30 ASHRAE Jour nal ashr ae.

or g Mar ch 2004
By Steven T. Taylor, P.E., Fellow ASHRAE, and Jeff Stein, P.E., Member ASHRAE
S SS SS
About the Authors
Steven T. Taylor, P.E., is a principal, and Jeff
Stein, P.E., is a senior engineer at Taylor Engi-
neering in Alameda, Calif.
electing the inlet size of a variable air volume (VAV) terminal box
requires the consideration of five factors: (1) pressure drop across
the box; (2) ability of the VAV box controller to measure and control
the desired minimum and maximum airflow setpoints; (3) first costs of
the VAV box, its installation, and controls; (4) noise generation; and (5)
space constraints.
schematic for a parallel fan-powered VAV
(FPV) box. For both box types, V
max
is
typically the design cooling airflow rate.
For VRH boxes, V
min
is typically selected
to be the largest of the following:
1. The airflow required to meet the de-
sign heating load at a supply air tem-
perature that is not too warm, e.g., 90F
( 32C). Warmer temperatures tend to
result in poor temperature control due to
stratification and short circuiting.
2
2. The airflow required to prevent
dumping and poor distribution. This
limit depends on the diffuser style and
sizing. Thirty percent is a common rule-
of-thumb but some research has shown
that lower rates are satisfactory.
6
3. The minimum required for ventila-
tion. Depending on the code one is de-
signing to, determining this rate can be
The first three considerations affect
energy costs and first costs and ideally
should be balanced to minimize life-
cycle cost (LCC). The last two consider-
ations represent installation and
application constraints that can limit the
LCC optimum selection. This article
summarizes a detailed analysis of VAV
box control and selection
1
and provides
VAV boxes sizing criteria that will mini-
mize LCC in typical applications.
VAV Controls and Airflow Setpoints
Figure 1 shows a single duct VAV box
using pressure independent control
logic. Pressure independent controls use
two cascading control loops. The first
loop controls space temperature; its out-
put is an airflow setpoint limited to a
range between the minimum airflow
setpoint (V
min
) and the maximum airflow
setpoint (V
max
). This setpoint is then sent
to the second control loop, which modu-
lates the VAV damper to maintain the box
airflow rate at setpoint.
V
max
and V
min
are shown in Figure 2, a
control schematic for a typical VAV re-
heat (VRH) box, and Figure 3, a control
The following article was published in ASHRAE Journal, March 2004. Copyright 2004 American Society of Heating, Refrigerating and Air-
Conditioning Engineers, Inc. It is presented for educational purposes only. This article may not be copied and/or distributed electronically or
in paper form without permission of ASHRAE.
Mar ch 2004 ASHRAE Jour nal 31
simple (e.g., Californias Title 24
3
) or it can be
complex due to varying supply air rates and out-
door air percentage (e.g., ANSI/ASHRAE Stan-
dard 62, Ventilation for Acceptable Indoor Air
Quality
4
).
For the FPV box, only the last issue typically
applies because the parallel fan operation ensures
sufficiently high supply air rates in the heating
mode and at the diffuser. Californias Title 24
3
and the latest version of ASHRAE Standard 62
4
allow the designer to take credit for the dilution
offered by the transfer air supplied by the FPV
fan. This allows the minimum primary airflow to
the zone, V
min
, to be reduced to very low levels,
even to zero in some cases depending on the de-
tails of the design.
The above three factors determine the lowest
required value of V
min
. On the high side, V
min
is
limited by energy codes to minimize reheat en-
ergy losses. Both Californias Title 24
3
and
ASHRAE Standard 90.1
5
limit V
min
to the largest
of:
30% of V
max
;
The minimum required for ventilation;
0.4 cfm/ft
2
(2 L/s per m
2
) of conditioned floor
area of the zone; and
300 cfm (142 L/s).
V
min
can have an impact on VAV box sizing
because pressure independent controllers have a
limit to how low their setpoint can be. The lowest
non-zero setpoint is a function of the character-
istics of the flow probe located at the box inlet
(Figure 1) and the accuracy of the transducer/
controller (see sidebar). Different box sizes us-
ing the same flow probes and
transducer/controllers will have approximately
the same minimum controllable velocity. Since
the minimum velocity is the same, the larger the
VAV box, the higher the minimum airflow setpoint. For FPV
systems that can meet ventilation codes at very low primary
airflow rates, the minimum setpoint allowed by the controller
can be higher than that required for ventilation. Thus, a larger
box serving a given zone may need a higher minimum
setpoint, which will result in higher central fan energy, re-
heat energy, and possibly cooling energy (for systems with-
out economizers).
This limitation typically does not affect VRH systems with
digital controls because the minimum setpoint for proper heat-
ing diffuser performance (e.g., 30% of V
max
) is generally much
higher than the lowest controller setpoint (see table in sidebar)
unless the box is significantly oversized. It can be a limitation
for VRH systems with pneumatic controls, which have rela-
tively high minimum setpoints, or with digital controls when
a dual maximum control strategy is used (see Reference 1
for a complete description of this strategy).
Pressure Drop and Fan Energy
The pressure drop across the VAV box at the zone design
Total Pressure
VAV Box
Static Pressure
From Supply
Duct Main
Differential
Pressure
Transmitter
Reset Input
DPT
DM
C
Pressure
Independent
Controller
T
To Space
Space
Thermostat
Figure 1: Typical VAV box controls.
100%
0%
P
e
r
c
e
n
t

