Energy and Buildings 32 2000. 8993 www.elsevier.

comrlocaterenbuild

Ventilation impact of a solar chimney on indoor temperature fluctuation and air change in a school building
Joseph Khedari ) , Boonlert Boonsri, Jongjit Hirunlabh
Energy Technology Diision, School of Energy and Materials, King Mongkuts Uniersity of Technology Thonburi, Bangmod, Rasburana, Bangkok 10140, Thailand Received 15 June 1999; accepted 12 September 1999

Abstract The aim of this research was to investigate, experimentally, both the feasibility of a solar chimney to reduce heat gain in a house by inducing natural ventilation and the effect of openings door, window and inlet of solar chimney. on the ventilation rate. The study was conducted using a single-room school house of approximately 25 m3 volume. The southern wall was composed of three different solar chimney configurations of 2 m2 each, whereas, the roof southern side included two similar units of 1.5 m2 each of another solar chimney configuration. Those configurations were built by using common construction materials. Experimental observations indicated that when the solar chimney ventilation system was in use, room temperature was near that of the ambient air, indicating a good ability of the solar chimney to reduce houses heat gain and ensuring thermal comfort. The air change rate varied between 815. Opening the window and door is less efficient than using solar chimneys, as temperature difference between room and ambient was higher than that obtained with solar chimneys. q 2000 Elsevier Science S.A. All rights reserved.
Keywords: Natural ventilation; School building; Tropical region; Field testing; Thermal comfort

1. Introduction Natural ventilation may result from air penetration through a variety of unintentional openings in the building envelope, but it also occurs as a result of manual control of buildings openings doors, windows. or when a building is equipped with a ventilation system like natural ventilation solar chimneys. Air is driven inrout of the building as a result of pressure differences across the openings, which are due to the combined action of wind and buoyancy-driven forces. Today, natural ventilation is not only regarded as a simple measure to provide fresh air for the occupants, necessary to maintain acceptable air-quality levels, but also as an excellent energy-saving way to reduce the internal cooling load of housing located in the tropics. Depending on ambient conditions, natural ventilation may lead to indoor thermal comfort without mechanical cooling being required.
) Corresponding author. Tel.: q 66-66-2-4708625; fax: q 66-66-24708623.

Many configurations of solar chimneys have been used widely in the past and many are being developed again today w14x. In our school, several studies were carried out in this topic since 6 years w58x. Four configurations of solar chimney, each with approximately 2 m2 of surface area, were built by using common construction materials: The Roof Solar Collector RSC. w5x, the Modified Trombe MTW. w6x, the Trombe Wall TW. w7x and the Metallic Solar Wall MSW. w8x. The description and dimensions of each configuration are given in Fig. 1 and Table 1. These configurations were integrated into the south-facing roof and facade of a single-room building of approximately 25 m3 volume. The room was located at the 12th floor of the School of Energy and Materials. Tests were run separately. It was found that all of these devices allow, on the one hand, to minimize the fraction of the solar flux absorbed by the dwelling acting, therefore, as a very good insulation material and, on the other hand, to induce a natural ventilation which improves thermal comfort. However, the resulting number of air changes ACH. was rather low varying depending on climate conditions between 3

0378-7788r00r$ - see front matter q 2000 Elsevier Science S.A. All rights reserved. PII: S 0 3 7 8 - 7 7 8 8 9 9 . 0 0 0 4 2 - 0

90

J. Khedari et al.r Energy and Buildings 32 (2000) 8993

Fig. 1. Dimensions of the single-room solar house.

Fig. 2., air at both inlet and outlet of four configurations of solar chimney Figs. 35.. Room temperature was measured at 27 3 = 9. positions, between 50 cm above floor to 2 m as shown in Fig. 6. Air velocities at the same position were measured by hot wire anemometers. The door of approximately 2 m2 surface area has two air grills located at the upper and lower parts of the door. Investigations were made for the several positions upper openedrlower closed, upper closedrlower opened and both opened.. The ambient temperature, wind velocity and wind direction were measured by an environmonitor model of RJ1412HPL; Type: 4X. located about 20 m from the house. The solar radiation was measured by an Eppley precision spectral pyranometer and pyranometer with shading ring w9x. Experimentation started at 9 AM and ended at 5 PM by recording data at 30-min intervals.

to 5, which is not sufficient to completely satisfy room occupants as a higher number is required above 20 ACH. for houses without any mechanical cooling device. So, to increase the ACH, all devices have to act simultaneously. This is the objective of this work: running the four configurations and discussing the resulting thermal comfort by means of three parameters: indoor air temperature and velocity and the number of ACH. However, as tests would have to be done on different days, a relativeness index was introduced. This index is the temperature difference between average room temperature and ambient. In addition, the effect of openings door, window, and position of free inlet of wall solar chimneys was also investigated.

