CONTENTS

TAB

Determining the Air Tightness of a Building Envelope . . . . . . . . . . . . . . . 4

Michael J. Smith

journal

Negative Building Pressure and High Humidity . . . . . . . . . . . . . . . . . . . . . . 7

Associated Air Balance Council

Testing and Confirming Life Safety Systems

Board of Directors and Officers

President
Ralph Meacham
Kahoe Air Balance Company
Executive Vice President
William K. Thomas, P.E.
Thomas-Young Associates, Inc.
Secretary/Treasurer
Rick L. Cox
Technical Air Balance, Inc.
Vice President/Eastern Zone-1
Patrick H. Kelly
American Testing Inc.
Vice President/Central Zone-2
William A. Derse
Professional System Analysis, Inc.
Vice President/Western Zone-3
Robert A. Conboy
American Air Balance Co., Inc.
Immediate Past President
Herman Woody Philhower
Penn Air Control, Inc.
Director, Canadian Chapter
Oscar Horvat
Executive Director
Kenneth M. Sufka
TAB Journal Editor
Sean M. McKenna
Editorial Office:
1518 K Street NW, Suite 503
Washington DC 20005
(202) 737-0202
TAB Journal is published quarterly by
the Associated Air Balance Council. It
is distributed free to AABC members
and by subscription to nonmembers at
$24 per year.

Steve Young
........................

10

Don Butler

Laboratory Airflow: Digital vs. Analog Controls . . . . . . . . . . . . . . . . . . . . 13

Dr. James Rock and Dr. Swiki Anderson

Humidity Control for Indoor Air Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

William R. Beckwith

A Proposed Method for Validating Cooling Coil Performance . . . . . . 20

Joseph Flood

Serie-ous Concerns with Series Fan Powered Boxes . . . . . . . . . . . . . 24

Gary L. Miller

A Model IAQ Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Michael A. Price

Cover illustration by Karen Barnes - Wood Ronsaville Harlin, Inc.

From the Publisher

This issue of TAB Journal features three articles that relate to building pressurization. Mike Smith reports on how his company utilized a depressurization method to determine the air tightness of a building envelope. The
purpose of this exercise was to prevent unpleasant odors from escaping into
the surrounding environment.
In another case history involving building pressurization, Steve Young
takes us through an investigation of a hospital to determine the source of
musty odors. In this case, building pressurization not only provided valuable clues during the investigation; it also helped to solve the problem.
Finally, Don Butler explains the importance of stairwell pressurization
when testing and confirming life safety systems. These are but a few examples of how building pressurization can affect the environment and the
health and safety of building occupants.

TAB Journal is an open forum for the

free expression of opinions and information. The views expressed are not necessarily those of AABC, its officers,
directors, or staff.

We are also pleased to feature Dr. James Rock and Dr. Swiki Anderson in a
far-ranging discussion regarding digital vs. analog controls. Also in this
issue are two articles on Indoor Air Quality, by William Beckwith and
Michael Price.

Letters, manuscripts, and other submissions are welcome. However, TAB

Journal accepts no responsibility for
unsolicited material.

Gary Miller provides us with an analysis of the pros and cons of series fan
powered boxes. Finally, Joseph Flood provides the reader with a new and
innovative approach to validating cooling coil performance. This article
should be of particular importance to all test and balance professionals.

All rights reserved. Copyright 1996

by the Associated Air Balance Council.

We hope you enjoy this issue of TAB Journal. Your comments are always
welcome!

B U I L D I N G P R E S S U R I Z AT I O N

Determining the Air Tightness of a Building Envelope

Michael J. Smith
Thermo Mechanical Services Ltd.

Introduction

s more and more companies,

communities, and individuals become
interested in outdoor air quality and the
environment, test and balance engineers will continue to be asked to
ensure that commercial buildings are
not leaking dangerous liquids or odors
into the air. In northern climates (especially northwest Canada), air tightness
and leakage testing are becoming more
commonplace due to the impact of
severe weather on buildings.
As leaks occur, cold outdoor temperatures and high winds freeze condensation on the building skin, often ruining
the brickwork. While more and more
building walls are deteriorating, governments, insurance companies, and
lending institutions are demanding
tests on buildings prior to completing
a transaction.
Recently, our firm was hired to conduct tests to find leaks in the building
envelope of a sewage sludge processing facility. This facility converts
sludge, the solid portion of the waste
stream that results from the treatment
of municipal sewage, into a dry granular residual material known as a
biosolid. Unpleasant odors are
emitted during the conversion process,
which uses indirect drying equipment
to vaporize the moisture and volatile
organic compounds in the sludge.
Since the facility is located in an environmentally sensitive location, the

development permit required that all

odors be contained inside a leakproof
building envelope.

Odor, gas, or liquid

leaks can be
an annoyance, an
environmental danger,
and ultimately, a
costly problem.
-Michael Smith

The building is adjacent to other structures and is in an area subject to high

and varying wind conditions. Because
of the stringent requirements and varying atmospheric conditions, the developers had to test the building envelope
to confirm its integrity. The air handling system in the building is capable
of containing all odors when the processing equipment is in operation. If
the building is shut down or in fail
mode, however, odors can escape into
the atmosphere.

Test Procedure
This facility is a large steel frame
structure encased in precast concrete
wall panels. There are many access
and intentional openings in the building envelope. We employed the
depressurization method using the
buildings own air handling systems to
determine the air tightness of the
building envelope. This method actually allows the test engineer to look

(from the inside) for leaks into the

building rather than looking for leaks
outside in a windy, less precise environment. The method is prescribed in
the Canadian National Standards for
all commercial buildings and has been
used successfully on many other projects.

Principal
The buildings exhaust air system is
the primary environmental assurance
system, as it draws all the buildings
air through an odor scrubber and
exhausts clean air into the atmosphere.
Air flows into the building through
loading doors, intentional openings for
motorized dampers, and the HVAC air
systems.
During testing performed while the
processing equipment was in operation, the odor scrubbers were set to
deliver full design constant-volume
airflows at a velocity required to
remove all odors (each scrubber can
move 50,000 CFM [23,600 l/s]).
Overhead loading doors were closed,
and the intentional openings at inlet
dampers were operated to maintain a
constant building pressure of 25 pa.
To determine the tightness of the
building envelope, all HVAC equipment was shut down. All overhead
loading doors were closed, and inlet
dampers were set to maintain a test
pressure of 50 pa. Smoke was then
generated and introduced to all building joints, connections, and uninten-

TAB Journal

tional penetrations in the building

envelope. Airflow rates and the corresponding pressure differences across
the building envelope were measured
and recorded. The air leakage characteristics of the test were determined by
plotting the airflow rates against the
corresponding pressure differences.

Test Apparatus and Setup

The pressure-measuring apparatus
(graduated 0 to 100 pa) was set up
outside the operating engineers con-

trol office. Due to possible wind tunnel effects, all exterior exposures were
fitted with exterior pressure taps connected to a cylindrical pressure-averaging container, then to the pressuremeasuring instrument.
The operating engineers office is
manned 24 hours a day, and any variations in space negative conditions can
be monitored and adjusted to maintain
a constant negative pressure inside the
building regardless of outdoor conditions.

Airflow through the odor scrubbers

was measured constantly and was corrected for temperature at test pressures.

Test Results
The test results indicated that some
openings, such as overhead doors,
equipment exhaust pipes, and maintenance access doors, could not be
sealed to a 100% leak-free condition.
To overcome this problem during normal operation, the air inlet dampers
had to be controlled by a space pres-

Sketch #1

BOILER STACK
WELD TO STACK
ROLLER SEAL
ROOF

Sealing Major
Leaks in the
Building Envelope
(See Sketches 1 7)

Stack Rain Cap Flashing

Sketch #2

Sketch #3

PARAPET FLASHING
PRE-CAST PANEL

HINGED DOOR

STEEL ROOF DECK

GASKET

REQUIRE SEAL

ROOF

Wall at Roof Joint

TAB Journal

Access Hatch

sure sensor to maintain the desired

negative pressure. It was recommended that an emergency exhaust system
consisting of an exhaust fan, charcoal
filters, and an emergency generator be
installed to maintain the building at a
25 pa pressure during a building
power failure or a maintenance shutdown procedure.

