Climate and building design - tradition, research and design tools

Energy and Buildings, 7 (1984) 55 - 69 55 Climate and Building Design -Tradition, Researchand Design Tools E. SHAVIV Faculty of Architecture and Town Planning, Technion -- Israel Institute of Technology, Haifa (Israel) SUMMARY Important design paramaters that influence the building thermal behavior and in particular natural cooling are discussed. A m o n g these design parameters are: ventilation, evaporation, proper shading, orientation o f the building and its proportions with respect to the orientation, the color o f the building's envelope and its conductivity, the thermal mass o f the building, night radiation to the sky and the stack effect. The latter is the katabatic and anabatic cooling. Different design tools aimed at the study o f the influence o f the climatological parameters on the form and characteristics o f buildings were developed in recent years. These tools help architects in designing houses with improved indoor thermal conditions without mechanical means, or with m i n i m u m energy consu mp tions. Several design tools, as well as design considerations and traditional constructions, are presented. Emphasis will be p u t on computer aided design tools. INTRODUCTION Building design embraces a large number of parameters that affect the thermal behavior of the building. These parameters include the climatological ones like the run of ambient temperature, as well as the amplitude of temperature variations, the intensity of the radiation incident on the exterior walls (short waves and long waves), the relative humidity and wind direction. While the climatological parameters are environmental variables and are not under human control (except for the choice of the location of the building), the second t y p e of parameters are the design variables which are under the control of the 0378-7788/84/$3.00 architect. These design parameters include the general layout of the building with respect to the orientation that influences, for example, the insolation and the impact of wind velocity. Other important design parameters are: the area of the outer envelope, the location of windows and their sizes, shading of windows and envelope, the color and texture of the building, the distances between the buildings and the existence of interior courts. The thermophysical properties of the building materials like conductivity and heat capacity affect the in-house conditions also. The general problem of predicting the thermal performance of buildings involves the interplay between a large number of parameters and complicated mathematics, and therefore finding the answer to such a problem by hand is very tedious. Accurate solutions to the mathematical equations that contain all the relevant design and climatological parameters can be obtained only by a computer. Many simulation models to predict the thermal behavior of buildings were developed in recent years: to mention just a few; the ESP developed by Clarke in Europe [1] and the PASOLE, BLAST, TRANSYS, DEROB or SERI-RES models [2 - 6] developed in the U.S.A. I will refer in this paper to a simulation model developed in Israel by Shaviv and Shaviv seven years ago [7 - 10]. This dynamic simulation model was developed for predicting the thermal behavior and energy consumption of full-scale buildings. The model can include most of the design and climatological factors affecting the building performance. The time-dependent equation for the heat flow through the walls is converted into an implicit scheme and solved numerically. Special effort has been devoted to producing a model capable of aiding the architect during the various steps of the © Elsevier Sequoia/Printed in The Netherlands 56 TABLE OF E N E R G E T I C S OF A L T E R N A T I V E S A I SPECIFICATION : : .A-~ : : : : : ENERGY C O N S U M P T I O N MONTH : Tl E R : Th ............................ 1 : 1B,8 12.1 77,4 : 1B.7 8 : 25.8 -Sl.6 16.2 : 25.7 ............................................................................................... TOTAL HEATING CONSUMPTION TOTAL COOLING CONSUMPTION T O T A L ENERGY CONSUMPTION ................................................. : 843.7 -5809.3 6653.0 : : A-3 1 A-4 : : : : : E R : TA E R : TA .L. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.9 77.4 : 1B.9 1.9 77.4 : 1B.B -59.7 16.2 : 25.7 -57.8 16.2 : 25.7 169.7 : 148.7 : -60Sl . 3 -5070.9 6221.0 : 5219.6 : ~ ..................................................................... : TOTAL HEATING SAVING : 0 0 : -674.0 79.9 TOTAL COOLING SAVING O 0 : 242.1 4.2 T O T A L ENERGY SAVING : 0 0 : -432.0 6.5 .......................................................................................................................... : : : -695.0 -738.4 -1433.4 B2.4 12.7 21,S : : : L.AL. HEATING SAVING 1 O 0 : -674.0 79.9 L.AL. C O O L I N G S A V I N G 0 0 : 242.1 4.2 L.AL, ENERGY SAVING : 0 0 : -432.0 6.5 .......................................................................................................................... : : : -21,0 -980.4 -1D01.4 12.4 16.2 16.1 : : : NO FAN MONTH : TA TM TL I TA TM TL : TA TM ............................................................................................ 8 : 27.4 2B.3 4.0 ; 27.4 28.3 4.0 : 27.3 2B.2 .......................................................................................................................... WITH FAN MONTH : TA TM TL : 8 : 26.6 27.7 4.0 : ............................................................................................ E ENERGY C O N S U M P T I O N IKWHI R SOLAR RADIATION TA TM 26.6 (KWH) 27.6 TL 4.0 : : TA AVERAGE TEMP TM TA 26.5 (OEG 27.5 CI TL 4.0 TL 4.0 : : : : E R : A-S TA 12.1 77.4 : 18.9 -55.9 10.0 : 25.7 ~ .......................... 843.7 - S O 12 . 