Hpac 870 Nytimeinstallation2

Underfloor-Air-Distribution Design Concepts

May 6, 2008
UFAD systems save energy and deliver better occupant comfort and indoor-air quality for certain applications

Although the number of applications for underfloor-air-distribution (UFAD) systems has increased over the years, the basic design concept for UFAD has not changed. UFAD systems utilize the space under an access floor as a supply-air plenum, taking advantage of thermal stratification to reduce energy use.

The American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) recommends that an occupied zone have a temperature of 73 to 77°F, a relative humidity of 25 to 60 percent, and a maximum velocity of 50 fpm for cooling or 30 fpm for heating. The key to a successful UFAD system is a UFAD diffuser's ability to mix room air into supply air rapidly at low velocities. Because supply air is introduced directly into an occupied zone, it is important that the air meet ASHRAE's recommendations and mix into the space rapidly. This usually is achieved with high-induction swirl diffusers.

Unlike an overhead-air-distribution system, which mixes supply air throughout a zone, a UFAD system mixes air only about 4 to 6 ft above floor level within an occupied zone. Because supply air is introduced directly into an occupied space at floor level, warmer supply air of about 65°F should be used to maintain comfort.

Airflow Requirements

There often is confusion about the use of warmer supply air when calculating a zone's airflow requirements in a UFAD system. Let's look at an example load equation:

British thermal units = 1.08 × cubic feet per minute × ΔT

1.08 = 4.5 (lb per hour) × 0.241 (specific heat of air)

It would appear that the cubic feet per minute would need to be increased to compensate for a reduced ΔT. However, this assumes that the load remains constant. When UFAD systems are utilized, the load is reduced because heat is not mixed from the top of a space into an occupied zone. A typical UFAD system will use the same cubic feet per minute as an overhead system.

Plenum Leakage

Air usually is supplied to an occupied zone through a pressurized plenum. A typical plenum has a depth of 12 to 18 in., with a supply pressure between 0.05 and 0.07 in. wg. However, plenum leakage sometimes can occur. Although supply air may leak from between raised-floor tiles, it already has reached the occupied zone and, therefore, is not of great concern. This can be addressed by offsetting carpet and floor tiles so that the carpet seals the edges of the tiles. If carpet is not being used, an application of sealant between floor tiles can reduce the potential for leakage.

The biggest challenge related to plenum leakage is penetrations in subfloor walls. Supply air can enter spaces between walls, resulting in wasted energy. Although it is critical to seal all of the penetrations in a subfloor when using a UFAD system, adequately specifying the sealing of a subfloor can be a problem when the related work does not fall under HVAC specification.

Eric Mitchell, PE, a senior associate with mechanical-engineering firm Flack + Kurtz, and his team tackled this challenge in the New York Times Building in New York City by coordinating closely with the building's architect. The engineering team and architect detailed the leakage requirements for the mechanical and architectural plans during the design phase of the project. Using leakage values from the floor-tile manufacturer before any openings for electrical floor boxes and diffusers were cut in the tiles, the engineering team tested each zone for leakage using a method similar to a duct-leakage test.

To simulate the carpet that was not yet on site, the engineering team sealed tile seams with duct tape. The floor then was tested for leakage, and leakage rates were determined at the proposed floor pressure of 0.08 in. wg per zone. Five-percent leakage of the total design flow to the space was measured. The tape was removed, and the leakage rates were measured again at the same pressure. The amount of air leaking between the tiles was determined by subtracting the leakage value of the first test (seamed) from that of the second test (unseamed). Several zones were tested, and a constant leakage value from the unseamed test was used as a benchmark for the other floors.

Five-percent leakage on the New York Times Building project is significantly less than what has been reported in older UFAD projects. During the construction phase, it is crucial that contractors understand the importance of these details. Numerous surveys should be planned prior to the close of a project to address leakage concerns associated with UFAD systems.

Energy Savings

One major benefit of UFAD systems is the potential for energy savings. A project's savings potential depends on the climate of the area in which the project is located. However, most projects can profit from at least some UFAD energy benefits. Energy savings in UFAD systems are related to higher supply temperatures and low pressure requirements.

