Volume 13, Number 7
- What Is Thermal Comfort, and What Factors Affect It?
- ASHRAE Standard 55
Expanding the Engineers' Comfort Zone: Working with Adaptive Thermal Comfort
The most common complaint facility managers hear from building occupants is that their office space is too cold. That would seem an easy enough problem to solve, except for the fact that the number two complaint is that it’s too hot. Different people, it turns out, are comfortable under different conditions, and keeping everyone comfortable at the same time is an elusive goal at best. Offices are considered thermally successful if only 80% of their occupants are reasonably comfortable at any given time. This is the goal laid out by the American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (ASHRAE®) in the industry’s gold standard of comfort—Standard 55, Thermal Environmental Conditions for Human Occupancy.
Humans have been creating spaces to induce thermal comfort for eons. The long-established components of comfort include air temperature, mean radiant temperature, humidity, and air velocity within the space, along with the personal factors of clothing insulation and activity level. But our understanding of what makes a space comfortable is still evolving, and these components, we’re discovering, represent only part of the puzzle of thermal comfort.
A collection of less tangible components also plays a role. These include the amount of control occupants have over the thermal conditions of their environments and their expectations stemming from the weather, the season, and even their own habits and culture. We expect outdoor temperatures to be colder during the winter, for example, and we actually shift our range of indoor comfort slightly to accommodate the corresponding shift outside.
Although it is not fully understood how important adaptability is relative to the traditionally recognized factors of thermal comfort, or how the various factors interact with one another, it is increasingly accepted that adaptive comfort plays some measurable role in determining the comfort of building occupants. Even ASHRAE’s Standard 55 now recognizes adaptability; the just-released 2004 version includes an expanded comfort zone for optional use in naturally ventilated spaces.
The adoption of a more adaptive comfort model is significant for the world of green building. If we can achieve occupant comfort with greater swings in temperature or humidity, then less energy is needed to condition the air. The acceptance of an adaptive model could also result in buildings that keep more people comfortable more of the time—which might increase productivity, keep people healthier, and have other benefits. This article takes a look at how our evolving understanding of comfort could shift our expectations for buildings, encourage the design of naturally ventilated buildings, and increase occupant comfort and productivity, all while reducing operating costs and saving energy.
What Is Thermal Comfort, and What Factors Affect It?
I can’t define it, you might say, but I know it when I feel it. Thermal comfort is hard to define and perhaps harder to measure. According to ASHRAE, thermal comfort is “the condition of mind that expresses satisfaction with the thermal environment.” Even ASHRAE recognizes the difficulty inherent in defining something so abstract: “It requires subjective evaluation,” their definition concludes.
Most of us would agree that 85°F (29°C) feels fine outdoors, in the shade, with a breeze blowing, but miserable in a sealed office building. We have known for a long time that thermal comfort is affected by far more than just air temperature. Air temperature, the degrees in Fahrenheit or Celsius of the ambient air, is the most obvious of the physical factors, one category of conditions that affect comfort. Also important is mean radiant temperature, the average temperature of all nearby surfaces, weighted according to the emissivity of those surfaces.
Another variable in the comfort equation is humidity. Absolute humidity refers to the amount of water vapor in the air, usually expressed as pounds of moisture per pound of dry air. As air temperature rises, it can hold more moisture, and relative humidity (RH) is used to express the amount of water vapor in the air as a percentage of the total amount it could hold at that temperature. Although the role of humidity at temperatures within the comfort range is relatively small, its influence on temperatures outside the comfort range can be great, making already warm temperatures unbearable.
The saying “It’s not the heat; it’s the humidity” highlights the role of moisture in thermal comfort. Since air with low RH can absorb more moisture, under these conditions sweat will evaporate from our skin more quickly, cooling us more effectively. As RH rises, sweating becomes a less effective means of cooling the body until, at 100% RH, the air can absorb no more moisture, rendering sweating ineffective. Since it encourages mold growth, high humidity also poses a threat to indoor air quality, sometimes at levels below the upper limits recommended for comfort. Although low humidity is never a problem for thermal comfort, it can cause other problems, such as dry eyes and static electricity.
