Energy and Buildings 37 (2005) 373381 www.elsevier.

com/locate/enbuild

Experimental investigation on air heating and natural ventilation of a solar air collector
X.Q. Zhai, Y.J. Dai, R.Z. Wang*
Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Shanghai 200030, China Received 26 March 2004; received in revised form 2 July 2004; accepted 27 July 2004

Abstract Solar air collectors are important components for solar energy utilization in green houses. In this paper, experimental studies were carried out regarding to a solar air collector (SAC), for which the length of air channel is 1500 mm, the width is 500 mm, and a variable air channel gap ranges from 100 to 500 mm. In the experiment, the uniform heat ux along the air channel is effected by three electric heating plates, which play a role as solar radiation. It is found that the temperature distribution of air and the induced natural air-ow rate are highly dependent on heat input, inclination angle, channel gap, etc. Experimental results indicate that the optimum inclination angle for the SAC is 458, under which a maximum natural ventilation rate can be created. Also found is that there exists an appropriate channel length, about 1 m in this study, beyond which the obtained heat and the natural ventilation rate cannot be increased drastically. Higher the volume of air-ow rate through the SAC, lower the temperature difference between inlet and outlet, consequently, it should be balanced between the air temperature rise and a suitable mechanical air-ow rate in order to obtain maximum heat. Additionally, theoretical analysis based on heat balance equations is testied to agree well with experimental results. # 2004 Elsevier B.V. All rights reserved.
Keywords: Solar air collector; Heating; Natural ventilation

1. Introduction It is important to exploit natural energy, as well as to choose reasonable architecture structures for saving energy, which will meet the sustainable development of architectures, energy and the environment. A solar house promotes a comfortable indoor environment utilizing solar radiation, one of the most important sources of heat gain in a building and through its surroundings. It is also considered as an appropriate architecture for developing countries to provide heating, cooling and ventilation and improve indoor air quality depending on seasons [1]. The traditional Chinese house, mainly in rural areas, with a gable roof, can be well designed to integrate a solar air collector (SAC) to form a structure available for solar thermal utilization. In summer, it is feasible to use the roof structure to induce natural ventilation. In winter, it can be employed for space heating
* Corresponding author. Tel.: +86 21 62933250; fax: +86 21 62933250. E-mail address: rzwang@mail.sjtu.edu.cn (R.Z. Wang). 0378-7788/$ see front matter # 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2004.07.006

by mechanical ventilation with a payment of small electricity consumption. In the past decades, building roof structures based on solar air collector utilization have attracted attention in various investigations. Most of them were concerned with natural ventilation. Khedari et al. [2] conducted an experimental study of a roof solar collector (RSC) made by using CPAC Monier concrete tiles on the outer side and gypsum board on the inner side, and concluded optimum dimensions of the studied unit as follows: short length about 100 cm; tilt at 308; and the space between upper and bottom plates equal to 14 cm. Khedari et al. [3] carried out another eld measurements of the same kind RSC, and the experimental results showed that large air gap and large and equal size of openings would induce the highest rate of air-ow rate. Hirunlabh et al. [4] proposed four different congurations of the RSC to maximize natural ventilation through numerical modeling. Another kind of building roof structure comprising solar thermal utilization is solar chimney, which is similar to those mentioned above except

374

X.Q. Zhai et al. / Energy and Buildings 37 (2005) 373381

Nomenclature A Ai Ao Cp Dh f hb hc ho hr H I L q Q T Tfp Tfp(x) V W Greek b b0 d e l j r DTfp cross-section area of SAC channel (m2) cross-section area of SAC inlet (m2) cross-section area of SAC outlet (m2) specic heat of air (J kg1 8C1) hydraulic diameter of the air channel (m) friction factor of the channel wall convective heat transfer coefcient between insulation and indoor air (W m2 8C1) convective heat transfer coefcient (W m2 8C1) overall heat transfer coefcient between top plate and indoor air (W m2 8C1) radiative heat transfer coefcient (W m2 8C1) channel gap (m) heat input (W m2) length of SAC (m) obtained heat (W) air-ow rate across SAC channel (m3 s1) temperature (8C) mean air temperature inside air channel (8C) sectional mean air temperature (8C) mean air velocity of SAC channel (m s1) Width of SAC (m) symbols inclination angle of SAC (degree) coefcient of expansion of air thickness (m) emissivity thermal conductivity (W m1 8C1) local loss coefcient density of air (kg m3) air temperature difference between inlet and outlet of the channel (8C)