To maintain comfort, UFAD systems use 65°F supply air. In areas with high humidity, supply air must be cooled to 55°F to reduce relative humidity to the ASHRAE-acceptable level of 25 to 60 percent. The 55°F air then must be mixed with the return air to achieve 65°F supply air for an occupied zone.

In areas with low humidity, it may be possible to provide 65°F supply air straight from an air handler. In this case, energy consumption is reduced by raising discharge temperature. Additionally, a UFAD system can utilize an extended economizer cycle with warmer supply-air temperatures to save energy. A study published by the Center for the Built Environment found that the energy savings associated with an extended economizer cycle can be between 15 and 30 percent.1

Because they move a large volume of air with low pressure drops, UFAD systems utilize fans with lower horsepower. Another study found that a UFAD system's fan energy savings can be greater than 20 percent when compared with a standard overhead system.2

For further energy savings, a concrete structural slab can be used to lower peak cooling demand. The use of an underfloor plenum as a supply duct allows the use of a building's thermal mass as an energy "flywheel." With an underfloor plenum ventilated with cool air at night, a building can be cooled enough to significantly lower the load during the early part of the day. However, the occupied space can be overcooled during the night, requiring additional morning warmup and expending more energy.

Climate and building operation are important considerations in the design of a UFAD system. In very humid climates, it may be necessary to operate an HVAC system 24 hr a day to maintain acceptable humidity levels. In these climates, shutting down a UFAD system at night could result in condensation in the underfloor plenum when the system starts up in the morning. This 24-hr operation has to be balanced against the potential for reduced energy use to determine if UFAD is an acceptable overall solution for a project.

Occupant Comfort

Improved occupant comfort is another key benefit of UFAD systems. Because supply air directly enters an occupied zone, UFAD diffusers must mix room air into the supply air rapidly at low velocities so an occupant does not feel drafts or cold temperatures.

An overhead system's occupied zone is defined as the space that exists from the floor to a height of 6 ft and that begins 1 ft away from the walls. The 1-ft area allows space for supply air to flow down the walls and mix with occupied-zone air. However, because they are installed in occupied zones, UFAD diffusers cannot perform the same way. Manufacturers define clear areas around UFAD diffusers, representing the space in which supply and return air is mixed. Outside this area, diffusers can reach temperatures and velocities recommended by ASHRAE for occupied zones.

The best measure of comfort for an air-distribution system is the Air Diffusion Performance Index (ADPI). The ADPI is a single-number means of relating temperatures and velocities in an occupied zone to occupants' thermal comfort. ADPI values are calculated by measuring temperatures and velocities within an occupied zone at four heights: 4 in., 24 in., 42 in., and 66 in. from the floor, as described in Appendix C of ASHRAE Standard 113-2005, Method of Testing for Room Air Diffusion. Effective draft temperature (Θ) then is calculated for each height:

Θ = (tx - tc) - 0.07(Vx - 30 fpm)

0.07 = recognized constant for turbulence
tx = local air temperature
tc = control temperature
Vx = local air velocity

The ADPI is the percentage of points within a temperature range of -3°F to +2°F from the four reference points with a room velocity of 70 fpm or less.

Although ADPI is designed for the cooling of mixed-air systems only, the concept of maintaining a space with an effective draft temperature in the range of -3°F to +2°F still is important for UFAD systems. UFAD systems supply small volumes of air throughout a space. A typical UFAD swirl diffuser supplies around 100 cfm of supply air, which is distributed evenly throughout a space. This creates a comfortable space outside the manufacturer-defined clear area. A study published by Lawrence Berkeley National Laboratory found that occupants' satisfaction with thermal conditions in buildings utilizing UFAD systems was well above average.3

An additional benefit of UFAD systems is that their diffusers typically are occupant-adjustable. Studies have found that occupants are more comfortable when they have some control over their environments. Occupants who do not have control over their environments are twice as sensitive to variations in temperature.2

Indoor-Air Quality

The potential for increased employee productivity is another benefit of UFAD systems. Several studies have tied improved indoor-air quality (IAQ) to improved employee productivity.

