RenewableEnergy Vol. 1, No. 5/6, pp. 595607, 1991 0960--1481/91 $3.00+.

00
Printed in Great Britain. © 1991 Pergamon Press plc

DESIGN ANALYSIS OF CORRUGATED AND FLAT

PLATE SOLAR AIR HEATERS

C. CHOUDHURY a n d H. P. GARG
Centre for Energy Studies, Indian Institute of Technology, Hauz Khas, New Delhi-110016, India

(Received 23 August 1991 ; accepted 12 September 1991)

AMtraet--A detailed theoretical parametric analysis has been made on corrugated and flat plate solar air
heaters of five different configurations. The analysis involves investigating the effects of the air flow
velocities (or the air channel depths) on the air temperature increment, the system efficiency and the
pressure drop experienced by the flowing air for different air channel lengths and different specific mass
flow rates of air. The design analysis and the curves discussed in the article are intended to enable a designer
to construct economical and efficient solar air heaters with technically logical air passage dimensions.

1. INTRODUCTION effectiveness of the system. Charters [5] in 1971

examined the optimization of aspect ratio of the rect-
The commercial acceptability of a solar air heating
angular flow passage from the point of view of mini-
system depends on its performance efficiency and cost-
mizing the cost for a fixed pumping power. Hollands
effectiveness. The key parameter responsible for
and Schewen [6] in 1981 studied the effect of the
increasing the efficiency of such a system is the overall
dimensions of the air flow passage on the heat transfer
heat transfer coefficient between the absorber and the
coefficient between the absorber and the flowing air
flowing air. This parameter, for a fixed specific mass
for fixed pressure drop values. They used rectangular
flow rate of air through the heater channel, can be
and triangular flow passage in plate-type air heating
varied by varying the air velocity or the air channel
collectors. Choudhury et al. [7] investigated the effects
depth which, in turn, causes a variation in the pressure
of the air flow passage dimension on the air tem-
drop experienced by the air stream in passing through
perature increment and the efficiency of an uncovered
the collector channel. If no upper bound is placed on
corrugated plate air heater. In the present study,
the pressure drop in the system, the convective heat
attempts have been made to examine theoretically the
transfer coefficient (between the absorber and the
effects of the air flow passage dimensions (length and
flowing air) can be made arbitrarily large by the pro-
depth) on the collector efficiency, the air temperature
per choice of the flow passage dimension. However, a
increment and the pressure drop in the system for
cost-effective design consideration of the system pro-
different fixed specific mass flow rates of air. Five
vides constraints on the pressure drop and hence on
different configurations of corrugated and flat plate
the flow passage depth of the system. If the material
solar air heaters were used.
cost of the collector is fixed, then for a fixed specific
mass flow rate of air, the economics of the system are
governed by the cost of the pumping power expended
2. THEORETICAL ANALYSIS
in the collector which depends on (increases with) the
pressure drop in the system. A consideration of these Figure l(a) (a-e) shows schematically the cross-
points suggests the existence of optimum air flow pass- sectional views of the solar air heaters investigated in
age dimension that corresponds to efficient and cost- the present work. In all the configurations, the air
effective design of the system. heaters are comprised of three plates. Plate 1, the
A review of solar heating literature [1-4] reveals upper-most plate, act as the cover in all the con-
that flat and corrugated plate solar air heaters have figurations, plate 2, the middle plate, is the inner cover
received considerable attention for space heating, in Types I, I1 and III and absorber in Types IV and
agricultural drying and to some extent, water heating V, and plate 3, insulated at the bottom, is the rear
applications. However, little has been published on plate which is the absorber in Types I, II and III and
the effect of the air flow passage dimension on the wooden sheet in Types IV and V. In all the con-
efficiency and the pressure drop and hence on the cost- figurations, the air flows through the passage between
595
596 C. CHOUDHURYand H. P. GARG

