Measured thermal comfort and sensation in highly transient environments.
Human thermal sensation and comfort are important topics in the design and operation of occupied spaces. Steady state thermal sensation and comfort has been widely studied and has robust predictive models available for designers. The thermal sensation and comfort responses to transient conditions have received much less attention. The human body rarely reaches steady state in daily activity and in many situations can experience rapid and large transient changes. For example, people moving in and out of transportation environments often experience large temperature variations. In many cases, good experimental data sets are difficult to obtain to validate modeling efforts. In this light, an experimental study documenting transient thermal responses with a number of measured physiological parameters was undertaken to fill this void. This paper reports on the transient thermal comfort and sensation reported by subjects while playing a video game. The video game is designed to access additional performance measures from the subjects and additional information can be found in Young et al. (2011).
Transient environments more accurately represent the thermal stresses a human must respond to each day. These changes in environmental temperatures require both behavioral and physiological accommodations to achieve a certain level of comfort. This has been documented in a number of experimental studies on transient thermal environments (Arens et al. 2006a, Chen et al. 2011, Guan et al. 2003a,b, Liu et al. 2014, Nagano et al. 2005, Zhang et al. 2004, Zhang et al. 2010a,b,c, Zhang et al. 2014). In these studies, thermal sensation, thermal comfort, or both were recorded from subjects with the use of questionnaires throughout the experiments. Thermal comfort is defined as 'that condition of mind which expresses satisfaction with the thermal environment' (ASHRAE 1966). "Thermal sensation is related to how people 'feel' and is therefore a sensory experience and a psychological phenomenon. It is not possible to define sensation in physical or physiological terms." (Parsons 2014, 83) It is important to understand both the definitions and scales used by the researchers. Chen et al. (2011), Liu et al. (2014), Nagano et al. (2005), and Zhang et al. (2014) all utilized the ASHRAE 7-point sensation scale. The scale ranges from -3 to +3 (cold, cool, slightly cool, neutral, slightly warm, warm, and hot). Arens et al. (2006ab), Zhang et al. (2004), and Zhang et al. (2010abc) used an extended ASHRAE 7-point scale and added +4 as "very hot" and -4 as "very cold" to accommodate extreme environments. Based on 109 human tests performed under non-uniform and transient conditions, Zhang et al. (2004) developed a predictive model of overall thermal comfort (BTCM), and presented the Berkeley comfort scale. "The comfort scale ranges from 'just comfortable' (+0), to 'comfortable' (2), to 'very comfortable' (4), 'just uncomfortable' (-0), to 'uncomfortable' (-2), to 'very uncomfortable' (-4). This comfort scale differs from that of most previous thermal comfort research in that it differentiates levels of comfort on the positive side as well as the negative side. The usual scale range includes 'intolerable', 'very uncomfortable', 'uncomfortable', 'slightly uncomfortable', and 'comfortable' " (Arens et al. 2006a).
Subjective sensation and comfort measurements should ideally be coupled with documentation of physiological responses to the thermal environments. The measured physiological responses are important when attempting to correlate sensation and comfort. Researchers have measured responses such as skin temperature, core temperature, heart rate, and sweat rate. Skin temperature is most commonly measured with the use of thermocouples. Core temperature was measured by few comfort researchers throughout the literature. Arens et al. (2006ab) and Zhang et al. (2010abc) used an ingestible thermometer pill to measure intra-abdominal core temperature. Tympanic measurements were used in the current study. Sweat is important in hot transient environments where sweating disturbs the equilibrium between moisture content of the clothing and relative humidity of the air (Havenith et al. 2008). Chen et al. (2011) tracked skin moisture levels through the use of a TEWL probe and a flat-head moisture monitor. Sweat rate was documented in the current study by weighing both subjects and clothing before and after the experiment and accounting for water consumed and expelled.
Analysis of transient environments is more complicated because rate of change of temperature, thermal overshooting, and lagging all play a role in the subjects' comfort. In the classic paper by Gagge et al. (1967) it was discovered that thermal sensation overshoots in the down-step conditions from warm to neutral environments and lags in the up-step conditions from neutral to warm environments. Guan et al. (2003b) further explains that this is partially due to the body's response to rate of heat gain. Positive heat gain indicates that the body is warming up and negative heat gain indicates that the body is cooling down. The human brain can ignore if body temperatures are warm or cool by responding to the rate of heat gain headed in the direction necessary to correct its condition. Guan et al. (2003b) defines this as "second-order anticipation effect". Thermal sensation is then attributed to cumulative heat gain, the rate of heat gain, and the change in rate of heat gain. "It is through this effect that we can have thermal sensations cooler than neutral even though the body is in a thermal state warmer than neutral and vice versa" (Guan et al. 2003b). For example, in an experimental case of air temperature dropping from 32 to 24[degrees] C (89.6-75.2 F), sensation of subjects initially dropped to levels lower than what was observed later when the thermal sensation vote returned to a steady-state, indicating the occurrence of a cold sensation overshot (Chen et al. 2011). This phenomenon is unique to transient environments and has been noted by others (Arens et al. 2006ab, Guan et al. 2003b, Liu et al. 2014, Nagano et al. 2005, Zhang et al. 2010c). Arens et al. (2006b) observed that the overshoot for comfort was more pronounced than that for sensation.
