©2011 ASHRAE. THIS PREPRINT MAY NOT BE DISTRIBUTED IN PAPER OR DIGITAL FORM IN WHOLE OR IN PART.

IT IS FOR
DISCUSSION PURPOSES ONLY AT THE 2011 WINTER CONFERENCE. The archival version of this paper will be published
in ASHRAE Transactions, Volume 117, Part 1.

LV-11-C061

Application Issues for Chilled Beam

Technologies

Boggarm S. Setty, P.E., ASHRAE Fellow

ABSTRACT
Recently, many discussions and papers relating to chilled beams have been published. Chilled beams usage has become
rampant along with Dedicated Outside Air Systems (DOAS). The paper discusses the design issues relating to filtration,
induction effect, certification and energy consumption. In most cases, only the energy savings relating to reduced air qualities
may not be real. The paper discusses the application issues for chilled beam technologies, its usages and cost effectiveness.

GENERAL
Chilled beam technologies rely on direct heat transfer within a space by circulating liquids (usually chilled or hot water)
through a heat transfer surface or coil that in-turn cools space through radiation and convection. In commercial applications,
in order to maximize cooling effects and to control moisture, chilled beam solutions can feature a separate ventilation
(Dedicated Outside Air, DOA) system, to accommodate not only code required air exchange rates, but to provide a means to
pressurize and dehumidify space conditions: Such configurations are often referred to as an Active Chilled Beam (ACB)
system. There are issues to consider in applying these systems.

BACKGROUND
Chilled beam technologies are not that new. Today’s ACB systems are adaptations of the floor and ceiling mounted
induction units used in the mid-1900s. Like the older induction units, coils are included to further temper induced air. Rather
than being only a heating coil, the ACB system offers either 2-pipe or 4-pipe cooling and heating capabilities. The older
induction technology became less favored in the late 1960’s - 1970 due to energy efficiency, maintenance issues, and high
initial cost. Early induction air units were often associated with high-pressure air supplies, which resulted in significant fan
energy use. It was also the case that surface temperatures within induction units could allow condensation on exposed
surfaces, and such also required regular maintenance to keep primary air nozzles clean. Induction systems were also
relatively expensive due to required high-pressure ductwork construction and costly components. By the mid-1970’s, the use
of such systems virtually stopped with the introduction of lower cost and more efficient Variable Air Volume (VAV) system
technologies.

NEW TECHNOLOGIES
The recent resurgence of interest of ACB technologies is based upon assessments that such systems are more energy
efficient: It is argued that because the majority of space conditioning energy is moved in water, not air, there are inherent
energy savings. Care must be taken in relating to performance expectations, as many assessments do not take into
consideration water flow (pump) energy use and its associated cost. Also, the energy needed to induce air movement in the
space is still felt back at the primary supply’s air handling unit – in this case, a DOA ventilation unit.

Current chilled beam technologies utilize linear horizontal air grilles, located at the ceiling plane, with air supply

1
delivered across the ceiling. Primary air supply “outlets” within the chilled beam do require less pressure than their induction
unit predecessors. Induction Nozzle Pressure (INP) is often referenced as low as 0.5 – 0.75 inches WG, but may require
higher values. This combination of ventilation air delivery and water energy movement provides the cooling and/or heating
energy required for space conditioning.

Such technology applications are perhaps considered in comfort space-conditioning applications that have minimal
latent loads and continuous operation. Deep row coils, which are normally associated with high latent load control, are not
supported by current ACB unit technologies, requiring the consideration of supplementary dehumidifying equipment when
addressing high latent loads.

Existing (older) buildings have additional challenges associated with heavy infiltration loads for spaces with envelope
or entrance elements. In these and other high latent load situations, chilled beams could have issues with temperature and
humidity control, and resulting condensation.

This paper specifically compares two air conditioning system applications: a conventional VAV system with Dedicated
Outdoor Air (DOA) and perimeter fan coil units, against an Active Chilled Beam (ACB) system. To be sure, there are
numerous variations in system configurations that can be associated with such systems. Also, there can be many space load
conditions besides the ones assumed; so this paper’s analysis is not intended to offer a definitive comparison of the two
technologies, but to show possible performance issues. To be somewhat definitive, our analysis did consider three different
city locations to address climate impacts, and different INPs were also considered.

