Renewabh, Energy Vol. 3. No. 6/7, pp. 795-799, 1993 Printed in Great Britain.

096~1481/93 $6.00+.00 .C 1993 Pergamon Press Ltd

TECHNICAL NOTE
Parameters of solar collectors tested with a KMITT solar simulator
T. KIATSIRIROAT a n d K. NA THALANG

School of Energy and Materials, King M o n g k u t ' s Institute of Technology Thonburi. Bangkok 10140, Thailand (Receh,ed 15 June 1992 ; accepted 31 July 1992) Abstract This paper reports on the experimental performances of flat plate solar collectors tested with a solar simulator under steady-state conditions, in terms o f collector efficiency, q, and ratio of temperature difference and solar radiation ( T ~ - T~)/IT. T~ was the effective heat-sink temperature of the tested collector and could be evaluated from temperatures of the collector's cover, ambient and light source panel (or infrared filter). Techniques for converting values of the collector's parameters, Frt Uco and FR (r:0~, obtained from the indoor tests to match outdoor results were demonstrated. The adjusted results agreed well with those of the outdoor data in the case of a collector having a flat glass cover. For a collector having a convex plastic cover, the estimated optical efficiency was lower than that of the outdoor result.

INTRODUCTION The steady-state approach to test solar collector performance has been widely used. However, it is realized that the procedure is time-consuming due to the variation of solar radiation flux. Clear sky periods for experiments are rarely available, especially in tropical regions. Thus to obtain a full set of data sometimes several weeks for testing are needed. A solar simulator is a light source from which the spectral output is close to real solar radiation and the radiation flux can be adjusted to different levels. Solar equipment can be tested indoors continuously and a whole set of experimental data can be achieved in a much shorter period when compared with that of outdoor testing. A solar simulator has been constructed at King M o n g k u t ' s Institute of Technology Thonburi ( K M I T T ) in Thailand and operated since 1988. The system, as shown in Fig. 1, consists of an air-cooled lamp array of 2 x 2 m 2, an infrared (IR) filter and a collector test rig. The lamp array m o u n t s 115 O S R A M H Q T - R 250 W / D L lamps which can generate radiation over 810 W / m 2 on the tested collector plane. The infrared filter consists of double 1.95 1.95 m 2 clear glass panes each of 6 m m thickness and 3 m m apart. Clean water is fed through the gap to control the filter temperature close to that of the surrounding ambient. Thus the effect o f long-wave radiation emitted from the filter to the tested collectors is alleviated. Under steady-state conditions, parameters of flat plate collectors are always plotted in terms of efficiency, r/, with the ratio of temperature difference over incidence radiation, ( T,~ - T,)/I r, The intercept on the axis and the slope of the curve are F~(t~) and FRU ~ which show optical performance and thcrmal loss of each collector, respectively. Some flat plate collectors were tested with the K M I T T solar simulator [1]. It was found that the radiation obtained was highly diffused, and there is an effect of thermal radiation exchange between the IR filter and the tested collector, thus some deviations from the outdoor test results were obtained. These observations were also found in some solar simulator laboratories which used a set of lamps as a light source [2.3].

The experiment described in this paper introduced an effective temperature, T~., which is the heal-sink temperature of the tested collector in the K M I T T solar simulator laboratory. This value is averaged from the collector's surrounding temperature and the temperature o f l h e IR filter. A technique to convert the indoor test results to be comparable with those of the outdoor test was also demonstrated.

Fig. I. The K M I T T solar simulator. 795

796
SOLAR

Technical Note
C O L L E C T O R E F F E C T I V E HEAT-SINK TEMPERATURE

UL,, - A B / ( A + B + C ) + U,. = A C i ( A + B + C ) : then

U~p~,,

At the K M I T T solar simulator laboratory, an IR filter is placed between the lamp array and the tested collector, as described in the previous section. Therefore, there are thermal losses from the collector to the surrounding ambient by convection and radiation, including thermal radiation loss to the IR filter. A diagram of thermal losses under steadystate conditions of a tested collector is shown in Fig. 2. Rate of thermal losses from upper part of collector per unit area
= h,:,p<(T v - T~) +h~.p_<.(Tr, - L )

q~, = F R(~c~)~lr -- FR UL,,( Tli -- T,) -- FR Uti(T,i -- Tin,). (7) FR is the ratio of actual useful heat to useful heat when the temperature of the whole absorber was at the inlet fluid temperature. Collector efficiency, ~/, then can be written as : q = q j l l = F R ( z ~ ) ~ - F R ( T , i - T,)/lr

= h<.~.,,(T<.- 7",) +h,..<_:,(T~ - T~) +h~.<.lii,(T,: - Tin,),

(1)

FRUi.r(7]i

Tm,)/Iv.

