerably more work.
before the division is made, such as in the
following example: Algebra
Algebra is not a separate branch of mathe-
3X4+sX2 12+10 22 matics but is merely a form of generalized
2 X 3+ 2+ 3- 6+ 2+ 3- 11 arithmetic in which letters of the alphabet and
occasional other symbols are substituted for
Problems of this type are very common in numbers, from which it is often referred to as
dealing with circuits containing several in literal notation. It is simply a shorthand meth-
ductances, capacities, or resistances. od of writing operations which could be spelled
The order of operations specified above does out.
not always meet all possible conditions; if a The laws of most common electrical phe-
series of operations should be performed in a nomena and circuits (including of course ra-
different order, this is always indicated by dio phenomena and circuits) lend themselves
parenthrri or bracket c. for example: particularly well to representation by literal no-
tation and solution by algebraic equations or
2+3 X4'=2+3 X 16-2+48=50 formulas.
(2 + 3) X 4'=5 X 4'=5 X 16 = 80
While we may write a particular problem in
Ohm's Law as an ordinary division or multi-
2+(3 X 4)2= 2 + 12'=2+ 144= 146 plication, the general statement of all such
problems calls for the replacement of the num-
In connection with the radical sign, brackets bers by symbols. We might be explicit and
may he used or the "hat" of the radical may write out the names of the units and use these
be extended over the entire quantity whose names as symbols:
root is to he extracted. Example:
volts = amperes X ohms
V4+5= + 5 = 2 + 5 = 7
Such a procedure becomes too clumsy when
V14+5) = V4+5- = 3
the expression is more involved and would be
unusually cumbersome if any operations like
It is recommended that the radical always be multiplication were required. Therefore as a
short way of writing these generalized rela-
extended over the quantity whose root is to be
extracted to avoid any ambiguity. tions the numbers are represented by letters.
Ohm's Law then becomes
Cancellation a fraction in which the
In
numerator and denominator
E - I X R
consist of several factors to be multiplied, con- In the statement of any particular problem
siderable labor can often be saved if it is the significance of the letters is usually indi-
found that the same factor occurs in both cated directly below the equation or formula
numerator and denominator. These factors using them unless there can be no ambiguity.
cancel each other and can he removed. Ex- Thus the above form of Ohm's Law would be
ample: more completely written as:
792 Radio Mathematics and Calculations THE RADIO
E = I X R Similarly, we have:
where E
I
= e.m.f. in volts
= current in amperes a (- b) = o - b
R = resistance in ohms When a minus sign is in front of an expres-
sion in brackets, this minus sign has the effect
Letters therefore represent numbers, and for of reversing the signs of every term within the
any letter we can read "any number." When brackets:
the same letter occurs again in the same ex-
pression we would mentally read "the same
- (2a
-+3b - (a b) =
-5c) = -2a -3b +5e
- a + b
number," and for another letter "another num-
ber of any value."
These letters are connected by the usual op- Multiplication. When both the multiplicand
erational symbols of arithmetic, +, x, -, and the multiplier are negative, the product is
positive. When only one (either one) is nega-
and so forth. In algebra, the sign for division
is seldom used, a division being usually written tive the product is negative. The four possible
as a fraction. The multiplication sign, x, is cases are illustrated below:
usually omitted or one may write a period
only. Examples: +X +=+ + X - _ -
- X + _ - - X - _ +
2XaXb =lob
2.3.4.50= 2X3X4X5Xa Division. Since division is but the reverse of
multiplication, similar rules apply for the sign
In practical applications of algebra, an ex- of the quotient. When both the dividend and
pression usually states some physical law and the divisor have the same sign (both negative
each letter represents a variable quantity which or both positive) the quotient is positive. If
is therefore called a variable. A fixed number they have unlike signs (one positive and one
in front of such a quantity (by which it is to negative) the quotient is negative.
be multiplied) is known as the coefficient.
Sometimes the coefficient may be unknown, yet
to be determined; it is then also written as a
letter; k is most commonly used for this pur-
pose.
-_ +
The Negative In ordinary arithmetic we
Sign seldom work with negative Powers. Even powers of negative numbers
numbers, although we may are positive and odd powers are negative. Pow-
be "short" in a subtraction. In algebra, how- ers of positive numbers are always positive.
ever, a number may be either negative or pos- Examples:
itive. Such a thing may seem academic but a
negative quantity can have a real existence.
We need only refer to a debt being considered - 2X-
-2'=- = -2X -2 = +4
2'
2X -2
a negative possession. In electrical work, how-
ever, a result of a problem might be a negative -2 = -8 = +4X
number of amperes or volts, indicating that the Roots. Since the square of a negative num-
direction of the current is opposite to the di- ber is positive and the square of a positive
rection chosen as positive. We shall have il- number is also positive, it follows that a posi-
lustrations of this shortly. tive number has two square roots. The square
Having established the existence of negative root of 4 can be either +2 or -2 for ( +2)
quantities, we must now learn how to work x ( +2) = +4 and ( -2) X ( -2) = +4.
with these negative quantities in addition, sub-
traction, multiplication and so forth. Addition and Polynomials are quantities
In addition, a negative number added to
positive number is the same as subtracting
a
a
Subtraction like 3ab2 + 4ab' 7a2b' -
which have several terms of
positive number from it. different names. When adding polynomials,
only terms of the same name can be taken to-
-
7 7
-3 gether.
(odd) is the same as 3
(subtract) 703+8ob' +3o'b +3
4 4
or we might write it o' - S ab' - b'
7 + (- 3) = 7 - 3 =4 8o' +3 ob' +3o'b -b' 3
www.americanradiohistory.com
HANDBOOK Division 793
them together. x xy
a X b is written os ab + x y
x y -xy
a X b' is written os ob' + xy
xy
--
Bracketed quantities are multiplied by a
single term by multiplying each term:
Factoring Very often it is necessary to sim-
alb +c +d) =ob +oc + ad plify expressions by finding a fac-
tor. This is done by collecting two or more
When two bracketed quantities are multi- terms having the same factor and bringing the
plied, each term of the first bracketed quantity factor outside the brackets:
is to be multiplied by each term of the second
bracketed quantity, thereby making every pos- 6ob + 3ac = 3a12b + cl
sible combination.
In a four term expression one can take to-
gether two terms at a time; the intention is to
la+b) Ic+dl = oc + ad + bc+ bd
try getting the terms within the brackets the
same after the factor has been removed:
In this work particular care must be taken
to get the signs correct. Examples:
30oc - 18bc + 10od - 6bd =
(a+b) - =o'+ab-ab-b'= 6c 15o - 313) + 2d 50 - 310 =
(a
-
b)
a' b'
15o - 3b1 16c + 2d)
la+b) (a + b)
a'+2ab+
=a'+ab +ob+b-
b
- Of course, this is not always possible and
the expression may not have any factors. A
similar process can of course be followed when
lo-b) (a - b) -o - ob -ob+b'- the expression has six or eight or any even
number of terms.
o'-2ab+b: A special case is a three -term polynomial,
which can sometimes be factored by writing
Division It is possible to do longhand divi- the middle term as the sum of twn terms:
sion in algebra, although it is
somewhat more complicated than in arithme- - 7xy + 12y' moy be rewritten os
tic. However, the division will seldom come x'- 3xy-4xy+ 12y'=
out even, and is not often done in this form. x Ix-3yi - 4y Ix - 3y1 =
The method is as follows: Write the terms of
the dividend in the order of descending powers
1z - 4y) (x - 3y)
of one variable and do likewise with the di-
visor. Example:
The middle term should be split into two
in such a way that the sum of the two new
a' a' l'.' = a' Any toot may be written in this form
b'Xb - b' -6 'b =b1/4 b' -b3/
es eT
= en The same notation is also extended in the
negative direction:
Similarly, dividing of powers is done by
subtracting the exponents.
a' = aoa =o
b-, - by.
1 1 C
_ 1/2
- - c 1.s 1
a' oo
or a'
a=
= o "-" = o' =a
Following the previous rules that exponents
b'
b'
bbbbb
bbb b' or b'
b'
=b "" - b' add when powers are multiplied,
1
tional power.
These examples illustrate two rules: (1) any
number raised to "zero' power equals one or
unity; (2) any quantity raised to a negative
='for (`)'ti = ou
power is the inverse or reciprocal of the same Removing radicals. A root or radical in the
quantity raised to the same positive power. denominator of a fraction makes the expres-
sion difficult to handle. If there must be a rad-
n = 1 o'- ical it should be located in the numerator
rather than in the denominator. The removal
Roots. The product of the square root of of the radical from the denominator is done
two quantities equals the square root of their by multiplying both numerator and denomina-
product. tor by a quantity which will remove the radi-
cal from the denominator, thus rationalizing it:
fo X VT)= ab
4 - v
-
Also, the quotient of two roots is equal to the 1
-
va- foX
1
3a
Note, however, that in addition or subtrac- In this case we must multiply
Va + Vb
tion the square root of the sum or difference is
not the same as the sum or difference of the
square roots.
numerator and denominator by V
same terms but with the second having the
the - v
/
opposite sign, so that their product will not
Thus, - V 4 = 3- 2= contain a root.
- =
1
but
Likewise
V9
+ fis
4 = 2.2361
not the same os Va + b f+f3a Da( Vei
(+xb)(fa-f?
-VW) aa( \fa
a
- V.*
HANDBOOK Powers, Roots, Imaginories 795
=iorjandi'orj`= -1 E = IR
Imaginary numbers do not exactly corre- One of the three quantities may be unknown
spond to anything in our experience and it is but if the other two are known, the third can
best not to try to visualize them. Despite this be found readily by substituting the known
fact, their interest is much more than academ- values in the equation. This is very easy if it
ic, for they are extremely useful in many cal - is E in the above example that is to be found;
culations involving alternating currents. but suppose we wish to find / while E and R
The square root of any other negative num- are given. We must then rearrange the equa-
ber may be reduced to a product of two roots, tion so that / comes to stand alone to the left
one positive and one negative. For instance: of the equality sign. This is known as solving
the equation for 1.
57= \r-1 \ro=i\r5-1 Solution of the equation in this case is done
simply by transposing. If two things are equal
f-a=io
or, in general then they must still be equal if both are multi -
plied or divided by the same number. Dividing
both sides of the equation by R:
10-"
It is, however, simpler in this case to use
the elimination method. Multiply both sides of
- .28x 10'x25x 10-0
1
- la
6.28 x 25
the first equation by two and add it to the
second equation:
=
6360 ohms
+ l0y =
A bias resistor of 1000 ohms should be by-
passed, so that at the lowest frequency the re-
6x
4x - IOy =
14
3
odd
actance of the condenser is 1 /10th of that of 10x = 17 x = 1.7
the resistor. Assume the lowest frequency to be
50 cycles, then the required capacity should
Substituting this value of x in the first equa-
have a reactancc of 100 ohms, at 50 cycles: tion, we have
C- 1
farads
5.1 + Sy = 7 .. 5y = 7 - 5.1 = 1.9 .'.
4 x 3.14 x 50 x,100 y = 0.38
lb' microforods
C 6.28 x s000
C = 32 fd.
f C
f _ 2nCX
1 In this simple network the current divides
through the 2000-ohm and 3000 -ohm resistors.
The current through each may be found by
Substituting known values using two simuli linear equations. Note
that the arrows indicate the direction of elec-
f - 2 x 3.14 x 0.03 x 10-4 x 100,000
1
cycles tron flow as explained on pogo 18.
f = 0.01884
cycles = 53 cycles An application of two simultaneous linear
equations will now be given. In Figure 3 a
These equations are known as first degree simple network is shown consisting of three re-
equations with one unknown. First degree, be- sistances; let it be required to find the currents
cause the unknown occurs only as a first power. I, and I, in the two branches.
Such an equation always has one possible so- The general way in which all such prob-
lution or root if all the other values are known. lems can be solved is to assign directions to
If there are two unknowns, a single equa- the currents through the various resistances.
tion will not suffice, for there are then an infi- When these are chosen wrong it will do no
nite number of possible solutions. In the case harm for the result of the equations will then
of two unknowns we need two independent be negative, showing up the error. In this sim-
simultaneous equations. An example of this is: ple illustration there is, of course, no such dif-
ficulty.
3x +Sy =7 4x -10y =3 Next we write the equations for the meshes,
in accordance with Kirchhoff's second law. All
Required, to find x and y. voltage drops in the direction of the curved
This type of work is done either by the sub-
stitution method or by the elimination method. arrow are considered positive, the reverse ones
In the substitution method we might write for
negative. Since there are two unknowns we
write two equations.
the first equation:
www.americanradiohistory.com
HANDBOOK Quadratic Equations 797
(b)
6000 1, + 9000 1, - 10 = 0
X_ -bfVb'-4ae
2a
e - o = b
www.americanradiohistory.com
HANDBOOK Logarithms 799
words, the question is "to which power must log 10 = log 10' = 1
could be used, and therefore there are many value of the logarithm of 20 is 1.30103... .
possible systems of logarithms. In practice we The part of the logarithm to the left of the
use only two bases: The most frequently used decimal point is called the characteristic, while
base is 10, and the system using this, base is the decimals are called the mantissa. In the
known as the system of common logarithms, case of 1.30103 . ., the logarithm of 20, the
or Briggs' logarithms. The second system em- characteristic is 1 and the mantissa is .30103 ..
ploys as a base an odd number, designated by
the letter e; e = 2.71828.... This is known Properties of If the base of our system is
as the natural logarithmic system, also as the Logarithms ten, then, by definition of a
Napierian system, and the hyperbolic system. logarithm:
Although different writers may vary on the
subject, the usual notation is simply log a for 10'' =a
the common logarithm of a. and log. a (or
sometimes In a) for the natural logarithm of or, if the base is raised to the power having an
a. We shall use the common logarithmic sys- exponent equal to the logarithm of a number,
tem in most cases, and therefore we shall ex- the result is that number.
amine this system more closely. The logarithm of a product is equal to the
sum of the logarithms of the two factors.
Common In the system wherein 10 is the
Logarithms base, the logarithm of 10 equals log ab = log a + log b
i; the logarithm of 100 equals 2,
etc., as shown in the following table: This is easily proved to be truc because, it
800 Radio Mathematics and Calculations T H E R A D I O
"
(DMmpC QOOmpD
m $m
NmN(DO O.+.+.Of npONO pmoNp vINNpM
g m!)N00 POMm+
()m
NNNNN rnrmm m(PVMMP
mCOm)O.~i
mmmmm rNmGi
mmmmm (I)(ND(mD .N.CI
m00mmm mmmmm cMirmi?
mmmmm cNO(nOthD 0~OWOOii
mmmmm
m
mmmOfm
1
N mmNpo
puDm[ Mp
nNrnn mom . C
Nnnmm
Cp D+ O.r
NrMONmOMmpnmro
.NMMP PNmOn rmOOO 0..NN MmNV c=. mmmOeD7.D
mmmmm Nmmrn
mmmmm MMppN mmmmm mmmmm
I
m
mppmN
ON NO NO)IDMO
PN MN mmmO.+
NNNNN r[nmm
Nm(DNN mNmM(D
(DMO)tDN
NNMP
mmmmh
mpODN
pN (DI a
NNO)pm MIONp
mPOfNO D.+[N[
C-mmm0 O....NN
((7NNPM ..mmMm NOpO)M
N[N[N L.....N mmmO.0)
MMPpN Nm(DII ONO)Mm
mmmD,Oi mmmmm mmmmm mmmmm
MONNN mmmmm
pNmC-p
pNNm[ mNNm
cnnnn mm000
ct-nmm
v)Mmpm ONN..m
M..m(OM ..mNNm
pNNAh
NmMC-O
mNO).f)N
NNMP
mNN..P
NMMMN
mpOwN
mmmmm
mChC-m
MC-Mm
Mmp
pN(Dm[ mmm
P..hMm
a
mNNNm
N..(D..m
O....NN
mmmmm
NNmN.. NMmN.. hMmpm MmmNM
NON..m
N Mp PN O m[ [m mm O....NN
000mm
NhNm..
MMPPN m..mON
mmmmm
NNNNM
Nmm[I
(D..(D.+
(DP..mp
mmmmm
NmSMO
(DONON
pNOpm
OpO)M[
mmmO)O)
mmmmm
N..mOp
mpmMN
nnnNn mmm00
Innmm mmmm mmmmm mmmm mmmmm MMppN
mmmmm NfOtD[L
mmmmm [mmmm
mmmmm
CNNh.. MNNPN mNOpC- 0......0 mNNmM N..p(Dm mpOOm NNNm(l)
M C`IOmNM NMmN.. mNmMm pONON ....NN O(O..mO NONmP mMmN(D
d'NN(DN mmm00 NNMP pN)f)m[ Immmm MMPpN Nmmml
ncnt n nr-nmm OMOmM mmmmm mmmmm Cmmmm
mmmmm
mNppmpM (OmmIN NmPm.. MNNNp NO(DN[ yNy.(Dm..M ppNNNyO (DNh.N
fA
W
J -
M
NNNNN rmmO
ppN(D nnNmm .NNM ym.N(N(pDr nmmmm
mmmmm OO..N
mmmmm l")MpPN
mmmmm NMMOM
mmmmm (D(D(m0 mmmmm
=
M
4
1-
..
O
NOmNm
pPN(DN
OmM
mO..Om
..mmp.. m(DMO(D
NNNNN hmmOO
n[nmm
ONm(ND
mNh..N Nmmmm mP..CN AOMmm
MOmMm
mmmmm
Nm(ONm N.M-.ICM]m
c
N..NC)m N..NNm MmPmp
NNMM PNNO(D nmmmm OO....N
MOMOM
NOAN MmMmP
m0p Om
mNONm
NMPpp
NNMm(D NNNs.