Conclusion
Following our findings and recommendations, all major leaks found in

the building envelope were sealed (see

Sketches 1 through 7) by a sheathing
contractor.
While the problems of high winds and
severely cold temperatures may be
unique to northwestern North
America, the importance of building
air tightness is universal. Odor, gas,
or liquid leaks can be an annoyance,
an environmental danger, and ultimately, a costly problem. Performing
a relatively simple test of a buildings

envelope can assure everyone involved

from the owner to the contractor to
the local or state government that
the surrounding community is safe
from whatever environment is contained inside the structure itself.

Sketch #5

Sketch #4

SEAL OPENING
SCREEN

1" CONDUIT

SILENCER

SEAL WITH
SILICONE

MOTORIZED
DAMPER

LIGHT

SEAL AROUND
MOTORIZED
INLET DAMPER
ROD

Inlet Louvers

Sketch #6

EXTERIOR WALL

Electrical Connection Seals

Sketch #7
INLET LOUVER

EXTERIOR WALL
SEAL

OVERHEAD DOOR
SCREEN
DUCT

NEOPRENE DOOR
SWEEP SEAL

MOTORIZED
SMOKE DAMPER
SEAL

ANGLE OR STRIP IRON

FULL LENGTH

Overhead Doors

EXTERIOR WALL

HVAC Intake Louver

TAB Journal

B U I L D I N G P R E S S U R I Z AT I O N

Negative Building Pressure and High Humidity

Steve Young
The Phoenix Agency

building under a negative

pressure (depressurized) is generally
undesirable, especially in a warm and
humid climate. Anyone who has studied indoor air quality problems is certainly aware that high humidities foster growth mediums for mold and
mildew. If you have ever walked into a
building in the summer and felt air
rushing in with you, then looked up
and noticed water drops on the lobby
air conditioning grilles, you have been
in a building that was out of balance
and depressurized.
The phenomenon is very simple
there is more air leaving the building
than is mechanically being brought in,
thereby inducing air from the outside.
The induction of the air is generally
indiscreet. When a vessel (in this case
a building) is under a pressure (in this
case negative), the pressure is exerted
equally on all sides. With a building
under a negative pressure, air will
infiltrate through every crack and
crevice in order to satisfy the negative
condition.
The building pressure is dependent
upon the differential volume of air (the
amount of air which must be brought
in to arithmetically neutralize the
excess air being exhausted) and the
area of the cracks, crevices, and openings through which the differential
volume of air must pass. For example,
a small differential volume with large
cracks (poor window seals) will result

TAB Journal

in a negative building pressure close to

neutral, but if you seal all of the windows without changing the differential
volume, the negative pressure will
increase. Clearly, the amount of pressure is not necessarily an accurate
indicator of a buildings condition.

The Hospital Case

Recently, we were asked to investigate
the cause of musty odors on the
ground floor of a hospital. This hospital is a very large, well-established
institution, and the ground floor
encompassed seven different buildings
that have been merged together over
the 50-plus years of service to the
community. The problem was at the
south end of the complex, which is
shaped like a T. This southernmost
building is single story and has a partial basement. Traveling south down
the main corridor towards the building, the musty odor became apparent,
even before reaching the fire doors
that separate this area of the hospital.
The corridor runs north to south across
the top of the T, with the base of the
T facing west.
As we walked south, the odor intensity
increased all the way to the south end
of the corridor, which terminated at an
exit. We then walked down the base of
the T to the west, where the odor did
not seem as strong (one of the problems with using your nose as an
instrument is that it becomes desensitized very quickly and your ability to

smell decreases with time). With this in

mind, we walked outside to observe the
exterior construction of the building.
The one-story building has a flat roof
with a soffit overhang. The soffit was
vented at one time, but we were told
that the hospital had sealed the vents
because the ceiling tiles inside became
heavy with moisture and were falling
out of the ceiling grid. The sealed sof-

With a building under

a negative pressure, air
will infiltrate through
every crack and crevice
in order to satisfy the
negative condition.
-Steve Young

fits had solved the problem for the

most part, but some warped tiles were
still visible. The south end of the T
has a vented crawl space, while the
north end and base of the T have a
full basement. The basement is used
for material supplies and is connected
to the other parts of the hospital basement through a corridor system. It is
not a true basement; it is more of a
lower level with a loading dock and
driveway to the material storage area.
The building is concrete block, and the
air conditioning systems consist of
chilled and hot water fan coil units
around the perimeter of the building; a

central unit serves the corridors and

intersection of the top of the T. This
building had been converted from a
pediatrics ward and is now used for
offices.

Looking for Pressure

Clues
Continuing the investigation of the
building, we inspected the crawl space
below the south end of the building
through an access door from the basement. Immediately after opening the
door, a strong musty odor was present.
It was obvious we had located the
source of the odor, but how was the
odor present even beyond the fire
doors to this building?
The chilled and hot water pipes run
through the crawl space and up to the
vertical fan coil units mounted on the
floor around the perimeter of the
building. There are gaps around the
pipe penetrations, along with other
cracks and gaps in the flooring system,
which allow for infiltration. However,
the cause of the infiltration needed to
be determined.
Taking a differential pressure at the
south doorway, we discovered a pressure of .003 from inside to outside.
The wind outside was relatively calm
and the pressure represented an average of ten consecutive readings. We
next took a differential pressure across
the closed fire doors that separate this
building from the rest of the complex
and discovered the main hospital had a
.004 pressure. The pressure had
increased when we closed the fire
doors because we had reduced the
opening through which the air could
infiltrate.
This testing revealed that the negative
pressure source was in the other area
of the hospital. We now knew where

the odor was coming from, how it was

entering the building, and why you
could smell the odor beyond the area
of the source.

and immediately discovered the source

of the problem. The original ventilation design used one power roof venti-

So what next? Obviously you treat the

source of contamination, but the main
mission had become location of the
source or sources of negative pressure
and reversing the condition. We
reported our findings to the director of
our investigation at the hospital and
highly recommended that the source of
the problem be located and corrected.
During our conversation, we discovered that the main entrance lobby and
employee entrance had noticeable condensation problems during the summer
months. This information only further
confirmed our opinion that the source
of the problem was in the main part of
the hospital.

We now knew where

the odor was coming
from, how it was
entering the building,
and why you could
smell the odor beyond
the area of the source.
So what next?

With the go-ahead to continue our

investigation, we began to measure the
outside air intake and exhaust air
quantities on all equipment serving the
ground floor. This was a time-consuming task, but when all of the measurements were complete, the tally indicated that the ground floor was positive,
almost 1500 CFM. How could the
ground floor have a net outside air
intake of 1500 CFM more than the
exhaust, yet be .0044?
The main part of the hospital has seven
stories, three service elevators, and two
passenger elevators. This building is
not tall enough to create a large stack
effect, but after we discovered that the
ground floor should be positive, we
measured the pressure from the ground
floor elevator lobby to the elevator
shaft. To our amazement, we measured
.0055 into the shaft.
We then went to the elevator penthouse, which serves all five elevators,

-Steve Young

lator at 3500 CFM to cool the equipment room. With the fan on the roof, a
30 x 30 intake is located about five
feet off of the floor to cross-ventilate
the room. In addition to this ventilation, two additional propeller exhaust
fans had been added to help cool the
penthouse.
To further help the cooling, a piece of
plywood had been placed over the 30
x 30 intake. With the outside air
source for make-up cut off, all the
exhaust air was being pulled up
through the top of the elevator shafts,
through the cable and electrical penetrations. We measured 8.42 square feet
of openings with an average velocity
of 855 feet per minute. The elevator
penthouse was being cooled by the
exhausted, conditioned air out of the
hospital.

Restoring Neutral
Pressure
As noted in the beginning of the article, a building under a negative pressure will induce air through every
crack and crevice to satisfy the pressure. In this case, the central elevator
shafts acted like a large plenum,
depressurizing each level equally. The
upper levels, however, are well sealed,

TAB Journal

leaving make-up air for the exhaust to

come from the lower levels where
there are openings to the outside. The
south end of the ground floor was the
achilles heel in this event, as the
infiltrated air carried the musty odor of
the crawl space.
As a temporary solution, the two propeller fans have been turned off, and
the plywood removed from the outside
air intake in the elevator penthouse.
This reversed the ground floor pressure to +.0013 w.c. We recommended
that the penthouse ventilation be

out, however, the cost of our investigation had a payback which was easy to
calculate.
Not only did we thwart possible IAQ
complaints, lawsuits, etc. which cannot have an estimated dollar amount,
we saved energy.
We measured approximately 7200
CFM of condition air being needlessly
exhausted out of the hospital complex.
It costs approximately $1.50 per CFM
to heat, cool, and dehumidify outside
air in a hospital. Since the facility
operates 24 hours per day, 365 days

per year, a quick calculation reveals

that we saved $10,800 per year, which
results in a payback in less than three
months for the cost of our services.
Not every case will have a payback
this fast, but correcting any depressurized building will have a payback,
even if it is only in saved building
materials or comfort. The overall
building balance is one of the most
important aspects of the building environmental envelope. Maintaining this
balance is one key element to an efficient, healthy building.

