time-of-day schedule is used to start and stop

the HVAC system. When a building is expected

to be unoccupied, the system is shut off and
the temperature allowed to drift away from the
occupied set point. The time at which the system
is to restart typically is set to
ensure that the indoor temperature
reaches the desired occupied set
point prior to occupancy on either
the coldest or warmest morning
of the year. As a result, for most days, the system
starts much earlier than needed. In turn, this
increases the number of operating hours and
system energy use.
An alternative approach is a strategy
called optimal start. This strategy
utilizes a building-automation system
(BAS) to determine the length of time
required to bring each zone from its
current temperature to the occupied
set-point temperature. The system
waits as long as possible before start-
R
ooftop variable-air-volume (VAV)
systems are used to provide comfort in
a wide range of building types and cli-
mates.
1
This kind of system consists of a packaged
rooftop air conditioner that serves several individ-
ually controlled zones. Each zone
has a VAV terminal unit that is
controlled by a temperature sensor.
This article will discuss HVAC-
system control strategies that can
be used to save energy in rooftop VAV systems.
OPTIMAL START/STOP
In some buildings, a simple time clock or
An applications engineer for Trane,
John Murphy, LEED AP, aids engineers
in the proper design and application
of HVAC systems. He is a member of
several American Society of Heating,
Refrigerating and Air-Conditioning
Engineers (ASHRAE) technical com-
mittees and a contributing author
of ASHRAE s recentl y publ i s hed
Advanced Energy Design Guide for
K-12 School Buildings.
By JOHN MURPHY, LEED AP
Trane
LaCrosse, Wis.
28 May 2008 HPAC Engineering
Control strategies can help turn
energy savings into operating-cost
savings and earn LEED credits
E n e r g y - S a v i n g S t r a t e g i e s f o r
Occupied
set-point
temperature
Unoccupied
set-point
temperature
System on System off Occupied hours
Optimal start Optimal stop
Zone temperature
Allows 2F
drift below
set point
Starts systems
as late as possible
Midnight 6 a.m. Noon 6 p.m. Midnight
Murphy FIGURE 1.
(Size: 27 picas.)
FIGURE 1. With an optimal-start strategy, a building-automation system waits
as long as possible before starting so each zones temperature can reach the
occupied set point just in time for occupancy.
VAV Systems
Rooftop
ing, so the temperature in each zone
can reach the occupied set point just in
time for occupancy (Figure 1).
This optimal starting time is deter-
mined using the difference between the
actual zone temperature and occupied
set point. It compares this difference
with the historical performance of the
zone warming up or cooling down.
The optimal-start strategy reduces
the number of system operating hours
and saves energy by avoiding the need
to maintain the indoor temperature at
the occupied set point even though the
building is unoccupied.
A related strategy is called optimal
stop. As mentioned previously, at the
end of an occupied period, the HVAC
system is shut off and the temperature
allowed to drift away from the occu-
pied set point. However, the building
occupants may not mind if the indoor
temperature drifts just a few degrees
before they leave for the day.
Optimal stop uses a BAS to deter-
mine how early heating and cooling
can be shut off for each zone so that
the indoor temperature drifts only a
few degrees from the occupied set point
(Figure 1). In this case, only cooling
and heating are shut off. The supply fan
continues to operate, and the outdoor-
air damper remains open to continue
ventilating the building.
The optimal-stop strategy also
reduces the number of system operating
hours, saving energy by allowing
indoor temperatures to drift early.
FAN-PRESSURE OPTIMIZATION
As cooling loads change, VAV termi-
nals modulate to vary airflow supplied
to the zones. This causes the pressure
inside the supply ductwork to change.
In many systems, a pressure sensor is
located approximately two-thirds of
the distance along the main supply
duct. The rooftop unit varies the
capacity of the supply fan to maintain
the static pressure in this location at a
constant set point. With this approach,
however, the system usually generates
more static pressure at part load than
necessary.
When communicating control-
lers are used on VAV terminals, it is
possi bl e to opti mi ze thi s stati c-
pressure control function to minimize
duct pressure and save fan energy.
Each VAV-unit controller knows the
current position of its air-modulation
damper. The BAS continually polls
these individual controllers, looking
for the VAV terminal with the damper
that is open the widest (Figure 2). The
supply fans set point then is reset to
provide just enough pressure so that at
least one damper is nearly wide open.
This results in the supply fan generat-
ing only enough static pressure to push
the required quantity of air through
this critical VAV-terminal unit.
This control strategy, which some-
times is called fan-pressure optimiza-
tion, has several benefits:
Reduced supply-fan energy use. At
part-load conditions, the supply fan is
able to operate at a lower static pressure
29 HPAC Engineering May 2008
Murphy FIGURE 2.
(Size: 27 or 41 picas.)
Supply fan
Static-pressure sensor
Communicating
building-automation
system
Variable-air-volume
terminals
FIGURE 2. The building-automation system looks for the VAV terminal with the damper
that is open the widest.
Murphy FIGURE 3.
(Size: 27 picas.)
S
t
a
t
i
c

