You are on page 1of 98

T 7.4.

4
Microwave
Propagation
in Waveguides
by Prof. H. J. Chaloupka

revised by Dipl.-Ing. Anton Oster
June 1998
LEYBOLD DIDACTIC GMBH
.
Leyboldstrasse 1
.
D-50354 Hrth
.
Phone (02233) 604-0
.
Fax (02233) 604-222
.
e-mail: info@leybold-didactic.de

by Leybold Didactic GmbH Printed in the Federal Republic of Germany
Technical alterations reserved
The sensitive electronics of the equipment contained in the present experiment litera-
ture can be impaired due to the discharge of static electricity. Consequently, electro-
static build up should be avoided (particularly by utilizing appropriate rooms) or
eliminated by discharging (e.g. at the panel frames or similar).
3
MTS 7.4.4
Contents
Table of contents
Equipment overview ................................................................................................................... 4
Symbols and abbreviations ......................................................................................................... 5
Preface .......................................................................................................................................... 7
List of experiments performed in this training unit
Ex1 The Gunn oscillator ......................................................................................................... 11
Ex2 The direct modulation of the Gunn oscillator ............................................................... 17
Ex3 The selective measurement amplifier ............................................................................ 19
Ex4 Attenuators ....................................................................................................................... 23
Ex5 The slotted measuring line.............................................................................................. 27
Ex6 The complex reflection coefficient ................................................................................ 37
Ex7 Matching and the Smith chart ........................................................................................ 43
Ex8 Reflection of a single-slot antenna ................................................................................ 53
Ex9 Measuring the permittivity.............................................................................................. 59
Ex10 The cross directional coupler ......................................................................................... 65
Ex11 The principle of the reflectometer.................................................................................. 71
Ex12 The cavity resonator ........................................................................................................ 77
Solutions ..................................................................................................................................... 81
Index ........................................................................................................................................... 97
4
MTS 7.4.4 Contents
Equipment overview
Equipment
Gunn oscillator 737 01 1 1 1 1 1 1 1 1 1 1 1 1
Gunn power supply with SWR-Meter 737 021 1 1 1 1 1 1 1 1 1 1 1 1
Coax detector 737 03 1 1 1 1 1 1 1/(2) 1/(2) 1 1 1 1
Transition waveguide/coax 737 035 1 1 1 1 1 1 1 1 1
PIN modulator 737 05 (1) (1) (1) 1* 1* 1* (1) (1) (1)
Isolator 737 06 (1) (1) (1) 1* 1* 1* (1) (1) (1)
Variable attenuator 737 09 1 1 1 1
Fixed attenuator 737 095 1 1 1
Moveable short 737 10 1
Slotted measuring line 737111 1 1 1 1 1
Waveguide 200 mm 737 12 1 1 1 1
Slide screw transformer 737 13 (1) (1)
3-Screw transformer 737 135 1 1 (1)**
Waveguide termination 737 14 1 1 2 1 2 2 2
Cross directional coupler 737 18 1 1 1 1 1
Set of 4 diaphragms with slits and holder 737 22 1 1 1 1 1 1 1 1
Waveguide propagations accessories 737 29 1 1 1 1 1 1
E-Field probe 737 35 (1)
Set of thumb screws M4 737 399 1 1 1 1 1 1 1/2** 1 1 1 2 1
Accessories
Digital storage oscilloscope 575 292 1 1 1 (1) (1) (1) (1) (1)
Coax-cables with BNC/BNC plugs, 2 m 501 022 2 2 2 2 2 2 3 3 3 2 2 2
Stand base MF 301 21 2 2 4 2 2 2 2 2 2 3 3 2
301 26 1 1 2 1 1 1 1 1 1 2 2 1
Support for waveguide components 737 15 1 1 3 2 2 2 2 2 2 3 3 2
XY-Yt-recorder 575 663 (1) (1) (1) (1) (1) (1)
() = recommended ** = when using 737 13
* = required for reproducible results ()** = alternative intermediate piece for 200 mm waveguide
Stand rod L = 25 cm, = 10 mm
M
T
S

7
.
4
.
4


E
x
p
e
r
i
m
e
n
t
s
E
x
4
:
A
t
t
e
n
u
a
t
o
r
s
E
x
1
:
T
h
e

G
u
n
n

o
s
c
i
l
l
a
t
o
r
E
x
2
:
T
h
e

d
i
r
e
c
t

m
o
d
u
l
a
t
i
o
n

o
f

t
h
e

G
u
n
n

o
s
c
i
l
l
a
t
o
r
E
x
5
:
T
h
e

s
l
o
t
t
e
d

m
e
a
s
u
r
i
n
g

l
i
n
e
E
x
6
:
T
h
e

c
o
m
p
l
e
x

r
e
f
l
e
c
t
i
o
n

c
o
e
f
f
i
c
i
e
n
t
E
x
7
:
M
a
t
c
h
i
n
g

a
n
d

t
h
e

S
m
i
t
h

c
h
a
r
t
E
x
8
:
R
e
f
l
e
c
t
i
o
n

o
f

a

s
i
n
g
l
e
-
s
l
o
t

a
n
t
e
n
n
a
E
x
3
:
T
h
e

s
e
l
e
c
t
i
v
e

m
e
a
s
u
r
e
m
e
n
t

a
m
p
l
i
f
i
e
r
E
x
9
:
M
e
a
s
u
r
i
n
g

t
h
e

p
e
r
m
i
t
t
i
v
i
t
y
E
x
1
0
:
T
h
e

c
r
o
s
s

d
i
r
e
c
t
i
o
n
a
l

c
o
u
p
l
e
r
E
x
1
1
:
T
h
e

p
r
i
n
c
i
p
l
e

o
f

t
h
e

r
e
f
l
e
c
t
o
m
e
t
e
r
E
x
1
2
:
T
h
e

c
a
v
i
t
y

r
e
s
o
n
a
t
o
r
5
MTS 7.4.4
Contents
Symbols and abbreviations
a : Display value of the selective measurement amplifier
a', s' : Geometric dimensions of the resonator cavity
a, b : Geometric dimensions of the rectangular waveguide
a
D
: Directivity
a
K
: Coupling attenuation
B : Susceptance
c : Velocity of light
C : Capacitance
DUT : Device under test
E

: Electric field vector


E : Electric field strength
E
TH
, U
TH
: Threshold field strength and Gunn voltage required to reach E
TH
f : Frequency
f
0
: Resonance frequency of the Gunn oscillator
f
c
: Cut-off frequency
f
r
: Frequency of the modulating wanted signal
G : Conductance
H : Magnetic field strength
H

: Magnetic field strength vector


RF : Radio frequency
i(t), u(t) : Current and voltage of a Schottky diode
I
Detekt
: Direct current through the Schottky diode of the detector
I
G
: Gunn current through the Gunn diode
I
S
: Saturation voltage of a Schottky diode
J : Electrical dipole
K : Constant
k : Coupling coefficient
L : Inductance
l, : Linear measure
M : Magnetic dipole
n : Transformation ratio
LF : Low frequency
P
in
: Input or feed power
P
rad
: Radiated power
P
rec
: Received power
Q
0
: Unloaded Q
r : Reflection coefficient (complex)
R : Actual resistance
s : Standing wave ratio
S
ij
: Elements of the scattering matrix (S)
T : Period duration
t : Time
u
: Mean value of a voltage
U
0
: Threshold value of a voltage-dependent switch
U
1
(t), U
2
(t), U
CL
(t): Voltages in the equivalent circuit diagram of a lock-in amplifier
U
D
: Demodulated signal at the output of the detector (low frequency)
U
G
: Gunn voltage at the Gunn diode
U
max
, U
min
, U : Maximum and minimum value of the demodulated receiving signal and the
resulting voltage range
U
n
(t), U
n,rms
: Noise voltage
6
MTS 7.4.4 Contents
U
T
: Temperature voltage
: Amplitude of a voltage
v
1
, v
2
: Drift velocities
v
ph
: Phase velocity
v
r
: Radial velocity
X : Reactance
x, z : Coordinates
Y : Admittance
Z : Impedance
Z
0
: Field characteristic impedance of free space, line impedance
: Geometric angle
: Phase constant
: Penetration depth
: Absolute error

r
,
r
' ,
r
'' : Relative permittivity

0
: Free space wavelength

g
: Guided wavelength
: Phase-shifts

r
,
r
' ,
r
'' : Relative permeability
: Angular frequency
7
MTS 7.4.4
Preface
Preface
The experiments in the present training system
are intended to achieve various training objec-
tives in parallel:
() The understanding of the physical effects
which are of significance in microwave
technology (for example, diffraction and
interference of electromagnetic waves,
Gunn effect).
() Acquiring knowledge of the function of
important components and systems of mi-
crowave technology and the principles
behind the methods of realizing these
functions by exploiting physical phenom-
ena (such as the function of a waveguide
directional coupler in microwave circuits
in black-box representation and the re-
alization of the directional coupler using
effects in electromagnetic coupling
through holes).
() Acquiring skills in measuring techniques
and principles to determine the properties
of microwave devices (example: measur-
ing the reflection coefficient using the re-
flectometer principle).
() Becoming familiar with the actual techni-
cal design of various components and us-
ing these in practical applications (exam-
ples: learning to correctly install a ferrite
waveguide isolator in a microwave circuit,
operation of a slide screw transformer for
matching of a load). The components for
carrying out experiments with the above
objectives are all designed in waveguide
technology. Dual-plate configurations
(unit MTS 7.4.3) are particularly suitable
because they can be easily disassembled
and are also extremely robust. One dis-
crepancy, however, had to be considered
when designing this training system: on
the one hand, waveguide technology is the
most suitable of all technologies for exper-
iment purposes; on the other hand, in the
field of radio-frequency circuit technolo-
gy, it is with few exceptions, being in-
creasingly overtaken by microwave inte-
grated circuits (MIC) utilizing microstrip
line or coplanar line technology.
The result of this discrepancy is the concept of
a teaching and training system based primarily
on waveguide technology, in which the physi-
cal phenomena and the fundamental principles
of technical elements and measuring methods
are given priority, independent of the particular
transmission line forms (waveguide, coaxial
line, microstrip line, etc.). Thus, for example,
the knowledge gained in experiments on Gunn
oscillators using waveguide technology will
enable students to understand an oscillator re-
alized using microstrip technology, as the fun-
damental principles of the interaction between
the various semiconductor elements and the
resonance circuit are the same. With the aid of
an extremely specialized waveguide element
such as the cross directional coupler, students
can become familiar with important effects in
electromagnetic fields, e.g. coupling through
small openings; in addition, the black box
behaviour of the cross directional coupler is
typical of a large class of different directional
couplers (hybrids in microstrip line technolo-
gy, coaxial-line couplers, etc.).
The target group of this teaching and training
system is students who have widely varying
levels of prior knowledge and/or experience.
Thus, the system is equally applicable in uni-
versity-level science instruction and in techni-
cal schools. One may assume that the manner
in which most of the experiments are presented
and carried out is suitable for the learning needs
of this broad target group, but that the contents
and the interpretation of the experiment results
within the framework of a theoretical struc-
ture must be treated differently in each case.
For this reason, this handbook contains some-
times even advanced material. This is in-
tended for university-level students, and may
be omitted when working through the experi-
ments.
8
MTS 7.4.4 Preface
endangered! DO NOT TOUCH! Discharge any
long cables before connecting them to these
components. This is carried out by connecting
them to the power supply unit.
Components which operate with strong perma-
nent magnets, e.g. isolators or circulators, must
be kept at a distance from magnetically con-
ductive materials. Avoid shaking or bumping
the equipment.
The flange surfaces should be treated with care.
Mechanically movable parts are to be carefully
lubricated from time to time. Do not allow any
oil or grease on electrical contacts or in the
waveguide.
Design of the microwave source
With the exception on experiment Ex1 a modu-
lated microwave source is required to perform
the microwave experiments in the MTS 7.4.4
training system. This source generates the mi-
crowave field needed for the load connected
downstream. Only through modulation is it
possible to perform frequency-selective detec-
tion of the demodulated receiving signal and
thus carry out effective noise suppression. For
the design of a modulated microwave source
there are two options shown in Fig. 0.3 and Fig.
0.5.
1. Direct modulation of the Gunn oscillator
2. Modulation with external PIN modulator
1. The direct modulation of the Gunn
oscillator
Direct modulation of the Gunn oscillator is one
way of performing modulation without addi-
tional equipment but also without the best re-
sults. As is shown in Experiment 1 the emitted
microwave power (and the emitted spectrum) is
severely dependent on the Gunn voltage. Fur-
thermore, these variables are also subject to
considerable manufacturer tolerance involving
the Gunn diode. Due to its dependency on the
Gunn voltage, the emitted microwave power
demonstrates discontinuities or irregularities.
In Experiment 2 it is shown that in the case of
amplitude modulation of the Gunn voltage
where the operating point (supply voltage) is in
the proximity of a discontinuity, a slight varia-
tion in the Gunn voltage leads to a considerable
The theoretical principles of the experiments
which are of a general nature may be found in
standard textbooks on the subject. However,
the experiment descriptions also contain some
of the theoretical principles in order to facilitate
the students application of general representa-
tions found in textbooks to the specific knowl-
edge required for each experiment.
This manual is structured as follows:
In the descriptions of the individual experi-
ments, the necessary theoretical principles are
discussed first. This is followed by a list of the
equipment required and a detailed description
of the individual experiment steps (experiment
procedure). The experiments are evaluated us-
ing a list of questions. Each experiment section
includes its own bibliography or, where suffi-
cient, a reference to the bibliography of another
experiment.
In order to facilitate evaluation and to provide
ideas for evaluating the results, empty tables
and diagrams are included in the list of ques-
tions.
Safety Instructions
Important!
Read this before putting the equipment
into operation!
Due to the low power level of the Gunn oscillator
(approx. 10 mW) there is absolutely no danger
for persons conducting the experiment, even in
the case of free space experiments. Neverthe-
less, the following rules should also be observed
in view of later work with stronger RF-sources:
It is imperative that any direct viewing into
the radiating aperture be avoided in all of
the experiments in which RF-power is ra-
diated. This applies for open waveguides,
for example, and especially all types of
antennas.
Disconnect the supply voltage whenever
modifications are made to experiment set-
ups in which waveguide components are
interchanged
Active elements in microwave components can
be destroyed by electrostatic discharge. Detec-
tors and varactor components are especially
9
MTS 7.4.4
Preface
Fig. 0.1: Emitted spectrum for a Gunn voltage
U
G
= 4 V
Fig. 0.2: Emitted spectrum for a Gunn voltage
U
G
= 8 V
variation in the emitted power and thus to a
large demodulated signal.
However, if the operating point is selected
close to a discontinuity (approx. 4V), one con-
sequence is that the emitted spectrum contains
many different frequency components. But due
to the fact that all of the microwave compo-
nents are designed for a frequency of 9.4 GHz,
and normally demonstrate a very narrow fre-
quency bandwidth, the measurement results for
this case are frequently insufficient and require
explanation.
Fig. 0.1 and Fig. 0.2 provide examples of the
emitted spectrum of the Gunn oscillator for
Gunn voltages of 4 V and 8 V respectively.
The emitted spectrum varies for discontinuities.
A purer spectrum (TE
101
mode at 9.4 GHz and
TE
202
mode at 18.8 GHz) is generally obtained
starting at a Gunn voltage higher than 8 V.
On its own the emitted microwave power
changes only very slightly in the range of
higher Gunn voltages (7 V up to 10 V) with the
variation of the Gunn voltage (flat characteris-
tic). This yields a demodulated signal which is
very small. This means that in the range where
the Gunn diode emits a pure spectrum (starting
at 8 V) direct modulation is possible with only
a very dimunitive amplitude deviation.
Add to this the further disadvantage that the
Gunn oscillator is subjected to reflections
which can alter its operating features, e.g. para-
sitic oscillations can be generated.
Due to the fact that the PIN-Modulator is only
included as a recommendation in the MTS 7.4.4
equipment set, the use of the PIN modulator is
dispensed with in the experiments whereever
possible. If you have a PIN modulator avail-
able, then we highly recommend that you use it
(see also subpoint 2) as this normally yields
better measurement results.
For the internal direct modulation of the Gunn
oscillator
connect the Gunn oscillator to the GUNN
socket in the GUNN POWER SUPPLY
section of the basic unit.
set the toggle switch in the PIN MODU-
LATOR section of the basic unit to
GUNN INT.
set the Gunn voltage to approx. 8 V (up to
10 V) using the U
G
controller.
The Gunn voltage is then superimposed by a
square-wave signal with a frequency of 976 Hz
and an amplitude of 300 mV, see Fig. 0.4.
GUNN-OSC.
737 01
737 01
Fig. 0.3: Connection sketch for direct modulation of the
Gunn oscillator. Switch set to GUNN INT.
1 0
MTS 7.4.4 Preface
Fig. 0.5: Connection sketch for modulation with external
PIN modulator and internal modulation signal.
Switch set to PIN INT
Fig. 0.4: Principle characteristic of the Gunn voltage on
the basic unit (GUNN output socket) in the case
of Gunn-internal modulation. U
G
represents the
supply voltage, which is set using the rotary
potentiometer U
G
.
2. Modulation with an external PIN
modulator
The experiment setup is performed as depicted
in Fig. 0.5.
The Gunn oscillator generates a continuous wave
power, i.e. its power output is constant in time.
The modulation is performed by a PIN modula-
tor connected downstream. Depending on the
control voltage this can have a variably high re-
flection and transmission coefficient. In order to
prevent any undesired reflections from reflecting
back to the Gunn oscillator, this is isolated from
the rest of the circuit by an isolator. The unit con-
sisting of the Gunn oscillator, isolator and PIN
modulator constitutes a typical configuration fre-
quently used in practical applications and is
known for its easy handling. Thus you avoid the
disadvantages of direct modulation performed
along a Gunn diode's characteristic (subject to
discontinuity) which is severely dependent on
manufacturer's tolerance. The modulated signal
is considerably better than direct modulation
because in this case the amplitude range only
depends on the control voltage of the PIN
modulator. As the Gunn diode can operate at a
fixed operating point, the emitted microwave
spectrum is not affected by the modulation. Ad-
ditional information can be taken from the in-
struction sheet for the PIN modulator.
If you have a PIN modulator at your disposal,
then we recommend using it for modulation to
obtain better measurement findings. The PIN
modulator can be modulated internally or ex-
ternally. For external modulation you need an
additional function generator. In this case a
measurement with the internal SWR meter in no
longer possible as this requires frequency and
phase synchronization.
1 1
MTS 7.4.4
The Gunn Oscillator
The Gunn oscillator
Principles
Microwave power, i.e. electromagnetic power
in the GHz frequency range, may be generated
using quite different physical phenomena.
Some examples of this are vacuum-tube oscil-
lators such as the klystron and the magnetron,
or semiconductor oscillators such as the FET
oscillator, the Gunn oscillator and the impatt
oscillator. Some simple experiments for under-
standing Gunn oscillators are described below.
Gunn effect
In some semiconductor materials, such as Gal-
lium Arsenide (GaAs), the mobility of the elec-
trons (= the quotient of the drift velocity v and
the electrical field strength E) decreases above
a threshold value E
TH
of the electrical field
strength (see Fig. 1.1, left side). This is because,
as the field strength increases more and more
electrons transfer to a state in which their ef-
fective mass becomes greater, thus decreasing
their velocity. For field strengths where E > E
TH
the electrons have a negative differential mobil-
ity, i.e. an increase in the field strength results
in a decrease in the drift velocity.
When the electrical field strength in a homoge-
neously doped GaAs block (no barrier layer!)
is greater than the threshold value E
TH
space
charge instabilities occur as a result of the
negative differential mobility.
While any random local surplus or deficiency
of electrons will disappear by itself when a
positive differential mobility is present, this sur-
plus or deficiency will increase under a nega-
tive differential mobility. In the upper right
section of Fig. 1.1, a random surplus of elec-
trons is assumed. The resulting increase in field
strength on the anode side leads to a decrease
in the drift velocity v
2
on the anode side relative
to the drift velocity v
1
on the cathode side (de-
crease in field strength). This causes bunch-
ing and yields a carrier enhancement layer.
The analogous effect occurs in the event of a
random deficiency of electrons, in which case
a depletion layer occurs (see Fig. 1.1, middle-
right section). When the enhanced and de-
pleted layers approach each other, they attract
each other and jointly pass through the diode in
the form of a domain (see Fig. 1.1, lower
right section). The field in the interior of the
domain can be so high that the field outside it
falls below the threshold value E
TH
. Thus no
new domain can be formed until the existing
one has disappeared at the anode.
If the Gunn element were not connected to a reso-
nator (tuned circuit), the frequency of the micro-
wave power generated would be determined by
the time it takes for the domains (velocity approxi-
mately 10
7
cm/s) to pass through the diode (transit
frequency). If, however, the Gunn diode is oper-
ated with a resonator, the resonator frequency can
be imposed upon the Gunn diode. There are
several operating modes here (delayed mode,
quenched mode, LSA mode).
The delayed domain mode is briefly described
here as an example. It occurs when the resonator
frequency is lower than the transit frequency. At
the moment in which the domain reaches the
anode, the momentary value of the diode voltage
(= bias voltage + RF voltage) is less than the
threshold value. The formation of a new domain is
Fig. 1.1 Principle of a Gunn diode
left: Drift velocity v of the electrons as a
function of the electrical field strength E for
GaAs.
right: Formation of a domin (cathode () left
and anode (+) right)
1 2
MTS 7.4.4 The Gunn Oscillator
delayed until the voltage exceeds the threshold
value, thus imposing the oscillator frequency of
the resonator on the Gunn element.
Design of the oscillator
One of the many possible resonator types in mi-
crowave technology is the rectangular cavity
resonator. Fig. 1.2, left side, shows a rectangu-
lar cavity enclosed on all sides by metal walls.
Just as in a resonant circuit built of a lumped
inductance and capacitance, oscillations of the
electrical field variables with a certain frequen-
cy (= resonance frequency f
0
) can also exist in a
cavity resonator. In this case, the energy is
stored alternately in the electrical and the mag-
netic field. While an (ideal) resonant circuit has
only one resonance frequency, the cavity reso-
nator has an infinite number of oscillation types
and resonance frequencies. Fig 1.2 (left) shows
the electrical and magnetic fields for the oscil-
lation type with the lowest resonance frequen-
cy (TE
101
resonance) at three different points in
time, at intervals of one quarter of the period
T = 1 / f
0
. A side view of the cavity resonator
and the variations of the electrical field strength
as a function of the coordinate z are reproduced
in the middle of Fig.1.2. The resonance fre-
quency of this oscillation type is calculated (for
air) according to the formula:
f
a s
0
2 2
GHz
15
1
( /cm)
1
( /cm)
= +
' '
(1.1)
More specific details on cavity resonators are
dealt with in Experiment Ex12.
The electromagnetic oscillations of the cavity reso-
nator are attenuated due to losses occurring in the
metal walls. After installing the Gunn element,
which transforms DC power into microwave
power, just enough microwave power is fed into
the resonator to compensate for the wall losses and
to achieve a continuous unattenuated oscillation.
To obtain a microwave oscillator, the resonator
must also have an opening through which
power can be fed to a load. In this case, the
Gunn element must generate not only enough
microwave power to compensate for the wall
losses, but also the greater amount of power
Fig. 1.2: Basic design of the Gunn oscillator
left: TE
101
oscillation in a rectangular cavity
resonator
middle: Dependency of the electrical field
strength E on the longitudinal coordinate z for
TE
101
resonance
right: Two possible configuratios for the Gunn
oscillators (1) Gunn diode (2) aperture
Configuration B is used in these experiments
and configuration A in MTS 7.4.6.
generally required by the load. Developing a
Gunn oscillator from a cavity resonator re-
quires (I) that the Gunn element be coupled to
the resonator, and (II) that the load be coupled
to the resonator. Fig. 1.2, right section, shows
two possible configurations.
In configuration (B), the Gunn element is cou-
pled to the resonator using a round metallic
post, and the load is coupled using an aperture
(hole or slot). From the sketch of the longitudi-
nal electrical field distribution in configuration
(B), we can see that the planes of the post axis
and the aperture may be regarded as short-cir-
cuit planes for the purpose of estimating the
resonance frequency.
In configuration (A) (Fig. 1.2, top right) the me-
tallic post fullfils both functions, i.e. the cou-
pling of the resonator to the Gunn element as
well as to the load.
In the present experiment, an oscillator con-
structed according to configuration (B) is used.
In contrast in experiments from MTS 7.4.6 a
mechanically adjustable oscillator according to
configuration (A) is assembled.
1 3
MTS 7.4.4
The Gunn Oscillator
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Fixed attenuator 737 095
1 Transition waveguide/coax 737 035
1 Coax detector 737 03
1 Thumb screws (2 each) 737 399
Additionally required equipment
1 Oscilloscope 575 29
1 XY recorder (optional) 575 663
2 Stand bases 301 21
1 Support for waveguide components 737 15
1 Stand rod 0.25 m 301 26
2 Coax-cables with BNC/BNC
plugs, 2 m 501 022
1 Slide caliper
Experiment procedure
1. Visually study the design of the disassem-
bled Gunn oscillator (see also the instruc-
tion sheet to the device).
1.1 Disassemble the Gunn oscillator by loosening
the quick-release thumb screws. (Disconnect
the back panel of the housing, diaphragm and
waveguide terminating piece)
1.2 Consider the design of the Gunn element and
compare it to Fig. 1.2 (right). Determine the
waveguide width a and the distance s
[see Fig. 1.2 right, configuration (B)] of
the post axis to the flange plane (= location
of the diaphragm) using a slide caliper.
Note:
Make sure that you DO NOT touch the Gunn
diode with the slide caliper.
2. Determine the dependency of the DC cur-
rent I
G
flowing through the Gunn element
and (in the case of oscillation) the micro-
wave signal generated by the supply volt-
age U
G
(when the diaphragm is attached).
2.1 Reassemble the Gunn oscillator dismantled in
part 1 of the experiment, i.e. reattach the
back panel of the housing and the waveguide
to the Gunn element module.
2.2 Screw the fixed attenuator and waveguide/
coax transition onto the open end of the
waveguide (Fig. 1.3).
The attenuator keeps the detector in
the square-law characteristic range (see
MTS 7.4.5 and Experiment 4 of this book)
and attenuates undesired reflections]
2.3 Set up the equipment configuration on the
lab bench using the stand rods.
2.4 Connect the basic unit (set the rotary knob for
the Gunn diode supply voltage U
G
to 0)
to the Gunn oscillator using the coaxial
cable.
2.5 Attach the coax-detector to the
waveguide/coax transition (making sure
here that the thread does not get jammed)
2.6 Connect the detector to the oscilloscope
(not contained in the measurement sta-
tion). Set the oscilloscope to DC and the
measurement range for the following
measurements to approx. 2 to 100 mV.
2.7 Switch on the Gunn power supply unit. In-
crease the supply voltage U
G
from 0 to 10 V
in steps of 0.5 V. At the same time read off
the Gunn element's DC current I
G
and the
receiving voltage U
D
proportional to the rai-
ated microwave power and enter the results
in columns 2 and 3 of Table 1.1. Draw the
findings in Diagram 1.1. Connect the points
representing the measurements from Table
1.1 with straight lines.
Note:
If an XY recorder is available, the function
I
G
= I
G
(U
G
) can also be recorded directly.
Fig. 1.3: Experiment setup
1 4
MTS 7.4.4 The Gunn Oscillator
In this case the appropriately designated
sockets X and Y of the Gunn power sup-
ply of the basic unit are connected to the X
and Y inputs of the XY recorder. Likewise
the function U
D
(U
G
) can be recorded. The
Y-input of the recorder must be connected
to the coax detector (Since U
D
is negative,
interchange the +- and input).
For the recording of the characteristic the
Gunn voltage is to be increased slowly by
turning the 10-turn potentiometer. You
can also use a digital storage oscilloscope
instead of the XY recorder. In this case an
additional triangular function generator (0
to 10 V) is required for wobbling the char-
acteristics. A very low wobble frequency
must be selected here, because the RE-
with diaphragm
with rear panel
without diaphragm
with rear panel
with diaphragm
without rear panel
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
U
G
mV
I
G
mA
U
D
mV
I
G
mA
I
G
mA mV
V
CORDER outputs X and Y have lowpass
filters. The advantage of a continuous in-
crease of U
G
as opposed to a step-by-step
increase (0.5 V steps) lies in the fact that
the steps (discontinuities) in the charac-
teristics are easier to discern.
3. Repeat part 2 of the experiment but with-
out the diaphragm.
3.1 To remove the diaphragm loosen the thumb
screws, take out the diaphragm and then re-
assemble the waveguide adapter (including
the waveguide/coax transition).
3.2 Repeat the experiments in accordance with
point 2.7. Enter the results for I
G
and U
D
in
columns 4 and 5 of Table 1.1
Table 1.1 a'= mm s'= mm
U
D
U
D
1 5
MTS 7.4.4
The Gunn Oscillator
2. Assuming that the active GaAs layer has a
thickness of 10 m, determine from the value
for the threshold voltage U
TH
(= voltage
above which the differential mobility be-
comes negative) the threshold value E
TH
of
the electrical field strength in kV/cm. The
voltage drop outside the active layer may be
ignored here, and you may assume a ho-
mogenous spatial distribution of the field
strength.
3. Assuming a domain velocity of 10
7
cm/s,
determine the transit frequency of the Gunn
element.
4. Explain the different responses obtained in
experiment parts 2 (with diaphragm and rear
panel), 3 (without diaphragm, with housing
rear panel) and 4 (with diaphragm, but
without rear panel).
4. Repeat part 2 of the experiment, but without
the rear panel.
4.1 Reinsert the diaphragm. Remove the rear
panel.
4.2 Repeat the experiment according to point
2.7. Enter the results for I
G
and U
D
in col-
umns 6 and 7 of Table 1.1
Questions
1. Calculate approximately the oscillator fre-
quency according to the Equation (1.1) speci-
fied above for the resonance frequency of
a rectangular cavity resonator. For this use
the geometrical data determined in experi-
ment part 1.2. Here you may assume a
TE
101
resonance with effective short-cir-
cuit planes at the location of the post axis
and the diaphragm (see also Fig. 1.2,
lower right).
Diagram 1.1: For the graphic representation of the measurements values from table 1.1
1 6
MTS 7.4.4 The Gunn Oscillator
Bibliography
Simple introductory works:
[1] F. Nibler u. a. : Hochfrequenzschaltungstechnik (Abschnitt 1.5). Expert-Verlag,
Sindelfingen 1984
[2] G. A. Acket, R. Tijburg und P. J. de Waard: Die Gunn-Diode.
Philips techn. Rdsch. 32, 394-404 (1971/72)
[3] J. Magarshack: Gunn-Effekt-Oszillatoren und -Verstrker.
Philips techn. Rdsch. 32, 424-431 (1971/72)
[4] Jansen: Entwurf von Gunn-Effekt Oszillatoren. Mikrowellen Magazin, 330-357 (1977)
Tiefergehende Darstellungen wissenschaftlichen Charakters:
[5] H.-G. Unger und W. Harth: Hochfrequenz-Halbleiterelektronik. Hirzel-Verlag,
Stuttgart 1972
[6] K. Kurokawa: Microwave Solid State Oscillator Circuits. In: M. Howes, D. Morgan (Eds.):
Microwave Devices, Device Circuit Interactions. Wiley, London 1976
[7] J. E. Carroll: Hot Electron Microwave Generators. Edward Arnold, London 1970
[8] F. Sterzer: Transferred Electron (Gunn) Amplifiers and Oscillators for Microwave Applications.
Proceedings IEEE, 59, 1155-1163 (1971)
[9] C. P. Jethwa und R. L., Gunshor: An Analytical Equivalent Circuit Representation
for Waveguide- Mounted Gunn Oscillators. IEEE Trans. Microwave Theory Tech.,
MTT-20, 565- 572 (1972)
[10] J. F. White: Simplified Theory for Post Coupling Gunn Diodes to Waveguides. IEEE Trans.
Microwave Theory Tech., MTT-20, 372-378 (1972)
1 7
MTS 7.4.4
Modulation of the Gunn Oscillator
Direct modulation of the Gunn oscillator
Principles
In order to investigate direct modulation of the
Gunn oscillator, you first have to delve into
microwave detection. The waveguide/coax
transition used in Experiment 1 converts the
propagating wave into a TEM wave. The coax-
ial detector serves to determine the electrical
field strength of this TEM wave. To do this
there is a miniature antenna structure located
inside the detector. This antenna supplies a
voltage which is proportional to the sought-af-
ter electrical field strength.
The voltage u(t) being applied to the terminals
of the antenna represents a microwave signal
(here approx. 9 GHz), which cannot be directly
displayed on a conventional oscilloscope.
A detector diode is used to generate a cohesive
LF signal which coincides with the envelope of
the microwave signal. Schottky diodes are par-
ticularly well-suited for this. They essentially
consist of a metal-to-semiconductor junction.
The following holds true for the relationship be-
tween the detector voltage u(t) and the detector
current i(t) in a Schottky diode
i t I
u t
U
( )
( )