ACTIVE CHILLED BEAMS

Chilled beams can be characterized as in-room terminal units. Because passive chilled beams do not provide for
ventilation, they can’t directly be compared an alternative to a VAV system. As an ACB system provides for ventilation, it
can be directly compared in function and performance as an alternative to a VAV system. Figure 1 represents a 100 percent
DOA system that provides primary air to ACB terminals. In this particular case, the DOA - ACB unit generates 4.5” WG
static pressure to provide an induction ratio of 3-4 at the terminal.

Figure 1 Integrated DOA - Active Chilled Beam System

2
 CEILING

HEATING/COOLING COIL

INDUCTION NOZZLE

Figure 2 Active Chilled Beam Terminal

As shown in Figure 2, an active chilled beam consists of a cooling/heating (i.e., two-pipe or four pipe) finned coil and a
primary air supply (Ventilation air) duct through the length of the beam with nozzles that induce room air across the coil and
diffuses it back into the room. Some ACB terminals may also include illumination functions. Note there are no filters to
clean secondary air.

CHILLED BEAM DESIGN ISSUES

Leakage
Air and moisture leakage through a building’s envelope can be high in both new and existing buildings. In older
buildings, the infiltration rate can be particularly high, likely to exceed 1.0 – 1.5 cfm/sf of wall area. In such cases,
significant rates of primary airflow through the ACB terminals would be needed to pressurize the building and to impede
water vapor transfer (i.e., water vapor transfer through infiltration). Chilled beams must address infiltration loads when
considering both heating-cooling operations, especially latent loads.

High Occupancy
Areas of potentially high occupant density must also be carefully addressed. The magnitude of the changes in sensible
and latent loads due to varying occupancy loads and increased ventilation requirements can be significant for spaces; such as
conference rooms, courtrooms, auditoriums and training rooms. In particular, during periods of high occupancy, the latent
cooling demand can dramatically increase. Subsequently, a larger primary airflow rate, at a lower dew point, could be
needed for both ventilation and moisture control. If the primary air flow is limited to a normal ventilation rate, additional
latent cooling will be required for unusual conditions, often requiring auxiliary air-conditioning equipment. Without such
latent load control, surface temperatures within the chilled beam may go below dew-point conditions, causing moisture
condensation, and potential dripping into the occupied space.

Space Air Distribution

Zone or space air distribution is critical to proper ventilation and occupant comfort. Chilled beam air delivery must be
designed such that necessary minimum static pressure is available to maintain sufficient air-circulation (e.g. having an
induction ratio of at least 3 to 4). In many cases, chilled beam manufacturers require 1/2” and 3/4” WG which are required to
create the needed induction effect in order to create proper air distribution.

Brake-horsepower
The difference in total brake-horsepower (BHP) between system comparisons must consider supply and return air fans
and heating hot water and chilled water pumps. Power related calculations must also address the rated airflow static pressure
drops and water pressure demands of terminal units and their impact to BHP needs. The considered VAV and the ACB

3
systems have different configurations and size requirements for these system components, and it is here that oversight may
occur by not considering all involved components or their required operating schedules. In addition, for both systems both
DOA fans and involved system water pumps must be scheduled to run year round if continuous pressurization and humidity
control is to be achieved. In some assessments, fans and pumps for chilled and hot water circulation are assumed to be “off”
during unoccupied periods: While such an operation can cause discomfort for the start-up of a VAV system, it can cause
moisture damage for an ACB system should high humidity conditions exist.

Response Time
The time delays and expected frequency of load changes in each zone and exposure also need to be considered, along
with the energy waste due to switch-over. For an ACB system, the heat loss or gains involved are often not considered, nor is
the possible moisture damage impact of restart-up.

Churn Rate
The occupancy churn rate (i.e., number of times per year that workstations or spatial requirements are changed) can also
cause start-up moisture problems for an ACB system. Time impacts can be significant when considering burden of
relocating both water piping and primary air ductwork. Modifying an ACB system can result in additional time delays for
drain-down, refilling and purging, and water leak corrections. Additionally, flexibility may be encumbered by the designed
beam terminal sizes, refitting drain pans, and relocating controls.

Reliability and Control Systems

Reliability and longevity of additional control systems can also be a problem for chilled beam systems. As with other
hydronic terminal units, e.g. Fan Coil Units (FCUs), chilled beam terminals have “self-contained microprocessor
controllers.” Possible condensation and accessibility issues may increase maintenance costs.

