- Email: [email protected]
An investigation of thermal comfort inside an automobile during the heating period
ARTICLE IN PRESS
Applied Ergonomics 36 (2005) 301–312 www.elsevier.com/locate/apergo
An investigation of thermal comfort inside an automobile during the heating period Omer Kaynakli, Muhsin Kilic Department of Mechanical Engineering, Faculty of Engineering and Architecture, Uludag University, 16059 Bursa, Turkey Received 3 March 2004; accepted 6 January 2005
Abstract This paper describes a combined theoretical and experimental study of thermal comfort during the heating period inside an automobile. To investigate the effects of thermal conditions on the human physiology and thermal comfort during the heating period, temperature, humidity and air velocity were measured at a number of points inside the automobile, so thermal conditions were accurately determined. The human body was divided into 16 sedentary segments, and the change of temperature was observed both experimentally and theoretically. During transient conditions of the heating period, heat and mass transfer between the human body and the interior environment of an automobile were simulated by a computational model, and predictions were compared with the measured data. It is shown that there is a good agreement between the model predictions and experimental results. By means of the present model, the effects of the fast transient conditions of the heating period on the sensible and latent heat transfer from the body, body segments skin temperatures and thermal sensation were investigated in detail. r 2005 Elsevier Ltd. All rights reserved. Keywords: Automobile; Heating period; Thermal comfort; Thermal sensation
1. Introduction The passenger compartment of an automobile is heated in the winter months by circulating hot engine coolant through a coolant-to-air heat exchanger that warms the compartment’s air. The heating system is designed to operate in conjunction with the air ventilating system to provide the desired air temperature. With progressive reductions in engine size, stemming from considerations of fuel economy, and corresponding reductions in the heat available for the passenger heating system, there is interest in the development of more effective systems to ensure passengers thermal comfort even in extreme conditions by considering market situation. It is difficult to achieve and maintain passenger thermal comfort under extremely cold driving Corresponding author. Tel.: +90 224 4429183; fax: +90 224 4428021. E-mail address: [email protected] (M. Kilic).
0003-6870/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.apergo.2005.01.006
conditions. Some auxiliary heating or cooling apparatus may greatly reduce the time needed to attain thermal comfort. But, power requirements associated with this apparatus are substantial. In the highly cold winter season, the heating period from the start up of the vehicle requires some time to reach steady-state conditions. During this period, conditions are highly nonuniform over the body of the occupant. The vehicle passenger experiences localized chilling due to contact with an initially cold seat or steering wheel, and nonuniform air velocities that vary depending on the location of the air registers and dashboard control settings. Thus, in addition to the air temperature, several other factors have a bearing on the thermal comfort of the passenger. Until the thermal comfort reached in the automobile compartment, the temperature and the humidity changes dramatically. Driver and passengers are greatly affected by these changes. This problem is needed to be solved with respect to human comfort, health and driving safety.
ARTICLE IN PRESS 302
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
Nomenclature surface area, m2 specific heat, J/(kg K) convective heat loss rate, W/m2 cold signal evaporative heat loss rate, W/m2 correction factor heat transfer coefficient, W/(m2 K) segment number air or fabric layers number conductive heat transfer coefficient, W/(m K) thermal load, W/m2 Lewis ratio, 1C/kPa body mass, kg mass flow, kg/(s m2) rate of metabolic heat production, W/m2 number of layers covering segment water vapor pressure, kPa heat transfer rate, W/m2 outer radius of fabric layer thermal or evaporative resistance, (m2 K)/W or (m2 kPa)/W RH relative humidity S heat storage, W/m2 t time, s (unless specified in minutes) T temperature, 1C TS thermal sensation V air velocity, m/s w skin wettedness W humidity ratio, kgH2O/kg dry air WSIG warm signal ^ W external work accomplished, W/m2 x thickness, m
A cp C CSIG E f h i j k L LR m _ m M nl p Q r R
Consequently, there is substantial interest in the development of more efficient techniques for achieving and maintaining passenger thermal comfort in an automobile environment. Human thermal comfort has been the subject of considerable previous study, and much of the available information was documented and codified in the literature (see ASHRAE, 1989; Parsons, 1993). Most of the studies have considered the thermal conditions are nearly uniform and steady over the entire body of occupant. Tanabe et al. (1994) investigated sensible heat loss from several parts of the human body by the use of a manikin. For each considered part of the body, total heat transfer coefficient and thermal resistance were found. Since their study was performed in constant temperature environment, it did not give any result about the thermal comfort conditions. The effects of thermal environment on the health, comfort and working efficiency of occupants was discussed separately by
Greek letters a Z
ratio of skin layer mass to total body mass permeation efficiency
Subscripts a act al b bl cd cl cr c dif e ex f int o r res rsw s shiv sk t
air, ambient activity air layer body blood conduction clothing core convection diffusion evaporation exhaled fabric interface between outer clothing surface and a solid (such as the seat or back support) operative radiation respiration due to regulatory sweating sensible shiver skin total
Parsons (2000). The discussion was confined to the factors of heat and cold, vibration, noise, and light. Other environmental factors and combined effects were also briefly considered in that study. Kaynakli et al. (2003a) presented a numerical model of the heat and mass transfer between the human body and the environment. In their study, the required environmental and personal conditions for satisfaction of the people obtained under steady-state conditions, and total sensible and latent heat losses, skin temperature, wettedness, predicted mean vote (PMV) and predicted percentage of dissatisfied (PPD) values were calculated via simulation. Kaynakli et al. (2003b) reported a study in which the human body is divided into 16 sedentary segments, a computational model of thermal interactions between each of the 16 body segments and the environment is developed. By the use of the model, skin wettedness and latent (sweating, diffusion) and sensible (conduction, convection, radiation) heat losses from
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
each body segment and whole body are calculated for both sitting and standing postures. Less attention appears to have been directed to comfort in an automobile, where conditions are highly nonuniform and transient over the body of the occupant. Burch et al. (1991a, b) studied thermal comfort conditions in an automobile under very cold winter season. In their study, the changes of the temperatures of interior and body parts contacted with solid surfaces were investigated during standard heating process in a very cold day (20 1C). During this period, the effects of the heat losses from body by conduction, convection and radiation on the thermal sensation (TS) were investigated. But, heat losses from the body segments, and their skin temperatures were not considered in their study. Chakroun and Al-Fahed (1997) presented a study of the temperature variation and thermal comfort inside a car parked in the sun during the summer months in Kuwait. They also considered the effect of using different combinations of internal covering on the PMV inside the car. Lee and Yoon (1998) investigated that the effects of ventilation mode on the distribution of air temperature and velocity in a 1/10 scale vehicle interior model during heating period experimentally. In the experiments, three different ventilation modes (panel-vent, foot-vent and hybrid-vent) were tested at the same flow rate. Aroussi and Aghil (2000) stressed that the need to improve the climatic comfort within passenger vehicles is critical not only to passenger comfort but also to their safety, and to make progress in this area, a good understanding of the flow behavior within the vehicle is required. But, in the study, the effects of air flow on the human body were not mentioned and any result relevant with the thermal comfort was not reached. Daanen et al. (2003) are experimentally investigated effects of warm, cold and thermoneutral environments on driving performance. They concluded that driving performance was affected from the cold and hot ambient conditions. Jones (2002) compared several thermal sensation model outputs with the measured data for a typical winter automobile warm-up condition and showed that the models differ widely in their predictions. Kaynakli et al. (2002) presented a computational model of heat and mass transfer between a human and the vehicle interior environment during heating and cooling periods. The model is based on the heat balance equation for human body, combined with empirical equations defining the sweat rate and mean skin temperature. Guan et al. (2003a) presented an experimental study to examine human thermal comfort under highly transient conditions in an automobile. They used an environmental chamber to simulate 16 typical winter and summer conditions. Thermal sensation modelling was discussed in their companion paper (Guan et al., 2003b). In their mathematical model, physiological and psychological factors were combined, and environmental and personal
303
parameters were used as inputs to determine the physiological responses. Guan et al. (2003c) presents a literature review on the current advances in thermal comfort modelling for both building and vehicle HVAC applications. This study presents an experimental study and a mathematical model of thermal interactions between a human and the interior environment of an automobile. The human body is divided into 16 sedentary segments, the change of temperature was observed both theoretically and experimentally. The model is based on the heat balance equation for human body, combined with empirical equations defining the sweat rate and mean skin temperature. Simulation has been performed by the use of transient conditions. The effects of the heating process on thermal comfort, with respect to temperature, relative humidity and air velocity inside the automobile are investigated. The details of the mathematical model are presented, and predictions obtained by the model are compared with experimental results in order to validate the present model.