j
(
,
\
,
(

,

,
,
]
]
]
]
S
T
exp 1 (2.1)
where U
T
stands for the temperature voltage (20
up to 30 mV) and I
S
stands for the saturation cur-
rent.
For low modulation levels | u(t) | << U
T
the
above equation can be approximated using the
two first terms of Taylor's series
i t
I
U
u t I
u t
U
( ) ( )
( )
+
j
(
,
,
\
,
(
(
S
T
S
T
2
1
2
(2.2)
A microwave voltage
u(t) = cos ( t) (2.3)
yields the following direct current based on the
square component of the above expression
I
I
U
u
Detect
S
T
2
2
1
4
( )

(2.4)
Due to the fact that is, on the other hand, pro-
portional to the amplitude of the electrical
field strength at the location of the probe, the
coax detector consisting of a dipole antenna and
detector diode supplies an output signal U
D
,
which for sufficiently small field strengths is
proportional to the square of the amplitude of the
electrical field:
U
D
= k
2
(2.5)
Here k is a constant with the dimension
m
2
V
The gain and measurement of small DC detec-
tor voltages is complicated by, among other
things, drift phenomena. For that reason it is
expedient to have a low-frequency AC signal
available at the output of the coax detector in-
stead of the DC voltage signal. This is made pos-
sible through amplitude modulation of the micro-
wave oscillator (in this case the Gunn oscillator).
Amplitude modulation of the Gunn oscillator's
microwave signal can be achieved by means of
a time-variable supply voltage. This becomes
clear from the results from Experiment 1 (Dia-
gram 1.1). There are no microwave oscillations
up to a supply voltage of approx. 4 V. Above
this threshold the amplitude of the microwave
signal is severely dependent on the supply volt-
age. Thus the amplitude of the microwave sig-
nal can be modulated as a function of time by
controlling the bias level U
G
of the Gunn oscil-
lator.
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Fixed attenuator 737 095
1 Transition waveguide/coax 737 035
1 Coax detector 737 03
1 Set of thumb screws (2 each) 737 399
Additionally required equipment
1 Oscilloscope 575 29
2 Stand bases 301 21
1 Support for waveguide
components 737 15
1 Stand rod 0.25 m 301 26
2 Coaxial cables with BNC/BNC
plugs, 2 m 501 022
1 8
MTS 7.4.4 Modulation of the Gunn Oscillator
Fig. 2.1: Basic characteristic of the Gunn voltage on the
basic unit (GUNN output socket) for Gunn-
internal modulation. U
G
represents here the
supply voltage, which is set via the rotary
potentiometer U
G
.
Experiment procedure
1. Amplitude modulation of the Gunn oscilla-
tor
1.1 Experiment setup as in Experiment 1 and
specified in Fig. 1.3.
1.2 In the MODULATION section of the
basic unit there is a switch setting (GUNN
INT) for the internal modulation of the
Gunn supply voltage. The Gunn voltage is
then superpositioned onto the square-wave
signal with the frequency 976 Hz. The am-
plitude of the signal has a value of 300 mV
PP
(see Fig. 2.1).
1.3 Consider the receiving signal U
D
(t) of the
detector for an amplitude-modulated micro-
wave signal when changing the supply volt-
age U
G
.
For this first connect the Gunn oscillator to
the basic unit (GUNN sockets) and set the
toggle switch to GUNN INT. Connect
the detector to the input of the oscillo-
scope (DC setting first).
1.4 Slowly turn the control knob for the Gunn
voltage from left to right (increasing the
supply voltage) and observe how the char-
acteristic of U
D
(t) (square-wave shaped
signal with DC component) changes.
1.5 Now set the oscilloscope to AC mode and
repeat point 1.4.
Questions
1. What can be observed in subpoints 1.4
and 1.5? How do you explain this re-
sponse?
1 9
MTS 7.4.4
Amplifier
The selective measurement amplifier
Principles
Figure 3.1 shows the dynamic characteristic
U
D
(t) of the output signal of the E-field probe. It
results from the superposition of the square-
wave shaped wanted signal of the frequency
f
r
(here f
r
= 976 Hz) with the noise signal U
n
(t).
If the rms value
u U t
n,rms n
2
= ( ) (3.1)
of the noise signal is larger than the amplitude of
the voltage step U = U
max
U
min
to be deter-
mined, than this determination is made consid-
erably more complicated, if not impossible
should no additional measures be taken on the
signal processing side. These considerations
lead to the sensitivity limits of the measurement
method.
In order to recognize which measures lead to an
increase of the sensitivity (lowering of the sen-
sitivity limits), it is advantageous to consider
the frequency spectrum of the signal U
D
(t). Fig-
ure 3.2 (above) depicts this frequency spectrum,
which consists of spectral lines at the frequencies
f
r
, 3 f
r
, 5 f
r
etc. belonging to the wanted periodic
signal and a continuous noise spectrum. The first
spectral line at f
r
belongs to the fundamental fre-
quency component.
U t u f t
D r
cos 2 ( ) ( ) = (3.2)
whereby is definitely related to U via
= 2 U/ (Fourier analysis).
If the received signal U
D
(t) is sent through a
narrow bandpass filter (see Fig. 3.2, middle)
with the center frequency of f
0
f
r
, then the
first spectral line and thus the fundamental fre-
quency component U
D
(t) is almost completely
retained. However, only a small part of the
noise spectrum remains which is determined by
the effective bandwidth of the filter. The signal-
to-noise-ratio is considerably increased by
means of this form of selective frequency filter-
ing. And this increase is even greater the nar-
rower the bandwidth of the filter. In the
corresponding dynamic signal characteristic
obtained downstream from the bandpass filter,
the fundamental frequency component remains
nearly the same in comparison to the input sig-
nal, but the rms value of the noise signal has
been reduced in proportion to the bandwidth of
the filter. Based on the principle expounded
upon until now you could replace the bandpass
filter with a narrow-band low-niose amplifier
and supply its output signal to an AC voltme-
ter. Thus, a frequency selective measurement
amplifier is obtained; a principle often applied
in the measurement of small signals.
However, if a still greater increase in the sensi-
tivity is desired, then we must also make it clear
that there is a limit to the reduction of the filters
Fig. 3.2: Frequency spectrum of the signals according to
Fig. 3.1 upstream (above) and downstream
(below) from a narrow-band bandpass filter.
1 Wanted signal
2 Noise
3 Frequency response of the filter
Fig. 3.1: Dynamic characteristic of the output signal of the
E-field probe with superpositioned noise voltage
(f
r
= 976 Hz)
u
min
u
max
U
D
(t)
U
2 0
MTS 7.4.4 Amplifier
Fig. 3.3 Design of the lock-in amplifier
1 Clock generator
2 Bandpass filter for suppesing the
harmonics
3 Amplifier
4 Phase-sensitive rectifier
(Synchronuos rectifier)
5 Low-pass filter
Fig. 3.4: Voltage controlled switch as phase-sensitive
rectifier (PSD). Switch set to position A, if
voltage U
Cl
(t) of clock generator is smaller than
threshold value U
0
. Otherwise in setting B.
for the case that U
1
(t) is in phase with the clock
signal U
CL
(t), is shown in Fig. 3.5 (right). If a
very narrow-band low-pass filter (e.g. band-
width 2 Hz) is arranged behind the PSD, this
filter supplies the mean value u
2
[DC voltage
component, see Fig. 3.5 (right)] to its output,
which is clearly in conjunction with
1
. In order
to conduct a more exact analysis of the lock-in
amplifier features, you can observe its re-
sponse to an input signal of any given fre-
quency f and phase in accordance with the
equation
U t u f f t
D
cos 2 ( ) = ( ) + ( )
(3.3)
Only if f equals f
r
or a multiple of f
r
in whole
numbers, is there a DC voltage value u
2
other
than zero at the output. A very high frequency
bandwidth using the principle dealt with up to
now. The narrower the bandwidth of the filter
is, the better its center frequency f
0
must coin-
cide with the clock frequency f
r
, so that the
wanted signal is not significantly attenuated by
the filter. Due to drift phenomena in the filter,
e.g. of a thermal nature or due to fluctuations in
the clock frequency, deviations between f
0
and
f
r
must be tolerated to a certain extent. Therefore,
the bandwidth of the filter cannot be reduced to
arbitrarily low values.
A solution to this problem can be found when
the clock signal, on which the voltage U
D
(t) to
be measured is based, is available and can thus
be used for the synchronization (f
0
joined
with f
r
) of the bandpass filter.
This basic idea is utilized in the so-called lock-
in amplifiers, the fundamental principle of
which is explained in the following paragraphs.
First a simple bandpass filter (with no extreme
demands on the bandwidth) eliminates the re-
ceived signal U
D
(t) from harmonics of a higher
mode and spectral noise components and then
the filtered signal can be amplified in a low-
noise, narrow-bandwidth amplifier. The result-
ing signal U
1
(t) (downstream from the
amplifier (3), see Fig. 3.3) is supplied together
with the signal U
CL
(t) of the clock generator to a
phase sensitive detector (PSD), sometimes
called a synchronous detector. Figure 3.4
shows a simple design for a PSD. Here the sig-
nal of the clock generator controls a switch so
that the output signal is alternatively identical to
the input signal U
1
(t) or the negative input sig-
nal U
1
(t). The resulting output voltage U
2
(t)
Fig. 3.5: Voltage characteristics of the single phase-
sensitive rectifier
Clock generator voltage U
Cl
(t) (above),
Input voltage U
1
(t) (bottom left),
Output voltage U
2
(t) (bottom right)
2 1
MTS 7.4.4
Amplifier
selectivity is achieved by selecting a very nar-
row bandwidth for the low-pass filter (= pro-
longed duration of the integration time). This
yields a considerable improvement in the sig-
nal-to-noise ratio (see above). Thus, extremely
weak signals can be detected in measurement
systems which are based on the lock-in ampli-
fier principle.
If it is true that f = f
r
, but 0 (phase-shift be-
tween the clock and receiving signal), then
u
u
2
=
2
cos
1


(3.4)
Thus, we obtain not only frequency- but also
phase-selectivity.
In the lock-in amplifier version considered up
until now the receiving signal U
D
(t) appears in
its baseband. Such systems are referred to as
homodyne and is the type of system integrated
into the basic unit of the existing training sys-
tem. It is designated here (equipment designa-
tion) as SWR Meter. The designation SWR
meter (standing wave ratio) comes from its
frequent use in determining the standing wave
ratio on transmission lines.
In improved systems of greater complexity the
received signal is first converted into a (fixed)
intermediate frequency (IF) and supplied to the
PSD (Heterodyne System).
Requiredequipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Waveguide 200 mm 737 12
1 Transition waveguide/coax 737 035
1 Coax detector 737 03
1 Set of thumb screws (2 each) 737 399
Additionallyrequiredequipment
1 Oscilloscope 575 29
4 Stand bases 301 21
3 Supports for waveguide
components 737 15
2 Stand rods 0.25 m 301 26
2 Coax cables with BNC/BNC
plugs, 2 m 501 022
Experiment procedure
1. Operationoftheselectivemeasurement
amplifier
1.1 Set up the experiment as specified in Fig.
3.6. Position the two waveguide ends op-
posite each other at a distance of approx.
Fig. 3.6: Experiment setup
4 cm (The waveguide axes are in perfect
alignment without transverse shift).
1.2 Connect the coax-detector to the input of
the SWR meter. Set the toggle switch for
modulation to GUN-INT. Connect the
Gunn oscillator to the GUNN socket of the
basic unit.
1.3 Presetting of the selective measurement
amplifier:
Set the gain selection switch of the selec-
tive measurement amplifier (SWR meter)
to 0 dB.
Turn the GAIN ZERO to far left stop.
1.4 Set U
G
for maximum pointer deflection on
the SWR meter (see Experiment 2).
(Note: in this experiment the spectral mode
distribution is not decisive, in this context
see also Design of the Microwave
Source in the preface). In the process
vary the gain using the selection switch
until the display is in the range from 0 to
5 dB. Afterwards set the SWR meter to
maximum pointer deflection by setting the
ZERO control knob to 0 dB.
2. Transverseshiftofthewaveguideendin
0.5cmsteps.
2.1 As shown in Fig. 3.7 shift the end of the
waveguide in a transverse manner in ac-
cordance with the values given in Table
3.1 and read off the respective values (in
dB) displayed on the selective measure-
ment amplifier and enter these measured
2 2
MTS 7.4.4 Amplifier
values into column 2 of the table. During
this measurement select an appropiate set-
ting for the gain factor of the selective
measurement amplifier and take this into
account in the result.
2.2 Now connect the detector to the oscillo-
scope (AC setting) and repeat experiment
point 2.1. This time determine the voltage
range U(x
0
) = U
max
(x
0
) U
min
(x
0
) of the
receiving signal and enter it into column 3
of the Table.
Note:If you have a BNC-T adapter (cat.
no. 501 091) and an additional RF cable
(cat. no. 501 022) at your disposal, you
can do experiment points 2.1 and 2.2 in
parallel.
3. Demonstratingthesensitivitygainwhen
usingtheselectivemeasurementamplifier
3.1 Move one of the waveguides until noise
makes it impossible to recognize the
square-wave signal on the oscilloscope.
Covering one waveguide aperture with
your hand should now have no effect on
the signal.
3.2 While maintaining the transverse shift as it
is, now connect the detector to the meas-
urement amplifier (INPUT socket) and set
the display of the SWR meter to a value
between 5dB und 0 dB by selecting a
suitable gain factor. Now cover one
waveguide aperture with your hand and
observe the display.
Questions
1. Based on column 3 of the table (U) deter-
mine the respective values
Fig. 3.7: Transverse shift of the waveguide axes (view
from above)
U x
0
( )
mV
a
dB
10 log
(0)
0

U x
U
( )
0 0.0 0.0
0.5
1
1.5
2
2.5
3
x
0
cm
Table3.1
10 log
(0)
10 log
(0) (0)
0 max 0 min 0
max min
=