PmpppppM mNh.m
mmmmm (()Nm(Dh
mmmmm nmmmm
yy.
mmmmm
OPiPmpm pO0)MpNpppM IN(DNN
pppp
ppNmC
NNNNN mmmc
ntnm NMM pNN(Dm I-mmmm OO....N
mmmmm NMMpp mmmmm Immmm
mmmmm (1)Nm(OI mmmmm
mmmmm
Q
(' Nmhmm O.NMpp NmNmm O.NMp N DNmm O.NMp NONCOM 0.MMV NDhmm
NNNNN mDDOD mDmAD NNNNN NNNNN mmmmm mmmmm mmm) mmmmm
O
J
W
U
4
H MpN
OM .r
OO...... ONNNm
Mn..pn
NNNNN
mm OmmD NM
MmO
NpNNm CpPpp
p
OMO Nmm
NpmC- mO..Mp (1)m[m0
NmN.Omm DMM.m
.CCN
NMy.7.N WC-ppppmm O..NMM
p(1)NNN NNNNm mmm(Dm WWWWW nnnnn
3
OI
n
PNmNM
mmm
mNOmN
NDOCm mNpm
mNNmm ..NmM nnMnm m.ppNM
MnNm ONPN mON
OO... ..NNNN NppCC
MMMMM PPpPp PNNN
DD
pmO
DNmO
NNNNN Omd C mMNN
Mm
MMCONCO
M
0
OM
n003)00
B
O
IL MNpNp MONmN
NppOMV
..NNM mN..mm NmmpM
MppNO NmMmm P(!)NAN N..PNM
NCNM..
.MOON mmmmm OpmmN NmONN
mC0NmN
p
MNNm mpO(D..
oNpNN NDNppm yN..mN
OO... mNp(Dm PPNN! Nmhmm
mm.NM NCy NNM
N(DOM(D
.NNNN MMMMM pppP NNfNN pWWWWW mmADD CNCN`
NNmMp MNONO mPN..N NNMmm MMmOm NOPNp
pm..Nh NMONm MmNpP
O(!)NN(D
H OC`MI ..NN..m mC)mem
0NNN ONpNm pC-mmmm mhmmNp C)..Om
(cQpDQ(DO.+
ONOO..W MCM7MMM w TpNN NN (D(D((MD(D (D(D(O(D(D tM-
OMV.V NmNmm m NNP mmmMM mmNNm O.mMN pOpNp .Nmmn pOMND
NmMC CPOp 0000N NMpp NmOMm m.NpN mNhNC NmNmm N.mNN
NDNPM
NmNN m.pDm
OOO.... ..NNNN
C
ONDm pp
MMMMM ONMN(D
ppPPp mm..NM pmNmm O..NMP
pp(I)NN N(()NNN mmm(DD mDm(DmO....NM
NNNNN
' m..mmM NNONm NPMP(D ..ONmm PNNPM mmNNP MCMNP .mMOM m..NNm
M ..OmpCIN CPO..Cm)(Dm ONpmm
pPPPP mmOM
MMMMM ONMN(m0 PpNNN (mnaMOOp) O..NMp (NOMmC) .mnM
'000.. ..NNNN N(lINNN (DA(Dmm mm(Dm(D rrC`C`C,
WMONM OmNMO
000mON mNNCmI MM(m`ImM pPp ppM( NN m(D(DmtD [I[N
www.americanradiohistory.com
HANDBOOK Logarithm Tables 801
was shown before that when multiplying to logarithm tables, it has become the rule that
powers, the exponents are added; therefore, the mantissa should always be positive. Such
notations above as -
1.918555 really mean
a X b = 10101 X 10108b = 10 (log.. "g b' (+0.918555 -1);and -2.981555 means
( + 0.918555 -
2). There are also some other
Similarly, the logarithm of a quotient is the notations in use such as
difference between the logarithm of the divi-
dend and the logarithm of the divisor. 1.918555 and 7.918555
www.americanradiohistory.com
HANDBOOK The Decibel 803
log x = 2.038
- 0.408 Therefore the gain is 10 x 1.48 = 14.8
decibels. The decibel is a logarithmic unit;
x = ontilog 0.408 = 2.56 when the power was multiplied by 30, the
power level in decibels was increased- by 14.8
The tuning ratio should be 2.56. decibels, or 14.8 decibels added.
804 Radio Mathematics and Calculations THE RADIO
TuDE STE, -Up
Example: In the circuit of Figure 8, the gain
GAIN. A RATIO = 3.5 I in the stage is equal to the amplification in the
tube and the step -up ratio of the transformer.
If the amplification in the tube is 10 and the
step -up in the transformer is 3.5, the voltage
gain is 15 and the gain in decibels is:
20 x log 35 = 20 x 1.54 = 30.8 dh
www.americanradiohistory.com
HANDBOOK
80
50
40
30
20
10
i--
a
tta
M_
MN
ra
"-_
mtttmmui1AlS
_1IttA
taiJ6/
I.i'r*-
imozZiNAM
it_!
1:JjII
T:1'
1111=.w
r4eTeTr[.><uAu.OMMI-
MINIIMINNIMI+tta
` ==MMel11isYfiT:1:/J
i1110i!-
.._.;.'I.i
-
. .
=
!
mil wow.
/S1iCJ
Decibel -Power Conversion
-
-tfa
m
It; 30
M
m
=I
... 20
t
NI
m
IO
.1
NE
=I
no
50
40
-10
805
60
Ilt
ENI
OM
z
-20
MO
20 -J
-J-30
>
iP!i!iT
MINIM
MIIII---1111111111
tNIMM=11111l111 1 J
ef.7!tll
r 30 w
J
W IMMIIIMM11110
-1-40 -40
It-1111-1111
--111-21111Mw
ttIl
INIIIIIM11111111..R
-50 11111111= -50
-60
-70
1R
m_7 11iitY/!/
-'2".=NeonlI
MENIIMINI111
.. en11:llleJiL
-60
-70
_
ItItiM1lt>11ONI_NlaLf\i0
. feiel7.d7i NM
-80 ta -80
. S
fsi'1 '
Figure 9.
CONVERSION CHART: POWER TO DECIBELS
Power levels between 6 mlcromicrowatts and 6000 watts may be referred to corresponding decibel
levels between -90 and 60 db, and vice versa, by means of the above chart. Fifteen ranges are
provided. Each curve begins at the same point where the preceding one ends, enabling uninterrupted
coverage of the wide db and power ranges with condensed chart. For example: the lowermost curve
ends at -80 db or 60 micromicrowatts and the next range starts at the same level. Zero db lev& is
taken as 6 milliwatts (.006 watt).
806 Radio Mathematics and Calculations THE RADIO
Converting Decibels It is often convenient Solution:
to Power to he able to convert a
decibel value to a pow-
-- 17.3
2.7 + 2.7
er equivalent. The formula used for this oper-
ation is -20 + 2.7
P = 0.006 X antilogg Nd, -20+2.7
l 10 = = -2.27
10
where P is the desired level in watts and d, Antilog -2.27 = 0.0186
the decibels to be converted.
To determine the power level P from a dec- 0.006 X 0.0186 = 0.000 16 watt or
1 1
ibel equivalent, simply divide the decibel value
by 10; then take the number comprising the
0.1116 milliwatt
antilog and multiply it by 0.006; the product Input voltages: To determine the required
gives the level in watts. input voltage, take the peak voltage necessary
Note: In problems dealing with the conver- to drive the last class A amplifier tube to max-
sion of minas decibels to power, it often hap- imum output, and divide this figure by the to-
pens that the decibel value -Nd,, is not tal overall voltage gain of the preceding stages.
divisible by 10. When this is the case, Computing Specifications: From the preced-
N,.,
the numerator in the factor - 10
must be ing explanations the following data can be
computed with any degree of accuracy war-
made evenly divisible by 10, the negative signs ranted by the circumstances:
must be observed, and the quotient labeled ac- (1) Voltage amplification
cordingly. (2) Overall gain in db
To make the numerator evenly divisible by (3) Output signal level in db
10 proceed as follows: Assume, for example, (4) Input signal level in db
that -N,,, is sonie such value as -38; to (5) Input signal level in watts
make this figure evenly divisible by 10, we (6) Input signal voltage
must add -2 to it, and, since we have added
a negative 2 to it, we must also add a positive
When a power level is available which must
be brought up to a new power level, the gain
2 so as to keep the net result the same.
required in the intervening amplifier is equal
Our decibel value now stands, -40 + 2.
to the difference between the two levels in dec-
Dividing both of these figures by 10, as in the
ibels. If the required input of an amplifier for
equation above, we have -
4 and +0.2. Put-
full output is -
30 decibels and the output
ting the two together we have the logarithm
-4.2 with the negative characteristic and the from a device to be used is but -45 decibels,
the pre -amplifier required should have a gain
positive mantissa as required.
The following examples will show the tech-
of the difference, or 15 decibels. Again this is
nique to be followed in practical problems.
true only if the two amplifiers are properly
matched and no losses are introduced due to
(a) The output of a certain device is rated
mismatching.
at -74 db. What is the power equivalent?
Solution:
Push -Pull To double the output of any cas -
No,
= -10 (not evenly divisible by 10) Amplifiers cade amplifier, it is only neces-
sary to connect in push -pull the
Routine: last amplifying stage, and replace the inter -
-- 74
6 +6
stage and output transformers with push -pull
types.
To determine the voltage gain (voltage ra-
-80 +6 tio) of a push -pull amplifier, take the ratio of
one half of the secondary winding of the push -
Net. -80 +6_
=--1-0-- -86
10
pull transformer and multiply it by the of
one of the output tubes in the push -pull stage;
antilog -8.6 = 0.000 000 04 the product, when doubled, will be the voltage
.006 X 0.000 000 04 = amplification, or step -up.
0.000 000 000 24 watt or
240 micro- microwott Other Units and When working with deci-
Zero Levels bels one should not im-
(b) This example differs somewhat from mediately take for granted
that of the foregoing one in that the mantissas that the zero level is 6 milliwatts for there are
are added differently. A low -powered amplifier other zero levels in use.
has an input signal level of -
17.3 db. How In broadcast stations an entirely new system
many milliwatts does this value represent? is now employed. Measurements made in
www.americanradiohistory.com
HANDBOOK Trigonometry 807
Figure 11. a
tangent A = cotangent A =
THE RADIAN. b a
A radian is an angle whose arc is exactly equal
to the length of either side. Note that the c
angle is constant regardless of the length of
secant A = cosecant A =
a
the side and the arc so long as they are equal.
A radian equals 57.2958 . Let us take a special angle as an example.
For instance, let the angle A be 60 degrees as
In trigonometry we consider an angle gen- in Figure 13B. Then the relations between the
erated by two lines, one stationary and the sides are as in the figure and the six functions
other rotating as if it were hinged at 0, Figure become:
12. Angles can be greater than 180 degrees and
even greater than 360 degrees as illustrated in a 1/2 v-3-
this figure. sin. 60 _ -
C
= = 1/2
of A when
a 1/z Nr3
A= (90 -B) tan 60 =
b 1/2
and when
B=(90 -A)
cot 60 =
1/2
1/2V ,
- _ = y3)/3
1
1
Two angles are supplements of each other sec 60 = b = = 2
when their sum is equal to'a straight angle, or 1/z
180 . A is the supplement of B and B is the c 1
X
AC
Figure 12.
AN ANGLE IS GENERATED BY TWO LINES, ONE STATIONARY AND THE OTHER
ROTATING.
The line OX is stationary; the line with the small arrow at the for end rotates in a counterclockwise
direction. At the position illustrated in the lefthandmost section of the drawing it makes an angle,
A, which is less than 90 and is therefore in the first quadrant. In the position shown in the second
portion of the drawing the angle A has increased to such a value that it now lies in the third
quadrant; note that an angle can be greater than 180 In the third illustration the angle A is in
.
the fourth quadrant. In the fourth position the rotating vector has made more than one completo
revolution and is hence in the fifth quadrant; since the fifth quadrant is an exact repetition of the
first quadrant. its values will be the some os in the lef thandmost portion of the illustration.
www.americanradiohistory.com
H A N D B O O K Trigonometric Relations 809
C=e
90'
b= I
Figure 13.
THE TRIGONOMETRIC FUNCTIONS.
In the right triangle shown in (A) the side opposite the ongle A is a, while the adjoining sides are
b and c; the trigonometric functions of the ongle A are completely defined by the ratios of the
sides a, b and c. In (B) are shown the lengths of the sides a and b when angle A is 60' and side c
is I. In (C) angle A is 45 ; a and b equal 1, while c equals VI
In (D) note that c equals a for a
right angle while b equals O.
cos A - sec A 1
-
1
tan 45 = 1 = cot 45 = = 1 and tan A
=cot A
sec 45 = =
cosec 45 =
- =
From the definitions also follows the relation
seco =- =1 cosec 0 =
a
C C
0
m
Angle
0
Sin
0
Cos.
1
Tan
0_
Cot
co
Sec. Cosec.
11
In general, for every angle, there will be defi- 30 Y2 1/2 V3 1/3 v , V , 2/3 v - 2
Figure 15.
ln this figure the sides a, b, and c are used
to define the trigonometric functions of angle
B as well as angle A.
- -
-a
has some of its functions negative:
-
cos = neg.
c c
sin A =
-a = neg. cos A =
b
=
tan A = = neg. cot A = = neg. pos.
-
b
c
-=
a
sec A =
C
= neg. cosec A = -=
a
pos. tan A =
-a
b
= neg. cot A -
-a neg.
sin A = -- c
a
= neg. cos A
-b
=- = neg.
Summarizing, the sign of the functions in
each quadrant can be seen at a glance from
Figure 17, where in each quadrant are written
== =
tan A = =bb = pos. cot A
-b the names of functions which are positive;
pos.
those not mentioned are negative.
SECOND
QUADRANT
THIRD FOURTH
QUADRANT QUADRANT
Figure 16.
TRIGONOMETRIC FUNCTIONS IN THE SECOND, THIRD, AND FOURTH QUADRANTS.
The trigonometric functions in these quadrants are similar to first quadrant values, but the
signs of the functions vary as listed in the text and in Figure 17.
www.americanradiohistory.com
HANDBOOK Trigonometric Curves 811
Figure 18.
SINE AND COSINE CURVES.
In (A) we have a sine curve
drown in Cartesian coordinates.
This is the usual representation
IA)
1l
0
i-
o
90
_-
180 270
a-
360
_
2'r
AllikikT
450
5
540'
1411P:=
.7r
630
C==n=
720'
>r
n-
I I
1
J7r
h-'
.rl_
270"
_.
360"
Z 7r
450"
I
540
jr
630"
'Al
PriI
v 7r
720'
} 7r
2
Graphs of Trigono- The nine a ate. When 3. Cos x = -cos (180 -x) or
metric Functions have the relation
we -cos (r -x)
y= sin x. where x is an -1. Cos x = cos (360 -x) or cos (27r -x)
angle measured in radians or degrees, we can The graph of the tangent is illustrated in
draw a curve of y versus x for all values of Figure 19. This is a discontinuous curve and
:he independent variable, and thus get a good illustrates well how the tangent increases from
:onception how the sine varies with the mag- zero to infinity when the angle increases from
nitude of the angle. This has been done in zero to 90 degrees. Then when the angle is
Figure 18A. We can learn from this curve the further increased, the tangent starts from
following facts. minus infinity going to zero in the second quad-
1. The sine varies between +1 and -1 rant, and to infinity again in the third quadrant.
2. It is a periodic curve, repeating itself after 1. The tangent can have any value between
every multiple of 27 or 360 + m and -
3. Sin x = sin (180 -x)
or sin (v -x) 2. The curve repeats and the period is it
0
i> ar .; en-
Figure 22.
Vectors may be added as shown in these
sketches. In each case the long vector repre-
sents the vector sum of the smaller vectors.
For many engineering applications sufficient
accuracy can be obtained by this method
COSINE which avoids long and laborious calculations.
Figure 21.
ANOTHER REPRESENTATION OF velocity of 50 miles per hour, we speak of a
TRIGONOMETRIC FUNCTIONS. vector quantity.
If the radius of a circle is considered as the Vectors, representing forces, speeds, dis-
unit of measurement, then the lengths of the
various lines shown in this diagram are numer- placements, etc., are represented by arrows.
ically equal to the functions marked adjacent They can be added graphically by well known
to them. methods illustrated in Figure 22. We can make
the parallelogram of forces or we can simply
The graphs of the secant and cosecant are draw a triangle. The addition of many vectors
of lesser importance and will not be shown can be accomplished graphically as in the same
here. They are the inverse, respectively, of the figure.
cosine and the sine, and therefore they vary In order that we may define vectors algebra-
-
from +1 to infinity and from -1 to infinity. ically and add, subtract, multiply, or divide
them, we must have a logical notation system
Perhaps another useful way of visualizing
the values of the functions is by considering that lends itself to these operations. For this
Figure 21. If the radius of the circle is the unit purpose vectors can be defined by coordinate
of measurement then the lengths of the lines systems. Both the Cartesian and the polar co-
are equal to the functions marked on them. ordinates are in use.
Trigonometric Tables There are two kinds of Vectors Defined Since we have seen how the
trigonometric tables. by Cartesian sum of two vectors is ob-
The first type gives the functions of the angles, Coordinates tained, it follows from Fig-
the second the logarithms' of the functions. ure 23, that the vector Z
The first kind is also known as the table of equals the sum of the two vectors x and y. In
natural trigonometric functions. fact, any vector can be resolved into vectors
These tables give the functions of all angles along the X- and Y -axis. For convenience in
between 0 and 45 . This is all that is necessary working with these quantities we need to dis-
for the function of an angle between 45 and
90 can always be written as the co- function
of an angle below 45 . Example: If we had to
find the sine of 48 , we might write
sin 48 = cos (90 -48 ) = cos 42
Tables of the logarithms of trigonometric
functions give the common logarithms (log,)
of these functions. Since many of these logar-
ithms have negative characteristics, one should
y.4 I
www.americanradiohistory.com
HANDBOOK Vectors 813
R = x, + j y,
Z = x, + j y,
equal to the sum of the squares of the sides Division has to be carried out so as to re-
adjoining the right angle. In Figure 23, OAB move the j -term from the denominator. This
is a right -angled triangle; therefore, the square can be done by multiplying both denominator
of OB (or Z) is equal to the square of OA and numerator by a quantity which will elimi-
(or x) plus the square of AB (or y). Thus the nate j from the denominator. Example:
absolute values of Z and R may be determined
as follows: jy, (x, +jy,) (x, - jy,)
Z == v(30
Z x, + jy, (x, + jy,) (x, - jy,)
- x,y,)
I I y'
Figure 26.