5 5856.3 : 0 O : 13.7 : 12.0 : -796.7 -796.7 27.2 TM 28.0 TA 26.4 TM MAX TEMP TL 4.0 TM : : TL : E 11.0 "58,B C) : : 39.1 7.6 11.6 : : : -330.0 356.5 26.5 39.1 7.1 .5 : : : TL : TA TM 27.3 2B.2 TA TL TIME : : -330.0 -440,2 -770.2 TM 27.4 4.0 : 26.S -,. ......................... (DEG R : 77,4 16.2 S13,7 -5369 . t SB82.B : 695.0467.5 : -58,4 1.2 : 636.7 12.2": TA : 1 27.S LAG 4.0 : TL : q.O : (HOURS) Fig. 1. An example of a comparison table obtained by the computer. A-1 to A-5 are different alternative designs of a solar house. A-1 is the first design, A-2 is an improvement in U-value of southern windows (and slightly enlarging these windows). A-3 is an improvement in reducing infiltration through the above windows. A-4 is an improvement in southern window shading during summer only and A-1 is an improvement in the U-value of the roof. The Table gives the energy consumption, the energy saving compared with the first design and with the last alternative design. It also gives the maximum temperatures in summer with or without operating a fan whenever the temperature outside is lower than inside the building. LIVING ROOM BEDROOM i • K IT C H E N BROTH ~RTH BEDROOM I BEDROOM N Fig.2. The layout of the house examined in the numerical examples and shown in Figs. 3, 4, 5, 9. design p r o c e s s , so as to p r o v i d e t h e r m a l comfort with minimal energy consumption, T h e results o f a s i m u l a t e d building are pres e n t e d in a graphical f o r m (see Fig. 3) a n d in a table that summarizes and compares different design a l t e r n a t i v e s (see Fig. 1). This f o r m o f presentation enables the architect to discover easily t h e crucial f a c t o r s in t h e therm a l p e r f o r m a n c e o f t h e building. Most of the simulation models developed for p r e d i c t i n g t h e t h e r m a l p e r f o r m a n c e o f buildings a p p l y various a p p r o x i m a t i o n s in t h e solution of the heat equation. The model d e v e l o p e d b y Shaviv a n d Shaviv [7 - 9] solves t h e h e a t e q u a t i o n very a c c u r a t e l y , y e t t h e c o n v e r g e n c e is fast a n d efficient. T h e treatm e n t o f long-wave r a d i a t i o n e n e r g y e x c h a n g e b e t w e e n t h e building and t h e sky, t h a t plays a d o m i n a n t role in passive solar cooling, is t r e a t e d carefully. Most o t h e r m o d e l s ignore it. The model predicts correctly the appearance o f d e w on various surfaces a n d the e f f e c t s o f n o c t u r n a l cooling. A n o t h e r p o i n t to stress is t h a t t h e c a l c u l a t i o n s o f d a y l i g h t a n d artificial lighting n e e d e d t o k e e p a p r e s c r i b e d level o f i l l u m i n a t i o n in the building are p e r f o r m e d d y n a m i c a l l y [11 ]. As a n y o t h e r s i m u l a t i o n m o d e l s , t h e m o d e l d e v e l o p e d b y Shaviv and Shaviv is m a i n l y an e v a l u a t i o n r a t h e r t h a n a design t o o l . This is b e c a u s e a s i m u l a t i o n m o d e l d o e s n o t yield an o p t i m u m s o l u t i o n . H o w e v e r , t h e results o f this s i m u l a t i o n m o d e l are p r e s e n t e d in an a t t e m p t t o d i r e c t t h e designer t o w a r d s t h e w e a k p o i n t s in t h e p r e s e n t e v a l u a t e d design alternative, so t h a t o n e can k n o w w h a t is t h e m o s t e f f e c t i v e design p a r a m e t e r , at each stage o f t h e design, t h a t s h o u l d b e c h a n g e d in o r d e r t o i m p r o v e t h e building (see ref. 8). We discuss t h e e f f e c t o f d i f f e r e n t design parameters on the thermal performance of the building. T h e e v a l u a t i o n o f t h e p e r f o r m a n c e 57 of the building was carried o u t by using the simulation model briefly described above. Traditional examples for the proper use of these design parameters in the past will be presented. The emphasis is p u t on design for passive cooling in hot climates. Some design tools to reach a good thermal design will be reviewed along the general discussion, especially if these are computer aided architectural design tools. VENTILATION Ventilation plays a triple role in controlling the in-house thermal conditions: (1) by mixing of outside and inside air. In summer, this mixing is desirable whenever the outside temperature is below the in-house one, and undesirable otherwise. It is also usually undesirable in winter: in this case, the mixing is called infiltration. (2) by creating a local air motion, which helps the evaporation of perspiration and improves the thermal comfort in summer. This local air motion is very important in hot, humid climates. (3) by removing excessive humidity. Often, the in-house relative humidity is higher than the outside one (particularly in winter). Ventilation can remove the excessive water vapor and prevent condensation {if such a likelihood is imminent). Let us discuss in more detail the first point mentioned, i.e., mixing of outside and inside air. The time lag of houses built from massive materials can be as long as 5 to 6 hours. Therefore, the internal temperature of the house, which is heated during the day, reaches its maximal value in the evening. In many hot regions, the temperature drops in the evenings and nights. A proper design of the openings should allow for effective inside-outside air mixing during these hours and in this way remove the in-house hot air. The effect of mixing on the in-house temperature can be quite large, in spite of the fact that the walls continue to emit heat into the house during the evening hours. This heat obviously accumulates in the walls during the day. Fast ventilation can overcome this heat flow into the house and reduce the in-house air temperature to almost the outside one. Note that in this case, the walls cool from both sides and the maximal temperature in the wall is somewhere inside it and not on the inner surface. Hence, the walls will start heating during the subsequent hot day from a lower temperature. Figure 3 demonstrates the above prediction by means of a simulation of the performance of a house situated in Herzlia on the sea, which is a suburb of Tel Aviv (see Fig. 2). The house is built from typical Israeli building materials, like hollow concrete blocks covered with plaster on both sides, and from concrete slabs and cement floor tiles. Thus, the house possesses a high thermal mass. The initial design of the house had large unshaded western windows. The simulation of this unshaded and unvented house (only 0.5 air change per hour} shows that the house operates in summer as a greenhouse. Although the maximum temperature outside is only 30.8 °C, the in-house temperature reaches 32.5 °C. Note that the maximal in-house temperature occurs at about 1 7 : 0 0 - 1 8 : 0 0 . At this time (18:00), the outside air temperature drops to 28 °C. A fast mixing between the inside and outside air