2. Mathematical modelling The human body generates heat, which covers a wide range, depending upon activity. Heat generated in the body for heavy work is approximately 10 times higher than resting situations (ASHRAE, 1989; Butera, 1998). To continue vital functions and in addition to that ensure comfort conditions, the heat generation of the human body must be transmitted to the environment. Thermal interaction of the human body with the environment for the core and the skin compartments can be written as follows: ^ Qres Qcr;sk , S cr ¼ M W
(1)
S sk ¼ Qcr;sk ðQcd þ C þ R þ E sk Þ,
(2)
^ is where M is the rate of metabolic heat production, W the rate of external work done by the muscles, Qres is the total rate of heat loss through respiration, Qcr,sk is the rate of heat exchange between the core and skin, Esk is the total rate of evaporative heat loss from skin. The total rate of sensible heat loss from skin can be divided to three parts namely conduction (Qcd), convection (C) and radiation (R) heat loss rates. Heat storage rate of the core and skin causes instantaneous temperature changes in each compartment. This effect may be expressed as follows: dT cr =dt ¼ S cr Ab =ðð1 aÞmcp;b Þ,
(3)
dT sk =dt ¼ S sk Ab =ðamcp;b Þ,
(4)
where a is the fraction of body mass concentrated in skin compartment, m is the mass of body, cp,b is the specific
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
304
heat of body, Tcr and Tsk are the core and skin temperature of the body, respectively. Convective and radiative heat loss from clothed body to environment is calculated as CþR¼
T sk T o , Rcl þ 1=½ðhc þ hr Þf cl
(5)
where Rcl is thermal resistance of clothing, fcl is the ratio of clothed to nude body area, hc and hr are the convective and radiative heat transfer coefficients, respectively. Operative temperature (To) can be defined as hr T¯ r hc T a To ¼ . hr þ hc
(6)
The radiative heat transfer coefficient (hr) was assumed to be 4.7 W/m2 1C for typical clothing systems and convective heat transfer coefficient (hc) was determined as follows (ASHRAE, 1989) hc ¼ 8:3V
0:6
.
(7)
In an automobile, a significant portion of the body surface is in contact with an initially cold seat and back support. This portion does not lose heat by convection and radiation. Conductive heat loss is calculated as Qcd ¼
T sk T int , xcl =kcl
(8)
where xcl and kcl are the thickness and thermal conductivity of clothing, respectively. Tint is the surface temperature of solids in contact with the body. The total latent heat loss from the skin (Esk) is given by E sk ¼
wðpsk;s pa Þ , ðRcl =Zcl LRÞ þ ð1=hc f cl LRÞ
(9)
where w is the skin wettedness, psk,s is the saturated water vapor partial pressure at the skin temperature and pa is the water vapor partial pressure in the ambient air, Zcl is permeation efficiency of the clothing and LR is the Lewis ratio which is the ratio of the evaporative heat transfer coefficient to the convective heat transfer coefficient. During respiration, the body losses both sensible and latent heat by convection and evaporation of heat on water vapor from the respiratory tract to the inhaled air. Convective (Cres) and evaporative (Eres) heat losses due to respiration are _ res ½cp;a ðT ex T a Þ þ hfg ðW ex W a Þ=Ab , C res þ E res ¼ m (10) _ res the mass flow rate of air inhaled, Tex and Ta where m are the exhaled air and the ambient air temperatures, respectively. Wex and Wa are the exhaled air and the ambient air humidity ratio, respectively. The heat of vaporization (hfg) is 2.43 106 J/kg.
When the ambient conditions change, body thermoregulatory control mechanisms act to maintain body temperature and heat balance to environment. The rate of heat generation of the body must be equal to the rate of heat loss from it. For this purpose, the body uses thermoregulatory signals. The temperature control signal equations used in this model are taken from Doherty and Arens (1988). The blood flow between the core and the skin can be expressed mathematically as _ bl ¼ ½ð6:3 þ 200WSIGcr Þ=ð1 þ 0:5CSIGsk Þ=3600, m (11) where WSIGcr and CSIGsk are warm signal from the core and cold signal from the skin, respectively. Changes in blood flow influence the relative masses of the skin and the core compartments. This effect can be calculated by _ bl þ 0:585Þ. a ¼ 0:0418 þ 0:745=ð3600m
(12)
Metabolic energy production due to shivering is related to the two signals by the expression: M shiv ¼ 19:4CSIGsk CSIGcr .
(13)
2.1. Thermal and evaporative resistance of clothing and air This model is based on the same approach that was used in the study of Olesen et al. (1988) in which the body was divided into 16 segments. The 16 body parts and their respective surface areas are listed in Table 1. In this model, the dry and evaporative resistance is calculated for each of 16 body segments. These calculated values are used for each body segment, which is treated as concentric cylinders. There are some difficulties in calculating the resistance to dry and evaporative heat transfer of clothing ensembles. The body is divided into 16 segments that are uniformly clothed. Each successive layer has a larger area for heat transfer. The heat flows from the body through alternating clothing and air layers. The total thermal resistance (Rt) and the total evaporative resistance (Re,t) for each segment can be calculated as follows McCullough et al., 1989): rði; 0Þ rði; nlÞ nl X rði; 0Þ rði; 0Þ þ Rf ði; jÞ þ Ral ði; jÞ , rði; j 1Þ rði; jÞ j¼1
Rt ðiÞ ¼ Ra ðiÞ
ð14Þ
rði; 0Þ rði; nlÞ nl X rði; 0Þ rði; 0Þ þ Re;f ði; jÞ þ Re;al ði; jÞ , rði; j 1Þ rði; jÞ j¼1
Re;t ðiÞ ¼ Re;a ðiÞ
ð15Þ
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
305
Table 1 Surface areas of the body segmentsa Body segments
Segment number
Surface area (m2)
Fraction of total body surface area (%)
Left foot Right foot Left fibula Right fibula Left thigh Right thigh Pelvis Head Left hand Right hand Left forearm Right forearm Left upperarm Right upperarm Chest Back Whole body
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0.062 0.062 0.14 0.14 0.16 0.16 0.08 0.18 0.05 0.05 0.062 0.062 0.077 0.077 0.185 0.204 1.751
3.5 3.5 8.0 8.0 9.1 9.1 4.6 10.4 2.9 2.9 3.5 3.5 4.4 4.4 10.6 11.7 100.0
a
Olesen et al. (1988).
where Ra and Re,a are the thermal and evaporative resistance of outer air layer, respectively. Thermal resistance of fabrics (Rf) used was hot-plate data from McCullough et al. (1989). Heat loss from the outer surface exposed to the environment is the result of both convection and radiation, hence the resistance of this layer can be expressed as Ra ¼
1 . hc þ hr
(16)
where the values of a and b are 0.0334 m2 kPa/W and 15 mm, respectively (McCullough et al., 1989). The preceding relationships can be used to determine the total heat loss for the whole body: Qs;sk ¼
16 X
½ðQcd þ C þ RÞsk ðiÞAðiÞ=Ab ;
(20)
i¼1
E sk ¼
16 X
½E sk ðiÞAðiÞ=Ab .
(21)
i¼1
The evaporative resistance of the outer air layer can be determined from the convective heat transfer coefficient and the Lewis ratio, and it can be written as Re;a ¼
1 . hc LR
(17)
Heat flows from the body parts through alternating clothing and air layers. In each air layer, there are parallel paths for dry heat flows, one by conduction through the air and one by radiation between the fabric surfaces. Thus, the thermal resistance of an air layer is given by Ral ¼
1 . hr þ ka =xa
(18)
The values of hr and ka were taken as hr ¼ 4:9 W=m2 1C and ka ¼ 0:024 W=m 1C (McCullough et al., 1989). Similar equations can be written for the evaporative resistance of an air layer. A relationship of the following form was taken as Re;al ¼ a½1 expðxa =bÞ,
(19)
2.2. Prediction of thermal comfort Thermal comfort is related to the physiological responses of people. For this purpose, comfort indices have been developed to quantify the degree of discomfort. To predict it, TS is given by Burch et al. (1991b) as TS ¼ ð0:303 expð0:036MÞ þ 0:028ÞL,
(22)
where L is thermal load on the human body. Thermal sensation is scaled as 5 extremely cold, 4 very cold, 3 cold, 2 cool, 1 slightly cool, 0 neutral, +1 slightly warm, +2 warm, +3 hot, +4 very hot and +5 extremely hot.