(8)

or
T~ = ( A T ~ + BT~, + C T . , O / ( A + B + C). where A = h,..p_+ h,.p ~, B = hc.p~+h ...... (2)

It can be seen that the efficiency now depends not only on T~ and T~ but also on Trot; therefore, a relation of q with ( T ~ - T~)/las the outdoor result seems to be imperfect. Equation (8) may then be developed in a form as : q = FR(r~)c -- FR UL~(7~ -- T~)/IT. (9)

C - hr.~_tilt,
h..... - 2 . 8 + 3 V ; V is wind speed in m/s. Rate of thermal losses from the back of the collector per unit area
-

T~ is called the effective heat-sink temperature of the collector, and UL~ is the effective heat loss from the collector to ambient at the temperature T+. Comparing eqs (8) and (9), we get : Fr~UL~ = FRUL, + FRUI, and T~ = (FRULjFRUL~OT, +(FRUu./FRUL~)T,,~. (11) (10)

U~+.,,(T~

T<,).

(3)

F r o m eqs ( I ) and (2). rate o f total heat losses per unit area, q~<>~can be written as :
qioss
=

(h~.p-~+h,.e~)(Tn-T~)+U

........( G -

T,,).

(4)

Substituting T~ from eqn (2) into eqn (4), q~,,~ = [A B/(A + B + C3 + Ub.p_~,](Tp- T,) +[AC/(A+B+C)](Tp7]~,,). (5)

When the ambient temperature T, and the infrared filter temperature Trot are measured, the effective temperature T~ could be evaluated from eqn (l l). In the case of the indoor test, from eqn (9), an experimental curve could be drawn in a form of a q vs (T,~-T,.)/II diagram instead of q vs (Tt~-T~,)/1T. The intercept on the q-axis is FR(z~)~ and its slope is --FRU~.
STEPS FOR CONVERTING 1NDOOR PARAMETERS TO OUTDOOR RESULTS

Therefore, the rate of useful heat-gain per unit area, q,, can be written as : q. = (z~)~/r - - ULa(Tp - - Ta) where
--

ULr(Tp -- T/lit),

(6)

"|'ill| ht,c-[ilt(Tc'Tfill )e ~ hl.c.~('l'c-T~)

NN~

h'c~(lc"l'~)
h r.,Wc('rp-+l') f

Perlbrmance parameters of collectors tested with the KMITT solar simulator might be different from those of the outdoor tests. There are two main factors, firstly, the light source consists of a set of lamps and thus most of the energy emitted is diffuse radiation; and secondly, there is thermal radiation exchange between the collector tested and the infrared filter. In the case of the outdoor tests, radiation loss from the collector is transferred to the surrounding ambient. With these problems, the two parameters, FR(r~)~ and FRULc will deviate from those obtained from the outdoor experiments. Thus it is necessary to develop a technique to convert the indoor results to match those of outdoor tests. The essential features of the method can be summarized. The optical parameter Outdoor tests are usually done in the midday hours, on clear days, with the beam solar radiation nearly normal to the collector. The transmittance absorptance term for the test conditions is approximately the normal incidence value, and is written as (z~),. In the case of indoor conditions, since the radiation obtained from the light source panel is highly diffused, the effective angle of incidence at the same transmittance as that of a beam radiation can be evaluated by [4] 0,. = 59.68--O.1388B+O.OOI497B 2, (12)

'<'">+\
+.,<.

\ )
Tp

~ h r,p.('I'p-T c)

Fluid /

5/11111/ !i/I/I/.
i ub,p.i(Tp.T. )

. ut + o , l e t .

"

Fig. 2. Directions of energy flow of the indoor tested collector.