can reduce the in-house temperature significantly. Figure 4 shows that ventilation at a rate of 10 air changes per hour can reduce the inside temperature to 29 °C. Physical modelling of air flow inside complex buildings and through exterior wall openings is often carried out by smoke experiments with models of buildings in wind tunnels, or in actual buildings. Tests of air flow patterns were carried out by Olgay [12] over t w e n t y years ago and published in his book Design with Climate. A summary of experimental results on air flow inside houses and their environments can be found in a recently published paper by Bowen [13]. Islamic architecture has many examples of design with proper attention to ventilation. The famous Rushaan that was designed for h o t and humid climates like that of Jeddah provides good in-house ventilation, prevents excessive glare and renders uniform and pleasant daylighting. The Rushaan is used as a daily living area but mostly as night-sleeping space that is ventilated from all sides, including the floor (see ref. 14). The Rushaan is built with a lot of soft-wood that absorbs the humidity and hence the air entering the house is drier than the air outside the building. Let 58 56+,~; ::: :::::::::::::::::::::::: 5 1(9000 4 900O 3 8000 2 7000 I~ 6000 54 ~ mn ° 32 30 .= ~o 28" on o 26" • - oo nm •n"• o " ~MRMRMM 44444 R 4 M " " ....: " MpIM M M I~MM I" • 4 , u . . . . . . . . . 'L Immmmmmm,4 ~" 0 ~ -, ~~ ,ooo k-3 R!lI - ~ooo -2 """ M 5000 ";, ~1L ~ : MM i _. M . ,, m"M M ~ MMMMM~MM I:~ '~ Fl "6 • 4444 I'I . ,~ ~I;|~,~ "MM mmc " " "• ~Immmmmm.4 _~ M . . • :...... ..... $Im***4 .•'• a ~o • "-• 24. ' : " " " . 14 " DOe0- * • ,~ o • pa 2o1 • • " 4 n° oo o I~°D° " o~m 2000 I ,ooo I~ m I0 m ,, ..... n m 6 " ,8. . . . . ..... A .............. ~5. . . . 2, ..... "'''''''" a2 3~ .... .... ,i8 .... i4 .............. .... 66 "=== .... ~6 .... "5 0 -~2 Hour Fig. 3. Results of the thermal performance of an unshaded and unrented house during three successive days, obtained by a simulation model. Stars denote the outside air temperature (°C) and squares denote the in-house air temperature. R denotes the power deposited by short-wave radiation penetration through the windows (kcal/ hi. The numbers and M denote the difference between the in-house air temperature and the inside wall temperature. T h e n u m b e r s are: (1) n o r t h ; (2) east; (3) s o u t h ; (4) west a n d (5) r o o f ; (M) mass. 5 I0000 4 9000 5 8000 2 7000 34 32 ]0 • •" i~I l l 26, ",.F'E - " . : z0 . o • lllm I * . ~,E . III~*; " . ~ i R %F.~ • Y ~ '( I[ * II . .-.~ . ~ ~ ~ L ~ ea . M~II R ~ 12 18 24 30 36 42 u J 0 5000 " -, ~ 4000 --2 ~00o "-3 2000 "-4 moo0 mA 12. 6 ~ 0 MMR All ma . "-..%. . . . . .. 0 - OoOoa ° N q[ ~ME [ 411111111 4 MNM 14 lllm i M ,8. o a m - • ...%. . . . . i • o O o O o J° N • ~4E Ig 4 1 1 1 1 1 1 1 b m -...%. . . . . i • • ooooo ° e 4E t, 2 4 . I l l l l ~ I • • "" n l l l oe o • 28 ,~ m lu'lllll "•Qe 48 54 60 G6 72 Hour Fig. 4. Results o f t h e t h e r m a l p e r f o r m a n c e o f a n u n s h a d e d a n d v e n t i l a t e d h o u s e . T h e s p e c i f i c a t i o n of this h o u s e is t h e same as in Fig. 2, b u t a fan t h a t p r o d u c e s 10 air changes p e r h o u r o p e r a t e s during all h o u r s in w h i c h t h e o u t s i d e air t e m p e r a t u r e is l o w e r t h a n t h e inside one. When t h e fan is on, t h e s q u a r e r e p r e s e n t i n g t h e in-house air t e m p e r a t u r e c o n t a i n s a m i n u s sign. 59 us also mention the Egyptian Malkaf [15] and the Iranian wind towers designed to capture the breeze and provide ventilation in hot climates [ 16 ]. EVAPORATION Evaporation can provide good cooling in hot and dry climates. A good example is the Arabian water jar. The water permeates through the porous jar and its subsequent evaporation cools the water remaining inside. The Egyptian Meshrabeyh provides a good example of design with attention to ventilation and evaporation in hot, dry climates [15]. Like the Saudi Arabian Rushaan, it provides good in-house ventilation, pleasant daylighting and prevents excessive glare. However, this window element is designed with niches around it. Water jars are placed in the niches and the evaporated water cools and humidifies the fresh air which enters through the window. The original role of the jars is as containers of cold water, and for this reason, the window element is called Meshrabeyh, which means a place for drinking. The Egyptian Meshrabeyh and the Saudi Arabian Rushaan are expressions of local venicular architecture and have evolved from the local need for privacy as required by Islamic tradition and social standards. A similar architectural element, the role of which is to reduce glare and supply good ventilation, can be found in Spanish and Indian architecture. The use of light lattice work lets the breeze in. When this opening with the lattice work faces an internal court with water pool or fountain, the penetrating air cools by evaporation and humidifies. Obviously cooling by evaporation is good only for hot and dry climates and cannot be applied in humid places. Numerical models capable of describing the phenomenon and indicating how to spread the water in various spaces of the house are unavailable. The problem is investigated experimentally to some extent, but building tradition in countries with hot climate provides many good design examples. SHADING Shading is one of the most important design parameters for achieving good inhouse climatic conditions in countries with hot dry or hot humid summers. Direct insolation provides the cheapest means for winter ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: 5 IO000 4 9000 3 8000 2 70O0 32 50 ° ''''''. • "''''' -'°°'° ° • ° m m .L .~ E ii! :i" ! ~- 6000 5000 . ~" 20. '-~ 18, 40o0 I,- "-2 3000 '-3 ZoO0 '-4 I000 16. • 14. 12 . " . = . = = = . ; ; : : : : : : : : : : . .=.'.'.=f=.=.=.=,=." : : : : : : : : : : : : . .'=.='f=.='."."'.'=.'=.'=~. : 6 12 18 24 30 36 42 48 54 : : : : 60 HOUr Fig. 5. T h e t h e r m a l p e r f o r m a n c e o f a s h a d e d a n d v e n t i l a t e d house. . ==ft.