Cover
Air flow L I Cover
perpendicular - - v

to plane of ~ Absorber
paper Insulation
(a) Type-I

F,.~"~ ..- _..-~-3 cOver 1 Cover

- Cover L-, J Cover
® ®
Absorber Y//////~///////////Z Absorber
1/¢///////////////'/'~" Insulalion Insu lot ion
(b) Type - II (c) Type-Ill

~,/ . _ ~ ~ Cover
Absorber
F
, ,] Cover
Absorber
® ®
Bookplate
,/'//////////////// eack plate
"///////////////z ,nsula,io
Insulation
(d) Type-IV (¢) Type V
Fig. 1(a). Schematic views of the solar air heaters.

hla

Plate 1
TI I h21
T2 I I Plate 2
Ta Q~h2f
Air th2 3 To
flow
', ~ -h3f

1
h3a
Fig. l(b). Coefficientsof various heat transfer in the solar air heaters.

the middle and the rear plate and provides the heat
transfer. The various heat transfer coefficients at
different components of the air heaters are shown in
Fig. 1(b). In the analysis, the heat transfer is con-
sidered to be steady state, equal ambient temperatures
are assumed at the front and the rear of the collector
and the inlet air temperature is assumed to be the
same as the ambient temperature. The energy balance
equations for the plates and the flowing air in these
configurations can be written as :
 Design analysis of solar air heaters 597

S,I+h2t(T2-T,) = h~a(T,-Ta) (1) heat transfer coefficient h23 and the conductive heat
transfer coefficient h3a were obtained by using the
S.,I = h2t ( T2 - T]) + h23( T2 - T3) + h2r( Tz - Tr)
standard heat transfer relations summarized in Duffle
(2) and Beckman [3].
The forced convective heat transfer coefficients h2r
S 3 I = h 32(T3 - T2) + h3f(T3 - Tr) +h3~(T3 - Ta)
and h3r were obtained by using the relation derived
(3) by Tan and Charters [8] which includes the effects of
the entrance and exit length in the air flow passage
()QrCf/W)(dTddy) = h2r(T-, - Tr) + h3r(T3 - Tr).
and which is given by :
(4)
h = Nu KID (7)
The boundary conditions are :
Nu = Nuoo (1 + MOLL) (8)
Tr=Ta at y=0
where
Tr=To at y=L.
Nuo-~ = 0.0182Re °'~ Pr °'4 (9)
In the above equations :
Re = p V D / p (10)
S1 ~ 1
M = 14.3(1og N - 7 . 9 ) (11)
S 2 = ~1~2
with
and
N=L/D if 0 < L / D < ~ 6 0
$ 3 " = -/71"c2~3.
= 60 if L / D > 60.
For the corrugated absorber (cover), since the actual
surface area absorbing (transmitting) the incident To account for the heat transfer to the flowing air
radiation is larger than the nominal area used for from the corrugated plate, the right-hand side of eq.
estimating the specific mass flow rate of air and since (7) is multiplied by the corrugation factor ft.
a part of the radiation is absorbed (transmitted) by
Pressure drop
multiple reflection on the absorber (glass) surface, the
effective value of a(~), the total solar absorptance When air flows through the channel in the air
(transmittance) for the corrugated absorber (cover) heater, owing to friction, the air pressure drops along
in the computation is assumed to be 0.95 (0.95) where the length of the flow channel. If the effects of the
as that for the flat absorber (cover) is 0.90 (0.90). channel entrance and exit, of surface roughness and of
Solving eqs (1)-(3) for TL, T2 and Z 3 and inte- the compressibility of the air are neglected, the pres-
grating eq. (4) (after substituting for T2 and T3) sure drop AP can be computed from the relation :
within the limits T f ~ T a to To as y ~ 0 to L, the outlet AP = 2f(pVZ)L/O (12)
air temperature To was obtained as :
where the friction factor f, for turbulent flow, is given
To = (Ti + Pts/P19) exp ( P I 9 L W / r h r C r ) - P t a / P I g . by [6, 9]:
(5) f = 0.059Re -°'2. (13)
The coefficients Ps are defined in the appendix. Numerical values of different parameters, i.e. To, r/
The efficiency of the collector was then obtained by etc., were computed corresponding to an ambient tem-
using the relation : perature of 300 K, solar flux of 900 W/m 2 and wind
speed of 1.5 m/s. It may also be noted that eqs (8)-
rhrCr(To -- Ta)
~/= I (6) (13) were used for computing h2f, h3r and AP for
turbulent as well as transition flow regimes and that
no theoretical results were obtained in the laminar
Heat transfer coefficients
flow region.
The radiative and the wind related convective heat
transfer coefficient h ~a, the radiative and the natural
convective heat transfer coefficient h~t, the radiative 3. R E S U L T S A N D D I S C U S S I O N