U x
U
U x U x
U U
( ) ( ) ( )
(3.1)
for the given transverse distances x
0
and
enter these into column 4.
2. Compare the findings in columns 2 and 3
of the table.
Comments:
If under point 2.2 you are already unable
to determine U(x
0
) in the range around
x
0
= 0 up to 2 cm (due to too much noise),
reduce the longitudinal distance from
4 cm to a smaller value, and repeat the en-
tire experiment.
However, if you have too high a signal
level you can also increase the longitudi-
nal distance and repeat the entire experi-
ment.
2 3
MTS 7.4.4
Attenunators
Attenuators
Calibrating the attenuator, measuring relative microwave power
side, or slide the attenuator vane into the field
from the side. The latter possibility is imple-
mented in the attenuator 737 09. The shift
movement of the attenuator vane is performed
backlash-free using a micrometer screw. In or-
der to attain low insertion loss the vane is insert-
ed in such a manner that the absorbing layer
lies against the waveguide wall (E = 0) for a
shift of x = 0 mm. Because of scarfing the re-
flection coefficient caused by the vane is kept
as small as possible. The attenuating vane itself
can consist of coated fiber glass, mica or plas-
tics like mylar or Kapton. Nickle-chrome or
tantalum alloys are deposited as layered coat-
ings in vacuum metallization processes. These
metal layers have inherent absorbtion properties
(as opposed to reflecting ones) as long as their
layer thicknesses are smaller than the penetrat-
ing depth of the electrical field at the desired
operating wavelengths. Resistance film cards
are manufactured with varying resistance per
square unit (surface resistance). Standard values
lie between 50 and 377 /square (Square means
that the resistance is the same for any random
square surface and is given a specific value. Thus
the resistance is dependent on the geometry, not
on the surface area). The surface resistance in an
attenuator in conjunction with the field distribu-
tion of the mode in question (often TE
10
) deter-
mines the attenuation characteristic as a function
of the shift x. Fig. 4.2 contains a cross-section
through the adjustable attenuator 737 09.
The calibration of the attenuator is performed in
the experiment using the power scale of the
SWR meter. The microwave power is con-
Definition and properties
Attenuators are among the linear, reciprocal
components of electrical lines (four-pole). They
are frequently realized like reflection-free
waveguide terminals in the form of dissipating
resistances. As such the operating principle
comprises the transformation of RF power into
thermal energy. With the exception of a few
high-load attenuators reciprocity always exists.
Only in the former is the input designed to take
more power than the output, which is why any
interchanging of the gates is not permitted. A dis-
tinction is drawn between fixed attenuators and
variable attenuators. Variable attenuators can be
adjusted mechanically or manufactured with
electronically controllable line components. The
electronic attenuators are designed using, for ex-
ample, PIN diodes. The PIN diode is used here
as an electrically controllable resistance for mi-
crowaves. For that reason a PIN diode attenua-
tor has variously high transmission and reflection
coefficients depending on the control voltage.
The function of the PIN diode is explained in
greater detail in Experiment Ex1 Principle of
the PIN modulator in MTS 7.4.5, for that rea-
son we will only focus here on passive attenuat-
ing elements. Fig. 4.1 presents 2 conventional
principles for the assembly of mechanically tune-
able attenuators.
The basic idea involves inserting an absorbing
medium into the waveguide. The rectangular
waveguide depicted in Fig. 4.1 guides the funda-
mental mode (TE
10
). You can either insert a
vane attenuator into the waveguide through a
middle slot along the waveguide's longitudinal
Fig. 4.1: On the principle of attenuation by means of
insertion with an attenuating vane
Fig. 4.2: Cross-section through the attenuator 737 09
2 4
MTS 7.4.4 Attenunators
verted into a LF signal with the coax detector.
As long as the coax detector is operated in the
range of square-law characteristic there is a
proportional relationship between its output
signal and the incidenting microwave power. A
correct reading of the dB scale is only possible
under these conditions. Generally speaking
when it comes to detectors, low power levels
are the precondition for the square law charac-
teristic range. Fig. 4.3 shows the principle char-
acteristic curve of the output signal versus the
microwave power.
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Fixed attenuator 737 095
1 Variable attenuator 737 09
1 Transition waveguide/coax 737 035
1 Coax detector 737 03
1 Set of thumb screws (4 each) 737 399
Additionally required equipment
2 Stand bases 301 21
2 Supports for waveguide
components 737 15
1 Stand rod 0.25 m 301 26
2 Coaxial cable with BNC/BNC
plugs, 2 m 501 022
Recommended
1 PIN modulator 737 05
1 Isolator 737 06
Experiment procedure
Note:
When using the PIN modulator and isolator
complete the experiment setup as specified in
Fig. 0.5 (Preface).
1. Calibration of the attenuator
Experiment setup as specified in Fig. 4.4
Note:
The fixed attenuator is required to attenu-
ate the microwave signal present at the
coax detector by approx. 10 dB. This is
how the detector is supposed to be operat-
ed in its square-law characteristics range.
Connect the coax detector to the SWR re-
ceiver INPUT.
Modulate the microwave signal (generally
performed by means of direct modulation
of the Gunn oscillator). Set the variable
attenuator to x = 0.00 mm. Calibrate the
display a of the homodyne SWR meter to
Fig. 4.3: Detector output voltage as a function of the
microwave power
1 Saturation range
2 Linear range
3 Quadratic range
4 Noise
Fig. 4.4: Experiment setup
2 5
MTS 7.4.4
Attenunators
0 dB using the gain selection switch V/dB
and the ZERO control knob. Do not make
further changes to the ZERO control knob
in the course of the experiment.
First check the validity of the square law
(the detector operates in the linear range).
Now increase the attenuation to 3 dB by
turning the micrometer screw (set x to
value specified on the attenuator). The dis-
play a of the power meter should now lie
in the range from 2.5 dB to 3.5 dB. You
will hardly be able to reach the ideal value
of 3 dB due to the unavoidably large dis-
persion of the detector's microwave diode.
Reset the attenuator to x = 0.00 mm. Now
set the attenuation to the values specified
in the table. Enter the measured values x
into Table 4.1 and plot the dependency
|a (x)| in a graph.
Questions
1. Which other components can be used to
reduce the microwave power to avoid
overloading the detector?
Table 4.1: Calibration of the attenuator
a / dB x / mm
0 0.00
1
2
3
4
5
6
7
8
9
10
12
14
16
18
20
Diagram 4.1: Plot of the values from Table 4.1
2 6
MTS 7.4.4 Attenunators
2 7
MTS 7.4.4
The Measuring Line
The slotted measuring line
Sampling the field in front of a short-circuit plate and a waveguide termination
Principles
Various line types for electromagnetic waves
and their technical function
Lines fulfill various functions in radio fre-
quency technology:
(a) They serve in the transmission of radio fre-
quency signals between remote locations.
Examples: radio link between the transmit-
ter and a remote antenna system, broadband
cable for signal distribution of a satellite re-
ceiver system to the individual subscribers
and underwater cables. Transmission lines
and directional radio links in free space are
frequently considered potential alternatives
in this function as a medium for communi-
cations transmission
(b) They also serve as circuit elements (in-
stead of capacitors and inductors) and in-
terconnecting lines for the realization of
passive microwave circuits (e.g. transmis-
sion-line filters).
In general, transmission lines can be consid-
ered as structures for guiding electromagnetic
waves. For this function a multitude of various
transmission line types are suitable.
The coaxial and two-wire line depicted in Fig.
5.1a) guides transverse electromagnetic waves
(TEM waves) and can also be used for any ar-
bitrarily low frequency.
In other transmission line types, however, the
wave propagation is dependent on the condi-
tion that the cross-sectional dimensions of the
line are at least about as large as a half free-
space wavelength (
0
/2). This leads to the result
that in the range of frequencies above approx.
1 GHz there are a larger number of transmis-
sion line types available than for low frequen-
cies. This is because it is only at the higher
frequencies where the dimension
0
/2 results in
handy cross-sections.
Figure 5.1 b) shows various planar transmission
line types, which are particularly well-suited for
the design of microwave integrated circuits
(MIC).
Transmission lines can be realized without any
metal conductors, see Fig. 5.1 d). The dielectric
Fig. 5.1: Various line types for electromagnetic waves
(1 = metal conductor, 2 = isolator).
Numbering from left to right.
(a) Coaxial and two-wire lines
(b) Planar line structures:
Microstrip line,
coplanar line, slotted line
(c) Waveguide: Rectangular, circular and
ridged waveguides
(d) Dielectric waveguides: Round and
rectangular-shaped surface waveguide;
optical waveguide
waveguides shown there are based on the prin-
ciple of total reflection of electromagnetic
waves at the boundary surfaces between the
insulators with higher to lower relative per-
mittivity
r
.
Necessary fundamentals drawn from elemtary
transmission line theory
For the description of line-bound wave propa-
gation we can begin our investigation with a
2 8
MTS 7.4.4 The Measuring Line
simple two-wire line on which the voltage and
current can be specified at any given location.
Subsequently the results can be generally ap-
plied as a model for any other homogenous
transmission line.
Fig. 5.2 shows a homogenous transmission line
structure represented symbolically by two dou-
ble lines. The electromagnetic state on this line
can be expressed by specifying the spatial and
time dependency of the instantaneous voltage
u(x,t) and current i (x,t).
First we will consider the case in which a single
wave propagates on the line in the +x direction.
The following holds true with = 2 f as the an-
gular frequency and
g
as the guided wavelength
(g = guide) and where the attenuation on the
line is disregarded
u x t t
x
, cos ( ) =

2
g
(5.1)
The phase velocity resulting here is
v f
ph
g
g
=

=

2
(5.2)
If Z
0
expresses the line's characteristic imped-
ance, then the following holds true for the corre-
sponding voltage u (x,t) and current i (x, t)
i x t
u x t
Z
,
,
( ) =
( )
0
(5.3)
In the description of time-harmonic processes
you can also make use of complex amplitudes.
As such an AC voltage of the form
u (t) = cos (t + ) can be expressed by a cor-
responding complex amplitude U = exp ( j)
and the following applies
u(t) = Re {U exp (j t)} (5.4)
Complex numbers are now specially marked
by an underline. Equations (5.1) and (5.3) are
given complex numbers with the form
U x U x j x ( ) = ( ) ( )
[ ]
exp
u
(5.5)
where U x ( ) =
Fig. 5.2: Wave propagation on a line
(a) Spatial dependency of the voltage u (x,t)
at two different time points
(b) Spatial dependency of the current i (x,t)
at two different time points
(c) Spatial dependency of the magnitude |U (x)|
and phase
u
(x) for the complex voltage
amplitude (phasor)
(d) Spatial dependency of the magnitude |I(x)|
and phase
I
(x) for the complex current
amplitude (phasor).
(a)
(b)
(c)
(d)
2p
Z
0
t= t d
i(x,t)

0
0
p
p
2p
f
u
(x)
f
I
(x)

dt
t= t d
t = 0
t = 0
Z
0
i(x,t)
|U(x)|
|U(x)|
|I(x)|
u(x,t)
u(x,t)
x
x
2
l
g

Z
0
and

u
g
( ) x
x
x = = 2 (5.6)
and as such the following is true for the current
I x I x j x ( ) = ( ) ( )
[ ]
exp
I
(5.7)
where
I x

Z
x x x ( ) = ( ) = ( ) =
0
und
I u

(5.8)
Figures 5.2 c) and d) show the spatial depend-
ency of the magnitudes |U (x)| and |I (x)| as well
as the phases
u
(x) and
I
(x) of the complex
voltage and current amplitude. In equations
(5.6) and (5.8) the following phase constant has
been introduced
2 9
MTS 7.4.4
The Measuring Line

v
= =
2
g
ph

(5.9)
as a second parameter to express line-bound
wave propagation.
Description of the field in front of a short-
circuit
If there is a short at the end of the transmission
line (x = 0) (see Fig. 5.3, above), the wave can
be described from the superposition of one
component travelling to the load u
+
(x, t) and
one reflected component travelling back from
the load u

(x, t).
As the total voltage at x = 0 must be identical to
zero (short-circuit!), it follows that:
u (x, t) = u
+
(x, t) + u

(x, t)
= cos (t x) cos(t + x)
= 2 sin (t) sin (x) (5.10)
The current of the reflected wave is related to
u

by virtue of (Z
0
) and thus it follows (see Fig.
5.3 b) that
i x t

Z
t x ( , ) cos =

( ) ( )
2
0
cos
(5.11)
The current and voltage now have distinct re-
sponses with respect to time and space. We are
dealing with a standing wave [see Fig. 5.3 c)
and d)]. The nodes and maxima of the u and i
characteristic are at locations unchanged with
respect to time. The nodes of the voltage are
shifted by
g
/4 with respect to the nodes of
the current. Moreover, there is a phase shift of
+ /2 between the voltage and current. This
means that u exhibits its extreme values exactly
when i (for each x) is zero and vice versa. If
these relationships are represented for the mag-
nitude and the phase of the complex amplitude,
then we obtain the results depicted in Figures
5.3 e) and f).
Wave propagation in a rectangular
waveguide
Generally a waveguide is a hollow tube in
which electromagnetic waves can propagate. A
Fig. 5.3: Standing wave on a line shorted on one end:
a) Voltage characteristic u
+
(x, t) of the wave
travelling to the load and u

(x, t) of the
reflected wave travelling back at the time
point t = T/8.
b) i
+
(x,t) and i

(x,t) at t = T/8.
c) Characteristic of the total voltage
u (x, t) = u
+
(x, t) + u

(x, t)
at 4 different points in time
d) i (x, t) = i
+
(x, t) + i

(x, t)
e) Spatial dependency of the magnitude |U (x)|
and phase
u
(x) of the complex
voltage amplitude
f) Spatial dependency of the magnitude |(x)|
and phase
I
(x) of the complex current
amplitude.
rectangular waveguide as shown in Figure 5.4
is a special form of waveguide. This rectangu-
lar waveguide is the object of a series of experi-
ments and should therefore be considered more
closely in theoretical terms. This is also be-
cause many of the results found for rectangular
waveguides also apply to other forms of
waveguides (e.g. circular waveguide).
If you consider the zig-zag reflection of a plane
uniform wave between the side surfaces (dis-
tance a) of a waveguide, you obtain a critical
frequency called the cut-off frequency, as of
which wave propagation becomes possible
3 0
MTS 7.4.4 The Measuring Line
f
c c
a
c
c, 0
= =
2
(5.12)
For the phase velocity a relationship is yielded

ph
0
2
1
2
=

c
a
(5.13)
A calculation of the guided wavelength
g
and
the phase constant amounts to

g
ph
0
0
2
1
2
= =

v
f
a
(5.14)

a
= =

2 2

g 0
0
2
1
2
(5.15)
For more information regarding the derivation
of these relationships we wish to refer to the
specified sources in the bibliography.
Another approach used for describing propa-
gation originates from the solution of Maxwell's
equations with the boundary conditions exist-
ing at the metal surfaces taken into account. But
do not be alarmed. Here only the result of this
analysis is reproduced for the fundamental
mode (TE
10
):
The electrical field has only one Cartesian com-
ponent. It points in the y-direction and so it fol-
lows that:
E x z
a b
U
a
x e
y
jz
, sin ( ) =



2
0

(5.16)
It is evident that E
y
is not only dependent on the
propagation coordinate z, but also on the trans-
verse coordinate x. As the tangential compo-
nents of the electrical field must vanish at the
surface of (ideal) conductors, it is true that
E
y
(x = 0, z) = E
y
(x = a, z) = 0.
Fig. 5.4: Rectangular waveguide
The field strength is maximum in the middle of
the waveguide at x = a/2, and it holds true that
E
a b
U
y 0
2
, max
=


whereby U
0
constitutes an arbitrarily intro-
duced voltage.
Its corresponding magnetic field has both a
transverse component H
x
(x, z) as well as a lon-
gitudinal component H
z
(x, t).
The transverse component H
x
has the same
phase as E
y
and also the same spatial depend-
ency, and so it follows that:
H x z

ab
U
a
x e
x
0
0
jz
2
sin ( , ) =

(5.17)
At every cross-sectional point z = const. the ra-
tio of E
y
to (H
x
) is equal and given by the char-
acteristic impedance
Z
E x z
H x z
a
0
y
x
0
1
=
( )
= =

( , )
( , )

120
0
2
2
(5.18)
The following applies for the longitudinal com-
ponent H
z
:
H x z j
a
ab
U
a
x e
z
z
0
0
j
1 2
cos , ( ) =

(5.19)
3 1
MTS 7.4.4
The Measuring Line
E-plane
E-Plane
T-plane
..
H

g
2
(a)
(b)
..
H

I
Fig. 5.6: Existence zones for various wavetypes capable
of propagating in the rectangular waveguide
(a = 2.25 b).
T-plane
II
I
II
I
Fig. 5.5: Field pattern of the fundamental mode
(TE
10
-wave) in the rectangular waveguide
(a) E and T fields
(b) Current density distribution on the
metal walls
Thus it is phase-shifted by /2 with respect to
the transverse components of E

and H

. In con-
trast to these it assumes at x = 0 and x = a its
maximum value (in terms of magnitude) at the
metal sidewalls. The spatial dependencies of
the field components reproduced in Equations
(5.16) to (5.19) are shown in Fig. 5.5. The wave
type under consideration till now is also re-
ferred to as TE
10
-mode. Here the TE stands
for transverse electric and 1 means that the
number of halfwaves in the x-direction is 1
while the index 0 means that the field is con-
stant in the y-direction. At higher frequencies
higher wave modes, namely TE
mn
-waves and
TM
mn
-waves are capable of propagation.
According to Fig. 5.6 the TE
20
mode is capable
of propagation starting from a frequency of
f
c2
= 2 f
c,TE10
, when there is a side ratio of
a/b = 2.25 so that a frequency octave is avail-
able for single-mode operation.
Fig. 5.7: Principle of the slotted measuring line (exploded
view)
1 Slit in the waveguide
2 Field probe = short rod-type antenna
3 Detector diode
Slotted measuring line
As will be demonstrated with particular care in
experiment 6, a reflection at the end of the
transmission line has the effect that maxima
and minima are formed in the spatial distribu-
tion of the field strength along the line. Based
on the ratio of the amplitude values (maximum/
minimum) and the locations of the maxima and
minima you can draw conclusions as to the
magnitude and phase of the reflection coeffi-
cient.
If to this purpose you wish to measure the dis-
tribution of the field strength along the trans-
mission line (slotted line), the following must
be taken into consideration:
(a) You should interfere as little as possible
with the electromagnetic field in the
waveguide. This requirement is met, if a
narrow slit is added in the center of the
wide section of the waveguide (see
Fig. 5.7 and compare to Fig. 5.5 below).
3 2
MTS 7.4.4 The Measuring Line
Fig. 5.8: Experiment setup
(b) The characteristic impedance Z
0
(see
Equation [5.19]) is slightly altered by this
modification. This can be compensated for
by slightly increasing the waveguide
width a in the region of the slit.
(c) A short probe (electrical dipole) as
shown in Fig. 5.7 supplies a voltage to its
output which is proportional to the trans-
verse component |E
y
| of the electrical field
strength. Thus behind the detector probe
(square-law rectification) you obtain a
voltage
U K E
y D
=
2
(5.20)
Here K is a constant with the dimension

m
2
V
Low reflection waveguide termination
By inserting a wedge-shaped absorber material
the power of the incident wave is nearly com-
pletely absorbed thus suppressing any reflec-
tion almost totally.
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Diaphragm with slit
2 x 15 mm 90 737 22
1 Slotted measuring line 737 111
1 Coax detector 737 03
1 Short-circuit plate, from
accessories 737 29
1 Waveguide termination 737 14
1 Set of thumb screws (2 each) 737 399
Additionally required equipment
1 Oscilloscope (optional) 575 29
1 XY recorder (optional) 575 663
2 Stand bases 301 21
2 Supports for waveguide
components 737 15
1 Stand rod 0.25 m 301 26
2 Coaxial cables with BNC/BNC
plugs, 2 m 501 022
Recommended
1 PIN modulator 737 05
1 Isolator 737 06
3 3
MTS 7.4.4
The Measuring Line
Experiment procedure
Note:
If you are using a PIN modulator and isolator
the experiment setup in Fig. 5.8 is supple-
mented as explained in the preface.
1. Set up the experiment arrangement in ac-
cordance with Fig. 5.8 (perhaps modify
according to Fig. 0.5)
Note:
The diaphragm slit serves as a frequency-
selective component (filter) to improve the
spectral purity of the guided wave.
2. Measurement with short-circuit plate
2.1 Attach the short-circuit plate at the open
end of the slotted measuring line.
2.2 Set the range switch V/dB of the SWR me-
ter to the most insensitive range.
Set the Gunn voltage to between 8 V and
9 V (Tip: first increase it to 10 V and then
adjust it back to the desired value), set the
modulation switch to GUNN-INT.
2.3 Place the slotted measuring line probe at
one end position. Now slowly push the
slide to the other end position and at the
same time search for the maximum detec-
tor voltage (always adjusting the gain to a
suitable level). Calibrate the maximum
value to 0 dB using the ZERO con-
troller. The maximum position is desig-
nated x
0
(read off the scale and note down)
and functions as the reference position.
2.4 Now shift the position of the probe in
2 mm steps (in the direction of the end po-
sition which is farthest removed from x
0
),
in other words to the corresponding posi-
tions |x
n
x
0
| = n 2 mm. Enter the values
measured for the detector voltage into the
second column of Table 5.1.
=

( )
20
0
log
max
U x x
U
U x x
U

( )
0
max
cos
2
0

x x
Display in dB
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
Minimum at
Table 5.1
Probe position
x x
0
in mm
Distance of the Minima
x / mm=_________

g
=_________ mm
3 4
MTS 7.4.4 The Measuring Line
You can conclude the measurement after
exceeding the position of the 2nd maxi-
mum (Display/dB 0).
2.5 Determine the distance x/mm between
two minima and enter the result in the last
line of Table 5.1. (Tip: By successively
increasing the gain (V/dB) in the minima,
you can determine the positions more pre-
cisely).
3. Measurement with reflection-free
waveguide termination.
3.1 The short-circuit plate is replaced by the
reflection-free waveguide termination.
3.2 Slide the probe along the slotted measur-
ing line over the entire range and note
down by which value (in dB) there is de-
viation in the characteristic, i.e. by which
value the level (a) has dropped in com-
parison to the short-circuit experiment.
3.3 As in point 2.3 move the probe over the
entire range and search for the location of
the maximum detector voltage and again
re-calibrate this value to 0 dB.
Display in dB
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
Probe position
x x
0
in mm
=

( )
20
0
log
max
U x x
U
U x x
U

( )
0
max
Table 5.2
3.4 Position the probe in 2-mm steps in the
same locations as in Experiment 2.4 (see
also Table 5.1), and enter the display val-
ues in Table 5.2.
Note:
If you have an XY recorder, a digital stor-
age oscilloscope or a CASSY interface at
your disposal you can also directly record
the measurement curve. For this purpose
the slotted measuring line is equipped with
an integrated displacement sensor.
Connect the IN socket of the slotted meas-
uring line to the X output (supplying |U
G
|)
of the basic unit (or with an external power
supply with 10 V DC).
Connect the X socket of the slotted meas-
uring line to the X input of the XY re-
corder (oscilloscope or CASSY).
Connect the Y input of the XY recorder
(oscilloscope or CASSY) to the AMP-OUT
socket of the basic unit.
The AMP-OUT socket supplies a linear output
signal whereby 0 V is about 20 dB and
approx. 4.5 V corresponds to 0 dB.
3 5
MTS 7.4.4
The Measuring Line
Furthermore, it is important to point out that the
integrated displacement sensor is not com-
pletely backlash-free (it does possess hyster-
esis), for that reason the curve should only be
plotted once in one direction.
Questions
1. Based on the distance between two
neighboring zero points (distance to the
minima) x/mm (see Table 5.1) determine
the wavelength
g
of the guided wave in
mm.
2. Using the value for
g
determined above
under point 1 with the waveguide width of
a = 22.9 mm taken into consideration, de-
termine the free-space wavelength
0
and
frequency f of the guided wave.
3. What is the mathematical value resulting
for the phase velocity v
ph
and the phase
constant according to the Equations
(5.13) and (5.15).
4. What is the cut-off frequency f
c
of the TE
10
mode in the given rectangular waveguide,
and from which frequency is the TE
20
wave capable of propagation? For this
also refer to Fig. 5.6.
5. In addition to the values from the measur-
ing instrument displays determined
through experimentation in experiment
2.4 calculate the ratio of the respective
voltages (field strength levels) to the max-
imum value and enter these in column 3 of
Table 5.1.
Note:
As described in Experiment 2 the detector
voltage is proportional to the square of the
received field strength (or here the voltage
from the transmission line model accord-
ing to Fig. 5.2). You take the logarithm of
the value indicated in the SWR meter dis-
play in accordance with the expression
10 log(U
D
/ U
D,ref
). From this you obtain
the relationship
Display
dB
where
=

( )
=
( )
20
0
0
log
max
max
U x x
U
U U x
6. According to Fig. 5.3 e) the voltage (field
strength) responds accordingly.
U x x
U
x x