= 66 72 - 60 space heating and hence is desirable. However, direct summer insolation must be reduced to a minimum in order to reduce the load on the air-conditioning system, or reduce the in-house air temperature when no mechanical systems are available. The effect of shading on energy consumption and comfort can be studied by a simulation model. Figure 5 demonstrates how the house in Herzlia behaves when its windows are shaded and intercept about 50% of the direct radiation. The maximal temperature dropped from 29 °C (maximal temperature obtained with ventilation only) to 26 °C. Shading devices have optimal sizes because they should be maximal in size to prevent high temperatures in summer and should be minimal in size to provide m a x i m u m winter insolation. Hence, a year-round calculation for finding the o p t i m u m shading devices should be carried out (cf. Shaviv [11]). In most cases, we find that external shading is the optimal one. External sunshades enhance the appearance of a building and give strong architectural expressions. Consequently, architects use sunshades to design an interesting external appearance. A known example is the BriSoleil designed by Le Corbusier. Most Israeli public houses are designed with different external sunshades. However, the geometry of these sunshades is not always the most suitable to prevent direct summer radiation and to allow winter insolation. A m e t h o d for the design of sunshades was developed by Shaviv [ 1 7 - 20]. The m e t h o d provides the architect with a special computer produced nomogram. The nomogram contains the entire family of possible sunshades that satisfy a given shading requirement, given window shape and orientation, and given hours and months when shading is required (see Fig. 6). The nomogram allows the architect to design sunshades of almost any shape (see Fig. 7), i.e., fixed shading [17] or movable ones [20], and can be obtained on a screen or a printer of a personal computer. The m e t h o d is based on the search for the length of an imaginary pole situated on the window and sufficiently long to cast the shade during the required period. By checking all possible locations for the pole, the entire window area is covered. The search for the minimal length of the pole can be calculated NEW WINDOL~?IF YES PRIN? 1,IF nO PRIhT o 7 1 GIUE DATR~(W,O,S,90,E,I,O,N,270) WALL AZ, R(WIDTH), H(HIEGHT), NA, NH, MI,M2,DM, HI, H2, DH, SCALE ? 315s 140s 1 4 0 s 3 , 3 . 4 ~ , 1,12, 16.1,85 SPECIAL OPTION,'~ IF YES PRINT I , I F NO PRINT e ~e WEST.e.00 WALLAMG, A SOUTH,ge.e 31S.000 140o0e0 ,4o.0eo H • HA NH 3 3 PIONTHS • 4 ~ HOURS • 12 16 EAs?.lBo.ee NORTH.270.00 1 i MAXIMUM SHADES NEEDED FOR WINDOa 140.oe/140.oe 916 616 616 IN WALL 31S.00 616 063.32114.28122.35 / 0 91, 0 g 0 57.1, ,1, 9 0 0 ,1.18 ,8.18 ,1, ~ Fig. 6. A n o m o g r a m shades. 91, 0 0 g ~'~ /~ / 0 0 a n d a t a b l e for d e s i g n i n g s u n - L30 130 83 65 56 28 65 43 22 28 22 Fig. 7. A n o m o g r a m f o r a n o r t h e a s t w i n d o w ( A ) a n d s e v e r a l g e o m e t r i c a l s o l u t i o n s (B - L). over a given day, given m o n t h or over the entire year. Besides providing the proper summer shading, the m e t h o d accounts for winter insolation and provides architecturally satisfactory forms (see Fig. 8). 61 120.13 @ 10 174,90 9 9 218-28 g 8 25?.]3 g 8 257.33 g 8 318,10 9 • 381,7~ g ? 42q.24 g 7 t29.26 9 • 100,61 8 !0 14SETS 9 g 214,4.4 9 8 214,44 g 8 254*48 g • ~'I B * I O g ? 357*70 9 • 357,?0 G ? 3 5 ?9 . ? 0? 80*49 g I0 116,00 9 9 171,55 9 5 190,86 9 ? 254. 48 q • 286,16 g • 286,16 9 ? 286ol8 g 7 286.16 9 • 60,37 9 i0 log*14 9 8 128,66 9 8 190.86 g ? 214,62 9 • 214,62 9 • 214.82 9 • 2 1 4 , 6~: 9 • 21 ~* .62 9 ? 40.24 g IO 85,78 @ 8 127,24 g ? 143.0S ¢; 7 143,08 9 • 163.08 9 • 143.08 9 • 143,08 9 ? 165,08 g ? 20.12 q 10 63.62 q ? 71,54 g • 71.54 g • 71.54 9 • 71,54 g 7 ?leSt 9 • •1.54 g • ?1 , 5 4 9 • 0.0 q 0 0 eO 9 0 O. 0 g 0 0,0 g 0 O. 0 9 0 0.0 g 0 0.0 g 0 000 g 0 0.0 9 0 257 U 3 5 A 286 214 A B C D B Jl II[llllllllhjlltlllllllllullltlllllh, o UIIIIVIIIlllIIIIIJrHIIIIIIIIUlIIIIUI 1 l ] !~ II !il [ II 230 "s its /~q ;I Q : C ;575. 575. $75.S75L [3 Fig. 8. A t a b l e o f d i s t r i b u t i o n o f 1 in t h e field o f t h e s o u t h e a s t w i n d o w (A). B (A - D) are d i f f e r e n t r e q u i r e d s h a d i n g devices f o r e q u a l w i n d o w s ( w h i c h is h a l f t h e area o f t h e given w i n d o w s h o w n in (A), b u t w i t h d i f f e r e n t p r o p o r t i o n s ) . C a n d D are t w o designed a l t e r n a t i v e s b a s e d o n p r o p o r t i o n s BD a n d B A. 62 ORIENTATION The orientation of the building affects the in-house thermal comfort in two ways: (1) by the amount of radiation incident on the building envelope; (2) by the local wind pattern, i.e. ventilation and infiltration. Olgay [12] describes the Baltimore m e t h o d which provides simple weighting of radiation and possible building cooling b y the dominant wind. The dominant wind is defined on the basis of intensity and direction. However, using a simulation model enables us to achieve an accurate result without resorting to annual or seasonal averages of climatic factors. Let us discuss in more detail the first point mentioned above, i.e., the influence of the insolation in the different orientations of the building on its thermal performance. The envelope of the building can absorb the solar radiation, and therefore is affected by its orientation. However, if the external wall is white, and well insulated, the effect is reduced. On the other hand, we expect a strong effect of the orientation of the openings on the insolation of the building, and consequently on the thermal behavior. Olgay determines the optimal building pro- 3 4~3 6 i : ' ~ ; : : : t ; : : : : : : : : : ; ; * ' : portions by considering energy conservation as controlled by heat conduction through the envelope on the one hand, and by solar exposure on the other hand. He proposes to present the effect of the temperature and radiation on buildings by means of a "sol-air" temperature. As a consequence, the square house with the minimal envelope area (per given floor area) is not the optimal one. The optimal proportions are somewhat elongated along the E-W axes, so that the southern and northern elevations are longer than the eastern and western ones. The exact proportions of the different elevations, as well as the orientation, depend on the local climatic conditions. No use of sol-air temperature is required when a detailed simulation model is applied. The handling of every factor like temperature, insolation, etc., is accurate and can be studied simultaneously. An example of such a calculation is provided in Fig. 9, where only the influence of radiation is included (and no wind). The house in this example is identical to the one shown in Fig. 2. Recall that the maximal temperature reached 32.5 °C. The house is unshaded and unventilated. However, it was rotated by 180 °C, namely, the large windows : : t : : : t ~ ; ~ : : ~ : "~'~: : : ~ ' : ; ; : ; ; : : : : : ; ~ : t ~ ; "~ : : ~ : : : : : : ;'~ 4~ "''''~°°° .....Ba,.,oo .....eooQ BO~O ° ° " m OOoDo O11°°°D° • OOOoD O~ODoOD ~m "ODOI D e O O DQ °a a" 000 o. 000 ~8~ OOOQ • 000 • OD • 3 8000 2 7000 t~ 6000 30~ 26+ (2 u. • • " I 09 0 0 0 6 • 24- o ~22 *.jJ"lJ,,,~." • ,,,~., •,, 20 m M ' IB . ~ , -, m, S~l~i,,.r, , ." .