Vectors can be trans formed from Cartesian
into polar notation as shown in this figure.
X
which flows due to the alternating voltage is
Figure 25.
not necessarily in step with it. The rotating
IN THIS FIGURE A VECTOR HAS
BEEN REPRESENTED IN POLAR
current vector may be ahead or behind the
INSTEAD OF CARTESIAN CO- voltage vector, having a phase difference with
ORDI NATES. it. For convenience we draw these vectors as
In polar coordinates a vector is defined by if they were standing still, so that we can indi-
a magnitude and an angle, called the vec- cate the difference in phase or the phase angle.
torial angle, instead of by two magnitudes In Figure 28 the current lags behind the volt-
as In Cartesian coordinates.
age by the angle e, or we might say that the
voltage leads the current by the angle e.
the vector Z has a magnitude 50 and a vector- Vector diagrams show the phase relations
ial angle of 60 degrees. This will then be between two or more vectors (voltages and
written currents) in a circuit. They may be added and
= 5O 6O subtracted as described; one may add a voltage
vector to another voltage vector or a current
A vector a + jb can be transformed into vector to a current vector but not a current
polar notatimi very simply (see Figure 26) vector to a voltage vector (for the same reason
that one cannot add a force to a speed). Figure
= o +jb = '1o' +b2 ton 'b 28 illustrates the relations in the simple series
In this connection tan' means the angle of
circuit of a coil and resistor. We know that the
which the tangent is. Sometimes the notation current passing through coil and resistor must
he the same and in the same phase, so we draw
arc tan b/a is used. Both have the same mean-
ing. this current I along the X -axis. We know also
A polar notation of a vector can be trans- that the voltage drop IR across the resistor is
formed into a Cartesian coordinate notation in in phase with the current, so the vector IR rep-
resenting the voltage drop is also along the
the following manner (Figure 27)
X -axis.
The voltage across the coil is 90 degrees
Z = pLA = p cos A + jp sin A
ahead of the current through it; /X must
A sinusoidally alternating voltage or cur- therefore be drawn along the Y -axis. E the
rent is symbolically represented by a rotating applied voltage must be equal to the vectorial
vector, having a magnitude equal to the peak sum of the two voltage drops, IR and /X. and
voltage or current and rotating with an angular we have so constructed it in the drawing. Now
velocity of 2.7f radians per second or as many expressing the same in algebraic notation, we
revolutions per second as there are cycles per have
second.
The instantaneous voltage, e, is always equal
= IR + jIX
to the sine of the vectorial angle of this rotat- I=IR+IX
ing vector, multiplied by its magnitude.
Dividing by I
e = E sin 2^ft Z = R + jX
The alternating voltage therefore varies with Due to the fact that a reactance rotates the
time as the sine varies with the angle. If we voltage vector ahead or behind the current
plot time horizontally and instantaneous volt- rector by 90 degrees, we must mark it with a
age vertically we will get a curve like those i in vector notation. Inductive reactance will
in Figure 18. have a plus sign because it shifts the voltage
In alternating current circuits, the current vector forwards; a capacitive reactance is neg-
www.americanradiohistory.com
HANDBOOK Graphical Representation 815
),
A
p COS A
Figure 27. Figure 28.
Vectors can be transformed from polar into VECTOR REPRESENTATION OF A
Cartesian notation as shown in this ligure. SIMPLE SERIES CIRCUIT.
The rlghthand portion of the illustration shows
the vectors representing the voltage drops In
ative because the voltage will lag behind the the coil and resistance illustrated at the left.
current. Therefore: Note that the voltage drop across the coil XL,
leads that across the resistance by 90 .
X. = + j 2 -fL
Graphical Representation
x,- = - j 1
2nfC Formulas and physical laws are often pre-
sented in graphical form; this gives us a
In Figure 28 the angle e is known as the
phase angle between E and I. When calculat-
"bird's eye view" of various possible conditions
due to the variations of the quantities involved.
ing power, only the real components count.
In some cases graphs permit us to solve equa-
The power in the circuit is then
tions with greater ease than ordinary algebra.
P = I (IR)
Coordinate Systems All of us have used co-
but IR = E cos 9 ordinate systems with-
out realizing it. For instance, in modern cities
P =El cos e we have numbered streets and numbered ave-
nues. By this means we can define the location
This cos e is known as the power factor of of any spot in the city if the nearest street
the circuit. In many circuits we strive to keep crossings are named. This is nothing but an
the angle e as small as possible, making cos e application of Cartesian coordinates.
as near to unity as possible. In tuned circuits, In the Cartesian coordinate system (named
we use reactances which should have as low a after Descartes), we define the location of any
power factor as possible. The merit of a coil point in a plane by giving its distance from
or condenser, its Q. is defined by the tangent of each of two perpendicular lines or axes. Figure
this phase angle: 30 illustrates this idea. The vertical axis is
called the Y -axis, the horizontal axis is the
Q = ton e = X/R X -axis. The intersection of these two axes is
called the origin, O. The location of a point,
For an efficient coil or condenser, Q should P, (Figure 30) is defined by measuring the
be as large as possible; the phase -angle should respective distances, x and y along the X -axis
then be as close to 90 degrees as possible, mak- and the Y -axis. In this example the distance
ing the power factor nearly zero. Q is almost along the X -axis is 2 units and along the Y-
but not quite the inverse of cos e. Note that in axis is 3 units. Thus we define the point as
Figure 29
1600
NO 1000
NMI-2o
2
=MI= fc
$00
\
:.
600 r
ME Figure 31.
R
3 REPRESENTATION OF A SIMPLE
4 FUNCTION IN CARTESIAN CO-
5 ORDINATES.
300,000
TH RD FOURTH In this chart of the function fkr =
QUADRANT- X meter
tit p
6 QUADRANT
distances along the X axis represent wave-
length in meters, while those along the Y
axis represent frequency in kilocycles. A curve
Y
such as this helps to find values between
Figure 30. those calculated with sufficient accuracy for
CARTESIAN COORDINATES. most purposes.
The location of any point can be defined by
its distance from the X and Y axes.
f is said to be a function of X. For every value
of f there is a definite value of X. A variable is
P 2, 3 or we might say x = 2 and y = 3. The said to be a function of another variable when
measurement x is called the abscissa of the for every possible value of the latter, or inde-
point and the distance y is called its ordinate. pendent variable, there is a definite value of
It is arbitrarily agreed that distances measured the first or dependent variable. For instance,
from 0 to the right along the X -axis shall be if y = 5x2, y is a function of x and x is called
reckoned positive and to the left negative. Dis- the independent variable. When a = 3b3 + 56'
tances measured along the Y -axis are positive -25b + 6 then a is a function of b.
when measured upwards from 0 and negative A function can be illustrated in our coordi-
when measured downwards from O. This is nate system as follows. Let us take the equa-
illustrated in Figure 30. The two axes divide tion for frequency versus wavelength as an
the plane area into four parts called quadrants. example. Given different values to the inde-
These four quadrants are numbered as shown pendent variable find the corresponding values
in the figure. of the dependent variable. Then plot the points
It follows from the foregoing statements, represented by the different sets of two values.
that points lying within the first quadrant have
both x and y positive, as is the case with the ftc.
point P. A point in the second quadrant has a 600 500
negative abscissa, x, and a positive ordinate, y. 800 375
This is illustrated by the point Q, which has 1000 300
till coordinates x = -4 and y = +1. Points 1200 250
in the third quadrant have both x and y nega- 1400 214
tive. x = -5 and y = -2 illustrates such a 1600 187
point, R. The point S, in the fourth quadrant 1800 167
has a negative ordinate, y and a positive ab- 2000 150
scissa or x.
In practical applications we might draw Plotting these points in Figure 31 and draw-
only as much of this plane as needed to illus- ing a smooth curve through them gives us the
trate our equation and therefore, the scales curve or graph of the equation. This curve
along the X -axis and Y -axis might not start will help us find values of f for other values
with zero and may show only that part of the of X (those in between the points calculated)
scale which interests us. and so a curve of an often -used equation may
serve better than a table which always has
Representation of In the equation: gaps.
Functions When using the coordinate system described
300,000 so far and when measuring linearly along both
f = axes, there are some definite rules regarding
www.americanradiohistory.com
HANDBOOK Representation of Functions 817
Figure 32.
Only two points are needed to define func-
tions which result in a straight line as shown
in this diagram representing Ohm's Law. Figure 33.
A TYPICAL GRID - VOLTAGE
PLATE -CURRENT CHARACTER-
ISTIC CURVE.
the kind of curve we get for any type of The equation represented by such o curve Is
equation. In fact, an expert can draw the curve so complicated that we do not use it. Data
for such a curve is obtained experimentally,
with but a very few plotted points since the and intermediate values can be found with
equation has told him what kind of curve to sufficient accuracy from the curve.
expect.
First, when the equation can be reduced to
the form y = mx + b, where x and y are the If the resistance were 4 ohms, we should get
variables, it is known as a linear or first degree the equation E = 4I and this also represents
function and the curve becomes a straight line. a line which we can plot in the same figure.
(Mathematicians still speak of a "curve" when As we see, this line also passes through the
it has become a straight line.) origin but has a different slope. In this illus-
When the equation is of the second degree, tration the slope defines the resistance and we
that is, when it contains terms like x' or y' could make a protractor which would convert
or ., y. the graph belongs to a group of curves, the angle into ohms. This fact may seem incon-
called conic sections. These include the circle, sequential now, but use of this is made in the
the ellipse, the parabola and the hyperbola. In drawing of loadlines on tube curves.
the example given above, our equation is of Figure 33 shows a typical, grid -voltage,
the form plate -current static characteristic of a triode.
xy = e, c being equal to 300,000 The equation represented by this curve is
rather complicated so that we prefer to deal
which is a second degree equation and in this with the curve. Note that this curve extends
case, the graph is a hyperbola. through the first and second quadrant.
This type of curve does not lend itself read- Families of curves. It has been explained
ily for the purpose of calculation except near that curves in a plane can be made to illustrate
the middle, because at the ends a very large the relation between tuo variables when one
change in a represents a small change in f and of them varies independently. However, what
vice versa. Before discussing what can be done are we going to do when there are three vari-
about this let us look at some other types of ables and two of them vary independently. It
curves. is possible to use three dimensions and three
Suppose we have a resistance of 2 ohms axes but this is not conveniently done. Instead
and we plot the function represented by Ohm's of this we may use a family of curves. We
Law: E = 21. Measuring E along the X -axis have already illustrated this partly with Ohm's
and amperes along the Y -axis, we plot the Law. If we wish to make a chart which will
necessary points. Since this is a first degree show the current through any resistance with
equation, of the form y = mx + b (for E _ any voltage applied across it, we must take the
y, m = 2 and I = x and b = 0) it will be a equation E = IR, having three variables.
straight line so we need only two points to We can now draw one line representing a
plot it. resistance of 1 ohm, another line representing
2 ohms, another representing 3 ohms, etc., or
I E
as many as we wish and the size of our paper
(line passes through origin) 0 0 will allow. The whole set of lines is then
5 10 applicable to any case of Ohm's Law falling
within the range of the chart. If any two of
The line is shown in Figure 32. It is seen to the three quantities are given, the third can be
be a straight line passing through the origin. found.
www.americanradiohistory.com
II
n
818 Radio Mathematics and Calculations THE RADIO
I //44
1 p/
i11MANMI 4
WPMA
I
N
AVERAGE PLATE CHARACTERISTICS
E5=6.3 v.
\I
.
n
to
NMI
I ..,/....
V4
P4M
_' A ;.
A.E
III1111
nvae.
,rLII
.:.........
1VIf!
IvvrI--
/ i
1 Itlll
,.,a
1
so
:
o
p /O IN
har70
.M
VOL
M
rase-
. NO NO er0 M IN
www.americanradiohistory.com
HANDBOOK
tube the loadline, that is the line showing
which part of the plate supply voltage is across
the resistor and which part across the tube for
any value of plate current. In our example, let
us suppose the plate resistor is 50,000 ohms.
Then, if the plate current were zero, the volt-
age drop across the resistor would be zero and
the full plate supply voltage is across the tube.
Our first point of the loadline is E = 250,
/ = 0. Next, suppose, the plate current were
1 ma., then the voltage drop across the resistor
2500
2000
1500
1000
900
100
700
600
300
200
omm
1111.mm
8
-.4
.
Logarithmic Scales
\IU
IIIIIIIIIIAIII
IiIIIIIIIIIIIIII
.==.
fa.1fam...e.
011111MNI....e
tsr;goa ra1 01111=NO......
1MINIMIMIONOMMMOMOSE
iMMEMENIMOUNNMMI
1111111MMMW111goI
Illlllli
mimminimn`ummi
111111111111=11111IN1MIIII
111111MMIIM11111111111
-..,,111'IIIIII
.,1"1""'III
111/111111111111111
11111111111111111
WAVELENGTH IN
8
METERS
Figure 37.
A LOGARITHMIC CURVE.
$
mimNIIIIIIII
8$
1.0
0.9
0.8
0.7
1000
0.6
0.5
0.4
0.3
0.2
100
0.1
0
15 10 5 0 S 10 15
Figure 38.
A RECEIVER RESONANCE CURVE.
This curve represents the output of a re-
ceiver versus frequency when plotted to linear r
coordinates.
curve a "wobbulator" will show. The curve -20 -10 0 -F10 +20
does not give enough information in this form
KC. OFF RESONANCE
for one might think that a signal 10 kc. off
resonance would not cause any current at all Figure 39.
and is tuned out. However, we frequently have A RECEIVER SELECTIVITY CURVE.
off resonance signals which are 1000 times as This curve represents the selectivity of a re-
strong as the desired signal and one cannot ceiver plotted to logarithmic coordinates for
the output, but linear coordinates for fre-
read on the graph of Figure 38 how much any quency. The reason that this curve appears
signal is attenuated if it is reduced more than inverted from that of Figure 38 is explained
about 20 times. in the text.
In comparison look at the curve of Figure
39. Here the response (the current) is plotted
in logarithmic proportion, which allows us to be used because now our unit is logarithmic.
plot clearly how far off resonance a signal has An example of full logarithmic paper being
to be to be reduced 100, 1,000, or even 10,000 used for families of curves is shown in the re-
times. actance charts of Figures 40 and 41.
Note that this curve is now "upside down ";
it is therefore called a selectivity curve. The Nomograms or An alignment chart con -
reason that it appears upside down is that the Alignment Charts sists of three or more sets
method of measurement is different. In a se- of scales which have been
lectivity curve we plot the increase in signal so laid out that to solve the formula for which
voltage necessary to cause a standard output the chart was made, we have but to lay a
off resonance. It is also possible to plot this in- straight edge along the two given values on
crease along the Y -axis in decibels; the curve any two of the scales, to find the third and
then looks the same although linear paper can unknown value on the third scale. In its sim-
www.americanradiohistory.com
HANDBOOK Polar Coordinates 823
RADIUS /
b VECTOR /
/ PNGtf;
JCOpP
e X
AXIS
Figure 43.
G o
THE LOCATION OF A POINT BY
Figure 42. POLAR COORDINATES.
In the polar coordinate system any point is
THE SIMPLEST FORM OF NOMOGRAM. determined by its distance from the origin
and the angle formed by a line drawn from
it to the origin and the O -X axis.
plest form, it is somewhat like the lines in Fig-
ure 42. If the lines a, b, and c are parallel and
equidistant, we know from ordinary geometry, by the angle A the vectorial angle. We give
that b = 1/2 (a + c). Therefore, if we draw a these data in the following form
scale of the same units on all three lines, start-
ing with zero at the bottom, we know that by P =3Lo
laying a straight -edge across the chart at any
place, it will connect values of a, b, and c. Polar coordinates are used in radio chiefly
which satisfy the above equation. When any for the plotting of directional properties of mi-
two quantities are known, the third can be crophones and antennas. A typical example of
found. such a directional characteristic is shown in
If, in the same configuration we used loga- Figure 44. The radiation of the antenna rep-
rithmic scales instead of linear scales, the rela- resented here is proportional to the distance of
tion of the quantities would become the characteristic from the origin for every
possible direction.
log b = 1/2 (log a + log c) or b =
By using different kinds of scales, different
units, and different spacings between the scales,
charts can be made to solve many kinds of
equations.
If there are more than three variables it is
generally necessary to make a double chart,
that is, to make the result from the first chart
serve as the given quantity of the second one.
Such an example is the chart for the design of
coils illustrated in Figure 45. This nomogram
is used to convert the inductance in microhen-
ries to physical dimensions of the coil and vice
versa. A pin and a straight edge are required.
The method is shown under "R. F. Tank Cir-
cuit Calculations.' later in this chapter.
www.americanradiohistory.com
HANDBOOK
- 20000 5
- 300
4 -4
--
-10000
8000
6000
3
- 200 .-
- 4000
3000
-3
-150
- 2000 2
-- 1000
- 800
600
-100
- -- 400
-
90
300
-
80
70
- 200
._ 2
- 80 -- 100
- 50
80
-60 .6
- 40
- 40
_--- 1.5
--
30
20- .6
- 30-
-t
-- 8
6
.5
- 20 -4
- 3
-15 - 2
.4
-
-8
1
-.8 .3 -.75
-- 10
--.4
-.3
-.2
-.1
.2
5
-- 0
-4
-3
826 Radio Mathematics and Calculations THE RADIO
ductance, the D/L ratio, and the diameter are given than the accuracy of the measurement
known, simply work backwards from the ex- permits. Thus, if the measurement is good to
ample given. In all cases, remember that the two places, we would write, for instance, 6.9
straightedge reads either turns and D/L ratio, which would mean that the true value is
or it reads inductance and diameter. It can somewhere between 6.85 and 6.95. If the meas-
read no other combination. urement is known to three significant figures,
The actual wire size has negligible effect we might write 6.90 which means that the true
upon the calculations for commonly used wire value is somewhere between 6.895 and 6.905.
sizes (no. 10 to no. 30). The number of turns In dealing with approximate quantities, the
of insulated wire that can be wound per inch added cipher at the right of the decimal point
(solid) will be found in a copper wire table. has a meaning.