( )
=

0
0
2
max
cos

g
[Note:
In Equation 5.10 the sinusoidal term is de-
fined as sin(x). However, here it is per-
mitted to convert to the cosine term be-
cause it only concerns a phase-shift and a
maximum is assumed to exist at x x
0
= 0].
The values resulting from this equation are
entered into column 4 of Table 5.1 and
then they are compared to the values in
column 3. Discuss any possible causes if
you notice regular deviations between the
values in the two last columns.
7. In conjunction with the values determined
experimentally under point 3.4 from the
measuring instrument display, calculate
the ratio of the voltage (field strength) to
the voltage maximum at the respective lo-
cation of the slotted measuring line. Enter
the values into the 3rd column of Ta-
ble 5.2.
8. Discuss the value obtained under experi-
ment point 3.2.
By what magnitude in dB do the measured
values from subpoint 3.4 vary?
3 6
MTS 7.4.4 The Measuring Line
Bibliography
[1] G. Megla: Dezimeterwellentechnik, Berliner Union, 1962
[2] Altmann: Microwave Circuits, Van Norstrand, 1964
[3] F.E. Gardiol: Introduction to Microwaves, Artech House, Dedham 1984
[4] M. Sucher, J. Fox: Handbook of Microwave Measurements. Polytechnic Press,
Brooklyn (NY) 1963
[5] C.G. Montgomery: Techniques of Microwave Measurement. McGraw Hill,
New York 1947
[6] R.E. Collin: Field Theory of Guided Waves. McGraw Hill, New York 1947
[7] R.F. Harrington: Time Harmonic Electromagnetic Fields. McGraw Hill,
New York 1961
[8] S. F. Adam: Microwave Theory and Applications. Prentice Hall, Englewood Cliffs, 1969
[9] H. Meinke, F. W. Gundlach: Taschenbuch der Hochfrequenztechnik. Springer - Verlag
1986
[10] E. Meyer, R. Pottel: Physikalische Grundlagen der Hochfrequenztechnik. Vieweg Verlag,
Braunschweig 1969
[11] H. Groll: Mikrowellen - Metechnik. Vieweg - Verlag, Braunschweig 1969
3 7
MTS 7.4.4
The reflection Coefficient
The complex reflection coefficient
Determining the reflection coefficient according to magnitude and phase
Principles
Voltage curve for random termination impedance
In Experiment 5 two special cases were studied,
namely the case where a line is terminated in a
short-circuit (r = 1) and a line which is termi-
nated with total absorbtion (r = 0). Now we will
describe in more detail the standard case where
the transmission line is generally terminated
with any given complex impedance Z (see Fig.
6.1(a)).
In this case we have the superpositioning of
two waves, a reflected wave U
0,
e
+jx
travel-
ling in the (x) direction and the wave U
0,+
e
jx
travelling in the (+x) direction, whereby the
(complex) reflection coefficient r is defined as
the ratio of the (complex) amplitudes of the
wave being sent back and the wave travelling
to the load at the point of reference (here x = 0):
r r e
U
U
= =

+
j
def
0,
0,

(6.1)
This results in a spatial dependency for the
complex amplitude of total voltage U(x) and
total current I(x) according to
U x U e r e ( ) = +
[ ]
+

0,
jx jx (6.2a)
I x
U
Z
e r e ( ) =
[ ]
+

0
0
,
jx jx
(6.2b)
But due to the fact that the ratio U(x)/I(x) at
x = 0 must be equal to the impedance Z, it is
true that
Z
U
I
r
r
Z =
( )
( )
=
+

0
0
1
1
0
(6.3)
Solving equation (6.3) for r yields the signifi-
cant relationship
r r e
Z
Z
Z
Z
= =

+
j
0
0
1
1
(6.4)
l
l'
Z
0
Z
0
~
(a)
(b)
(c)
I(x)
|U(x)|
|U(x)|
|r|=r
~
|U(x)|
Z
0
x
o
R
U
max
U
min
DX
l
g
2
Z
x
Fig. 6.1: Voltage distribution along a transmission line
(a) Random complex termination impedance Z
(b) Equivalent circuit for Z (R and line with a
length of l)
(c) Shorted transmission line
If we consider the distribution of the magnitude
|U(x)| along the transmission line, then it fol-
lows from equation (6.2 a) that
U x U r r x ( ) = + + + ( )
+ 0
2
1 2 2
,
cos
(6.5)
Part a of Fig. 6.1 shows the voltage distribu-
tion, where it is true that
U U r
U U r
max 0,
min 0,
1
1
= +
=
+
+
( )
( )
and
(6.6)
The voltage standing wave ratio (VSWR = s) is
given by:
s
U
U
r
r
= =
+

def
max
min
1
1
(6.7)
3 8
MTS 7.4.4 The reflection Coefficient
and
r
s
s
=

+
1
1
(6.8)
Determining the complex reflection coefficient
from the measurements performed with the
slotted measuring line
According to Eq. (6.8) the magnitude |r| of the
reflection coefficient can be calculated from the
standing wave ratio s determinable using the
slotted measuring line.
The phase of the reflection coefficient can be
determined by comparing the location x
1
of the
minimum value of |U(x)| to the location x
0
of
the zero point or minimum when the transmis-
sion line is shorted.
In this context first consider Fig. 6.1 b). Here
the voltage distribution is extrapolated beyond
location x = 0 up to the next maximum value.
In conjunction with this maximum value there
is a positive real resistance R > Z
0
where
R
r
r
Z =
+


1
1
0
(6.9)
The line segment with the length l, terminated
with R, has the same input impedance Z as the
original circuit. Thus, for the phase it is true
that

= =
4
2
g
l l (6.10)
On the other hand, due to l + l =
g
/4 (distance
of the minimum to the maximum) and
x + l =
g
/2 (generally: x + l = n
g
/2) the
following relationship applies for l
l x n = ( ) 2 1
4
g

(6.11)
and therewith

/
rad
g
4 2 1 = + ( )
x
n or

= +
=
720 180
0, 1, 2,...
g
x
n ( ) (6.12)
Fig. 6.2: For the determination of s and |r| from the
measurement of the node width l.
To determine you can proceed as follows:
(a) You determine a location x
0
for the zero
point or minimum when the transmission
line is shorted.
(b) Afterwards you determine a location x
1
for
the minimum when the line is terminated
with an unknown impedance.
(c) x constitutes the shift from x
1
with re-
spect to x
0
. For this x is counted as posi-
tive, if you must shift from the minimum
in the direction of the termination (Z) (see
also Fig. 6.1).
The determination of the magnitude |r| out of s,
i.e. the ratio of the maximum to the minimum
amplitude, becomes inexact as soon as the am-
plitude ratio (= s) becomes somewhat greater
than 5. It is here namely that slight measure-
ment errors arise through the overload of the
measuring amplifiers, deviation in the diode
characteristic when measuring the maximum
and on account of noise during the measure-
ment of the minimum. These errors can be
avoided to a great extent, if instead of
measuring the voltage quotients s = U
max
/U
min
you determine the width of the voltage
minimum (node width). If according to Fig. 6.2
you express the so-called node width at
U x U ( ) /
min
= 2 (corresponding to 3 dB) with
l, the exact result is:
1
sin
1
g
2
g
s
l
l
=

sin
(6.13)
3 9
MTS 7.4.4
The reflection Coefficient
Fig. 6.3: Complex number plane
(a) the reflection coefficient r = | r | exp(j) and
(b) for the normed impedance Z/Z
0
= R/Z
0
+ j X/Z
0
.
: Short-circuit : Matching
Fig. 6.4 Smith chart
For l/
g
<< 1 this can be approximated with
sin << 1 for << 1
yielding
1
g
s
l

(6.14)
The value |r| is extrapolated from s using Equa-
tion (6.8).
Smith chart
The reflection coefficient r and the normalized
impedance Z/Z
0
are complex numbers depicted
accordingly in separate number planes in Fig.
6.3. If according to Equation (6.3) the corre-
sponding values of Z/Z
0
for the reflection coef-
ficient r are entered into the complex number
plane (Fig. 6.3 a) and this is done in the form of
lines of the constant real component R/Z
0
and
constant imaginary component X/Z
0
, you ob-
tain the Smith Chart depicted in Fig. 6.4. It is
well suited for the rapid determination of Z/Z
0
for a given value of r (and vice versa). Fur-
thermore, it is very useful as explained in
more detail in Experiment 7 for describing
various properties of transmission line net-
works. In Fig. 6.4 the reflection coefficient is
entered for Z/Z
0
= 0.5 + j 1.4. By dimensioning
the length |r| relative to the radius of the outer
circle you obtain |r| 0.72 and by measuring
the angle you arrive at 67. The determina-
Z
Z
0
Re
Z
Z
R
Z
0 0

=
j im
Z
Z
j
x
Z
0 0

=
tion of Z/Z
0
at a given value of r is performed in
a similar manner.
The Smith chart can also be used for admit-
tances Y Z
0
= Z
0
/Z. The corresponding complex
number is then (r). Then if it is true that
Y/Y
0
= Y Z
0
= 0.5 + j 1.4, the corresponding re-
flection coefficient is:
0.72 exp [ j (67180)] = 0.72 exp ( j 113).
4 0
MTS 7.4.4 The reflection Coefficient
Note:
As is shown in Experiment 3 of MTS 7.4.5,
any random reflection coefficient (load) can
be realized using the combination (series con-
nection) of a variable attenuator, a variable
phase-shifter and a short-circuit plate. Here the
attenuator primarily influences the magnitude
and the phase-shifter affects the phase-angle of
the reflection coefficient.
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Diaphragm with slit
2 x 15 mm 90 737 22
1 Slotted measuring line 737 111
1 Coax detector 737 03
1 Short-circuit plate,
from accessories 737 29
1 Sample holder, from accessories 737 29
1 Sample made of graphite(*),
from accessories 737 29
1 Waveguide termination 737 14
1 Set of thumb screws (2 each) 737 399
Additionally required equipment
1 Oscilloscope (optional) 575 29
1 XY recorder (optional) 575 663
2 Coaxial cables with BNC/BNC
plugs, 2 m 501 022
2 Stand bases 301 21
2 Supports f. waveguide components 737 15
1 Stand rod 0.25 m 301 26
Recommended
1 PIN modulator 737 05
1 Isolator 737 06
Experiment procedure
1. Set up the experiment configuration as
specified in Fig. 6.5.
Note:
The diaphragm with slit again serves as a
filter to improve spectral purity.
2. Calibration
Screw on the short-circuit plate to the end
the slotted measuring line to function as
the device under test (DUT). Determine
the location x
0
/mm of the first minimum
(zero point) in front of the short-circuit
plate and enter the findings in Table 6.1.
3. Measurement of the sample
3.1 Assemble a measurement object A con-
sisting of a sample holder with graphite
material sample (see Fig. 6.5) and a reflec-
tion-free waveguide termination. This
measurement object is fastened onto the
open end of the slotted measuring line.
3.2 Determine the standing wave ratio
(VSWR) s with the help of the scale on the
frequency-selective voltmeter. To do this,
calibrate in the voltage maximum to
0 dB using the range switch and the
(*) : In order to be able to compare the measured values of this experiment with those from Experi-
ment 11, the samples used here should be stored so as not to mix them up with any other exist-
ing samples.
Fig. 6.5 Experiment setup
4 1
MTS 7.4.4
The reflection Coefficient
ZERO controller. Shift the position of
the slotted measuring line probe to the first
minimum in front of the load. This is
where you read off the VSWR value and
enter it into Table 6.1.
3.3 Determine the location x
1
of the minimum
and enter this value into Table 6.1
3.4 Preliminary evaluation:
Determination of |r| from s with the aid of
Equation (6.8) and extrapolation of from
x = (x
1
x
0
) and Equation (6.12). Enter the
values into Table 6.1.
3.5 Determining |r| by means of the node
width l. For this proceed with measure-
ment techniques corresponding to Fig. 6.2
to determine the node width (3 dB beyond
U
min
). Calculate s according to Equation
(6.13) and |r| according to Equation (6.8).
Enter the findings into Table 6.2
Tip:
Adjust the value of the minimum to 3 dB
using the gain selection switch and the
ZERO controller, and find the positions
(to the left and right of the minimum),
where the display assumes a value of 0 dB.
Questions
1. Determine the value of the normalized im-
pedance Z/Z
0
for the measurement object
A (see Table 6.1) using the Smith chart
(approximately with Fig. 6.4 or more ac-
curately with Fig. 7.7 of the subsequent
experiment).
Table 6.2
l/mm
s |r|
Measurement
object A
Table 6.1
DUT
Location of the
Minimum/mm
s |r|
Short-circuit plate 1 180
Measurement
object A

/degree
4 2
MTS 7.4.4 The reflection Coefficient
Bibliography
[1] S.J. Algeri, W.S. Chiung, L.A. Stark: Microwaves Made Simple -The Workbook.
Artech House, Norwood 1986
[2] G.J. Wheeler: Introduction to Microwaves. Prentice Hall, New Jersey
[3] T. Moreno: Microwave Transmission Design Data. Dover Publ., New York
[4] C.G. Montgomery: Technique of Microwave Measurements. McGraw Hill
[5] S. F. Adam: Microwave Theory and Applications. Prentice - Hall, Englewood Cliffs, 1969
[6] T. S. Laverghetta: Modern Microwave Measurements and Techniques.
Artech House, Norwood 1988
[7] T. A Johnk: Engineering Electromagnetic Fields and Waves. Wiley & Sons,
New York 1975
4 3
MTS 7.4.4
The Smith chart
Matching and the Smith chart
Principles
Preliminary considerations
If the input impedance Z of a one-port is not in
agreement with the characteristic impedance Z
0
of the transmission line this results in a reflection
characterized by the reflection coefficient
r r e
Z
Z
Z
Z
= =

+
j
0
0
1
1
(7.1)
For various reasons this reflection is undesirable
for most applications. Some of the reasons are
listed here:
(a) If the transmission line is fed from a gen-
erator with the available power P
av
and this
generator is matched to the line (see Fig-
ure 7.1, part 1), then the real power ab-
sorbed by the load is
P = P
av
(1 |r|). (7.2)
This shows that the maximum power is
only supplied to the load when it is per-
fectly matched; (|r| = 0).
(b) Because of non-linear effects, the exist-
ence of a wave returning to the generator
can lead to changes in the operation char-
acteristics of the generator, e.g. to a fre-
quency shift and to parasitic oscillations at
different frequencies (fulfilling the condi-
tion for self-excitation through reflection).
(c) Compared with the case of |r| = 0 an en-
hanced field strength (standing wave) is
caused by the interference of the reflected
wave with the one travelling forward.
Thus, the danger of disruptive discharges
(high electric field strengths) is associated
with the transport of relatively high power
levels.
The danger described in point (c) does not exist in
this training system because only low power lev-
els are used here.
Basic function of a lossless matching element
and the matching condition
For a given frequency one can generally match
any one-port with |r| < 1 to the characteristic
impedance Z
0
by positioning a two-port (=
matching network) upstream in series between
the reflecting one-port and the transmission line.
If a lossless two-port (ideally) is used for the
matching network, then its general function is to
compensate the given reflection coefficient r to
zero by adding additional reflections. Section 2 of
Fig. 7.1 initially demonstrates the general situa-
tion. By connecting the network N (two-port) in
series upstream, the reflection coefficient r is
transformed into the reflection coefficient
~
r .
The aim is to achieve correct matching by select-
ing the right parameters of the lossless linear
two-port, i.e. to achieve
~
r = 0.
The condition to be fulfilled here (matching con-
dition) can be deduced in general, that means
without any special knowledge regarding the con-
struction of the two-port N. Here a consideration
of only one parameter of the two-port is suffi-
cient; namely the backward reflection coefficient
, the definition of which is shown in section 3 of
Fig, 7.1. Here, it is assumed that port 1 (1-1') of
the two-port is terminated reflection-free (i.e.
the internal impedance of the source is Z
0
). is
then the effective reflection coefficient felt by
a wave travelling to port 2 (2-2') from right to
left. (Bear in mind that r applies to a wave arriv-
ing from the left.)
You can demonstrate that it is sufficient for
matching when is chosen to be the conjugate
complex value of of the reflection coefficient r
to be matched (see below):
= r* = |r| e
j
(matching condition)
(7.3)
Section 4 of Fig. 7.1 shows the resulting situation
for this specific case. At port 1 the reflection re-
sults in
~
r = 0 (matching). Between port 2 and the
mismatched (|r| 0) one-port you obtain the
superpositioning of a wave with the power
P
av
/(1 |r|) travelling to the one-port while the
reflected wave has the power P
av
|r|/(1 |r|).
4 4
MTS 7.4.4 The Smith chart
The total power results from the difference of
the power of both waves and equals P
av
. Thus, as
desired, the total available power is provided to
the one-port (for a lossless matching network).
Matching element according to the principle of
the slide screw transformer
According to the findings from the previous
section there are two different ways to explain
how a matching element works:
(A) The explanation dealing with the transfor-
mation of a given reflection coefficient
r = |r| e
j
to the value
~
r = 0 using a match-
ing element.
(B) The explanation involving the setting
= |r| e
j
of the matching element con-
nected reflection-free at the input-side.
According to Figure 7.2 a slide screw trans-
former consists of a homogeneous waveguide
section designed with an obstacle which can
be adjusted in terms of location (variable posi-
tion setting
0
) and magnitude. If the obsta-
cle is in the form of a metal post as in Figure
7.2 (left) with an adjustable penetration depth of
h < b, this can be represented in the equivalent
circuit diagram by a shunt capacitance C (see
Fig. 7.2, right) as long as h is sufficiently small
compared to the height b of the waveguide. C is
zero for h = 0 and increases with h. If h is only
slightly smaller than b, one obtains a series
resonant circuit, whereas, if the post touches the
opposite side, it responds like a shunt inductance.
In the case of the slide screw transformer here
we can always assume that we are dealing with a
shunt capacitance.
Now we shall first consider the transformation of
a random reflection coefficient r = |r| e
j
in the
Fig. 7.1: General considerations regarding the problem of
matching
1 Transmission line with matched generator
and mismatched load.
Corresponding complex amplitudes from
wave travelling to and reflected from load.
2 Series connected two-port
transforms reflection coefficient r
into the value
~
r .
3 On the definition of the backwards
reflection coefficient G of the two-port
4 Ratios for matching, i.e.
two-port N is the ideal non-dissipative
matching network.
f: Phase displacement through Z
g: Phase displacement through the
matching network
Fig. 7.2: Technical construction of a slide screw transformer (left) and
corresponding equivalent circuit diagram (right).
P
av
r e P
j
av

P
av
~
r P
av
j P
e e
r
av
j
2
j
2
1

j P
r e e
r
av
j
2
j
2
1

P
av
4 5
MTS 7.4.4
The Smith chart
matching point
~
r = 0 (Explanation A according
to above case distinction).
In Fig. 7.3 point 1 specifies the random reflec-
tion coefficient r = |r| e
j
(in the example:
|r| = 0.605 and = 210). To be able to read
the admittances out of the Smith chart (expedi-
ent for parallel circuits), the transition to r
(point 2) is carried out through inversion at
the matching point (r = 0). In the example
Y/Y
0
= 2 + j 1.9 of the normalized admittance
belongs to the value of r (for practice please
verify in Fig. 7.3).
Based on this preliminary step it is now easy to
explain the function of the slide-screw trans-
former. Only the phase of the reflection coeffi-
cient but not the magnitude is changed by the
waveguide section with the length
0
(phase
rotation by the angle, 720
0
/
g
in clockwise
Fig. 7.3: Matching of a random reflection coefficient r = |r| e
j
(where |r| < 1)
with the aid of a slide screw transformer:
1 Reflection coefficient r to be matched
2 Point for r (admittance representation)
3 Negative reflection coefficient r
1
after transformation by the waveguide
section of the arbitrary length
0
.
4 Negative reflection coefficient
~
r (resp.
~
Y /Y
0
) after parallel connection of a
positive susceptance B of random magnitude.
Matching case: 5 and 6represent r
1
and
~
r for the correct selection
0
=

0
and B = B for the line length and susceptance.
2
.0
3
.0
0 , 3 5
0
,
2
3
0
,
2
2
0
,
1
9
0
,
1
8
0
,
1
7
0
,1
6
0
,1
5
0,14
0,13 0,12
0,11
1
.
0
4.0
5.0
2
.
0
3
.
0
4
.
0
5
.
0
0
0
,
2
5
1
0
0
,
2
4
2
0
3
0
4
0
5
0
6
0
7
0
80
90
100
1
1
0
0
,1
0
1
2
0
0
,
0
9
1
3
0
1
4
0
1
5
0
1
6
0
1
7
0
1
8
0
0
1
9
0
j
j
0
,
4
9
0
,
0
0
0
,
0
1
0
,
0
2 G
e
n
e
r
a
t
o
r
0
,
0
3
0
,
0
4
0
,
0
5
0
,
0
6
0
,
0
7
0
,
0
8
2
0
0
2
1
0
2
2
0
2
3
0
2
4
0
2
5
0
2 6 0
2 7 0
0 , 3 8
0 , 3 9
0
, 4
0
0
, 4
1
0
,
4
2
0
,
4
3
0
,
4
4
0
,
4
5
0
,
4
6
0
,
4
7
L
o
a
d
0
,
4
8
2 8 0
2
9
0
3
0
0
3
1
0
3
2
0
3
3
0
3
4
0
0
,
2
8
0
,
2
9
0
,
3
0
0
,
3
1
0
,
3
2
0
,
3
3
0
, 3
4
0
, 3
5
0 , 3 7
3
5
0
0
,
2
7
0
,
2
6
0
.
2
1
0
,
2
0
l
Smith Chart
0
.
8
0
.
6
0
.
5
0
.
4
0
.
3
0
.
2
0
.
1
20
10
1
0
2
0
1
.
5
+
o
Z X
o
Z
R
0.2
0.4
0.6
1,0
0.8
0 . 2
0 . 2
0 . 4
0
, 6
0
, 8
1
, 0
0
, 4
0
,
6
0
,
8
1
,
0
0
.4
0.2
0
.6
0
.8
1
,0
1
.
5
1
.
5
10
2
0
0 . 1
0 . 1
0
. 2
0 . 2
0
.
4
0
. 4
0
.
5
0
. 5
0
.
6
0
.
6
0
.
7
0
.
7
0
.
8
0
.
8
0
.
9
0
.
9
1
,
0
1
,
0
0
. 3
0 . 3
2
.
0
3
.
0
4
.
0
5
.0
X Z
o
-
l
F
5
3
l
g
x
0
720
l
g
x
0
720
Y
0
B
r
r
r
Y
0
B
Y
0
Y
4
6
1
2
r
~
1
j
j
4 6
MTS 7.4.4 The Smith chart
direction). Thus, one obtains the value of r
1
given by point 3 in Figure 7.3. The parallel con-
nection of j B corresponds to the addition of an
imaginary admittance and consequently the
value of the reflection coefficient changes into
~
r (point 4) along the locus of a constant real
component (in this example Re (Y/Y
0
) = 0.4).
In order to achieve matching, i.e. r = 0, the
length
0
=
0
must be selected in accordance
with Figure 7.3 so that r
1
is located on the locus
Re (Y/Y
0
) = 1 (point 5 where
0

0.118
g
).
Under this prerequisite there exists a value
B = B for which
~
r is located at the matching
point (point 6) (in the example: B/Y
0
1.5).
Predicated on the explanations and descriptions
found in the previous section an alternative ex-
planation suggests itself for the function of the
slide screw transformer (explanation B accord-
ing to the case distinction above). In this pres-
entation we will be referring to the circuit found
on the left of the reference plane (Symbol ) and
not to the circuit found on the right as above. If
the slide screw transformer is terminated reflec-
tion-free at port (1- 1' in Fig. 7.4), then
~
= 0
(see Fig. 7.4) also applies to the left of the shunt
capacitance.
1
is obtained to the right of the
shunt capacitance C, whereby the magnitude
|
1
| can be altered by varying C (penetration
depth of the post) between 0 and values approx-
imating 1. Here the phase of
1
also changes,
but as a function of the magnitude |
1
|. A phase
variation independent of the magnitude is possi-
ble by changing the location
0
. As such the phase
condition for matching specified above
= r* = |r| e
j
can always be fulfilled by varying C and
0
.
Matching element according to the principle of
the multi-screw transformer (2- or 3-screw
transformer)
There are two variable parameters in the slide
screw transformer dealt with above, namely the
position
0
and the penetration depth h of the
post (screw). In the case of multi-screw trans-
formers we dispense with an adjustment of the
screw position (x-coordinate). To realize this
the number of screws is greater than 1 and each
of the two or three screws can be adjusted inde-
pendently of each other in terms of their penetra-
tion depth.
The lower part of Fig. 7.5 shows the equivalent
circuit diagram for a 2-screw transformer. The
positions (
0
and ) of the screws are fixed but
the penetration depth and therefore both
susceptances (B
1
and B
2
) can be adjusted inde-
pendently.
The negative value r of the reflection coeffi-
cient at the end of the transmission line (r) is
given by point 1 (bear in mind: the same value
as in Fig. 7.3). The waveguide section of the
fixed length
0
transforms r into point 2 (same
magnitude, phase rotated by [
0
/
g
] 720 in
the clockwise direction). By means of a parallel
connection of j B
1
one obtains a value corre-
sponding to point 3. The waveguide section of
fixed length transforms point 3 into point
4 (same magnitude, rotation of the phase by
(/
g
) 720 in the clockwise direction) and fi-
nally the parallel connection of j B
2
transforms
point 4 into point 5. In order for point 5 to be
located in the matching point (r = 0), B
1
must be
Fig. 7.4: How the slide screw transformer works (alternative explanation to Fig. 7.3).
=