The computed values of the efficiency for different

specific mass flow rates of air through the channel in
* S 3 for Types IV and V collectors is zero. the Type I air heater are plotted against the flow
598 C. CHOUDHURYand H. P. GARG

velocity in Fig. 2. With an increase in the air mass
flow rate, the curves show an increase in the efficiency,
which is very rapid at lower flow rates and less rapid
at higher flow rates. In addition, as may be noted, for
a fixed specific air flow rate and a fixed collector
length, the efficiency increases with increase in the air
velocity, initially rapidly and then slowly. This can be
attributed to the fact that larger air velocities with
shallower air channel depths cause larger heat transfer
to the flowing air (Fig. 3) resulting in higher efficiency
of the system. However, the dependence of efficiency
on air velocity is more predominant at higher mass
flow rate than at lower flow rates. The relationship
between the air temperature increment and the system
efficiency for different air flow rates and air velocities
for the five different configurations of the air heaters
is shown in Figs 4-8. The curves illustrate that with
increase in the temperature increment of the flowing
air, the efficiency of the air heaters decreases. It is also
revealed that the speed of the air is most important in
determining the efficiency at higher air flow rates and
the temperature rise at lower flow rates. Therefore,
the temperature rise of the flowing air can be increased
by decreasing the air flow rate per unit area and the
loss in efficiency at lower flow rates can be avoided by
increasing the air speed or by decreasing the channel
depth in the air heater until the pressure drop experi-
enced by the flowing air is the maximum that can be
tolerated.
The variation in efficiency with the air channel
depth for different collector lengths is displayed in
Figs 9-13 for different specific mass flow rates of air
through the heaters. A comparison of efficiencies in
these figures indicates that at a fixed specific air flow
rate and a fixed air velocity, the system efficiency
decreases with increase in the air channel length. This
effect is more predominant at lower air velocities. The
curves also illustrate that for a fixed specific air flow
rate and a fixed air channel length, the efficiency
increases with increased air velocity (or decrease in