For example, the aforementioned Lawrence Berkeley National Laboratory study found that U.S. businesses could save as much as $58 billion in lost sick time and $200 billion in improved worker performance if IAQ were upgraded.3 Another study found that by reducing allergens, respiratory illnesses, and sick-building syndrome, businesses could save $43 billion to $235 billion in lost employee productivity.4

UFAD systems can provide improved IAQ and help contribute to increased productivity by supplying air directly to an occupied space. A study published in Atmospheric Environment used computational-fluid-dynamic modeling to determine the particle-removal performance of UFAD systems and ceiling and sidewall supply overhead return systems.5 The model found that UFAD systems have the best particle-removalefficiency and the greatest potential to reduce cross-contamination. The Lawrence Berkeley National Laboratory study found that carbon-dioxide pollutant-removal efficiency was 13 percent higher in UFAD systems than that expected from overhead systems.3 These results suggest UFAD systems can reduce occupant exposure to generated pollutants by 13 percent.

UFAD-System Challenges

One challenge for UFAD systems is how to handle the high load of a building perimeter without sacrificing occupant comfort. The perimeter typically is the most difficult area of an underfloor system to design. In many of the first U.S. UFAD projects, the perimeter was handled through the use of fan-powered terminals ducted to linear bar diffusers. However, this construction can have complications. First, although UFAD fan terminals are designed to fit within a raised-floor pedestal grid, they may require periodic maintenance, requiring access to a fan terminal that may interfere with a building's layout.

Another challenge is that the throw of linear bar diffusers can be very long, about 15 to 20 ft. Not only does the long throw from the grille waste energy by mixing the air above a stratified layer into an occupied zone, it can stream up perimeter windows and across the ceiling, where it can drop into the occupied zone and cause discomfort for occupants near the perimeter. Many building owners who use this system have designed the perimeter as a walkway so that high velocity and long throws do not affect building-occupant comfort.

Better designs that can handle the high loads of perimeters without sacrificing occupant comfort are available. George Karidis, vice president and director of mechanical engineering for SmithGroup Inc., first used this concept on the Visteon Village project in Van Buren Township, Mich. The perimeter cooling system utilizes pressurized airflow under a plenum, eliminating fan-powered terminals and their related challenges in the raised-floor plenum. Linear bar diffusers use aperture plates to create induction in supply-air streams, providing an engineered plume that does not break the stratified layer of a UFAD system.

On subsequent projects, SmithGroup has utilized a linear-bar-diffuser heating system that uses induction and natural currents to heat the perimeter without reheating supply air. This system meets ANSI/ASHRAE/IESNA Standard 90.1, Energy Standard for Buildings Except Low-Rise Residential Buildings, which notes that air should not be cooled and heated simultaneously.


Much has been learned about UFAD systems since they first gained popularity in the United States during the mid-1990s. UFAD systems are not right for every building, but they can save energy and provide improved occupant comfort and IAQ for the right application.


1) Webster, T., Bauman, F., & Ring, E. (2002). Supply fan energy use in pressurized underfloor air distribution systems. Berkeley, CA: Center for the Built Environment.

2) Bauman, F. (1999, October). Giving occupants what they want: Guidelines for implementing personal environmental control in your building. Paper presented at World Workplace 99, Los Angeles, CA.

3) Fisk, W., Faulkner, D., & Sullivan, D. (2004). Performance of underfloor air distribution: Results of a field study. Berkeley, CA: Lawrence Berkeley National Laboratory.

4) Fisk, W. (2000). Health and productivity gains from better indoor environments and their implications for the U.S. Department of Energy. Berkeley, CA: Lawrence Berkeley National Laboratory.

5) Zhang, Z. & Chen, Q. (2006). Experimental measurements and numerical simulations of particle transport and distribution in ventilated rooms. Atmospheric Environment, 40, 3,396-3,408.

Director of marketing and product development for Titus, Jenny Abney is preparing for the Leadership in Energy and Environmental Design accreditation exam. She has a bachelor's degree in physics and a master's degree in business administration. Her background provides a mix of technical knowledge and business acumen in bringing products to market.