C
a
p
a
c
i
t
y
Hot Water
Valve
Cold
Supply Air
V
max
V
min
Heating Loop
Deadband
Cooling Loop
Parallel Fan
100%
0%
P
e
r
c
e
n
t

C
a
p
a
c
i
t
y
Hot Water
Valve
Cold
Supply Air
V
max
V
min
Heating Loop
Deadband
Cooling Loop
Figure 2: VAV reheat control diagram.
Figure 3: Parallel fan-powered VAV control diagram.
32 ASHRAE Jour nal ashr ae. or g Mar ch 2004
airflow rate will affect the design pressure required
by the supply air fan, provided the box is in the
critical path or index run, which is the airflow
path that has the highest overall pressure drop.
Assuming this to be the case (more on this later),
the greater the pressure drop across the box, the
greater the fan power. VAV boxes with smaller in-
lets will have higher pressure drops and, hence,
will result in higher fan energy.
VAV box pressure drops can be expressed in
terms of both static pressure drop and total pres-
sure drop, which are related by Equation 1:
VP SP TP + =

+ =
2 2
4005 4005
out in
v v
SP

+ =
2 2
2
4005
4005
4
WH
Q
D
Q
SP
(1)
where
TP is the total pressure drop
SP is the static pressure drop
VP is the velocity pressure drop
v
in
and v
out
are the inlet and outlet velocities
Q is the airflow rate
D is the box inlet diameter
W and H are the inside (clear) width and height of the
box outlet (outside dimensions less insulation thickness)
The static pressure drop across standard commercial VAV
boxes is always lower than the total pressure drop since the
velocity at the box inlet is much higher than the outlet ve-
locity, resulting in static pressure regain. But the total pres-
sure drop is the true indicator of the fan energy required to
deliver the design airflow through the box since the fan has
to generate both the pressure and velocity at the box inlet.
Therefore total pressure drop, not static pressure drop, should
be used to evaluate and select VAV boxes. Using static pres-
sure drop can be misleading since VAV boxes from different
manufacturers may have different outlet dimensions, and,
hence, different outlet velocity pressures. Unfortunately,
most VAV box manufacturers do not list total pressure drop
in catalogs. If not, it can be calculated using Equation 1.
Simulations
The previous discussion established the fundamentals of
how VAV box selection affects energy usage. To calculate the
magnitude of the impact, a prototypical Oakland, Calif. of-
fice building was simulated using the DOE-2.2 computer
program. Local utility rates were modeled with resulting
blended rates on the order of $0.13/kWh and $0.59/therm.
VAV systems with variable speed supply fans were modeled
using the two VAV box types shown in Figures 2 and 3. For FPV
boxes, V
min
was set to the larger of 0.15 cfm/ft
2
(0.8 L/s per m
2
)
and the lowest box setpoint (see sidebar), which varies by box
inlet size. V
min
was set to 30% for VRH boxes. First costs were
determined by averaging contractor costs for VAV boxes from
two popular manufacturers and adding a 25% contractor
markup. Installation costs were assumed to be the same for all
boxes. The cost differences for inlet reducers and discharge
plenums were ignored, mainly because there was no easy way
to account for them. Life-cycle costs were calculated over a
15-year life using a discount rate of 8% and 0% escalation
rates for both electricity and gas.
Results
The incremental life-cycle cost impact of VAV boxes sized for
0.3 in. w.g. to 0.8 in. w.g. (75 Pa to 200 Pa) TP at design airflow
rates are shown in Figures 4 and 5 for VRH and FPV systems,
respectively. The results indicate that for optimum life cycles
costs, VRH boxes should be sized for 0.5 in. w.g. to 0.6 in. w.g.
(125 Pa to 150 Pa) total pressure drop while FPV systems should
be sized for 0.6 in. w.g. to 0.7 in. w.g. (150 Pa to 175 Pa) total
pressure drop. However, LCC differences among the options and
systems are quite small, only a few cents per square foot per year.
This is because first costs favor smaller boxes while energy
costs favor bigger boxes, and both effects are small. Energy
savings are small because the design pressure drop only occurs
at peak conditions; pressure drop and fan energy drop quickly
(nearly with the square and cube of flow, respectively) at part-
load conditions. Life-cycle costs for FPV systems favor smaller
0.3 in. 0.4 in. 0.5 in. 0.6 in. 0.7 in. 0.8 in.
Design Box Total Pressure
FPV System
$80
$60
$40
$20
$0
Figure 5: Incremental life-cycle cost for FPV system as a function of VAV
box total pressure drop.
I
n
c
r
e
m
e
n
t
a
l

L
C
C
$
/
1
,
0
0
0

f
t
2
0.3 in. 0.4 in. 0.5 in. 0.6 in. 0.7 in. 0.8 in.
VRH System
Design Box Total Pressure
$60
$40
$20
$0
Figure 4: Incremental life-cycle cost for VRH system as a function of
VAV box total pressure drop.
I
n
c
r
e
m
e
n
t
a
l