1
4
0
0 e
j

= r* =|r| e
j
4 7
MTS 7.4.4
The Smith chart
selected precisely so that point 3 is located on
the auxiliary circle K. The auxiliary circle K is
obtained from circle K (Re {Y/Y
0
} = 1) by rota-
tion around the matching point (rotation angle
[[/
0
] 720 counterclockwise). As you can
conclude from Figure 7.5, not every reflection
coefficient |r| < 1 can be matched using the 2-
Fig. 7.5: Matching using a 2-screw transformer.
screw transformer.
However, this can be achieved using the 3-
screw configuration. But in this case there are
several solutions for B
1
, B
2
and B
3
for a given
value of r. Consequently, we shall dispense
with a more detailled theoretical investigation
of the 3-screw transformer.
l
g
x
0
720
l
g
Dx
720
G
e
n
e
r
a
t
o
r
L
o
a
d
2
.0
3
.0
0 , 3 5
0
,
2
3
0
,
2
2
0
,
1
9
0
,
1
8
0
,
1
7
0
,1
6
0
,1
5
0,14
0,13 0,12
0,11
1
.
0
4.0
5.0
2
.
0
3
.
0
4
.
0
5
.
0
0
0
,
2
5
1
0
0
,
2
4
2
0
3
0
4
0
5
0
6
0
7
0
80
90
100
1
1
0
0
,1
0
1
2
0
0
,
0
9
1
3
0
1
4
0
1
5
0
1
6
0
1
7
0
1
8
0
0
1
9
0
j
j
0
,
4
9
0
,
0
0
0
,
0
1
0
,
0
2
0
,
0
3
0
,
0
4
0
,
0
5
0
,
0
6
0
,
0
7
0
,
0
8
2
0
0
2
1
0
2
2
0
2
3
0
2
4
0
2
5
0
2 6 0
2 7 0
0 , 3 8
0 , 3 9
0
, 4
0
0
, 4
1
0
,
4
2
0
,
4
3
0
,
4
4
0
,
4
5
0
,
4
6
0
,
4
7
0
,
4
8
2 8 0
2
9
0
3
0
0
3
1
0
3
2
0
3
3
0
3
4
0
0
,
2
8
0
,
2
9
0
,
3
0
0
,
3
1
0
,
3
2
0
,
3
3
0
, 3
4
0
, 3
5
0 , 3 7
3
5
0
0
,
2
7
0
,
2
6
0
.
2
1
0
,
2
0
l
0
.
8
0
.
6
0
.
5
0
.
4
0
.
3
0
.
2
0
.
1
20
10
1
0
2
0
1
.
5
+
o
Z X
o
Z
R
0.2
0.4
0.6
1,0
0.8
0 . 2
0 . 2
0 . 4
0
, 6
0
, 8
1
, 0
0
, 4
0
,
6
0
,
8
1
,
0
0
.4
0.2
0
.6
0
.8
1
,0
1
.
5
1
.
5
10
2
0
0 . 1
0 . 1
0
. 2
0 . 2
0
.
4
0
. 4
0
.
5
0
. 5
0
.
6
0
.
6
0
.
7
0
.
7
0
.
8
0
.
8
0
.
9
0
.
9
1
,
0
1
,
0
0
. 3
0 . 3
2
.
0
3
.
0
4
.
0
5
.0
X Z
o
-
l
j
Y
0
B
1
j
Y
0
B
2
1
2
3
4
5
r
K
K'
l
g
Dx
720
4 8
MTS 7.4.4 The Smith chart
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Diaphragm with slit
2 x 15 mm 90 737 22
1 Isolator 737 06
1 PIN modulator 737 05
1 Slotted measuring line 737 111
1 Coax detector 737 03
1 Cross directional coupler 737 18
1 Transition waveguide / coax 737 035
1 Waveguide 200 mm 737 12
1 3-screw transformer 737 135
1 Short-circuit plate, from
accessories 737 29
1 Sample holder, from accessories 737 29
1 Graphite sample, from accessories737 29
Fig. 7.6 Experiment setup
2 Waveguide terminations 737 14
1 Set of thumb screws (10 each) 737 399
Additionally required equipment
1 Oscilloscope (optional) 575 29
1 XY recorder (optional) 575 663
3 Coaxial cables with BNC/BNC
plugs, 2 m 501 022
2 Stand bases 301 21
2 Supports for waveguide
components 737 15
1 Stand rod 0.25 m 301 26
Recommended
1 Slide screw transformer 737 13
1 Set of thumb screws (2 each) 737 399
4 9
MTS 7.4.4
The Smith chart
Notes:
In the experiment a cross directional coupler
is used (Coupling diaphragm with 2 cross-
shaped holes) to measure the reflected wave.
The cross directional coupler is an arrange-
ment made up of 2 waveguides, which are
connected together vis--vis coupling holes.
A portion of the travelling and reflecting
wave can be detected at the output ports of
the coupling waveguide. The exact function
of the cross directional coupler is explained
in Experiment 10.
The coax detector is connected alternately
to the slotted measuring line and the cross
directional coupler. Naturally, it is also
possible and even recommended to use
two detectors.
Since this experiment responds with par-
ticular sensitivity to parasitic modes you can
only expect reproducible results when using
a PIN modulator (see also Design of the
Microwave Source from the preface).
Experiment procedure
1. Calibration
1.1 Set up the experiment in accordance with
Fig. 7.6. First attach the short-circuit plate
to the open end of the line.
1.2 Use the slotted measuring line to deter-
mine the location x
0
of the first (right)
field strength minimum.
Enter this value into Table 7.1.
1.3 Replace the short-circuit plate with the
measurement object A from Experiment 6,
consisting of a sample holder with graph-
ite sample and waveguide termination and
determine the location of the first (again
from the right) minimum x
1
/mm and en-
ter the value into Table 7.1 a).
2. Matching using the 3-screw transformer
2.1 Insert the 3-screw transformer between the
reflecting one-port (measurement object A
from Experiment 6) and the 200 mm
waveguide.
2.2 Connect the coax-detector to the cross di-
rectional coupler to measure a portion of
the reflected microwave. Here be sure to
select a suitable gain factor V/dB.
2.3 Make an attempt to achieve matching by
adjusting all three screws one after the
other in succession. This means a mini-
mizing of the reflected microwave power
(ideally: no reflected wave). Here the best
procedure is to find the screw which has
the greatest effect on the reflected wave. To
do this turn one of the three screws (in-
creasing the penetration depth) while
keeping the other two respective screws
set to a penetration depth of 0. Then set the
screw with the greatest effect to its opti-
mum setting (local minimum of the re-
flected wave) and then turn the other
screws in succession to further diminish
the reflected wave. At the same time you
have to suitably adapt the gain factor to be
able to determine the optimum setting of
the matching element.
Note: If you are unable to obtain a distinct
response for any of the screws, then it
might prove useful to turn the 3-screw trans-
former in the configuration (i.e. interchang-
ing the ports), because it is designed slightly
asymmetrically. This gives you a different
phase angle for the screws.
2.4 Now replace the coax detector again with
the slotted measuring line and check the
transforming effect on the standing wave
ratio (reset the maximum to 0 dB). The
standing wave ratio should tend towards the
value 1 (s 1, i.e. |r| 0, ideally:
s = 1, |r| = 0). You have attained good
matching if you reach a standing wave ratio
under s < 1.1. In order to rule out any in-
fluence of the cross directional coupler on
the reflection coefficient, you can also con-
nect the matching transformer plus the
measurement object directly to the slotted
measuring line (i.e. remove the cross direc-
tional coupler and the waveguide section).
2.5 Afterwards measure the backward reflec-
tion coefficient of the 3-screw trans-
former without any further adjustment to
the screws. To do this
() Remove the reflecting single-port.
() Reverse the connection of the 3-screw
transformer (reversing the ports) to the
slotted measuring line.
() Equip the open end of the 3-screw
transformer with the reflection-free
waveguide termination (see also
Fig. 7.6).
Read off the standing wave ratio s and the
value of x
1
. Enter the findings in Table 7.1 b).
5 0
MTS 7.4.4 The Smith chart
3-screw transformer
Table 7.1
Short-circuit plate x
0
= ________ mm
a) Measurement object without 3-screw transformer
VSWR
s
b) Calibrate to s 1, i.e. |r| 0 by successively turning all the screws
Backward reflection coefficient (see Fig. 7.1)
VSWR
s
3. Matching using the slide screw transformer ( optional )
Minimum at
x
1
/mm
r
s
s
=

+
1
1

=

180 720
1 0
x x
g
3.1 In accordance with Fig. 7.6 insert the slide
screw transformer between the reflecting
one-port (measurement object A from Ex-
periment 6) and waveguide 200 mm.
Connect the coax detector to the cross di-
rectional coupler.
3.2 Minimize the reflected wave by succes-
sively adjusting the longitudinal position
x and the penetration depth h. In general
there are several combinations you can use
to achieve matching.
3.3 Again mount the coax detector on the slot-
ted measuring line and check the effect on
the standing wave ratio (reset the maxi-
mum to 0 dB). You have again reached
good matching if s < 1.1 (i.e. r< 0.05). To
rule out any influence from the cross di-
rectional coupler on the reflection coeffi-
cient, you can once again connect the
matching transformer plus measurement
object to the slotted measuring line (i.e.
remove the cross directional coupler and
the waveguide section).Use a table similar
to table 7.1a.
3.4 The backward reflection coefficient of
the slide screw transformer is determined
for matching. The measurement procedure
is analogous to that shown in experiment
point 2.5. Determine the standing wave
ratio s and x
1
. Use a table similar to Table
7.1b.

=

180 720
1 0
x x
g
=

+
s
s
1
1
Minimum at
x
1
/mm
5 1
MTS 7.4.4
The Smith chart
Questions
1. For measurement object A calculate the val-
ues of r = |r| e
j
corresponding to the meas-
urement data.
2. Determine the complex value of the
backward reflection coefficient of the 3-
screw transformer based on the measure-
ment data in Table 7.1 b). Check the
validity of Equation (7.3).
(Optional)
3. Determine the complex value (| | and

)
of the backward reflection coefficient from
the measurement data. Test the validity of
Equation (7.3) by comparing the value de-
termined for with the value of r. Com-
pare the value determined for using the
slide screw transformer with the respective
value determined using the 3-screw trans-
former.
Bibliography
See Experiment 5
5 2
MTS 7.4.4 The Smith chart
Fig. 7.7: Smith chart
5 3
MTS 7.4.4
Reflection
Reflection of a single-slot antenna
Fundamentals
Preliminary remarks
In experiments 6 and 7 the partially absorbing
and reflecting one-port is used as an example
for an incorrectly matched load and its
matching with the aid of a matching network.
In the present experiment we will consider a
case which frequently occurs in practice,
namely the mismatched antenna.
An antenna can generally be considered a spe-
cial form of a wave-type converter. In the case
of a transmitting antenna, the power P
in
sup-
plied by the generator arrives in the form of a
guided wave at the input of the antenna. The
function of the antenna is to convert the power
P
in
as completely as possible into a free-space
wave (with a given directional dependency =
directional pattern). An incomplete conversion
(P
rad
< P
in
) results from a reflection at the an-
tenna input (r) and/or dissipation (efficiency
< 1) within the antenna (conversion into ther-
mal energy). The radiated power P
rad
resulting
is
P r P
rad in
=
( )
1
2
(8.1)
In the case of a receiving antenna, it supplies
receiving power to the load or consumer
equalling
P r P
rad rec,max
=
( )
1
2
(8.2)
whereby P
rec, max
is the maximum available re-
ceiving power.
From equations (8.1) and (8.2) you can discern
that incorrect matching (|r| > 0) leads to both a
reduction in the radiated power as well as a
drop in the power actually received (= reduc-
tion of the signal-to-noise ratio).
In the following section we will look into this
special type of antenna, namely the slot an-
tenna.
Principle of the slot antenna
For reasons of a didactic nature we shall now
expound on the principle of the slot antenna
step by step.
Here we will first consider a plane metal plate
of infinite extent, into which a slot (width h, in-
finitely long) has been mounted as specified in
Figure 8.1. Such a structure represents a special
form of a double transmission line (slotted
transmission line), on which guided waves can
propagate in both directions (parallel to the
slot). The field pattern of the guided wave of the
slot is depicted in part 1 of Figure 8.1.
It is assumed in the next step of our explanation
that the slot according to Fig. 8.1 (part 2, left)
only has a finite length w. Thus, it represents a
transmission line resonator short-circuited on
both ends. According to theory, without any
excitation a field can only exist on a transmis-
sion line of the length w, when w is an integer
multiple of
g/2
(here equal to a free-space wave-
length of
0
). Hence, the following results for
H
E h
h
w
|E
0
|
|E|
E
(1) (2) (3)
Fig. 8.1: On the principle of the slot antenna:
(1) Field pattern of the wave in a slotted line
(2) Slot guide resonator (= at the end of the shorted slotted guide)
and corresponding distribution of the electric field strength along the slot
(3) Slot antenna fed via waveguide
left: Slot perpendicular to the electrical field lines of the TE
10
fundamental mode
right: Slot at an oblique angle to the electrical field lines of the TE
10
mode
5 4
MTS 7.4.4 Reflection
the lowest resonance frequency (w = l
0
/2, f = c/
l
0
):
f
c
w
0
2
=

(8.3)
The part on the right of Fig. 8.1, part 2, shows
the distribution of the electrical field strength
along the slot. The maximum at the middle of
the slot is denoted

E
0
.
In the case of external excitation of the electro-
magnetic field in the slot, there results a non-
vanishing field for each excitation frequency f
where

=

+
( )
+


E
j Q
f f
f
E
0
0
0
0
1
2
1 2

c
c
(8.4)
Here

E
1
stands for the amplitude of the ex-
cited field (presupposing a suitable definition).

c
is the coupling coefficient and Q
0
is the un-
loaded Q of the transmission line resonator.
c
and Q
0
are among other things functions of the
slot width h.
The dependency of interest here is that for a
given coupling coefficient
c
(degree of exci-
tation)

E
0
reaches maximum at f = f
0
and is se-
verely reduced, if f deviates radically from f
0
.
There are various possibilities for feeding the
slot antenna. Frequently a coaxial line is con-
nected in a suitable form to the slot (see the rel-
evant literature in the bibliography).
Here we will study the case of a waveguide
feed. Fig. 8.1 (part 3) demonstrates this princi-
ple. The dominant wave (TE
10
mode) propa-
gates in the rectangular waveguide; its
electrical field is polarized parallel to the nar-
row sides of the waveguide. The electrical field
of the guided wave is excited in the slot by this
field. If the exciting frequency f is in agreement
with f
0
, then we find a slot width w, so that it is
true that
c
= 1 for f = f
0
and thus P
rad
= P
in
and
|r| = 0.
In the slot antennas used in this experiment the
conditions f = f
0
and =
c
are more or less se-
verely violated so that it is true that |r| 0.
Therefore, these antennas require the use of a
matching element.
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Isolator 737 06
1 PIN modulator 737 05
1 Slotted measuring line 737 111
1 Coax detector 737 03
1 Cross directional coupler 737 18
1 Transition waveguide / coax 737 035
1 Waveguide termination 737 14
1 Waveguide 200 mm 737 12
1 3-screw transformer 737 135
1 Diaphragms with slits incl. holder 737 22
1 Set of thumb screws (6 each) 737 399
Additionally required equipment
1 Oscilloscope (optional) 575 29
1 XY-recorder (optional) 575 663
3 Coax cable with BNC/BNC
plugs, 2 m 501 022
2 Stand bases 301 21
2 Supports f. waveguide components 737 15
1 Stand rod 0.25 m 301 26
Recommended
1 E-field probe 737 35
1 Slide screw transformer 737 13
1 Digital oscilloscope 575 292
1 XY recorder 575 663
Notes:
As in experiment 7 a cross directional cou-
pler is used here (coupling diaphragm
with 2 cross-shaped holes) to measure the
reflected wave. For more detailled infor-
mation please refer to Experiment 7 or Ex-
periment 10.
Instead of the 3-screw transformer (with
200 mm waveguide) the slide screw trans-
former can also be employed. The experi-
ment points can be carried out in like
fashion.
To be able to obtain reproducible findings,
you need to use the PIN modulator with
isolator. If you only use the oscillator with
internal modulation, normally the Gunn
oscillator is so severely affected by reflec-
tions that it generates high-frequency
modes with the result that the experiment
objective can no longer be achieved.
5 5
MTS 7.4.4
Reflection
Experiment procedure
1. Set up the experiment arrangement in ac-
cordance with Fig. 8.2. Set the screws of
the 3-screw transformer to a penetration
depth of approx. 0 (i.e. no effect from the
matching transformer).
(Optional: Set up the E-field probe as re-
ceiving antenna approx. 30 cm in front of
the open end of the waveguide. Keep the
intermediate space free of objects which
might cause scattering.)
2. Investigating mismatching using the three
different slot antennas (versions A, B and
C according to Fig. 8.3).
2.1 Attach slot antenna A to the open end of
the waveguide.
2.2 Connect the measurement amplifier with
detector to the slotted measuring line.
2.3 Determine the standing wave ratio s and
enter the value into Table 8.1. Calculate
the magnitude |r| of the reflection coeffi-
cient and enter the value into Table 8.1 a).
2.4 Repeat measurements 2.1 to 2.3, but re-
place slot antenna A with the slot antenna
Fig. 8.2: Experiment setup
Fig. 8.3: Dimensions and alignment of the three slot
antennas used here
(1) Metal cover with slit (= slot antenna)
(2) Waveguide flange
(C)
(B)
(A)
(2)
(1)
2
10
2
15
2
1
5
45
versions B resp. C (see Fig. 8.3). Enter the
findings for s and |r| in Table 8.1 a).
5 6
MTS 7.4.4 Reflection
Note:
Large standing wave ratios should be deter-
mined according to Fig. 6.2 and Equations 6.13
resp. 6.14. However, the applicability of these
methods are limited by the fact that l becomes
increasingly smaller with increasing s and con-
sequently more and more difficult to determine
using measuring techniques. Alternatively the
standing wave ratio can also be computed from
the level at the minimum a
min
(bear in mind the
adjustment of the gain selection switch V/dB).
With a
max
= 0 dB (maximum level set to 0 dB)
the standing wave ratio can be calculated ac-
cording to the following equation:
s
E
E
U
U
s
a a
a
= = =
=

max
min
max
min
max min
min
D
D
dB
dB
10
20
20
10
Optional (when using the E-field probe) :
2.5 Establish the connection of the measure-
ment amplifier to the E-field probe. Work
through points 2.6 and 2.7 using the slot
antennas A, B and C.
2.6 Rotate the E-field probe around its own
axis ( = 0, 45 and 90) (Attention:
leave the receiving dipole of the E-field
probe on the symmetrical axis of the
waveguide). At the same time observe the
receiving voltage. Determine =
opt.
for
maximum receiving voltage. Enter the
value of
opt.
in Table 8.1 b).
2.7 Select a calibrated (clear) setting for the
measurement amplifier e.g. ZERO to the
far right limit. Read off the receiving volt-
age (relative unit in dB). Enter the result
into Table 8.1 b) together with the setting
of the range switch V/dB.
3. Matching the slot antennas A and deter-
mining the radiated power (relative meas-
urement).
3.1 Reattach the slot antenna A to the open
end of the waveguide.
3.2 Connect the coax-detector to the cross di-
rectional coupler to measure the reflected
wave.
3.3 Successively adjust the 3-screw trans-
former (see Experiment 7 point 2.3, to
minimize the reflected wave (i.e. match-
ing). Set to the achievable maximum
matching (a value of s < 1.5 should be
reached; very good matching is reached
when s < 1.15).
3.4 Reconnect the detector to the slotted meas-
uring line and determine the standing
wave ratio (here again set the display to 0
dB at the maximum). Enter the values for s
and |r| in Table 8.2 a). (If necessary you
can remove the cross directional coupler
and waveguide section again, see Experi-
ment 7.)
Optional (when using the E-field probe) :
3.5 Establish the connection of the measure-
ment amplifier to the E-field probe.
3.6 Verify the optimum angle =
opt
from
point 2.6. Transfer the value of
opt
into
Table 8.2 b).
3.7 Select the calibrated setting for the meas-
urement amplifier (ZERO to the far right
limit). Read off the receiving voltage (rela-
tive unit in dB). Enter the result into Table
8.2 b) together with the setting of the range
switch.
4. Matching of the slot antenna B and deter-
mining the radiated power (relative meas-
urement).
4.1 Replace slot antenna A with slot
antenna B.
4.2 Repeat experiment points 3.2 to 3.4 for
this case.
Optional (when using the E-field probe) :
4.3 Repeat the experiment points 3.5 to 3.7 for
this case.
5. Matching the slot antenna C and deter-
mining the radiated power (relative meas-
urement).
5.1 Replace slot antenna B with slot
antenna C.
5.2 Repeat experiment points 3.2 to 3.4 for
this antenna type. At the maximum attain-
able matching you should achieve s < 2.
Optional (when using the E-field probe) :
5.3 Repeat experiment points 3.5 to 3.7 for
this antenna type.
Optional (when using the E-field probe) :
6. Calculate the absolute level a= a V,
(here a and a are negative and V > 0). De-
termine the maximum level a
max
. Then
compute the value for a a
max
5 7
MTS 7.4.4
Reflection
Table 8.1
a) Standing wave ratio and reflection coefficient without matching.
Antenna type s |r|
A
B
C
Optional :
b) Receiving level at the E-field probe without matching.
Antenna type
A
B
C
Table 8.2
a) Standing wave ratio and reflection coefficient
with maximum matching
Antenna type s |r|
A
B
C
Optional :
b) Receiving level at the E-field probe with maximum matching
Antenna type
A
B
C
The absolute level computed is: a= a V
The maximum level is a
max
= ________ dB

opt
V/dB a/dB a/dB (a a
max
)/dB

opt
V/dB a/dB a/dB (a a
max
)/dB
5 8
MTS 7.4.4 Reflection
Questions
1. Why is there a considerably higher reflec-
tion coefficient resulting for version C of
the slot antennas (see Fig. 8.3) than for
versions A and B?
2. Why is the matching of version C with the
aid of a matching transformer less sensible
from a technical point of view?
Optional:
3. In parts 2.6, 3.6 etc. of the experiment pro-
cedure the polarization of the electrical
field was experimentally determined for
the radiated waves (
opt
). What conclusion
could be drawn from this result?
4. Compare the magnitude of the receiving
signal for versions A and B. Why do sig-
nals of nearly the same magnitude result,
although the reflection coefficients are
considerably different (measurements 2,
Table 8.1)?
5. Assess the matching based on the level dif-
ference a a
max
while keeping in mind
for which case a
max
is reached.
Bibliography
See Experiment 5
5 9
MTS 7.4.4
Measurment of the permittvity
Measurement of the permittivity
Determining the complex dielectric constant of material samples from reflection coefficient measurements
Fundamentals
The interaction of the transparent materials is
characterized by complex permittivity

r
=
r
j
r
=
r
(1 j tan

) (9.1)
(
r
= real component of
r
, tan

= dielectric loss
factor) and the complex permeability

r
=
r
j
r
=
r
(1 j tan

). (9.2)
These material parameters are generally de-
pendent on the frequency.
For various technical problems and their solu-
tions knowledge of these material parameters
are required:
(a) If the materials are used for the construction
of microwave components, then the exact
values of
r
and
r
are required as input data
for computer codes in computer aided design
(CAD). Examples: low-loss dielectrics as
substrate materials for planar circuits, for
example, in microstrip technology and for
the realization of dielectric resonators for os-
cillator stabilization (DRO), and strongly ab-
sorbing materials for attenuating elements.
(b) Materials for the construction of radomes for
antenna systems and absorber materials for
reducing the radar cross-section of targets.
(c) The complex dielectric constant can fre-
quently be used for the determination of an-
other physical quantity. Thus the moisture
content of a material, for example, (sand,
coal, chipboards, tobacco etc.) modifies the
real as well as the imaginary part of the ef-
fective dielectric constant. Consequently, it
is possible to continuously monitor moisture
content using microwaves.
(d) Should the material be heated with micro-
waves (cooking, drying, etc.), the complex
dielectric constant and its change must be
known, in order to optimize the correspond-
ing microwave system.
(e) In physics and chemistry one can draw
conclusions as to material composition
from the position and width of the ab-
sorption spectra in the frequency re-
sponse of
r
or
r
. If a material sample of a
given shape is placed in a waveguide, the
desired information on the value of
r
or
r
is generally contained in the measurable
reflection coefficient and transmission co-
efficient. For this the following considera-
tions apply:
) Only for a geometrically simple
sample does a simple mathematical
expression apply for the relationship
between the measurable (reflection
coefficient and eventually the transmis-
sion coefficient) and the sought after
quantities
r
and
r
) The dimensions of the samples and
the selection of the measurement
quantities must ensure a sufficient
sensitivity on behalf of the meas-
urement quantities to changes in the
material parameters.
) If both
r
as well as
r
are unknown,
at least two independent (complex)
measurement quantities must be
determined. It is frequently known
beforehand that the material has a
value of
r
= 1. In this case one
measurement quantity would be
sufficient for the determination of
r
.
In the following we shall presuppose that

r
= 1, so that only
r
needs to be determined.
Figure 9.1a shows the possible measurement con-
figuration in which a material sample with the
length L with an unknown value of the (complex)
relative permittivity
r
fills up the rectangular
waveguide completely.
The short-circuit plane coincides with the back-
side of the material sample for the position l= 0
of the moveable short , and thus the equivalent cir-
cuit diagram (b1) of the transmission line applies.
Thus the following is true for the normalized input
impedance:
Z
Z
a
a
j
L
a