There is unfortunately no standardized sys-
tem of writing approximate figures with many
Significant Figures ciphers to the left of the decimal point. 69000
does not necessarily mean that the quantity is
In most radio calculations, numbers repre- known to 5 significant figures. Some indicate
sent quantities which were obtained by meas- the accuracy by writing 69 x 10' or 690 x 10!
urement. Since no measurement gives absolute etc., but this system is not universally em-
accuracy, such quantities are only approximate ployed. The reader can use his own system, but
and their value is given only to a few signifi- whatever notation is used, the number of sig-
cant figures. In calculations, these limitations nificant figures should be kept in mind.
must be kept in mind and one should not fin- Working with approximate figures, one may
ish for instance with a result expressed in more obtain an idea of the influence of the doubtful
significant figures than the given quantities at figures by marking all of them, and products
the beginning. This would imply a greater ac- or sums derived from them. In the following
curacy than actually was obtained and is there- example, the doubtful figures have been under-
fore misleading, if not ridiculous. lined.
An example may make this clear. Many am- 603
meters and voltmeters do not give results to 34.6
closer than 1/4 ampere or 1/4 volt. Thus if we 0.120
have 21/4 amperes flowing in a d.c. circuit at
63/4 volts, we can obtain a theoretical answer 637.720 answer: 638
by multiplying 2.25 by 6.75 to get 15.1875
watts. But it is misleading to express the an- Multiplication:
swer down to a ten -thousandth of a watt when
the original measurements were only good to 654 654_
1/4 ampere or volt. The answer should be ex- 0.342 0.342_
pressed as 15 watts, not even 15.0 watts. If
1308 19612
we assume a possible error of 1/8 volt or am-
2616 26!16
pere (that is, that our original data are only
correct to the nearest 1/4 volt or ampere) the 1962 11308
true power lies between 14.078 (product of 223.668 answer: 224 224
21/8 and 65/8) and 16.328 (product of 23/8 and
67/8). Therefore, any third significant figure It is recommended that the system at the
would be misleading as implying an accuracy right be used and that the figures to the right
which we do not have. of the vertical line be omitted or guessed so as
Conversely, there is also no point to calcu- to save labor. Here the partial products are
lating the value of a part down to 5 or 6 sig- written in the reverse order, the most impor-
nificant figures when the actual part to be used tant ones first.
cannot be measured to better than 1 part in In division, labor can be saved when after
one hundred. For instance, if we are going to each digit of the quotient is obtained, one fig-
use 1 resistors in some circuit, such as an ure of the divisor be dropped. Example:
ohmmeter, there is no need to calculate the
value of such a resistor to 5 places, such as 1.28
1262.5 ohm. Obviously, 1% of this quantity
is over 12 ohms and the value should simply
527 673
527
be written as 1260 ohms. $3 146
There is a definite technique in handling 106
these approximate figures. When giving values 5 713
obtained by measurement, no more figures are 40
www.americanradiohistory.com
HANDBOOK 827
Appendix
STANDARD COLOR CODE- RESISTORS AND CAPACITORS
AXIAL LEAD RESISTOR
-
INSULATED FIRST RING SECONDRING THIRD RING DISC CERAMIC RMA CODE
UNINSULATED BODY COLOR END COLOR DOT COLOR
BROWN- INSULATED COLOR FIRST FIGURE SECOND FIGURE MULTIPLIER 5 -DOT 3 -DOT
BLACK -NON-INSULATED BLACK
BROWN
0
1
0
I
NONE
0
- CAPACITY
RED 2 2 00 MULTIPLIER
ORANGE 3 3 ,000 TOLERANCE
YELLOW 4 4 0.000
TOLERANCE
GREEN 5 S 00.000 TEMPERATURE
-MULTIPLIER BLUE B B 000.000 COEFFICIENT
tST 2ND SIGNIFICANT FIGS VIOLET 7 7 0,000,000
GRAY B B 00,000,000
WIRE-WOUND RESISTORS NAVE IST
DIGIT BAND DOUBLE WIDTH. WHITE 9 9 000,000,000
RADIAL LEAD DOT RESISTOR S- ROTRADIAL LEAD CERAMICCAPACITOR EXTENDED RANGE TC CERAMIC HICAP
rMULTIPLIER CAPACITY
TOLERANCE
ggqt(((0777t
Ir,ITI E'77'TT''CCAPACITY
lI 0
FIGURE CT 011144X1 COEFF LIIE1111Ib1
- 1ST FIGURE TOLERANCE TOLERANCE
MULTIPLIER MULTIPLIER
TC MULTIPLIER
RADIAL LEAD (BAND) RESISTOR BY -PASS COUPUNG CERAMIC CAPACITOR AXIAL LEAD CERAMIC CAPACITOR
AM
TOLERANCE IST FIGURE
FIGURE
I
DII[.IlInG
I
TOLERANCE
V(OPT
E
Ill
MULTIPLIE
i I
TOLERANCE
MULTIPLIER
WHITE (RMA)
2N0} SIGNIFICANT FIGURE
JAN B
2019 TOL.
RATED 300 V.O.C.
CLASS -
TOLERANCE 1STDIGIT
-
BLACK(JAN) MULTIPLIER 1940 RNA MULTIPLIER MULTIPLIER III DIGIT
CODE
CLASS TOLERANCE SIGNIFICANT FIG. 3RD DIGIT
123T
RMA 5 -DOT CODE (OBSOLETE RMA 6 -DOT (OBSOLETE) RMA 4 -DOT (OBSOLETE)
Ilje. IST
IPLI ER Ir 3RDJ SIG. FIGURE
- LULTIPLIER 3R0 WORK. VOLTAGE
_[ -2 ROUT
I
4
T ZND SIG. FIG.
WORK. VOLT. MULTIPLIER
Mi
MULTIPLIER
MULTIPLIER
TOLERANCE 2ND
REAR TOLERANCE SIG. FIGURE
TOLERANCE
WORN.
VOLT. BLANK
WORKING VOLTAGE 1ST }
MOLDED PAPER TYPE CAPACITORS
TUBULAR CAPACITOR MOLDED FLAT CAPACITOR JAN CODE CAPACITOR
NORMALLY STAMPED COMMERCIAL CODE
FOR VALUE IST}
2ND
SIGNIFICANT FIGURE
WORKING VOLTS
SILVER 1ST
2ND
if SIGNIFICANT FIG.
MULTIPLIER BLACK
TOLERANCE
Flail
I %I?IIIII,
IsI
2NTD} SIG. VOLTAGE FIG.
BODY
-MULTIPLIER / MULTIPLIER
/
1S 2NOl SIGNIFICANT TOLERANCE
A 2 -DIGIT VOLTAGE RATING INDICATES MORE THAN --IST J FIGURE CHARACTERISTIC
900 V. ADD 2 ZEROS TO END OF 2 DIGIT NUMBER.
J}r
(,.) : EL
: T , ____ ----- L__ 1
LO+.ss T SEC 1io 1.04-73 n 5(07 02 3.4..-55 T 3E01,674 . ,. r.ss 17 sEC ou
1200
5 .5 -- 4-- 8000
.300 --
e
6
9
--.-
6
.1
e
'9
,
1400
1500
_
100 Y- _ 6000
70.00
i
-
10 1.0
leoo
- W .900 U
+
<
+
,71 2000 =.1=-' 5000
r
U
5 5 N
22.
V
20 2.0 2 2500 000
'
<
V -1-
>
30 3.0
a
2
3000 -+
Cr
t 3000
0 .0 W
W
50 5.0 V 000 ---r-
60 6.0
'
70 7.0 W
90 6.0 5000 2000
'
90 9.0
_'
00 10.0
3
}}
W 6000
1500
150 -6-75.0 1000
O
-22
1
200 20.0
8000
9000
J
300 30.0
10000 1000
L C
www.americanradiohistory.com
829
HANDBOOK
w
V
Z
< r M No o a 0 0 o O N o 0 0 0 0 0 0 00 - 999 0 o O O
N O ^ r- N N f1
- Y in h -- A rf o o G N 0
N A o - N a - 0-
O in 0
N - -
Q D a
o 0
O N 0 O O en
Z -
X
3 Q
O
0
O
0 0
O O
0 0O 0
O N
0 0 0 4 0
0 0 - -- 0 0
N N N N Q O
0 0 0 4 0 0
0 o 0 o
Q O Q 0
- 0-
000000004 0 0 V
o
N
0 0 O
0
N N N
V 0 0 0
0 0 0
N N N N
I ) )
3elllllllllllllllllIIIIIII
CD
IX
w
a
?
Z a 0 0 a 0 0 0 o a 0 0 b a 0 0 a 0 0
D
vr 1-
CC
O Qq
I--
O -I< -IN MI< CV -INN r)
O
Z 0.s2
O w
Z U
D Z
a 0 O4 p..-N
N O
a 0 N
0 0 aA r0 P.o oO O
o o oN o 0 o i
O
j O - N
O
N
N
.
- -
Q
O o p p - N d 0
N
- N d o o 0
0
N
\W
1.
in J
Q Z
X Jin
D
0 F
O
F
-
1- 1- F 1-
0 0 0 o NIn Q 0O 0O 0- 0- NMI 0O 0
1- 1-
0 0 0 N 0 0 0 0 0
O o - -- MI O O O
H
- Nn
F- 1- 1- 1_ 1- 4- 1- 1- F F- 1- 1-
4W
O
J
a
O O O O
v < O Q 'v v o o r) o 0 0 0 0 0 0 0 0 0 0 0 0 0 0
iv
>
QCD
\O
aQ W
N I1 o r o a O N e) 4 o o Q J~
J
)n V
4 M I I O9 0g M I I 0n 0g I O9 I M I i09 I p 0q
CO
Q Q
).
UpW
Cr
W
CL
ZZV O 0 0 O
-
O Q 0 0 O
-
4) 01 4 o o O
_ 0
-
4 0 o O
- -
ID 01
LL1
1-
~2
?4
JpJ
O-
F.
O
Z
ua,
Qy
3W
_J
--IN oI Iv _
0?
U
830 THE RADIO
WdFa- OWOWM.OrP.O-a,OWaW-.00-nr
B 8 ddNWN.NNOWNMWWe-O--MMMOddO-u7d,0-PONFMOaO.
m e mow- O- .00WOVINNVIaV)N--Ndn-NppONWNNOFOdN- Wr
NPVIMOWOdN-OPWrOVIMddMMNNNN+r.-I--.O.00
a. F.O M VId d M M N N N N
J ),
i 0
H V e roWdMdF-rM-dFl9NM0-rMeOc
F-OrOVIWOraNdlnrNFM-WOVfMNN.+rr000000000000
tmmU<
V: C V1 d VI r N n M -W b N d M N N - -
N d M N N r-- -
U tot
dMaMNWOVIF
q
E
U
tO po ON-00
ba OMaNWO FWdP
......
ndM00
O-W-drdPO--V'VI.ObFONWFd
NNMdVIWONOO.nNO-NNNOdF-O`Od-dNFMOOOOd0-
N.n .
as.
ON ---NNMdW.OWCMOO.OM-N.OMVINr-ANMMNppppP
. . . .
.-rNNMdNOWOM.-..OMNMFY`O
-
NNMdNOW0
`O.OPWa.nMM
U a d O W W r F F P O W r W O- d N M W N P r d W F F W F F a ANN
d I i
I I
I
I
-NMMd.OnP-NWMarOWdo-NNOMAMabMrOOON
NNMdV)rP-NVIrr rWM.OaN
Q
-..
- NNNM.P.O.OFP
O
NNO
d F n- W F- M N N P O d
P PNPW P N W O N W- VIW d d a M d N d W N d n
p
M M dA n a
O N- oo O M O- F r M.O M r d W F- O a F F r N r- O
^ NNMYVI.OWOVINO-d-MOMOOWdNP-
NNMdNbWr-Oi.MiO.ON-NV)MdN Y`. OMd O
O PO O
T
Q
r
NNMdN.OW-.OMN NM
Q
a)
W Ci OM 1 i
Vl
IIIIIIIIImPNWNNMCa9dnPOInMtiIIIIIIIIII
e9
- -
r- N N N M 0
`
rR W I I I I I I I I IOMNOOW.OF1dnV NaOFOF.eWInb$O
--- -NMMdV)rWM.O..PMrMP`ONNMMFF
- -NNMMd1n.OFWWO
op
I I I/ v
3 E
;
d .1e
3:
f - WMFWP$M$$SI
-- NNMMdNVrWaO^
.,NCPadInNO..Voon-$pp I I I
0
;
.21'
G) . u
/1 U -WPWPOWFd-WWWOO.O.O.OWONWNNOMONpMOOF 0
^
1
I I I I I FFWaONMd.OWPNNMMMCWdNNbbhrWW ,
0
U U dNMMVIWNWPaOOW-AVI.ONONdNVf.OMO Ofr
VCI
I I I I I IrWaOrNdVrPPdn00N0
V+dV0On0PO-rM
S
!
IU
C
PNOddFN.OMdO.OWMO
I d? IIIIIIIIIIIIIIINMg3:.2MbWP e 1
VJ
V 8
F opU
.o.o.OrOV)OWO.NOdd-OMMFOaFOV) 94
B
w
I I I I I I I
FWaONMNOWrM.OPMFr.p
N N NN MMd dN N
W N-O-MFMWNWdppN
F W P O- N d N P
0
OO
3 r
V-`.
I'M
(,
PFddSVIN-L'""'
000pOOOOOdOWFFMWdWNONpp..dOd-POOaMOaryO,VINP
-d0
M.00.OMOWOVIdlMNN---
WOVIdMNN---
NVIOVfNOFONMMN---
`O-F.-NWVIOMNVI--NV)O.ONOW.ONMNN--..
-dNOd-UfWFV)OnN.M-.INnOWh00 9 QQ
oy
E
0
d-OWFN000.0 NFOOdOyOMNNWOVM
M.OdMPOMVIdPnWPOOWNMWPMN-aaNONOPPOMVf- pq S
O S c a F P d- N d W d- O O- d F O O N r W N N O F N d N r OWFFOVfNdMMM v 0
Q /. WVINOWOdN-OPWF.OV)V)dMMNNNN
NNNM
..\i
.!