~*"T;' . + + | • " Itl ,', .[,II • • ,', . m " 24 30 2000 +4 IOOO II n 18 "+3 M ii 12 p, M M I0 3000 ,', q ii 6 III -2 m m M E+ ~- ~ m . Jl 14 ~iSl l l ] " ""+]I" i " ,+"T;' ~ |II ~ -I ~ 4 0 0 0 C~ ~ M " • m,V~'TT, M m ' ,t--/ • ,Is,,,,,.r., . 16. 12 .w"',J,,,J • ~ I 36 42 " 48 54 60 66 "" 5 0 72 Hour Fig. 9. The thermal p e r f o r m a n c e o f the unshaded and u n r e n t e d house s h o w n in Figs. 2 and 3, rotated by 180 °. 63 TABLE 1 Comparison of energy consumption of four identical apartments with different orientations (kWh) Specification S &E S&W N&E N&W Total heating consumption 2388 2735 3729 3981 Total cooling consumption 4843 5087 4599 4749 Total energy consumption 7231 7822 8328 8730 of the living room faced the east. The different orientation is sufficient to reduce the maximum temperature by 2 °C, i.e., from 32.5 ° to 30.5 °C. The building is a high thermal mass structure and hence has a long time lag and the m a x i m u m temperature occurs in the afternoon. If at this time the building receives a lot of direct or diffuse radiation owing to the large west-facing windows, the temperature will be higher than if the large windows were facing the east. In the last case, the two factors operate in opposite directions and therefore the m a x i m u m temperature is reduced. Another check was carried out on a typical Israeli apartment house complex with four apartments per story, designed in a suburb of Tel Aviv [21]. Each of the apartments has two orientations: S & E , S & W , N & E and N & W. The energy consumption in all four apartments was calculated in winter and summer. The results are given in Table 1. The results indicate that minimal energy for winter space heating is required by the S & E apartment. On the other hand, minimal energy for summer cooling is required by the N & E apartment. However, the minimum annual energy consumption was found for the S & E apartment. Obviously, the results m a y change with the climatic conditions. Similarly, a check on the preferred orientation for passive solar collectors was carried out [22]. The conclusion is that the c o m m o n belief that the exact south is the optimal solution for winter and summer is correct for a climate like that of Tel Aviv. However, in places where the summer is particuarly h o t and the winter n o t too severe, the SSE orientation is a better one. Furthermore, if winter is severe and summer mild, the optimal orientation is SSW. These results are particularly true for massive buildings with large thermal masses. The maximal temperature in such buildings is obtained in summer in the afternoon hours. A southwest window allows direct solar insolation during these hours and hence, the maximum temperature is highest. If mechanical means are used, a room with a southwest window will need a larger system than one with a southeast window. On the other hand, if the winter is very cold, and the building has a large thermal mass, the thermal mass will store the energy for later use. A southwest window will collect more energy which is stored for evening and night use, and thus economize on night heating. The above results were obtained by considering the effect o f radiation only. No reference to the wind was made. If the dominant wind in summer is from the west, it can clearly affect the above conclusions, especially if the building under consideration is an office building where good thermal performance in summer should be achieved only in the mornings and until 15:00. COLOR The color of the building's envelope affects the absorption of short-wave radiation. Black paint absorbs most of the short-wave radiation and heats up, while white paint reflects most of the radiation and hence remains cooler. Table 2 summarizes the results obtained for a building with a 20-cm thick concrete roof painted black or white. The results indicate a 2 °C difference in the maximal temperature in favor of the white roof. Similar results were obtained for the average air temperature. The effect of the color of the roof decreases with the increase in roof insulation or roof shading. However, the effect of the color of the roof is more significant than that 64 TABLE 2 Comparison of temperatures in summer of two identical buildings with black or white roof (°C) Specification Black roof White roof Ambient temp. Average room air temp. Maximum room air temp. Minimum room air temp. Temp. swing Time lag (h) Maximum ceiling temp. Minimum ceiling temp. Temp. swing 27.3 30.0 25.0 5.0 4 35.1 23.2 11.9 26.5 28.1 24.9 3.2 5 28.3 25.2 3.1 27.0 31.2 32.0 8.2 of the external walls; and of all outside walls, the color o f the east and west walls are more effective than the southern and n o r t h e r n ones. This is because the last two walls are the least insolated in summer. The f r eq u en t use of white in Mediterranean architecture as a means to reduce summer heating stems from the fact that the houses are n o t well insulated. The white paint creates excessive glare which annoys pedestrians and creates man y reflections between buildings. This undesirable shortcoming of the white color can be overcome to some e x t e n t by painting white the roofs and the eastern and western walls only. Darker colors can be used for the n o rth er n and southern walls, since t he y receive only small amounts of direct solar radiation in summer. Obviously, pr ope r insulation or shading of the buildings allows complete freedom in the choice of outside colors. NOCTURNAL RADIATION One o f the mos t promising possibilities for cooling in h o