3. Experimental setup Although, vehicles have smaller volume than home or office environments, they have greater heat loss or gain. Most automobiles have a heating, ventilation, air
ARTICLE IN PRESS 306
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
conditioning (HVAC), device to control thermal environments of passenger compartment. In general, the HVAC system causes a complicated three-dimensional unsteady turbulent flow and temperature variations in the vehicle interior. The ventilation flow could be affected by several parameters such as the shape and location of the air-vents (inlet and outlet), the flow rate and the disposition of the passengers, etc. Nonuniform air and temperature distribution may cause localized discomfort of the occupants in the vehicle. In order to estimate the thermal comfort level, accurate information on the thermal environment is essential. The present experimental study was performed on a 1991 model Toyota Corona Sedan automobile equipped with a 2000 cc engine (see Fig. 1a). The automobile was instrumented with sensors in the passenger compartment to measure the air temperature, velocity and relative humidity on the locations shown in Figs. 1b and c. Equipment used in temperature, velocity and relative humidity measurements are shown in Fig. 2. A series of tests and measurements were performed in the automobile while parked in cold and cloudy winter conditions in February in Turkey. Air temperature measurement data were collected every 3 s by means of 16 temperature sensors located in
Fig. 2. Experimental apparatus: (a) thermometer system, (b) hygrometer, and (c) anemometer.
Fig. 1. (a) The automobile used in experiments, (b) measurement locations from top view, (c) measurement locations from side view.
the car as shown in Figs. 1b and c. Since the large number of sensors were used in measurements and the data were automatically saved to the hard drive, a data acquisition system integrated with a PC was employed to control the complicated task. Relative humidity and air velocity were measured in 1 min intervals. In order to find out the effects of thermal environments on the human body in the automobile cabin, skin temperatures were measured on the 10 different locations of the body (i.e. foot, fibula, thigh, pelvis, chest, back, head, hand, forearm, upperarm). In order to calculate the conductive heat loss from the body, interface temperatures of the body contact with solid surfaces, which are seat and
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
307
Table 2 Measurement instruments Measuring range
Accuracy
Temperature/humidity measuring: Testo 454 Multi-function measuring instrument Temperature probe 0y+70 1C Humidity probe
0y+100%RH
Velocity measuring: AIRFLOW Developments Limited Velocity probe 0y15 m/s
back support, were also measured during the experiments. Specifications of the equipments used in the experiments are presented in Table 2. In the beginning of each experiment, the human subject entered the automobile cabin where inside temperature was equal to outside ambient temperature and the standard heating system of the automobile was operated. Outside ambient conditions were approximately 0 1C temperature and 73% relative humidity. During the heating period, required measurements were taken in the automobile compartment and on the test subject. Results of the measurements are given in the Experimental results section. In order to determine the value of the experimental error, uncertainty analysis was carried out using equations proposed by Moffat (1988). Maximum uncertainties in experimental results were found to be within 73%. In addition, it must be mentioned that measurements were taken in a parked automobile with only the driver inside. The automobile in motion or in an automobile with passengers may affect the measurements.
70.4 1C (0y+49.9 1C) 70.5 1C (+50y+70 1C) 72%RH (0y+9.9%RH) 71%RH (+10y+90%RH) 72%RH (+90.1y+100%RH) 72% FSD at 20 1C and 1013 mbar
Fig. 3. Comparison of body heat losses in the heating period.
4. Results and discussion In order to validate the mathematical model, the results obtained from the model, are compared with the experimental measurements of Burch et al. (1991a, b). In their experimental study interior air has been heated from 20 to 20 1C in a half-hour time period. They measured air temperature, humidity, mean radiant temperature, seat, back support and steering wheel surface temperatures. By the use of their experimental data, they presented heat loss from the body and TS as shown in Figs. 3 and 4. Figs. 3 and 4 also show the comparisons between the present mathematical model results and the results of Burch et al. (1991b). At the beginning of the heating period, the automobile interior temperature is very low, therefore heat losses are very high, and the thermal sensation value is about 4.5. After the heating system starts to work, heat losses from
Fig. 4. Comparison of thermal sensation during the heating period.
the body decrease with increasing interior temperature, hence the thermal sensation value increases. It can be seen that the agreements between the results are very good. Mathematical model results obtained by the use of the data presented in Burch et al. (1991b): Initial core temperature T cr ¼ 37 1C; initial skin temperature T sk ¼ 34 1C; average thermal resistance of clothing 1.5clo
ARTICLE IN PRESS 308
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
(0.23 m2K/W), metabolic rate due to activity M act ¼ 136 W (75 W/m2), relative humidity 35% and mean radiant temperature T¯ r ¼ 0:94T a 1:38: 4.1. Experimental results Experiments were conducted for the heating period, which started when the interior temperature was about 0 1C. Fig. 5 shows the change of temperatures at front and back sides of the cabin during 1 h time period. The front and back side temperatures are the arithmetic mean values of the measured first eight points (1–8) and the last eight points (9–16) shown in Figs. 1b and c, respectively. The difference between the front and back side temperatures is not too much during all the heating period, the front side temperature is slightly higher than the back side one. It can be also seen from Fig. 5 that the outside temperature stays almost constant about 0 1C. Variation of the air temperatures at vertical locations at front and back side with time during a warm-up period in an automobile is shown Fig. 6. Air temperature difference between the head and foot levels rises up to 5 1C. Air temperatures on foot level are lower than air temperatures on head level both at front and at back sides in the automobile. Variations of the inside air and surface average temperatures and relative humidity with time during a warm-up period are shown in Fig. 7. Since the temperature of the surfaces and materials used in the automobile rises slower than the interior air temperature, there is a difference between the mean temperatures during the heating period. The mean temperature of the surfaces is lower than the interior air mean temperature. Relative humidity decreases with increasing temperature of the air inside the automobile. At the beginning of the heating period relative humidity is about 73%, it drops sharply in 30 min, and then its magnitude stays about 20%.
Fig. 6. Variation of the air temperatures at vertical locations with time during the heating period, (a) at front side, (b) at back side.
Fig. 7. Variation of the average inside air and surface temperatures and relative humidity with time during the heating period.
4.2. Inputs used in automotive application of the model
Fig. 5. Variation of the air temperatures with time during the heating period.
The following empirical relations are derived by the use of present measured data of air temperature, relative humidity and surface temperature inside the automobile, and they are used in the computer program in order to simulate the conditions of the environment and
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
309
passengers during normal warm-up period in winter driving. In the following equations t represents the elapsed time in minutes from start-up. Mean air temperature: ( 1:177 lnðtÞ þ 2:034 Ta ¼ 10:306 lnðtÞ 13:824
for 0otp7; for t47:
(23,24)
Mean radiant temperature: T¯ r ¼ 1:769 expð0:0871T a Þ.
(25)
Mean relative humidity: RH ¼ 73:018 expð0:0257tÞ.
(26)
Surface temperatures of solids in contact with the body (Tint) were: Seat: T int ¼
(
11:253 lnðtÞ þ 22:983 for 0oto3; 1:885 lnðtÞ þ 25:643 for tX3:
Back support: ( 10:729 lnðtÞ þ 22:352 for 0oto3; T int ¼ 2:751 lnðtÞ þ 22:642 for tX3:
Fig. 8. Average body skin temperature in the heating period.