oy
h m
col
O4 +NMdNOrWPp-NMdN.OFWPNMMtM9MtMlMMMMa)
0p
L
C7 f
www.americanradiohistory.com
HANDBOOK 831
TO CHANGE
TO OPERATOR
FROM
I -F TRANSFORMERS
PLATE LEAD BLUE
E l- LEAD RED
GRID (OR DIODE ) LEAD GREEN
A -VC (OR GROUND) LEAD BLACK
AUDIO TRANSFORMERS
PLATE LEAD (PR /. ) BLUE OR BROWN
B t LEAD (PR/.) RED
GRID LEAD (SEC.) GREEN OR YELLOW
GRID RETURN (SEC.) BLACK
Index
832
filament supply, 614 all -band center -loaded, 139 directive, 476
grid bias, 614 auxiliary trimmer, 137 -138 Franklin, 481
grid circuit, 654 -605 10- meter, 538 -139 Franklin collinear, 439
plate fad, 654 multiband, 447-455 parasitic, feed systems for, 516 -523
radio -frequency vacuum -tube, 147 -168 center -fed, 449 -450 delta match and T-match, 520 -521
RC, general characteristics, 127 end -fed Hertz, 448 driven element with center feed, 521
r -f folded fiat -top, 450 -451 folded -element match calculations,
grid -circuit considerations, 158 -159 low- frequency discone, 452 519 -520
grounded -grid, 119 -161 molter, 451 gamma match, 521 -522
plate- circuit considerations, 149 -110 single -wire -fed, 452 -453 matching stub, 522-523
plate -circuit coupling, 149 -150 3/4 -wave folded doublet, 447 -448 rotary -link coupling, 521
power, cathode -follower, 161 -162 two -band Marconi, 448 -449 stacked -dipole, 481 -483
power, class -C, 150 -156 vertical, 453 -455 collinear, 482 -483
power, plate- return, 161 -162 multielement vhf beam, 105 -111 double extended Zepp, 483
screen- circuit considerations, 119 eight -element tiltable array, 106-508 in -phase spacing, 481
tuned cathode circuit, 166 -168 long yagi, 111 out-of-phase spacing, 481 -482
resistance, 103 screen beam for 2 meters, 509 -111 spaced half -wave, 483
resistance- capacitance coupled, 124 -127 three-element beam, 505 -506 stacked yagi, 517
response, RC, 127 physical length, 412 -413 three-element, rotary -beam, 514-516
sweep, oscilloscope, 173 polarization, 412 element spacing, fly
tank circuit, Q of, 155 vhf and uhf, 495 -496 length of parasitic element, 515 -516
tetrode, linearity curve of, 164 radiation from, 410 -411 materials for elements, $18
tetrode pi- network, 659 -662 requirements, vhf and uhf, 493 -496 more than three elements, 516
grounded -screen, 661 -662 changeover, 494 stacking of yagi arrays, 516
inductively tuned tank circuit, 662 insulation, 495 tuning, 528.530
transistor i -f, 114 polarization, 495 -496 matching to the antenna
Tribander linear, 692 -697 radiation angle, 494 -495 transmission line, 528 -529
circuit, 692 -696 radiator cross section, 49f raising and lowering, 530
transmitter adjustment and transmission lines, 494 unidirectional driven, 523 -525
tuning, 697 resonance, 414 -415 feed methods, 523 -525
transmitter control circuits and space -conserving, 444 -446 stacked broadside arrays, 523 -525
power supply, 696 -697 broad -band dipole system, 446 unidirectional parasitic end -fire, 512 -514
- riode power, operation twin -lead Marconi, 446 Assymetrical speech, 699 -700
characteristics of, 133 -135 supports, 461 -464 Atom, 23, 24
triode, single -ended, 133 -131 A -frame mast, 461 Atomic structure of germanium and
types of, 122 -123 guy wires, 462 silicon, 100 -101
vacuum -tube, 120 -146 insulation, 463 -464 Audio
vertical, oscilloscope, 170 -171 painting, 463 circuitry, transistor, III -114
video-frequency, 127 -128 trees as supports, 463 detector, and control circuits, 213 -216
voltage, PNP transistor, 111 TV masts, 461 -462 a-c loading of second detector, 254
833
Balanced -line bridge, 765 Bridge- rectifier supplies, IVS, 731 -733 keying, 403 -404
Balanced modulator, 336 -347 Bridge -T oscillator, 192 modulation, 293-294
circuits, 343, 347 Bridge -type standing -wave indicators, installation, 291
diode, 336 760 -762 tetrodes, 294
for mechanical filter, 343 Bridge -type v.t.v.m., 710 types, 78
Balun, broadband 30 -ohm, 431 Broadband dipole system, 446 Cathode- coupled phase inverter, 131.132
Bandpass characteristic of low -frequency Broadband 10 -ohm balun, 431 Cathode- driven amplifier, 121, 662 -667
SSB filter, 339 Broadcast interference, 374 -391 choice of tubes, 666
Bandspread circuits, 201 Broadside and end -fire arrays, construction, 664 -661
Bandwidth, 423 combination, 491 -492 distortion products, 663 -664
modulation, 278 Broadside arrays, 483 -488 feedthrough power, 662 -663
SSB, 333 bi- square, 486 -487 neutralization, 666-667
Base -collector waveform, 113 bobtail bidirectional, 487 -488 tuning, 661 -666
Base electrode, 101 cubical quad, 487 Cathode -driven high - triodes, 166
Basic filters, 75 double- Bruce, 481 -486 Cathode follower, 122 -123
Bas -attenuation chart, 304 six -shooter, 487 amplifiers, use of, 144 -141
Bass suppression, 304 stacked, 323 -521 circuits, 269
Battery bias, 262-263 stacked dipoles, 483 -484 in r -f stages, 141
BCI, 374-391 X- array, 484 modulator, 285.287
Beam antennas, vhf multielement, 501 -311 "Building -out" modulation transformers, power amplifiers, 143 -141
eight -element tiltable array, 106 -108 302, 303 r -f power amplifier, 161 -162
long Yagi, 111 Butterfly circuit, uhf, 222 -223 Cathode -modulated stage, excitation, 294
screen beam for 2 meters, 109 -111 Button -mica capacitors, 361 Cathode -ray oscilloscope, 169 -184
three-element beam, 101.106 Bypass capacitors, 360 -361 Cathode -ray tube, 94 -97
Beam -deflection tube, 317 operation, 94 -97
Beam -power tube, 87 -90 Cavities, resonant, uhf, 221 -222
Beam tetrodes C Center -fed antenna systems, tuner for,
neutralizing, 248 471 -473
parasitic', 370-372 Capacitance Center -fed feed -point impedance, 416 -417
plate modulation, 293 calculations, 36, 37 Center -fed half -wave horizontal
Beams, rotary, 112 -132 grid -cathode symbol for, 120 antennas, 417 -441
Beat -frequency oscillators, 212 -213 grid plate, symbol for, 120 delta- matched and standard doublets, 440
Beat oscillator and product detector, 228 input, 271 doublets with quarter -wave
Beta, d -c, 107 input, formula, 121 transformers, 439
Beta cutoff frequency, 107 input, symbol for, 120 multiwire doublets, 439 -440
Beta formula, 106 interelectrode, 86, 121 off -center -fed doublet, 440 -441
BFO adjustment, 226 neutralization of, 121 -122 tuned doublet, 437 -439
B -H curve, 41 output, symbol for, 120 Center -fed multiband antenna, 449 -410
Bias plate- cathode, symbol for, 120 Center -fed Zepp antenna, 437
automatic, 263 static interelectrode, 121 Center -loaded, all -band, mobile antenna,
battery, 262 -263 stray, 206 539
cathode, 123, 263 stray circuit, 206 Center -tap keying, 402
cutoff, 81, 120 tank- circuit, 253 Ceramic- dielectric capacitor, 161
fixed, 123 unit of, 34, 31 Changeover, antenna, vhf and uhf, 494
forward, 102 Capacitance charts for tank Q, 255, 256 Characteristic curve
grid, 123 Capacitance meter, transistorized, 737 -758 emitter, 109
grid -leak, 123 Capacitance variation, unction, 102 In vs. Ep, 88
grid -leak for high -ji triodes, 127 Capacitive boost coupling, 386 triode, 83 -84
reverse, 102 Capacitive coupling, 265 -266 Characteristic impedance, two -wire lines,
safety, 262 -263 Capacitive reactance, 12 430
self, 262 Capacitors, 33 -39, 360 -362 Charge transport, 106
Bias configuration for transistor, 111 button -mica, 361 Chassis material, 367
Bias feed, parallel and series, 267 bypass, 360-361 Checking for parasitic oscillations, 372-373
Biasing methods, 123 ceramic -dielectric, 361 Chokes, r -f, 267 -268, 364
Bias -shift Heising modulator, 301 -307 coaxial, 361 Circuit
Bias -shift modulator, operating doorknob, 361 a -c, 46
characteristics, 101 electrolytic, 38, 39 a -c bridge, 716
Bias supply filter, 733 -735 autodyne converter, 115
for grounded -grid amplifiers, 166 in -c and d -c circuits, 37, 38
a automatic- volume-control, 111
separate, 261 in series, a -c, 38 balanced -modulator, 347
shunt -regulated, 718 in series, voltage rating, 38 handspread, 203
Bias systems, types of, 122 mica -dielectric, 360 -361 butterfly, uhf, 222 -223
Bidirectional rotatable arrays, 121 series inductance, 360 cathode -follower, 269
Bishop noise limiter, 217 -218 series resistance, 360 clamping, 188
Bi- square broadside array, 486 -487 shunt resistance, 360 counting, 190.191
Bistable multi vibrator, Fccles- Jordan, 118 stored energy, 35 differential keying, 406 -408
Blanketing, 384 variable air, 361 -362 equivalent, transistor, 109
Blanking circuit, oscilloscope, 171 -172 variable vacuum, 362 i -f tuned, 206 -211
Bleeder resistors, 731 Carrier elimination, 332 beat -frequency oscillators, 212 -213
Bleeders, safety, 401 -402 circuits, 336 -338 crystal filters, 209, 211
Blocked -grid keying system, 404 Carrier, reinserted, 335 mechanical filter, 211 -212
Blocking, 376 Carrier tube, 294 Miller effect, 208
Blocking oscillator, 190 Carrier -wave distortion, 309 practical filters, 209
transistor, 118 Carrier -wave pattern, 180 rejection notch, 210 -211
Body static, 143 Cascaded voltage -amplifier, 127 shape factor, 207 -208
Bombardment cathode, 79, 81 Cascode amplifier, 202 transformers, 206 -207
Breakdown ratings of plate spacings, 257 Cathode, 75, 77 variable-selectivity filters, 209 -210
Bridge bias, 123, 263 limiting, 183 -188
balanced -line, 765 bombardment, 79, 81 audio peak, 183 -187
measurements, 751 -757 coupling, 129 -130 diode limiters, 185
SWR, 472 current, 88, 90 grid limiters, 187 -188
Bridge rectifier, 714 heater, 78, 80 -81 Loftin- White, 133
834
www.americanradiohistory.com
low -frequency equivalent transistor, 109 Clipper circuits, 301 -302 stages. 162
magnetic, 40 Cli pper-filter, 298 Controlled -warmup tube, 99 -102
multivoltmeter, 741 action, 299 Controlling transmitter pcwer output, 395-
one -way RTTY, 326 Clipper, gate, series -diode, 187 )97
parallel- resonant, 61 Clipper speech amplifier, 142 Controls, trans- receiver, 143.547
parallel RLC, 60 -62 Clipping, speech, effects of, 140 Conductance, conversion, 70, 92
peak -voltage measurement, 749 Clipping, speech, low -level, 700 Conductors, 24
phase- modulation, 317 Closed -loop feedback system, 193 Conversion conductance, 92
push- button transmitter control, 399 Classifications, transistor, 107 -108 Converter
push -pull, 137 Coaxial capacitors, 361 "antioverload," for 50 MHz, 363 -567
regenerative detector, 193 Coaxial circuits, coupling into, 221 adjustment, 566 -547
resonant, 62 -68 Coaxial line, 432 -413 circuit description, 164
r -f feedback, 270 -276 Code, International Morse, 15 -22 construction, 564 -366
series -resonant, 62 Code practice set, 20, 22 audio -frequency, 327 -328
series RLC, 56 -60 Code, teleprinter, 325 -326 audio RTTY, 321
SSA generation by phasing, 343 Coefficients of tetrodes and pentodes, circuit, autodyne, Ill
tank, 61 -6g 89, 91 412 -MHz low -noise, 394 -604
transistor oscillator, 117 Coercive force, 41 i -f amplifier
835
www.americanradiohistory.com
construction, 97 let, 440 control -grid, in grounded-grid stages,
electromagnetic, 96 -97 Depletion control, 119 162
operation, 97 Design and installation of mobile equip- grid, symbol for, 121
phosphors, 99 ment, 533 -117 Distortion
screen material, 97 -99 Detection or demodulation, 194 -195 amplitude, 124 -125
trace, 97, 99 autodyne detector, 191 frequency, 123
Crystal radiotelegraphy reception, 195 in amplifiers, 123 -125
current, 237 radiotelephony demodulation, 194 intermodulation, 168
cuts, orientation, 235 Detector, audio, and control circuits, 213- modulated- amplifier, 304
doped, 101 216 phase, 123 -124
galena, 100 a -c loading of second detector, 214 products, SSB, 349
heating and fracture, 237 audio amplifiers, 215 -216 products due to nonlinearity, 310
holders, 236 automatic volume control, 214 Dissymmetry, speech waveform, 280
lattice filter, 341 avc in bfo- equipped receivers, 214 Divider, frequency, 190 -191
ovens, 236 detectors, 213 -214 D layer, 426
overtone -cut, 237 signal -strength indicators, 214-215 Duct formation, 424
piezoelectric, 209 Detectors, 195, 213 -214 Doherty
rectifiers, 224 autodyne, 195 linear amplifier, 289, 296
switching, 240 circuits, regenerative, 195 operating efficiency, 294 -297
Crystal and coil tables for vhf exciten, envelope, schematic, 183 modulated amplifier, 294 -297
622 Frrmodyne, 197 Dome audio phase -shift network, 344, 345
Crystal -diode wavemeter, 379 product, 227 -228, 316 -357 Donor impurities, 101
Crystal filters, 341 double -triode, 228 Doorknob capacitors, 361
i -f, 22f -226 triple -triode, 228 Doped crystal, 101
circuits, 209, 211 ratio, 321 -322 Dot multivibrator, 648
Crystal oscillator, 234 -242 second, loading of, 214 Double -Bruce array, 481 -486
1763, 240 Deviation Double conversion, 202-203
keying, 240 f -m, 310 Double -diode paired modulator, 338
tuned -plate, 238 measurement, 318-319 Double extended Zepp, 483
tuning, 239, 240 of single -sideband signals, 332 Double- resonance curve, 181
Crystal oscillator circuit, 237 -242 Devices Double- resonance effect, push -pull tank
all -band, 240 SCR (silicon controlled rectifier), 104 circuit, 367
harmonic, 238 -239 semiconductor, 100 -119 Double -sideband supp d- carrier signal,
pentode, 238 -240 Diamond antenna, 479 -481 338
Cubical quad antenna, 487 Dielectric Double skeleton cone antenna, 498
Current breakdown, 36 Driven arrays, unidirectional, 123 -525
alternating, 46 constant, 36 feed methods, 123 -52f
amplification formula, transistor, 106 materials, 31-36 stacked broadside arrays, 123 -121
crystal, 237 Differential keying circuits, 406 -408 Driven element with center feed, parasitic
feed, 418 -419 Differentiator, RC, 69 -71 arrays, 121
flow, 25 Diode, 80 -83 Driver -stage calculations, 142
in a triode, 82, 84 balanced modulator, 336 Double- triode product detector, 228
in tetrodes and pentodes, 88 -91 characteristics, 80, 82 -83 Doublers, frequency, 211
formula, collector, 113 color codes and markings, 101 Doubling and tripling circuits, varactor,
forward, 102 keyer for frequency shift, 329 103
gate, SCR, 104 limiters, 181 Doublet
ground. 364 -365 mixers, 90 -92 delta- matched and standard, 440
leakage, 102 superheterodyne receiver, 200 off- center -fed, 440 -441
Curtain arrays, 483 operation, series, 723 -724 3/4 -wave folded, 447 -448
Cutoff bias, 81 PN junction, 102 -101 tuned. 437 -439
symbol for, 120 ring modulator, 337 -338 multiwire, 439 -440, 416 -417
Cutoff frequencies, 107 steering, 118 with quarter -wave transformers, 419
alpha, 107 tunnel, 102 -103 Drift space, 91
beta, 107 zener, 104 Drift velocity of hole, formula, 101
transconductance, 107 Dipole system, broadband, 446 Dropping- resistor screen supply, 264
Dipole, twin -lead folded, 439 Dual -band, folded gat-top, antenna, 410-
Direct-coupled cathode multivibrator, 188 431
D Direct coupling, 130 Dummy load, 713 -714
Direct-current circuits, 23 -41 assembly, 754
Dash flip -flop, 610 Direct current, pulsating, fl 2- kilowatt, 753
D -c Direct f-m circuits, 312-316 DX
amplifiers, 132 -133 Directive arrays, 476 aurora -type, 424
beta, 107 Directive high- frequency antennas, 474- operation, 14
circuit, capacitance in, 37, 38 492
grid -supply voltage, symbol for, 120 Directivity
plate -supply voltage, symbol for, 120 antenna, 418 -421 E
Decibel, 803 -807 end -fire, 488 -491
De-emphasis, 323 flat -top beam, 489 -490 Eccles-Jordan bistable multivibrator, I18,
circuit, 323 triplex beam, 491 189 -190
Deflection horizontal, 412, 418 -419 Effective Q, 213
coil, 96, 99
electromagnetic, 99
electrostatic, 98
of radiation, 411
vertical, 412, 419
Discone antenna
Ex -
Effective value of a.c., 10
1p curve, 163
Eight -element tiltable array, 506 -108
polarity switch, oscilloscope, 170 low- frequency, 412 E layer, 426
Demodulation or detection, 194 -195 vhf uhf, 498 -500
and Electric filters, 74 -76
autodyne detector, 195 Discontinuities, line, 434 -43f design, 76
radiotelegraphy reception, 195 Discriminator Electrical
radiotelephony demodulation, 194 Foster -Seeley, 320-321 quarter -wave line, 296
Delta match and T- match, parasitic arrays, Tnris, 320 relationships, fundamental, 24 -33
120 -521 vector diagram, 321 symbols for transistors, 106
Delta- matched antenna system, 416 voltage-frequency curve, 320 units, fundamental, 24 -27
Delta- matched doublet and standard doub- Dissipation Electrical equivalent, r -f choke, 363
836
www.americanradiohistory.com
Electrode, focusing, 91 F swing, 310
Electrolytic capacitors, 38, 39 terms, 310
Electrolytic conduction, 71 Fading, 427 -428 transmission, narrow -band, 311 -312