t climates is by nocturnal radiation cooling. The emission by any object is proportional to the t e m per a t ur e to the f our t h power. The incident radiation from t h e {night) sky can be represented by an equivalent radiation (or brightness) temperature. If the equivalent sky radiation t e m p e r a t u r e (to distinguish from the ambient air t e m per a t ur e) is lower than the surface t e m p e r a t u r e of the outside envelope of the building, t he net heat exchange between the building and the sky will induce its cooling. Shaviv and Shaviv, 1977 [7, 9] suggested a definition of the radiation t em pe r at ure 27.0 31.2 8.2 of the surrounding {Trad) as the t e m p e r a t u r e of a black b o d y which emits a radiative energy flux equal t o t hat of the sky and the ground incident on the wall. Since the r o o f and the vertical walls see different parts of the sky and the ground, t h e y will exchange heat with black bodies of different temperatures. Table 3 provides results for Tra d as seen by a r o o f and a vertical wall. The temperatures are based on radiative fluxes measured by Manes [23] and are calculated according to Shaviv and Shaviv. The differences between the r o o f and the walls are remarkable. At night the difference between the ambient air t em perat ure and the radiant t e m p e r a t u r e may be 28 °C for the r o o f and 24 °C for a vertical wall (if no ot her building is around). Thus, the sky behaves as a very cool heat reservoir. The situation reverses at midday and the vertical wall sees a radiation t e m p e r a t u r e which is higher than the ambient air temperature, while the radiative t em perat ure seen by the r o o f is still lower than the air temperature. Hence, the r o o f can always cool faster than the vertical walls. The latter can even be heated by the radiation, depending on the light reflection properties, of the ground. An experimental and theoretical study was carried out recently at Lawrence Berkeley L a b o r a t o r y by Martin Marlo [24], and Berdahl and Fromberg [25], and in San Antonio by Clark and Allen [26]. Measurements of long-wave radiation emitted by the sky were carried out at six different locations in the U.S.A. A simple correlation function was found between the long-wave sky emissivity and the t e m p e r a t u r e of the dew point. A good agreement was found between these results and those of Shaviv. 65 TABLE 3 Results o f Trad for August, as seen b y a r o o f a n d a vertical wall (°C) Hour Ambient temp. Tra d r o o f Tra d wall 3.00 6.00 9.00 12.00 15.00 18.00 21.00 24.00 23.2 23.9 29.4 31.2 30.7 28.4 25.3 24.0 --3.5 4.9 13.6 18.6 11.3 6.9 --3.5 --3,5 --0.4 6.9 26.1 35.4 27.8 14.6 1.5 0.3 When using a polyethylene sheet to protect the radiative surface from cooling by convection, dew might form on its surface and change the long-wave transmission properties, as was found by Givoni [27]. If the problem is a severe one, a sufficiently tilted r o o f may be advantageous. The direct way to use the night sky cooling can be found in traditional architecture in h o t countries. Roofs were designed so that people could sleep on them at night and cool down. A more modern way to use the cold night sky indirectly is by a solar roof pond, as was designed by Hay in the Atascadero Building in California. In solar r o o f ponds, the thermal mass which is the water pond is cooled at night by the nocturnal radiation. The cold thermal mass can keep the in-house temperature low during the day. THE INFLUENCE OF THE BUILDING'S MATER I A L S ON ITS P E R F O R M A N C E Two physical properties of the building's materials affect its thermal performance, firstly by the ability to conduct heat, and secondly by the ability to store heat. It is important to have an insulated external envelope (poor heat conductor). The outside walls also have the role to delay the transfer of heat from the outside into the inside space. The implication of this property is t h a t the building will heat slowly in summer and will reach its maximal temperature only during the late hours when the outside air temperature is low. With good ventilation, the heat which flows from the walls into the inside air can be removed. The capability to store energy also helps in winter, since energy can be stored in the walls from one sunny winter's day to the next cloudy day; thus wall storage can slow down the cooling and help overcome relatively severe short cold periods. However, simulation runs on the computer show that for a climate like that found in Haifa or Tel Aviv, the thermal mass is even more important in summer to reduce maxim u m in-house temperature without mechanical means, than it is needed to store the passive solar energy in winter. Walls are usually made of several layers with different thermophysical properties. Attention should be paid to the correct order of the layers. Simple, steady state calculations ignore the effect of order. However, dynamic models indicate (see Fig. 10) that a house made of walls in which the insulation is on the outside with a layer of thermal mass on the inner side cools slower than a house in which the order is reversed. When the insulating layer is on the inside, it reduces the effect of the thermal mass of the walls. Only little energy storing is possible this way. The thermal mass can be located in the outside envelope or in internal elements like internal partitions, floors and ceilings, stairs, etc. The effect of the location of the thermal mass on the behavior of the building was checked in the following calculations: four identical structures were evaluated under identical conditions. The only differences we re: Case A : Light external wall and light inner partitions (L-L). Case B: Heavy outside wall and light inner partitions (H-L). The thickness of the external wall was determined in such a way as to yield 66 TEMPERRTURE '° • • - . ; ~tqlllll~ilPlPk I 2 bRV5 ~ Fig. 10. The in-house temperatures obtained in an unheated building. The dark line denotes the behavior of the building with insulation on the outside while the light line is with insulation inside the external wall. TABLE 4 Comparison of energy consumption of four identical buildings with different amounts of thermal mass located in the outside wall and/or inner partitions (kWh) Case Specification (external-internal) A L-L B H-L C