0
0
2
0
2
0
0
2
1
2
2
2
2

j
(
,
\
,
(

j
(
,
\
,
(

j
(
,
\
,
(
,

,
,
,
]
]
]
]
]


r
r
tan
(9.3)
6 0
MTS 7.4.4 Measurment of the permittvity
Z
A
/Z
0
can be determined from the reflection
measurements and
0
/2a as well as L/
0
are
known variables. Consequently, the sought-
after value of
r
can be determined from Equa-
tion (9.3). However, Equation (9.3) is a complex
transcendental equation, which can only be
solved numerically (zero position search pro-
gram) for
r
.
Thus, a different approach should be taken here
and, in addition, the normalized input impedance
should be used for the open-circuit case. We then
obtain the open-circuit case for the position
l=
g
/4 of the moveable short, and with the
equivalent circuit according to (b2) in Figure 9.1
the result is
Z
Z
a
a
j
L
a

0
0
2
0
2
0
0
2
1
2
2
1
2
2

j
(
,
\
,
(

j
(
,
\
,
(

j
(
,
\
,
(
,

,
,
,
]
]
]
]
]


r
r
tan
(9.4)
If you form the product from Z
A
and Z
B
, then the
tan-function cancels out, and you obtain an ex-
pression which can be solved directly for the de-
sired value
r
.


r

j
(
,
\
,
(
+
j
(
,
\
,
(
j
(
,
,
\
,
(
(
0
2
0
2
2
1
2 a a
(9.5)
whereby is used as an abbreviation for the ex-
pression

Z
Z Z
0
2
A B
(9.6)
Z
A
and Z
B
can be calculated from the values r
A
and r
B
of the reflection coefficient determined
using the slotted measuring line via
Z
Z
r
r
Z
Z
r
r

0 0
1
1
1
1

A
A
B
B
and
(9.7)
or determined using the Smith chart.
4
l
g
l =
(3) (3)
r
a
Z
B
Z
d
Z
0
L r
A
Z
A
Z
d
Z
0
L
l = 0
(b2) (b1)
r 0 0
e e m
l
L (a)
(2)
(1)
(3)
Fig. 9.1: (a) Measurement configuration with
material sample (1) with the length L
and moveable short (2) connected
downstream.
Distance of the short-circuit plane from
rear side of material sample is l.
Reference plane for measurement is (3).
(b) Equivalent circuit diagram for trans. line
for computation of the input impedances
Z
A
and Z
B
for the short-circuit (l = 0) and
open-circuit case (l =
g
/4).
Requiredequipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Diaphragm with slit
2 x 15 mm, 90 737 22
1 Isolator 737 06
1 PIN modulator 737 05
1 Slotted measuring line 737 111
1 Coax detector 737 03
1 Short-circuit plate 737 29
1 Sample holder 737 29
1 Material sample of polystyrene 737 29
1 Material sample of graphite 737 29
1 Moveable short 737 10
1 Set of thumb screws (2 each) 737 399
Additionallyrequiredequipment
1 Oscilloscope (optional) 575 29
1 XY recorder (optional) 575 663
3 Coax cables with BNC/BNC
plugs, 2 m 501 022
2 Stand bases 301 21
2 Supports f. waveguide components 737 15
1 Stand rod 0.25 m 301 26
1 Caliper
6 1
MTS 7.4.4
Measurment of the permittvity
Note:
To be able to obtain reproducible results,
you need to use the PIN modulator with iso-
lator, because in this experiment monomode
operation is indispensible. Reproducible re-
sults can only be guaranteed with a PIN
modulator.
Experiment procedure
1. Setupandcalibrationmeasurement
1.1 Set up the experiment according to Fig.
9.3. Arrange the measurement object ac-
cording to Fig. 9.1 and Fig. 9.2.
1.2 Attach the short-circuit plate to the open end
of the slotted measuring line.
1.3 Determine the position x
0
of the first mini-
mum (from the left) on the slotted measur-
ing line and enter the value into Table 9.1.
Determine the position x
1
of the 3rd mini-
mum and enter it into Table 9.1.
Fig. 9.3: Experiment setup
(1) Short-circuit plate
(2) Meas. object corresponding to Fig. 9.2 (see also Fig. 9.1)
(3) Moveable short-circuit
Fig. 9.2: Measurement object for Experiment 9
1.4 Calculate the guided wavelength

g
x x
0 1
.
2. Determiningshort-circuitandopen-cir-
cuit input reflection coefficient for
sampleI(polystyrene)
2.1 Insert sample I (polystyrene, black) into the
sample holder. On the positioning of the sam-
ple in the holder please refer to Fig. 9.2. Use a
caliper to measure the distance l
H
to the right-
hand edge of the sample holder (l
H
should
amount to just about 10.0 mm exactly).
6 2
MTS 7.4.4 Measurment of the permittvity
2.2 Attach the sample holder (with sample) to
the open end of the slotted measuring line
(see Fig. 9.3).
2.3 Attach the moveable short to the open side
of the sample holder.
2.4 Adjust the moveable short (x
k
) so that the
position of the short-circuit plane coincides
with the rear side of the sample (l =
g
/2, i.e.
x
k
=
g
/2 l
H
). Note: Read x
k
from the mi-
crometer of the variable short.
2.5 Based on the standing wave ratio s
A
and the
position of the minimum x
A
(first from the
left), determine the short-circuit reflec-
tion coefficient:
r r e '
'
A A
j

according to magnitude and phase (refer-


ence plane = front surface of the sample).
Enter into Table 9.1. (Note: see Table 9.1
for the individual mathematical steps)
2.6 Shift the moveable short further by one
quarter of a guided wavelength (
g
/4) so
that the open-circuit plane coincides with
the rear side of the sample.
2.7 Based on the standing wave ratio and the
position of the minimum determine the
open-circuit reflection coefficient
r r e ' '
'
B B
j

according to magnitude and phase and


enter the values into Table 9.1.
3. Determiningtheshort-circuitandopen-
circuitinputreflectioncoefficientforsam-
pleII.(non-magneticabsorbermaterial
madeofsyntheticresinwithgraphite,
color:anthracite)
Analogous to 2.1 to 2.7 determine
r r e ' ' ' '
' '
A A
j

and
r r e ' ' ' '
' '
B B
j

Enter the values into Table 9.1.


General notes
To avoid big mistakes when determining
the dielectric constant or permittivity, the
phase measurements have to be performed
with great care (precise values of x
A
, x
B
and x
0
). Furthermore, the guided wave-
length must be known precisely (deter-
mine using the slotted measuring line) and
the moveable short (x
k
) must be set pre-
cisely (so that l =
g
/2 or 3
g
/4, see Fig.
9.1).
Tip: For the determination of the mini-
mum, the gain should be successively in-
creased to find the location with more
precision. An even finer setting is possible
by a slight tapping, i.e. moving the car-
riage with a pen.
Questions
1. In accordance with the equations
Z
Z
r
r
Z
Z
r
r

0 0
1
1
1
1

A
A
B
B
and
calculate the normalized short-circuit and
open-circuit impedances for samples I
and II and enter the values into Table 9.1.
(Alternatively: Determine using the Smith
chart, see Fig. 9.4).
2. Calculate the expression (see also Equa-
tion 9.6)

Z
Z Z
0
2
A B
for samples I and II and enter the values
into Table 9.1.
3. Determine the complex dielectric con-
stants from sample I and II according to


r

j
(
,
\
,
(
+
j
(
,
\
,
(
j
(
,
,
\
,
(
(
0
2
0
2
2
1
2 a a

(
0
= free-space wavelength,
a = waveguide width = 22.8 mm).
Enter the final results into Table 9.1.
6 3
MTS 7.4.4
Measurment of the permittvity
Sample I (Polystyrol) Sample II (Absorber material)
from the Smith chart
or mathematically
from the Smith chart
or mathematically
VSWR s
A
x
A
/mm
r
s
s
A


+

1
1

A
A
g


180 720
0
x x
r r j
A A A A
+ ( ) cos sin
VSWR s
B
x
B
/mm

B
B
g


180 720
0
x x
r
s
s
B


+

1
1
r r j
B B B B
+ ( ) cos sin
Z
Z
r
r
B B
B 0
1
1

Z
Z Z
0
2
A B

r
='
r

r
''
Z
Z
r
r
A A
A 0
1
1

tan

=
r
''/
r
'
Table9.1
x
0
= ________ mm x
1
= ________ mm

g
= ________ mm
0
= ________ mm
6 4
MTS 7.4.4 Measurment of the permittvity
Fig. 9.4: Smith chart
Bibliography
[1] M. Sucher, J. Fox: Handbook of Microwave Measurements. Polytechnic Press,
Brooklyn NY, 1963
[2] A. R. von Hippel: Dielectric Materials and Applications. J. Wiley & Sons,
New York 1954
[3] F. E. Gardiol: Introduction to Microwaves. Artech House, Dedham MA, 1984
6 5
MTS 7.4.4 The cross directional coupler
The cross directional coupler
Fundamentals
General properties of waveguide (directional)
couplers
The cross directional coupler is a special type of
directional coupler. Thus, it makes sense to follow
with a general explanation applicable to the func-
tion of all types of waveguide couplers and their
most important parameters. An (unconnected)
waveguide coupler is a reciprocal four-port, which
is also ideally loss-free. Fig. 10.1 illustrates the re-
sponse of the ideal directional coupler. If a wave is
fed exclusively into port 1, its effective power P
1
is distributed to port 2 (primary path) and port 4
(coupling path). If the power exiting port 4
amounts to k
2
P
1
, then the power must be (1 k
2
)
P
1
at port 2 because of the lossless property of
waveguide port 2 (conservation of energy). Ideally
there is no power exiting port 3, i.e. it is decou-
pled (isolation path). If, on the other hand, a wave
is fed into port 2, port 4 is decoupled and the power
fed is distributed to port 1 (primary path) and port 3
(coupling path). The magnitude of k (coupling
coefficient) can vary in amplitude depending on
the design of the directional coupler. The following:
a
k
= 20 log k [dB] (10.1)
is referred to as coupling loss. If you consider the
definitions provided above and furthermore the
relationships between the S-parameters resulting
from the loss-free property of the waveguide (see
Experiment 5 from MTS 7.4.5 or the bibliography
provided there), we obtain the following scattering
matrix for the ideal waveguide coupler.
S e
k jke
k jke
jke k
jke k
( )

,
,
,
,
,
,
,
]
]
]
]
]
]
]
]
]
j
j
j
j
j

0 1 0
1 0 0
0 0 1
0 1 0
2
2
2
2
(10.2)
Here, in addition to the coupling coefficient k, there
are also the phase angles and as parameters
( is the phase rotation in the primary guide and
+ are the phase rotations in the coupling guide).
Fig. 10.2 shows one of the primary applications of
waveguide couplers, namely the separate
measurement in front of a one port (e.g. antenna)
of the wave propagating to and reflected from the
load.
The signal exiting at port 3 is only proportional to
the wave P
2,in
= |r| P
2,out
reflected by the load
(here the antenna), while a signal proportional to
the wave propagating to the load appears at port 4.
The following applies:
P
P
k P
k P
r k
3,out
4,out
2
2,in
2
1,in
2
2

(1 )
(10.3)
When the value of k
2
is known, the reflection co-
efficient |r| can be derived from the relationship
between P
3,out
and P
4,out
. The configuration shown
Fig. 10.2: Use of the directional coupler as a reflectometer Fig. 10.1: Principal response of a directional coupler
1
1
1
1
2
2
2
2
3
3
3
3
4
4
4
4
6 6
MTS 7.4.4 The cross directional coupler
in Fig. 10.2 is referred to as a reflectometer. In
Experiment 11 the cross directional coupler is used
as a reflectometer.
Contrary to the statements above, in real direc-
tional waveguide couplers decoupling via an isola-
tion path (e.g. from port 1 to port 3 in Fig. 10.1) is
not total. If a wave is only fed in at port 1, power is
still obtained at port 3; P
3,out
> 0. The ratio of
undesired power at port 3 (P
3,out
) to desired power
at P
4,out
at port 4 is one of the ways of determining
the quality of a waveguide coupler. This is called
the directivity factor. The following applies for
directivity:
When feeding at port 1
a P
P
D 4,out
3,out dB
10 log (10.4)
When feeding at port 2
a P
P
D 4,out
3,out dB
10 log
High quality directional couplers are expensive
and can have a directivity of over 50 dB, whereas
simple waveguide couplers average around
20 dB.
The cross directional waveguide coupler as a
special form of directional waveguide coupler
Up until now the response of a directional
waveguide coupler has only been dealt with as a
kind of black box, i.e. only its operation within a
circuit. However, nothing has been mentioned
about the physical effects which are needed to
exploit this behavior. There are directional
couplers for all conventional transmission lines in
microwave engineering (coaxial lines, microstrip
lines, hollow waveguides). These couplers employ
a wide variety of different principles to realize
directivity (the separation of the incident and
reflected wave). Examples for this include the
directivity employed in coupling TEM lines running
parallel and the interference of waves in hollow
waveguides, which are coupled together via holes.
In the following the principle of the cross
directional coupler is studied in more detail. In
accordance with Fig. 10.3 the cross directional
coupler is made up of two hollow waveguides
arranged at a 90 angle as specified in Fig. 10.3, so
that they have a common wall (overlapping
square surface). Coupling is carried out via one
or two coupling holes in this common wall. For a
closer examination of the theoretical aspects of
this phenomena please refer to the literature
specified in the bibliography. Only a few important
findings are dealt with here.
Electric and magnetic coupling through a
hole in a metal wall
In the upper part of Fig. 10.4 a cutaway section of
a closed metal wall is shown, on whose underside
an electrical (E) and a magnetic field (H) is found.
This means that:
1. An electric and magnetic field exist on the
underside of the closed metal wall.
2. Penetration of the electric and magnetic field
through a coupling hole.
3. Description of the coupling via an electric (J)
and a magnetic (M) dipole.
Fig. 10.5: gives a schematic depiction of the cross
directional coupler with a cylindrical coupling hole.
Where:
Fig. 10.3: Principal design of a waveguide (cross direc-
tional) coupler and including port numbering
Fig. 10.4: Explanation of the hole coupling
6 7
MTS 7.4.4 The cross directional coupler
Fig. 10.5: Cross coupler with a circular coupling hole with
a diameter of 2 R
Fig. 10.6: Wave excitation when feeding wave into port 1
3 2 1
1 2
3
4
d
a
J
M
X
M
Y
1 Transversal hollow waveguide section
2 Coupling hole
3 Reference planes (port 3) for the phases of
the S-parameters
We are assuming that the TE
10
wave propagates
from port 1 to port 2. If you now consider sepa-
rately the effect of the x- and y-components of the
magnetic dipole and the effect of the electric di-
pole, see Fig. 10.6, you arrive at the following con-
clusions:
The partial magnetic waves at port 3 completely
cancel each other out, while the magnetic waves
at port 4 superposition each other constructively.
Thus, if only magnetic coupling were present, the
four-port would respond like an ideal directional
coupler. However, there are still the partial waves
excited by the electric dipole (electric coupling).
Here, two partial waves of the same magnitude
are formed for port 3 and port 4. The complete
decoupling of port 3 thus fails due to the partial
waves caused by the electric coupling.
Measures to increase directivity over that
attained with a single round coupling hole
Deviation from this ideal directional coupler results
when electric coupling is added. If an increase in
directivity is desired, the electric coupling must be
reduced. One suitable way of realizing this is by
substituting the round coupling hole with the cross-
shaped hole. Compared to round-shaped holes, in
cross-shaped coupling holes (see Fig. 10.7 centre)
the magnetic penetration predominates more
than the electric penetration. As the experiment
will show a substantial increase in directivity is
achieved by exchanging the round coupling hole
with the cross-shaped hole. An added improve-
ment is achieved by using two cross-shaped holes
instead of one (see Fig. 10.7, right). Due to the in-
teraction between the two holes, the electric
coupling is weakened.
The cross coupler included in the training system is
comprised of several dismountable parts and there
are various options between different alternative
hole configurations for coupling. Therefore, the
improvements attained when going from one round
hole to one cross-shaped hole and from two round
holes to two cross-shaped holes can be verified ex-
perimentally.
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Diaphragm with slots
2 x 15 mm, 90 737 22
1 Variable attenuator 737 09
1 Cross directional coupler 737 18
1 Transition waveguide/coax 737 035
1 Coax detector 737 03
2 Waveguide terminations 737 14
1 Set of thumb screws (8 each) 737 399
Additionally required equipment
2 Coax cables with BNC/BNC
plugs, 2 m 501 022
3 Standbases 301 21
3 Supports for waveguides 737 15
2 Stand rods 0.25 m 301 26
Fig. 10.7: Measures to increase directivity
6 8
MTS 7.4.4 The cross directional coupler
Fig. 10.8 Experiment setup
2.1 Install the diaphragm with round hole into the
cross coupler. Here it is important to focus on
the installation direction (e.g. 4 to 4) (see
Fig. 10.9, above).
2.2 Connect port 1 of the cross coupler to the
open end of the variable attenuator (instead
of the measurement head in Fig. 10.8,
above).
2.3 To measure the magnitude of the transmis-
sion coefficient (|S
21
|) to port 2 you must
equip ports 3 and 4 with a reflection-free
waveguide termination and connect a meas-
urement head (transition waveguide/coax
with coax detector) to port 2. The display of
the SWR meters supplies |S
21
| in dB. Enter
the result in Table 10.1.
2.4 For the measurement of |S
31
| connect meas-
urement head to port 3 and reflection-free
terminations to ports 2 and 4. Enter result in
Table 10.1.
2.5 For the measurement of |S
41
| connect meas-
urement head to port 4 and reflection-free
terminations to port 2 and port 3. Enter result
into Table 10.1
Recommended:
1 PIN modulator 737 05
1 Isolator 737 06
Experiment procedure
Note:
When using the isolator and PIN modulator modify
the experiment setup in Fig. 10.8 according to the
preface!
1. Calibration of the experiment setup
1.1 First set up the measuring system as speci-
fied in Fig. 10.8, above, without the cross
directional coupler. Set ZERO to the far
right, the gain level V/dB to approx. 10 dB
(resp. 15 dB). With the aid of the variable
attenuator calibrate the display of the SWR
meter to 0 dB. This setting is no longer
changed during the experiments.
2. Measurement of some S-parameters of the
cross directional coupler with 1 round
coupling hole.
6 9
MTS 7.4.4 The cross directional coupler
Fig. 10.9: On the numbering of the ports of the cross-
coupler (here, for example, port 4 is marked)
2.6 Now make connections for wave feed via
port 2, i.e. connect port 2 to the open end of
the variable attenuator. The transmission co-
efficients |S
12
|, |S
32
| and |S
42
| are determined
as in 2.3 to 2.5. Enter the results in Table 10.1.
3. Measurement of some S-parameters of the
cross coupler with 1 cross-shaped hole.
3.1 Exchange the diaphragm with round hole for
the diaphragm with cross-shaped hole.
3.2 Measure the transmission coefficients |S
21
|,
|S
31
|, |S
41
|, |S
12
|, |S
32
| and |S
42
| as in 2.3 to 2.6.
Enter the values in Table 10.1
4. Measurement of some S-parameters of the
cross directional coupler with 2 round
holes.
4.1 Exchange the current diaphragm for the dia-
phragm with 2 round holes.
4.2 Measure the transmission coefficients as set
forth under 2.3 to 2.6. Enter the results in
Table 10.1
5. Measurement of some S-parameters of the
cross directional coupler with 2 cross-
shaped holes.
5.1 Exchange the current diaphragm for the dia-
phragm with 2 cross-shaped holes.
5.2 Measure the transmission coefficients as in
2.3 to 2.6. Enter the results in Table 10.1
Exercises
1. In Fig. 10.2 the coupling loss attenuation of
the directional coupler is assumed to be 20 dB
(20 dB-coupler) and the power fed into the
port is P
1,in
= 1 W.
Specify the values for P
1,out
, P
3,out
and P
4,out
as well as P
3,out
/ P
4,out
, if
) |r| = 0.0
) |r| = 0.5.
2. Determine the directivity for the 4 different
configurations of the cross directional coupler
and enter these into Table 10.1.
Note
Directivity:
When feeding from port 1
a
S
S
D
log
j
(
,
,
\
,
(
(
20
31
41
or when feeding from port 2
a
S
S
D
log
j
(
,
,
\
,
(
(
20
42
32
Here the S-parameters are not to be used
logarithmically.
If when determining the S-paramters of the
isolation path you reach the limits of the
measurement amplifier, then try to enhance
the modulated signal by varying the Gunn
voltage (see Experiment 2. But take care:
Monomode operation must be maintained).
Otherwise, we would highly recommed using
a PIN modulator (with isolator).
7 0
MTS 7.4.4 The cross directional coupler
Table 10.1
|S
21
| / dB
|S
31
| / dB
|S
41
| / dB
|S
12
| / dB
|S
32
| / dB
|S
42
| / dB
|a
D
| / dB
|a
D
| / dB
1 Round hole
1 Cross-shaped
hole
2 Round holes
2 Cross-shaped
holes
Feed
via
port 1
Feed
via
port 2
Port 1 :
Port 2 :
7 1
MTS 7.4.4 Principle of the reflectometer
Principle of the reflectometer
As demonstrated in Experiment 10, incident and
reflected waves can be separated with the aid of a
directional coupler (here specifically the cross di-
rectional coupler). Thus the directional waveguide
coupler constitutes an important component in the
measurement of reflection coefficients. This
special measuring technique is called
reflectometry as opposed to the reflection coeffi-
cient measurement using the slotted measuring
line. The greatest advantages over the slotted
measuring line method are; one, it takes considera-
bly less time to conduct the measurement and two,
it can be used in automatic sweep measurements.
Various measurement principles of
reflectometry
(1)Use of only one coupling guide and compar-
ing the result with reflection at the short
In the measurement configuration according to
Fig.11.1, port 4 remains terminated reflection-free
throughout the entire measurement procedure.
Thus, reflection-free termination can be an integral
part of the directional coupler in the practical
design of a reflectometer. The wave travelling via
the feed-through guide from port 1 to port 2
(weakening of the amplitude by 1
2
k ) is re-
flected at the measurement object with the un-
known reflection coefficient r. A portion of the
wave incidenting port 2 with the amplitude:
a r b r k a
2 2
2
1
1 = =
(11.1)
is transmitted to port 3 via the coupling guide so
that the magnitude of the wave exiting at port 3 is
given by:
b k a k k r a
3 2
2
1
1 = =
(11.2)
The coupling coefficient k of the directional cou-
pler is known in advance, but not the amplitude |a
1
|
of the wave incidenting at port 1. For that reason
this measurement only suffices to determine |r|.
Thus a reference measurement is performed with
a short (r = 1) at port 2. For this you obtain from
Eq. (11.2)
Fig. 11.1 Measurement configuration 1
r' = 0
4
3
1 2
a = r b
2 2
b
2
b
3
b
1
a
1
DUT
r
~
b k k a
3
2
1
1 =
(11.3)
and consequently by forming the ratio we obtain
the desired magnitude of the reflection coefficient
b
b
r
3
3
~
= (11.4)
However, the Equation (11.4) only applies under
the assumptions made above; i.e.
() Ideal directional coupler with infinitely high
directivity and free of reflection at its ports.
() The wave b
1
exiting port 1 is not reflected by
the circuit connected at port 1.
If condition () is violated, we have a situation
(due to multiple reflection) where in the reference
measurement with short circuit, the wave entering
port 1 is different from the wave entering port 1 in
the measurement of the device under test (DUT).
Under the assumption that an ideal directional cou-
pler is present, a precise mathematical analysis
supplies the equation
b
b
r
k r e
k r e
3
3
2
2
1 1
1 1
~
=
+

