Off-Peak Building Control

Considerations Utilizing CO2

Based Demand-Controlled
Ventilation (DCV) with Large
Packaged Rooftop Units

By: Bill Timmons

Mark Tozzi
Carrier Corporation
Syracuse, New York

September, 2000
Introduction

This white paper will highlight operational

challenges associated with part-load control of
packaged rooftop equipment and discuss
innovative solutions to maintain comfort and
save substantial amounts of energy while
simultaneously meeting the intent of ASHRAE’s
ventilation standard, Std. 62-99. In order to set
the tone for the premise of this topic, the
following observations are made.

While air-conditioning equipment is always

sized to maintain desired indoor room conditions
at peak design conditions, the majority of actual
operating hours occur at off-peak conditions. For
a packaged rooftop product, for instance, this
peak design condition usually occurs at the
summer design dry bulb temperature.

Building envelope loads, that is load components Figure 1

associated with the outdoor conditions (such as
transmission and solar), are only one component Without special controls you would have to
of the total cooling load. Internal loads such as assume that the design occupancy exists for all
lights, people and equipment are transient in zones at all times during the day. Why? Because
nature and fluctuate based on occupant usage With a control system that cannot re-set the
patterns rather than due to external weather outdoor air damper position in response to the
conditions. zone occupancy levels, you must assume the
maximum occupancy levels for each zone and
Following ASHRAE’s Standard 62-99 pre-program these occupancy schedules ahead of
Ventilation Rate Procedure, design ventilation time into the control system. This means that
rates are based on the total number of people in since you are guessing at the occupancy levels,
the space as well as their associated activity rather than actually measuring occupancy, you
levels. However, due to the transient nature of must err on the side of conservancy and
people in the building, that is people move from overestimate the number of people in any given
space to space during the day, the ventilation space. This results in over-ventilating and wastes
requirements for a given space will vary energy. In our conference room example, we
throughout the day as the number of people and would be supplying five times as much
their associated activity levels change. For ventilation as required if there were only two
instance, a conference room designed for 12 people in the room (10/2 = 5).
people may contain only 10 people at certain
times and be packed with 15 people on other A well-engineered HVAC system must be able
occasions. These same people may later move to to modulate its capacity to closely match the
other areas in the building, as indicated in Figure continually changing loads in the building.
1. Variable air volume (VAV) systems were
created such that as the building cooling loads
The total number of people in the building may change, the corresponding equipment capacity
not change, in this case, but the occupancy level may be modulated to closely match the actual
in each zone varies dramatically throughout the building cooling load. The end result was
day. How would you design a ventilation system improved comfort and substantial energy
to handle this scenario? savings. Given the fact that the cooling loads in a
building change throughout the day, or from
season to season, would we ever design a system
with a constant cooling capacity? Of course not!
With the exception of small systems with single

Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

3
steps of capacity, most larger commercial local CO2 conditions to determine occupancy
equipment has two or more cooling capacity levels as people enter and leave the zones
stages. Hypothetically, how well would a single throughout the day. This allows for a reduction
50-ton system, with one step of capacity, control in the amount of ventilation air that needs to be
a building? Not very well! The system would introduced into the building and yields
constantly cycle on and off attempting to significant energy savings over other ventilation
maintain setpoint. control strategies. A more detailed description of
DCV is included later in this discussion.
Continuing with the VAV system analogy,
would it be prudent to introduce design
quantities of ventilation air into a building that is Smoke & Mirrors
not fully occupied? Of course not! But that’s
exactly what many HVAC systems deliver…. One particular HVAC manufacturer offers an
design ventilation rates during all occupied elaborate and proprietary control algorithm and
times. Why? Because unless you know how complicated outdoor air damper device that
many people are in each zone of the building at claims to perform continuous, real-time
all times you must provide design ventilation calculations of the ASHRAE recommended
rates to cover the worst case scenario, that is 100 ventilation rates. This calculation is based on the
percent occupancy. Of course you must design ventilation requirements of the “critical zone”,
and select the equipment at peak occupancy and that is the zone with the highest required
peak design temperatures….but this doesn’t percentage of ventilation air. This manufacturer
mean that that you have to operate the system claims that their controls continuously and
that way all the time. automatically adjust the ventilation quantity to
ensure compliance with ASHRAE Std. 62. This
sounds good at first glance. Using ASHRAE’s
Design vs. Operating own Equation 6.1 from Standard 62 to calculate
Conditions the ventilation air quantities would seem to meet
the intent of Standard 62 wouldn’t it? Sure it
As mentioned previously, a dichotomy exists would, but what if the actual occupancy levels in
because the equipment must be sized at design the space are greater than or less than the
conditions but rarely operates at those assumed design occupancy levels? If the actual
conditions. Controlling the quantity of occupancy levels are less than design levels,
ventilation air introduced into a building based (which is the case over 90% of the time) then
on assumed maximum occupancy levels for each you waste energy by over-ventilating the space.
zone is not very accurate, nor is it the most cost- On the other hand, if the actual occupancy levels
effective method of controlling ventilation air. are greater than design levels (which may happen
Sizing the ventilation system for design occasionally) then you under-ventilate the space
occupancy without a feedback mechanism to tell and violate ASHRAE Std. 62 requirements.
you how you are doing is a risky design strategy. Sure, the control system that calculates Eq. 6-1
The good news is, achieving acceptable IAQ, may sometimes deliver the proper amount of
and at the same time saving lots of energy, no ventilation….but only when the actual space
longer requires a complex, sophisticated and occupancy levels exactly match the assumed
expensive HVAC system. Direct digital (DDC) design values.
controls are now available factory-installed on
relatively inexpensive, packaged rooftop unit
systems that provide excellent control of IAQ at CO2 Sensors - People Meters
the individual zone level, as well as save
substantial amounts of energy, especially when People exhale carbon dioxide (CO2 ) in quantities
compared to more complex, built-up type HVAC proportional to their activity level. The higher
systems. their activity levels, the higher the CO2
production. As a matter of fact, unless you’re
A control strategy known as Demand Controlled talking about a brewery, people are the
Ventilation, or DCV, has been developed, which predominant contributor of CO2 in a building.
meets ASHRAE Std. 62-99 requirements. This Another fact: humans exhale CO2 at
control strategy uses zone level measurement of concentrations of approximately 40,000 ppm.

Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

4
The average concentration of CO2 in outdoor air strategy offers the best of both worlds: precise
is approximately 350 ppm. control of ventilation rates plus the added benefit
of dramatic energy savings over constant
The latest revision of ASHRAE Standard 62- ventilation or ASHRAE Equation 6.1 based
1999, Appendix D, contains a detailed control systems. What is DCV and why is DCV
description of how to calculate differential CO2 better? Since people are generally the only
levels between indoor and outdoor air and how producers of CO2 in a building, it is very easy to
these values may be used to predict the level of measure the CO2 level in each zone of the
occupancy in the space. Using ASHRAE’s building as well as changes in these CO2 levels
Ventilation Rate Procedure, designers must over time. The concept of correlating the number
supply a certain ventilation air rate (CFM) per of people in the building with the indoor CO2
person based on a maximum occupancy level. levels is not new and has been demonstrated for
Seems simple enough doesn’t it? Count the several years now with a high degree of
maximum number of people in the building, then accuracy. Since ASHRAE states that we must
size the ventilation system to meet the worst-case deliver a prescribed quantity of ventilation air
occupancy levels. This is where the problems per person, and we know the number of people
begin. in the space (from the CO2 level), we can
continually adjust the outdoor air damper
Some manufacturers have added CO2 position to maintain the required quantity of
measurement in the return air duct as a "fail- ventilation air for each individual zone in the
safe" mechanism so that they may override building.
outdoor damper position if CO2 levels exceed a
certain setpoint. This is analogous to using a
return air temperature sensor and trying to Part-Load Profile
control individual zone temperatures. It cannot
be done! Any sensor mounted in the return air A representative 100-ton building cooling load
duct measures "average" values. In order to profile is shown in Figure 2.
control ventilation at the zone level, you must
measure CO2 at the zone level. A simplified method of representing a building
load profile is to construct a straight line between
the cooling and heating design points. Note that
DCV Advantages the maximum cooling load occurs at the summer
A Demand Controlled Ventilation (DCV) design temperature and that the minimum

140
120
100
Capacity (tons)

80
60
40
20
0
55 60 65 70 75 80 85 90 95
Bldg. Load
Ambient T emp.
Equip. Capacity

Figure 2 - Unit Capacity & Load vs. Ambient Temp.

Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

5
cooling load occurs where the load profile line does this mean? Cycling! Cycling of
meets the horizontal (temperature) axis. Loads compressors.
below this point are heating loads and are not
shown on this graph for simplification. When a compressor starts, there is a significant
in-rush of electrical current to the compressor
Another interesting phenomenon occurs. As the motor windings. This places thermal and
outdoor ambient temperature drops, equipment mechanical stresses on the compressor
capacity actually increases as the outdoor components, primarily the motor windings. The
(condensing) temperature drops, as represented end result is premature compressor wear. What
by the dashed line in Figure 2. This means that other effects does this excessive cycling cause?
as the building load is dropping, the equipment With only 4-steps of capacity control, you are
capacity is increasing. This is exactly the effectively banging on and off 25-tons of cooling
opposite of what we would like to see happen. capacity at a time. This results in a significant
Ideally, we would like the equipment capacity to change in the leaving coil conditions every time
drop in direct proportion to the building load. To a compressor or a bank of compressors turns on
achieve a reduction in equipment capacity, as the and off. The end result? Wide fluctuations in
building load drops, manufacturers offer multiple room temperature and humidity levels. Not
steps of capacity control. This is very critical as exactly what you want for a VAV system that is
outside air loads are highly variable due to controlling based on leaving air temperature
changes in ambient temperature and fluctuating (LAT).
quantities of ventilation air required.
The solution? Add more steps of capacity control
Figure 3 illustrates the part-load operation of a to more closely match the unit’s capacity to the
typical, large tonnage rooftop unit. building load. Figure 4 illustrates the same
condition with 9-steps of capacity control.

120
100%
100

Oversized (cycles)
Undersized (cycles)
Capacity (tons)

80
75%

60
50%
40
25%
20

0
55

59

63

67

71

75

79

83

87

91

95

A m b ie n t T e m p . 4 - S t e p C a p a c it y
Load

Figure 3 - Capacity Stages vs. Building Load for 4-step RTU

Notice that there are four discrete capacity steps Notice how much closer the unit with 9 steps of
as represented by the stair-step shape of the capacity matches the actual building load profile.
capacity line. Also notice that the unit capacity
exceeds the actual building cooling load a How can you get additional steps of capacity?
significant amount of time. Finally, notice that Either by adding more compressors or by using
the equipment capacity is below the actual compressors with capacity control. The most
cooling load a significant amount of time. What common capacity control device is called a
cylinder unloader. A cylinder unloader basically

Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

6
cuts-off the flow of refrigerant to some of the one theater may be showing a very popular
compressor's cylinders, thereby reducing the movie and be completely full, while the theater
capacity of the compressor. This is analogous to next door is showing a less popular movie and
making an 8-cylinder car engine operate more has a 50% or less occupancy level. Other likely
fuel efficiently by using only 4 cylinders, when candidates for DCV would be applications such
needed, instead of continuously braking and then as conference rooms, gymnasiums, school
pressing the accelerator to maintain desired cafeterias and any area that has widely
speed. The end result of having these additional fluctuating occupancy patterns.
capacity steps is much closer tolerances for the For the movie theater analysis, there were 10
LAT, room conditions and relative humidity and screens served by a total of 180-tons of variable
a significant reduction in wear to the air volume (VAV) packaged rooftop units. This
compressors. analysis considered the movie theaters only and

120

100%
100
90%
80%
80
70%
Capacity (tons)

60%
60
50%
40 40%
30%
20
20%

0
55

58

61

64

67

70

73

76

79

82

85

88

91

94
Ambient Temp. 9-Step Capacity
Load

Figure 4 - Capacity Stages vs. Building Load for 9-step RTU

Potential Energy Savings not the open common areas such as concessions,
hallways and front lobby areas.
The primary advantages to providing more steps
of capacity control is enhanced comfort, reduced For this analysis three different scenarios were
compressor wear due to less cycling and more analyzed. In the Baseline Case, each theater is
precise control of the space temperature and ventilated based on worst case (100 percent)
humidity levels, however there are significant maximum occupancy levels at all occupied times
energy saving opportunities also available when (constant ventilation strategy) and served by
combined with a Demand Controlled Ventilation VAV packaged rooftop units with only 4-steps
(DCV) strategy. of capacity control. Although this assumption is
an unlikely scenario, it was deliberately chosen
To quantify the costs & benefits of DCV, a 10- to establish a benchmark for maximum energy
screen multiplex cinema located in Dallas, Texas consumption.
was chosen for this analysis. This application is a
prime candidate for applying a DCV strategy as The second scenario, Alternate #1, was exactly
people loads fluctuate drastically throughout the like the first except a slight diversity was taken
day as well as from individual theater to theater. for the people, in anticipation that the theaters
The afternoon matinee may only have a 30-40% would not be filled to 100 percent maximum
occupancy level while the Friday evening occupancy levels except at night. In other words,
premier of the new blockbuster movie will have since we don't know, and can't measure, the
a packed house (100 percent occupancy). Also, number of people in the theater we must assume

Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

7
maximum predicted occupancy levels that would higher, worst-case (100 percent) design values
be expected during a typical day. This is the way rather than the values that would be measured by
a control system based on the ASHRAE Eq. 6-1 a CO2 sensor located in each theater. The
equation operates. difference in people occupancy levels is the
primary component of the annual energy savings
The third scenario, Alternate #2, simulated a achieved by using the DCV ventilation strategy
DCV control strategy utilizing CO2 sensors to over the other ventilation strategies.
measure anticipated changes in occupancy levels
throughout the day, in conjunction with VAV
packaged rooftop units with 9-steps of capacity
Energy Simulation
control.
An hourly computer simulation program was
The weekend people schedule is shown in Figure
used to calculate the design cooling and heating
5. The weekend occupancy schedules were
loads and also to simulate the annual operation
assumed to be slightly higher than those during
of all three scenarios and to calculate annual
the week, as would be expected. For
operating costs.
simplification, and since occupancy varies
dramatically at different times, all theaters were
It is important to understand that with a DCV
assumed to have identical occupancy schedules
control system we are still sizing our equipment
throughout the day.
and ventilation system for the worst-case, 100
percent design occupancy levels. However, by

120

100

80

60
%

40

20

0
11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour
% People (weekend, w/DCV) % People (weekend, max)
% People (weekend, 100%)

Figure 5 - Weekend occupancy schedules

Notice the difference in the people schedules actually measuring the number of people, and the
between the “max” values and the “DCV” changes in occupancy over the course of the day
values. The “DCV” values are lower because the (by measuring the changing CO2 levels), we are
outside air quantity will exactly match the able to throttle our ventilation air damper and
number of people in the theater. The “max” reduce the amount of outside air brought into the
values are higher because we must provide building. In addition, since all theaters are on a
higher quantities of ventilation air (essentially common duct system, and assuming there are
over-ventilate) since we don’t know the number times where some theaters are nearly empty (low
of people. In other words, with the “max” values CO2 levels) while others are nearly full (higher
we are supplying ventilation rates closer to the CO2 levels), the resulting mixed air CO2 level is
Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

8
reduced. By utilizing fresh air from adjoining (lights, people, equipment, etc.) and all system
spaces with lower CO2 levels, often we do not loads, such as ventilation air, at their highest
have to open the outside air damper to satisfy values. We will focus the remainder of our
one zone that requires additional ventilation. discussion on the ventilation air component of
This results in significant energy savings as the cooling load because this is where we have
indicated in the next section. the greatest potential for savings using the DCV
strategy.
The cooling load components for the theater on a
design day are indicated in Figure 6. This analysis assumes that the occupants are the
primary source of indoor air contaminants in the
building. These people-generated contaminants
are also referred to as bioeffluents. There are
Other Wall/Roof
2%
actually two components that make-up the total
12% required ventilation air quantity. The first
component is referred to as the occupant
Ventilation ventilation rate (e.g. 15 CFM/person), and is
35% listed in Table 2 of ASHRAE Std. 62-99. The
Lights
second component, sometimes referred to as the
23%
base ventilation rate, is a minimum
recommended value to dilute any building-
generated contaminants such as cleaning
products, fumes from construction materials or
building furnishings (carpet, wallpaper and
People furniture). In other words, if there were no
28% people in the building you would still need to
supply the base ventilation rate to dilute any
building-generated contaminants. The base
Figure 6 - Cooling Load Components ventilation rate is generally assumed to be
between 15-50 percent of design, depending on
As you can see from the chart, the design cooling the age of the building. A new building may
load is comprised of approximately 1/4 lights, have higher amounts of fumes due to
1/4 people, 1/3 ventilation, and the remainder is construction and an older building will have little
the roof and wall transmission and other “off-gassing” of fumes from building materials.
miscellaneous loads. A base ventilation rate of 20 percent of the
design ventilation rate is a reasonable
As indicated, the ventilation air is the single assumption for an existing building and was used
largest cooling load component and thus offers in this particular analysis.
the most opportunity for potential energy
savings. At design conditions you are seeing the
worst-case scenario with all internal loads