Water Damage
Disruptions to occupants can be a major issue if water leakage occurs. As hot and chilled water supply and return lines
will typically be circulated through several chilled beams in a zone or area, the disruption to one terminal involves a large
area. In this context, consider that the ACB distribution of piping spans 70 to 80 percent of the ceiling area, as opposed to a
VAV systems more localized VAV box locations.

Building Pressurization
As addressed earlier, building pressurization and dehumidification should be considered for both systems. While a
comfort and (worse case) a mold control issue could present itself for a VAV application, ACB systems can cause liquid
moisture damage if allowed to operate below dew-point conditions. As such any start-up regiment for an ACB system must
involve an early drying-out of building air by the early and separate operation of the DOA system, with chilled water going to
ACB terminals only after space dew-point temperatures have been lowered below ACB terminal temperatures.

Filtration
VAV systems provide the necessary 10 MERV filtration through the building’s DOA and VAV air handling unit filters.
However, for an ACB system, only the primary air is filtered: Induced secondary air at terminals is not filtered. As such, an
ACB system may be difficult to achieve the necessary 10 MERV filtration performance without auxiliary space filtration
units.

4
 An additional maintenance issue for an unfiltered ACB terminal is that the chilled beam/cooling coils will pick up more
dust/lint and mold particles from the occupied space than comparable terminal units: Such could cause additional cleaning
efforts to protect occupant health.

Certification
VAV systems and their components can receive certification by a third party, such as the Air Conditioning, Heating,
and Refrigeration Institute (ARHI), relating to its performance. However, for ACB systems, certification is typically
available only through the manufacturer.

Chilled Water Temperature

Chilled water temperature can be adjusted to save energy by allowing it to rise during part load. Also full cooling
chillers can be used without any mechanical cooling to save energy.

Further Operation and Maintenance Issues

As discussed earlier, to avoid condensation on chilled beam surfaces, the air dew point temperature in each space must
be maintained below the coldest surface temperature of the chilled beams. Operations and maintenance staff must accurately
and constantly monitor the indoor air dew points and chilled beam surface temperatures in each separate space. On a rise in
room dew point temperature, swift action by operational staff is needed to prevent condensation by quickly reducing the
room dew-point temperature or warming the supply water temperature to the chilled beams.
As such, design of a chilled beam system must include the ability of the operations staff to be informed, under all
operating conditions, that building pressurization is maintained.

Related to operations and maintenance, the design of a chilled beam system must specifically provide for ease of
regular inspection and maintenance of valves, piping, insulation, terminal unit connections, and terminal heat transfer surface
areas of the chilled beams. An ACB design must also include concise maintenance procedures, and inspection schedules.

Design of an ACB system must be then reconciled with the anticipated expertise of the operations and maintenance
staff available to the building. The design decision-making process must work with owners to insure required operations and
maintenance training.

COMPARISON OF VAV AND ACTIVE CHILLED BEAM DESIGN FEATURES

The considered ACB system was described earlier and is shown in Figure 1. The ACB system uses 3,000 cfm
ventilation primary air to induce a total air movement of 10,000 CFM. Each ACB terminal features a 4-pipe system
(heating/cooling) coil.

The compared VAV system alternate consists of a “Dedicated Outdoor Air (DOA) unit, connected to a floor-based
VAV Air Handling Unit (AHU). A perimeter FCU system is also provided for perimeter zones. This integrated DOA and
VAV system is fairly conventional where precise ventilation control is needed. Refer to Figure 3.

Involved DOA systems must typically be capable of providing air at a minimum of 50°F dew-point, considering both
the latent loads generated in the space and the loads from the outside air. In certain cases, recovery heat transfer could also be
used, thereby occasionally satisfying ventilation needs with reduced tempering.

5
 Figure 3 Integrated DOA - Variable Volume System
For the comparison of system performance, a 10,000 square foot office was modeled using an hourly building energy
use program. For the VAV system, a 10,000 cfm system provides heating/cooling with a 20% ventilation air from the DOA
unit. Air is filtered to 10 MERV. Supply air ducts run in the ceiling spaces and connect to approximately 10 VAV boxes.
FCUs are provided to accommodate building skin loads and perimeter zone space conditioning.

MODELING ASSUMPTIONS
The simulated building is a simple rectangle, as shown in Figure 4, representing a typical office space facility.