L-H D H-H Total heating consumption Total cooling consumption Total annual consumption 1034 805 1839 532 621 1153 584 698 1282 326 575 901 a heat conductivity coefficient (U-value) equal to that of the light wall in case A. Case C: Light outside wall as in case A and heavy inner partitions (L-H). The heat storing capability of the inner partition wall is equal to that of the heavy outside wall of case B. Thus, cases B and C have the same U-value and thermal mass. The only difference is that in case B the thermal mass is located in the external envelope while in case C it is in the inner partitions. Case D: Heavy outside wall as in case B and heavy inner partitions (H-H). This means t h a t the U-value of the external walls is the same as in all other cases, but the thermal mass is twice as much as in cases B and C. The results are shown in Table 4. We can conclude that: (a) summer and winter energy savings are larger when the thermal mass is in the outside walls {case B compared with case C). However, the differences are rather small. Therefore, if traditional design calls for light insulated structure as the solution for the building skin, the thermal mass can be arranged inside the building, by concrete slabs or walls, and by using heavy materials for floor covering. (b) the addition of thermal mass improves the performance of the building in winter and in summer. The marginal saving, however, decreases with the increase of the thermal mass. The effect of thermal mass does n o t appear at all in steady state calculations. Only dynamic models include the thermal mass and can predict its effects, including the effects of the ratio of surface to volume and the distribution of thermal mass in the house, possibilities n o t discussed in this work. Examples of buildings with thermal mass can be found in cliff dwellings in New Mexico or Mesa Verde in Colorado, where the cave openings are to the south. Another example is the Indian Pueblos t h a t were built from heavy adobe bricks. The Mediterranean building is mostly heavy in order to preserve the night coolness during daytime. We also find t o d a y that there is a tendency to return to the idea of cave dwelling as is seen in the earthsheltered houses. 67 THE STACK EFFECT Differences between internal and external air temperatures induce air m o t i o n when the hot air rises. When this m o t i o n is used for ventilation, it is called the stack effect, or thermal chimney. Ancient buildings in h o t countries used to have high ceilings to provide space for the h o t rising air. The use of domes with openings for release of the h o t air increases the effect. The tradition in countries with h o t and dry climates is to build the houses very close to one another and with an internal court. The proximity of the outside walls reduces solar insolation of the outer envelope, while the inner court can serve as a passive solar element for cooling. At night, the rooms which face the inner court are kept open. The ambient air cools, sinks into the court and penetrates into the rooms. This is known as the katabatic cooling [28]. The court itself cools also by radiative exchange with the sky. The buildings contain large thermal masses which cool at night by the cold air that penetrates the buildings. The rooms are kept closed during daytime. The house, which was cooled during nighttime, preserves its comfortable temperature. The temperature of the house starts to rise in the evening and as the ambient temperature is already low at this time, ventilation is required. However, the h o t air in the court that was heated during the day rises and creates air m o t i o n which ventilates the house quickly. This p h e n o m e n o n is known as the anabatic cooling [28]. Note that houses with inner courts are good only in h o t and dry climates. In a h o t ht~mid climate, the distance between the houses must be large so as to allow the prevailing wind to reach every house and ventilate it. The stack effect is used in a Trombe wall. Due to the black color of the wall, the air between the wall and the glazing heats up and rises. If the upper part of the Trombe wail is open to the outside and the lower part of the Trombe wall is open to the room, the reduced pressure will suck air from the room into the Trombe wall. Open windows on the northern cool side of the house will let fresh cool air penetrate the building. Another form of using the stack effect can Fig. 11. A t h e r m a l c h i m n e y e f f e c t to i n d u c e ventilation. be seen in the greenhouse windows used by Shaviv [29]. The windows can be opened halfway so that the upper and lower parts are open (see Fig. 11). The air inside the greenhouse heats up and rises. This motion reduces the air pressure -- the Bernoulli effect. Consequently, air is dragged from outside the house. When the northern windows are open, a pleasant ventilation is achieved. U R B A N O U T D O O R SPACES - - SITE L O C A T I O N Town planning and the relation between buildings have a non-negligible effect on the thermal comfort inside the building because of the following reasons: (1) One building can shade its neighbor and thus deprive it of winter passive solar heating, while creating a favorable summer microclimate. This feature explains the very dense building in h o t climates. (2) The dominant wind can be stopped by building perpendicular to the prevailing wind direction. Alternatively, the houses can be directed to the dominant breeze if so desired. ( 3 ) T h e ambient air temperature in a densely built area can be quite affected by the total thermal mass of the buildings. Los (1983) has developed a simulation model to investigate the interplay between m a n y site location factors such as ground and air temperatures, wind velocity, sky radiation, etc. and their effect on the temperature in the space between buildings [30]. The model allows a study to be made of how the proportions of the space module, and how the shape factors of the space, i.e. the rela- 68 tions between the heights and distances between buildings, affect the microclimate. The results are presented in a form of matrix of temperatures, which represents the temperatures at ground level and various heights in the space between buildings as well as the outside surface temperature of the buildings. The above reference discusses several