Air movement can play an important role in the thermal comfort of man and beast. A breeze on a humid summer day can make a significant difference to one’s thermal comfort. Recent strategies for improving energy-efficiency in buildings attempt to take account of the cooling effects of air movement from natural ventilation. When the building envelope is closed for air conditioning, local air movement is kept below 40 ft/min. This ignores the option of increased air movement to reduce the cooling energy in air conditioned space. This paper explores opportunities for saving energy by utilizing the effects of indoor air movement.

  1. Cooling energy savings in air conditioned space from elevated air speed

The current edition of ANSI/ASHRAE Standard 55-2004 Thermal Environmental Conditions for Human Occupancy (ASHRAE, 2004), provides for limited increases of summer thermostat temperature settings by increased local air speed. Figure 1 is derived from Figure 5.2.3 in the Standard 55-2004.

The curves of equal heat loss from the skin for combinations of operative temperature and air movement are referenced to the upper limit of the comfort zone (PMV= +0.5). Limits of 160 fpm and 5.4ºF are set for sedentary activity, 1.0 to 1.3 met. Large individual differences in preferred air speed
requires that occupants have personal control of air speed in increments of 30 ft/min.

The Standard states that it is acceptable to interpolate between these curves. Air speed is more effective at offsetting increases in temperature when mean radiant temperature is greater than the mean dry bulb air temperature.

It should be noted that there are two errors in Figure 5.2.3 of the Standard. The “18°C” should read “18°F” and there is a scaling error between the fpm and m/s scales.

Five separate curves are provided to accommodate temperature differences of -18°F, -9°F, 0.0°F, +9°F, and +18°F between mean radiant temperature, tr , and mean dry bulb air temperature, ta. The writer fitted equations to the portion of the curves limited to sedentary activity of 160 fpm and 5.4°F for 1.0 met to 1.3 met and 0.5 to 0.7 clo.

The writer also fitted equations to the portion of the curves for activity beyond the sedentary limits. Cooling effect limits for these equations fitted to curves in Figure 5.2.3 in the Standard 55-2004 were 300 fpm and 8°F.

2.1 Curve for tr – ta = 0.0 K

For tr – ta = 0.0°F, an air speed of 160 fpm permits a thermostat set point increase of 4.4°F limit for light sedentary activity (1 to 1.3 met) and 0.5 to 0.7 clo.

V = 40 + 6.8”t 1.85 (1)
Where V is the mean relative air speed in fpm and ”t is the cooling effect in °F.

In most thermostatically controlled air hvac in Surfside Beach conditioned spaces, wall, ceiling and floor surfaces temperatures are close to air temperature. That is tr – ta = 0°F. Conditions when tr — ta is not zero include spaces with poorly insulated windows, walls or ceilings where the outer surface is exposed to direct solar radiation or cold winter conditions.

2.2 Curve for tr – ta = +9°F

For tr – ta = +9°F an air speed of 160 fpm permits a thermostat set point increase of 5.4°F limit for light sedentary activity (1 to 1.3 met) and 0.5 to 0.7 clo.

V = 40 + 1.26”t 2.85 (2)
Where V is the mean relative air speed in fpm and ”t is the cooling effect in °F.

2.3 Curve for tr – ta = +18°F

For tr – ta = +18°F an air speed of 126 fpm permits a thermostat set point increase of 5.4°F limit for light sedentary activity (1 to 1.3 met) and 0.5 to 0.7 clo.

V = 40 + 1.28”t 2.7 (3)

  1. Beyond Sedentary Activity limits

The Standard is not clear on constraints for the portions of the curves up to 89°Fand 300 fpm, beyond the limits set for sedentary activity. Studies have measured the cooling effect of air movement up to 600 fpm in warm climate conditions (Khedari et al, 2000, Tanabe and Kimura, 1994, and Scheatzie et al, 1989). Air movement higher than 160 fpm is used in air conditioned gymnasia and shopping malls to augment cooling of occupants. The writer has fitted equations to the portion of the curves for activity beyond the sedentary limits

For tr – ta = 0.0°F an air speed of 300 fpm indicates the thermostat set point increase could be 6.6°F at activity levels higher than 1.3 met.