Electromagnetic CRT, 96 -97 Feedback, 192 -193 tuning and operation, 680 -682
Electromagnetic deflection, 97 -99 amplifier, 145 -146 Focus coil, 96, 99
Electromagnetism, 39 -43 gain and phase shift in, 145 -146 Focusing electrode, 91, 97 -98
Electron, 23 -24, 71 neutralization, 274 -276 Foster-Seeley discriminator, 320 -321
emission. 75 parasitic oscillations, 274 Folded dipole, twin -lead. 439
energy dissipation, 81 relationships, 145 Folded doublet, 3/4 -wave, 447 -448
flow, 25 circuit, shunt, 146 Folded-element match calculations, para-
gun. 91, 97 -98 closed -loop, 193 sitic arrays, 119 -120
Electron- coupled multivibrator, 188 types of, 146 Folded flat -top dual -band antenna, 450 -431
Electron- coupled oscillator, 231 -232 voltage, vector relationship, 274 Forward bias, 102
Electronic conduction, 71 Feed methods, unidirectional driven ar- Forward current, 102
Electronic keyer for RTTY, 328 rays, 523 -125 Four -tone SSE envelope, 335
Electronic test equipment, 744 -772 Feed -point resistance, 411 Franklin antenna array, 481
Electron -ray tube, 99, 101 -102 Feed systems Franklin collinear array, 439
Electron tube at vhf, 90 -91 for parasitic arrays, 516 -523 Franklin oscillator, 233, 234
Electron -tube types, 76 delta match and T-match, 520 -521 Fremodysse detector, 197
Electrostatic cathode -ray tube, 91 driven element with center feed, 521 Frequency
Electrostatic deflection, 91 -96, 98 folded -element match calculation, 519- alpha cutoff, 107
Electrostatics, 33, 34 520 beta cutoff, 107
Element, metallic and nonmetallic, 100 gamma match, 521 -522 conversion problems, 349 -310
Emission matching stub, 122-523 conversion systems, SSB, 346 -350
cathode, 77 rotary -link coupling, 521 critical, 426
electron, 71 Q- section, 419 -460 cutoff, 107
equation, 79, 81 Feedthrough power, 210 detector, 319 -320
photoelectric, 75 -77 Field -effect transistor (FET), 118 -119 distortion, 123
secondary, 81, 82 If -watt clipper -amplifier, 704 -706 divider, 190-191
.pace- charge- limited, 82, 83 Filament doubler, 231
:hermionic, 71 oxide -coated, 80 doubler, push -push, 252
Emitter characteristic curve, 109 reactivation, 77 effects, resistor, 359
Emitter electrode, 105 temperatures, 78, 79 measurements. 758 -719
End -fed tungsten, 78, 79 modulation. 308 -324
half -wave horizontal antennas, 436 -437 Filament -type tubes, 76 multiplier, 250 -213
Hertz antenna, 436 -437, 448 Filament types, 76 multiplier, 6)60, 624
End -fire Filter spectrum, 46, 47
arrays, unidirectional parasitic, 112 -114 assembly, 72 -74 spotter, 100 -kHz, 719
directivity, 488 -491 basic, 72 transconductance cutoff, 107
flat -top beam, 489 -490 capacitors, 733 -731 tripler, push -pull, 252 -213
triplex beam, 491 choke coils, 731 Frequency bands, amateur, 13 -11
End -fire and broadside arrays, combina- click, 402 Frequency -impedance characteristic, r -f
tion, 491 -492 crystal, 209, 211 choke, 363
Egvelope crystal, i -f, 221 -226 Frequency -shift
detector, schematic, 183 crystal lattice, 341 keyer, diode, 329
modulation, 277 half -wave, 382 -383 keying, 323, 328 -329
Epitaxial planar transistors, 111 high -pass, 74 -76, 37f, 376 Front -end alignment, 226
Equations, gain high -pass transmission line, 375 FSK, 325, 328 -329
for pentode r -f amplifiers, 15I low -pass, 74 -76, 379 -384 F, layer, 421
for RC- coupled amplifier, 126 mechanical, 211 -212, 341 Fachs antenna, 436 -437
Equipment, mobile, design and installation method for SSB generation, 338 Full -wave
of, 533 -117 operation, 75 limiter, 220
post -modulator, 303 rectifier, 713 -714
Equipment, speech and a -m, 698 -706
Equivalent circuit
power line, 216 rectifier, v.t.v.m., 752
practical, i -f circuits, 209 Fundamentals, radio receiver, 194 -228
pentode RC-coupled amplifier, 126
resistance -capacitance, 711 Fusion process, 105
quartz crystal, 231
variable -selectivity, 209 -210
resistor, 319
Filter sections, 75, 76
series circuit, 60 -61
G
constant -k, 71, 76
transistor, 109
m- derived, 75, 76
Error cancellation, 193 Gain
Filter -system resonance, 711 -713
Error signal, 193 100 -watt amplifier,
671 -682 equations for pentode r -f amplifiers, Ill
Excitation of cathode -modulated stage, 294 circuit,676 -677 equations for pentode RC- coupled ampli-
Exciter construction, 677 -680 fier, 126
6 and 2 , 620 -624 design, 676 measurement, parasitic. 373
adjustment, 623 -624 Fixed bias, 123 power, transistor, 108
circuit, 621 Flat -top beam, 489 -490 resistance, transistor, 108
construction, 622 -623 Flip -flop, dash, 610 voltage, transistor, 108
frequency multiplier, 624 Flip -flop multivibrator, oscilloscope, 171 Gain and phase shift in feedback ampli-
SSB 175 -watt, 624 -631 Floating paraphase phase inverter, 131 fiers, 145 -146
circuit, 624 -626 Flux density, 40 Gain -bandwidth product, 107
dial assembly, 627 -629 Flywheel effect, 67 Galena crystal, 100
layout, 626 -627 F-nl, 308 -324 Gamma match, 417
main bandswitch assembly, 627 deviation, 310 parasitic arrays, 121 -522
wiring and testing, 629 limiters, 322 Gases, inert, 100
SSB filter, 353 -355 modulation index, 310 Gas tubes, 97 -100
"Ten -A," 353, 354 modulation index, effect, 312 Gate clipper, series -diode, 187
transistor, 117 percentage modulation, 310 Gate current, SCR, 104
transistor SSB, 313, 315 reception, 197 Gr, 92
Exciters and transmitters, low -power, 620- receiver design, 322 -324 General -purpose triode class -B modulator,
6x1 side frequencies, 310 702 -704
Extrinsic, 101 signal, slope detection, 319 Generation
alternating current, 47, 48 H Hot- cathode phase inverter, 131
radio -frequency energy, 229 -268 Hysteresis, coercive force, 41
SSB signals, 338 -346 Half -wave
Generator, sweep oscilloscope, 172 -173 antenna, physical length, 412-413
Generator "whine," 542 -543 center -fed horizontal antenna, 437 -441
Germanium delta- matched and standard doublets,
N (negative) -type, 101 440 Idler loop, 103
P(positive) -type, 101 doublets with quarter -wave trans- I -f
Germanium and silicon atomic structure, formers, 4)9 alignment, 221
100 -101 multiwire doublets, 439 -440 alignment, receiver, with oscilloscope,
gm, 91 off- center -fed doublet, 440 -441 179 -182
Graphical representation, 811-123 tuned doublet, 437 -439 amplifier coupling, 207
Grid end -fed horizontal antennas, 436.437 amplifier, superheterodyne receiver, 198
excitation, 243 -244 filter, single -section, 382, 383 passband, 208
excitation voltage, peak, symbol for, rectifiers, 713 response curve, 1111
www.americanradiohistory.com
static, 121 Kirchhoff's Laws, 31, 32 formula, 137
Interstage coupling, 263 -267 Klystron, 91 -95 symbol for, 121
inductive, 266 power, 93 -94 Loading, tube input, 149
push -pull transformer, 128 reflex, 94 -91 Local loop supply, 321, 329
Intercoupling of ground currents, 365 Location of keyed stage, 402
Interference Locating noise sources, 344
broadcast, 374 -391 Loftin -White circuit, 133
classifications, 384 -387 Long -wire radiators, 476 -477
hi -fi, 391
Logarithms, 798 -803 Long Yagi antennas, SII
Lagging phase, 52, 53 Loop, idler, 103
image, 389, 390
Lamb noise limiter, 217 lower sideband, 332
isolating, 378 -379
television, 374 -391 Lacy -fl antenna system, 484 Low -frequency
Lead filtering systems, 366 discone antenna, 452
Irtermodulation distortion, 168
Lead inductance, effects, 90 -91, 9) equivalent circuit, transistor, 109
Internal leads, 366
Intrinsic resistivity of silicon, 101 leads, high -voltage, 362 parasitic oscillations, 368 -369
Intrinsic semiconductor, 101 Leakage current, 102 parasitic suppression, 368
Leakage reactance, 73 Low-level speech clipping, 700
Inverse voltage, 102
Inverter, phase, 130 -132 Leading phase, 52, 53 Low -noise transistor preamplifier, vhf, 116
cathode- coupled, 131 -132 Lerre -A'rs file alternator, 536 Low -pass filters, 74 -76, 379-384, 390
Boating paraphase, 131 Left -hand rule, 39 construction, 310 -381
hot- cathode, 131 Length of parasitic element, SI5 -516 using, 381 -383
voltage- divider, 132 Licenses, amateur, 12 Low -power transmitters and exciters, 620-
Iron -core inductors, 43 Limiter 631
Ion, definition, 25 audio noise, 218 low-range ohmmeter, 747
Ionization density, 425 circuit, alternative, 219 -220
electronic for RTTY, 328 er supply, 696 -697 marching stubs, 457
transistor, 648 -651 Utility 2- kilowatt, 682 -692 multiwire doublets, 456 -457
circuit, 648 -650 circuit, 682 -685 Q- section feed, 459 -460
construction, 651 construction, 685 -688 quarter -wave matching transformers, 459
dash flip-flop, 650 external anode tetrodes, 688 -690 standing waves, 455.456
tone oscillator, 650 -651 tube combinations, 690.691 T match, 457
Keying tuning and adjustments, 691 -692 voltage feed, 458
audio frequency -shift, 325 -330 Linear r -f transformers, 459 Matching stubs, 457, 322 -523
cathode, 403 -404 Linearity curve of set rode amplifier, 164 parasitic arrays, 522.523
center -tap, 402 Linearity tracer, block diagram, 182 Matching to the antenna transmission line,
circuits, differential, 406 -408 Linearity tracer, oscilloscope, 183 -184 528 -529
crystal oscillator, 240 Linearity traces, 184 Matching transformers, quarter -wave, 459
frequency -shift, 325 Lines, short, tuning, 220 Mathematics and calculations, 784 -126
grid -circuit, 404 -405 Lines, tuned or resonant, 433 -434 MBK, 325
make -break, 325 Link coupling, 266 -267 M- derived section, 75, 76
screen -grid, 405 -406 Link, variable, coupling, 466 Measurement
transmitter, 401 -409 l.issajous figures, oscilloscope, 175 -178 alternating current and voltage, 747
Kilohertz, 47 determination of phase angle, 178 balanced transmission line, 764 -765
Kilowatt oscilloscope settings, 175 -176 circuit constants, 754 -755
dummy load, 754 phase -difference patterns, 176 -178 deviation, 318 -319
CCS power supply, 743 Litz wire, 64 frequency, 758 -759
linear amplifier for six meters, 668 -675 L- matching networks, 258 modulation, 279
adjustment, 673 -673 L- network impedance, 258 power, 752 -754
a -m linear operation, 675 Load impedance, symbol for, 121 with a bridge, 751 -757
circuit, 668 -669 Load line, 83 -84 Mechanical filter, 211 -212, 341
construction, 670 -671 class -B a -f amplifier, 141 Mechanics of modulation, 278 -210
metering and suppression circuits, 670 triode, 85 -86 Mechanism of conduction, 101 -102
plate -circuit assembly, 671 -673 Load resistance Medium- and low -range ohmmeter, 744-
839
747 double- diode, 338 push -pull, 241 -246
Megahertz, 47 general -purpose triode class -B, 702 -704 r -f amplifiers, 244 -250
Mercury -vapor tubes, 98, 100, 718 grid -bias, 283 screen -grid amplifiers, 247 -250
Metallic element, 100 reactance -tube, 312 -316 shunt, 246 -247
Meter, grid -dip, 129 series cathode, 294 -291 techniques, 276
Meter shield, 361 tetrode, 698 -699 tests, 273
Metering and suppression circuits, 670 35 -watt, 114 Neutralizing
Meteors and bursts, 429 vacuum- diode, 339 procedure, 247 -250
Mica -dielectric capacitors, 360 -361 zero -bias triode, 706 single -ended tetrodes, 249 -250
Micromho, 83 Monitoring circuit for modulated -wave Nodal point, 266
Microphones and circuits for mobile equip- pattern, oscilloscope, 180 Noise
ment, 541 Monitoring transmitter performance factor, 148.149
Microwave tubes, 93 -96 with oscilloscope, 178 -179 ignition, 141 -542
Miller effect, 121, 208 modulated -wave pattern, 179 sources, locating, 144
Miller oscillator, 238 trapezoidal pattern, 179 Noise and images, mixer, 200 -201
Misilarlo,coil material, 160 waveforms, 178 -179 injection voltage, 200
Minimum instantaneous plate voltage, sym- Monitor scope, 769 -770 triode mixers, 200
bol for, 120 MSCL, 81 Noise limiter, audio, 218
Mismatch, plate circuit, 143 Mu (), 81 -82, 90 Bishop, 217-218
Mixer M.u.f., 13, 426 Lamb, 217
diode, 90, 92 Mu factor, grid -screen, 88 peak, 216 -217
diode, superheterodyne receiver, 200 Multee antenna, 451 peak, practical, 218 -219
noise and images, 200 -201 Multiband antennas, 447 -430 Noise recording, sun, 604
injection voltage, 200 center-fed, 449 -410 Noise suppression, 216 -220
triode mixers, 200 end -fed Hertz, 448 alternative limiter circuit, 219 -220
pentagrid, as a product detector, 217 folded flat -top, 450 -451 noise limiter
pentode, 92 low- frequency discone, 452 audio, 218
stages, SSB, 346 -310 multee, 401 Bishop, 217-218
triode, 90, 92 single- wire -fed, 412 -453 full -wave, 220
twin triode, 346 -347 3/4 -wave folded doublet, 447 -448 Lamb, 217
unit for 432 MHz, 198 two -band Marconi, 441.449 peak, 216 -217
Mixer and converter tubes, 89 -92 vertical, 413 -415 practical peak, 218 -219
Mobile Multivibrator, 234 TNS, 220
antennas for, 537 -139 Multicavity magnetron, 96 power -line filters, 216
all -band center- loaded, 139 Multielement vhf beam antennas, 50f -Iii Nondirectional vertical array, 498
auxiliary trimmer, 537 -138 eight -element tiltable array, 106 -108 Nonlinearity, distortion products, 310
10- meter, 038 -539 long Yagi, 111 Nonresonant transmission lines, 430 -433
converter, transistorized, 133 -535 screen beam for 2 meters, 109 -111 Nonsinusoidal waves, 68, 69
equipment design and installation, 133- three -element beam, 105 -106 Notation, polar, 04, 11
157 Multiplier Notch, rejection, 210 -211
power sources, 531 -137 frequency, 201 -203 Null circuit, Q- multiplier, 226
three -phase, 136 6360, frequency, 624 Nuvistor converter, 560 -063
transistor power supplies, 113 -006 sweep -speed, oscilloscope, 172 adjustments, 062 -163
operation, 013 -104 Multirange meters, 746 circuit description, 560 -061
rating, 104, 101 Multivibrator construction, 161 -162
self- starting oscillators, 11f bistable, Eccles- Jordan, 118 Nuvistor two -meter converter. 067-072
transformer, 110, 116 direct- coupled cathode, 188 adjustments, 170 -171
Modulated -amplifier distortion, 304 dot, 648 antioverload modification, 571 -172
Modulated -wave p , monitoring cir- Eccles-Jordan, 189 -190 circuit description, 567 -068
cuit for, 180 electron -coupled, 188 construction, 168 -169
Modulation, 308 -310, 698 -701 flip-flop, oscilloscope, 171
amplitude, 277 -307 Multivoltmeter circuits, 745
assymetrical speech, 699 -700 Multiwire doublets, 439 -440,
bandwidth, 277
406 -407 0
Mutual conductance, 82 -84
capability, 279 -280 Mutual inductance, 42 Off- center-fed doublet, 440 -441
cathode, 293 -294 Ohm's Law, 27
envelope, 277 alternating current, 11
factor, 279 application, 27 -33
N
frequency, 308 -324 impedance, 64
high -level splatter suppressor, 700 -701 Narrow -band f -m transmission, 311-312 magnetic circuits, 40 -41
increasing effective percentage of, 699 Nature of SSB signal, 334 -331 Ohmmeter, low- range, 747
low -level speech clipping, 700 nbfm, 311 -312 171 -watt SSB exciter, 624 -631
index, f -m, 310 405 -kHz adapter, 323-324 circuit, 624 -626
measurement, 279 Negative and positive clamping dial assembly, 627 -629
mechanics, 278 -280 circuits, 187 layout, 626 -627
percentage, 277 -279 Negative -grid characteristics, triode, 83 main bandswitch assembly, 627
graphical determination, 279 Negative -grid oscillator, 230 wiring and testing, 629
phase, 316 -319 N (negative) type germanium, 101 I -KW IVS power supply, 728 -729
plate and screen, 292 -293 Negative -resistance oscillator, 233, 234 One -way RTTY circuit, 326
screen grid, 284 -281 Nonmetallic element, 100 Open line, two -wire, 430 -433
suppressor grid, 288 -289 NPN transistor, 101 Operating parameters, 137
system, input, 219 -293 Network Operating region, 81
system, plate, 289-290 pi -L, 261 -262 Operational circuit using 10 -kHz
tetrode, 691 -699 pi- matching, 219 -260 sideband filter, 340
transformer calculations, 140 -141, L- matching, 218 Operator, j, 13, 14
291 -292 Neutralization, 271 -276 Optimum angle of radiation, 419
ariable -efficiency, 281 beam tetrodes, 248 Orientation of crystal cuts, 231
Modulator check, 271 Oscillator, 229
balanced, 336 -337, 343 coil, 246-247 AFSK, 329
cathode- follower, 281 -287 grid, 240 blocking, 190
circuit, SSB, Jr., 352 interelecerode capacitance. 121 -122 blocking, transistor, 111
class -B, practical aspects of, 142 procedure, feedback amplifier, 274 -276 bridge -T, 192
840
www.americanradiohistory.com
Clapp, 232 plate feed, 267 hot- cathode, 131
classes, 230 resistance, 28 voltage- divider, 132
code -practice, 20, 22 resonance, 63, 66 Phase modulation, 316 -319
Colpiffs, 231 resonant circuit, 63 Phase shift
crystal, 234 -242 RLC circuits, 60-62 correction, 298 -301