(27,28)
(29,30)
Other constants are used in the mathematical model taken as follows: Average thermal resistance of clothing Rcl ¼ 1clo ð0:155 m2 K=WÞ; metabolic rate due to activity 75 W/m2, the surface area of the nude body 1.751 m2 (see Table 1), ratio of clothed to nude body area f cl ¼ 1:1: Local air velocities on the body of sitting passenger were measured under warm-up conditions, as shown in Table 3. 4.3. Comparisons of the measurements and model results Skin temperature results obtained from the present measurements and simulations are given comparatively in Fig. 8. As it can be seen from Fig. 8, both results are in a good agreement. At the beginning of the heating period, the automobile interior is cold, hence mean skin temperature decreases. Then, mean skin temperature increases with the increasing automobile inside temTable 3 Local air velocities on the body Region
Air velocity (m/s)
Head Trunk Right shoulder Left shoulder Right knee Left knee Right ankle Left ankle
0.25 0.13 0.25 0.35 0.30 0.12 0.60 0.75
Fig. 9. Hand and forearm skin temperatures in the heating period.
perature. Mean skin temperature gives an idea of thermal comfort, but it is not enough to determine comfort level of the occupant. Although the mean body skin temperature is in the required range, the temperatures of the extremities such as hand, foot and face or naked parts of human body may stay on undesired values. For example, the maximum change of temperature takes place in hands, which are unclothed body segments (Fig. 9). At the beginning, the hand skin temperature is 20.4 1C, and then it rises 30 1C with the increasing automobile inside temperature. Pelvis is the least affected segment from the environmental conditions (see Fig. 10). Since the clothing thermal insulation of the pelvis is greater than other body segments, the skin temperature of the pelvis is high and its temperature does not change too much in the heating period. As it can be seen in Fig. 11, the temperature changes of the fibula and thigh are lower than those of the hands. Variations of sensible and latent heat losses from the whole body during the heating period are shown in Fig. 12. In early minutes, convective and radiative heat losses are high since the temperature inside the automobile and the interior surface temperatures are very
ARTICLE IN PRESS 310
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
Fig. 10. Pelvis and chest skin temperatures in the heating period.
through the skin (Edif). Evaporation of sweat secreted due to thermoregulatory control mechanisms (Ersw) is equal to zero because sweat generation does not occur. Considering the Figs. 13–15, the same conclusions can also be drawn for chest, fibula and thigh. Since the hand is naked and the forearm has light clothing, heat loss from these segments rapidly decreases with increasing automobile inside temperature as can be seen from Fig. 13. The pelvis has high clothing thermal insulation according to other body segments, therefore both sensible and latent heat losses are low, as seen in Fig. 14. When both Figs. 15 and 14 are examined together, it can be observed that this segment is not too much affected from the environmental conditions. Generally, latent heat loss is lower than sensible heat loss for all body segments. Variation of TS values in the warm-up period is given in Fig. 16. In the beginning of the heating period, the body heat loss is greater than the generated energy because automobiles, inside temperature is about 0 1C. For this reason, a thermal sensation index has been started from low values (approximately 2). And then,
Fig. 11. Fibula and thigh skin temperatures in the heating period.
Fig. 13. Heat losses from hand and forearm in the heating period.
Fig. 12. Heat losses from the whole body in the heating period.
low. Since temperature difference between the skin and ambient decreases with time, these heat losses decrease, too. Evaporative and respiratory heat losses are not severely affected by the environment and stay at about 15–20 W during the period. Latent heat loss from the skin is caused from the natural diffusion of water
Fig. 14. Heat losses from chest and pelvis in the heating period.
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
Fig. 15. Heat losses from fibula and thigh in the heating period.
311
model, and predictions were compared with the measured data. By means of the present model, the effects of the fast transient conditions of the heating period on the sensible and latent heat transfer from the body, body segments skin temperatures and thermal sensation were also investigated. In view of the fact that the occupant is exposed to highly nonuniform environmental conditions, it is shown that there is a good agreement between the model predictions and experimental results, and the accuracy of the predictions obtained with the all body segments separately and body-averaged results is remarkable. As a result of this study, it is believed that the methodology and the model introduced would enable the designer to test and to better optimize the air conditioning system’s ability to meet comfort requirements, while minimizing the need for expensive and time-consuming experiments.
References
Fig. 16. The change of thermal sensation during the heating period.
thermal sensation has improved with increasing inside temperature and interior surface temperatures.
5. Conclusions In this paper, a combined theoretical and experimental study of thermal comfort inside an automobile during the heating period is presented. The effects of thermal conditions on the human physiology and thermal comfort during the heating period are examined in details. A theoretical simulation model has been developed for automobile interior thermal comfort and evaluation. The human body is divided into 16 sedentary segments, the change of temperature was observed both theoretically and experimentally. Temperature, humidity and air velocity were measured at a number of points inside the automobile, so thermal conditions were accurately determined. During transient conditions of the heating period, heat and mass transfer between the human body and the interior environment of an automobile were simulated by a computational
Aroussi, A., Aghil, S., 2000. Characterisation of the flow field in a passenger car model. Opt. Diagnostics Eng. 4 (1), 1–15. ASHRAE Handbook—Fundamentals, American Society of Heating, Refrigeration and Air-Conditioning Engineers, Atlanta, 1989, p. 29 (Chapter 8). Burch, S.D., Pearson, J.T., Ramadhyani, S., 1991a. Experimental study of passenger thermal comfort in an automobile under severe winter conditioning. ASHRAE Trans. 97, 239–246. Burch, S.D., Ramadhyani, S., Pearson, J.T., 1991b. Analysis of passenger thermal comfort in an automobile under severe winter conditioning. ASHRAE Trans. 97, 247–257. Butera, F.M., 1998. Principles of thermal comfort (Chapter 3). Renewable Sustain. Energy Rev. 2, 39–66. Chakroun, C., Al-Fahed, S., 1997. Thermal comfort analysis inside a car. Int. J. Energy Res. 21, 327–340. Daanen, H.A.M., van de Vliert, E., Huang, X., 2003. Driving performance in cold, warm, and thermoneutral environments. Appl. Ergonomics 34, 597–602. Doherty, T.J., Arens, E., 1988. Evaluation of the physiological bases of thermal comfort models. ASHRAE Trans. 94 (Part 1), 1371–1385. Guan, Y., Hosni, M.H., Jones, B.W., Gielda, T.P., 2003a. Investigation of human thermal comfort under highly transient conditions for automotive applications—Part 1: experimental design and human subject testing implementation. ASHRAE Trans. 109, 885–897. Guan, Y., Hosni, M.H., Jones, B.W., Gielda, T.P., 2003b. Investigation of human thermal comfort under highly transient conditions for automotive applications—Part 2: thermal sensation modeling. ASHRAE Trans. 109, 898–907. Guan, Y., Hosni, M.H., Jones, B.W., Gielda, T.P., 2003c. Literature review of the advances in thermal comfort modeling. ASHRAE Trans. 109, 908–916. Jones, B.W., 2002. Capabilities and limitations of thermal models for use in thermal comfort models. Energy Build. 34, 653–659. Kaynakli, O., Unver, U., Kilic, M., 2002. Simulation of thermal comfort heating and cooling periods in an automobile compartment. Proceedings of the Automotive Technologies Congress, 24–26 June, Bursa, Turkey, pp. 127–135 (in Turkish). Kaynakli, O., Unver, U., Kilic, M., 2003a. Calculation of thermal comfort zones with the ambient parameters. Proceedings of the
ARTICLE IN PRESS 312
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
First International Exergy, Energy and Environment Symposium, 13–17 July, Izmir, Turkey, pp. 769–773. Kaynakli, O., Unver, U., Kilic, M., 2003b. Evaluating thermal environments for sitting and standing posture. Int. Com. Heat Mass Transfer 30 (8), 1179–1188. Lee, S.J., Yoon, J.H., 1998. Temperature field measurement of heated ventilation flow in a vehicle interior. Int. J. Vehicle Design 19 (2), 228–243. McCullough, E.A., Jones, B.W., Tamura, T., 1989. A data base for determining the evaporative resistance of clothing. ASHRAE Trans. 95 (2), 316–328.
Moffat, R.J., 1988. Describing the uncertainties in experimental results. Exp. Thermal Fluid Sci. 1, 3–17. Olesen, B.W., Hasebe, Y., de Dear, R.J., 1988. Clothing insulation asymmetry and thermal comfort. ASHRAE Trans. 94 (1), 32–51. Parsons, K.C., 1993. Human Thermal Environments. Taylor & Francis Ltd., London, UK. Parsons, K.C., 2000. Environmental ergonomics: a review of principles, methods and models. Appl. Ergonomics 31, 581–594. Tanabe, S., Arens, E.A., Bauman, F.S., Zang, H., Madsen, T.L., 1994. Evaluating thermal environments by using a thermal manikin with controlled skin surface temperature. ASHRAE Trans. 100 (1), 39–48.