V = 40 + 2.52”t 2.5 (4)

Limits for Equation 4 are 160 fpm to 300 fpm and 4.4 F to 6.6 F

For tr – ta = +9ºF an air speed of 276 fpm permits a thermostat set point increase of 8ºF at activity levels higher than 1.3 met.

V = 40 + 5.7”t 1.8 (5)

Limits to Equation 5 are 160 fpm to 280 fpm and 5.4ºF to 8ºF .
For tr – ta = +18ºF an air speed of 211 fpm indicates the thermostat set point increase could be 8ºF at activity levels higher than 1.3 met.

V = 40 + 6.3”t 1.59 (6)

Limits for Equation 6 are 132 fpm to 209 fpm and 5.48ºF to 8ºF.

  1. Estimating Cooling Energy Savings

The electrical US utility corporation Exeloncorp (2005), suggests that domestic air conditioning cooling costs can be reduced by 3% to 4% for each ºF that the thermostat setting is raised in summer.

Occupants can offset an increased thermostat setting of 4.7ºF by providing 160 fpm of low-cost air flow from circulator fans and enjoy normal comfort while saving air conditioning operating cost. On the basis of the Exeloncorp (2005) recommendation, an increase in the thermostat setting of 4.7ºF would provide cooling energy savings from 14% to 19%. In gymnasia where higher air movement is acceptable the savings from a thermostat increase of 8ºF could be from 24% to 32%. A detailed analysis of reduction in residential cooling loads due to air flow was performed for six US cities in a variety of climate zones (Byrne and Huang, 1986)

  1. Comparison of fans and room air conditioners

A detailed comparison of the energy required to maintain the same thermal comfort in a 141.5 ft2 bedroom in Townsville, Hope (2003), was conducted using a 55 inch diameter residential ceiling fan and a VF100C Carrier window/wall room air conditioner, sized for the room by engineers at the local distributor. The measured rate of power consumption of a 55 inch diameter ceiling fan operating at its top speed was 0.068kW or 0.48 W/ft2 of floor area. This is 8.7% of the power used by the room air conditioner to achieve the same thermal comfort. The rate of power consumption of the window/wall room air conditioner was 0.78 kW, or 5.51 W/ft2 of floor area. This is 11.5 times the power used by the ceiling fan.

  1. Destratification

In heated spaces in winter, indoor air tends to stratify with the hottest, less dense, air accumulating under the roof due to the gravity force. This condition creates two problems. Firstly the hottest air is not contributing to the thermal comfort of occupants near floor level, and secondly, it creates a high temperature difference between the underside of the roof and the exterior of the roof that increases heat losses through the roof.

Destratification is the process of thoroughly mixing indoor so that air temperature near the floor is the same as the air temperature under the roof, or no more than 2ºF difference. This is done using circulator fans. In a typical US distribution warehouse with a 30 ft high ceiling, the seasonal heating energy savings from effective destratification is around 20% to 30%. To be effective about one half of the total volume of air in the space needs to be moved from ceiling level to floor level per hour.

To be effective in destratification the fan should be no more than 1 diameter below the ceiling and the jet from the fan must impact on the floor in order to achieve effective circulation. Jets from ceiling fans have an effective throw of 5 to 6 diameters.

In large buildings with high ceilings such as churches, industrial buildings or distribution warehouses, a large volume of air needs to be circulated. In order to avoid complaints of drafts from occupants, the local air velocity at head height needs to be kept less than 40 ft/min.

Circulator fans are much more energy-efficient at low speeds, so large diameter, slow moving, fans are well suited for destratification. One 24 ft diameter industrial ceiling fan operating at top speed of 42 rpm uses 1.67 kW of electrical power but only 0.06 kW operating at 14 rpm its peak efficiency. At 42 rpm this fan delivers around 337,700 cfm of air and 76,670 cfm at 14 rpm. An added benefit of operating large fans at low speed compared to smaller fans at higher speeds is the reduction in fan noise. Large slow moving fans are virtually silent.