Technical Note where B is the slope of the collector. With the K M I T T indoor test, the tested collector is oriented parallel to the I R filter and the light source panel. T h u s B = 0 and 0~ becomes 59.68 ~- 6 0 . With the value of 0~ = 6 0 , a relation between the indoor transmittance absorptance product (zc0,, and that of the normal incidence as the o u t d o o r condition could be estimated from Fig. 3 [5]. In the case of the single glass cover, we get ( r ~ ) j ( z ~ ) , - 0.83 or FR(r~)e/FR(r~)n = 0.83. With this technique, as the FR(r:0~ from the indoor experinaent is estimated, the result can be converted to match that of the o u t d o o r s by : F~(r~), = [ ( r ~ ) n / ( r ~ ) c ] F R ( r ~ ) e . In the case of the single glass cover, Fa(rc0, , = (I/0.83)FR(r~ L.
T h e loss

797

radiosity at the cover temperature. With the values of T~, T~tt, T~, and Tp, q*~ and FRU[ can be evaluated.

EXAMPLES
Two types of flat plate collectors having water as a working fluid were tested. The first one consisted of a black copper absorber of 1.318 1.09 m= and a single glass cover. The second one had a block rubber absorber with a convex plastic cover. Its aperture area was 1.015 x 0.615 m -~.

Outdoor data
The o u t d o o r data of both collectors at a wind speed of 3 m/s are s h o w n below.

(13)

Black copper collector.

r / = 63.1 - 8 9 2 . 5 ( T t ~ - T~,)/1, (%) Fa(r:0,, = 0.63, FRUI = 8.93 W/m-~K at T, = 34 C.

parameter

The loss parameter FFtUL~ which is obtained from the indoor result can be converted to match that of the o u t d o o r s by:

Black rubber colh, ctor.

q = 6 8 . 7 - 1364(T, -- T,,)/lv (%) FR(rc0, = 0.69,

FR U~ ,/FRt',~ - UL/Uu,
TD]. (14)

FRUL = 13.64W/m-~Kat 7~, - 30 'C.

Figure 4 shows the performance curves of the collectors under o u t d o o r conditions.

T,.)/(Tp -

F~,UL

(q,,,~/qt,~D[(T p -

is the loss parameter obtained as if the collector is tested under o u t d o o r conditions at the temperature T,,. The terms q~o~and q~*~are the rates of heat loss from the collector under o u t d o o r and indoor conditions, respectively, which can be evaluated as follows :

Indoor experiments Black copper collector. The indoor conditions for the black
copper collector were as follows : /'hi, - 39.5'C, 7", = 3 5 C , wind speed 3 m/s.

q,,,~ - ~:~a[( T~ + 273) 4 _ ( T~ + 273) 4] +h~x.,,(T,-- l~,)+qb~,.k (15)

and

q*~

(E~ - J~)/[(l - ~;~,)/~,~1+ho.~_otL- 7:~)+ q ~ .

(16)

Thermal emittances of the infrared filter and the cover of the collector = 0.85 ; the configuration factor of the collector -IR fitter =0.89. The effective temperature was calculated, with the details given in Appendix A, to be 36.YC while the average cover temperature during the experiment was 43.9 C. The relation between the collector efficiency and the value ( T , ~ - T ~ . ) / l r was

Ebc and Jc are radiation emissive power of blackbody and

rl = 51.04-788A(T,~- 7..)/IT
FR(r=) ~ = 0.51 and

(%)

FRUL,. = 7.88W/m -~K.

0.9 0.8
O. 7

70 60 F o ~ o Covers
~"

0, r
O.l 0 tO 20 30 40 50 0i

\\\V

3o

10

I
60 70 80 90 0 0.01

I
0.02

I
0.03

I
0.04

( T r i - T , ) I I T ( C m z/W ) Fig. 4. O u t d o o r collector performance curves.

t:ig. 3. Typical (r~)b/(rCt)n curves for glass covers.

798

Technical Note

From Appendix B, the rate of heat loss from the collector under both indoor and outdoor conditions was estimated to be : qt*~ = 160.22 W/m ~' and qlo~= 199.62 W/m-' respectively.