electron- coupled, 231, 232 tube circuits, 268 oscillator, 192
Fraeklin, 233, 234 Parameters network, low-Q, 344
Hartley, 189, 231 operating, 137 Phase shift and gain in feedback
magnetron, 93 -93 vacuum -tube, 120 -122 amplifiers, 143 -146
Miller, 238 symbols for, 120 -121 Phasing
negative -grid, 230 Parasitic audio -frequency, 344 -341
phase -shift, 192 arrays, feed systems for, 316 -323 method for SSB generation, 338
Pierre harmonic, 237 -239 delta match and T- match, 120 -321 radio-frequency, 344
relaxation, 117 -118, 188 driven element with center feed, 321 system for SSE generation, 342 -346
reflex kystron, 92 folded-element match calculations, Phosphors, CRT, 99
resistance -capacitance, 191 -192 319 -320 Photocells, 81, 82
self- controlled, 229 -234 gamma match, 321 -322 Photoelectric emission, 73 -77
self- starting, transistor, 333 matching stub, 322 -323 Pi- coupled final amplifiers, components,
stability, 234 rotary -link coupling, 321 239
tone, 610 -631 element, length of, 313 -316 Pierce oscillator, 238
T.N.T., 231 gain measurement, 373 crystal oscillator circuit, 237 -231
T.P.T.G., 231 resonances, 366 -367 harmonic circuit, 237 -239
transistor, 116 -111 suppression, low- frequency, 368 Piezoelectric crystal, 209
transition, 233 suppression, tetrodes, 372 Piezoelectric effect, 234 -236
triode -overtone, 240 -242 Parasitic oscillations Pi -L network, 261 -262
two -cavity klystron, 92 beam tetrodes, 370 -372 harmonic attenuation, 261
two -terminal, 233 checking, 372 -373 Pi- network, 260
Wien -bridge, 191 in r -f amplifiers, 367 -369 amplifiers, tetrodo, 659 -662
beat- frequency, 212 -213 in feedback amplifiers, 274 grounded -screen, 661 -662
Oscillator circuit low -frequency, 361 -368 inductively tuned tank circuit, 662
pentode, 238 -240 triode, 369 -370 component chart, 260
Pierre, 237 Passband, i -f, 208 coupling, 446
transistor, 117 Passband, mechanical filter, 341 matching, 219 -260
Oscilloscope, 169 -114 Passive audio phase -shift network, 343 output capacitor, peak voltage
accelerating potential, 173 Psttern, carrier -wave, ISO rating, 260
block diagram, 169 Patterns, trapezoidal, oscilloscope, 110 Q, 260
deflection polarity switch, 170 Peak Planar triode, 93
display of waveforms, 173 -173 a-c v.t.v.m., 741 Plate
linearity tracer, 183 -184 envelope power, 333 dissipation, symbol for, 120
Lissajous figures, 171 -171 fundamental plate current, symbol for, efficiency, amplifier, calculating, 133 -116
determination of phase angle, 171 120 symbol for, 121
oscilloscope settings, 173 -176 grid excitation voltage, symbol for, 120 feed, parallel, 267
phase-difference patterns, 176 -178 inverse voltage, 102 feed, series, 267
monitoring transmitter performance noise li , 216 -217 potential, instantaneous, symbol for, 120
with, 178 -179 plate voltage, symbol for, 120 power input, symbol for, 120
modulated -wave pattern, 179 voltage measurement circuit, 749 power output, symbol for, 120
trapezoidal pattern, 179 voltage rating, output capacitor, 260 resistance, 82 -84, 91
waveforms, 171 -179 Peaked wave, 69 symbol for, 120
power supply, 173 Pentagrid return amplifier, 122 -123
receiver i -f alignment, 179 -182 converter, 90 return r -f power amplifier, 161 -162
single- sideband applications, 112 -184 mixer circuit for SSB, 346 spacing, breakdown ratings, 237
sweep amplifier, 173 mixer as a product detector, 227 supply voltage, d -c, symbol for, 120
sweep generator, 172 -173 Pentode, 87 -89 voltage, minimum instantaneous,
sweep -speed multiplier, 172 amplifiers, single- ended, 133 -137 symbol for, 120
time-base circuitry, 171 -172 characteristic curves, 18 voltage, peak, symbol for, 120
blanking, 171 -172 coefficients, 91 voltage, static, symbol for, 120
sweep- trigger, 171 crystal oscillator circuits, 231 -240 voltage swing, class-C, 234
trapezoidal patterns, 180 current flow, 90, 91 Plate-and -screen modulation, 292 -293
vertical amplifier, 170 -171 mixers and converters, 90 -92 Plate -cathode capacitance, symbol for, 120
Outlet strips, 394 plate resistance, 91 Plate-circuit
Out -of -phase spacing, stacked dipole, power amplifier, operating considerations, r -f amplifiera, 149 -130
411 -482 characteristics of, 133 -137 coupling, r -f amplifiers, 149 -130
Outtut capacitance, symbol for, 120 RC- coupled amplifier, equivalent efficiency, 233
Output characteristics of junction circuits for, 126 impedance matching, 143
transistors 109 r -f amplifier, 201 mismatch, 143
Ovens, crystal, 236 stages, RC- coupled, 123 -126 Plate current, e8
Overloading, TV -set, 374 -377 PEP, 333 angle, frequency multiplier, 231 -232
Overmodulation, 386 linear amplifier, 682 -692 ge, characteristics of 813 tube, 117
Overtone oscillators, triode, 240 -242 2 -KW supply, 729 -731 average, symbol for, 120
Overtone -cut crystals, 237 Percentage, modulation, 277 flow, one -half angle of, symbol for, 121
Oxide- coated filaments, 80 Percentage modulation, f -m, 310 maximum instantaneous, symbol for, 120
Perforator, tape, 330 peak fundamental, symbol for, 120
Permeability, 40, 41 Plate-current saturation, 81 -83
P Phantoms, 386, 387 static, symbol for, 120
Phase Plate-modulated class -C amplifier, 243
Parallel angle, 32, fl Plate modulation, 289 -290
bias feed, 267 distortion, 123 -124 average power, 289
circuit, impedance curve, 63 lagging, 32, 33 class -B, 290, 291 -293
feed, 268 leading, f2, 33 of beam tetrodo, 293
inductance for bass suppression, 304 Phase inverter of screen -grid tube, 293
inductors, 42 cathode-coupled, 130 -132 Heising, 290 -291
operation, 268 floating paraphase, 131 systems, 289 -290
841
Plumber's delight, rotatable arrays, 85-watt, transistor, 157 Quadrupler, voltage, 719 -720
527-328 IVS, 723 -728 Quarter -wave matching transformers, 419
Pm, 316 -319 1 -KW, 743 Quartz crystal, equivalent circuit, 235
PN junction diode, 102 -101 1 -KW IVS, 728 -729 Quartz crystal oscillators, 234-242
PNP transistor, 103 oscilloscope, 173 Quench methods, 196 -197
voltage amplifiers, Ill PEP 2 -KW, 729 -731
Point -contact transistor, 103 requirements, 707.713
construction of, 106 current- rating considerations, 708 -709 R
Point- contact transistor and vacuum -tube filter- system resonance, 711 -713
characteristics, comparison of, 108 resistance -capacitance filters, 711 Radian notation, 49
Polarity ripple considerations, 709 -711 Radiating elements, rotatable arrays,
inversion, 83 -86 voltage regulation, 709 326 -327
reversal, grid and plate, 87 SSB transceiver, 733 Radiation, 410 -431
switch, deflection, 170 special, 736 -739 angle, effect of feed system on, 494.491
Polarization, 411 regulated supplies-VP. tubes, 736 -737 angle of, vhf and uhf antennas, 494
antenna, 412 shunt- regulated bias supply, 738 directivity, 411
speech waveform, 280 voltage -regulated, 737 -738 from an antenna, 410 -411
vhf and uhf antennas, 495-496 33 -watt, transistor, 116 -137 harmonic, 377 -379
Polarized antennas, horizontally, transformerless, 719 patterns, vertical, 420 -421
vhf and uhf, 496 -497 transistor, mobile unit, 113 -316 resistance, 411 -412
Polar notation, 54, 33 operation, 113 -134 resistance and feed -point impedance,
Polyphase rectification circuits, 717 rating, 114 -111 411 -418
P (positive) type germanium, !Of self -starting oscillator, 111 resistance, effect of height, 46
Positive and negative clamping circuits, transformer, 111 -116 Radiator cross section, vhf and uhf
187 Power transformer, 113 -116 antennas, 491
Positive -grid characteristics of triode, 83 Power transistor circuit, audio, Radiators, long -wire, 476 -477
Post-modulator filter, 303 class-A, 113 Radio -frequency
Potential, accelerating, oscilloscope, 173 Practical filters, i -f circuits, 209 amplifiers, 242 -268
Power advantage, SSB over a -m, 332 -333 Practical peak noise limiter circuits, chokes, 267 -268
Power 218 -219 energy, generation of, 229-265
control, 393-397 Preamplifier, vhf low -noise transistor, 116 phasing, 344
energy, and work, 33 Pre -emphasis, 323 power amplifiers, I30 -168
gain of a transistor, 108 Prefixes, 47 Class -B, 116 -I19
grid -driving, symbol for, 120 Premium tubes, 98 Class -AB1, 163 -161
input, plate, symbol for, 120 Product detector, 227 -228, 336-317 vacuum -tube amplifiers, 147 -168
in resistive circuits, 32, 33 and beat oscillator, 228 Radio receiver fundamentals, 194 -228
klystron, 93, 94 double-triode, 228 Radiotelegraphy reception, 195
measurement, 732 -734 triple- triode, 228 Radiotelephony demodulation, 194
peak envelope, 333 Product, gain -bandwidth, 107 Radioteletype systems, 321-330
rating, transistor, 334 -333 Propagation, 410-435 Ratio detector, 321 -322
relations in speech waveforms, 291 of radio waves, 421 -422 Ratio, signal -to- noise, 112
sources, mobile, 133 -337 tropospheric, 423 amplifier general characteristics, 127
systems, 392 -397 Protective circuits for tetrodes, 264 -261 amplifier response, 127
Power amplifier Protective interlocks, 401 coupled pentode stages, 121 -126
design, 652-654 Pulsating, d.c., 11 coupled triode stages, 123
circuits, 633.634 Push- button transmitter control circuit, differentiator, 69 -71
tubes, 652-653 399 integrator, 69, 71
pentode, operating characteristics of, Push -pull RC and RL time constant, 43
135-137 audio amplifiers, 137 -138 RC and RL transients, 43-45
push -pull class -A triode, 138 circuits, 137 Reactance, 11 -33
r -f cathode -follower, 161 -162 class-A triode power amplifier, 138 calculations, 824 -826
r -f plate -return, 161 -I62 frequency tripler, 212, 253 capacitive, 12
triode, operation characteristics of, link coupling, 267 feed -point impedance, 411
111 -113 neutralization, 241 -246 frequency chart, audio, 821
cathode -follower, 143 -143 operation, 268 frequency chart, radio, 822
class AB, r -f, 163-165 stages, tank circuits, 216 inductive, 52
class -B audio -frequency, 138 -143 tetrode amplifiers, 656-659 leakage, 73
class-B radio -frequency, 136 -139 transformer intent age coupling, 128 tank -capacitor, 235
high- frequency, 632 -697 triode amplifiers, 614 -616 tank -inductor, 211
radio -frequency, 130 -168 circuit layout, 611 tube linearity checker, 315
r -f class -C, 130 -136 construction, 611 -616 tube modulators, 312 -316
Power line excitation requirements, 611 Receiver
filters, 216 filament supply, 614 a- c /d -c, 387
pickup, 388, 389 grid bias, 634 adjustment, 224.226
wavetrap, 389 grid circuit, 654-655 alignment,
Power output plate feed, 614 front -end, 226
formula, 113 -117 i -f, 225
tube circuits, 268
plate, symbol for, 120 Push -push frequency doubler, 212 superheterodyne, 221
Pou rrntvt, 397 -398 TRF receiver, 223
Power supply, 707 -749 BPO, 226
components, 733 -736 0 i -f with crystal filter, 225-226
bleeder resistors, 735 instruments, 224 -226
filter capacitors, 733 -731 Q circuts, uhf, 223
filter choke coils, 733 amplifier tank circuit, 131 design, f-m, 322 -324
circuits, standard, 714 -720 circuit, 64 -66 deluxe NBA, 604-619
mercury -vapor tubes, 718 effect of loading, 217 alignment, 616 -619
polyphase, 717 effective, 213 circuit, 606 -608
single- phase, 716 -717 parallel- resonant circuit, 63 -68 construction, 609 -612
transformerless, 719 pi- network, 260 r -f circuits, 613-616
voltage- doubler, 719 tank- circuit, 213 -218 wiring, 612 -613
voltage- quadrupler, 719 -720 Q- multiplier, 226 -227 fundamentals, radio, 194-228
design, 739 -743 Q- section feed system, 439 -460 i -f alignment with oscilloscope, 179 -112
842
www.americanradiohistory.com
rectification, 388 harmonic, 414 Sawtooth wave, 69
SSB, 356 parasitic, 366-367 Scattered reflections, 428 -429
.uperheterotlyne, 197 -200 Resonant S. hem a tic
advantages of, 198 cavities, uhf, 221 -222 Antennascope, 765
arithmetical selectivity, 198 circuits, 62 -68 bridge for balanced lines, 765
choice of intermediate frequency, 198 frequency, 62, 63 capacitance meter, 717
converter stage, 198.200 lines, 433 -434 Hrathkif Ins-13, 751
diode mixers, 200 Response curve, i -f, 181 high -voltage peak voltmeter, 748
principle of operation, 197 -198 Response, RC amplifier, 12' kilowat dummy load, 754
superregenerative, 191 -197 Reverse bias, 102 low -range ohmmeter, 747
wiring, trans- receiver, 550 -311 R -f monitor scope, 769
Receivers and transceivers, 518 -619 amplifier 100 -kHz frequency spotter, 719
Receiving accessories, 226 -228 grid- circuit considerations, 158 -159 rcflrctometer, 761
product detector, 227.228 grounded -grid, 159 -161, 250 r -f probe, 712
Q- multiplier, 226 -227 neutralization, 244 -210 SWR indicator, 760
Receiving converter, RTTY, 326 -328 parasitic oscillation, 367 -369 transistor checker, 771
Reception pentode, 201 v.t.v.m., 751
f -m, 197, 319 -324 gain equations for, Ill SCR (silicon controlled rectifier)
radiotelegraphy, 195 plate- circuit considerations, 149 -I50 devices, 104
RTTY, 326.328 plate -circuit coupling, 149 -150 Screen, 86 -87
SSB signals, 353-357 screen -circuit considerations, 119 circuit considerations, r -f amplifiers, 119
Rectified a.c., 50, 51 tuned cathode circuit, 166 -168 lead inductance, cancelling, 249.250
Rectifiers choke, 267 -268, 364 materials, CRT, 97
crystal, 224 equivalent, 363 supply, dropping -resistor, 264
selenium and silicon, 720 -721 frequency and impedance supply, separate, 264 -265
silicon controlled (SCR), 104 characteristic, 361 supply, series, 264
tubes, mercury- vapor, 718 circuitry, 114 wattage -limiting resistor, 265
Rectifier circuit, 713 -714 feedback, 269 -276 Sc reco beam for 2 meters, 109 -5II
843
www.americanradiohistory.com
Sideband transceiver, 200 -watt 3 -band, in-phase, stacked dipole, 491 Standard power- supply circuits, 714 -720
583 -593 out -of- phase, stacked dipole, 481 -482 mercury -vapor tubes, 718
circuit description, 583, 585 Spectrum, frequency, 46, 47 polyphase, 717
construction, 585 -587 Speech single -phase, 716 -717
testing and alignment, 587 -590 amplifier, clipper, 142 transformerless, 719
transmitter alignment, 590 -593 and a -m equipment, 698 -706 voltage- doubler, 719
Sidebands, 277 -278 clipping, 297 voltage -quadrupler, 719 -720
relative proportion, 279 effects of, 140 Standing wave, 433, 455 -456
Side frequencies in the modulator stage, 287 indicator, bridge -type, 760 -762
a -m, 309 low -level, 700 on a resonant antenna, 411
f -m, 310 waveform ratio and reflection coefficient
Signal, error, 193 amplitude, 280 relation, 761
Signal- frequency tuned circuits, 203 -206 dissymmetry, 280 Standing waves on a transmission line,
bandspread tuning, 205 polarization, 280 434
frequency range selection, 205 power relations, 291 Start space, 325 -326
impedance, and Q, 204 Splatter, 280, 297 Static
input resistance, 204 suppressor body, 543
stray circuit capacitance, 206 action, 301 grid current, symbol for, 120
superheterodyne tracking, 204 -205 high -level, 300, 700 -701 grid voltage, symbol for, 120
Significant figures, 826 Split- stator plate neutralization, 245 interelectrode capacitance, 121
Signal, safety, 400 Sporadic -E layer, 427 plate current, symbol for, 120
Signal- strength indicators, 214 -215 Spurious missions, 374, 387 wheel, 542
Signal- to-noise ratio, 182 Square wave, 68, 69 Station and Operator Licenses, 12, 13
Silicon SSB envelope, 335 Steering diodes, 118
crystal noise generator, 767 -769 test, 71 -72 Stop pulse, 325 -326
intrinsic resistivity of, 101 SSB Stratospheric reflection, 424 -425
supplies for SSB, 724-728 advantage with selective fading, 335 Stray circuit capacitance, 206
Silicon and germanium atomic structure, applications, commercial, 331 -332 Stub -fed Zepp -type radiator, 437
100 -101 bandwidth, 333 Stubs, matching, 457
Silicon and selenium rectifiers, 720 -723 distortion products, 349 Substitution method, 755
Silicon controlled rectifier (SCR), 104 envelope Sterba curtain array, 484
Sine wave, 48 four -tone, 335 Sunspot cycle, 427
Single -band SSB transceiver, 572 -583 square -wave, 335 Sun noise recording, 604
alignment, 580 -581 three -tone, 334 Superheterodyne
circuitry, 573 -571 exciter, 175 -watt, 624 -635 alignment, 225
final adjustment and neutralization, 581 circuit, 624 -626 receivers, 197 -200
layout and assembly, 575 -578 dial assembly, 627 -629 advantages of, 198
transmit operation, 581 -583 layout, 626 -627 arithmetical selectivity, 198
wiring, 578 -580 main bandswitch assembly, 627 choice of intermediate frequency, 198