p
r
e
s
s
u
r
e
Surge
Duct
static-pressure
control
Potential energy savings
Fan-pressure
optimization
1,000 rpm
Airow
800 rpm
900 rpm
FIGURE 3. Fan-pressure optimization. At part-load condidtions, the supply fan is able to
operate at a lower static pressure and consume less energy.
and consume less energy (Figure 3).
Lower sound levels. The supply
fan does not generate as much static
pressure and typically generates less
noise. In addition, with lower pressures
in the supply duct, the dampers in the
VAV terminals open wider, resulting in
less regenerated noise.
Reduced risk of fan surge. With the
fan operating at a lower pressure when
delivering reduced airflow, the fan
operating point is kept farther away
from the surge region (Figure 3).
Flexibility of sensor location.
Because this strategy uses the position
of VAV dampers to reset the pressure
set point, the static-pressure sensor
can be located anywhere in the supply
duct. It even can be located at the fans
discharge, allowing it to be installed
inside the rooftop unit and tested at
a factory. In this location, it also can
serve as a duct high-pressure sensor,
protecting ductwork from damage in
the event that a fire damper closes.
SUPPLY-AIR-TEMPERATURE RESET
In a VAV system, it is tempting to
raise the supply-air temperature at
part-load conditions to save compres-
sor and/or reheat energy. Increas-
ing supply-air temperature reduces
compressor energy because it allows
a compressor to operate at a warmer
suction temperature. The correspond-
ing higher suction pressure reduces
compressor lift, reducing the power
required.
An air-side economizer is more
beneficial when supply-air temperature
is reset. When outdoor air is cooler than
the supply-air-temperature set point,
compressors are shut off and outdoor-
and return-air dampers modulate to
deliver the desired supply-air tempera-
ture. A warmer supply-air-temperature
set point allows the compressors to be
shut off sooner and increases the num-
ber of hours in which the economizer
is able to provide all of the necessary
cooling.
When supply airflow has been re-
duced to a VAV terminals minimum
setting, raising supply-air temperature
decreases the use of reheat in those
zones that have very low cooling loads.
However, because the supply air is
warmer, zones that require cooling will
need more air to satisfy the cooling
load. This increases supply-fan energy.
Finally, in non-arid climates, warmer
supply air means less dehumidification
at coils and higher humidity levels in
zones. If dehumidification is a concern,
use caution when implementing this
strategy. Table 1 contains some general
guidance on determining when this
strategy can provide the most benefit.
2
Supply-air-temperature reset should
be implemented so that it minimizes
overall system energy use. This requires
consideration of the tradeoff among
compressor, reheat, and fan energy, as
well as the impact on space humidity
levels. These competing issues often
are balanced by first reducing supply
ai rfl ow, taki ng advantage of the
significant energy savings that come
from unloading the fan. Once fan
airflow has been reduced, the supply-air
temperature can be raised to minimize
reheat energy and enhance the benefit
of the air-side economizer.
Figure 4 is an example of a supply-
air-temperature-reset strategy that
is based on a changing outdoor dry-
bulb temperature. When the outdoor
temperature is warmer than 70F, no
reset takes place, and the supply-air-
temperature set point remains at the
design value of 55F. When it is that
warm outside, outdoor air provides
30 May 2008 HPAC Engineering
R O O F T O P V A V S Y S T E M S
Murphy TABLE 1.