The cost of our

investigation had
a payback which was
easy to calculate. Not
only did we thwart
possible IAQ complaints,
lawsuits, etc. which
cannot have an
estimated dollar
amount, we saved
energy.
-Steve Young

revised by turning one of the propeller

exhaust fans into a supply fan, which
would maintain a high air exchange,
while keeping the pressure close to
neutral, thus eliminating the effect on
the building.

Real Savings
This experience shows one example of
how a depressurized building causes
problems that can be far reaching.
Fortunately, there were no formal
complaints or sick employees. The
problem was identified in a one-day
investigation, but it took us over a
week to track down and locate the
source of the problem. As it turned

TAB Journal

LIFE SAFETY

Testing and Confirming Life Safety Systems

Don Butler
Butler Balancing Company

ife safety systems are an integral

but often forgotten part of the building
after construction is complete and the
customer has occupied the building.
Life safety is a term that applies not
only to commercial buildings, but also
to homes. How often do we hear of
tragic house fires where smoke detectors were either absent or without batteries? Although our industry does not
perform a vast amount of residential
work, we should remember that
portable, battery-operated smoke
detectors are a simple but vital part of
a residential life safety system. By
changing the batteries on a regular
basis and activating the test button, a
simple version of a life safety check is
performed on a residence.
Commercial life safety systems, however, are much more complex. A commercial life safety system can include
smoke evacuation fans, stair tower
pressurization fans, fire and smoke
damper systems, the sprinkler system,
and the alarm signal to emergency
response personnel. As technology
moves at a record pace, however,
problems arise in the form of bugs in
the operating systems or smaller staffs
as computers replace people.

Benefits Of Testing
As in the test and balance of an HVAC
system, the life safety system needs to
be checked by a qualified independent
firm. As has often been said of the

10

testing and balancing of HVAC systems, life safety checks are also the
cheapest insurance policy owners can
have for their systems (of course, an
insurance company may offer a better
rate to the owner if it knows that the
life safety system has been confirmed
as operable). Local fire departments
are also pleased when systems are
checked, and they tend to be more
cooperative when they feel that the
building owner cares about accident
prevention and safety.

As has often been said

of the testing and
balancing of HVAC
systems, life safety
checks are also the
cheapest insurance
policy owners can
have for their systems.
-Don Butler

Proper testing of life safety systems is

also imperative because readings from
the system can sometimes be misleading. For example, the computer might
indicate that the VAV is operating and
the airflow is correct. A look at the
box shows that the velocity sensor is
reading the airflow at the box, but
closer inspection might reveal that the
downstream duct work is incomplete
or that the flex has blown off and the
air is blowing into the ceiling.

Another example is the signal from a

fan in the penthouse and its corresponding signal to the command center. The fact that the signal in the command center indicates the fan is running properly is not enough. The signal may be coming only from the
motor; the belt may have broken, or
there may be excessive duct leakage.
The point is that the owner should be
able to prove that he or she had the
system tested in an honest attempt to
prevent disaster. The system could still
fail, possibly (though not likely) within hours of a proper test, but making
an attempt to provide maximum safety
is the owners and, by extension, the
test engineers responsibility.

Testing Methods
Initial


Review the design documents,

confirming with the owner that the
drawings are up to date with changes.
Also review the control sequence and
verify which systems are part of life
safety.

Ask the original design engineer

about the system and discuss your
function as a test engineer. Often there
are matters that need special attention,
such as a punch list item that may not
have been completed by the installer
or design readings from a balancing
report that have yet to be addressed.
The design engineer may also know of
a change in the sequence that was
never put into the final documents. He

TAB Journal

or she may even wish to witness the

tests to see how the product is working
a few years after installation (design
engineers take the manufacturers
word that the system will work as
promised in the literature, and they are
often eager to see it in operation).


Talk to your customers and tell

them what you plan to do. Ask to have
their personnel take you through the
building or have them assist you in
certain tests. After all, because it is
their building, they can tell you the little items that may take you hours to
find (if you find them at all). If the
building is a high rise of 30 or 40
floors, you will need their help simply
to confirm that the doors remain
closed during the tests.

Obtain the building codes for the

area that you are testing. If you are in
a building that was built prior to the
latest standards, confirm the pressures
that you are expected to use for pass
or fail.

On-Site Testing


Verify that the systems are completely installed. If the system you are
testing is in a stair tower, walk the
tower from top to bottom. Look for
the obvious missing door sweeps, gaping holes due to pipe penetrations, or
debris or material stored in the tower
or behind an emergency door. Visually
inspect the fan; again, look for the
obvious, such as a cracked fan belt,
improper duct connections, torn flex,
or signs of leakage (such as dirt lines
coming from the joints). Remember,
you have not started the system yet.
You are only looking for obvious
problems.

Next, call the local fire department to inform them of your test and

TAB Journal

to avoid a false alarm that could be

triggered by the fans operation. Give
a rough estimate of your time frame
and tell them that someone is ready to
give them verification should an actual
alarm go off due to a fire in another
part of the building. Also notify them
when you are finished and the system
is back to its normal mode.


With the cooperation of the building personnel, energize the system.

For stair pressurization, we do the following, unless specifically instructed
otherwise:

Check the fan for amperage

design versus actual, rpm actual, and
sheave pitch diameter. This establishes
a written record of the readings in case
there is a problem down the road.

elderly person would be unable to

open the door under actual operating
conditions. With the permission of the
owner, we asked a person walking
down the hall to try to open the door
while the fan was running. She was
glad to help, and she opened it with
ease. We confirmed not only that the
system worked to design specifications, but also that it could be entered
under operating conditions.)

Smoke Evacuation
The smoke evacuation system can
consist of several types of design,
including a separate exhaust system
dedicated solely to the removal of
smoke in case of an emergency and an
air handling system that, when activated in emergency mode, will go to full
relief, drawing from the return system.
The latter is the most difficult to test.

As a starting point, test the top,

middle, and basement floors. This will
indicate an immediate problem,
depending on where the fan discharge
is located. If the number of floors
equate to it, also test every third floor
from top to bottom. The test is intended simply to measure for the positive
pressure differential from the tower to
the door leading to the tenant space.
Confirm that, if relief dampers are in
the tower, they are relieving the pressure at the proper limit.
Once these initial tests have been completed successfully, proceed to the
floors tested and check the door opening force. People in a panic, especially
the very young and the elderly, must
be able to open the doors against the
pressure in the tower.
(As a side note, we once tested a large
hotel with a member of the local fire
department. Although the system had
passed, he was concerned that an

Testing of any smoke evacuation system is problematic for several reasons.

First, an actual smoke test is difficult
and usually undesirable, especially in
a hotel environment. Smoke tests are
also performed at room temperature;
in a real fire, the heat factor could certainly alter the characteristics of the
smoke. Second, putting the air handling system under full smoke removal
conditions for any period of time can
cause discomfort for tenants.
Depending on outside conditions, the
protection of coils also needs to be
considered.
For the actual testing, perform all of
the fan testing as described earlier in
the stair pressurization test for recordkeeping purposes and carry out the
same precautions as in the pressurization tests.

11

Smoke Tests
First, evaluate the area to be tested for
the amount of smoke to be generated.
If we are simply igniting a smoke candle in a hall or room, we set the candle
in a metal bucket, which is then set in
a larger bucket filled with water to
help absorb the heat. It is a good idea
to keep a bucket of water nearby to
douse the candle if the amount of
smoke is too great or if the smoke is
still being generated after the test is
completed. Also confirm that your
candles are non-toxic and will not
damage the area.
I recommend having a video camera
and stop watch. Whether the time limit
is 30 seconds, 1 minute, or 3 minutes
(which most fire departments consider
the maximum), you need to be accurate for the actual test. If time documentation is critical, I may hold a
newspaper up in front of the camera to
verify the date or ask a representative
from outside of my firm to verify the
correct date on film. Some may consider this overkill, but documentation
in this type of testing is critical.
When an existing building is being
smoke tested, careful consideration
must be given to the owner and tenants. If notification has not already
been given, try to avoid this test. A
panic could result if people see or
smell the smoke and are unaware of
the test. (An interesting example of this
is the first test that our organization
ever conducted. It was in a new tunnel
on the Pennsylvania Turnpike. Smoke
removal was to be confirmed prior to
opening, and the testing was being
performed the morning of the opening
ceremonies. We were able to record
the smoke removal with and without
the fans running. Everything went

12

well, except that during the final test,

someone removed the barricades at
one end of the tunnel and let the traffic
through. Although the story is amusing, a terrible disaster could have
occurred simply because of a lack of
proper safety precautions.)