L
C
C
$
/
1
,
0
0
0

f
t
2
Mar ch 2004 ASHRAE Jour nal 33
boxes than VRH systems because siz-
ing affects the minimum airflow
setpoint as well as pressure drop, so
reheat energy increases when boxes
are oversized.
To test the sensitivity of model-
ing assumptions, parametric runs
were made with varying sizing as-
sumptions (undersizing and
oversizing), controller minimum
setpoint, weather, occupancy
schedule, internal loads, supply air
temperature, utility rates, and win-
dow area. In all cases, the LCC opti-
mum pressure drops were very
similar to those shown in Figures 4
and 5.
Table 1 shows performance data
for a particular VAV box manufacturer based on total pres-
sure drops of 0.5 in. w.g. and 0.6 in. w.g. (125 Pa and 150 Pa),
including the pressure drop across a hot water reheat coil. At
these total pressure drops, noise is not likely to be an issue
for this particular line of VAV boxes for most applications, as
indicated by the NC levels from radiated noise predicted by
the selection software. For other manufacturers, higher de-
sign pressure drops, and for noise sensitive applications, this
may not be the case.
Sizing of Non-Critical Zones
As noted previously, box pressure drop only affects fan
energy if the box is in the critical path, the airflow path
requiring the most fan pressure. Arguably then, the sizing
criteria suggested by Figures 4 and 5 apply only to these
zones, while VAV boxes closer to the fan hydraulically, where
excess pressure is available, could be sized for a greater
pressure drop. However, we recommend that a consistent
pressure drop sizing criterion be used for all boxes regard-
less of location for following reasons.
First and foremost, it is much simpler. Designers would
not have to determine where boxes are located along the
duct mains before sizing them. Automated sizing programs
or spreadsheets could then be used much more easily.
Undersizing boxes to absorb excess pressure is limited
by noise constraints. As pressure drop increases, noise genera-
tion increases.
Energy codes
3,5
require that static pressure setpoints used
for fan capacity control be reset to satisfy the box requiring
the most static pressure, and as loads shift throughout the
day and year the most demanding box will change.
Figures 6, 7, and 8 are images of VAV box zone demand at
different times of day for an office building in Sacramento,
Calif. All three images are taken on the same day, Aug. 5,
2002. At 7 a.m., Zone 14 on the southeast corner of the build-
ing has the most demand. Later that morning at 9 a.m., Zone
36 in the interior of the building experiences the most de-
mand. At 5 p.m., the high demand has shifted to Zone 30 in
the northwest corner. Throughout the period monitored (the
better part of a year), the peak zone changed throughout the
floor plate, including both interior and perimeter zones. Hence,
the zone requiring the most static pressure could vary through-
out the day. If fan static pressure is reset to meet the require-
ments of only the zone requiring the most pressure, and if
boxes close to the fan are undersized to dissipate excess pres-
sure that is available at design conditions, then fan pressure
and fan energy would increase when these boxes become the