G
j
G
j

(11.5)
instead of Eq. (11.4). Here r
G
is that reflection co-
efficient which is present for wave b
1
leaving port
1. It can be seen from equation (11.5) that |r
G
|
must be kept as small as possible to avoid meas-
urement errors. This can be accomplished by, for
example, connecting an isolator or an attenuator
with a 10 dB in series. If it is assumed that
r
G
= 0 but the directivity factor is finite as in the
case of all real waveguide couplers, then using
equation (11.4) we obtain a maximum error of
7 2
MTS 7.4.4 Principle of the reflectometer

def
=
D
b
b
r
a
3
3
20
10
~
max



(11.6)
which for a directivity of a
D
= 60 dB is smaller than
0.001, but at a directivity of only 20 dB amounts to
approx. 0.1.
(2) Utilizing two coupling paths
In the measurement configuration according to
Fig. 11.2 a) with one directional coupler, the waves
exiting from port 3 and port 4 are measured and
their amplitudes compared. The following holds
true for ideal directional couplers:
|b
4
| = k |a
1
|
and
b k k r a
3
2
1
1 =
and thus
b
b
k r
3
4
= 1
2
(11.7)
Since k is unknown, Eq. (11.7) can serve to deter-
mine |r|. In this procedure a reference measure-
ment with a short is not needed and a reflecting
circuit (r
G
0) does not create any fault source (as
|a
1
| is the same for the measurement of |b
3
| and
|b
4
|). In the case of less than ideal directional cou-
plers the use of Eq.(11.7) also leads to errors in the
determination of |r|. When using two separate
directional couplers in the configuration according
to Fig. 11.2 b) coupling errors have less of an
impact on the final result (naturally the coupling
coefficients of the two directional couplers must
be known).
(3) Prospects for more modern measurement
methods of reflectometry
In the previous investigations it was assumed that
only the magnitudes of the waves exiting ports 3 or
4 could be measured and that furthermore only the
magnitude of r could be determined. Nevertheless,
not only the amplitudes but also the phases of the
waves can be compared as, for example, in the
configuration according to Fig. 11.2.
For this you can use a so-called network analyzer,
which is a special kind of dual channel receiver, in
which the signals in both channels are converted to
a lower frequency range while maintaining their
amplitudes and phase relationships. This frequency
conversion can be carried out with mixing or
sampling. In the low frequency range the ampli-
tudes and phase relationships of the signals can be
determined by electronic means. One of the great
advantages of using the network analyser is the
possibility of correcting errors arising from compo-
nents with less than ideal characteristics (e.g. finite
directivity of directional couplers) in the mea-
surement circuit. For this several calibration
measurements are first conducted on known com-
ponents and the results entered into a computer.
Since the faulty characteristics of the measure-
ment circuit are also contained in the results of the
calibration measurements, they can be corrected
using suitable algorithms.
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Diaphragm with slit
2 x 15 mm, 90 737 22
Fig. 11.2: Measurement configuration 2
a) Measurement with a directional coupler
b) Measurement with two directional couplers
7 3
MTS 7.4.4 Principle of the reflectometer
1 Variable attenuator 737 09
1 Cross directional coupler 737 18
1 Transition waveguide/coax 737 035
1 Coax detector 737 03
1 Waveguide termination 737 14
2 Sets of thumb screws (12 each) 737 399
For the assembly of the measurement object
or DUT the following is required:
1 Sample holder 737 29
1 Absorbing material sample
(graphite) 737 29
1 Waveguide termination 737 14
1 Waveguide 200 mm 737 12
or
3 screw transformer (*) 737 135
Additionally required equipment
2 Coax cables with BNC/BNC
plugs, 2 m 501 022
3 Stand bases 301 21
3 Supports for waveguide 737 15
Fig. 11.3 Experiment setup
(*) : The 3-screw transformer serves only as an intermediate piece to permit attachment. For this all screws must
be set to a penetration depth of 0. In this case the thumb screw requirement is reduced to 8 each (1 set).
3
1
2
4
737 14
737 14
737 29 737 12
or
737 135
737 29
Short circuit plate
737 18
737 14 737 14
7
3
7
1
8
737 09 737 22 737 01
GUNN-OSC.
737 01
2 x 15 mm
90
Gunn
WG
coax
SC
ATT
Holder +
sampler
7 4
MTS 7.4.4 Principle of the reflectometer
2 Stand rods 0.25 m 301 26
Recommended:
1 PIN modulator 737 05
1 Isolator 737 06
Experiment procedure
Note:
When using the isolator and PIN modulator the
experiment setup specified in Fig. 11.3 must be
changed in accordance with the preface!
1. Set up a reflection-free one port as the de-
vice under test, DUT).
1.1 In Experiment 6 the reflection coefficient of
a DUT was determined using the slotted
measuring line. Use the same DUT to permit
a comparison of the reflectometer measure-
ment with the slotted measuring line measure-
ment. In accordance with Fig. 11.4 use the
sample holder 1, the absorbing graphite sam-
ple 2 and a reflection-free waveguide termi-
nation 3 for set up of the DUT.
2. Measurement of the reflection coefficient
using only one coupling path and com-
parison with reflection at the short-circuit
plate.
2.1 Set up the measurement circuit as specified
in Fig. 11.3. Use the coupling plate with 2
cross-shaped holes in the cross coupler.
Note:
The attenuator set to approx. 10 dB should be
used to reduce the reflection coefficient for
waves propagating out of port 1 and re-
flected backwards to the microwave source.
Port 4 is terminated reflection-free and the
measurement head (transition waveguide/
coax with coax detector) is connected to port
3.
2.2 Calibration measurement with short-circuit
plate.
For this attach the short-circuit plate to port
2. After the power supply voltages have been
switched on, calibrate the display of the SWR
meter (port 3) to 0 dB using the ZERO
control knob. (Set the modulation to Gunn-
Int., U
G
to 8 V)
2.3 Replace the short-circuit plate with the DUT.
Read off the display a (port 3) in dB (a < 0).
Fig. 11.4: Assembly of the device under test (DUT)
1Sample holder
2Graphite sample
3Waveguide termination
According to Eq. (11.4) a represents the
reflection coefficient in dB, i.e.:
a / dB = 20 log |r|.
Calculate |r| and enter the value into Table
11.1.
3. Measurement of the reflection coefficient
when using 2 coupling paths
3.1 The DUT remains connected to port 2.
3.2 First terminate port 3 reflection-free by re-
placing the measurement head (transition
waveguide/coax to the coax detector) with
the reflection-free waveguide termination.
Attach the measurement head to port 4. Set
the display a of the SWR meter (port 4) to a
value of 0 dB using the ZERO control knob
and keep this setting steady.
3.3 Reverse the connection configuration of ports
3 and 4, i.e. connect reflection-free termina-
tion to port 3. Read display a of the SWR
meter off in dB (a < 0). According to Eq.
(11.7) the following applies:
a/dB = 10 log (1 k
2
) + 20 log |r|
20 log |r|
Enter the values determined for |r| in Table
11.1.
Table 11.1
7 5
MTS 7.4.4 Principle of the reflectometer
Experiment
point 2
Experiment
point 3
Questions
1. Compare all the values determined for |r| to
each other and with the results from Experi-
ment 6 (measurement principle with slotted
measuring line).
2. Compute the ratio of |b
4
| / |b
3
| from the con-
figuration in Fig. 11.2 b). The coupling coeffi-
cients are k
1
and k
2
. What is the ratio for the
special case k
1
= k
2
= k?
a/dB
|r|
7 6
MTS 7.4.4 Principle of the reflectometer
7 7
MTS 7.4.4
The cavity resonator
The cavity resonator
Fundamentals
First of all a completely enclosed cavity resona-
tor is considered with (theoretically) an ideally
conductive metallic surface. In this resonator
unattenuated electromagnetic oscillations can ex-
ist at discrete frequencies (= resonance frequen-
cies). Fig.12.1 shows the TE
101
resonance in a
rectangular waveguide resonator with the fre-
quency:
f
l
0
2 2
GHz
15
1
a
cm
1
cm
=

(12.1)
Due to the losses in the metallic walls of real reso-
nators, natural oscillations are attenuated. One
measure for attenuation in a cavity resonator is its
unloaded Q value:
Q
W
P
0
0
V
=

(12.2)
Here W is the stored energy,
0
= 2 f
0
is the
(angular) resonance frequency and P
V
is the
power dissipation. Fig. 12.1 shows the electro-
magnetic field in the cavity resonator at various
points in time (T = 1/f
0
). In order to couple a
resonator to a microwave circuit, you can use a
diaphragm with aperture, for example, as illus-
trated in Fig. 12.3a. Fig. 12.3b shows the corre-
sponding waveguide equivalent circuit diagram
in which the resonator is represented by a short-
circuited waveguide and the diaphragm with ap-
erture by a shunt inductance jL = jX. Fig. 12.4
shows Z
0
/X as a function of the relationship of the
aperture diameter d to the waveguide width a
(22.86 mm). If you mathematically convert the
parallel connection of the resistance Z
0
and jX
into a series connection, the result is an equiva-
lent circuit diagram for X << Z
0
as specified in
Fig. 12.2.
In accordance with the transformation ratio:
n
Z
X
=
0
(12.3)
the resistance Z
0
is transformed to lower resist-
ance Z
0
= Z
0
/n. Thus, the diaphragm with aper-
ture can also be understood as a transformer. If we
suppose that the reactance jX is taken (or ignored)
Fig. 12.1: TE
101
resonance in a rectangular cavity resona-
tor
Fig. 12.2: Series equivalent circuit diagram for resonators
with coupling diaphragm
7 8
MTS 7.4.4 The cavity resonator
Fig. 12.3: Development of an equivalent circuit diagram
for the cavity resonator coupled via a diaphram
with aperture
Fig. 12.4: Normalized transformation ratio for the
coupling diaphragm Z
0
/X. corresponds
here to the free-space wavelength
0
Required equipment
1 Basic unit 737 021
1 Gunn oscillator 737 01
1 Diaphragm with slits
2 x 15 mm, 90 737 22
1 Variable attenuator 737 09
1 Cross directional coupler 737 18
1 Transition waveguide/coax 737 035
1 Coax detector 737 03
1 Waveguide termination 737 14
1 Accessories waveguide propagation 737 29
1 Moveable short 737 10
1 Set of thumb screws (8 each) 737 399
Additionally required equipment
2 Coax cables with BNC/BNC
plugs, 2 m 501 022
2 Stand bases 301 21
2 Supports for waveguides 737 15
1 Stand rod 0.25 m 301 26
Recommended:
1 PIN modulator 737 05
1 Isolator 737 06
together with the waveguide, we obtain the
equivalent circuit depicted in Fig. 12.3 c). At ap-
proximately the resonance (angular) frequency
2 f
0
the
g
/2 long waveguide can be designed by
a resonance circuit with:
1
2
1
2
0
0 0
0
2
c
2
0
c
0

=
f
L C
L
Z Z X
0
and
R
X
Q
V
c
0
= (12.4)
Finally, we obtain for resonance =
0
the
equivalent circuit diagram according to Fig.
12.3 e).
7 9
MTS 7.4.4
The cavity resonator
Fig. 12.5: Experiment setup
Experiment procedure
Note:
When using the isolator and PIN modulator
modify the experiment setup shown in Fig. 12.5 in
accordance with the preface!
1. Set up the experiment according to Fig.
12.5. Use the coupling plate with 2 cross-
shaped holes in the cross directional cou-
pler. Set the attenuator to approx. 10 dB
(see Experiment 11, subpoint 2.1).
2. Calibration
Determination of the reflection coefficient is
performed like in Experiment 11, point 1,
i.e. using only one coupling path and then
comparing the results to the reflection
ocurring during short-circuit.
Thus for calibration insert the moveable
short with the diaphragm d = 6 mm into the
experiment setup and set the moveable short
to 0 mm (i.e. implement short in front of
port 2). The voltage displayed for this at the
SWR meter corresponds to the reflection
8 0
MTS 7.4.4 The cavity resonator
coefficient |r| = 1 and can thus be used for
calibration. Set the SWR meter to 0 dB us-
ing the ZERO control knob. As a further
reference for the matching case you can in-
sert a reflection-free waveguide termina-
tion instead of the moveable short.
3. Resonator measurements
Insert the moveable short with diaphragms
(one after the other d = 6, 7, 9, 10 mm).
Change the position of the moveable piston
and observe how the reflection coefficient is
dependent on the setting of the micrometer.
Determine the minimum of |r| = |r|
min
for
each diameter of the diaphragm aperture.
Enter the values into Table 12.1.
Using Fig. 12 .4 carry out the exact deter-
mination of the value n corresponding to
each aperture diameter d and enter it into
the table. (First compute the parameter
value
0
/a, (f 9.4 GHz, a = 22.86 mm),
but bear in mind the logarithmic scaling).
Sketch the measured values of |r| as a
function of n/n
0
. Here n
0
is the value cor-
responding to |r| = 0. Assume that you
achieve matching at an aperture diameter
of d = 8 mm (standard diaphragm aperture
of the Gunn oscillator).
Questions
1. Compute the value of the transformation ra-
tio n = n
0
using the equivalent circuit dia-
gram according to Fig. 12.2 e) so that the
reflection coefficient becomes r = 0. For
this use approximately X
c
= Z
0
and enter
the result in the form n
0
= f (Q
0
).
2. Compute and plot the characteristic magni-
tude of the reflection coefficient |r| in dB as a
function of n/n
0
in the range 0.1 to 2.5.
3. Estimate the unloaded Q
0
of the resonator
using the experiment results according to 3
and the relationship between the aperture
diameter d and the transformation ratio n in
Fig. 12.3.
Table 12.1
n
0
= ___________
6
7
9
10
r
min
dB
r
min
d / mm
n
n
n
0
8 1
MTS 7.4.4 Solutions
Solutions
Experiment 1 sample solution
1.3 Waveguide width: a = 23 mm
s = 21.8 mm
2.7 See Table 13.1.1 and subsequent
diagrams
3.1-4.2 see Table 13.1.1
Table 13.1.1
with diaphragm
with rear panel
without diaphragm
with rear panel
with diaphragm
without rear panel
0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.5 0.0 44.3 0.0 47.3 0.0 46.5
1.0 0.0 90.0 0.0 90.0 0.0 90.5
1.5 0.0 126.6 0.0 126.0 0.0 126.2
2.0 0.0 153.3 0.0 152.9 0.0 152.7
2.5 0.0 169.2 0.0 169.1 0.0 168.7
3.0 0.0 174.3 0.0 174.0 0.0 174.1
3.5 0.2 156.8 0.0 168.0 0.0 167.4
4.0 19.6 152.9 0.0 163.7 0.0 162.3
4.5 40.1 148.4 0.0 159.6 0.0 159.8
5.0 55.0 145.8 0.0 158.0 0.0 154.5
5.5 66.5 144.5 33.5 151.4 0.0 152.6
6.0 76.0 143.0 30.0 149.3 0.0 151.1
6.5 82.1 142.2 19.7 147.9 0.0 150.3
7.0 86.5 141.0 7.2 147.2 63.2 144.6
7.5 90.1 140.0 0.0 147.3 69.2 143.7
8.0 94.3 138.6 0.0 147.2 74.8 143.1
8.5 95.6 137.3 0.0 146.4 79.8 142.0
9.0 97.6 135.8 0.0 145.1 84.7 140.4
9.5 99.4 134.6 0.0 142.8 90.4 138.4
10.0 100.0 133.1 0.0 141.0 94.7 136.4
U
G
mV
I
G
mA mV
I
G
mA
I
G
mA mV
V
U
D
U
D
U
D
8 2
MTS 7.4.4 Solution
Diagram 13.1.1:
Gunn current and detector voltage as a function of the
Gunn voltage in the case with diaphragm and with rear
wall.
A component which emits power can be consid-
ered as a negative resistance in the equivalent cir-
cuit diagram. This can also be discerned in the
characteristic of the Gunn oscillator. Starting with
the voltage at which power is emitted, the charac-
teristic of the Gunn current demonstrates a nega-
tive differential resistance (negative slope).
Diagram 13.1.2:
Gunn current and detector voltage as a function of the
Gunn voltage in the case without diaphragm and with rear
wall.
Diagram 13.1.3
Gunn current and detector voltage as a function of the
Gunn voltage with diaphragm but without rear wall.
Notes :
The characteristics are subject to strong compo-
nent tolerances (primarily the Gunn diode). Con-
sequently the characteristics shown here are only
meant as examples for orientation.
Sudden discontinuities can arise in the detector
voltage. During these discontinuities the emitted
spectrum of the oscillator varies. For small Gunn
voltages a broad spectrum of frequencies are emit-
ted, while normally only two frequencies exist [a
dominant mode at approx. 9.4 GHz (TE
101
) and a
mode normally attenuated in amplitude by
20 dB at approx. 18.8 GHz (TE
202
)] for higher
voltages. (See also Design of the Microwave
Source from the preface).
Frequently differences can be distinguished as to
whether the characteristic is taken for rising Gunn
voltage or for falling Gunn voltage U
G
. The above
mentioned discontinuities or sudden steps only
arise in the characteristic of falling U
G
in the
range of low Gunn voltage hysteresis effect.
This can be attributed to the fact that more energy
is needed to generate mode distribution (higher
U
G
), than is required to maintain it.
If high reflections are allowed to have an
unattenuated impact on the Gunn diode, then its
general operating response is severely altered. A
measurement curve which has been recorded
without the attenuator has been depicted in Dia-
gram 13.1.4. Extremely disadvantageous in this
case was the fact that a pure spectrum only set in
right under 10 V after the last discontinuity. If af-
terwards you reduce the Gunn voltage again,
mode distribution is maintained until just above
5 V. For that reason it is always favorable to
work with high Gunn voltages, i.e. first increase
the Gunn voltage to 10 V then subsequently set
it to the desired value (e.g. 8 V).
Diagram 13.1.4:
Gunn current and detector voltage as a function of the
Gunn voltage for the case of strong reflections
8 3
MTS 7.4.4 Solutions
Answers
1.
f
a
s
0
2 2
GHz
15
1
cm
1
'
cm
GHz
j
(
,
\
,
(
+
j
(
,
\
,
(

'
. 9 48
2.
U
TH
= 4 V E
TH
= 4 10
3
kV/10
3
cm
= 4 kV/cm
3.
f
s
s

10
10
10
1
10
7
3
10
cm
cm
GHz
/
4. In the current experiment we will use config-
uration B from Fig. 1.2 for the resonator op-
eration of the Gunn element. The resonance
field is mainly located between the stud for
the Gunn element (part 1 in Fig. 1.2) and the
diaphragm with aperture. This means that if
only the housing's rear wall is removed, the
characteristics compared to the case with
rear wall and diaphragm do not change that
much. The situation is different if the dia-
phragm is removed, as now there is no cav-
ity resonator coupled to the Gunn element.
In the case without the diaphragm (experi-
ment part 3) there are only relatively small
discontinuities in the falling part of the
I
G
(U
G
) curve and the radiated microwave
power is also only relatively low. In the case
with the diaphragm (Experiment 2 and 4) a
cavity resonator exists. Here the radiated mi-
crowave power is considerably higher than
in experiment part 3. The onset of the oscil-
lation can be noticed by the powerful and
abrupt drop (discontinuity) in the I
G
(U
G
)
characteristic at U
G
4 V.
Experiment 2 sample solution
Answers
1. In the case of direct modulation of the Gunn
diode, modulation is carried out along the
static characteristic U
D
(U
G
) from Experi-
ment 1, Diagram 13.1.1 (i.e. Table 13.1.1).
If the characteristic has sudden discontinui-
ties, a small change in U
G
results in a con-
siderable change in the emitted microwave
power. This can be observed in points 1.4
and 1.5 (in this context refer also to Design
of the Microwave Source from the pref-
ace).
Normally the maximum receiving signal lies
at a supply voltage of approx. 4 V, meaning
the voltage at which the Gunn diode begins
emitting microwave power and where the
static characteristic begins showing the
greatest discontinuity.
Note:
At a voltage of approx. 4 V the Gunn diode
emits a whole series of frequencies (see pref-
ace). Because spectral purity is normally de-
sired in the following experiments, the
working point must be set to approx. 8 V (or
higher) for Gunn internal modulation. There
the static characteristic is normally very flat
and thus the modulated signal (AC compo-
nent) only very small.
If you have a PIN modulator at your dis-
posal, use it instead as it provides a consid-
erable improvement for generating a mod-
ulated microwave signal.
Diagram 13.2.1 graphically demonstrates the rela-
tionship of the receiving voltage as a function of
the Gunn voltage during direct modulation. For
this the detector was connected to the input of the
SWR meter and the output signal U
Amp Out
of the
SWR meter observed. The maximum value was
calibrated to approx. 0 dB on the scale of the
SWR meter. The diagram confirms the technical
explanation given above.
The maximum from Diagram 13.2.1 coincides
with the discontinuities from Diagram 13.1.1.
Diagram 13.2.1
Output signal of the SWR meter as a function of the Gunn
voltage during direct modulation of the Gunn oscillator
8 4
MTS 7.4.4 Solution
Experiment 3 sample solution
Table 13.3.1
Experiment 4 sample solution
Table 13.4.1: Calibration of the attenuator
a / dB x / mm
0 0.00
1 1.28
2 1.75
3 2.05
4 2.32
5 2.55
6 2.74
7 2.91
8 3.04
9 3.17
10 3.3
12 3.53
14 3.74
16 3.97
18 4.23
20 4.59
: Noise
Responses
1. see column 4 from Table 13.3.1
2. A comparison between columns 2 and 4
shows real good agreement.
This shows that in the current level range a
description of the display a can be obtained
with the relationship
a
U U
U U

( )
,

,
,
,
]
]
]
]
]
10 log
max min
max min
ref
0 0.0 25.6 0.0
0.5 0.9 21 0.86
1 3.0 13 2.94
1.5 6.2 6.2 6.16
2 11.0 2 11.07
2.5 16.5
3 22.0
x
0
cm
a
dB
U x
0
( )
mV
10 log
(0)
0

,
,
]
]
]
]

U x
U
( )
Diagram 13.4.1
Graphic display of the values from Table 13.4.1
(1): Measurement curve recorded with a microvoltmeter
from Rohde&Schwarz
(2) and (3): Measurement curves recorded with two
different coax detectors
The reference values on the attenuator were deter-
mined using a microvoltmeter.
8 5
MTS 7.4.4 Solutions
Deviations from these values are based on the
characteristics of the respectvie detector.
1) Verifying the square law
Specified value on the attenuator used:
3 dB at x
0
= 2.12 mm
Meas. value at x
0
: 3.2 dB
Answer
1. The power can be further reduced using
The out coupling via a slotted measur-
ing line (coupling attenuation of approx.
20 dB). The slotted measuring line is
terminated with low-reflection wave-
guide termination.
The use of a circulator (see MTS 7.4.5
Experiment 8): Feeding at port 1, con-
nect e.g. a 3-screw transformer plus
waveguide termination to port 2 and
couple this out to port 3. With the 3-
screw transformer (+ waveguide con-
nection) a reflection coefficient can be
set. The smaller it is, the lower the
power coupled out at port 3.
The use of a 3-screw transformer as an
inserted two-port to generate a larger
reflection coefficient (around 1 in terms
of magnitude) so that only a small
portion of the power is transmitted.
Disadvantage:
The reflections return to have an
impact on the microwave source.
Experiment 5 sample solution
Experiment procedure
2.4 see Table 13.5.1
2.5 x = 22 mm
3.2 x 6 dB
3.4 see Table 13.5.2


( )
20
0
log
max
U x x
U
U x x
U

( )
0
max
cos
2
0

g

j
(
,
,
\
,
(
(
x x
Display in dB
0 0 0.997 1.000
2 0.4 0.956 0.959
4 1.5 0.837 0.841
6 3.7 0.656 0.655
8 10.4 0.303 0.415
10 16.6 0.149 0.142
12 11.1 0.279 0.142
14 5.5 0.529 0.415
16 3.1 0.702 0.655
18 1.7 0.826 0.841
20 0.7 0.925 0.959
22 0.1 0.987 1.000
24 0.8 0.915 0.959
26 2.1 0.786 0.841
28 5.1 0.556 0.655
Minimum at 10.34 17.7 0.131 0.093
Probe position
x x
0
in mm
Distance of the minima
x / mm= 22

g
= 44 mm
Table 13.5.1
8 6
MTS 7.4.4 Solution
Table 13.5.2
Display in dB
0 0.5 0.940
2 0.2 0.979
4 0.0 1.000
6 0.3 0.971
8 1.0 0.894
10 1.3 0.861
12 1.5 0.845
14 1.6 0.835
16 1.8 0.812
18 1.9 0.804
20 1.7 0.820
22 1.4 0.852
24 1.2 0.867
26 1.3 0.865
28 1.3 0.856
Probe position
x x
0
in mm


( )
20
0
log
max
U x x
U
U x x
U

( )
0
max
Answers
1. From x = 22 mm there follows
g
= 44 mm,
because the maxima and minima have a dis-
tance of
g
/ 2.
2. Relationship between
g
and the free space
wavelength
0
according to Equation (5.14):

g
0
0
2
1
2

j
(
,
\
,
(
a
Resolving the equation with respect to
0
at a = 22.9 mm results in

0
2
44
1
44
45 8
31 7
+
j
(
,
\
,
(

+
j
(
,
\
,
(

g
g
2
1
2
mm
mm
a
.
.
and
f
c

0
0
300
31 7
9 46
.
. GHz GHz
There is very good agreement with the
base frequency of the Gunn oscillator
(9.4 GHz).
3.

ph
0
2
1
2
s

j
(
,
\
,
(

c
a
c
m
0
0
8
1 386 4 16 10 . .