that the outer side is replaced by glazing. Bouchair [5] have done some experiments to testify the applicability of solar chimneys in improvement of indoor thermal environment by promoting natural ventilation in summer. Afonso and Oliveira [6] compared the behavior of a solar chimney with a conventional one. Their results showed that there was a signicant increase in ventilation rate with the solar chimney. Theoretical and experimental studies on the natural ventilation induced by solar irradiation were also done by Awbi and Gan [7], Chen et al. [8], Ong [9], and Bansal [10], etc. The aforementioned investigations were almost based on natural ventilation of building roof structure. However, heating in winter is the same necessary as ventilation in summer for most regions of China. Hence, the practical roof structure with the integration of SAC developed in China should be made to implement function of heating by mechanical ventilation, besides passive cooling by natural ventilation. In this paper, natural ventilation effected by SAC with an uniform heat ux on a single wall was experimentally investigated for different channel gaps, heat inputs and different SAC inclination angles. In addition, its air heating performance was studied at different air-ow rate across the SAC channel with the channel gap of 200 mm. The obtained heat and the natural ventilation air-ow rate of the SAC were theoretically analyzed according to heat balance equations, and were compared with the experimental results.

2. Experimental set-up Fig. 1 shows the schematic diagram of the experimental set-up and sectional view of air channel. The dimensional sizes for tested SAC was (length width channel gap) 1500 mm 500 mm 100500 mm. The SAC was supported by a bracket and could be tilted with different angles. The bottom and two sidewalls were all made of wood plank with thickness of 15 mm. The top cover, which was used to adjust the channel gaps by inserting it into ve different slots on the two side walls, was also a wood plank with dimensional sizes 1500 mm 500 mm 3 mm. To create an uniform heat ux at the bottom of the air channel, three stainless steel electric heating plates were arranged in series, acting as the heated surface. Moreover, a voltage regulator was used to change the input power. In order to supply heat for a given space in winter, mechanical ventilation mode was adopted by adding an upper cap over the air channel and connecting it with a fan through air ducts. The cap was, however, unnecessary for natural ventilation mode, it was removed when natural ventilation was needed. Temperatures at different points were monitored using Ttype thermocouples with the certainty of measurement of 0.5 8C, as shown in Fig. 2. There were six T-type thermocouples that were xed on the heating plates to

Subscript 1 asbestos board 2 bottom plate 3 glass wool blanket 4 top cover f air inside channel o outlet of SAC r indoor air Dimensionless terms Nu Nusselts number Pr Prandtl number Ra Rayleigh number Re Reynolds number

X.Q. Zhai et al. / Energy and Buildings 37 (2005) 373381

375

experimental unit, and V is the cross-sectional mean air velocity. It should also be noted that in the following, the difference of temperature between the measured and the indoor condition (Tr represents indoor air temperature), instead of practical value of air temperature, is applied in the analysis.