most demanding during off-design conditions.
Conclusions
Life-cycle costs over a wide range of economic and operat-
ing assumptions were found to be minimized when VRH boxes
(and other box types with high minimum volume setpoints)
were sized for a maximum total pressure drop of 0.5 in. w.g. to
0.6 in. w.g. (125 Pa to 150 Pa). Similarly, life-cycle costs for
FPV boxes (and other box types with low minimum volume
setpoints) were minimized when boxes were sized for a maxi-
mum of 0.6 in. w.g. to 0.7 in. w.g. (150 Pa to 175 Pa) total
pressure drop. It is important to note that these sizing criteria
apply to the total pressure drop, not the static pressure drop,
across the box including reheat coils.
Acknowledgments
The research for this article was funded by a Public Interest
Energy Research (PIER) grant from the California Energy Com-
mission. The authors would like to acknowledge the input and
work of other members of our research team, including Cathy
Higgins of the New Buildings Institute, Mark Hydeman of
Taylor Engineering, and Erik Kolderup and Tianzhen Hong of
Eley Associates.
Table 1: VAV reheat box maximum airflow rates.
Inlet
Size
(in.)
Outlet
Width
(in.)
Outlet
Height
(in.)
Total Pressure Drop = 0.5 in. w.g. Total Pressure Drop = 0.6 in. w.g.
Static
Pressure
Drop (in.
w.g.)*
Velocity
pressure
drop (in.
w.g.)
Max
cfm
Radiated
NC
Static
Pressure
Drop (in.
w.g.)*
Velocity
Pressure
Drop (in.
w.g.)
Max
cfm
Radiated
NC
4 12 8 0.08 0.42 230 24 * * * *
5 12 8 0.15 0.35 333 24 0.18 0.41 360 24
6 12 8 0.24 0.25 425 24 0.29 0.31 470 25
7 12 10 0.25 0.25 580 23 0.30 0.30 640 24
8 12 10 0.33 0.17 715 23 0.36 0.24 790 24
9 14 13 0.27 0.23 930 20 0.32 0.28 1,030 23
10 14 13 0.32 0.18 1,100 23 0.38 0.22 1,210 25
12 16 15 0.32 0.17 1,560 23 0.39 0.21 1,720 24
14 20 18 0.31 0.19 2,130 22 0.37 0.23 2,350 23
16 24 18 0.32 0.18 2,730 25 0.39 0.21 3,010 26
* For the 4 in. inlet size, 230 cfm is the maximum rate allowed by the manufacturer's selection * For the 4 in. inlet size, 230 cfm is the maximum rate allowed by the manufacturers selection
program.
References
1. Hydeman, et al. 2003. Advanced Variable Air Volume Sys-
tem Design Guide, California Energy Commission, Sacramento,
Calif., September. This guide was developed as a part of the
New Buildings Institutes Public Interest Energy Research
(PIER) Program for the California Energy Commission. Free
copies of the guide are available at www.newbuildings.org/
pier.
2. 2001 ASHRAE HandbookFundamentals, Chapter 32.
3. California Energy Commission. 2001. Title 24 Building
Energy Efficiency Standards, California Code of Regulations,
Part 6.
4. Addendum 62n to ANSI/ASHRAE Standard 62-2001, Ven-
tilation for Acceptable Indoor Air Quality.
5. ANSI/ASHRAE/IESNA Standard 90.1-2001, Energy Stan-
dard for Buildings Except Low-Rise Residential Buildings.
6. Bauman F.S., et al. 1995. Thermal Comfort with a Variable
Air Volume (VAV) System. Center for Environmental Design
Research, University of California, Berkeley.
Figure 6: Site 3 VAV box demand (7 a.m., Aug. 5, 2002).
Figure 7: Site 3 VAV box demand (9 a.m., Aug. 5, 2002).
10 11 12 13 14
35 36 15
33 34 38 37B 39 16 17
9
3
2
1
31 26 25 24 22 21 20
32
8
29 27
30, 74% 28 23
19
4
7 6 5
North
18
37A
10 11 12 13 14
35 36, 71% 15