(In Word document. Size: 27 picas.)
Conditions that favor using
supply-air-temperature reset
Mild climates that have many hours when
outdoor dry-bulb temperature is below 60F.
Systems with variable-air-volume terminals
that have minimum airow settings higher
than 30 percent of design airow.
Efcient air-distribution-system design.
Systems with interior zones that have varying
cooling loads.
Conditions that reduce the potential to save
energy with supply-air-temperature reset
Hot climates that have relatively few hours when
the outdoor dry-bulb temperature is below 60F.
Systems with efcient part-load fan modulation.
Inefcient air-distribution-system design.
Systems with interior zones that have varying
cooling loads.
Comments
In this type of climate, the economizer-related
benets typically outweigh the impact on
increased supply-fan energy.
Higher minimum airow settings increase the
likelihood that reheat will be needed to avoid
overcooling the zones.
Low pressure losses result in less of a
fan-energy penalty for the higher airows that
result from raising supply-air temperature.
The variable loads might allow these zones to
still be satised with a warmer supply-air
temperature.
Comments
In this type of climate, the economizer-related
benets are minimal.
Using a variable-frequency drive on the supply
fan and/or the fan-pressure-optimization control
strategy may make it more advantageous to
reduce supply airow to save fan energy, rather
than increase airow and save compressor
energy.
High pressure losses result in a signicant
fan-energy penalty for the higher airow that
accompanies a warmer supply-air temperature.
These zones will require near-design airow
during all occupied hours, limiting the ability to
raise supply-air temperature. Investigate using a
separate system to serve these zones.
TABLE 1. Guidance concerning supply-air-temperature reset.
little or no cooling benefit for econ-
omizing. In most zones, the cooling
load is likely high enough that reheat is
not required to prevent overcooling. In
addition, the colder (and drier) supply
air allows the HVAC system to pro-
vide sufficiently dry air to the zones,
improving part-load dehumidification.
When the outdoor temperature is
between 60F and 70F, the supply-
air-temperature set point is reset at a
2-1 ratio. That is, for every 2F change
in outdoor temperature, the set point
is reset 1F. In this range, supply-air-
temperature reset enhances the benefit
provided by the economizer and it is
likely that some zone-level reheat can
be avoided.
Finally, when the outdoor tem-
perature is colder than 60F, no fur-
ther reset occurs, and the supply-air-
temperature set point remains at 60F.
Limiting the amount of reset to 60F
allows the system to satisfy the cooling
loads in interior zones without needing
to substantially oversize VAV terminals
and ductwork.
Alternatively, some systems reset the
supply-air-temperature set point based
on the temperature in the critical
zone. (This is the zone at greatest risk
of overcooling, which would require
activating local reheat.) In these
systems, a BAS monitors the temper-
ature in all zones, finding the critical
zone that is closest to the heating set
32 May 2008 HPAC Engineering
R O O F T O P V A V S Y S T E M S
Murphy FIGURE 5.
(Size: 27 or 41 picas.)
Communicating
building-automation
system
Lounge
Restroom
Time-of-day schedule
Mechanical
room
Storage
Ofce
Vestibule
CO
2
Corridor
Occupancy sensor
Reception area
E
l
e
v
a
t
o
r
s
Ofce
Occupancy
sensor
Conference room
CO
2
Computer room
FIGURE 5. Zone-level demand-controlled-ventilation strategies.
Murphy FIGURE 4.
(Size: 27 picas.)
S
u
p
p
l
y
-
a
i
r
-
t
e
m
p
e
r
a
t
u
r
e