3. Formulation of the mathematical model The equations governing the performance of the system for either air heating mode or natural ventilation mode are formulated by coupling the energy balance equations of the components of the SAC with those for the determination of coefcients under different operation modes, making the following assumptions: (1) The SAC operates under steady-state conditions. (2) Thermal inertia of SACs components is negligible. (3) Under a given condition, the operating temperatures of SAC components and mean air temperature in air channel are all assumed to be uniform. (4) Temperature of the air varies only in the direction of the ow. (5) Heat ow through both the top cover and bottom plate is one-dimensional and in the direction vertical to air ow. (6) No leakage occurs from the air channel. 3.1. Energy balance equations The energy balance equations for the various components of the system are given below:For top cover: hr24 T2 T4 hc4f T4 Tfp ho4r T4 Tr For air channel: Cp rQTfo Tr hc2f WLT2 Tfp hc4f WLT4 Tfp (5) For bottom plate: I hc2f T2 Tfp ho2r T2 Tr hr24 T2 T4 (6) (4)

Fig. 1. Schematic diagram of the experimental system and sectional view of air channel. (a) Schematic diagram of the experimental system. (b) Sectional view of air channel (all units are in mm).

measure the local temperatures. The air temperature inside the channel was recorded by six T-type thermocouples arranged on a thin wooden pole, which can be xed at any position inside the experimental unit. The data were recorded at every 20 s interval in a data logger (Keithley 2700). The velocity of air at inlet and outlet of the channel was measured by a hot bulb anemometer (Type ET3-2A) for which the certainty of measurement is 0.02 m s1 within the range from 0.05 to 1.0 m s1. In Fig. 1, a three-dimensional coordinate system was set up with respect to the experimental unit. Here, Tf(x, y, z) represents air temperature at any point in the channel. As shown in Fig. 2, all the measuring points were located on six different sections along the air-ow direction. The distance from the channel inlet to these six sections were 50, 330, 610, 890, 1170, and 1450 mm, respectively. It is found that variation of air temperature along the zdirection is very small, thus the mean air temperature at any section, in which the measuring points have been distributed, can be calculated by: R W Tf x; y; 250 dy Tfp x H (1) WH The measured air temperature difference between inlet and outlet of the air channel is important for determining obtained heat in air heating mode and natural ventilation airow rate in natural ventilation mode. It can be expressed as: DTfp Tfp 1450 Tfp 50 The air-ow rate, Q, can be obtained as follows: Q WHV (3) (2)

where W and H are the width and channel gap of the

Fig. 2. Illustration of locations of thermocouples.

376

X.Q. Zhai et al. / Energy and Buildings 37 (2005) 373381

The obtained heat of the SAC can be expressed as: q Cp rQTfo Tr 3.2. Determination of coefcients in the mathematical model Among the coefcients of the mathematical model described above, all of them can be determined in the same way, except for air-ow rate across SAC channel, Q, mean air temperature inside the channel, Tfp, and convective heat transfer coefcient, hc, which should be estimated according to different operation modes. To begin with, the air-ow rate, Q, for air heating mode accomplished by mechanical ventilation is completely regulated by the fan and ow control valve. However, natural ventilation air-ow rate in the air channel is mainly determined by two aspects, which are stack pressure built up in the air channels and the pressure losses at the inlet, outlet and along the RSC channels. The energy conservation equation can be expressed as: rgb0 L sin bTfp Tr ji rQ=Ai 2 rQ=Ao 2 jo 2 2 L rQ=A2 f Dh 2 (7)

In contrast, it is suggested by Hirunlabh et al. [14] for natural ventilation mode: Tfp 0:75Tfo 0:25Tr (12)

Other coefcients for both air heating and natural ventilation are calculated as mentioned below. Overall heat transfer coefcient between bottom plate and the ambient air may be obtained from: ho2r 1 d1 =l1 d2 =l2 d3 =l3 1=hb (12)

where hb is the convection heat transfer coefcient between the insulation and the indoor air, according to Forson et al. [15], the values of hb range from 12.5 to 25 W m2 8C1). Overall heat transfer coefcient between top cover and the ambient air may be computed as: ho4r 1 d3 =l3 d4 =l4 1=hb (13)

The radiation heat transfer coefcient between top cover and bottom plate may be predicted by: s273 T2 2 273 T4 2 273 T2 273 T4 1=e2 1=e4 1

(8)

hr24

(14)

where the pressure loss coefcients are recommended by Sandberg [11] as ji = 1.5, jo = 1.0, f = 0.056, respectively. As for convective heat transfer coefcient, hc, the fully developed ow condition is assumed in the analysis of air heating mode for calculating forced convective heat transfer coefcient. According to Niles et al. [12], the following correlation can be adopted: Nu 0:0333Re0:8 Pr 1:3 (9)

The physical properties of air are assumed to vary linearly with air temperature. Basic parameters for calculation are listed in Table 1. According to this mathematical model, obtained heat for air heating mode and natural ventilation air-ow rate for natural ventilation mode can be gained, respectively.