2
0 143 1 43
1 1

g
mm cm . .
4.
8 7
MTS 7.4.4 Solutions
f f
c
a
f f
c c
c c1
m / s
m
GHz
GHz

1
0
8
3
2
2
3 10
2 22 9 10
6 55
2 13 1
.
.
.
5. See 3rd column from Table 13.5.1
6. See 4th column from Table 13.5.1
The measured values deviate from the calcu-
lated ones (sometimes they are larger, some-
times they are smaller). Possible causes lie in
the field distortions caused by the probe.
7. See Table 13.5.2
8. According to Figures 5.2 and 5.3 in the case
of short-circuit termination U
max
is twice as
high as the voltage of the interfering waves,
i.e. a = 6 dB, this corresponds to a volt-
age ratio of 0.5.
If the measured value deviates from this,
the reason is that due to multiple reflections,
the strength of the wave incidenting at the
reflection-free termination differs from the
wave incidenting at the short.
In this measurement example the voltage
varies by 1.9 dB according to the measure-
ment values from Table 13.5.2.
Note:
According to the standing wave ratio
s
U
U

max
min
,
at 1.9 dB this yields a standing wave of
s 10 1 24
1 9
20
,
.
In the best case you can expect a variation of
approx. 0.4 dB. This corresponds to a stand-
ing wave of s = 1.05, i.e. a reflection coef-
ficient of r = 0.025.
Fig. 13.5.1
Sample measurement to point 2.4 and 3.4, detector voltage
logarithmically plotted
Fig. 13.5.2
Sample measurement to points 2.4 and 3.4, detector
voltage plotted linearly
Fig. 13.5.3
Sample measurement to points 2.4 and 3.4, detector
voltage plotted logarithmically.
Spectrum with TE
10
and TE
20
modes.
8 8
MTS 7.4.4 Solution
Fig. 13.5.4
Sample measurement to points 2.4 and 3.4, detector
voltage plotted logarithmically.
Experiment setup with PIN modulator and isolator.
Fig. 13.5.5
Sample measurement to points 2.4 and 3.4, detector
voltage plotted logarithmically.
Use of the diaphragm with slit 2 x 10mm 90. Dominant
frequency at 14.1 GHz.
The curves have been recorded using the CASSY
program and except for curve 13.5.2 the logarithm
taken subsequently.
Fig. 13.5.1 and 13.5.2 (plotted logarithmcially
and linearly) are examples where the spectrum is
dominated by the TE
10
mode, meaning we obtain a
quasi-pure spectrum. This cannot always be at-
tained due to equipment tolerances and local dis-
continuities in the experiment setup. In the rela-
tively compact setup where the total length of the
installation lies in the area of only approx. 5
g
,
the local discontinuities (interference) have a di-
rect impact because they can influence the com-
ponents of the experiment setup almost unattenu-
ated (Gunn oscillator, detector etc.).
Fig. 13.5.3 shows the case where besides the
TE
10
mode also the TE
20
mode predominates. A
prediction of the precise characteristic is hardly
possible as the amplitude relationship of the in-
dividual modes is not exactly known and the
phase angles of the modes are also unknown.
The function characteristic shows a periodicity,
but due to the short measurement line, only a
small portion of this period can be seen. Which
section this is depends primarily on the unpre-
dictable phase shift between the individual
modes, which constantly vary from experiment
setup to experiment setup.
You can expect less problems with the spectral
purity when using the PIN modulator to modulate
the microwave source (in this context also refer to
Design of the Microwave Source from the Pref-
ace). Fig. 13.5.4 shows a measurement curve re-
corded using the measurement setup by adding
the PIN modulator and isolator.
By using the diaphragm with slit 2 x 10 mm 90
instead of the diaphragm with slit 2 x 15 mm 90,
a higher frequency can be separated for a suitable
setting of the Gunn voltage (U
G
4 V, i.e. at a dis-
continuity of the function U
D
(U
G
) from
Expriment 1). Here the Gunn oscillator must be
internally modulated (i.e. without the PIN modu-
lator). The diaphragm with slit 2 x 10 mm 90
represents a filter for frequencies between 13 to
17 GHz. The success of the experiment primarily
depends on which frequencies the Gunn oscilla-
tor generates at the discontinuities. It is required
that only a dominant mode is generated in the
transmission band of the diaphragm with slot. If
this is not the case, you will not obtain similar re-
sults as found in Fig. 13.5.5.
Experiment 6 sample solution
to 2.) See the first line of Table 13.6.1
to 3.2) and 3.3) See the second line in Table 13.6.1
to 3.4) See Table 13.6.1
to 3.5) See Table 13.6.2
8 9
MTS 7.4.4 Solutions

DUT
Location of the
Minimum/mm
s |r|
Short-circuit plate 69.3 1 180
Measurement
object A
76.8 4.2 0.615 5.73
x = 7.5 mm
r
s
s


+
1
1
Table 13.6.1
Note:
The phase primarily depends on the position of
the sample in the sample holder. It becomes
smaller when the sample is moved closer to the
measurement line.

+ 720 180
g
x
l/mm s |r|
Measurement object A 3.4 4.28 0.621
Table 13.6.2

g
= 44 mm Answers
to 1.
Starting from the complex reflection coefficient
r 0.62 e
j57
the related termination impedance results at
Z
Z
j
0
0 85 1 48 + . .
read off the Smith-Chart (similar to Fig. 6.4)
mathematically there results a value of
Z
Z
r
r
j
0
1
1
0 868 1 467
+

+ . .
x U
x U
s
l
l
1
2
2 75 1
2 78 5
min
min
.
.
sin
( )

( )

+
j
(
,
,
\
,
(
(
j
(
,
,
\
,
(
(
mm
mm
1
sin
2
g
g

/degree
9 0
MTS 7.4.4 Solution
Experiment 7 sample solution
3-screw transformer
Table 13.7.1
Short-circuit plate x
0
= 59.8 mm
a) Measurement object without 3-screw transformer
VSWR
s
4 67.5 0.6 54
b) Calibrate to s 1, i.e. |r| 0 (ideally:
s 1, |r| 0) by successively turning all
the screws.
(in the test measurement the optimum of the
calibration was at s = 1.06, i.e. |r| = 0.03)
c) Backwards reflection coefficient (see Fig.
7.1)
VSWR
s
3.9 74.8 0.6 65.4
Minimum at
x
1
/mm


+
s
s
1
1



180 720
1 0
x x
g
The value of

corresponds approximately to
the negative value of f = 54 from the above
measurement of the measurement object
Slide screw transformer
Table 13.7.2
x
0
= 59.8 mm (see above)
a) Matching (s 1.1)
VSWR
s
2.25 71.8 0.385 16.4
Minimum at
x
1
/mm


+
s
s
1
1



180 720
1 0
x x
g
The value of

is only in partial agreement in


the sample solution (here too the magnitude of
is not the same). This can be attributed to the
fact that in this case the optimum has not been
reached, although matching is very good. Ac-
cording to theory the slide screw transformer has
several constellations in which matching is
guaranteed. Furthermore, matching can only
be achieved by setting the metal stud to a very
deep penetration depth in the waveguide. The
result is that the stud no longer functions like a
simple shunt capacitance.



180 720
1 0
x x
g
Minimum at
x
1
/mm
r
s
s


+
1
1
9 1
MTS 7.4.4 Solutions
Fig. 13.7.1
Recording of the detector voltage via the slotted measuring
line when terminated using
(a) short, (b) measurement object A from Experiment 6
Fig. 13.7.2
Detector voltage recorded via the slotted measuring line
when using the 3-screw transformer
(a) no matching, (b) matching at s = 1.06
Fig. 13.7.3
Detector voltage recorded via the slotted measuring line
when using the slide screw transformer
(a) no matching (b) matching at s = 1.05
It is shown that the 3-screw transformer and the
slide screw transformer are well suited to match a
load to a source. However, finding the optimum
setting for the transformers is not always easy and
there are several combinations particularly with
the slide screw transformer. A calculation of the
parameters needed for matching is not possible
without some effort because in practice the metal
studs do not act like simple shunt capacitances.
Experiment 8 sample solution
Table 13.8.1
a) Standing wave ratio and reflection coefficient. No matching.
Antenna type s |r|
A 1.8 0.286
B 4.2 0.615
C 14.1 0.868
Optional :
b) Receiving level at the E-fiel probe in unmatched case. No matching.
Antenna type
A 0 20 3 23 0
B +45 25 2 27 4
C 0 35 0 35 12

opt
V/dB a/dB a/dB (a a
max
)/dB
9 2
MTS 7.4.4 Solution
Table 13.8.2
a) Standing wave ratio and reflection coefficient for matching.
Antenna type s |r|
A 1.2 0.091
B 1.3 0.13
C 1.8 0.286
Optional :
b) Receiving signal level at the E-field probe for matching.
The absolute level is calculated at: a= a V
Antennentyp
A 0 20 3.8 23.8 0.8
B +45 20 3.8 23.8 0.8
C 0 25 1.5 26.5 3.5

opt
V/dB a/dB a/dB (a a
max
)/dB
ated. As a consequence of this, most of the
power must be reflected.
2. With a (theoretically) ideal matching net-
work it is possible to compensate for any re-
flection coefficient |r| < 1 (see also
Fundamentals from Experiment 7).
However, in the case of compensation (i.e.
matching) of a one-port with a reflection co-
efficient in close proximity to 1 you have
the following disadvantages:
(a) Compensation has extremely nar-
row bandwidth.
(b) The loss mechanisms (tranforming
power into thermal energy) which are
always present in real matching net-
works have even greater impact, the
higher the reflection coefficient |r|
needing compensation.
Thus, in accordance with fact (b), although
the input reflection coefficient is zero when
matching for |r| 1, the power of the
incidenting wave is to a great extent not
transmitted further to the load (|r|), but in-
stead converted into heat in the matching
network.
3. The polarisation of the radiated field is not
determined by the direction of the E-field of
the guided wave, but by the direction of the
E-field in the slit. In the case of the slit ro-
tated by 45 (version B), also the direction of
the E-field and thus the direction of the
equivalent magnetic current are rotated
The maximum level here is a
max
= 23 dB (an-
tenna version A, unmatched)
Answers
1. The resonance frequency f
0
of the slit con-
figuration is
f
c
w f
c
w
0
0
0
0
2 2

where
Thus you obtain the following resonance fre-
quency for cases A and B where w = 15 mm
f
w
0
300
2
10
(A,B)
GHz mm

/
whereas in case C with w = 10 mm you ob-
tain
f
0
15
(C)
GHz

When the frequency of the exciting wave


is f 9.4 GHz, the resonance frequency in
cases A and B lies in close proximity to
f, resulting in strong excitation (reso-
nance step-up) of the field in the slit. In
contrast, case C has a very weak field ex-
cited in the slit due to the considerable de-
viation between f and f
0(c)
. Because the
radiated power is proportional to the
square of the magnitude of the electric
field strength in the slit, only a very small
portion of the incidenting wave is radi-
9 3
MTS 7.4.4 Solutions
(see Fig. 8.1 and 8.2).
4. In both cases, matching measures are car-
ried out to compensate for the reflection so
that more power is available to the antenna
for radiation.
5. It is demonstrated that when matching an-
tenna version A (Table 13.8.2 b) ) the receiv-
ing signal level drops, i.e. the advantages
attained through matching (here from s =
1.8 to s = 1.2) are cancelled out again be-
cause the transformer does not operate
loss-free. On the other hand, the matching
of antenna B results in the desired effect.
In the case of antenna C it is demonstrated
that the matching described in question 2, is,
on the one hand, not easy to set (narrow
bandwidth) and the losses in the trans-
former can be significant.
Sample I (Polystyrol) Sample II (Absorber material)
158 19.4
22.1 25.7
0.987 0.9
212 154.3
0.84 j 0.52 0.813 + j 0.39
from the Smith chart
or mathematically
0.007 j 0.29 0.054
178 11.2
30.6 21.7
0.989 0.836
75.5 218.6
0.25 + j 0.96 0.654 j 0.521
from the Smith chart
or mathematically
0.015 + j 1.29 0.1 j 0.347
2.69 j 0.033 11.83 j 0.55
1.86 j 0.017 6.524 j 0.281
0.009 0.043
VSWR s
A
x
A
/mm
r
s
s
A


+

1
1

A
A
g


180 720
0
x x
r r j
A A A A
+ ( ) cos sin
VSWR s
B
x
B
/mm

B
B
g


180 720
0
x x
r
s
s
B


+

1
1
r r j
B B B B
+ ( ) cos sin
Z
Z
r
r
B B
B 0
1
1

Z
Z Z
0
2
A B

r
= '
r
j
r
''
Z
Z
r
r
A A
A 0
1
1

tan

=
r
''/
r
'
Experiment 9 sample solution
Table 13.9.1
x
0
= 24.1 mm x
1
= 68.9 mm

g
= 44.8 mm
0
= 32 mm
9 4
MTS 7.4.4 Solution
Equation for determining the free-space wave-
length

0
g
g
2
1
2

+
j
(
,
\
,
(
a
It is shown that the absorber material has a
greater dissipation factor than the polystyrene
sample, and a considerably larger relative per-
mittivity
r
as well.
Note :
The values obtained for
r
and
r
(i.e. tan

e
) can vary due to manufacturer toleranc-
es. The values provided in the sample so-
lution are only meant as examples.
When measuring the standing wave ratio
of sample I the problem could arise (de-
pending on the coax detector) that the
level cannot be detected in the minimum
because the detector supplies a signal
which is too small and the gain of the
measurement amplifier does not suffice to
make it measurable.
Furthermore, in the case where standing
waves exist with a considerable magni-
tude, it proves difficult to determine the
position and the level of the minimum.
The measurement is then also burdened
with even greater inaccuracies.
If excessive measurement errors arise, an
indicator for this is when you obtain nega-
tive values for
r
. (Keep an eye on the
sign in
r
=
r
j
r

r
= Im(
r
) ).
Experiment 10 sample solution
Table 10.1
1 Round hole
1 Cross-shaped
hole
2 Round holes
2 Cross-shaped
holes
Feed 0 0 0 0
via 38.5 43 36.5 38.8
port 1 27.3 24.4 22.1 18
Feed 0 0 0 0
via 26.7 24.4 21.9 17.8
port 2 36 41.5 41 44
Port 1 : 11.2 18.6 14.4 20.8
Port 2 : 9.3 17.1 19.1 26.2
|S
21
| / dB
|S
31
| / dB
|S
41
| / dB
|S
12
| / dB
|S
32
| / dB
|S
42
| / dB
|a
D
| / dB
|a
D
| / dB
Note: directivity
a
S
S
a S S
S S
D
D
log
log
dB dB

j
(
,
,
\
,
(
(

+

20
20 20
31
41
31 41
41 31
log
It is demonstrated that when the ports are inter-
changed the coupling coefficients are compa-
rable (|S
41
| |S
32
|), while the levels at the
isolation paths differ considerably. This is due
to the fact that the level at the isolation path
(relatively low) responds sensitively to
asymmetries. Even slight asymmetries cause
differences in the levels.The absolute measure-
ment error becomes larger, the smaller the sig-
nal being measured becomes (i.e. the greater
the gain of the measurement amplifier be-
comes).
9 5
MTS 7.4.4 Solutions
Exercises
1. a
k
= 20 dB k = 0,1 (Eq. 10.1)
P
2,out
= (1 k
2
) P
1,in
= [1 (0.1)
2
] 1 W
P
2,out
= 0.99 W
) and )
P
4,out
= k
2
P
1,in
= 0,01 W
) |r| = 0 P
1,out
= 0, P
3,out
= 0
P
3,out
/ P
4,out
= 0
) |r| = 0.5 P
2,in
= |r|
2
P
2,out
= 0.52 0.99 W
P
2,in
= 0.2475 W
P
1,out
= (1 k
2
) P
2,in
= 0.245025 W
P
3,out
= k
2
P
2,in
= 0.002475 W
P
3,out
/ P
4,out
= 0.2475
Sample (energy balance): Dissipated power ab-
sorbed by the four-port
P
diss
= P
1,in
+ P
2,in
(P
1,out
+ P
2,out
+ P
3,out
+ P
4,out
)
= [1 + 0.2475 (0.245025
+ 0.99 + 0.002475
+ 0.01)] W
= (1.2475 1.2475) W = 0 W
Experiment 11 sample solution
Table 13.11.1
Experiment
point 2
3.6 0.66
Experiment
point 3
4.2 0.616
Supplement to experiment point 3
If you want a more accurate calculation and not
disregard k, then k can be computed out of |s
41
|
from Experiment 10. It holds true that:
k
S
10
41
20
.The result out of the sample solution
from Ex10 is k = 0.126 ;thus our calculation is
r
a k

( )
10 0 62
10 1
20
2
log
.
Answers
1. Sample solution from Experiment 6 re-
sulted in |r| = 0.615 resp. 0.621.
The values measured here from 0.66 resp.
0.616 are in very good agreement, but
even larger deviations are acceptable due
to the limited measurement accuracy.
2.
b k a
b k k a r k
4 1 1
3 1
2
2
2
1 2
1 1


from this it follows that
b
b
r
k
k
k k
b
b
r k k k k
3
4
2
1
1
2
2
2
3
4
2
1 2
1 1
1

( ) for
a/dB |r|
9 6
MTS 7.4.4 Solution
Experiment 12 sample solution

0
/ a 32 mm / 22.86 mm 1.4
d/mm d/a n
6 0.262 21
7 0.306 12.5
8 0.35 8
9 0.394 5.5
10 0.437 3.8
d = 8 mm and d/a = 0.35 n
0
= 8
Table 13.12.1
6 1.5 0.84 21 2.63
7 6.4 0.48 12.5 1.56
9 9.5 0.33 5.5 0.69
10 5.75 0.52 3.8 0.48
d / mm
n
n
n
0
r
min
dB
r
min
1.
r
Z Z
Z Z
n R Z
n R Z
n
Q
n
Q
r n n
Q

+

0
0
2
0
2
0
2
0
2
0
0
0
1
1
0
V
V
for

2.
r
n
n
n
n

j
(
,
,
\
,
(
(

j
(
,
,
\
,
(
(
+
0
2
0
2
1
1
3. Q
0
= n
0

where n
0
8 Q
0
201
Diagram 13.12.1
Graphic representation of | r(n/n0) | in dB and measurement
points from Table 13.12.1
97
MTS 7.4.4
Index
Index
Coplanar line ..................................................... 27
Coupling attenuation ......................................... 65
Coupling coefficient ................................... 54, 65
Coupling path .................................................... 65
Cross directional coupler ........................... 65, 71
Short-circuit plane...................................... 12, 59
Moveable short ................................................. 59
L
Transmission line equivalent circuit diagram .... 59
Transmission line resonator .............................. 53
Line wavelength ................................. 28, 34, 54
Leitungswellenwiderstand .................. 28, 39, 43
Optical waveguide ............................................ 27
Lock-in amplifiers ............................................. 20
M
Material parameter ........................................... 59
Multi-screw transformer ................................... 46
Slotted measuring line ....................................... 27
Metal semi-conductor junction.......................... 17
N
Network analyzer ............................................. 72
Zero position search program ........................... 60
Wanted signals .................................................. 19
O
Surface waveguide ........................................... 27
P
Phase velocity................................................... 28
Phase constants ................................................ 28
Phase shift ........................................................ 40
Phasen selectivity ............................................. 21
R
Space-charge instabilities ................................. 11
Noise ................................................................. 19
Rectangular waveguide ...................... 23, 29, 54
Reflectometer ............................................ 65, 71
Reflection coefficient ............ 23, 37, 43, 59, 71
Directivity ......................................................... 66
Directivity factor ............................................... 66
Directional waveguide coupler .................. 65, 71
Backwards reflection coefficient ..................... 43
S
A
Amplitude modulation ....................................... 17
Matching condition............................................ 43
Matching network...................................... 43, 44
Matching point .................................................. 44
Enrichment layer ............................................... 11
Antennas .................................................... 17, 53
C
Cut-off frequency ............................................. 21
D
Attenuation........................................................ 23
Attenuator ......................................................... 23
Detector probe.................................................. 31
Dielectric waveguide ........................................ 27
Relative permittivity ................................... 27, 59
Differential mobility .......................................... 11
Dipole ......................................................... 31, 66
Dipole antenna ................................................ 17
3-screw transformer ......................................... 46
Drift velocity ..................................................... 11
Primary path ..................................................... 65
E
Effective mass .................................................. 11
F
Enhanced field strength .................................... 43
Free space wavelength .............................. 27, 54
Frequency-selective measurement amplifier .... 19
Frequency selectivity ........................................ 21
G
Slide-screw transformer ................................... 44
Critical frequency ............................................. 29
Gunn effect ....................................................... 11
H
Guided wave ..................................................... 17
Cavity resonator ............................................... 12
I
Integrated microwave circuit MIC ................... 27
Isolation path..................................................... 65
K
Node width........................................................ 38
Coaxial line................................................. 54, 66
98
MTS 7.4.4 Index
Slot antenna ...................................................... 53
Slot guide.................................................... 27, 53
Schottky diode .................................................. 17
Signal-to-noise ratio ................................... 19, 53
"Single-mode" operation.................................... 31
Smith chart ................................................. 39, 43
Ridged waveguide ............................................ 27
Standing wave ratio ................................... 38, 49
Scattering matrix............................................... 65
Synchronous rectifier ........................................ 20
T
TEM wave ................................................. 17, 27
Temperature voltage......................................... 17
Transit frequency ............................................. 11
Transmission faktor ..................................... 9, 59
Unloaded Q....................................................... 54
V
Depletion layer .................................................. 11
Delayed domain mode ...................................... 11
Four-port, reciprocal ......................................... 65
VSWR .............................................................. 38
W
Wall losses ........................................................ 12
Wavetype converter ......................................... 53
Real power .......................................... 32, 43, 65
Z
Two-wire line ................................................... 27

You might also like