Applied Ergonomics 36 (2005) 301–312 www.elsevier.com/locate/apergo
An investigation of thermal comfort inside an automobile during the heating period Omer Kaynakli, Muhsin Kilic Department of Mechanical Engineering, Faculty of Engineering and Architecture, Uludag University, 16059 Bursa, Turkey Received 3 March 2004; accepted 6 January 2005
Abstract This paper describes a combined theoretical and experimental study of thermal comfort during the heating period inside an automobile. To investigate the effects of thermal conditions on the human physiology and thermal comfort during the heating period, temperature, humidity and air velocity were measured at a number of points inside the automobile, so thermal conditions were accurately determined. The human body was divided into 16 sedentary segments, and the change of temperature was observed both experimentally and theoretically. During transient conditions of the heating period, heat and mass transfer between the human body and the interior environment of an automobile were simulated by a computational model, and predictions were compared with the measured data. It is shown that there is a good agreement between the model predictions and experimental results. By means of the present model, the effects of the fast transient conditions of the heating period on the sensible and latent heat transfer from the body, body segments skin temperatures and thermal sensation were investigated in detail. r 2005 Elsevier Ltd. All rights reserved. Keywords: Automobile; Heating period; Thermal comfort; Thermal sensation
1. Introduction The passenger compartment of an automobile is heated in the winter months by circulating hot engine coolant through a coolant-to-air heat exchanger that warms the compartment’s air. The heating system is designed to operate in conjunction with the air ventilating system to provide the desired air temperature. With progressive reductions in engine size, stemming from considerations of fuel economy, and corresponding reductions in the heat available for the passenger heating system, there is interest in the development of more effective systems to ensure passengers thermal comfort even in extreme conditions by considering market situation. It is difficult to achieve and maintain passenger thermal comfort under extremely cold driving Corresponding author. Tel.: +90 224 4429183; fax: +90 224 4428021. E-mail address: [email protected] (M. Kilic).
0003-6870/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.apergo.2005.01.006
conditions. Some auxiliary heating or cooling apparatus may greatly reduce the time needed to attain thermal comfort. But, power requirements associated with this apparatus are substantial. In the highly cold winter season, the heating period from the start up of the vehicle requires some time to reach steady-state conditions. During this period, conditions are highly nonuniform over the body of the occupant. The vehicle passenger experiences localized chilling due to contact with an initially cold seat or steering wheel, and nonuniform air velocities that vary depending on the location of the air registers and dashboard control settings. Thus, in addition to the air temperature, several other factors have a bearing on the thermal comfort of the passenger. Until the thermal comfort reached in the automobile compartment, the temperature and the humidity changes dramatically. Driver and passengers are greatly affected by these changes. This problem is needed to be solved with respect to human comfort, health and driving safety.
ARTICLE IN PRESS 302
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
Nomenclature surface area, m2 specific heat, J/(kg K) convective heat loss rate, W/m2 cold signal evaporative heat loss rate, W/m2 correction factor heat transfer coefficient, W/(m2 K) segment number air or fabric layers number conductive heat transfer coefficient, W/(m K) thermal load, W/m2 Lewis ratio, 1C/kPa body mass, kg mass flow, kg/(s m2) rate of metabolic heat production, W/m2 number of layers covering segment water vapor pressure, kPa heat transfer rate, W/m2 outer radius of fabric layer thermal or evaporative resistance, (m2 K)/W or (m2 kPa)/W RH relative humidity S heat storage, W/m2 t time, s (unless specified in minutes) T temperature, 1C TS thermal sensation V air velocity, m/s w skin wettedness W humidity ratio, kgH2O/kg dry air WSIG warm signal ^ W external work accomplished, W/m2 x thickness, m
A cp C CSIG E f h i j k L LR m _ m M nl p Q r R
Consequently, there is substantial interest in the development of more efficient techniques for achieving and maintaining passenger thermal comfort in an automobile environment. Human thermal comfort has been the subject of considerable previous study, and much of the available information was documented and codified in the literature (see ASHRAE, 1989; Parsons, 1993). Most of the studies have considered the thermal conditions are nearly uniform and steady over the entire body of occupant. Tanabe et al. (1994) investigated sensible heat loss from several parts of the human body by the use of a manikin. For each considered part of the body, total heat transfer coefficient and thermal resistance were found. Since their study was performed in constant temperature environment, it did not give any result about the thermal comfort conditions. The effects of thermal environment on the health, comfort and working efficiency of occupants was discussed separately by
Greek letters a Z
ratio of skin layer mass to total body mass permeation efficiency
Subscripts a act al b bl cd cl cr c dif e ex f int o r res rsw s shiv sk t
air, ambient activity air layer body blood conduction clothing core convection diffusion evaporation exhaled fabric interface between outer clothing surface and a solid (such as the seat or back support) operative radiation respiration due to regulatory sweating sensible shiver skin total
Parsons (2000). The discussion was confined to the factors of heat and cold, vibration, noise, and light. Other environmental factors and combined effects were also briefly considered in that study. Kaynakli et al. (2003a) presented a numerical model of the heat and mass transfer between the human body and the environment. In their study, the required environmental and personal conditions for satisfaction of the people obtained under steady-state conditions, and total sensible and latent heat losses, skin temperature, wettedness, predicted mean vote (PMV) and predicted percentage of dissatisfied (PPD) values were calculated via simulation. Kaynakli et al. (2003b) reported a study in which the human body is divided into 16 sedentary segments, a computational model of thermal interactions between each of the 16 body segments and the environment is developed. By the use of the model, skin wettedness and latent (sweating, diffusion) and sensible (conduction, convection, radiation) heat losses from
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
each body segment and whole body are calculated for both sitting and standing postures. Less attention appears to have been directed to comfort in an automobile, where conditions are highly nonuniform and transient over the body of the occupant. Burch et al. (1991a, b) studied thermal comfort conditions in an automobile under very cold winter season. In their study, the changes of the temperatures of interior and body parts contacted with solid surfaces were investigated during standard heating process in a very cold day (20 1C). During this period, the effects of the heat losses from body by conduction, convection and radiation on the thermal sensation (TS) were investigated. But, heat losses from the body segments, and their skin temperatures were not considered in their study. Chakroun and Al-Fahed (1997) presented a study of the temperature variation and thermal comfort inside a car parked in the sun during the summer months in Kuwait. They also considered the effect of using different combinations of internal covering on the PMV inside the car. Lee and Yoon (1998) investigated that the effects of ventilation mode on the distribution of air temperature and velocity in a 1/10 scale vehicle interior model during heating period experimentally. In the experiments, three different ventilation modes (panel-vent, foot-vent and hybrid-vent) were tested at the same flow rate. Aroussi and Aghil (2000) stressed that the need to improve the climatic comfort within passenger vehicles is critical not only to passenger comfort but also to their safety, and to make progress in this area, a good understanding of the flow behavior within the vehicle is required. But, in the study, the effects of air flow on the human body were not mentioned and any result relevant with the thermal comfort was not reached. Daanen et al. (2003) are experimentally investigated effects of warm, cold and thermoneutral environments on driving performance. They concluded that driving performance was affected from the cold and hot ambient conditions. Jones (2002) compared several thermal sensation model outputs with the measured data for a typical winter automobile warm-up condition and showed that the models differ widely in their predictions. Kaynakli et al. (2002) presented a computational model of heat and mass transfer between a human and the vehicle interior environment during heating and cooling periods. The model is based on the heat balance equation for human body, combined with empirical equations defining the sweat rate and mean skin temperature. Guan et al. (2003a) presented an experimental study to examine human thermal comfort under highly transient conditions in an automobile. They used an environmental chamber to simulate 16 typical winter and summer conditions. Thermal sensation modelling was discussed in their companion paper (Guan et al., 2003b). In their mathematical model, physiological and psychological factors were combined, and environmental and personal
303
parameters were used as inputs to determine the physiological responses. Guan et al. (2003c) presents a literature review on the current advances in thermal comfort modelling for both building and vehicle HVAC applications. This study presents an experimental study and a mathematical model of thermal interactions between a human and the interior environment of an automobile. The human body is divided into 16 sedentary segments, the change of temperature was observed both theoretically and experimentally. The model is based on the heat balance equation for human body, combined with empirical equations defining the sweat rate and mean skin temperature. Simulation has been performed by the use of transient conditions. The effects of the heating process on thermal comfort, with respect to temperature, relative humidity and air velocity inside the automobile are investigated. The details of the mathematical model are presented, and predictions obtained by the model are compared with experimental results in order to validate the present model.