Airflow Tests
First, remember that damper operation,
control sequence of operation, etc.,
must still be considered.
Airflows can be confirmed under
proper mode by either traverse or terminal readings; testing at both is recommended. A traverse will confirm
the capacity, but the bottom line here
is whether or not the space is removing the air. This is where duct leakage
plays a major part in whether a system
passes or fails the test.
Keep in mind that the return may have
throttled terminals due to a prior balance. This should not be a problem if
the total CFM required is equal to the
return requirements. If there is a difference, however, the customer may
require some readjustment to the system so that both will be satisfied.

of long-term operation; this is precisely why annual checks are so critical. If

the customer simply wants readings
without the fan information and much
of the preliminary data, state that this
was how the testing was contracted.
Send documented problems to the
customer immediately, even if the
problems are noted in field report form
prior to the final report. Remind your
customers that you are not trying to
condemn them or their business
directly; in truth, you may actually be
saving them thousands, if not millions,
of dollars that they would have to pay
if the life safety system failed to operate during a disaster.
Life safety testing is time consuming
and requires patience from everyone,
including you, your employees, the
building owner, and tenants. Working
around a customers schedule can
cause problems and lost time; some of
the testing may have to be conducted
during off hours and may require a
time and material contract. The work
itself is often tedious; however, the
potential payoff is an operable system
that will perform properly under truly
trying circumstances.

Documentation
Times, dates, system conditions, operating modes, and the names of any
witnesses should be carefully documented. When selling this type of
work, stress that your recommendations to owners will carry more weight
than recommendations from their own
maintenance personnel, whose recommendations may simply go on a to
do list and be completed as part of
the overall maintenance plan.
Most importantly, stress that the testing is accurate for the day of the test,
but that you are not giving a guarantee

TAB Journal

IN BALANCE

Laboratory Airflow: Digital vs. Analog Controls

Dr. James Rock
Texas A&M University

and

Dr. Swiki Anderson

Accu*Aire Controls Inc.

The following is a discussion

between two experts on an
issue of interest to the test and
balance industry. It is mediated
by the editor of TAB Journal.
Dr. Rock is a professor in the
Nuclear Engineering Department
at Texas A&M University.
Dr. Anderson is vice president
of Accu*Aire Controls Inc.,
a manufacturer of laboratory
and temperature airflow controls.

It is worth noting that the world is

analog (that is, continuous). Every
digital system has an analog to digital
converter to change sensor signals
(input) into digital numbers, a digital
computer in the middle, and a digital
to analog convertor (output) to move
the damper or change the motor speed.

TAB Journal: Today well be discussing laboratory ventilation equipment, and our emphasis will be on the
design and maintenance of VAV flow
controllers for hoods equipped with
sash position sensors. There are two
groups of manufacturers: those who
advocate using analog control loops
and those who sell only digital controllers. First, lets define the terms
analog and digital.

Rock: Digital is more complex, but

complexity is not always bad, and digital need not be a lot more expensive
in first cost. The mass production of
the necessary digital components has
reached the point where they are
priced as commodities. Manufacturers
are able to choose between digital and
analog components for almost any part
of the ventilation control system.

Rock: An analog system is one that

functions on a continuous basis. A
digital system functions on a discrete
basis. The analog system continuously
measures flows or pressures in a duct
and commands appropriate smooth
changes in damper position or motor
speed to control exhaust flow. A digital system converts its measurements
into a number for a computer. The
computer typically creates another
number that sets one of a few allowed
damper positions or motor speeds to
produce an acceptable approximation
to the desired set-point flow.

TAB Journal

TAB: You make it sound like the digital system is more complex than the
analog system.

TAB: We seem to be talking about an

important concept. What, exactly, is a
controller, or a control system?
Rock: A ventilation controller or control system moves dampers or changes
fan speeds to automatically maintain
flow somewhere in the system. There
are two classes of control systems
found in laboratory ventilation applications. Well call them feed-forward
(or open-loop) and feed-back (or
closed-loop) controllers.
An important subset of the open-loop
controller is the model-predictive con-

troller, which uses a model of the

physical world to achieve desired control. Feed-back controllers usually
include open-loop components, and
high-quality open-loop components
are of the model-predictive variety.
Anderson: Youre right, Jim, but
some of us need examples. I think
what youre saying is that open-loop
control is implemented in mechanisms. Two examples of open-loop
controls come to mind. First, the
Mitco spring and cone type valve, in
which the geometry of the valve cone
and seat attempt to make flow through
the valve proportional to actuator position. Second, the early attempts to
control hood exhaust flow with sash
position were realized by connecting a
rack and pinion gear to the hood sash
with a mechanical cam to linearize
exhaust flow through a single-bladed
damper.
Rock: Great, Swiki. The Mitco valve
is an excellent example of a modelpredictive variety of the open-loop
control. The model is implemented in
the shape of the cone and valve seat.
Your second example perfectly illustrates the concept of a feed-forward
(or open-loop) control. This is used to
control flow at one point in the system
based on a sensor located somewhere
else.
Your example is perfect: a rack
attached to the laboratory hood sash
moves with the sash, and a pinion gear

13

turns a shaft. At the other end of the

shaft, an eccentric cam arrangement
moves the damper the right amount to
obtain the desired flow. Feed-forward,
or open-loop, systems can be designed
to achieve any kind of commanded
flow patterns.
Anderson: Im biased. I have never
seen an open-loop control work well
in a laboratory airflow control system.
In practice, we see overloaded filters,
misaligned cams, corroded damper
and fan blades, and slipping and misaligned belts. Connect a feed-forward
device to position actuators in such a
system, and the wrong flow results!
Rock: Thats why we teach our students to prefer the feed-back controller.
The feed-back controller uses a flow
sensor in the exhaust duct that sends a
signal to be compared with the setpoint flow.
With feed-back control, we no longer
worry about the calibration of the flow
controller. In fact, the feed-back controller can correct for modest amounts
of dirt, corrosion, slipping belts, etc.,
on flow control mechanisms. With a
reliable measure of face velocity, feedback controllers can correct for limited
mechanical malfunctions until the next
scheduled preventive maintenance!
Anderson: I agree, but I also realize
that it isnt practical to measure and
control based on hood face velocity.
If we installed a thermal anemometer
rake in the hood sash opening to measure face velocity, no one could use
the hood. Theyd poke their eyes out.
Instead, we have to measure sash position and use that to control both
exhaust flow and, in an indirect way,
face velocity.

14

Measured Flow

Exhaust
Flow
Sensor

Flow
Error

Predict
Flow
Set-point
Move
Actuators

Fan Speed
& Damper

Sash
Position
Sensor

Figure 1. This sketch shows the different kinds of controllers working together to control hood
flow. The Sash Position Sensor feeds a model-predictive controller to create the exhaust flow setpoint. The Exhaust Flow is subtracted from the Set-Point Flow to create the closed-loop controller
error. The error feeds forward to move damper position and change motor speed to maintain setpoint flow. When exhaust flow is too large, the damper is closed and the fan is slowed; when
exhaust is too small, the dampers are opened and the fan is accelerated.

This means that feed-back controllers

depend on the TAB technician, who
must use careful face velocity measurements to calibrate the sash position sensors. In practice, this is done
with the feed-back controller turned
on. The sash is opened, and the controller is adjusted to produce the
desired face velocity. Once that calibration is done, the technician checks
face velocity as the sash is closed to
ensure that the controller maintains
proper face velocity. Since the setpoint flow is controlled by sash position, the proper functioning of the control loop should be checked for calibration at least annually per NFPA 45.
TAB: How does all this relate to the
question about analog vs. digital
equipment?