Annual Operating
Annual HVAC Total Cost
Case Name Description Energy Cost Savings over
Baseline Case
Constant Ventilation (100%)
Baseline Case with 4-steps of capacity $53,833 $0
ASHRAE Eq. 6-1 control
Alternate #1 system with 4-steps of $45,889 $7,944 (15%)
capacity
Demand Controlled
Alternate #2 Ventilation (DCV) with $40,101 $13,732 (25%)
9-steps of capacity

Table 1 - Annual Operating Cost Results

Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

9
Operating Cost Savings resulting in a reduced part-load energy efficiency
ratio (EER).
Annual operating costs for each scenario are
indicated in Table 1.
Analysis Assumptions
To summarize the economic results: Alternate #2
(ASHRAE Eq. 6-1 control ventilation strategy) External static pressure on the RTUs was 2.0 in.
did save approximately 15 percent over the w.g.
Baseline Case, however Alternate #3, the DCV
strategy, resulted in an additional 10 percent Ventilation air quantities were selected per
energy savings, for a total savings of 25 percent. ASHRAE Std. 62-99 for a theater at 7-1/2
CFM/person. This is one-half of the ASHRAE
To reiterate, this additional savings in annual Std. 62-99 Table 2 value (15 CFM/person) as we
operating costs was attributed to two things. are taking credit for intermittent occupancy and
First, the reduction in ventilation air quantity due the fact that theaters are continuously occupied
to the DCV strategy. Second, the higher amount for periods of less than three hours (per Std. 62-
of cycling of compressors for the system with 99, paragraph 6.1.3.4).
only 4-steps of capacity control resulted in a
reduction in efficiency at part load conditions. Internal lighting levels were assumed to be 3.0
This reduction in efficiency occurs due to the Watts/sq. ft. Miscellaneous electric was assumed
lower saturated suction temperature (SST) that to be 1.0 Watt/sq. ft. to account for equipment,
occurs when a system is oversized. As illustrated projectors, etc.
previously in Figure 6, with only 4-steps of
capacity the system operates in an oversized The electric power rate was assumed to be a flat
condition much of the time. The reduced SST rate of $0.10/kWh. The cost for natural gas was
requires the compressor to work harder since the assumed to be $0.70/therm.
compressor "lift" is greater during these times,

Summary & Conclusions

For this particular building, simulated in Dallas, Texas, the DCV system with nine steps of capacity control
yielded an annual operating cost savings of approximately 25 percent over the constant ventilation system.
In addition, the DCV system yielded 10 percent savings over the ASHRAE Eq. 6-1 based control system,
which had only four steps of capacity control.

Let’s reiterate the advantages of using DCV combined with additional steps of cooling capacity:

• DCV is an ASHRAE Std. 62-99 approved method

• Results in significant energy savings when compared to constant ventilation or ASHRAE Eq. 6-1 type
systems
• First cost savings; since controls are available with relatively inexpensive packaged rooftop unit
equipment
• Zone CO2 levels may be monitored and data-logged for trending as well as for documenting
compliance with ventilation codes
• Does not require expensive and complex flow measuring stations and dampers
• Reduces over or under ventilation of the zones during varying occupancy times
• Tighter control of LAT, (especially in VAV systems), when compared to systems with fewer steps of
capacity control
• Improved comfort levels in the space
• Less cycling of compressors, thereby improving the long-term reliability and life of the equipment
• By unloading compressors, rather than cycling them, the RTU operates at an optimum saturated
suction temperature (SST) and compression ratio. Without these additional capacity steps the RTU will
operate at lower SST and higher compression ratios, and as a result compressor efficiency decreases
dramatically at part-load.

Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

10
 References

1. Hourly Analysis Program, v. 3.22; Carrier Corporation; Syracuse, NY

2. ASHRAE Standard 62-99; Ventilation for Acceptable Indoor Air Quality; American Society of
Heating, Refrigerating and Air-Conditioning Engineers, Inc.; Atlanta, GA; 1999.
3. Schell, Mike B, Turner, Stephen C., P.E. and Shim, R. Omar. “CO2-Based Demand-Controlled
Ventilation Using ASHRAE Standard 62: Optimizing Energy Use and Ventilation”; ASHRAE
Transactions Symposia; TO-98-21-1; American Society of Heating, Refrigerating and Air-
Conditioning Engineers, Inc.; Atlanta, GA; 1998.

Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

11
 833-015 09/00

Off-Peak Building Control Considerations Utilizing CO2 Based Demand-Controlled Ventilation (DCV) with Large Packaged Rooftop Units

12