4. Experimental results and analysis Two working modes, namely, air heating and natural ventilation were investigated here (Plates 1 and 2). The difference between the two modes was the use of a fanpower mechanical ventilation system for the air-heating mode. 4.1. Air heating Experiments were carried out with uniform wall heat ux at 120, 210, 330, 480, 650 W m2, respectively. The channel gap is 0.2 m, and the inclination angle for the tested unit is 458. Here, a fan drives the mechanical ventilation through the air channel and the air-ow rate is changed with a ow control valve within the range of 0.010.08 m3 s1. Fig. 3 describes the variations of air temperature along xdirection, as well as y-direction. Particularly, the heat ux is 650 W m2, and the air-ow rate inside the channel is 0.06 m3 s1. It is found that the highest air temperature occurs close to the electric heating plates, then, the air temperature descends gradually away from the electric heating plate. However, on account of the radiation from the

With regard to natural ventilation, convective heat transfer coefcient, hc, may be predicted according to Hollands et al. [13]: " #  1708sin 1:8b1:6 1708 Nu 1 1:44 1 1 Ra cos b Ra cos b " 1=3 # Ra cos b 1 (10) 5830 where the meaning of the + exponent is that only positive values of the terms in the square brackets are to be used (i.e., use zero if the term is negative). It should be noted that the correlations given by Eq. (10) does not cover the range of tilt angles from 75 to 908, however it is recommended that the 758 correlation be used to represent the vertical case adequately. Mean air temperature in the air channel for heating mode is given by: Tfp Tfo Tr 2 (11)

X.Q. Zhai et al. / Energy and Buildings 37 (2005) 373381 Table 1 Basic parameters for calculation d1 d2 d3 d4 l1 l2 l3 l4 ha ji jo f ec ep 0.005 0.015 0.01 0.003 0.12 0.048 0.043 0.048 15 1.5 1.0 0.056 0.85 0.82

377

Fig. 3. Air temperature distributions along x and y directions with the heat input of 650 W m2 and air-ow rate of 0.06 m3 s1.

electric heating plates, temperature rises near the top cover, which results in a somewhat higher temperature at the cover surface. Also found is that the temperatures at different distances from the electric heating plates all increase along the length of air channel.

The inuences of heat input on the sectional mean air temperatures are plotted in Fig. 4, with a constant air-ow rate of about 0.06 m3 s1. Experimental results indicate that the air channel should not be too long, because the air temperature rise would become weak after a specic length is reached. In this study, the suitable value for air channel is about 0.6 m. It is obvious that the larger the heat input, the higher the air temperature inside the air channel will be. Fig. 5 depicts the variations of sectional mean air temperatures with different air-ow rates at I = 480 W m2. With the increase in the air-ow rate, the sectional mean air temperature falls off and the curve tends to be smooth, which results in a relatively small temperature difference between the inlet and the outlet. Also can be deduced that it is unnecessary to just lengthen the SAC to attain the required heat especially when the air-ow rate is beyond 0.065 m3 s1. The obtained heat through the SAC is determined by airow rate and temperature difference between the inlet and outlet. In the experiment, the air-ow rate was regulated by the fan and the ow control valve. The obtained heat can be calculated as follows: q Cp rQ DTfp (15)

Plate 1. Natural ventilation mode of the experimental unit.

where q represents obtained heat, r and Cp are air density

Plate 2. Air-heating mode of the experimental unit.