The following sections document the methodology used in the current study. This is followed by a presentation of the results and a discussion highlighting the significant findings.
The following research used two environmental chambers at the Institute for Environmental Research at Kansas State University. The primary chamber (5.5 x 7 x 3.8 m) (20 x 23 x 12 ft) was equipped with a video game, desk, and chair. The second chamber was a preconditioning chamber which contained dressing rooms and instrumentation equipment. The environmen for these two rooms can be individually controlled.
The preconditioning chamber was set to 25[degrees]C (77 F). The temperature in the primary chamber was set to follow a transient profile during the measurement phase of the experiments as shown in Figure 1. The first 25 minutes of the test increased the temperature from 25[degrees]C (77 F) to 40[degrees]C (104 F). Then, temperature remained constant until the subject finished the first three (out of five) levels of the video game. At this point the temperature decreased back towards 25[degrees]C (77 F) for the remainder of the game. The exact final temperature of the chamber depended on the amount of time required for the subject to complete the final two levels.
The relative humidity for the chamber was set at a constant 55%. Figure 1 also shows that some subjects experienced transient profiles with significant anomalies.
A total of 31 healthy males ranging in age from 19 to 34 participated in the experiments. Subjects weighed between 65-100 kg (143-220 lb) and had a height between 1.70-1.95 m (5.58-6.40 ft). Subjects were free of chronic disease with no history of heat-related illness, respiratory illness, skin disorder, or known allergy to adhesive tape. Subjects were required to refrain from the use of caffeine, nicotine, and alcohol for at least 24 hours prior to the experiment.
Upon entering the preconditioning chamber, subjects were sized for the correct KSU ensemble which consisted of a blue oxford shirt and khaki pants. The ensemble has an intrinsic insulation value of 0.70 clo and an intrinsic evaporative resistance value of 18.74 Pa x [m.sup.2]/W (1.24 psi x [in.sup.2] x hr/Btu). Before they changed into the ensemble, subjects were weighed in a pair of athletic shorts and the KSU ensemble was weighed. This step was repeated following the experiment. Next, the subjects sat in a chair as skin temperature thermocouples were positioned on the body. Temperature was measured in seven locations: forehead, right scapula, right upper chest, right upper arm, right lower arm, right anterior thigh, and right calf. A heart rate monitor was also situated with a chest strap on the subject.
Before the experiment began, subjects were briefed and instructed to drink 236 ml (1 cup) of water. They then entered the preconditioning chamber which was set to 25[degrees]C (77 F). Subjects were weighed along with their KSU ensemble clothing. Once instrumentation of the subject was complete, the 45 minutes of preconditioning continued as they moved into the primary chamber. After this time was reached, subjects completed a resting oxygen consumption and metabolic rate analysis with tympanic temperature measured before and after. An average value across all subjects was 1.19 METs. Next, instructions were given for the video game that subjects played for the duration of the test. The video game consisted of 5 levels. Once the game began, the chamber temperature increased from 25[degrees]C (77 F) to 40[degrees]C (104 F) as seen in Figure 1. Then, temperature remained constant until the subject finished level 3. At this point the temperature began to decrease at the same rate until 25[degrees]C (77 F) was reached. The final temperature of the chamber depended on the amount of time required for the subject to complete the final two levels but as seen in Figure 1 the profile is nearly symmetrical for most subjects. At the end of each level, tympanic temperature was measured and subjects indicated their thermal sensation and comfort through a questionnaire. After completion of the final level, subjects completed a final resting oxygen consumption and metabolic rate analysis followed by a final tympanic temperature reading. An average value across all subjects was 1.14 METs. Finally, subjects were weighed separately from their KSU ensemble clothing and debriefed.
A data acquisition system was used to measure 7 skin temperatures, two dry bulb temperatures and a dew point temperature in the chamber. This system reads, displays, and stores each of the instrument readings during testing. Dry bulb temperatures were measured with type K thermocouples and skin temperatures were measured with type T thermocouples. The dew point temperature was measured with a hygrometer. The subject's tympanic temperature was measured using a hand-held infra-red (IR) thermometer before and after the completion of each level. Heart rate was measured with a typical heart rate chest strap. Oxygen consumption and metabolic rate was measured with a Metabolic Measuring System.
