Studies by Webb

W ebb [50] w hile w orking at the U niversity of M alaya in Singapore, c o n d u cted an o th e r field stu d y u sing 14 local S in g ap o rean s a n d 2 E uropeans w ho had lived there for m ore than 20 years. Besides taking readings of physical m easurem ents, the participants voted on a 9-point interval com fort scale also know n as the extended Bedford scale'. The term s 'extrem ely hot' an d extrem ely cold' w ere a d d e d to the tw o extremes of the scale. The participants carried out their daily routine as occupants of the houses (which were of different types) and buildings.

W ebb fo u n d th at m axim um com fort w as fo u n d at a m ean Effective T em perature of 26.2°C w hen the m axim um percentage of votes w ere registered. Table 2.10 shows the full results of W ebb's study. The range of comfort, according to Webb, lies between votes of 2.5 and 5.5 which refers to 25°C to 27.2°C. H ow ever, the range of comfort is usually considered to be betw een 'com fortably w arm ' and 'com fortably cool' on the com fort scale betw een the votes of 3.0 to 5.0 inclusive. W hen the latter range of comfort is used, the tem perature is found to lie betw een 25.3°C to 26.7°C.

The upper limit of comfort as found by Webb is sim ilar to the u p per lim it of neutrality of the Europeans and Asians in Ellis's study. It is interesting to note th at Ellis and Webb used therm al sensation and therm al com fort scales respectively.

In a separate analysis, m ultiple regression was used to obtain the following relation, know n as the Singapore Index or the Equatorial C om fort Index (E.C.I.). The index in term s of air tem perature, T, vapour pressure, P and air velocity, V, is as follows:

COMFORT VOTE VOTE RANGE MEAN EFFECTIVE TEMP [°C] COMFORTABLY COOL 2.5 - 3.5 25.1 - 25.8 COMFORTABLE 3.5 - 4.5 25.8 - 26.6 COMFORTABLY W A R M 4.5 - 5.5 26.6 - 27.3

Table 2.10 : Mean Effective Tem perature Range for the 3 m iddle Categories of the Comfort Scale [50]

0 = 0 .5 7 4 T + 0 .4 8 8 P - 0 .2 3 1 V v + 2 1 .2 3

The Equatorial Comfort Index is also w ritten by Webb in term s of dry-bulb tem perature, T d , w et-bulb tem perature, T w , and air velocity as follows:

0 = 0 .4 4 7 Td + 0 .5 5 3 T w - 0 .2 3 1 V v

The coefficients of the variables suggest th at a sim plification of the equation is possible. The simplification will result in errors betw een 0.2 to 0.4°F (0.1 to 0.2°C) in most cases. The simplified form of the ECI, according to Webb is as follows:

0 = 4 ( T d + T w ) - 0.25 Vv

w here 0 = the index in °F T = air tem perature °F

P = vapour pressure in mm Hg V = air velocity in ft/m in Td = dry-bulb tem perature in °F T w = wet-bulb tem perature in °F

The correlation coefficient betw een therm al sensation an d the dry-bulb tem p era tu re is 0.52. The m axim um percentage of subjects w ho are com fortable is 68 percent, w hich is at 26°C. The n o m o g ram of the Equatorial Comfort Index is shown in Figure 2.12.

l3(fF 120'F no°F mo'F LU en 3 y~ < cn L Ü Q_ Z LU CD _J 3 cn > en a 9CTF 8CTF ( 2 3 4 " :) 75" 75>Y).i'c)70 (2&7°C) 8 0 ' (STJZ'c^gpY 95"F 3 5 T ) ( 2 0 r c ) 85"F/ 7 8 . 7 'F ■90°F , ■bz?c) ■8S°F ■8CfF ^ 2&rc) -75 T

Figure 2.12: The Equatorial Comfort Index N om ogram developed by Webb [51]

It is im p o rtan t to note that since W ebb used a comfort scale, his results represent comfort rather than neutrality.

2.8.3 Studies by Lim and Rao

Lim an d Rao [52] conducted field surveys of pre-university stu d en ts in their classes an d 'haw ker centres' (cooked food stalls) in Singapore. A C om fort M eter w as also used to provide Predicted Mean Votes w hile the subjects voted on the Thermal Sensation scale.

The authors suggested that the neutral E.C.l. developed by W ebb m ay be lo w e re d from 26°C to 25°C. At 25°C, the m axim um p ercen tag e of neutrality assessment, which was 59 percent, was obtained. The PMV from the C om fort M eter also gave a neutrality zone aro u n d 25°C. The results obtained by Lim and Rao generally agree w ith Webb's, despite the fact that E.C.l. is a comfort index and PMV is a neutrality index.

2.8.4 Climate C ham ber Study by de Dear et al.

W hile all the studies conducted in the equatorial region have been of the survey type, a recent study has been carried out using the therm al chamber m ethod. The experim ents were carried out by de Dear, Leow and Ameen [53] at the N ational University of Singapore.