Fig. 4. Variations of sectional mean air temperatures along air-ow direction with different heat inputs at a constant air-ow rate of 0.06 m3 s1.

378

X.Q. Zhai et al. / Energy and Buildings 37 (2005) 373381

Fig. 5. Variations of sectional mean air temperatures with different air-ow rates at a constant heat input of 480 W m2.

Fig. 7. Variations of obtained heat with different air-ow rates at a constant heat input of 480 W m2.

and specic heat capacity at mean air temperature inside the channel, respectively. The experimental value of obtained heat is discussed and compared with the predicted value in terms of Eq. (7) as follows. As can be seen from Fig. 4, with an uniform air-ow rate, the air temperature difference between inlet and outlet goes up with the increase of heat input, which results in an enhancement of obtained heat according to Eq. (15). Fig. 6 shows the variations of obtained heat with different heat inputs when the air-ow rate is 0.06 m3 s1. It is found that the obtained heat rises continually with the increase of heat ux, indicating that the air heating effect of the SAC can be improved by the increment of solar irradiation. Also found is that the predicted values are rationally consistent with the experimental results, both the experimental values and the predicted values of obtained heat almost increase linearly with heat inputs under otherwise identical conditions. Fig. 7 compares the impact of air-ow rate on heating effect at I = 480 W m2. With the increase in the air-ow rate, the obtained heat initially goes up rapidly with the increment of air-ow rate and then the increase stops when the air-ow rate reaches 0.06 m3 s1. This is due to the simultaneous reduction of temperature difference between the inlet and outlet with the increment of air-ow rate.

Meanwhile, the increase of predicted values tends to slow down with further increase of air-ow rate. Using Eq. (7), predictions were made for the obtained heat of the SAC investigated in this work. From Figs. 6 and 7, the predicted values are found to be in reasonable agreement with the experimental results. Compared with the experimental results, predicted values seem to be a little higher, which is resulted from overestimation of convective heat transfer coefcients inside the air channel based on fully developed ow assumption. 4.2. Natural ventilation In this working mode, experiments are mainly carried out when the inclination angle is 458, the channel gap is 200 mm, and the heat input is 650 W m2. For the purpose of developing optimum structure, experiments are carried out for different channel gaps and inclination angles. Fig. 8 describes the air temperature at 10, 20, 50, 100, 150, 180, 190 mm from the electric heating plate along the air-ow direction when the heat input is 650 W m2. It is found that the air temperatures generally rise along the airow direction. Also found that the temperature distributions across the channel gap are not uniform, with higher air

Fig. 6. Variations of obtained heat with different heat inputs at a constant air-ow rate of 0.06 m3 s1.

Fig. 8. Air temperature distributions along x and y directions with the channel gap of 200 mm, inclination angle of 458 and heat input of 650 W m2.

X.Q. Zhai et al. / Energy and Buildings 37 (2005) 373381

379

temperature near the electric heating plates. Generally speaking, air temperatures fall off with increase of the distance from the electric heating plates. However, similar to the air-heating mode, temperatures rise when approaching the top cover as a result of the radiation from the electric heating plate. Fig. 9 shows the sectional mean air temperatures along the air-ow direction when the heat input changes at 120, 210, 330, 480, 650 W m2, respectively. It is seen that the sectional mean air temperatures move up with the increase of heat inputs. Both Figs. 8 and 9 indicate that, initially, the air temperatures go up gradually along the length of air channel, and reach their peak values at the section 1200 mm or so from the inlet, and then fall off near the outlet of the experimental unit. This is due to the occurrence of back ow at the outlet. The descending tendency of temperature is more obvious with higher heat input. Similar observation was reported by Bouchair [5]. In addition, because the air temperatures generally tend to go up with the increment of the SAC length, natural ventilation can be enhanced with longer SAC. However, as shown in Fig. 9, the air temperature rise gradually slows down with increment of the length of the experimental unit, especially when the length reaches 1 m or so. This means that the instantaneous efciency of heat collection tends to decrease with the increase of SAC length. As a result, the amount of air-ow rate induced by a longer SAC would be lower than that induced by two units of the SAC, with a total length equal to that of the longer unit. The most important performance for natural ventilation mode is natural ventilation air-ow rate induce by the SAC, then, the experimental values of natural ventilation rate obtained from Eq. (3) are discussed and compared with predicted values according to Eq. (8) as mentioned below. From experimental results shown in Fig. 9, it is obvious that higher heat input causes higher mean air temperature inside the SAC under otherwise identical conditions, and thus a higher stack pressure to induce natural ventilation. It