  1. Estimating Destratification Energy Savings

A recommended method for estimating heating energy savings from destratification is to determine the lumped seasonal heat transfer rate for the building envelope and determine the difference in heat loss before and after destratification (Pignet and Saxena, 2002).

The lumped seasonal heat transfer rate for the building envelope in Watts can be calculated using:

A x U = qbd / (ti -to) (7)

Where: A is the surface area of the building envelope in ft2; U is the lumped heat transfer coefficient for the building envelope in Btu/ft2.h.ºF; qbd is the rate of heat loss through the building envelope in Btu/h before destratification; and ti -to is the average heating season indoor to outdoor air temperature difference in ºF.

The total heat lost from the building is the sum of heat released from furnaces plus heat released in the space from other sources such as lighting, people, machinery or manufacturing processes. The heat released from the furnaces can be determined from the fuel bills for the season, the caloric value of the heating fuel and the system efficiency. The caloric value of natural gas is around 1000 Btu/ft3. The time used in these calculations is the heating season associated with the measured fuel consumption.

Forced air furnaces with flues have efficiencies around 0.7. Radiant heaters without flues have an efficiency of 0.8. Electrical heaters have an efficiency of 1.0. Heat from other sources is estimated in the normal way as set out in HVAC handbooks (ASHRAE, 2005).

With the overall heat loss U x A for the heating season before destratification determined, the reduction in heating after destratification, qad can be determined from:

qad = U x A x (tibd – tiad) (8)

Where: qad = Reduced heat load after destratification in Btu/hr; U = Lumped time-averaged heat loss rate for the building envelope in Btu/hr.ft2.ºF ; A = Surface area of the building envelope, ft2; tibd = Heating season average indoor air temperature before destratification,,°F;; This depends on vertical temperature profile. This should be measured on site because the shape of the temperature profile can vary substantially depending on type of heaters, their height above floor level, and how ventilation is provided; tiad = Heating season average indoor air temperature after destratification, °F. This is taken as the thermostat set point as the indoor air temperature throughout the space is close to uniform after destratification.

The reduced heating load due to destratification can be converted into a quantity of fuel taking into account the efficiency of the heating system and the caloric value of the fuel. The heating fuel cost saving typically between 20% and 30% is calculated using the unit cost of fuel.

  1. Thermal comfort in Non-air Conditioned Space

The ANSI/ASHRAE 55-2004 Standard offers a method for determining an acceptable range of indoor operative temperature in occupant-controlled, naturally conditioned spaces. Occupant-controlled, naturally conditioned spaces are defined as spaces where thermal conditions of the space are regulated primarily by the occupants through opening and closing windows. These are spaces with no refrigerated air conditioning, radiant cooling, or desiccant cooling. Fans can be used when natural ventilation does not provide sufficient air movement.
In such spaces, occupants have different expectations of thermal comfort and accept wider ranges of thermal conditions in both winter and summer than occupants of air conditioned spaces. This method is intended for climates where mean monthly air temperatures fall in the range of 50°F to 92°F. This method is generally described as the Adaptive Model (de Dear and Schiller (2001).

Using the adaptive approach, the first step is to determine the average monthly temperature for each month of the cooling season for the location. In ventilated buildings without air conditioning, temperature for operative comfort toc, is based on mean monthly outdoor air temperature tout, and can be calculated using the following equation (ASHRAE, 2005).

toc = 66 + 0.255(tout – 32) (9)

The comfort zone range of operative temperature to satisfy 80% of acclimatized people can be read of a graph in the Standard or by adding and subtracting 6.3 ºF to the operative comfort temperature.

With a mean daily air temperature of 83.6ºF in the city of Houston during July, toc = 66 + 0.255(83.6 -32)= 79.2 ºF. The thermal comfort zone to satisfy 80% of people in July is then 72.9ºF to 85.5ºF.