33 34 38 37B 39 16 17
9
3
2
1
31 26 25 24 22 21 20
32
8
29 27
30 28 23
19
4
7 6 5
North
18
37A
10 11 12 13 14, 85%
35 36 15
33 34 38 37B 39 16 17
9
3
2
1
31 26 25 24 22 21 20
32
8
29 27
30 28 23
19
4
7 6 5
North
18
37A
Figure 8: Site 3 VAV box demand (5 p.m., Aug. 5, 2002).
Determining
Minimum
Airflow Setpoint
0%
50%
90%
Damper
% Open
0%
50%
90%
Damper
% Open
0%
50%
90%
Damper
% Open
One limitation on the minimum airflow setpoint for
the VAV box is the controllability of the box. VAV box
manufacturers typically list a minimum recommended
airflow setpoint for each box size and for each stan-
dard control option (e.g., pneumatic, analog electronic,
and digital). However, the actual controllable mini-
mum setpoint is usually much lower than the box
manufacturers scheduled minimum when modern digi-
tal controls are used.
The controllable minimum is a function of the de-
sign of the flow probe and the accuracy and precision
of the digital conversion of the flow signal at the con-
troller.
The flow probe is installed in the VAV box inlet or
outlet and provides an air pressure signal that is pro-
portional to the velocity pressure of the airflow through
the box. Flow probes, which are typically manufac-
tured and factory installed in the VAV box by the box
manufacturer, are designed to provide accurate signals
even when inlet conditions are not ideal (e.g., an el-
bow close to the inlet) and to amplify the velocity
pressure signal to improve low airflow measurement.
The amplification factor varies significantly by VAV
box manufacturer and, to a lesser extent, by box size.
The greater the amplification factor, the lower the con-
trollable minimum. The VAV box manufacturer must
balance this benefit with other design goals such as
minimizing cost, pressure drop, and noise.
The box controller must convert the velocity pres-
sure signal from the probe to a control signal. To make
this conversion, digital controls include a transducer
to convert the velocity pressure signal to an analog
electronic signal and an analog-to-digital (A/D) con-
verter to convert the analog signal to bits, the digital
information the controller can understand. For stable
control around a setpoint, the controller must be able
to sense changes to the velocity pressure that are not
too abrupt. One controller manufacturer recommends
a minimum setpoint equating to at least 14 bits for
stable control.
The steps to calculate the controllable minimum air-
flow rate (V
m
) for a particular combination of VAV box
and VAV box controller are as follows:
1. Determine the velocity pressure sensor setpoint,
34 ASHRAE Jour nal
VP
m
in inches of water that equates to 14 bits. This will vary
by manufacturer. For several manufacturers who were con-
tacted, VP
m
can be as low as 0.004 in. w.g. (1 Pa). This will
require a 10-bit (or higher) A/D converter and a 0 to 1 in. w.g.
(0 to 250 Pa) or 0 to 1.5 in. w.g. (0 to 375 Pa) range transducer.
Use of an 8-bit A/D converter or a transducer with a wider
range can result in a stable control setpoint 0.01 in. w.g. (2.5
Pa) or higher.
2. Calculate the velocity pressure sensor amplification fac-
tor, F, from the manufacturers measured Q at 1 in. w.g. (250
Pa) signal from the VP sensor as follows:
2
1in.
4005