Fig. 10. Variations of natural ventilation air-ow rate with different heat inputs at the channel gap of 200 mm and inclination angle of 458.

could be further understood from Fig. 10, which depicts the air-ow rate through a 458 inclined SAC with different heat inputs and the same channel gap of 200 mm. The ascending trend of curve indicates that a higher natural ventilation rate by the SAC can be induced with higher solar irradiation. It can be seen that, for the experimental unit, the air-ow rate reaches 0.0285 m3 s1 at a 500 W m2 heat input, which corresponds to 0.038 m3 s1 m2 per unit area of electric heating plate. It is shown from experiments that, at otherwise identical conditions, the mean air speed of the channel decreases with the increase of channel gap, however, the air-ow rate appears to increase continuously with an increment of channel gap. This can be proved from Fig. 11, which shows the air-ow rate through a 458 inclined SAC with different channel gaps and the same heat input of 650 W m2. No optimum channel gap was found either by experiments or by simulations. Fig. 12 describes the impact of inclination angles on the air-ow rate through the SAC of 200 mm channel gap at I = 650 W m2. Theoretically, higher inclination angle causes higher stack pressure for driving natural ventilation, however, based on Hollands et al. [13], under otherwise identical conditions, the convective heat transfer coefcient

Fig. 9. Variations of sectional mean air temperatures along air-ow direction with different heat inputs at the channel gap of 200 mm and inclination angle of 458.

Fig. 11. Variations of natural ventilation air-ow rate with channel gaps at the heat input of 650 W m2 and inclination angle of 458.

380

X.Q. Zhai et al. / Energy and Buildings 37 (2005) 373381

Fig. 12. Variations of natural ventilation air-ow rate with inclination angles at the heat input of 650 W m2 and channel gap of 200 mm.

inside the air channel decreases with the increase of inclination angle, which will result in the reduction of natural ventilation air-ow rate. Predicted values indicate that natural ventilation air-ow rate tends to peak at about 608 inclination angle, meanwhile, also can be shown is that predicted value of 458 inclination angle is lower than that of 608 inclination angle by about 4.5%. Similar to the variation of predicted values, experimental results show that, initially, the natural ventilation air-ow rate enhances with the increase of inclination angle, and then it reaches the peak value at about 458 inclination angle, nally, it begins to fall off, which indicates that the decrease of convection heat transfer overwhelms the increase of stack pressure after the inclination angle is beyond 458. From Figs. 1012, it is shown that the predicted natural ventilation air-ow rates reasonably coincide with the experimental results. The deviation between them can be due to two reasons: (1) the air velocity measurement has not taken into account the boundary layer effect, which may lead to the decrease of experimental air-ow rate; and (2) theoretical analysis has not consider the variation of pressure loss coefcients at inlet and outlet of the air channel under different conditions, which may result in the predicted values either on the low side or on the high side of experimental values.

which indicates that solar irradiation is predominant in strengthening air heating and natural ventilation for a SAC. (2) For a SAC under the air-heating mode, it should be balanced between the air temperature rise between inlet and outlet and a suitable mechanical air-ow rate in order to obtain maximum heat. (3) In the studied cases, it is shown that the rise of air temperature along the experimental unit length tends to be decelerated when the unit length exceeds 600 mm under the air-heating mode. The similar phenomenon occurs when the length exceeds 1 m under the mode of natural ventilation. Accordingly, the air temperature distributions inside the channel under two operating modes show that the length of the SAC should not be too long. Taking two modes into account, it is recommended that the length of about 1 m is efcient in enhancing the air heating and natural ventilation performance. (4) For the SAC investigated in this work, the natural ventilation air-ow rate increases continuously with increasing channel gap from 100 mm to 500 mm under otherwise identical conditions. Experimental results also show that the optimum inclination angle for natural ventilation is 458.