Given the long term average monthly outdoor air temperature for Houston TX in July is 83.6ºF, this presents the average need for a cooling effect from air movement in January of 83.6ºF – 79.2ºF or 4.4ºF to restore the operative temperature to the norm. The question now is how much air movement is needed to achieve a cooling effect of 4.4ºF? Using the data from Khedari et al (2000), for a warm humid climate with a relative humidity of 75% indicates 87 fpm is needed for a 4.4ºF cooling effect.

  1. Cooling effects of air movement in naturally conditioned spaces

The US Naval Medical Command (1988) in a chapter on relieving heat stress published data on the relative cooling effect of air movement Figure 7. These data do not provide a quantitative cooling effect but are useful in that they indicate the maximum cooling effect occurs with air movement around 1,500 fpm.

In naturally conditioned space, there is no control of humidity. As the cooling effect of air movement in warm environments relates to evaporative cooling from sweating, it has been shown that as humidity increases, the cooling effect of air movement decreases. The reduced cooling effect is much greater in warm humid environments when air movement needed for thermal comfort exceeds 295 fpm, Figure 6 (Khedari et al, 2000). It is important to use cooling effect data derived from local climate and cultural conditions. These data will better reflect the thermal comfort expectations of local people taking into account local dress and typical levels of metabolic activity.

A variety of approaches have been taken by researchers to quantify the cooling effects of air movement. Cooling effects of air movement can effective in hot arid environments were evaporative cooling of the skin is not encumbered by high humidity (Scheatzle et al, 1989).
Another equation derived from several studies (Szokolay, 1998) that is widely used for estimating the cooling effects of air movement from 40 ft/min to 400 ft/min is:

”t = 10.8((V/197.85)-0.2)-1.8((V/197.85)-0.2)2 (11)

Where V is in ft/mim and ”t is in ºF.

Using this equation, air movement of 400 ft/min provides a cooling effect 13.7 ºF. This is equivalent to Khedari et al cooling effect for 400 ft/min at 57% relative humidity in Thailand.

  1. Indoor air movement for livestock

Dairy farmers have learned from university studies that thermally comfortable cows’ milk production, reproductive health and growth are much better than those of cows subjected to summer heat stress (Sanford, 2004). During hot summer periods dairy farmers have installed small high speed circulator fans to achieve the recommended air movement of 177 ft/min to 433 ft/min. Ten 36 inch diameter fans operating at 825 rpm use 3.73 kW of electrical energy. Farmers have found they can replace 10 of these 36 inch diameter fans with a single 24 ft diameter fan operating at 42 rpm that uses only 1.6 kW of electrical energy while providing the same air movement. Additional cooling can be achieved in drier climate regions using misting water sprays for evaporative cooling.

  1. Discussion

All the descriptions of air movement described so far in this document have referred to the average velocity of air movement. Olesen (1985) refers to a study by Fanger and Pedersen of the chilling effect of winter draughts. It was observed in the study that the chilling effect of gusting air flow reached a peak around a gust frequency of 0.5Hz.

More recently researchers in China (Xia et al,2000) repeated these studies inwarm, humid conditions with temperatures ranging from 79ºF to 87ºF and relative humidity between 35% and 65%. These experiments showed that the preferred gust frequency for cooling air movement was between 0.3Hz and 0.5Hz. Approximately 95% of subjects preferred gust frequencies below 0.7Hz. Natural breezes and air flow from large low-speed circulator fans have a significant portion of their energy spectral density around this frequency of 0.5Hz. Olesen (1985) suggested the use of an equivalent uniform air velocity, Table 1, to account for this effect but this enhanced cooling effect has not been specifically accounted for in cooling effects of air movement to date.

  1. Conclusions

Current air conditioning design provides for uniform air temperature and humidity throughout a space, with imperceptible local air movement in the occupied zone of less than 40 ft/min. This conventional design is based on air conditioning heating and cooling loads that ignore the substantial savings to be gained from increased indoor air movement from circulator fans.