70

--

The absorber temperature, Tp, was normally higher than that of the working fluid by about 3 5'C. While the average temperature of the water during the experiment was 58.7~C, by assuming 5'C higher, the average absorber temperature, Tr, then became 63.7' C. With the values ofq~,,~, qk,~ and Tp, * the heat loss parameter of the indoor experiment, FRUL~, could be converted to that of the outdoor result at the ambient temperature of 34~C by :

50

"~,~

.v.o 30 ~ ~ ' u t d r ~ ; u e =

FR UL/ FR ULe = (q,,~/q*,~J[( Tp - Te)/(Tp - 7",,)]

- 199.62/160.22)[(63.7 -- 36.3)/(63.7 -- 34)] = 1.149. For FRUL~ = 7.88 W/m~' then FRUL = 1.149X7.88 = 9.06 W/m: K. The optical parameter obtained from the indoor experiment could be converted to match that of the outdoor by using eqn (13), thus: FR(Z~)n = FR(r~L/0.83 = 0.51/0.83 -- 0.614. Therefore, the evaluated outdoor performance of the copper collector could be :

t0 -
I 0 I I I I

" : , ~.. ~~ o

0.01

0.02

1 0.03

l~

0.04

(Tri-'l")/IT (Cruz/W) Fig. 6. Comparison of indoor and outdoor results of the rubber absorber collector.

FR(zcOc = 0.504

and

FRULe =

10.45 W/m-" K.

q = 61.4--906(Tf~--T,)/IT

(%),

With the method described previously, the performance curve could be adjusted to match that of the outdoor condition :

which agreed well with the actual outdoor data. Comparisons of the indoor test results with the outdoor data are shown in Fig. 5.

r1 = 60.7--1429(Tt~--T,)

(%).

Black rubber collector.

T,, = 39.4'C, T~, = 34C, wind speed 3 m/s.

Compared with the actual data, the heat loss term agreed well with that of the actual outdoor experiment, whereas the optical term was lower since the plastic cover was convex and not flattened. The results are shown in Fig. 6. CONCLUSIONS The following conclusions can be drawn from the present investigations. (1) For indoor collector testing, the parameter curve of the tested collectors are proposed to be expressed in a form of r/ and (T,-T~)/IT, where T~ is the effective heat-sink temperature of the collectors. (2) This paper describes techniques to convert the collector parameters obtained from the indoor results to match those of outdoor conditions. It was found that the results in the case of the collector having a flat cover agreed well with those the actual outdoor experiments.

The collector eqn for indoor conditions was : ~/= 5 0 . 4 - I 0 4 5 ( L ; - TJ/Iv

(%)

70

a "~

tdoor

Acknowh, dgement--This work was supported in part by the

Third World Academy of Sciences (STWS). NOMENCLATURE

20 l0 -

I 0

I o.01

I 0.o2

I 0.03

I 0.04

( T r I - T ) / I T (Cm 21W) Fig. 5. Comparison of indoor and outdoor results of the copper absorber collector.

A i = collector area, m 2 A2 = infrared filter area, m 2 E~, E~ = emissive power of blackbody at the collector's cover temperature, W/m 2 Eh_, = emissive power of blackbody at the infrared filter temperature, W/m 2 Eh~ = emissive power of blackbody at the surrounding temperature, W/m 2 F~ 2 = shape factor 1-2

Technical Note F~ 3 = FR = h~ = h~ = I~ = shape factor 1-3 heat removal factor convective heat transfer coefficient, W / m 2 K radiative heat transfer coefficient, W / m 2 K solar radiation incidence on the collector plane,
W/m 2

799

The radiation exchanges at the collector's cover and the infrared tilter can be written as :

A~8,(Ebt-J,)/(I-e.t)+

A,Fj_2(Jz-J~)
+.-41Ft_3(Eb3-J]) : 0

J~, J~ = ratiosity of the collector's cover, W/m 2 J2 = ratiosity of the infrared filter, W/m 2 q~,,~ = rate of heat loss from the collector under outdoor conditions, W/m 2 q*~ = rate of heat loss from the collector under indoor conditions, W/m 2 ambient temperature C Td, T3 To T~ = collector's cover temperature, 'C L = effective heat-sink temperature of the collector,
'C

A2E2(E~,2 - J2)/(l --sz) + A ,F,_2( J, - J 2 )