Rock: The simple fact is that all components of any ventilation control system can be either analog or digital.
Take a look at my sketch (Figure 1).
The sash position sensor creates the
set-point for exhaust flow. If the flow
is too low, the controller sends a signal
that moves the actuator to increase the
exhaust flow. If the flow is too high,
the controller sends a signal to reduce
the exhaust flow. When the flow is set
properly, no actuator motion is commanded.
The nice thing about this concept is
that the controller doesnt need to
know anything about how far to move
the actuator; it only needs to know
which way to move it. Ultimately,
cost and performance must determine
which type of system must be used.

TAB Journal

Table #1

Analog vs. Digital Flow Controllers

Similarities

Differences
Analog

Digital

Both have Analog sensors

Continuous control

Intermittent control

Exhaust flow is Analog, not Digital

Hierarchy of nested
control loops

Sequential servicing of
control loops

Make-up supply air is Analog

Diagnose with
multimeter

Diagnose with software

Both use Analog (physical) actuators

Repair by simple
module replacement

Repair by module and

software replacement

Both can be applied in model-predictive,

feed-forward, or feed-back applications

Technology stable;
long life expectancy

Technology rapidly
evolving

Table 1

TAB: Be specific. When should each

type of controller be used?
Rock: The closed-loop (or feed-back)
controller is the best option. It is more
tolerant of system changes, including
normal wear and tear, than the other
options. In some designs, including
the one I sketched, the closed-loop
performance is improved by including
model-predictive or open-loop subsystems or components. In such systems,
the TAB technician should calibrate
the open-loop valve linkages prior to
adjusting the controller itself.
TAB: Dr. Anderson, I understand your
company makes an analog control system. Why did you choose analog
rather than digital?
Anderson: We selected analog over
digital controllers for the following
reasons:

TAB Journal

1. While digital hardware and software

are part of a rapidly changing technology, analog technology is stable.
When digital technology changes
which happens about every three
yearsmaintenance and repair
become serious and expensive problems.
2. Calibration and commissioning of a
controller require some way to read
hood sash position and flow to set the
controller. With a digital system, this
is accomplished with a special-purpose
computer terminal indirectly reading
something through a series of components. With the analog system, setup
and calibration are accomplished using
an ordinary digital volt meter and simple hand tools. The TAB technician
can read the signals from each transducer.
3. While digital is very flexible and
can be reprogrammed by vendors, it
isnt necessary in a laboratory hood

control loop. Control functions and

gains dont change in a simple hood
loop.
4. We build our analog controllers
with a very small parts count, resulting
in lower first cost and commissioning
costs than for competing digital controllers.
5. We have continuous monitoring of
the flow set-point with an analog control system. With some digital systems, the cycle times are far too long
for satisfactory control response.
Analog eliminates this.
6. For safety and reliability, we have a
dedicated analog controller for each
control loop.
7. Our modular analog control system
is designed to control both the room
supply air and the hood(s) exhaust air
to ensure continuous negative pressure
in the room.

15

8. Building systems have a long life

expectancy. Because the control is
modular, it allows flexibility to adapt
to any system configuration. Building
operators do not like to change control
hardware or software every two to
three years, but with some digital systems this is exactly what happens

From what youve

both said, it appears
that you agree there
is an optimal solution:
single closed-loop
analog controllers
monitored by a digital
building system.
-TAB

when a part that is no longer produced

needs to be replaced. Look at whats
happening with software.
Rock: Swiki, some buildings systems
are already digital. In fact, most modern buildings are designed around digital systems, and there are valid engineering reasons for this:
1. The parameters in a digital system
are easily set and do not drift or age,
as occurs in most analog systems.
2. Non-linear predictive models are
easily implemented in digital systems,
often involving only simple look-up
tables. Such models have been accomplished in analog systems using
mechanical cams and complex, operational amplifier circuit boards.
3. A building-wide digital network
makes it relatively easy to modify portions of a system as needs change.
Typically, a change in a few numerical
constants is all thats required for the

16

digital system to recognize new components (such as an additional hood)

or new conditions (such as revised
temperature or humidity set points).
4. The central computer can be easily
programmed to control both the hood
exhaust flow and the room supply airflow to maintain room-negative pressure.
5. If the building is so large that the
delay between sequential servicing of
rooms causes a loss of control, it is
easy to install a computer chip in each
room for local control.
6. Every component of a feed-back
building control system, including the
open-loop and feed-forward, modelpredictive subsystems, can be realized
in digital form. There is no longer a
true need for analog components.
In light of this information, does your
analog approach limit your companys
market to older buildings?
Anderson: All of our control modules
are designed to interface with any
building digital system. This combination produces the best of both
worlds: inexpensive continuous local
control and intermittent digital monitoring. The digital monitoring can be
programmed to give a rapid warning
whenever a control loop needs attention.
This encourages just-in-time preventive maintenance to prevent hazardous
incidents. If a digital system can read
010 Volt DC signals, interface is
easy. Analog to digital interface cards
are widely available, so we routinely
interface our controllers with all kinds
of digital systems in older and new
buildings.

Rock: Do you agree that an all-digital

system can do the job?
Anderson: Yes, but with a warning.
The first cost of analog components is
lower, the maintenance is easier, and
components are widely available.
Analog is cheaper to install, easier to
diagnose and repair, and functions better than a subroutine programmed into
a full-building digital system.
TAB: From what youve both said, it
appears that you agree there is an
optimal solution: single closed-loop
analog controllers monitored by a digital building system.
Were out of space for this issue.
Perhaps in future issues of TAB
Journal, you can return to discuss the
questions youve raised here. The
debate regarding analog and digital
controllers, as well as the classes of
control systems found in laboratory
ventilation applications, is of great
interest to our readers.
Rock: Certainly.
Anderson: I look forward to the
opportunity.

DC

TAB Journal

INDOOR AIR QUALITY

Humidity Control for Indoor Air Quality

William R. Beckwith
American Heat Pipes, Inc.

Introduction

igh humidity can have a detrimental effect on many aspects of

indoor environments. Building materials, books, and records deteriorate
rapidly in humidity levels above 60
percent due to biodegradation. High
humidity also increases electrical costs
by reducing the temperature required
for occupant comfort. Fungal growth
in humidities above 70 percent can
have many detrimental health effects
and is often an underlying cause of
many air quality disorders that result
in indoor air quality (IAQ) problems.
Moist environments (above 70 percent
rh) allow the proliferation of microbiological contamination that may cause
lethargy, allergic reactions, and disease
in people, and airborne pathogens may
survive freely in air that has a high relative humidity. Recently, the World
Health Organization identified microbiological contamination in buildings
as the fifth greatest threat to humans.
The effectiveness of air filters is also
reduced in high humidity, while the
offgassing of chemicals such as
formaldehyde and the perception of
odors all increase. Because suitable
growth substrates and adequate nutrients are generally available in properly
maintained modern buildings, the most
practical way to control microorganisms is to maintain reduced indoor
humidity. In fact, the ASHRAE IAQ
Standard 62-1989; Section 5.12 states
that the relative humidity in all ductwork should be kept below 70 percent.

TAB Journal

Control of indoor fungal growth can

prevent a wide variety of serious
health impacts for building occupants.

Problems with DX Systems

There is nothing inherently wrong with
DX systems. However, problems will
arise when the system is required to
meet performance criteria of which it is
not capable. Clearly, oversizing usually
results in high humidity levels. This is
because the system satisfies the thermostat quickly and does not operate
long enough to remove sufficient moisture. Adding a humidistat will do little
to improve comfort conditions. A continuous fan operation with ventilation
air further compounds the problem.
Proper sizing will usually provide sufficient humidity control, but only at peak
conditions (typically 95 degrees and
sunny outside with all the lights on).
The cooling capacity of DX equipment
is rated at 95 degrees outside air on
the condensing coil. When the outside
air is cooler, the systems capacity
increases while the cooling load
decreases. This results in an oversized
system and a need for reheat. An oversized system can also result from certain energy conservation measures
such as lighting retrofits and adding
thermal insulation, both of which
reduce cooling loads.
Prior to the mid-1970s, humidity was
controlled by using a reheat system. In
a reheat system, air is overcooled with

an air conditioner in order to condense

excess water vapor and is then reheated to a temperature suitable for introduction into the occupied space using
supplemental heat such as resistance
heaters or hot water. In effect, the
reheat adds an artificial, sensible heat
load to the space and achieves humidity control with higher capital and
operating costs.
This paper will introduce a technology
that offers one possible solution for
enhancing the dehumidification of DX
equipment while avoiding the high
cost of traditional reheat systems. The
DX enhancement (DXE) technology
increases the moisture removal of DX
equipment and provides virtually free
reheat using heat pipes.