=
Q
A
F
where A is the nominal duct area (ft
2
), equal to:
2
24

=
D
A
where D is the nominal duct diameter (in.).
Figure 9 shows an example of a VAV box flow probe perfor-
mance. The data on the right size of the graph are the airflow
rates at 1 in. w.g. for various neck sizes (shown on the left).
For example using this figure, this manufacturers probe for
an 8 in. (0.2 m) inlet box senses a 1 in. w.g. (250 Pa) signal at
702 cfm (331 L/s).
3. Calculate the minimum velocity v
m
for each VAV box
size as:
F
VP
v
m
m
4005 =
where VP
m
is the magnified velocity pressure setpoint from
Step 1.
4. Calculate the minimum airflow setpoint allowed by the
controls (V
m
) for each VAV box size as:
Figure 9: Sample VAV box inlet sensor performance chart,
cfm vs. velocity pressure signal.
Inlet Size
22 in.
16 in.
14 in.
12 in.
10 in.
9 in.
8 in.
7 in.
6 in.
5 in.
4 in.
10,000
1,000
100
10
c
f
m
0.01 0.1 1
Flow Probe Velocity Pressure Signal (in. w.g.)
7,000
3,665
2,806
2,062
1,432
1,160
916
702
515
358
229
c
f
m

a
t

1

i
n
.

S
i
g
n
a
l
A v V
m m
=
Table 2 shows the minimum airflow setpoint for the VAV
box probe depicted in Figure 9 with a controller capable of a
velocity pressure setpoint of 0.004 in. w.g. (1 Pa) and 0.01 in.
w.g. (2.5 Pa).
Nominal Inlet
Diameter, in.
Area, ft
2
cfm at Q
1 in.
Sensor Reading
Amplification
factor
Min. VP Sensor Reading =
0.004 in. w.g.
Min. VP Sensor Reading =
0.01 in. w.g.
Minimum
Velocity, fpm
Minimum Flow,
cfm
Minimum
Velocity, fpm
Minimum Flow,
cfm
D A F v
m
V
m
v
m
V
m
4 0.087 229 2.33 166 14 263 23
5 0.136 358 2.33 166 23 263 36
6 0.196 515 2.33 166 33 263 52
7 0.267 702 2.33 166 44 263 70
8 0.349 916 2.33 166 58 263 92
9 0.442 1,160 2.33 166 73 263 116
10 0.545 1,432 2.33 166 91 263 143
12 0.785 2,062 2.33 166 130 263 206
14 1.069 2,806 2.33 166 177 263 281
16 1.396 3,665 2.33 166 232 263 367
22 2.64 7000 2.28 168 443 265 700
Table 2: Sample calculation of box minimum flow.