examples of fundamental unit layouts. A design tool to help in achieving the allowed location of buildings and their recommended maximum heights so as to ensure solar rights to all buildings was developed by Arumi in 1979 [31]. The model provides a nomogram of all possible solutions to the problem of how to place a given building in the space between buildings, without shading it or the existing buildings. The approach recalls that of Shaviv, 1975 [17] in the design of sunshades. CONCLUSIONS Extensive work has been devoted to the analysis of the design of a single building, including development of very accurate dynamic simulation models. However, very little has been accomplished from an urban planning point of view, especially in the area of developing computer models for accurate analyses. The pioneering work of Los [30] should be continued and further developed. Another point to mention is that most design tools are evaluation tools rather than design tools, in particular, the c o m p u t e r codes that were developed. One should first design and then use these codes to evaluate the thermal performance of the buildings and their surroundings. The different approach to the design and evaluation models can be found in the works of Shaviv [17] and Arumi [31]. In these works, the computer is used to first develop a design tool and then the design tool is applied in creating a better building or neighborhood from the thermal point of view. However, in both works mentioned above, only a few design parameters were dealt with. These pioneering works and approaches should be followed and further developed to include most of the design parameters that influence the thermal performance of the buildings as a comprehensive computer aided design tool and not only as a simulation tool that evaluates a given design. REFERENCES 1 J. A. Clark, ESP. System Documentation Set, Abacus Publication, University of Strathclyde, 1982. 2 R. D. McFarland, PASOLE: A General Simulation Program for Passive Solar Energy, Informal Report, LA-7433-MS, Los Alamos Scientific Laboratory, NM, 1978. 3 Building Loads Analysis and System Thermodynamic Program (BLAST), AD-AO48982/3ST 1977, AD-AO48734/8ST 1977, Construction Engineering Research Lab., U. S. Army. 4 S. A. Klein et al, T R A N S Y S . A Transient Simulation Program, EESR Report 38, Solar Energy Lab., Univ. of Wisconsin-Madison, 1977. 5 F. N. Arumi, Dynamic Energy of Building (DEROB), Numerical Simulation Lab., School of Architecture, Univ. of Texas, Austin, TX, July 1979. 6 L. Palmiter, T. Wheeling and R. Judkoff, SERIRES, Solar Energy Research Institute - Residential Energy Simulator, Solar Energy Res. Inst., CO, 1982. 7 E. Shaviv and G. Shaviv, A Model for Predicting Thermal Peformance o f Buildings, WP ASDM-8, Faculty of Architecture, Technion, Israel, 1977. 8 E. Shaviv and G. Shaviv, Modelling the thermal performance of buildings, Build. Environ., 13 (1978) 9 5 - 1 0 8 . 9 E. Shaviv and G. Shaviv, Designing of buildings for minimal energy consumption, Computer Aided Design J., 10 (4) (1978) 239 - 247. 10 E. Shaviv, Building design for passive energy conservation, Proc. Int. Conf. on the Application o f Computers in Architecture, Building Design and Urban Planning, PARC 79, Berlin, 1979, Online Conference Ltd., Uxbridge, U.K., pp. 135 - 144. 11 E. Shaviv, On the Determination o f the Optimal Shading Factor o f Windows, CAD 80, IPC Science and Tech. Press, Brighton, 1980. 12 V. Olgyay, Design with Climate, Princeton Univ. Press, Princeton, NJ, 1963. 13 A. Bowen, Design guidelines on lateral airflow through and around buildings, in S. Yannas (ed.), Proc. 2nd Int. Conf. on Passive and Low Energy Architecture (PLEA), Crete Pergamon Press, 1983. 14 A. Salloum, "EL RAWASHIN" of Jeddah, Proc. 2nd Int. PLEA Conf., Crete, Pergamon Press, 1983. 15 S. Elawa, Housing Design in Extreme Hot Arid Zones with Special Reference to Thermal Performance, Dept. of Building Science, University of Lund, Sweden, 1981. 16 M. N. Bahadori, Passive cooling systems in Iranian architecture, Sci. Am., 238 (2) (1978) 144 - 154. 17 E. Shaviv, A method for the design of fixed external sun shades, Build. Int., 8 (2) (1975) 121 - 150. 18 E. Shaviv, The Determination o f the Form o f Windows and Sun Shades in a Hot Climate, CAD 78, IPC Science and Tech. Press, Brighton, 1978. 19 E. Shaviv, Sunshades as a passive cooling element, Proc. 5th National Passive Solar Conf., AS/ISES, 69 Philadelphia, PA, 1981. 20 E. Shaviv, A design tool for determining the form of fixed and movable sunshades, A S H R A E Annual Conv., Atlanta, 1983. 21 E. Shaviv, Energy Conservation in a Residential Building in Ra'anana, A report submitted to the Israeli Ministry of Energy, 1979, (in Hebrew). 22 E. Shaviv, The influence of the orientation of the main solar glazing on the total energy consumption of a building, Solar Energy J., 26 (5) {1981) 453 - 454. 23 A. Manes, A. Teitelman and I. Fruehling, Solar Radiation and Rad&tion Balance, Series A, No. 25, Ministry of Transport, Israel Meterological Service Res. Div., Bet Dagan, 1970. 24 M. Martin and P. Berdahl, Characteristics o f Infrared S k y Radiation in the United States. Lawrence Berkeley Laboratory Report, University of California, Berkeley, 1983. 25 P. Berdahl and R. Fromberg, A n Empirical Method for Estimating the Thermal Radiance o f Clear Skies, Report #LBL-12720, LBL, Berkeley, CA, 1981. 26 G. Clark and C. P. Allen, The estimation of atmospheric radiation for clear and cloudy skies, Proc. 2nd National Passive Solar Conf., AS/ISES, Philadelphia, PA, Vol. 2, 1978, p. 676. 27 B. Givoni, Cooling of buildings by passive systems, Proc. 1st. Passive and Hybrid Cooling Conf., AS/ISES, Miami, FL, 1981, p. 558. 28 K. Talib, Review of climatic design concepts and details in traditional architecture in various climatic z o n e s - - S a u d i Arabia, in S. Yannas (ed.), Proc. 2nd Int. PLEA Conf., Crete, Pergamon Press, 1983. 29 E. Shaviv, Window sunspace systems as heating and cooling components, Proc. 2nd Int. PLEA Conf., Crete, Pergamon Press, 1983. 30 S. Los, The design of urban outdoor space: A bioclimatic approach, Int. Conf. on Non-Conventional Sources, Trieste, Giugno, Trieste, 1983, pp. 20 - 25. 31 F. N. Arumi, Computer-aided energy design for Buildings, in D. Watson (ed.), Energy Conservation Through Building Design, McGraw-Hill, 1979.