The environmental conditions in the chamber were set by two primary variables: the dry bulb temperature and the wet bulb temperature. Little to no radiant load was present during the test so the mean radiant temperature will not be documented. Whole body thermal comfort and sensation were reported according to the Berkeley scales as defined in the introduction.
Results and Discussion
Four primary variables are presented in this section: chamber air temperature, area-weighted average skin temperature, sensation votes, and comfort votes. Sensation and comfort votes were taken at the end of each level of the video game while the chamber air temperature and skin temperature were monitored continuously. Subjects paused playing the game at the end of each level while the tympanic temperature was measured then subjects voted on sensation and comfort. Although not as desirable as a truly symmetric time and temperature profile, this method of evaluation was required as part of the evaluation built into the video game. Figure 2 displays sensation vote for each subject broken by game level.
Data. The chamber temperature profile was delayed for subject 7. Subject 4 skipped the quiz at the end of level 2. As a result, there was no time stamp for conclusion of level 2 and start of level 3. This subject was excluded from Figure 3 for that reason. Subject 24's chamber air temperature increased above 40[degrees]C (104 F). Subject 24 was included in Figures 1, 2, and 3. Subject 24's level one data was sufficient, but all other level data were removed from Figures 4, 5, and 6 due to the overshoot in chamber temperature. The sensation vote seen on the right axis of Figure 2 provides interesting information. For example, at the end of level 1, only one subject voted neutral and the rest were slightly warm or above. The highly transient nature of the responses can be seen by looking at how the data changes as levels increase. Figure 3 presents the are-aweighted average skin temperature vs. time and sensation vote for the subjects. There was a wide variation in skin temperature for each level. For example, skin temperatures at the beginning of level one ranged from about 32 to 34.5[degrees]C (89.6 to 94.1 F). Subject 22 also had cooler skin temperatures. Reasoning behind this is unclear. This subject was included in Figures 1, 2, and 3 and excluded from Figures 4, 5, and 6.
Additional analysis is needed to recognize the effect of transient conditions on the sensation and comfort votes. Figure 4 shows the transient effect on sensation vote by grouping the data according to rising and falling chamber temperatures. The lines are a linear fit to the data and are included for illustrative purposes and are not predictive in nature. Although one must be careful of drawing firm conclusions without statistical analysis, certain features can be identified in Figure 4. For example, between an average skin temperature of 34[degrees]C (93.2 F) and 34.5[degrees]C (94.1 F), all subjects experiencing rising temperatures voted slightly warm or above while most subjects experiencing falling temperatures voted neutral or cooler. This highlights the value of measuring skin temperature while taking subject sensation votes. Figure 5 shows comfort votes in the transient environment. In contrast to Figure 4, no clear difference exists in comfort votes during rising and falling chamber temperatures. Figure 6 highlights an interesting occurrence in the data by plotting area-weighted average skin temperature vs. chamber air temperature. Subject skin temperatures are clearly lower during rising chamber temperatures as compared to falling temperatures. This demonstrates the importance of measuring skin temperature, chamber air temperature, and the rate of change.
Figure 4 highlights that the rate of change plays an important role in sensation prediction. Arens et. al (2006) found that thermal comfort overshoot was more pronounced than that for sensation. For the current study, thermal sensation overshoot during whole-body transients was more pronounced than that for comfort. The reason for this difference is not well understood.
The current study provides needed experimental data on transient thermal sensation and comfort. The subjects experienced a rapid but realistic temperature change with the rise and fall both lasting about 20 minutes. Subjects were playing a video game during the study. The results indicated significant overshoot on thermal sensation but found a limited effect on comfort. Subject core and skin temperature was also monitored during the study allowing more detailed comparisons with models.
Arens, Edward, Hui Zhang, and Charlie Huizenga. 2006 a. "Partial- And Whole-Body Thermal Sensation and Comfort--Part I." Journal Of Thermal Biology 31: 53-59.
Arens, Edward, Hui Zhang, and Charlie Huizenga. 2006 b. "Partial- And Whole-Body Thermal Sensation and Comfort--Part II: Non-Uniform Environmental Conditions." Journal Of Thermal Biology 31: 60-66.
ASHRAE. 1966. Thermal Comfort Conditions. New York: ASHRAE Standard 55.66.
Chen, Chen-Peng, Ruey-Lung Hwang, Shih-Yin Chang, and Yu-Ting Lu. 2011. "Effects Of Temperature Steps on Human Skin Physiology and Thermal Sensation Response." Building And Environment 46: 2387-2397.
Gagge, A.P., J.A.J. Stolwijk, and J.D. Hardy. 1967. "Comfort And Thermal Sensations and Associated Physiological Responses at Various Ambient Temperatures." Environmental Research 1 (1): 1-20.