Cost Factors
There are two operating cost factors
involved with reheat. First is the obvious electrical power consumption by
the electric heat strips, or fuel consumption for a hot water system. Less
obvious is the additional power consumption required by the air conditioner to remove the reheat heat load.
Besides increased annual operating
costs, capital costs are higher.
Overcooling the moist air and handling the supplemental heat load
requires an air conditioner two to five
times larger than a unit that does not
overcool and actively reheat.

17

Heat Pipe Operating

Principle
Heat pipes are efficient, passive
devices that transfer heat. In its simplest form, a heat pipe is a sealed tube
that has been evacuated, charged with
a precise amount of refrigerant, and
sealed. The heat pipe is then positioned so that one end is in a heat
source and the other end is in a heat
sink. The refrigerant on the warm side
absorbs heat, boils, and flows to the
cold side. Here it condenses, releases
heat and returns to the warm side by
gravity to complete the process. This
boiling and condensing continue as
long as a temperature differential
exists. No pumps, blowers, or power
inputs are required for the heat pipes
operation.

DXE Operating Principle

DXE (see Figure 1) uses the heat contained in the DX systems liquid line
(subcooling) and hot gas line (desuperheating) to provide controllable reheat
and increase moisture removal. A heat
pipe is used to transfer heat from the
warm liquid line of the refrigeration
system (subcooling) to the cold air
leaving the evaporator to provide free
reheat. The subcooling
of the liquid refrigerant
Filter
increases the cooling
capacity at the evaporator, resulting in an
increase in moisture
air
removal. In many DXE
systems, the liquid
refrigerant is subcooled
directly by passing it
through a coil in the
supply air. After subcooling, the liquid
refrigerant flows into
the evaporator. The subcooling/reheating step
can be continuous or
controlled. In addition,

References

a heat pipe is used to transfer heat

from the hot gas discharge line of the
compressor (desuperheating) to the
cold air leaving the evaporator to provide additional reheat in a controlled
manner to provide humidity control at
low cooling load conditions. DXE is
applicable to both recirculation and
100 percent outside air units.

1. Morey, P., et al., Biological

Contaminants in Indoor Environments,
ASTM STP 1071, Baltimore, MD, 1990.
2. Toth, C., Microbials in the Overall
Context of Indoor Air Quality
Investigation, Proceedings of First
Annual IAQ Conference and Exposition,
NCIAQ, Tampa, FL, 1992 pp. 255-259.

To demonstrate the performance of the

technology, we will look at a ten-ton,
EER 9.0, DX system operating at
3,750 CFM using ASHRAE design
conditions of 80 degrees dry bulb and
67 degrees wet bulb.
Without DXE, the system performs ten
tons of cooling at the evaporator and
removes 29.2 pounds of moisture per
hour. With DXE, the system performs
11.8 tons of cooling at the evaporator
and removes 41.2 pounds of moisture
per hour due to the subcooling
process. In addition, the subcooling
provides the equivalent of 6.3 KW of
free reheat. Finally, another 7.0 KW of
controllable reheat is available from
the desuperheat reheat heat pipe to
provide humidity control at low cooling load conditions.

3. Kay, J.G., et al., Biological Pollutants in the Indoor Environment, Indoor

Air Pollution. Vol. 78, pp. 13-24, 1991.
4. Samet, J.M., et al, Health Effects and
Sources of Indoor Air Pollution, Part II,
Am Rev Respiratory Diseases, 137:221242, 1988.
5. Arundel, A.V., et al., Indirect Health
Effects of Relative Humidity in Indoor
Environments, Environmental Health
Perspectives, Vol. 65, pp. 351-361, 1986.
6. Burge, H., et al., Sick Building
Syndrome: A Study of 4,373 Workers,
Ann Occupational Hygiene, 31:493, 1987.

SCADR
Coil

DX Coil
SubCool
Reheat
Circuit

Desuperheat
Reheat
CLosed Loop

Fan
air

control valve
suction
line

control valve

Heat
Exchanger

hot gas

Heat
Exchanger

Compressor

to compressor
liquid line

Condenser

Figure 1

18

TAB Journal

TAB Journal

19

SYSTEM COMMISSIONING

A Proposed Method for Validating Cooling

Coil Performance
Joseph Flood
Flood & Sterling, Inc.

Introduction

VAC system commissioning is

one of todays hot topics in our
industry. Many of us believe that a
system that has been balanced correctly will in fact have been commissioned. Stated alternately, the system
will perform in accordance with the
design intent or those deficiencies preventing design compliance will be
identified. There will also be an adequate record of such compliance within the balancing report.
One of the requisites for correct balancing is a means to validate cooling coil
performance. Obviously any deficiency
in airflow delivered to a controlled
space is a directly proportional deficiency in the ability of the system to match
the cooling load. The point is that airside deficiencies are unforgiving.
Alternately, water flow shortages can
be very forgiving; in heating with 200
degree hot water based upon a design
20 degree temperature drop, 50 percent of the design flow will result in
about 90 percent total heat output. To
a more limited degree, this forgiving
characteristic exists when using cooling water.
In a conventional chilled water application, if chilled water flow is short by
10 percent of design flow, the effect of
the shortage will be unidentifiable and
it will likely perform satisfactorily.

20

There is a minority in our industry that

maintains, If I deliver the design
intended flow to a piece of heat transfer equipment, I have successfully
executed my water balancing responsibilities.
This thought process is mistaken
because such water balancing responsibilities are not complete until it has
been demonstrated that the designintended heat exchange is taking
place. This is why the AABC cooling
coil report form requires performance
test data of the entering and leaving
water temperatures and the entering
and leaving conditions of the air (both
wet and dry bulb).
Cooling coil performance validation
presents some unique problems for the
test and balance engineer. For most of
the country, cooling problems involve
moisture removal (dehumidification)
as well as temperature depression
(sensible cooling). In addition, it is
virtually impossible to perform field
testing at design entering air conditions.
Thus, it would be useful if we had an
acceptable method to relate tests executed at field found conditions to the
design stipulated conditions similar
to the cooling tower performance procedure developed by George Young.

Hedges & McElgin Method

In the late 1940s, Berkleigh Hedges
and John McElgin jointly developed a
precise cooling coil selection procedure for ASHRAE and the Nesbitt
Company. This procedure took into
account the difference in motive force
between wet and dry surfaces.
This method notes that the driving
force for heat exchange for wet surfaces is the difference between the
total heat of the air and the total heat
at the apparent surface temperature.
Stated another way, the heat exchange
rate is based upon the difference in the
enthalpy of the air and the enthalpy at
saturation of the surface temperature.
This method culminates in an equation
for surface size selection as follows:
Aw = (Ht x s)/(fa x M.H.D. x F.A.)
These terms are defined as follows:
Aw is the airside area required per
square foot face area. It is another
way of defining the combined coil
depth in rows and the number of fins
per foot. It is also the solution to the
coil selection problem for all wetted
surface.
Ht is the total heat exchange from
the air to the water.

TAB Journal

Ht= 4.5 x CFM x (h1 h2)

h1= Total Heat of Entering Air - BTU/lb.
h2= Total Heat of Leaving Air - BTU/lb.

s is the specific heat of moist air =

0.245
fa is the sensible heat transfer coefficient between the air and the surface.
M.H.D. is the mean heat difference
between the air and the surface.
F.A. is the face or finned area of
the cooling coil.
This equation provides a rational basis
for evaluating the results of field executed capacity tests and comparing the
results to the design data. The total
heat transference can be calculated

It would be useful if
we had an acceptable
method to relate tests
executed at field
found conditions
to the design stipulated
conditions similar
to the cooling tower
performance procedure
developed by
George Young.
-Joseph Flood

separately for the design data and for

the test data. The specific heat of
moist air is the same for design and
test conditions.
The sensible heat transfer coefficient
(fa) is defined by the face velocity of
the air. Since the system has been
adjusted, the air velocity at field test
conditions approximates the air velocity at design conditions; therefore, fa is
constant. Later, I will provide correc-

TAB Journal

tions for minor deviations in both air

and water flows. Likewise, the face
area remains constant for both the
field test and design conditions.
It follows that the coil size (rows deep
and fin density) is also constant for
design and field test conditions. If we
eliminate the unchanged element, we
can write a simplified equation as
follows:
Htd/M.H.D.d = Htt/M.H.D.t
As a result, we can say that the ratio of
the design total heat exchange divided
by the mean heat difference should
equal the ratio of the test total heat
exchange divided by the test mean
heat difference (these values should be
readily available to the test and balance engineer). It also follows that if
you divide the derived test ratio by the
calculated design ratio, an acceptable
result would be one or greater.
If that result is greater than one (1), it
indicates that said coil is performing
better than the specification required.
Conversely, a result less than one (1)
indicates performance less than the
specified requirement. I would regard
all results greater than 0.95 as in compliance because some averaging has
been done to define surface temperatures over normally expected selection
ranges.