+A2F2-3(Eb3 - J 2 )
= 0

8j = 0.88, e2 =0.88. F~.2 = 0.89, F2_~ = 0.34, TI = 43.9~C, T2 =39.1 C, T~=35C,

A I = 1.318x 1.09m -~ A2 = 1.95x 19.5m: El_ 3 = 0.1l F2. 3 = 0.66

Ehl = 5.67 10-~(T~ +273) 4 = 571.84W/re" E h 2 = 5 . 6 7 x 1 0 ~(T2+273) 4=538.12W/m 2 E~,3=5.67I0-S(T3+273) 4=510.25W/m 2

water inlet temperature, 'C 7]iu, T2 = infrared tilter temperature, C T~= absorber temperature, "C U b back loss coefficient of the collector. W / m 2 K effective heat loss coefficient, W/m2K ULe B = slope of the collector, degree emittance of the collector's cover 8c,~ I emittance of the infrared filter 82 (r~)o = transmittance absorptance product of the collector under indoor conditions transmittance-absorptance product of the collector under outdoor conditions

Ji = 567.32W/m 2,

J2 = 537.09W/m2

hr.l-2 = hrx-lilt = A IFI-~.(JI 2

- Je)/[(A t ( T~ -- T2)] = 5.605 W/m -~K h~. i-3 = h .... = A tFi-3(Jt -Eh3)/[(A ~ ( T t - T2)] = 0.705 W/m ~ K FRULjFR UL~ = (11.8+0.705)/(11.8+0.705+ 5.605) = 0.69
F R ULf/F RULc = 5.605/(11.8 + 0.705 + 5.605) = 0.31

Subscripts
c - a = cover-surrounding ambient

T~ = 0.69 x 35+0.31 x39.1 = 36.YC. A P P E N D I X B: EVALUATION OF T H E R M A L LOSS FROM THE C O P P E R COLLECTOR Indoor conditions ; Tc = 43.9~C, L = 35"C, T~ = 36.3 C.

c-tilt - cover infrared filter.

REFERENCES 1. T. Kiatsiriroat, S. Thepa and V. Suntornrungsun, Solar collector testing under solar simulator. Tech. J. Enqineering Institute o f Thailand 1, 38-46 (1989). 2. D. Govaer, Indoor collector testing with an incandescent simulator. Solar Energy 40, 363-368 (1988). 3. Anon, A unique solar simulator for collector testing. Solar Progress 12, 7-9 (1981). 4. J. A. Duffle and W. A. Beckman, Solar Engineering o f Thermal Processes. Wiley, New York (1988). 5. S. A. Klein, Calculation of the monthly-average transmittance absorptance product. Solar Energy 23, 547-551 (1979).

ql....... = 8, (Eb, - J , ) / ( 1 - 8 , ) +h~.c_,(Tc - T~) = 0.88(571.84- 567.3174)/(1-0.88) + 11.8(43.9- 35) = 138.1857W/m 2

q~ck = k ( T , -

T~)/x;

x = insulation thickness = 0.0508 m k = insulation thermal conductivity

A P P E N D I X A: E V A L U A T I O N OF T H E COLLECTOR'S EFFECTIVE TEMPERATURE, Te, FOR T H E C O P P E R COLLECTOR

= 0.035 W/m K) = 0.035(63.7-35)/0.0508 = 22.03 W/m 2 q*~ = 138.1857+22.03 = 160.22W/m 2. When the surrounding ambient is a blackbody at 34'C, q,...... p = e,a[(T~ + 2 7 3 ) 4 - ( T , +273)4]+h~.~-,(L- T~) =0.88x5.67x10 s[(43.9+273) 4

T~ = ( FR UL,,/ Fr~U J Td + (FR Uu/FR UL~)Ttitt FR ULA/FR Uu. = (hc.c-~,+ h~.c_,)/(h ..... + h~.~.,,+ h~.c_nl 0 FR ULr/FR UL~ = (h~.~.~lt)/(h ...... + h~.~_~,+ h,.~-~lt)
h...... - 2 . 8 + 3 V ; V=3m/s; h..... = l l . 8 W / m Z K .

- (34+ 273) 4] + 11.8(43.9- 34) = 176.82 W/m 2 qb~ck = 0.039(63.7-- 34)/0.0508 = 22.8 W/m 2

An enclosure for radiation exchange consists of three surfaces : (1) collector's cover, (2) infrared filter and (3) surrounding ambient which is assumed to be a blackbody.

qlo~, = 176.82+22.8 = 199.62W/m 2.