A fourth variable is air velocity. Fans and breezes make us feel cooler not because they introduce cooler air to the space, but because they move air across our skin, causing heat loss by convection and inducing evaporative cooling. The effect of air velocity on comfort varies with the other factors, but ASHRAE Standard 55 predicts that air moving at 100 fpm (0.5 m/s) can offset temperatures of 2°F to 4°F (1°C to 2°C) above the normal comfort zone, and an air velocity of 250 fpm (1.2 m/s) can offset temperature increases of 4°F to 10°F (2°C to 5.5°C). In his book Heating, Cooling, Lighting: Design Methods for Architects (see EBN ), Norbert Lechner gives a frame of reference: a gentle breeze outdoors is around 900 fpm (4.5 m/s), he says. Indoors, 200 fpm (1 m/s) is the upper limit for comfort in air-conditioned spaces, and a good speed for natural ventilation in hot, dry climates; 400 fpm (2 m/s) is good for natural ventilation in hot, humid climates, according to Lechner.
The fact that airflow can expand our comfort range is behind the design of “dogtrots,” or open hallways, in American southern-climate buildings. A dogtrot takes advantage of the Venturi effect to speed air through the narrow portions of the building, thereby making the occupants feel cooler. Air velocity can make us uncomfortable for other reasons, however; if the velocity is too great, for example, it will blow papers from desks. Both the 1992 and 2004 versions of ASHRAE Standard 55 call for an upper airspeed limit of 160 fpm (0.8 m/s), significantly lower than the recommendations of Lechner and others.
Personal factors constitute a second category of comfort elements. Clothing insulation is perhaps the most intuitive factor in the comfort equation. When we’re cold, we don warmer clothing; when we’re hot, we remove layers or opt for loose, breezy fabrics. Clothing insulation is measured in the unit “clo,” defined in amazing (and amusing) detail; clo levels range from 0.01 for a bra to 0.69 for a long, thick, long-sleeved wrap robe. Clo levels are additive, and a standard men’s business suit has been set at 1.0 clo, including the combined values for “briefs; broadcloth, long-sleeve shirt; single-breasted suit jacket; tie; straight, long fitted trousers; calf-length socks; hard-soled shoes.” Furniture should also be considered; typical padded office chairs insulate us as we work.
Although clothing level can expand the thermal comfort zone, making us comfortable in a wider scope of conditions, it also limits the comfort zone when occupants have little or no choice over the clothes they wear. Uniforms, cultural standards, and fashions all limit our control over our clothing. As Lechner points out, “We could save countless millions of barrels of oil if men wore three-piece suits only in the winter and women wore miniskirts only in summer.” This is precisely the point former President Carter made by wearing a cardigan on national television during the 1970s’ energy crisis. “If it’s cold,” he said, “turn down the thermostat and put on a sweater.”
The second personal factor, activity level, is also familiar; physical activity raises the body’s rate of energy production, or metabolic rate, keeping us warm at otherwise chilly temperatures or making us uncomfortably warm at temperatures that might be perfectly comfortable if we were sitting still. Metabolic rates are measured in units called “mets”’—one met is 18.4 Btu/h·ft2 (58.2 W/m2), and typical met levels are 0.7 for sleeping, 1.0 for sitting quietly, and as high as 8.7 for playing active sports. Note that the metabolic rate varies significantly from one person to another.
Physical and personal factors interact in complex ways, affecting the comfort of building occupants. But thermal comfort—and designing spaces to make people comfortable—is more complex than even all of this suggests. “Occupants drive comfort much more than the environment in which you place them,” says Walter Grondzik, of Florida A&M University. Engineers are just starting to recognize the role of adaptive factors, the third component of the thermal comfort conundrum. Gail Brager, Ph.D., associate professor in the Department of Architecture at the University of California at Berkeley and vice chair of the ASHRAE Standard Project Committee 55, and Richard de Dear, Ph.D., faculty member in the Division of Environmental and Life Sciences at Macquarie University in Sydney, Australia, have led the effort to convince ASHRAE to recognize adaptive comfort. In their 2000 ASHRAE Journal article “A Standard for Natural Ventilation,” they describe three elements of adaptive thermal comfort.