Guan, Yanzheng (Don), Mohammad H. Hosni, Ph.D., Byron W. Jones, Ph.D., P.E., and Thomas P. Gielda, Ph.D. 2003 a. "Investigation Of Human Thermal Comfort Under Highly Transient Conditions for Automotive Applications- Part 1: Experimental Design and Human Subject Testing Implementation." ASHRAE Transactions 109 (Part 2): 885-897.
Guan, Yanzheng (Don), M. H. Hosni, B. W. Jones, and T.P. Gielda, 2003 b. "Investigation Of Human Thermal Comfort Under Highly Transient Conditions for Automotive Applications-Part 2: Thermal Sensation Modeling." ASHRAE Transactions 109 (Part 2): 898-907.
Havenith, George, Mark G. Richards, Xiaoxin Wang, Peter Brode, Victor Candas, Emiel den Hartog, Ingvar Holmer, Kalev Kuklane, Harriet Meinander, and Wolfgang Nocker. 2008. "Apparent Latent Heat of Evaporation from Clothing: Attenuation and 'Heat Pipe' Effects." Journal Of Applied Physiology 104 (no. 1): 142-149.
Liu, Hong, Jianke Liao, Dong Yang, Xiuyuan Du, Pengchao Hu, Yu Yang, and Baizhan Li. 2014. "The Response of Human Thermal Perception and Skin Temperature to Step-Change Transient Thermal Environments." Building And Environment 73: 232-238.
Nagano, Kazuo, Akira Takaki, Megumi Hirakawa, and Yutaka Tochihara. 2005. "Effects Of Ambient Temperature Steps on Thermal Comfort Requirements." Int J Biometeorol 50: 33-39.
Parsons, Ken. 2014. Human Thermal Environments the Effects of Hot, Moderate, and Cold Environments on Human Health, Comfort, and Performance. 3rd. Boca Raton, Fla.: CRC Press/Taylor & Francis.
Young, M. E., Webb, T. L., & Jacobs, E. A. 2011. "Deciding When to 'Cash in' When Outcomes Are Continuously Improving: An Escalating Interest Task." Behavioural Processes 88: 101-110.
Zhang, H., C. Huizenga, E. Arens, and D. Wang. 2004. "Thermal Sensation and Comfort in Transient Non-Uniform Thermal Environments." Eur. J. Of App. Physiology 92 (6): 728-733.
Zhang, Hui, Edward Arens, Charlie Huizenga, and Taeyoung Han. 2010 a. "Thermal Sensation and Comfort Models for Non-Uniform and Transient Environments: Part I: Local Sensation of Individual Body Parts." Building And Environment 45: 380-388.
Zhang, Hui, Edward Arens, Charlie Huizenga, and Taeyoung Han. 2010 b. "Thermal Sensation and Comfort Models for Non-Uniform and Transient Environments, Part II: Local Comfort of Individual Body Parts." Building And Environment 45: 389-398.
Zhang, Hui, Edward Arens, Charlie Huizenga, and Taeyoung Han. 2010 c. "Thermal Sensation and Comfort Models for Non-Uniform and Transient Environments, Part III: Whole-Body Sensation and Comfort." Building And Environment 45: 399-410.
Zhang, Wencan, Jiqing Chen, and Fengchong Lan. 2014. "Experimental Study on Occupant's Thermal Responses under the Non-Uniform Conditions in Vehicle Cabin during the Heating Period." Chinese Journal Of Mechanical Engineering 27 (2): 331-339.
Michael Young, Ph.D.
Steven Eckels, Ph.D.
Caption: Figure 1: Chamber air temperature vs. time for each subject
Caption: Figure 2: Chamber air temperature vs. time and sensation vote
Caption: Figure 3: Area-weighted average skin temperature vs. time and sensation vote
Caption: Figure 4: Sensation votes grouped by rising chamber temperature (levels 1,2,3) and falling temperature (levels 3,4,5) vs. area-weighted average skin temperature with a 95% confidence interval
Caption: Figure 5: Comfort votes grouped by rising chamber temperature (levels 1,2,3) and falling temperature (levels 3,4,5) vs. area-weighted average skin temperature with a 95% confidence interval
Caption: Figure 6: Area-weighted average skin temperature grouped by rising chamber temperature (levels 1,2,3) and falling temperature (levels 3,4,5) vs. chamber air temperature with a 95% confidence interval
|Printer friendly Cite/link Email Feedback|
|Author:||Eckels, Erin; Young, Michael; Schlabach, Meredith; Eckels, Steven|
|Date:||Jan 1, 2017|
|Previous Article:||Energy demands for commercial buildings with climate variability based on emission scenarios.|
|Next Article:||Predicting envelope and micro cogeneration design conditions for future climates.|