Problems and New

Solutions
There are two problems associated
with this method which must be
resolved. The first is the need to determine the surface temperature at the
inlet and the outlet sides of the cooling
surface, then to determine if the inlet
surface temperature is lower than the
dew point of the entering air. If the
inlet surface temperature is below the
dewpoint of the entering air, then con-

Have an
opinion?
An interesting
case study?
A new method?
Tell us about it.
TAB Journal welcomes
submissions for publication. TAB Journal is
published quarterly by
the Associated Air
Balance Council. Send
letters or articles to:
Editor
TAB Journal
1518 K Street, NW
Suite 503
Washington, DC
20005

21

densation is occurring and the entire

surface is wet. Performing these calculations is a complicated process, but
personal computers now make the task
a simple one.

conditioning ranges, the maximum

error is 2.5 percent. The leaving surface temperature uses the same equation except that it employs LAWB and
EWT.

The entering and leaving surface temperatures are dependent upon the coil
face velocity and the velocity of the
water in the tubes. As the air face
velocity increases, the coil surface
temperature approaches the air temperature; likewise, as the water tube
velocity increases, the surface temperature approaches the water temperature.

For example, when the entering wet

bulb temperature is 78 degrees (100
percent O.A.) and the leaving water
temperature is 55 degrees, the surface
temperature is 64 degrees and the calculated surface temperature is 63.86
degrees. This error is less than the normal error common to nomographs. This
95 percent performance is an acceptable tolerance for field evaluation.

Fortunately, the overwhelming majority of cooling coils are selected within

very narrow windows of performance.
Airside face velocities are universally
selected in the range from 400 FPM to
650 FPM. The vast majority are selected in the more narrow range of 475 to
550 FPM. The same is true for tube
velocities, which fall into the range of
3 to 6 FPS. Again, water velocities are
more likely in the range of 4 or 5 FPS.
These ranges are sufficiently narrow to
permit defining the surface temperature as the average temperature
between the air wet bulb temperature
and the water temperature minus .84
times the natural log of the temperature difference. Stated as an equation,
it looks like this:
Ent. Surf. Temp. = ((EAWB + LWT)
/2) (0.84 x logn(EAWB LWT))
where:
EAWB = Entering Air Wet Bulb
Temperature
LWT = Leaving Water Temperature
This equation has been developed
empirically from a nomograph developed by the authors. In normal air

22

Condensation & M.H.D.

The alternate problem associated with
this method is a means to define the
dew point temperature of the entering
air mixture to assure that condensation
is indeed occurring. Here again, an
empirical formula has been developed
based solely on a standard psychrometric chart which provides reliable results
in the ranges where the procedure
applies. This algorithm is as follows:
DPT = (EADB (1.347 x (EADB
EAWB))) x 0.97123
where:
EADB = Entering Air Dry Bulb
Temperature
AWB = Entering Air Wet Bulb
Temperature
The maximum error over the normal
range is less than five percent.
However, this formula is only used to
confirm a wetted coil condition and
does not enter into the validation equations. Since the smallest likely fraction
of a coil to be dry would be 0.1 (that
would be 1 dry row of a 10 row deep
coil), an error of 1/2 magnitude would
be negligible.

The only other consideration is the

M.H.D., the Mean Heat Difference,
which is the mean log difference
between the differential enthalpy at the
entering side of the coil surface and
the differential enthalpy at the leaving
side of the coil surface. The M.H.D. is
expressed by the following equation:
Dw = (Dl Ds)/(2.3 Log10(Dl/Ds))
where:
Dw = the logarithmic mean
Dl = the larger difference
Ds = the smaller difference
In summary, for wetted surface in the
normal air conditioning ranges (95DB,
80WB; 70DB, 61WB), the method
calculates a surface constant for both
the design and the test data. Theoretically, these constants should be equal
if the test result is to confirm the
design parameters.

Deviations & Corrections

If the water flow rates are within 20
percent of the design values, any deviation in performance is corrected by
changes that occur naturally to the
temperature of the water. These
changes will then be reflected in the
surface temperature and Mean Heat
Difference, and no correction factors
are necessary.
The same cannot be said for deviations
in airflow. The airside coefficient (fa)
is affected by the air face velocity
according to the following formula:
fa = 0.0807 x (FV ^ 0.7665)
where:
FV = Face Velocity and is proportional to CFM

TAB Journal

Therefore, we can write an equation to

calculate performance even with airflow deviations as follows:
P.F. = (CFMt x delta ht)/(MHDt x
((CFMt/CFMd)^0.7665))
divided by
(CFMd x delta hd)/(MHDd)
The effect of this correction for airflow may likely produce a result
where the test airflow is less than the
design airflow but the performance
factor equals unity or very close to it.
That means that the coil would produce the design results if the airflow
could be increased to match the design
requirement.
Earlier in the text, I had stated that airflow deficiencies are proportional to
cooling capacity shortage. This proposed procedure therefore recommends reducing the performance factor in direct proportion to any airflow
shortfall. The final equation should
look like this:
P.F. = (CFMt x delta ht)/(MHDt x
((CFMt/CFMd)^0.7665))
divided by
(CFMd x delta hd)/(MHDd)
multiplied by
(CFMt/CFMd)
This will produce a meaningful result
for the owner as he considers the
acceptability of the finished product.
Some of these calculations may seem
cumbersome, but in a computer algorithm such as Lotus 123, they are quite
simple.
Figure 1 is a printout of the finished
result using Lotus 123 for the calculations. I would be happy to provide the
calculations by cell address and a
look-up table for enthalpy values to
anyone who is interested.

COIL CALC. COOLING

DESIGN CONDITIONS

ENTHALPY
EWT
LWT
EADB
EAWB
LAWB
EADP
LAST
EAST

42.0
58.7
80.0
68.5
55.5
62.7
46.6
61.7

16.17
25.58

BTUH = 540396.9

AIR ENT
32.84 AIR LV

23.52 LG DIFF
SM DIFF
18.42 LOG DL/DS
27.55
LMTD =

3.83
3.94
3.94
3.83
0.012297

Surface
Constant
143784.5

3.758380

CFM
12885
Coil is all WETTED
FIELD TEST RESULTS

ENTHALPY
EWT
LWT
EADB
EAWB
LAWB
EADP
LAST
EAST

43.8
53.5
77.0
61.8
53.0
54.9
46.5
55.9

17.05
22.3

BTUH = 323912.9

AIR ENT
27.69 AIR LV
22.02 LG DIFF
SM DIFF
18.42 LOG DL/DS
23.71
LMTD =

Surface
Constant
123403.3

2.84
2.59
2.84
2.59
0.040018
2.624831

CFM
12695
Coil is all WETTED
Coil Performance Index = 0.86
Coil Performance Index = 0.848534
Corrected for CFM

DESIGN

TEST

42.0 EWT
58.7 LWT
80.0 EADB
68.5 EAWB
55.5 LAWB

43.8
53.5
77.0
61.8
53.0

Coil Performance Index = 84.9%

Note: This enthalpy has been copied and
is adequate for air work.
Figure 1

TAB Journal

23

S E R I E S FA N P O W E R E D B O X E S

Serie-ous Concerns with Series Fan Powered Boxes

Gary L. Miller
Engineered Air Balance Co., Inc.

eries Fan Powered Boxes (FPBs)

are being installed in many projects
today. This type of terminal unit offers
initial benefits to the owner and
designer, including constant volume
air distribution, a single primary duct
system, and a lower installation cost
than a dual duct system.
There are potential drawbacks, however, to the basic design and applications
of Series FPBs. Many box manufacturers and designers specify Series
FPB volume control to be either a
Solid State speed controller/Silicone
Control Rectifier (SCR) applied to the
fan motor, a three-speed switch, or a
combination of the two methods.
While SCRs offer a range of volume
control, many problems can occur.