T hirty-tw o college students, half males and half females, particip ated in tw o and half-hour sessions. They w ore the standard KSU uniform of 0.6 d o value an d m aintained sedentary activity th ro u g h o u t the experim ent. They took p a rt one by one rather than on a group basis as is com m on in o th er cham ber studies. The p articip an ts filled in a q uestionnaire sheet w hich co n tain ed in stru ctio n s and p ro ced u res for the ex p erim en t an d v o ted on th e T herm al Sensation an d Preference R ating Scales. The p articip an ts an sw ered the questionnaire sheet and the scales every 10 m inutes. According to the preference rating of the participant (which w as conveyed to the technician via an intercom ), the tem p e ra tu re of the cham ber w as adjusted im m ediately in accordance w ith the participant's preference.

The m ean of the 32 participants' preferred tem peratures w as found to be 25.4°C and at 95% confidence interval, the estim ate of this m ean ranged from 25.0°C to 25.8°C. The preferred tem peratures of the m ales and the fem ales w ere not statistically significant.

W hen the p articip an ts w ere at their preferred tem perature, the average vote is -0.3. This value is closer to the neutral vote of 0.0 than the slightly cooler vote of -1.0. The au th o rs calculated th at the -0.3 vote, w hen converted into tem perature, gives a neutral tem perature of 26.4°C, which is IK h ig h e r th an the p refe rred tem p eratu re. The m ean p re fe rre d tem perature of 25.4°C is very close to Webb's E.C.I.

The authors also quoted a field study [54] they conducted in a high-rise b u ild in g in Singapore. The n eu tral tem p eratu re obtained was 28.5°C, w hich is 2K w arm er than th at p red icted by the PMV m odel for the hum idity, air speed, clothing and metabolic rate observed in the survey.

The authors com m ented that the discrepancy in the results betw een the cham ber m ethod and the field studies can be attributed to the difference b etw een 'n eu trality ' and 'preference'. The au th o rs concluded th at the preferred tem perature of Singaporeans is not significantly different from the preferred tem peratures of the people from the tem perate or tropical clim ates.

de Dear, Leow and Ameen carried out a follow-up study [55] to investigate th e acceptability of am bient tem p eratu res th at are w arm er th an the preferred tem perature. Two hum idity levels, 35% and 70% w ere used at six different tem peratures that ranged from 25°C to 30°C. The participants v o ted on th e T herm al S ensation scale an d the C om fort scale an d responded 'yes' or 'no' to a question on acceptability of the condition.

U sing a 20% dissatisfaction criterion, the u p p er lim it of the acceptable com fort zone at 70% relative hum idity was established at 27.6[°C] while the corresponding value at 35% was 27.9[°Cj. N ote that the tem peratures are alm ost sim ilar so the participants w ere not sensitive to the different hum idities. ASHRAE [16], prescribed an acceptable tem perature of 26.5°C at 50% relative hum idity, which is 1.5K cooler.

Finally, M arkus and Morris [56] who had w orked in Malaysia, attem pted to explain w hy the therm al com fort of tropical people is different from tem perate people. M arkus and Morris observed that tropical people exhibit a slow er pace of m ovem ents, which w ould lead to low er m etabolic rates. This m ight suggest that in everyday life tropical people can tolerate higher am bient conditions.

In sum m ary, there is a large discrepancy betw een the neutral tem peratures of the field study which includes high-rise buildings, as carried out by Lim and Rao and a similar study by de Dear et al. The difference betw een the neutral tem peratures is 3.5°C. The field studies carried out in the 1950s by Ellis and Webb show close sim ilarity in the u pper comfort limit.

The clim ate cham ber stu d y by de D ear et al determ in ed the Preferred T em p eratu re of the Singapore students to be 25.4*C. Table 2.11 gives a sum m ary of the major studies carried out in the equatorial region.

of y

EQUATORIAL STUDIES TEMP. [C] COMMENTS

ELLIS - TROPICAL SEAS 22.2 - 26.7 EFFECTIVE TEMP., EUROPEANS ON

W ARSHIP

ELLIS - SINGAPORE 22.2 - 26.1

21.7-27.2

EFFECTIVE TEMP. FOR EUROPEAN EFFECTIVE TEMP. FOR ASIAN , FEILD STUDIES

WEBB - SINGAPORE 25.0 - 27.2 EFFECTIVE TEMP. FOR

SINGAPOREANS, FIELD STUDIES

WEBB - SINGAPORE 26.0 DRY-BULB TEMPERATURE,

SINGAPOREANS, FIELD STUDIES

LIM - SINGAPORE 25.0 FIELD STUDIES AND PMV FROM

COMFORT METER

DE DEAR - SINGAPORE 25.4 PREFERRED TEMP., THERMAL

CHAMBER

DE DEAR - SINGAPORE 28.5 NEUTRAL TEMP., FIELD SURVEY

DE DEAR - SINGAPORE 27.6 UPPER ACCEPTABLE TEMP. AT 70%

REL. liU M . THERMAL CHAMBER Table 2.11 : Sum m ary of Equatorial Region Studies