Acknowledgements This work is supported by the state Key Fundamental Research Program under the Contract no. G2000026309 P.R.C., and the Shanghai Commission of Science and Technology under the Contract no. 03DZ12012.

Reference
[1] S. Kumar, S. Siaha, N. Kumar, Experimental investigation of solar chimney assisted bioclimatic architecture, Energy Conversion and Management 39 (56) (1998) 441444. [2] J. Khedari, J. Hirunlabh, T. Bonnag, Experimental study of a roof solar collector towards the natural ventilation of new houses, Energy and Buildings 26(2) (1997) 159164. [3] J. Khedari, W. Mansirisub, S. Chaima, N. Pratinsong, J. Hirunlabh, Field measurements of performance of roof solar collector, Energy and Buildings 31(3) (2000) 171178. [4] J. Hirunlabh, S. Wachirapuwadon, N. Pratinthong, J. Khedari, New congurations of a roof solar collector maximizing natural ventilation, Building and Environment 36(3) (2001) 383391. [5] A. Bouchair, Solar chimney for promoting cooling ventilation in Southern Algeria, Building Service Engineering Research and Technology 15 (2) (1994) 8193. [6] C. Afonso, A. Oliveira, Solar chimneys: simulation and experiment, Energy and Buildings 32(1) (2000) 7179. [7] H.B. Awbi, G. Gan, Simulation of solar-induced ventilation, Renewable Energy Technology Environment (1992) 4. [8] Z.D. Chen, P. Bandopadhayay, J. Halldorsson, C. Byrjalsen, P. Heiselberg, Y. Li, An experimental investigation of a solar chimney model with uniform wall heat ux, Building and Environment 38 (2003) 893906. [9] K.S. Ong, A mathematical model of a solar chimney, Renewable Energy 28 (7) (2003) 10471060.

5. Conclusion Using electric heating plate to simulate solar heat collecting plate, experiments were carried out to study the air heating and natural ventilation performance of a solar air collector. Mathematical model based on heat balance equations was testied to be available for the prediction of obtained heat during air heating mode as well as air-ow rate during natural ventilation mode. As a brief summary, it is wished to emphasize the signicant points of this work in the following: (1) Both obtained heat and natural ventilation air-ow rate almost increase linearly with the increase of heat inputs,

X.Q. Zhai et al. / Energy and Buildings 37 (2005) 373381 [10] N.K. Bansal, Solar air heater applications in India, Renewable Energy 16 (1999) 618623. [11] M. Sandberg, Cooling of building integrated photovoltaics by ventilation air, in: Proceedings of HybVent Forum99, First International Oneday Forum on Natural and Hybrid Ventilation, the University of Sydney, Darlington, New South Wales, Australia, 28 September 1999, pp. 1018. [12] P.N. Niles, E.J. Caxnegie, J.G. Pohl, J.M. Cherne, Design and performance of an air collector for industrial crop dehydration, Solar Energy 20 (1979) 1928.

381

[13] Hollands KGT, T.E. Unny, G.D. Raithby, L. Konicek, Free convective heat transfer across inclined air layers, ASME Journal of Heat Transfer 98 (1976) 189193. [14] J. Hirunlabh, W. Kongduang, P. Namprakai, J. Khedari, Study of natural ventilation of houses by a metallic solar wall under tropical climate, Renewable Energy 18 (1999) 109119. [15] F.K. Forson, M.A.A. Nazha, H. Rajakaruna, Experimental and simulation studies on a single pass, double duct solar air heater, Energy Conversion and Management 44 (2003) 12091227.