3. Results We remind that the ability of the solar chimney to offer thermal comfort is expressed by the difference dT . between indoor and ambient temperature. The lower dT is, the higher the rate of ventilation, meaning higher ACH. The relativeness index was introduced in order to have a tool for comparison of performance as testing this naturally ventilated room was undertaken on different days corresponding to different ambient conditions as shown Fig. 7. This, of course, cannot overcome this problem but it will allow us to formulate general and subjective conclusions. As mentioned in Section 1, the focus of this paper is on to see how effective such devices would be in aiding ventilation and reducing interior air temperature.

2. Methodology Thermocouples of type K were used to measure the temperature of incoming ambient air through doorrwindow
Table 1 Configurations of solar chimney Configuration dimension, cm. TW W: L:T :100:200:25.9.

Detail of materials from outside to inside of the house. Glass Air gap Masonry Fibre glass Plywood Masonry Air gap Gysum board Glass Air gap Zinc sheet Fibre glass Plywood CPAC monier Air gap Gypsum board

Dimension cm. W: L:T :100:200:0.5 W: L:T :100:200:14.5 W: L:T :100:200:8 W: L:T :100:200:2.5 W: L:T :100:200:0.4 W: L:T :100:200:8 W: L:T :100:200:14 W: L:T :100:200:0.9 W: L:T :100:200:0.5 W: L:T :100:200:14.5 W: L:T :100:200:0.07 W: L:T :100:200:2.5 W: L:T :100:200:0.4 W: L:T.rpiece:33:42:1.5 W: L:T :200:150:14 W: L:T :200:150:14

MTW W: L:T :100:200:22.9. MSW W: L:T :100:200:17.97.

RSC W: L:T :200:150:16.4.

J. Khedari et al.r Energy and Buildings 32 (2000) 8993

91

Fig. 2. Measured positions at the door and window. ^. Air temperature; v . air velocity.

3.1. Indoor temperature fluctuation Fig. 8 shows that when all openings were closed door and window grills and free inlet of solar chimney., the temperature difference between room average of 27 points, see Fig. 6. and ambient increased rapidly with time to reach a maximum of 68C around 5 PM. The corresponding

indoor temperature is, therefore, too high at about 408C. It should be pointed out that after sunset around 6 PM. this temperature difference is still important; because of the heat stored during daytime, leading to uncomfortable feeling of warmth at the beginning of the evening. This will force the occupants to turn on the air conditioning, with a very high cooling load. When the rooms common openings door and window. were open which is commonly known as one side ventilation this temperature difference decreased but is still relatively important, around 48C. When the solar chimney opening was opened wall chimneys opened at 1 m above floor, Fig. 5., corresponding to a total of 6 m2 of solar chimney surface area 3 m2 of RSC and 1 m2 of each wall configuration., this temperature difference decreased. The maximum difference was about 38C which occurred around 4 PM. The same observation was obtained when the wall chimneys were opened at 0.04 m above floor Fig. 4., corresponding to a total of 9 m2 of solar chimney surface area 3 m2 of RSC and 2 m2 of each wall configuration.. In other terms even though tests were undertaken on different days the ventilation induced by solar chimneys reduced room overheating relativeness index. by about 50%.

Fig. 3. Measured positions on the RSC. ^. Air temperature; v . air velocity.

Fig. 4. Measured positions on each type of walls with inlet openings at 0.04 m above floor. ^. Air temperature; v . air velocity.

92

J. Khedari et al.r Energy and Buildings 32 (2000) 8993

Fig. 7. Ambient conditions during the days of experiments.

flowrate. The small difference between measured incoming air and outgoing air indicate good accuracy of measurement. The corresponding induced number of air changes varied between 10 to 15. This quite high air change rate led to a lower indoor temperature, resulting from a reduction in the rate of heat stored within the house.

Fig. 5. Measured positions on each type of walls with inlet openings at 1 m above floor. ^. Air temperature; v . air velocity.

3.2. Induced number of air changes Due to the buoyancy driven force, the air is continuously induced through the house with a rate that depends, mainly, on intensity of incident solar radiation. Fig. 9 shows an example of hourly variation of induced air

Fig. 8. Comparison of the difference between average room and ambient temperatures for different room conditions. I. All openings closed on May 14, 1998; ^. door and window opened on June 29, 1998; =. solar chimney SC areas6 m2 ., lower door and window opened on July 8, 1998..

Fig. 6. Positions of air temperature and velocity measurement inside the house and the vertical and horizontal planes.

Fig. 9. The induced incoming and outgoing air flowrate through the room SC area s6 m2 , lower door and window opened on July 8, 1998..