Figure 4 Modeled Space

The following input data serves to further define assumptions within the model.

• Lighting: 1W/ft2; 80% load to space; Schedule per 90.1-2007.

• Misc. Loads: 2 W/ft2.
• People: 143 ft2/person; 250 Btu/h sensible and 200 Btu/h latent per person; Schedule per 90.1-2007.
• Ventilation: Per 62.1-2007 with 5 cfm per person and 0.06 cfm/ft2; Schedule per 90.1-2007.
• Thermostat: Cooling set-point at 75/81 °F and heating set-point at 72/64 °F; 50% RH.
• Ceiling/Floor: Adiabatic.
• Wall U-value: U-0.064 per 90.1-2007.
• Windows: 33% WWR on all exterior walls; U-0.55 and SC-0.46 per 90.1-2007.
• Height: Floor-to-ceiling height of 10 ft.
• Infiltration: 0.3 ACH for perimeter zones; 0.1 ACH for core zone.
• Economizer (air-side): Dry bulb, 65 °F.
• Minimum Room Relative Humidity: 30%.
• Chilled Water Reset: per Appendix G.
• Supply Air Reset: for VAV system.
• Static Pressures for VAV: 4” supply; 2” return.
• Static Pressures for Fan Coil: 0.5” supply.
• Static Pressures for DOA for Fan Coil: 1.5” total static including supply fan and ret./exh. fan.
• Air-to-Air Energy Recovery: Total-energy wheel at 74% sensible effectiveness and 71% latent effectiveness.

6
 • Chilled Beams Heating: 4-pipe.
• Static Pressure Drop in the Beam: 0.5”or 1” or 2” or 2.5” simulated.
• Static Pressures for ACB: Supply 4.5” WG is assumed.
• Sensible Cooling Capacity for Chilled Beams: 40.27 Btu/h per cfm of primary air flow.
• Auxiliary Cooling Coil Losses to Plenum for the Chilled Beams: 5%.
• BHPs assume 90% motor efficiency.

RESULTS
Computer simulation results are presented for three different cities representing three different climate zones:
• Billings, Montana (Zone 6B)
• Los Angeles, California (Zone 3B)
• New-York, New York (Zone 4A)

Each set of findings is grouped to reflect these climate issues, but also to consider different ACB terminal INP conditions
(i.e. 0.5, 1.0, 2.0, and 2.5 inches WG).

EUI** Supply Fan Ret/Exh Fan Ventilation CHW Pumps HW Pumps

(kBtu/SF/Yr.) BHP (MBtu) BHP (MBtu) BHP (MBtu) BHP (MBtu) BHP (MBtu)
Billings VAV 65.9 10.5 84.3 4.2 27.8 0.7 6.6 1.3 2.8 0.2 0.3
0.5” ACB ACB 70.3 9.0 122.5 5.3 79.2 / / 1.0 2.2 0.1 0.3
INP* Difference -4.4 +1.5 -38.2 -1.1 -51.4 / / +0.3 +0.6 +0.1 0.0
Billings VAV 65.9 10.5 84.3 4.2 27.8 0.7 6.6 1.3 2.8 0.2 0.3
1.0” ACB ACB 71.8 10.0 134.8 5.3 79.7 / / 1.0 2.2 0.1 0.3
INP* Difference -5.9 +0.5 -50.5 -1.1 -51.9 / / +0.3 +0.6 +0.1 0.0
Billings VAV 65.9 10.5 84.3 4.2 27.8 0.7 6.6 1.3 2.8 0.2 0.3
2.0” ACB ACB 74.9 11.7 159.9 5.3 81.0 / / 1.0 2.3 0.1 0.3
INP* Difference -9.0 -1.2 -75.6 -1.1 -53.2 / / +0.3 +0.5 +0.1 0.0
Billings VAV 65.9 10.5 84.3 4.2 27.8 0.7 6.6 1.3 2.8 0.2 0.3
2.5” ACB ACB 77.0 12.7 173.2 5.4 82.0 / / 1.0 2.3 0.1 0.3
INP* Difference -11.1 -2.2 -88.9 -1.2 -54.2 / / +0.3 +0.5 +0.1 0.0
L.A. VAV 56.6 8.4 81.3 3.5 29.7 0.6 6.6 1.3 6.5 0.1 ~0
0.5” ACB ACB 66.9 8.0 107.8 4.7 71.4 / / 1.0 3.5 0.1 0.2
INP* Difference -10.3 +0.4 -26.5 -1.2 -41.7 / / +0.3 +3.0 0.0 -0.2
L.A. VAV 56.6 8.4 81.3 3.5 29.7 0.6 6.6 1.3 6.5 0.1 ~0
1.0” ACB ACB 67.9 8.7 118.5 4.7 71.7 / / 1.0 3.5 0.1 0.2
INP* Difference -11.3 -0.3 -37.2 -1.2 -42.0 / / +0.3 +3.0 0.0 -0.2
L.A. VAV 56.6 8.4 81.3 3.5 29.7 0.6 6.6 1.3 6.5 0.1 ~0
2.0” ACB ACB 69.9 10.2 140.0 4.7 72.2 / / 1.0 3.6 0.1 0.2
INP* Difference -13.3 -1.8 -58.7 -1.2 -42.5 / / +0.3 +2.9 0.0 -0.2
L.A. VAV 56.6 8.4 81.3 3.5 29.7 0.6 6.6 1.3 6.5 0.1 ~0
2.5” ACB ACB 71.0 11.1 150.8 4.7 72.4 / / 1.0 3.7 0.1 0.2
INP* Difference -14.4 -2.7 -69.5 -1.2 -42.7 / / +0.3 +2.8 0.0 -0.2
New York VAV 56.4 8.6 74.4 3.2 27.0 0.7 6.6 1.3 3.8 0.1 0.1
0.5” ACB ACB 67.5 7.8 107.7 4.6 71.3 / / 1.0 2.5 0.1 0.3
INP* Difference -11.1 +0.8 -33.3 -1.4 -44.3 / / +0.3 +1.3 0.0 -0.2