Equipment & Application

Problems
One concern is that SCRs have the
ability to slow the fan down below the
manufacturers recommended RPM,
even in cases when the SCR has a
built-in, factory-established limit. It is
often possible to completely stop the
fan rotation by adjusting the SCR, or
to adjust the RPM below the manufacturer-required minimum. This minimum RPM must be maintained for
bearing lubrication and starting torque.
If the RPM is too low, the fan motor
will not have enough torque to restart
(which is also true of fan motors with
capacitors). If the fan is rotating too
slowly, the lubricating film on the

24

bearings will not be maintained, and

the life of the bearings will be shortened.

higher amps (and particularly above

nameplate amps) it stands to reason
that the additional heat produced by

Another problem often encountered is

overamping of the fan motor. One
puzzling phenomenon is the fact that
SCRs can cause the current load to
rise as the RPM declines. The increase
in amperage usually occurs between
the high and mid-range adjustment of
the SCR. Unlike variable speed drives
applied to central fan systems, SCRs
are not an energy-efficient method of
controlling fan volume. This is particularly true with three-speed motors.

How does the rating

of the motor change
simply by putting it
in a box? A motor
should never be
overamped, but...why
use a system that
consumes more
energy while producing
equal or less volume?

On some projects, the low or medium

speed nameplate amps can be exceeded but the box manufacturer allows it
because it was within the high speed
nameplate amps (even though the
motor had three separate windings).
Some manufacturers indicate only the
high speed amps on the outside of the
box instead of documenting all three
amp ratings listed on the motor.

the motor shortens its life. The use of

an SCR alters the sine wave of the
current through the motor and can
cause the motor to see an increase in
motor noise, an increase in motor temperature, and a decrease in motor reliability. Still, this type of control is
widely used.

How does the rating of the motor

change simply by putting it in a box?
A motor should never be overamped,
but even if the nameplate amps are
exceeded, why use a system that consumes more energy while producing
equal or less volume?
Another issue is equipment life, which
is even more important than bearing
life. Under the above conditions of

-Gary L. Miller

Some Series FPB applications have

other problems such as:

Floating RPM - Some box
fans RPM increase as the primary air
entering the box is reduced, actually
causing a more stable operation when
the pressure entering the fan becomes
negative. Box fan CFM can increase
or decrease depending on several
installation factors: amps decline; fan
static pressure rises (because the fan

TAB Journal

inlet becomes more negative); and the

RPM rises. When the box becomes
neutralized and primary equals secondary, the CFM usually declines; the
amps rise; and the RPM declines.
If the box must be balanced by reducing the RPM near the minimum 500
600 range, the fan can stall or literally
stop while operating. In some cases,
FPBs at too low of an RPM reverse
rotation while operating by over-pressurizing the box with too much primary air. This scenario is clearly the
exception but could happen following
problems with the control of the primary AHU.


Design engineers often specify

boxes that are oversized for the
required capacity and assumed to
operate quieter than a smaller box at
higher velocity and RPM. This causes
the three-speed switch to be on low
speed and the SCR to be greatly
reduced in order to achieve design
CFM. This slows down the fan further,
thereby using more amps and dropping
below the minimum RPM.
Manufacturers submit the box performance at a certain static pressure and
when the static pressure is different, so
is the performance. Yet there is no way
to predict what the static pressure will
be after the balancing dampers are
throttled in order to achieve design
CFM at the grilles. Most FPB fans are
chosen in the 0.30 to 0.40 range,
while a majority of the actual static
pressure conditions are in the 0.05 to
0.20 range. Again, this is due to
slowing the fan instead of throttling a
discharge damper.

When an SCR fails, the replacement box must be rebalanced. Unlike

motor replacement, the SCR has been
finely adjusted to achieve design

TAB Journal

CFM. Most maintenance personnel are

not trained or equipped to properly
balance a Series FPB.

Three-Speed Switch and

Discharge Damper
There is clearly a gap between what
the motor manufacturers (and in some
cases, the box manufacturers) are saying and what specifying engineers are
hearing. I am convinced that the best
volume control on Series FPBs is a
three-speed switch and a discharge
damper. Use of the three-speed switch
ball parks the fan volume, and the

I am convinced
that the best volume
control on Series FPBs
is a three-speed
switch and a
discharge damper.

sees the box discharge damper

because the box is either negatively
pressurized or neutral. With series
fans, the primary fan sees the system only up to the secondary fan.
What is past that point in the system
makes no difference.

Conclusion
At some point, the gap between recommended operating conditions from
manufacturers, design engineers, and
actual field operating conditions must
be closed. If Series FPBs are to be
used on projects, at least allow them to
operate at the most energy efficient,
low maintenance, and most reliable
method possible: using a three-speed
switch with a discharge damper.

-Gary L. Miller

discharge damper is used to achieve

design CFM while reducing the actual
amps. The discharge damper increases
the fan total pressure while reducing
the motor brake horsepower and fan
CFM, yet maintains the minimum
motor RPM. (It should be noted that
three-speed switches controlling 115volt fan motors offer approximately 10
percent volume reduction from high to
low speeds, compared to 277-volt fan
motors that offer approximately 25
percent reduction.)
The use of a discharge damper to save
energy is a method some people have
resisted. Conventional wisdom says
that the pressure loss across the
damper is a waste of energy, while
reducing the fan RPM with a SCR
saves energy. The truth is actually just
the opposite. The primary AHU never

25

INDOOR AIR QUALITY

A Model IAQ Case Study

Michael A. Price
MAP Environmental Services

ndoor air quality (IAQ) investigators are frequently called into buildings to conduct investigations as a
result of complaints from occupants
with symptoms such as eye and nose
irritation, fatigue, headaches, sore
throats, and general respiratory problems. The client generally assumes
that these symptoms are due to one
indoor airborne pollutant that is present within the building.
However, identical symptoms may be
found in individuals exposed to
formaldehyde, carbon monoxide, and
ozone. In addition, similar symptoms
are reported by individuals with allergies to specific fungi such as
Aspergillus and Cladosporium.
Symptoms can also be caused by
exposure to household dusts, cotton
fibers, and fibrous glass fragments.
An ever-increasing and similar problem is encountered due to low relative
humidities. Figure 1 depicts the overlapping symptomology of indoor airborne pollutants.
Since the cause of an IAQ-related
symptom can vary, it is generally not
productive to search for a single pollutant as the culprit of IAQ problems.
We advocate a solutions-based
approach to IAQ investigations that
examines all major sources within a
building, along with a thorough evaluation of the ventilation system to
ensure it is clean, balanced, and operating properly. We are convinced that

26

improving ventilation rates, upgrading

filters, and cleaning up the air distribution systems along with managing
strong pollutant sources will eliminate
the majority of indoor pollution problems. Such changes usually improve
worker productivity, enhance staff
morale, and reduce absenteeism.
The following case study illustrates
the importance of ensuring that a solutions-based approach is applied to the
initial IAQ investigation and the subsequent IAQ remediation.
In March, 1995, an initial IAQ investigation was performed at a small commercial office located in an office
industrial park in Northern Virginia.
The investigation began with an initial
survey of the office space to identify
obvious pollutant sources and building
configurations that could influence air
quality. Figure 2 provides a list of
common pollutant sources found during
IAQ investigations of office buildings.
Since one of the objectives of the
investigation was to assess maintenance standards, personnel responsible
for the maintenance of the building
were interviewed.
It is important to note that it is not our
practice to conduct detailed occupant
questionnaires when conducting an
IAQ investigation. Rather, we only
request enough data from occupants
or, preferably, their supervisors, to

provide clues as to the likely cause of

an IAQ problem.

Ventilation System
Inspection
Because IAQ is, in part, a function of
outdoor air quality, examination of the
ventilation system usually starts with a
check of the outside air intake. In this
case, the outside air intake was found
to be inadequate for the number of
occupants in the space. Airflow measurements and calculation of ventilation rates based on occupancy found
the outside air ventilation rate to be
less than 5 cubic feet per minute
(CFM), per person. The outside air
intake was also positioned directly
above the loading dock area.
A detailed inspection was made of the
buildings air distribution system,
including the internal components and
chambers of the air handling units, the
air supply ductwork, and return air
systems. Where necessary, access was
gained to the ductwork by the installation of a small access port for viewing
with a video inspection system. The
air handling units were found to be
contaminated with dust, debris, and
visible biological contamination on the
internal insulation and within the condensate drain pans. The ductwork was
contaminated with a general buildup
of dust and debris to a point were
staining was evident around the diffusers from contamination.

TAB Journal