The most well-documented and widely accepted of these newly recognized variables is behavioral adaptation, or control. Behavioral adaptation, according to Brager and de Dear, includes both conscious and unconscious actions that we take to adjust our thermal environment. Examples include changing clothes or activity levels (which are also considered personal factors), as well as adjusting the environment itself, by turning on a fan or opening a window, for example. “In a naturally ventilated building, occupants are more in the driver’s seat,” says Grondzik. “Given some choices and opportunities, people are willing to adapt and expand the comfort zone.”
The second component of adaptive comfort is physiological adaptation, also called acclimatization. Physiological adaptation refers to biological changes caused by “prolonged exposure to characteristic and relatively extreme thermal conditions,” according to Brager and de Dear. People physically adapt to hot climates, for example, by beginning to sweat at lower temperatures. While acclimatization may come into play in extreme conditions, Brager and de Dear report that it is insignificant for typical office conditions.
The third component of adaptive comfort is psychological adaptation, which “refers to an altered perception of, and reaction to, physical conditions due to past experience and expectations,” according to Brager and de Dear. These expectations—relating to weather, season, routine, and culture—actually shift our feelings of comfort. If the weather is cool, for example, building occupants are comfortable at a slightly lower range of temperatures.
Another aspect of psychological adaptation is our dislike of unchanging conditions, which results in lethargy and listlessness—the phenomenon called thermal boredom. In his foreword to the 1996 Sustainable Design Guide of the Japan Institute of Architects, Amory Lovins, cofounder and CEO of Rocky Mountain Institute, complains that “the typical Western mechanical engineer would strive to eliminate every … pesky trace of variability with thermostats and humidistats and photosensors, to render the human experience uniform and constant down to the last lux of light and molecule of air—as if people were dead machines, not dynamic organisms.” In his 1980 book Indoor Climate, Donald McIntyre captures the importance of fluctuating indoor temperatures: “It can be argued,” he says, “that achieving a steady optimum temperature is akin to finding the most popular meal at the canteen and then serving it every day.”
“Actually, the variety of human responses that get involved in the adaptive comfort concept has been studied for a long time in the human factors literature,” notes James Wise, Ph.D., CEO of Eco·Integrations/Integral·Visuals, Inc. and associate professor in the Department of Psychology and the Environmental Sciences and Regional Planning Program at Washington State University’s Tri-Cities campus. He says that while these ideas have been ignored for too long, “anytime you can get anyone in engineering to pay attention to anything that comes out of psychology, it’s a major victory.”
ASHRAE Standard 55
Since it was first introduced in 1966, ASHRAE Standard 55 has defined thermal comfort in the U.S. and Canada. It applies to most buildings—residential and commercial, new and existing—in which occupants are engaged in “light, primarily sedentary activity.” “I think the ASHRAE standard is extremely important,” says Wise. “It’s something every engineering and architectural building professional has learned in school, and it has become de facto accepted as current best practice around thermal comfort.” Though it’s not part of any building code, it often provides both leasing guidelines and legal cover. “What ASHRAE says is important, because, right, wrong, or indifferent, it is often deemed a reasonable benchmark. It gives designers a reference for discussions with a client,” says Paul Anseeuw, vice president at Keen Engineering. That doesn’t mean, however, that architects always design to stay within the ASHRAE guidelines. “All design is about managing risk,” says Anseeuw, adding, “You need to know what the waterline is.”
The 1992 version of Standard 55, with its expectation of narrowly controlled, constant conditions, establishes a benchmark that is impossible to meet with natural ventilation in almost any climate. “Plus or minus 75°F [24°C] was a good place to start the conversation if you were designing a conventional system, but it was an awkward place to start the conversation if designing a naturally ventilated building,” notes Anseeuw. Standard 55–1992 was designed for buildings with mechanical conditioning systems, based on laboratory tests of people’s comfort. As Brager and de Dear point out, “People were considered passive subjects of climate change in artificial settings, and little consideration was given to the broad ways they might naturally adapt to more wide-ranging thermal environments in realistic settings.” The model of thermal comfort used in that standard, they explain, “cannot account for the complex ways people interact with their environments, modify their behaviors, or gradually adapt their expectations to match their surroundings.” It does not, in other words, account for human adaptation.
To research the adaptive theory of comfort, Brager and de Dear examined 21,000 sets of raw data collected from 160 office buildings (both mechanically and naturally ventilated) on four continents. Measurements ranged from questionnaire responses to estimates of clothing insulation to specific information about the indoor and outdoor conditions. This research uncovered a number of intriguing trends.