2. Mathematical modelling The human body generates heat, which covers a wide range, depending upon activity. Heat generated in the body for heavy work is approximately 10 times higher than resting situations (ASHRAE, 1989; Butera, 1998). To continue vital functions and in addition to that ensure comfort conditions, the heat generation of the human body must be transmitted to the environment. Thermal interaction of the human body with the environment for the core and the skin compartments can be written as follows: ^ Qres Qcr;sk , S cr ¼ M W
(1)
S sk ¼ Qcr;sk ðQcd þ C þ R þ E sk Þ,
(2)
^ is where M is the rate of metabolic heat production, W the rate of external work done by the muscles, Qres is the total rate of heat loss through respiration, Qcr,sk is the rate of heat exchange between the core and skin, Esk is the total rate of evaporative heat loss from skin. The total rate of sensible heat loss from skin can be divided to three parts namely conduction (Qcd), convection (C) and radiation (R) heat loss rates. Heat storage rate of the core and skin causes instantaneous temperature changes in each compartment. This effect may be expressed as follows: dT cr =dt ¼ S cr Ab =ðð1 aÞmcp;b Þ,
(3)
dT sk =dt ¼ S sk Ab =ðamcp;b Þ,
(4)
where a is the fraction of body mass concentrated in skin compartment, m is the mass of body, cp,b is the specific
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
304
heat of body, Tcr and Tsk are the core and skin temperature of the body, respectively. Convective and radiative heat loss from clothed body to environment is calculated as CþR¼
T sk T o , Rcl þ 1=½ðhc þ hr Þf cl
(5)
where Rcl is thermal resistance of clothing, fcl is the ratio of clothed to nude body area, hc and hr are the convective and radiative heat transfer coefficients, respectively. Operative temperature (To) can be defined as hr T¯ r hc T a To ¼ . hr þ hc
(6)
The radiative heat transfer coefficient (hr) was assumed to be 4.7 W/m2 1C for typical clothing systems and convective heat transfer coefficient (hc) was determined as follows (ASHRAE, 1989) hc ¼ 8:3V
0:6
.
(7)
In an automobile, a significant portion of the body surface is in contact with an initially cold seat and back support. This portion does not lose heat by convection and radiation. Conductive heat loss is calculated as Qcd ¼
T sk T int , xcl =kcl
(8)
where xcl and kcl are the thickness and thermal conductivity of clothing, respectively. Tint is the surface temperature of solids in contact with the body. The total latent heat loss from the skin (Esk) is given by E sk ¼
wðpsk;s pa Þ , ðRcl =Zcl LRÞ þ ð1=hc f cl LRÞ
(9)
where w is the skin wettedness, psk,s is the saturated water vapor partial pressure at the skin temperature and pa is the water vapor partial pressure in the ambient air, Zcl is permeation efficiency of the clothing and LR is the Lewis ratio which is the ratio of the evaporative heat transfer coefficient to the convective heat transfer coefficient. During respiration, the body losses both sensible and latent heat by convection and evaporation of heat on water vapor from the respiratory tract to the inhaled air. Convective (Cres) and evaporative (Eres) heat losses due to respiration are _ res ½cp;a ðT ex T a Þ þ hfg ðW ex W a Þ=Ab , C res þ E res ¼ m (10) _ res the mass flow rate of air inhaled, Tex and Ta where m are the exhaled air and the ambient air temperatures, respectively. Wex and Wa are the exhaled air and the ambient air humidity ratio, respectively. The heat of vaporization (hfg) is 2.43 106 J/kg.
When the ambient conditions change, body thermoregulatory control mechanisms act to maintain body temperature and heat balance to environment. The rate of heat generation of the body must be equal to the rate of heat loss from it. For this purpose, the body uses thermoregulatory signals. The temperature control signal equations used in this model are taken from Doherty and Arens (1988). The blood flow between the core and the skin can be expressed mathematically as _ bl ¼ ½ð6:3 þ 200WSIGcr Þ=ð1 þ 0:5CSIGsk Þ=3600, m (11) where WSIGcr and CSIGsk are warm signal from the core and cold signal from the skin, respectively. Changes in blood flow influence the relative masses of the skin and the core compartments. This effect can be calculated by _ bl þ 0:585Þ. a ¼ 0:0418 þ 0:745=ð3600m
(12)
Metabolic energy production due to shivering is related to the two signals by the expression: M shiv ¼ 19:4CSIGsk CSIGcr .
(13)
2.1. Thermal and evaporative resistance of clothing and air This model is based on the same approach that was used in the study of Olesen et al. (1988) in which the body was divided into 16 segments. The 16 body parts and their respective surface areas are listed in Table 1. In this model, the dry and evaporative resistance is calculated for each of 16 body segments. These calculated values are used for each body segment, which is treated as concentric cylinders. There are some difficulties in calculating the resistance to dry and evaporative heat transfer of clothing ensembles. The body is divided into 16 segments that are uniformly clothed. Each successive layer has a larger area for heat transfer. The heat flows from the body through alternating clothing and air layers. The total thermal resistance (Rt) and the total evaporative resistance (Re,t) for each segment can be calculated as follows McCullough et al., 1989): rði; 0Þ rði; nlÞ nl X rði; 0Þ rði; 0Þ þ Rf ði; jÞ þ Ral ði; jÞ , rði; j 1Þ rði; jÞ j¼1
Rt ðiÞ ¼ Ra ðiÞ
ð14Þ
rði; 0Þ rði; nlÞ nl X rði; 0Þ rði; 0Þ þ Re;f ði; jÞ þ Re;al ði; jÞ , rði; j 1Þ rði; jÞ j¼1
Re;t ðiÞ ¼ Re;a ðiÞ
ð15Þ
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
305
Table 1 Surface areas of the body segmentsa Body segments
Segment number
Surface area (m2)
Fraction of total body surface area (%)
Left foot Right foot Left fibula Right fibula Left thigh Right thigh Pelvis Head Left hand Right hand Left forearm Right forearm Left upperarm Right upperarm Chest Back Whole body
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0.062 0.062 0.14 0.14 0.16 0.16 0.08 0.18 0.05 0.05 0.062 0.062 0.077 0.077 0.185 0.204 1.751
3.5 3.5 8.0 8.0 9.1 9.1 4.6 10.4 2.9 2.9 3.5 3.5 4.4 4.4 10.6 11.7 100.0
a
Olesen et al. (1988).
where Ra and Re,a are the thermal and evaporative resistance of outer air layer, respectively. Thermal resistance of fabrics (Rf) used was hot-plate data from McCullough et al. (1989). Heat loss from the outer surface exposed to the environment is the result of both convection and radiation, hence the resistance of this layer can be expressed as Ra ¼
1 . hc þ hr
(16)
where the values of a and b are 0.0334 m2 kPa/W and 15 mm, respectively (McCullough et al., 1989). The preceding relationships can be used to determine the total heat loss for the whole body: Qs;sk ¼
16 X
½ðQcd þ C þ RÞsk ðiÞAðiÞ=Ab ;
(20)
i¼1
E sk ¼
16 X
½E sk ðiÞAðiÞ=Ab .
(21)
i¼1
The evaporative resistance of the outer air layer can be determined from the convective heat transfer coefficient and the Lewis ratio, and it can be written as Re;a ¼
1 . hc LR
(17)
Heat flows from the body parts through alternating clothing and air layers. In each air layer, there are parallel paths for dry heat flows, one by conduction through the air and one by radiation between the fabric surfaces. Thus, the thermal resistance of an air layer is given by Ral ¼
1 . hr þ ka =xa
(18)
The values of hr and ka were taken as hr ¼ 4:9 W=m2 1C and ka ¼ 0:024 W=m 1C (McCullough et al., 1989). Similar equations can be written for the evaporative resistance of an air layer. A relationship of the following form was taken as Re;al ¼ a½1 expðxa =bÞ,
(19)
2.2. Prediction of thermal comfort Thermal comfort is related to the physiological responses of people. For this purpose, comfort indices have been developed to quantify the degree of discomfort. To predict it, TS is given by Burch et al. (1991b) as TS ¼ ð0:303 expð0:036MÞ þ 0:028ÞL,
(22)
where L is thermal load on the human body. Thermal sensation is scaled as 5 extremely cold, 4 very cold, 3 cold, 2 cool, 1 slightly cool, 0 neutral, +1 slightly warm, +2 warm, +3 hot, +4 very hot and +5 extremely hot.