7
New York VAV 56.4 8.6 74.4 3.2 27.0 0.7 6.6 1.3 3.8 0.1 0.1
1.0” ACB ACB 68.8 8.7 118.5 4.6 71.9 / / 1.0 2.5 0.1 0.3
INP* Difference -12.4 -0.1 -44.1 -1.4 -44.9 / / +0.3 +1.3 0.0 -0.2
New York VAV 56.4 8.6 74.4 3.2 27.0 0.7 6.6 1.3 3.8 0.1 0.1
2.0” ACB ACB 70.8 10.2 140.0 4.7 72.4 / / 1.0 2.6 0.1 0.3
INP* Difference -14.4 -1.6 -65.6 -1.5 -45.4 / / +0.3 +1.2 0.0 -0.2
New York VAV 56.4 8.6 74.4 3.2 27.0 0.7 6.6 1.3 3.8 0.1 0.1
2.5” ACB ACB 71.8 11.0 150.9 4.7 72.8 / / 1.0 2.6 0.1 0.3
INP* Difference -15.4 -2.4 -76.5 -1.5 -45.8 / / +0.3 +1.2 0.0 -0.2
ACB INP = Active Chilled Beam Induction Nozzle Pressure.
**EUI = Energy Use Index, (kBTU/GSF/Yr.)

As can be seen, for the systems defined, the VAV system proved to be more energy efficient than the ACB system.
Considering the earlier described characteristics of the two systems, the following Comparison Table results…

Comparison Table

VAV System with Active Chilled Beam

Features
DOAS with DOAS
Filtration Yes No
EUI Low High
Operation and Maintenance Low High
Controllability Yes Difficult
Comfort Yes -
Churn Yes Difficult

CONCLUSION
Potential limitations of the chilled beam have been identified within this paper. Some application conclusions are
summarized as follows:
1. Operational limitations, especially those related to the potential of a colder-than-dewpoint surface temperature
located within the building, can create the possibility of condensation drips into occupied space.
2. The potential for disruption to tenants is a major concern for any modification to refit space, or repair ACB
terminals.
3. There is an inability of chilled beams to effectively condition high-occupancy load spaces or other spaces with
high latent heat gains. This limitation could also apply to accommodating buildings with high infiltration
rates, especially older buildings.
4. There is a lack of nationally-recognized standard certification programs for rating the capacity and
performance of the chilled beams or radiant panels.
5. Secondary air is not filtered and could create long-term unhealthy conditions without proper system
maintenance.
6. In all cases, modeled energy use of the VAV system is less than the ACB system.
7. In considering an ACB system, the owner and design team must work together to assure adequate operating
staff training.