First, Brager and de Dear found that occupants of mechanically ventilated buildings were twice as sensitive to temperature changes as those in naturally ventilated buildings. “Such a finding suggests that people in air-conditioned buildings have higher expectations for thermal consistency,” they explain, “and quickly become critical if thermal conditions diverge from these expectations.” In other words, we can become addicted to air conditioning. “In contrast, people in naturally ventilated buildings seem to demonstrate a preference for a wider range of thermal conditions,” they add.
Brager and de Dear also found that while the predictions of comfort that are used in Standard 55–1992 have worked well for mechanically conditioned buildings, they have been woefully inaccurate when applied to naturally ventilated buildings. Thermal satisfaction in naturally ventilated buildings, they found, “varies widely from predictions made by the present laboratory-based standard.” Something was missing. “Probably about 50% of the variance is due to measurable factors such as changes in clothing and air velocity,” Brager estimates. “The other half comes from changes in expectations,” or psychological adaptation.
Brager has another bone to pick with Standard 55: “It says that a space is acceptable if 20% of the people are complaining,” Adopting the adaptive comfort theory could help improve this situation. “The only way you can raise the bar,” she says, “is by giving people personal control.”
ASHRAE Standard 55–2004
Revised roughly once each decade since its release in 1966, Standard 55 was updated in June 2004. Standard 55–2004 represents a major shift in our understanding of thermal comfort. “The old standard assumed that any person exposed to a given set of conditions responded the same way, regardless of the context,” Brager told EBN. The adaptive comfort option in the new version, according to Brager, “allows for a wider range of conditions that would float with outdoor temperatures.” It can be applied to whole buildings or to portions of buildings, and it is optional—designers may rely on the more familiar model even for naturally ventilated spaces if they wish.
This new method for determining acceptable thermal conditions applies to a limited range of situations. First, the optional method applies only to spaces where the occupants are nearly sedentary and have the freedom to adapt their clothing to the thermal conditions. The thermal conditions of the space must be regulated primarily through the opening and closing of windows; the windows must open to the outdoors and must be readily operable and adjustable by the occupants. Mechanical ventilation is allowed, but only in situations where windows provide the primary means of regulating the thermal conditions of the space. Mechanical heat is also allowed, but the adaptive comfort method applies only when the heating system is not in operation.
The new method does not apply, however, to any space that has a mechanical cooling system, even during times when that system is not being used. That means that the new method does not apply to mixed-mode spaces, where natural ventilation is used during swing seasons but supplanted by mechanical air-conditioning during the summer. While this may appear overly restrictive, some evidence suggests that it may be appropriate. According to Peter Alspach, an engineer with Arup in San Francisco, a U.K. study found that mixed-mode buildings rate lower in occupant satisfaction than either fully naturally ventilated or fully mechanically conditioned buildings. One theory for this result is that occupants of mixed-mode spaces come to expect the consistent conditions provided by the mechanical system and are then dissatisfied when windows are open and conditions fluctuate more. “Over time I would like to see the applicability of the adaptable comfort model expanded,” Brager told EBN, “but I was not arguing hard for that with the 2004 version. It’s important to move slowly and to get more data on those building types.”
Figure 2 shows the acceptable operative temperature ranges for naturally conditioned spaces, as laid out in Standard 55–2004. Note that the range of both acceptability and preference rises in accordance with outdoor air temperature. Since it is based on field studies, not laboratory simulations, the adaptive comfort model automatically takes into account factors such as humidity, air velocity, and occupants’ clothing. “The adaptive model is much simpler,” Brager said, “because it doesn’t have so many inputs.”
Not everyone agrees that this simpler approach is dependable, however. Architect and engineer Dan Nall of Flack & Kurtz Consulting Engineers in New York City notes that Brager and de Dear’s data doesn’t adequately cover temperate humid climates (including the United States’ Eastern Seaboard and Midwest). He argues that factors such as humidity and available breezes need to be factored into the equation, since under humid or still air conditions opening windows won’t do much to make occupants more comfortable.