s
e
t

p
o
i
n
t
,

d
e
g
r
e
e
s

F
a
h
r
e
n
h
e
i
t
55
56
57
58
59
60
61
Outdoor dry-bulb temperature, degrees Fahrenheit
50 55 60 65 70 80 75
FIGURE 4. A supply-air-temperature-reset strategy based on a changing outdoor dry-bulb
temperature.
point. A rooftop unit then resets the
supply-air-temperature set point to
prevent the critical zone from needing
to activate reheat.
When considering using supply-air-
temperature reset in a rooftop VAV
system, first analyze the system to de-
termine if the savings in compressor
and reheat energy will outweigh the
increase in fan energy. If higher space
humidity levels are a concern, consider
disabling the reset function when it is
humid outside or provide a humidity
sensor in one or more zones to override
the reset function whenever indoor hu-
midity exceeds some maximum limit.
For interior zones with near-constant
cooling loads during occupied periods,
design airflows should be calculated
based on the warmer, reset supply-air
temperature (60F in the example from
Figure 4). While this may require larger
VAV terminals and ductwork, it allows
R O O F T O P V A V S Y S T E M S
33 HPAC Engineering May 2008
Unoccupied Humidity Control
A
VAV system typically dehumidifies effectively over a wide range of operating
conditions because it continues to deliver cold, dry air at part-load conditions.
As long as supply-air-temperature reset is used with caution and reheat is available
for those VAV terminals that have high minimum-airflow settings or experience low
cooling loads, a VAV system typically will provide supply air at a dew point that is low
enough to prevent elevated indoor humidity levels during occupied periods.
However, controlling humidity levels always is a priority, not just when a building is
occupied. When indoor humidity rises too high during unoccupied times, one option is
to turn on the rooftop unit and dehumidify the recirculated air to about 55F. However,
there typically is little sensible load in the zones during these periods, so delivering
cold air will result in overcooling. Reheat coils in the VAV terminals, and possibly a
boiler and hot-water pumps, will need to be activated.
An energy-saving alternative is to equip the rooftop unit with hot-gas reheat. When
after-hours dehumidification is needed, the rooftop unit turns on and diverts hot
refrigerant vapor from the compressor through a refrigerant-to-air heat exchanger
located in the air stream following the evaporator coil. Sensible heat is transferred
from the hot refrigerant to reheat the dehumidified air leaving the evaporator.
This strategy uses heat recovered from refrigeration circuits to reheat centrally
and saves energy by avoiding the use of new energy to reheat remotely at VAV
terminals.
the use of supply-air-temperature reset
during cooler weather, while still
providing the necessary cooling to
weather-independent interior zones.
It also is best to design a rooftop
VAV systems air-distribution system
for low-pressure losses and use the
fan-pressure-optimization strategy to
minimize the fan-energy penalty that
accompanies a warmer supply-air
temperature.
VENTILATION OPTIMIZATION
In a typical VAV system, the roof-
top unit delivers fresh outdoor air to
several individually controlled zones.
Demand-controlled ventilation (DCV)
involves resetting intake airflow in
response to variations in zone popu-
lation. One approach to optimizing
ventilation in a multiple-zone VAV
system is to combine the various DCV
strategies at the zone level (using each
where it fits best) with ventilation reset
at the system level. Ventilation reset
involves resetting intake airflow based
on variations in system ventilation
efficiency.
For this approach, carbon-dioxide
(CO
2
) sensors are installed only in
zones that are densely occupied and
experience widely varying patterns of
occupancy. For the example building in
Figure 5, CO
2
sensors are installed only
in the conference room and lounge.
These zones are the best candidates for
CO
2
sensors and provide the biggest
bang for the buck. These sensors reset
the ventilation requirement for their
respective zones based on measured
CO
2
.
While DCV strategies commonly
employ CO
2
sensors, occupancy sensors
or time-of-day (TOD) schedules also
can be used. Zones that are less densely
occupied or have a population that
varies only a little (such as private
offices, open-plan office spaces, or many
classrooms) probably are better suited
for occupancy sensors. In Figure 5, each
of the private offices has an occupancy
sensor to indicate when the occupant
is present. When unoccupied, the con-
troller lowers the ventilation require-
ment for the zone. Occupancy sensors
are relatively inexpensive, do not need
to be calibrated, and already are used in
many zones to control lights.
Finally, zones that are sparsely
occupied or have predictable oc-
cupancy patterns can be controlled
34 May 2008 HPAC Engineering
R O O F T O P V A V S Y S T E M S
There is
real potential
to save energy
in rooftop VAV systems
through optimized
system-control
strategies.
Murphy FIGURE 6.
(Size: 27 or 41 picas.)
Outside
air
Rooftop unit with controls
Reset outdoor airNow
Supply
air
Return
air
CO
2
Occupancy
sensor
Occupancy
sensor
CO
2
Communicating building-automation system
hew outsideair set oint
Direct-digital controls/
variable-air-volume terminals
Required ventilation
Actual rimary airNow
FIGURE 6. System-level ventilation-reset strategy.
efficiently through the use of a TOD
schedule. The schedule can indicate
when the zone normally will be occu-
pied/unoccupied or be used to vary the
zone ventilation requirement based on
anticipated population.
These various zone-level DCV
strategies can be used to reset the
ventilation requirement for their
respective zones for any given hour.
The zone-level control then is tied
together using ventilation reset at the
system level (Figure 6).
In addition to resetting a zones
ventilation requirement, the controller
on each VAV terminal continuously
monitors the primary airflow being
delivered to the zone. A BAS periodi-
cally gathers this data from all of the
VAV terminals and solves the venti-
lation-reset equations (prescribed
by ANSI/ASHRAE Standard 62.1,
Ventilation for Acceptable Indoor Air
Quality) to determine how much
outdoor air must be brought in at the
rooftop unit to satisfy all of the zones
served. Finally, the BAS sends this
outdoor-airflow set point to the roof-
top unit, which modulates a flow-
measuring outdoor-air damper to
maintain the new set point.
In a direct-digital-control/VAV
system, this strategy is fairly easy to
implement because the necessary real-
time information already is available
digitally. Combining DCV at the zone
level with ventilation reset at the system
level has many benefits, including:
Ensuring that zones are ventilated
properly without requiring a CO
2