Single -ended wiring and testing, 629 converter stage, 198 -200
amplifier neutralization circuits, 244 frequency -conversion systems, 346 -350 diode mixers, 200
pentode amplifiers, 135 -137 generation principle of operation, 197 -198
tetrodes, neutralizing, 249 filter method, 338 tracking, 204 -203
triode amplifiers, 133 -135 phasing method, 338 Superregenerative receivers, 195 -197
Single -sideband phasing system, 342 -346 Supports, antenna, 461 -464
applications, oscilloscope, 182-184 Jr., 352 -353 A -frame mast, 461
receivers and adapters, 356 modulator circuit, 352 guy wires, 462
signals, deviation, 332 pentagrid mixer circuit, 346 insulation, 463 -464
transmission, 331 -357 signal, nature, 334 -335 painting, 463
Single -phase rectification circuits, 716 -717 signals trees as supports, 463
Single -swing blocking oscillator, 190 generation, 338 -346 TV masts, 461 -462
Single -tone observations, oscilloscope, 183 reception, 355 -357 wire, 463
Single -wire antenna tuner, 469 -471 silicon supplies for, 724 -728 Suppression
bridge calibration, 470 stability requirements, 335 high -angle radiation, 420 -421
tuner adjustments, 471 transceiver, single -band, 572.583 noise, 216 -220
tuner construction, 470 alignment, 580 -581 alternative limiter circuit, 219 -220
Single- wire-fed antenna, 452 -453 circuitry, 573 -575 limiter
Six -band trans -receiver, 544 -513 final adjustment and neutralization, audio, 218
adjustment, 551 -553 581 Bishop, 217 -218
circuitry, 545 layout and assembly, 175 -578 full -wave, 220
construction, 547 -550 transmit operation, 581 -583 Lamb, 217
controls, 545 -547 wiring, 578 -580 peak, 216 -217
receiver wiring, 550-551 transceiver supply, 733 practical peak, 218 -219
Six -meter TVI filter, 382 transmitter-exciter, deluxe HBT-200, TNS, 220
Six- shooter broadside array, 487 635 -648 power line filters, 216
Skin effect, 64 circuit, 638 noise, vehicular, 541 -544
Skip distance, 426 -427 construction, 641 -642 body static, 543
Skywave, 422 transmitter generator "whine," 542 -343
Slide -back v.t.v.m., 747 alignment, 644 -648 ignition noise, 341 -342
Slope detection of f -m signal, 319 wiring, 643 -644 locating noise sources, 544
Slotted line, 759 -760 vfo assembly, 642 -643 voltage regulator "hash," 542
Space, 325 Stability requirements, SSB, 335 wheel static, 142
Space charge, 82, 87 Stacked Suppression and metering circuits, 670
effect, 80, 82 broadside arrays, 523 -525 Suppressor, 87
limited current, 80 dipole arrays, 481 -483 Suppressor -grid modulation, 288 -289
limited emission, 82, 83 broadside arrays, 483 -484 Suppressor, splatter, high- level, 700 -701
Space -conserving antennas, 444 -446 collinear, 482 -483 Surface wave, 422
broadband dipole system, 446 double extended zepp, 483 Susceptance, 61
twin -lead Marconi, 446 in -phase spacing, 481 Sweep
Space wave, 422 out-of -phase spacing, 481 -482 amplifier, oscilloscope, 173
Spacd half -wave antennas, 483 spaced half -wave, 483 gneerator, oscilloscope, 172 -173
Spacing yagi arrays, 517 Sweep -speed multiplier oscilloscope, 172
844
www.americanradiohistory.com
Sweep -trigger circuit, oscilloscope, 171 T.N.T. oscillator, 231 voltage gain, 108
Switching, crystal, 240 Tone oscillator, 650 -651 Transistorized
SWR bridge, 472 Tools, 773 -775 capacitance meter, 757 -758
SWR indicator, twin -lamp, 764 T.P.T.G. oscillator, 231 exciter, 117
Synchros and selsynr, 531 Trace, CRT, 97 mobile converter, 133 -531
Tracer, linearity, oscilloscope, 113 -184 Transit time, effect, 91, 96
Traces, linearity, 184 Transition
T Tracking, superheterodyne, 204 -205 oscillator, 233
Transceivers region, 107
Tank -capacitor reactance, 255 and receivers, 558 -619 Transmission -line, 410 -435
Tank circuit, 65 -68 power supply, 558 -733 characteristics, 431
amplifier, Q of, 155 200 -watt, 3 -band sideband, 583 -193 circuits, uhf, 220
arrangements, 211 circuit description, 583 -181 filters, high -pass, 375
capacitance, 213 construction, 585 -587 impedance, 430
efficiency, 67 -68 testing and alignment, 187 -190 matching system, Collie's, 460 -461
flywheel effect, 67 transmitter alignment, 590 -593 matching to the antenna, 128 -529
push -pull stages, 256 Transconductance, 82 -84 nonresonant, 430 -433
Q, 253 -258 conversion, symbol for, 120 ribbon and tubular, 432 -433
Tank -inductor reactance, 255 cutoff frequency, 107 standing waves, 434
Tank Q, capacitance charts, 255, 216 grid -plate, 89, 91 vhf and uhf, 494
Tape perforator, 330 symbol for, 120 Transmitters
T -D unit, 329 Transformation and exciters, low- power, 620 -651
Teleprinter, 326 impedance, 73 control methods, 397 -398
code, 325 -326 ratio, 72 design, 358 -373
Teleprinting, 325 Transformer distributor unit, 329
Television interference, 374 -391 calculation, modulation, 140.141, 291 -292 exciter, SSB, Deluxe HBT -200, 635 -648
Temperature coupling, 128 circuit, 638
coefficient, resistor, 318 -359 i -f, 206 -207 construction, 641 -642
filament, 78, 79 matching, quarter -wave, 419 transmitter
"Ten -A" exciter, 353, 314 power, 551 -556 alignment, 644 -648
10 -meter mobile antennas, 533 -519 qu -wave, doublets with, wiring, 64) -644
439
Termax- Woodyard types, 71 -74 vfo assembly, 642-643
grid -modulation system, 219 Transformerless power supplies, 719 fundamental, tuned traps, 375
modulated amplifiers, 294 Transients, 68, 69 keying and control, 392 -409
Terminal unit, 326-327 RC and RL, 43 -41 loading problem, 464
Test Transistor output coupling adjustment, 461 -466
equipment, electronic, 744 -772 action, 101 -108 performance, monitoring with
for neutralization, 273 and vacuum -tube configurations, oscilloscope, 178 -179
probes, 712 comparison of, 110 modulated -wave pattern, 179
setup for para,itics, 372 audio amplifier, two -stage, 112 trapezoidal pattern, 179
square wave, 71, 72 bias configurations, 111 waveforms, 178 -179
Tetrode, 88, 89 blocking oscillator, 118 Transport, charge, 106
amplifiers characteristics, 108 -110 Trans -Receiver, six -band, 144 -553
linearity curve of, 164 checker, 770 -772 adjustment, 151.153
push -pull, 616 -619 circuitry, 110 -118 circuitry, 545
cathode modulation, 294 classifications, 107 -108 construction, 547 -150
characteristic curves, 88 configurations, 109 -110 controls, 145 -147
coefficients, 19, 91 electrical symbols, 106 receiver wiring, 110-111
current flow, 88 -89, 90, 91 epitaxial planar, 111 Trap
modulators, 698 -699 equivalent circuit, 109 a-c line, 388
zero -bias, 706 field-effect, 118 -119 high -Q isolating, S32
pi- network amplifiers, 659 -662 i -f amplifiers, 114 isolating, three -band beams, 531 -132
grounded -screen, 661 -662 Transistor Trapezoidal patterns, oscilloscope, 180
inductively tuned tank circuit, 662 junction, output characteristics cf, Traveling -wave tube, 91, 94, 96
plate resistance, 91 105, 109 Trarss discriminator, 320
protective circuits, 264 -265 keyer, 648 -611 TRF receiver alignment, 225
screening, 371 circuit, 648 -650 Triangle, impedance, 11
tubes, 86 -87 construction, 611 Tribander linear amplifier, 692 -697
Thermionic emission, 71, 77 dash flip -flop, 610 circuit, 692 -696
35 -watt tone oscillator, 610 -611 transmitter adjustment and tuning, 697
modulator, 114 low- frequency equivalent circuit, 109 transmitter control circuits and power
transistor power supply, 556 -117 NPN, 105 supply, 696 -697
Thoriated- tungsten filament, 71 -80 operation, 553 -154 Trigger amplifier, oscilloscope, 171
Three -band beams, rotary, 531 -532 oscillators, 116 -118 Trigonometry, 807 -812
Three-element array, rotary beams, 514 -516 PNP, 101 Trimmer, antenna, auxiliary, 537 -538
element spacing 111 point -contact, 105 -106 Triode
length of parasitic element, 111 -516 power gain, 108 amplifiers
materials for elements, 511 power rating, 154 -511 push -pull, 614 -656
more than three elements, 116 power supply circuit layout, 611
stacking of yagi arrays, 516 85 -watt, 117 construction, 651 -616
Three -element beam antenna, 105 -106 31 -watt, 516 -557 excitation requirements, 655
3/4 -wave folded doublet, 447 -448 mobile unit, 553 -156 filament supply, 614
Three -halve, power law, 82, 83 operation, 553 -114 grid bias, 614
Three -tone SSB envelope, 334 rating, 554 -555 grid circuit, 654 -615
Thyratron, 91, 100 self -starting oscillators, 111 plate feed, 614
Tiltable array, vhf, 106 -108 transformer, 115 -556 single-ended, 133 -135
Time -base circuitry, oscilloscope, 171 -172 preamplifier, vhf low- noise, 116 cathode -driven high-A, 166
blanking, 171 -172 resistance gain, 108 characteristic curves, 83 -81
sweep-trigger, 171 self -bias system, Ill current flow, 82, 14
Time constants, RC and RL, 43 SSB filter exciter, 313 -311 grid action, 81, 83
T match, 417 types of, 101 -106 high -a, grid -leak bias for, 127
TNS limiter, 220 unipolar, 118 -119 load line, 81, 86
mixer, 90, 92, 200 TV tube combinations, 690 -691
negative -grid characteristics, 83, 81 blocking, 376 tuning and adjustments, 691 -692
overtone oscillators, 240 -242 channels, vhf, 378
parantic oscillations, 369 -370 coverage, high -band, helical beam
planar, 95 antenna, 502 V
positive -grid characteristics, 85 masts, 461 -462
power amplifier, operation set overloading, 374 -377 Vacuum capacitors, 362
characteristics of, 133 -135 TVI, 374 -391 Vacuum -diode modulator circuits, 339
stages, RC- coupled, 125 filter, six -meter 382 Vacuum -tube
Triple- triode product detector, 228 types, 374 amplifiers, I20 -148
Triplex beam, 491 TVI -proof inclosures, 777.778 radio- frequency, 147 -168
Tripling and doubling circuits, viractor, Twin -lamp SWR indicator, 764 characteristics, 108
103 Twin -lead circuits, special, 185 -193
Tropospheric propagation, 423 folded dipole, 439 classification, 98, 100, 101
Tubes Marconi antenna, 446 constants, 121
at very -high frequencies, 91 -96 Twin noise squelch limiter, 226 keyers, 403
beam -deflection, 357 Twin- triode mixer, 346 -347 parameters, 120 -122
beam -power, 87.90 Two -band Marconi antenna, 448-449 principles, 75 -99
cathode -ray, 94 -99 Two -cavity klystron oscillator, 92 polarity inversion, 85 -86
characteristics, application, 84 -68 Two -element beam yagi type, 512 -514 voltmeter, 748 -752
circuits, 268 200 -watt -band sideband transceiver,
3 Vacuum -tube and transistor configurations,
clamp, 264 583 -593 comparison of, 110
controlled- warmup, 99, 102 circuit -description, 183.585 V antenna, 477 -479
electron -ray, 99, 101, 102 construction, 581 -587 Varactors, 103 -104
gas, 97 -500 testing and alignment, 587 -590 Variable
input loading, 149 transmitter alignment, 590 -593 air capacitors, 361-362
"magic -eye," 99, 101 2-KW PEP supply, 729 -731 vacuum capacitors, 362
mercury- vapor, 98, 100 Two -meter converter, nuvistor, 567 -572 Variable-efficiency modulation, 281
microwave, 93 -96 adjustments, 570 -571 Variable -link coupling 466
mixer and converter, 89 -90, 92 antioverload modification, 571 -572 Variable -ratio autotransformers, 397 -398
pentode, 87 circuit description, 567 -568 Variable -selectivity filters, 209 -210
premium, 98 construction, 168 -569 Variable -a tubes in r -f stages, 150
remote -cutoff, 87, 89 Two- terminal oscillator circuits, 233 Variac, 397-398
screen -grid, 86.87 Two -tone SSB envelope, vector Variation of Q with frequency, 64, 65
special microwave, 91 -94 representation, 334 Varicaps, 102
tetrode, 86 -87 Two -wire open line, 430 -433 Vector, 48 -49, 812 -815
thyratron, 98 diagram, discriminator, 321
traveling -wave, 94, 96 relationship of feedback voltage, 274
types, 76 V representation of two -tone SSB
vhf, 223 -224 envelope, 334
voltage -regulator, 98 -100 Uhf rotating, 48, 49
Tuned electron tubes, 90 -91 Vehicular noise suppression, 541 -544
cathode circuit, r -f amplifier, 166 -168 receiver design, 220 -224 body static, 143
circuits butterfly circuit, 222.223 generator "whine," 542 -143
coupled, 349 coupling into lines and coaxial ignition noise, 541 -842
i -f, 206 -213 circuits, 221 locating noise sources, 544
beat -frequency oscillators, 212 -213 crystal rectifiers, 224 voltage regulator hash, 542
crystal filters, 209, 211 resonant cavities, 221 -222 wheel static, 542
mechanical filter, 215 -212 transmission line circuits, 220 Velocity modulation, 91
Miller effect, 208 tuning short lines, 220 Vertical
practical filters, 209 vhf tubes, 223 -224 amplifier, oscilloscope, 170 -171
rejection notch, 210 -211 Uhf and vhf antenna, 493 -311 antenna, half -wave, 441
shape factor, 207 -208 corner- reflector and horn -type, 502 -503 antenna, multiband, 453 -455
transformers, 206 -207 helical beam, 500 -502 directivity, 412 -419
variable-selectivity filters, 209 -210 horizontally polarized, 496 -497 radiation patterns, 420-421
selective, 347 -349 requirements, 493 Vertical plane directional characteristics,
signal- frequency, 203 -206 changeover, 494 418
bandspread tuning, 205 insulation, 495 Vertical- polarized antennas, vhf and uhf,
frequency range selection, 205 polarization, 495 -496 497 -498
impedance and Q, 204 radiation angle, 494 -495 double skeleton cone, 498
input resistance, 204 radiator cross section, 495 nondirectional, 498
stray circuit capacitance, 206 transmission lines, 494 Vfo
superheterodyne tracking, 204 -205 vertical -polarized, 497 -498 40- meter, 579
doublet antenna, 437 -439 double skeleton cone, 498 transmitter controls, 234
lines, 433 -434 nondirectional, 498 Vhf
plate crystal oscillator, 238 Unidirectional bands, If
traps, 375 driven arrays, 523 -325 beam antenna, mul tielement, 505 -511
Tuner for center -fed antenna feed methods, 523.325 eight -element tiltable array, 506 -508
systems, 471 -473 stacked broadside arrays, 523 -521 long yagi, 511
Tungsten filament, 78, 79 parasitic end -fire arrays, 112 -514 screen beam formeters, 109.511
2
Tuning Unipolar transistor, I18 -119 three -element beam, f01-506
array, 528 -130 Unit low -noise transistor preamplifier, 116
matching to the antenna transmission of capacitance, 34, 31 par si tics, elimination, 369 -371
line, 528 -529 of inductance, 42 tubes, 223 -224
raising and lowering, 530 Unity coupling, 266 TV channel frequencies, 378
bandspread, 205 Unloaded Q, 67 Vhf and uhf antenna, 493 -III
capacitor air gap, 258 Upper sideband, 332 corner -reflector and horn -type, 502 -503
crystal oscillator, 239, 240 Utility 2- kilowatt linear amplifier helical beam, 500 -502
grid -bias modulated stage, 283 -284 (U -2), 682 -692 horizontally polarized, 496 -497
short lines, uhf, 220 circuit, 682 -681 horizontal rhombic, 503 -505
TU (terminal unit), 326 -327 construction, 685 -688 requirements, 493 -496
Tunnel diode, 102 -103 external anode tetrodes, 688 -690 changeover, 494
www.americanradiohistory.com
insulation, 491 Voltage regulator Wheel static, 142
polarization, 491 -496 hash, 142 Whip antenna, 139
radiation angle, 494 -493 tube, 91 -100 "Whine," generator, 142-143
radiator cross section, 491 Voltage -variable junction diode, 102 Wien -bridge oscillator, 191
transmission lines, 494 Voltohmmeters, 741 -746 Wire, antenna, 463
vertical -polarized, 497 -491 Volume control, automatic, 214 Workshop practice, 773 -783
double skeleton cone, 491 VOX circuitry, 401 -409
nondirectional, 491 V.t.v.m.
Video amplifier circuit, compensated, 127 a -c, 741 X
Video -frequency amplifiers, 127 -128 bridge -type, 750
Voltage Heatbleit IM -13, 750 -712 X- array, 414
amplifier, cascaded, 127 modern, 750 -712
amplifiers, PNP transistor, III slide -back, 747
avalanche, 102 T
decay, 102
feed, 411 w Yagi, long, vhf antenna, 111
gain of a transistor, 108 Yagi arrays, stacking of, 116-117
Water -pipe grounds, 443 -444 Yagi -type two -element beam, 112 -$14
gain per stage, audio amplifiers, 123
Wave composition, 61, 69
peak- inverse, 102
Waveform
rating of capacitors in series, 3$
base -collector, 313 z
regulation, 709
oscilloscope, 173 -171
quadrupler, 719 -720
Wave interference with height, 423 Zener
Voltage and current, 744 -747
Wavemeter, crystal-diode, 379 control region, 104
Voltage divider
Waves diode, 104
a-c, 61 -62 knee, 104
nonsinusoidal, 61, 69
calculations, 30 -31 standing, 413 -416 Zepp
phase inverter, 132 Wavetrap antenna system, 437
Voltage -equalizing resistors, 31 circuit, 381 center -fed, 437
Voltage -doubler circuits, 719 power line, 389 double -extended, 483
Voltage -regulated power supplies, 737 -731 Wheatstone Bridge circuits, 711 Zero-bias tetrode modulators, 706
847
www.americanradiohistory.com
www.americanradiohistory.com V