3. Experimental setup Although, vehicles have smaller volume than home or office environments, they have greater heat loss or gain. Most automobiles have a heating, ventilation, air
ARTICLE IN PRESS 306
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
conditioning (HVAC), device to control thermal environments of passenger compartment. In general, the HVAC system causes a complicated three-dimensional unsteady turbulent flow and temperature variations in the vehicle interior. The ventilation flow could be affected by several parameters such as the shape and location of the air-vents (inlet and outlet), the flow rate and the disposition of the passengers, etc. Nonuniform air and temperature distribution may cause localized discomfort of the occupants in the vehicle. In order to estimate the thermal comfort level, accurate information on the thermal environment is essential. The present experimental study was performed on a 1991 model Toyota Corona Sedan automobile equipped with a 2000 cc engine (see Fig. 1a). The automobile was instrumented with sensors in the passenger compartment to measure the air temperature, velocity and relative humidity on the locations shown in Figs. 1b and c. Equipment used in temperature, velocity and relative humidity measurements are shown in Fig. 2. A series of tests and measurements were performed in the automobile while parked in cold and cloudy winter conditions in February in Turkey. Air temperature measurement data were collected every 3 s by means of 16 temperature sensors located in
Fig. 2. Experimental apparatus: (a) thermometer system, (b) hygrometer, and (c) anemometer.
Fig. 1. (a) The automobile used in experiments, (b) measurement locations from top view, (c) measurement locations from side view.
the car as shown in Figs. 1b and c. Since the large number of sensors were used in measurements and the data were automatically saved to the hard drive, a data acquisition system integrated with a PC was employed to control the complicated task. Relative humidity and air velocity were measured in 1 min intervals. In order to find out the effects of thermal environments on the human body in the automobile cabin, skin temperatures were measured on the 10 different locations of the body (i.e. foot, fibula, thigh, pelvis, chest, back, head, hand, forearm, upperarm). In order to calculate the conductive heat loss from the body, interface temperatures of the body contact with solid surfaces, which are seat and
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
307
Table 2 Measurement instruments Measuring range
Accuracy
Temperature/humidity measuring: Testo 454 Multi-function measuring instrument Temperature probe 0y+70 1C Humidity probe
0y+100%RH
Velocity measuring: AIRFLOW Developments Limited Velocity probe 0y15 m/s
back support, were also measured during the experiments. Specifications of the equipments used in the experiments are presented in Table 2. In the beginning of each experiment, the human subject entered the automobile cabin where inside temperature was equal to outside ambient temperature and the standard heating system of the automobile was operated. Outside ambient conditions were approximately 0 1C temperature and 73% relative humidity. During the heating period, required measurements were taken in the automobile compartment and on the test subject. Results of the measurements are given in the Experimental results section. In order to determine the value of the experimental error, uncertainty analysis was carried out using equations proposed by Moffat (1988). Maximum uncertainties in experimental results were found to be within 73%. In addition, it must be mentioned that measurements were taken in a parked automobile with only the driver inside. The automobile in motion or in an automobile with passengers may affect the measurements.
70.4 1C (0y+49.9 1C) 70.5 1C (+50y+70 1C) 72%RH (0y+9.9%RH) 71%RH (+10y+90%RH) 72%RH (+90.1y+100%RH) 72% FSD at 20 1C and 1013 mbar
Fig. 3. Comparison of body heat losses in the heating period.
4. Results and discussion In order to validate the mathematical model, the results obtained from the model, are compared with the experimental measurements of Burch et al. (1991a, b). In their experimental study interior air has been heated from 20 to 20 1C in a half-hour time period. They measured air temperature, humidity, mean radiant temperature, seat, back support and steering wheel surface temperatures. By the use of their experimental data, they presented heat loss from the body and TS as shown in Figs. 3 and 4. Figs. 3 and 4 also show the comparisons between the present mathematical model results and the results of Burch et al. (1991b). At the beginning of the heating period, the automobile interior temperature is very low, therefore heat losses are very high, and the thermal sensation value is about 4.5. After the heating system starts to work, heat losses from
Fig. 4. Comparison of thermal sensation during the heating period.
the body decrease with increasing interior temperature, hence the thermal sensation value increases. It can be seen that the agreements between the results are very good. Mathematical model results obtained by the use of the data presented in Burch et al. (1991b): Initial core temperature T cr ¼ 37 1C; initial skin temperature T sk ¼ 34 1C; average thermal resistance of clothing 1.5clo
ARTICLE IN PRESS 308
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
(0.23 m2K/W), metabolic rate due to activity M act ¼ 136 W (75 W/m2), relative humidity 35% and mean radiant temperature T¯ r ¼ 0:94T a 1:38: 4.1. Experimental results Experiments were conducted for the heating period, which started when the interior temperature was about 0 1C. Fig. 5 shows the change of temperatures at front and back sides of the cabin during 1 h time period. The front and back side temperatures are the arithmetic mean values of the measured first eight points (1–8) and the last eight points (9–16) shown in Figs. 1b and c, respectively. The difference between the front and back side temperatures is not too much during all the heating period, the front side temperature is slightly higher than the back side one. It can be also seen from Fig. 5 that the outside temperature stays almost constant about 0 1C. Variation of the air temperatures at vertical locations at front and back side with time during a warm-up period in an automobile is shown Fig. 6. Air temperature difference between the head and foot levels rises up to 5 1C. Air temperatures on foot level are lower than air temperatures on head level both at front and at back sides in the automobile. Variations of the inside air and surface average temperatures and relative humidity with time during a warm-up period are shown in Fig. 7. Since the temperature of the surfaces and materials used in the automobile rises slower than the interior air temperature, there is a difference between the mean temperatures during the heating period. The mean temperature of the surfaces is lower than the interior air mean temperature. Relative humidity decreases with increasing temperature of the air inside the automobile. At the beginning of the heating period relative humidity is about 73%, it drops sharply in 30 min, and then its magnitude stays about 20%.
Fig. 6. Variation of the air temperatures at vertical locations with time during the heating period, (a) at front side, (b) at back side.
Fig. 7. Variation of the average inside air and surface temperatures and relative humidity with time during the heating period.
4.2. Inputs used in automotive application of the model
Fig. 5. Variation of the air temperatures with time during the heating period.
The following empirical relations are derived by the use of present measured data of air temperature, relative humidity and surface temperature inside the automobile, and they are used in the computer program in order to simulate the conditions of the environment and
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
309
passengers during normal warm-up period in winter driving. In the following equations t represents the elapsed time in minutes from start-up. Mean air temperature: ( 1:177 lnðtÞ þ 2:034 Ta ¼ 10:306 lnðtÞ 13:824
for 0otp7; for t47:
(23,24)
Mean radiant temperature: T¯ r ¼ 1:769 expð0:0871T a Þ.
(25)
Mean relative humidity: RH ¼ 73:018 expð0:0257tÞ.
(26)
Surface temperatures of solids in contact with the body (Tint) were: Seat: T int ¼
(
11:253 lnðtÞ þ 22:983 for 0oto3; 1:885 lnðtÞ þ 25:643 for tX3:
Back support: ( 10:729 lnðtÞ þ 22:352 for 0oto3; T int ¼ 2:751 lnðtÞ þ 22:642 for tX3:
Fig. 8. Average body skin temperature in the heating period.