Recent ASHRAE acceptance of an adaptive thermal comfort model clearly shows that people who live in air conditioned houses, drive air conditioned cars, work in air conditioned offices impair their natural thermal comfort adaptation. This impairment results in unnecessarily high summer cooling loads.

Where naturally conditioned buildings are acceptable, indoor thermal comfort can be achieved with substantial energy savings by better utilization of indoor air movement.

The cooling effect of air movement has been well established by a number of researchers. There remains a need for further research on the cooling effects of air movement on building occupants to accommodate activity levels beyond 1.3 met, higher air velocities for non-sedentary activity, and lighter clothing levels than 0.5 clo. This research is needed in both air conditioned and naturally conditioned spaces.

Research on the cooling effects of air movement has been presented in many forms. The chart produced by Khedari et al (2000) is one the better formats. Further research is needed to develop a form which presents data in a way that makes it more easily used by engineers to improve energy efficiency with increased indoor air movement.

The same circulator fans used to enhance summer thermal comfort can be used to destratify indoor air to save heating energy in winter. This particularly applies to commercial or industrial spaces with high ceilings.


ASHRAE (2005) ASHRAE 2005 Handbook of Fundamentals, ASHRAE, Atlanta, GA. Page 26.11.

ASHRAE (2004) ANSI/ASHRAE Standard 55-2004 Thermal Environmental Conditions for Human Occupancy. ASHRAE, Atlanta, GA.

Byrne, S. and Huang, V.(1986) The impact of wind-induced ventilation on residential cooling load and human comfort. ASHRAE Trans. Vol.92, Pt. 2, 793-802.

de Dear, R. and Schiller Brager, G. (2001) The adaptive model for thermal comfort and energy conservation in the built environment. Int. J. Biometeorology, 45: 100-108.

Exeloncorp (2005) Controlling Temperatures is accessible on the internet at:

Fountain, M. (1995) An empirical model for predicting air movement preferred in warm office environments. Standards for thermal comfort: Indoor air temperatures for the 21st century. Edited by F. Nicol, M. Humphreys, O. Sykes and S. London, Roaf, E & F Spon. pp. 78-85.

Hope, P (2003) Energy efficiency ratings: Implications for the building industry in the humid tropics. Master in Tropical Architecture dissertation, Australian Institute of Tropical Architecture, James Cook University, Townsville, Australia, pp. 377.

Khedari, J., Yamtraipat, N., Pratintong, N. and Hinrunlabbh, J. (2000) Thailand ventilation comfort chart. Energy and Buildings, Vol. 32, pp. 245-249.

Naval Medical Command (1988) Manual Of Naval Preventive Medicine, Chapter 3, page 3-7. Accessible on the internet at:

Pignet, Tom and Saxena, Umesh (2002) Estimation of energy savings due to destratification of air in plants, Energy Engineering, Vol 99, No. 1, 69-72.

Sanford, S. (2004) Energy conservation in agriculture: Ventilation and cooling systems for animal housing. University of Wisconnsin Cooperative Extension publication A3784-6, pp.3.

Scheatzle, D., Wu, H. and Yellott, J.(1989) Extending the summer comfort envelope with ceiling fans in hot, arid climates. ASHRAE Trans. Vol.100, Pt. 1, 269-280.

Szokolay, S. (1998) Thermal comfort in the warm-humid tropics, Proceedings of the 31st Annual Conference of the Australian and New Zealand Arch. Science Association, Uni. of Queensland., Brisbane, Sept.29-Oct.3, pp. 7-12.

Tanabe, S and Kimura, K. (1994) Importance of air movement for thermal comfort under hot and humid conditions. ASHRAE Trans. Vol. 100, Pt. 2, 953-969.

Xia, Y., Zhao, R. and Xu, W. (2000) Human thermal sensation to air movement frequency. Reading, UK. Proceedings of the 7th International Conference on Air Distribution in Rooms. Vol.1, pp. 41-46.