sensor in each. CO
2
sensors are used
only in the zones in which they can
have the most benefit. This minimizes
installation costs and avoids the peri-
odic calibration and cleaning required
to ensure proper sensor operation. For
the other zones, occupancy sensors and
TOD schedules can be used to reduce
ventilation.
Enabling documentation of actual
ventilation-system performance. The
VAV controllers communicate each
zones ventilation airflow to the BAS,
even for the zones that do not have
CO
2
sensors. The BAS can be used to
generate reports showing ventilation
airflow in every zone, hour by hour.
Using system-level ventilation-
reset equations that are defined in
an industrywide standard. Using
equations from ASHRAE Standard 62
improves the ability to defend the
control strategys use.
CONCLUSION
The impact of any energy-saving
strategy on the operating costs of a
specific building depends on climate,
building usage, and utility costs.
Building-analysis tools can be used to
analyze these strategies and convert
energy savings to operating-cost savings
and help make financial decisions.
Figure 7 shows the potential energy
savings of using these various strate-
gies in an office building with a typical
rooftop VAV system. The optimized
system uses the optimal-start, supply-
air-temperature-reset, and ventilation-
optimization strategies discussed in
this article. In addition, the supply
fan is controlled based on fan-pressure
optimization, rather than a constant set
point in the ductwork. The optimized
rooftop VAV system would reduce
HVAC energy use by about 30 percent
if the building were located in Atlanta
and Los Angeles and by 33 percent if it
were located in Minneapolis.
There is real potential to save en-
ergy in rooftop VAV systems through
optimized system-control strategies.
This savings can reduce a building
owners operating costs and help earn
points toward Leadership in Energy
and Environmental Design certifica-
tion.
REFERENCE
1) Murphy, J. (2007). Rooftop VAV
systems applications engineering manual
(SYS-APM007-EN). La Crosse, WI:
Trane.
2) California Energy Commission.
(2003). Advanced variable air volume
system design guide. Sacramento, CA:
California Energy Commission.
For past HPAC Engineering feature
articles, visit www.hpac.com.
35 HPAC Engineering May 2008
R O O F T O P V A V S Y S T E M S
Murphy FIGURE 7.
(Size: 27 picas.)
H
V
A
C

e
n
e
r
g
y

c
o
n
s
u
m
p
t
i
o
n
,

p
e
r
c
e
n
t
a
g
e

o
f

b
a
s
e
0
20
40
60
80
100
Optimized
system
controls
Base
system
Minneapolis Los Angeles Atlanta
FIGURE 7. HVAC energy savings from optimized system controls.