(27,28)
(29,30)
Other constants are used in the mathematical model taken as follows: Average thermal resistance of clothing Rcl ¼ 1clo ð0:155 m2 K=WÞ; metabolic rate due to activity 75 W/m2, the surface area of the nude body 1.751 m2 (see Table 1), ratio of clothed to nude body area f cl ¼ 1:1: Local air velocities on the body of sitting passenger were measured under warm-up conditions, as shown in Table 3. 4.3. Comparisons of the measurements and model results Skin temperature results obtained from the present measurements and simulations are given comparatively in Fig. 8. As it can be seen from Fig. 8, both results are in a good agreement. At the beginning of the heating period, the automobile interior is cold, hence mean skin temperature decreases. Then, mean skin temperature increases with the increasing automobile inside temTable 3 Local air velocities on the body Region
Air velocity (m/s)
Head Trunk Right shoulder Left shoulder Right knee Left knee Right ankle Left ankle
0.25 0.13 0.25 0.35 0.30 0.12 0.60 0.75
Fig. 9. Hand and forearm skin temperatures in the heating period.
perature. Mean skin temperature gives an idea of thermal comfort, but it is not enough to determine comfort level of the occupant. Although the mean body skin temperature is in the required range, the temperatures of the extremities such as hand, foot and face or naked parts of human body may stay on undesired values. For example, the maximum change of temperature takes place in hands, which are unclothed body segments (Fig. 9). At the beginning, the hand skin temperature is 20.4 1C, and then it rises 30 1C with the increasing automobile inside temperature. Pelvis is the least affected segment from the environmental conditions (see Fig. 10). Since the clothing thermal insulation of the pelvis is greater than other body segments, the skin temperature of the pelvis is high and its temperature does not change too much in the heating period. As it can be seen in Fig. 11, the temperature changes of the fibula and thigh are lower than those of the hands. Variations of sensible and latent heat losses from the whole body during the heating period are shown in Fig. 12. In early minutes, convective and radiative heat losses are high since the temperature inside the automobile and the interior surface temperatures are very
ARTICLE IN PRESS 310
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
Fig. 10. Pelvis and chest skin temperatures in the heating period.
through the skin (Edif). Evaporation of sweat secreted due to thermoregulatory control mechanisms (Ersw) is equal to zero because sweat generation does not occur. Considering the Figs. 13–15, the same conclusions can also be drawn for chest, fibula and thigh. Since the hand is naked and the forearm has light clothing, heat loss from these segments rapidly decreases with increasing automobile inside temperature as can be seen from Fig. 13. The pelvis has high clothing thermal insulation according to other body segments, therefore both sensible and latent heat losses are low, as seen in Fig. 14. When both Figs. 15 and 14 are examined together, it can be observed that this segment is not too much affected from the environmental conditions. Generally, latent heat loss is lower than sensible heat loss for all body segments. Variation of TS values in the warm-up period is given in Fig. 16. In the beginning of the heating period, the body heat loss is greater than the generated energy because automobiles, inside temperature is about 0 1C. For this reason, a thermal sensation index has been started from low values (approximately 2). And then,
Fig. 11. Fibula and thigh skin temperatures in the heating period.
Fig. 13. Heat losses from hand and forearm in the heating period.
Fig. 12. Heat losses from the whole body in the heating period.
low. Since temperature difference between the skin and ambient decreases with time, these heat losses decrease, too. Evaporative and respiratory heat losses are not severely affected by the environment and stay at about 15–20 W during the period. Latent heat loss from the skin is caused from the natural diffusion of water
Fig. 14. Heat losses from chest and pelvis in the heating period.
ARTICLE IN PRESS O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
Fig. 15. Heat losses from fibula and thigh in the heating period.
311
model, and predictions were compared with the measured data. By means of the present model, the effects of the fast transient conditions of the heating period on the sensible and latent heat transfer from the body, body segments skin temperatures and thermal sensation were also investigated. In view of the fact that the occupant is exposed to highly nonuniform environmental conditions, it is shown that there is a good agreement between the model predictions and experimental results, and the accuracy of the predictions obtained with the all body segments separately and body-averaged results is remarkable. As a result of this study, it is believed that the methodology and the model introduced would enable the designer to test and to better optimize the air conditioning system’s ability to meet comfort requirements, while minimizing the need for expensive and time-consuming experiments.
References
Fig. 16. The change of thermal sensation during the heating period.
thermal sensation has improved with increasing inside temperature and interior surface temperatures.
5. Conclusions In this paper, a combined theoretical and experimental study of thermal comfort inside an automobile during the heating period is presented. The effects of thermal conditions on the human physiology and thermal comfort during the heating period are examined in details. A theoretical simulation model has been developed for automobile interior thermal comfort and evaluation. The human body is divided into 16 sedentary segments, the change of temperature was observed both theoretically and experimentally. Temperature, humidity and air velocity were measured at a number of points inside the automobile, so thermal conditions were accurately determined. During transient conditions of the heating period, heat and mass transfer between the human body and the interior environment of an automobile were simulated by a computational
Aroussi, A., Aghil, S., 2000. Characterisation of the flow field in a passenger car model. Opt. Diagnostics Eng. 4 (1), 1–15. ASHRAE Handbook—Fundamentals, American Society of Heating, Refrigeration and Air-Conditioning Engineers, Atlanta, 1989, p. 29 (Chapter 8). Burch, S.D., Pearson, J.T., Ramadhyani, S., 1991a. Experimental study of passenger thermal comfort in an automobile under severe winter conditioning. ASHRAE Trans. 97, 239–246. Burch, S.D., Ramadhyani, S., Pearson, J.T., 1991b. Analysis of passenger thermal comfort in an automobile under severe winter conditioning. ASHRAE Trans. 97, 247–257. Butera, F.M., 1998. Principles of thermal comfort (Chapter 3). Renewable Sustain. Energy Rev. 2, 39–66. Chakroun, C., Al-Fahed, S., 1997. Thermal comfort analysis inside a car. Int. J. Energy Res. 21, 327–340. Daanen, H.A.M., van de Vliert, E., Huang, X., 2003. Driving performance in cold, warm, and thermoneutral environments. Appl. Ergonomics 34, 597–602. Doherty, T.J., Arens, E., 1988. Evaluation of the physiological bases of thermal comfort models. ASHRAE Trans. 94 (Part 1), 1371–1385. Guan, Y., Hosni, M.H., Jones, B.W., Gielda, T.P., 2003a. Investigation of human thermal comfort under highly transient conditions for automotive applications—Part 1: experimental design and human subject testing implementation. ASHRAE Trans. 109, 885–897. Guan, Y., Hosni, M.H., Jones, B.W., Gielda, T.P., 2003b. Investigation of human thermal comfort under highly transient conditions for automotive applications—Part 2: thermal sensation modeling. ASHRAE Trans. 109, 898–907. Guan, Y., Hosni, M.H., Jones, B.W., Gielda, T.P., 2003c. Literature review of the advances in thermal comfort modeling. ASHRAE Trans. 109, 908–916. Jones, B.W., 2002. Capabilities and limitations of thermal models for use in thermal comfort models. Energy Build. 34, 653–659. Kaynakli, O., Unver, U., Kilic, M., 2002. Simulation of thermal comfort heating and cooling periods in an automobile compartment. Proceedings of the Automotive Technologies Congress, 24–26 June, Bursa, Turkey, pp. 127–135 (in Turkish). Kaynakli, O., Unver, U., Kilic, M., 2003a. Calculation of thermal comfort zones with the ambient parameters. Proceedings of the
ARTICLE IN PRESS 312
O. Kaynakli, M. Kilic / Applied Ergonomics 36 (2005) 301–312
First International Exergy, Energy and Environment Symposium, 13–17 July, Izmir, Turkey, pp. 769–773. Kaynakli, O., Unver, U., Kilic, M., 2003b. Evaluating thermal environments for sitting and standing posture. Int. Com. Heat Mass Transfer 30 (8), 1179–1188. Lee, S.J., Yoon, J.H., 1998. Temperature field measurement of heated ventilation flow in a vehicle interior. Int. J. Vehicle Design 19 (2), 228–243. McCullough, E.A., Jones, B.W., Tamura, T., 1989. A data base for determining the evaporative resistance of clothing. ASHRAE Trans. 95 (2), 316–328.
Moffat, R.J., 1988. Describing the uncertainties in experimental results. Exp. Thermal Fluid Sci. 1, 3–17. Olesen, B.W., Hasebe, Y., de Dear, R.J., 1988. Clothing insulation asymmetry and thermal comfort. ASHRAE Trans. 94 (1), 32–51. Parsons, K.C., 1993. Human Thermal Environments. Taylor & Francis Ltd., London, UK. Parsons, K.C., 2000. Environmental ergonomics: a review of principles, methods and models. Appl. Ergonomics 31, 581–594. Tanabe, S., Arens, E.A., Bauman, F.S., Zang, H., Madsen, T.L., 1994. Evaluating thermal environments by using a thermal manikin with controlled skin surface temperature. ASHRAE Trans. 100 (1), 39–48.