Modifying Standard 55 to include even an optional adaptive comfort model proved a long and trying process. “It was hard for some people to get their minds around the adaptive concept,” Brager told EBN. “Anything that would promote designing a building without mechanical systems is a bit threatening, or uncomfortable, because you don’t have control.” Alison Kwok, of the University of Oregon, put the problem more directly: “This is the HVAC industry defining comfort in naturally ventilated buildings—see the irony?”
Naturally ventilated buildings are hard to argue with—everyone wants operable windows and a connection to the outdoors—but in most North American climates, and in most cities anywhere, getting them to work is a challenge. Relying on cross-ventilation is not easy, says Grondzik: “What occurs in the building is complicated. It’s hard to achieve what the magic arrows in the diagrams indicate. You also need to deal with dust, pollen, and noise issues.” For commercial buildings, fire code requirements for smoke control add another challenge.
In the eastern half of North America, hot, humid conditions make it impossible to achieve even the adaptive-comfort-prescribed conditions without some mechanical cooling. The adaptive comfort model is applicable even in humid climates, says Guy Battle of the U.K. firm Battle McCarthy, “but it won’t necessarily reduce your peak load and allow you to downsize your equipment.” Nevertheless, Battle notes, in most of the Eastern U.S. there are significant shoulder seasons in which natural ventilation and adaptive comfort can be used.
Battle is disappointed that the adaptive model in Standard 55 is limited to strictly naturally ventilated spaces. “It seems that it should apply to mixed-mode spaces,” he says, adding, “Even in the U.K., we have done only one fully naturally ventilated building.”
In spite of its restricted applicability, the new version of Standard 55 should profoundly affect the design of naturally ventilated buildings and their occupants. “It’s a very radical change,” says Brager. ASHRAE’s adoption of the adaptive comfort model should bring greater legitimacy to naturally ventilated buildings and result in greater reliance on natural ventilation. This, in turn, should yield energy savings, as mechanical systems are relied on less.
For buildings with conventional HVAC systems, the release of Standard 55–2004 won’t mean much—at least initially. Recognition of adaptive comfort could eventually change the way comfort is determined in all buildings, however. If the new model proves effective in naturally ventilated spaces, a future version of Standard 55 could include expanded comfort zones for mixed-mode spaces or spaces in which occupants have control, even if that control is over mechanical rather than natural ventilation. An increasingly common example of such a space is an office with raised-floor air distribution and a diffuser at each workstation.
Johnson Controls, Inc. uses this theory to promote their Personal Environments® systems, or “environmentally responsive workstations” (ERWs)—systems that offer individuals control over temperature, airspeed, lighting, and acoustics. They cite Rensselaer Polytechnic Institute’s 1992 case study of West Bend Mutual Insurance Company in West Bend, Wisconsin, which found that workers in a new building equipped with ERWs had 16% higher productivity. “Our best estimate is that ERWs were responsible for an increase in productivity of about 2.8%,” according to the researchers.
Even with the new adaptive thermal comfort standard, Byron Stigge of Buro Happold, who practiced in the U.K. and now works in New York, despairs of getting more attention on optimized system design. “In theory, we love the adaptive thermal comfort method and have used it plenty in the U.K. Realistically, in the U.S. we have really struggled with clients and architects,” he says. “We find clients requesting indoor design conditions which are fixed for summer and for winter. Their perspective typically is that they will operate the building as they see fit, but they need the ability to cool and heat spaces during peak conditions.”
In the rush to complete a schematic design, “important things like comfort are often sidelined,” Stigge says. When it comes to setting goals, “it takes 30 seconds to write down your indoor design criteria of 75°F (24°C), as opposed to a three-day ‘kumbaya’ session with the client where you convince them to take their shirts off. That’s how a lot of people see it,” he concludes. Once buildings are designed around the principles of adaptive comfort, though, occupants are likely to respond favorably.
Beyond Standard 55
Although ASHRAE is the most common standard for thermal comfort in the U.S. and Canada, some designers have long been acknowledging human adaptation in their design of naturally ventilated buildings, without the blessing of ASHRAE.