J. Khedari et al.r Energy and Buildings 32 (2000) 8993 Table 2 Indoor air temperature and velocity fluctuation inside the house SC area s 6 m2 , upper door opened on June 20, 1998. Plane number Time: 10 AM Temperature 8C. Vertical 1 V1. Vertical 2 V2. Vertical 3 V3. Horizontal 1 H1. Horizontal 2 H2. Horizontal 3 H3. 31.92 31.90 32.01 32.01 32.02 31.80 Velocity mrs. 0.06 0.04 0.02 0.02 0.04 0.06 Time: 2 PM Temperature 8C. 35.28 35.37 35.47 35.26 35.41 35.46 Velocity mrs. 0.09 0.05 0.03 0.02 0.06 0.08

93

3.3. Thermal comfort For a free running building, the two main thermal comfort parameters are the temperature and velocity of room air. From Section 3.2, we found that with 69 m2 surface area of solar chimney, the indoor temperature is about 238C above that of ambient, which is relatively acceptable provided that there is sufficient air motion to decrease resultant temperature. Table 2 shows that the average air velocity near the door plane V1, see Fig. 6. was the highest one, because of the incoming outdoor air through the upper door opening the lower door and window grills were closed.. At plane V3, the air velocity was too small as there are no openings. Along horizontal planes, the air velocity increased with increasing vertical height. Thus, at the living level, around 1 m above floor, there is continuous air motion induced by the buoyancydriven force resulting from the four solar chimney ventilators used here. Regarding thermal comfort, the induced air motion of about 0.04 mrs cannot satisfy occupants as with temperature of about 3537, a higher air velocity is needed, about 2 mrs w10x. This air motion could be increased by increasing the number of units of solar chimneys on roof, eastern and western walls and by installing several free openings at the northern facade of room.

velocity could be increased by increasing the surface area of solar chimney. Finally, it is our opinion that natural ventilation seems to be feasible and viable. This, of course, awaits a largescale testing in order to see how effective such a ventilation system would be in a real building.

Acknowledgements The authors are grateful to the National Research Council of Thailand for providing partial fund in this research work.

References
w1x H.B. Awbi, Design consideration for naturally ventilated building, Renewable Energy Int. J. 5 1994. 10811090, Part 2. w2x G. Gan, A parametric study of Trombe wall for passive cooling of buildings, Energy and Buildings 27 1998. 3743. w3x N.K. Bansal, Solar chimney for enhanced stack ventilation, Building and Environment 28 3. 1993. 73377. w4x A.K. Sharma, N.K. Sodha, M.S. Gupta, Very-thermal wall for coolingrheating of buildings in composite climate, Int. J. Energy Res. 13 6. 1989. 733739. w5x J. Khedari, J. Hirunlabh, T. Bunnag, Experimental study of a roof solar collector towards the natural ventilation of new houses, Energy and Buildings 26r2 1997. 159164. w6x J. Khedari, C. Lertsatitthanakorn, N. Pratinthong, J. Hirunlabh, The modified Trombe wall: a simple ventilation means and an efficient insulating materials, Int. J. Ambient Energy, 1998, pp. 104110. w7x J. Khedari, S. Kaewruang, N. Pratinthong, J. Hirunlabh, Natural ventilation of houses by a Trombe wall under the climatic conditions in Thailand, Int. J. Ambient Energy 20 2. 1999. 8594. w8x J. Hirunlabh, W. Kongduang, P. Namprakai, J. Khedari, Study of natural ventilation of houses by a metallic solar wall under tropical climate, Renewable Energy Int. J. 18 1999. 109119. w9x B. Boonlert, Study of effect of openings on indoor temperature and air changes of room naturally ventilated by solar chimney, Thesis, Master of Engineering, Energy Management Technology Program, King Mongkuts University of Technology Thonburi, 1998, pp. 196. w10x Y. Nuparb, Development of ventilation comfort chart for Thailand, Thesis, Master of Engineering, Energy Management Technology Program, King Mongkuts University of Technology Thonburi, 1999, pp. 180.

4. Conclusion With four solar chimney ventilators of surface area of only 6 to 9 m2 corresponding to a ratio of 0.240.36 m2rm3 SC area to house volume. the average rate of ventilation ACH. between 12 AM to 2 PM was about 15, corresponding to 1.6 to 2.5 timerh per unit surface area of solar chimney. This induced ventilation reduced room overheating, expressed by a relativeness index defined as the temperature difference between average room temperature and ambient, by about 50%; Room temperature was about 238C above ambient. Regarding thermal comfort, the induced air motion of about 0.04 mrs at the living level, around 1 m above floor, cannot satisfy occupants as with indoor temperature of about 35378C, a higher air velocity is needed. This air