Victor Olgyay may have been the first to link indoor human comfort with outdoor climate, in his now-classic 1963 book Design with Climate: A Bioclimatic Approach to Architectural Regionalism. Olgyay’s model, based on studies undertaken in a moderate American climate, defines a comfort zone for a person at rest, with the variables of temperature and relative humidity (see a simplified version of his bioclimatic chart, Figure 3). Some designers still rely on this model, even though “using Olgyay requires designers to stick their necks out,” Grondzik told EBN. “The upcoming change to 55 will allow more people to look at the ASHRAE standard—people who in the past have ignored the standard, referring instead to Olgyay’s work for passive buildings.” Grondzik stopped short of predicting the demise of Olgyay’s model, though. “Olgyay is still the definitive model for naturally ventilated buildings,” he said.
LEED and Thermal Comfort
The U.S. Green Building Council’s LEED® Rating System for New Construction and Major Renovations (LEED-NC) deals with thermal comfort in several ways. The most direct is in Indoor Environmental Quality credit 7.1, “Thermal Comfort: Compliance with ASHRAE 55–1992.” For this point, LEED accepts two different standards for thermal comfort. For mechanically ventilated buildings, LEED currently references Standard 55–1992, as indicated by the credit’s name. For naturally ventilated buildings, however, LEED references Appendix C of the Collaborative for High Performance Schools (CHPS) Best Practices Manual. This Appendix, “A Field-Based Thermal Comfort Standard for Naturally Ventilated Buildings,” by Brager and de Dear, lays out the adaptive comfort model on which ASHRAE Standard 55–2004 was based.
This two-pronged approach was introduced with LEED-NC v2.1, in May 2003. Prior to that modification, all buildings were expected to comply with Standard 55–1992 in order to achieve EQc7.1. According to Hernando Miranda, vice chair of the EQ technical advisory group (TAG), the change was made because the TAG was concerned that, “since ASHRAE 55 was developed to address mechanically conditioned buildings, not naturally ventilated ones,” credit 7.1 had a “mechanically conditioned bias.”
Though naturally ventilated buildings now have an easier time achieving EQc7.1, it’s worth noting that EQc7.2, “Thermal Comfort: Permanent Monitoring System,” is out of their reach. This point can be achieved only through the incorporation of active user controls over temperature and humidity, and “because humidity control is not achievable in naturally ventilated buildings,” the Reference Package explains, they are ineligible. In a separate credit, Indoor Environmental Quality credit 6, LEED recognizes the benefits of occupant control by providing a point for access to operable windows and lighting controls (in perimeter spaces), and another point for individual control of airflow, temperature, and lighting (in nonperimeter spaces).
Thermal comfort is also a factor in Energy & Atmosphere credit 1, “Optimize Energy Performance,” which accounts for far more points than any other LEED credit. To prevent designers from claiming extra energy savings by using nonstandard set-points for indoor temperatures, LEED defines allowable temperature ranges. LEED’s “bounding comfort parameters” are reasonably consistent with ASHRAE Standard 55’s conventional guidelines, although they don’t account for humidity or other factors. They are more restrictive than the adaptive thermal comfort option in Standard 55–2004, however. For example, LEED allows indoor temperatures to exceed 85°F (29°C) no more than 20 hours each year, while the adaptive model indicates that 90% of users near operable windows will be comfortable at temperatures up to 87°F (31°C), as long as outdoor temperatures are higher than 92°F (33°C).
It remains unclear when and how USGBC will incorporate the new version of ASHRAE Standard 55 into LEED. The 2004 version of Standard 55 will likely replace the 1992 version for use in mechanically ventilated buildings—this change could be expected with the next update to LEED. If Standard 55–2004 replaces CHPS Appendix C for use in naturally ventilated buildings, however, some freedom will be lost, since ASHRAE’s definition of a naturally ventilated building excludes mixed-mode spaces (those that can be mechanically cooled if necessary), while the CHPS version does not.
Although the long-established components of comfort—air temperature, mean radiant temperature, humidity, air velocity, clothing insulation, and activity level—all play critical roles in determining the human response to environments, they fail to tell the whole story. The theory of adaptive thermal comfort completes the comfort puzzle by taking into account behavioral, physiological, and psychological adaptation.
Wise is especially optimistic: “If one builds into the situation a variety of opportunities for occupants to effectively cool or warm themselves in ‘adaptive’ ways, people will do so.” That, in turn, can lead to tremendous operational cost savings and energy savings in addition to greater occupant comfort. “If the adaptive comfort movement leads to a little bit more sanity in how companies handle employees’ thermal comfort adaptations, I’m all for it,” he says.
– Jessica Boehland with