80
V=Smls
m/s

~~~
3mls
.~.._.. 2mls
60 _ lmls

~-0
f L = 8m

20

0 I I I I

8O
V= 5mls
., /,mls
3rots
2mls
I mls
6G

t. 0

20

I I I
0 50 100 150 200
m (kglh m2 )
Fig. 2. Et~ciency as function of the specific mass flow rates of air for Type I air heater•
 Design analysis of solar air heaters 599

t.O
r~ = 100 k g l h m 2 /L=2m
/-.,m
30 6rn
8rn

20

t0
,e -

.~ 0 I I I l I •
>

"~ ,'-0 - ~ = 200 kg/h m2,.~

100 kg/h m_~
2
.c [ L = 2m / j 50kglhm

30

20

10

o I
o 1 2 3 5 6
V(mls)

Fig. 3. Effect of air velocity on the coet~cient of convective heat transfer between corrugated absorber and
flowing air.

air channel depth). This effect is more predominant
for longer air flow channels with higher specific air
flow rates.
If no upper bound is placed on the air speed, the
efficiency of the systems can be increased arbitrarily
by the proper choice of the air channel dimension.
However, as has been mentioned earlier, too large an
air velocity and too small an air channel depth result
in excessive pressure drop in the system and hence in
excessive fan running costs. A consideration of this
point suggests the existence of air flow channel with
optimum dimensions which would correspond to
efficient and cost-effective designs of the systems.
Figures 14 and 15 show the variation of pressure
drop in the air flow passage as function of air passage
depth for different heater lengths and different specific
mass flow rates of air through the heater channel. An
analysis of the values in different figures suggests that
at 50 kg/h m 2 specific flow rate of air in a 6 m long
air channel, by increasing the air velocity from 2 to 4
m/s (or decreasing the air channel depth from 3.5 to
1.75 crn), the efficiency increases by 3-6% in different
collector types (Figs 9-13) whereas the pressure drop
rises from 10 to 70 Pa (Fig. 14), thus increasing the
pumping power expended by about 7 times. Similarly,
at 150 kg/h m 2 air flow rate for a 6 m long collector,
the increase in air velocity from 2 to 4 m/s (i.e. decrease
in air channel depth from 10.5 to 5.2 cm) increases
the efficiency by 4-7% (Figs 9-13) whereas the pres-
sure drop and hence the pumping power by more than
7 times (Fig. 15).
To obtain systems with low pressure drops and
improved efficiencies, proper dimensions of the air
flow passages, such as their length and depth, can be
chosen from the sets of curves as plotted in Figs 14
and 15, if fixed values of the pressure drop and the
air flow rate (by requirement of the air temperature
increment) are assumed at the beginning of the opti-
mization procedure. For this purpose, depending on
the dimension of the space available for installation,
600 C. CHOUDHURY and H. P. GARG

l-
90

~E "' L =8m

80

rnl s o ,~. , 4 "S ~ o

3 mls ~ ~
70
2,~Is
this
~ ~ ~
I[nl$ t ~ ~'Y
60 I rnl$ ~f t ///

7 50 I I I I I 1 I I I I

90

~ L =6m
-c E "~
80
- .-'rn,,~~ o o~ 6"/

70 2m1% ,~ l
In, is

6O
~ , '
~01
0.0
I
0.01 0.02
I
0.03
I I
0.0"
I
0.05 0.0
I
0.01 0.02
I
0-03
I 1
0.0~
.,.L
0.05
( T o - - T a ) ! (~C m 2 / W )

Fig. 4. Efficiency as a function of air temperature increment per unit incident flux for different air velocities
in the Type I air heater.

90

~ L :Am
E E L =Bin

80- ,c E
V e ~ C3 c~

4 -.,/$ ~ =.- ~ •
70
i!
6(3

50 I 1 I I I I I ,I I I

c-- 90

~ L = 2m
..~ E ~ L=6m
80--
crP ,,c" '%

7C

6O

50 I I [ I I I I, I I ,L
O.(3 o.m o.o2 o.o3 o.o~ o,o5 o.o 0.01 0.02 0.03 0.04 0.05
(To--T a )li(°C m21W )

Fig. 5. Efficiency as a function of air temperature increment per unit incident flux for different air velocities
in the Type II air heater.
 Design analysis of solar air heaters 601

80-
.c
'•E ~,
..~ E *
L = 8m

70 "°",Is'~ L~ o ",7 V= S ~ m x
< 22" - ~ .+<'

60
I",1$ ~ I/
I m/s ~ W

~0 I I I ~ U 1 I I [ !

-c E '~ L = 2m '~ L 6m

7O 4",15 -- k

"Is ~ ~ ,,o~
2 ",Is It
60

7 0.0
I
0.01 0.02
I I
0.03
i
0.0~ 0.05 0.0
I
0.01
{To-'re) / l (°£ m21W)
I
0.02
I I
0.03 0.04 0.05

Fig. 6. Efficiency as a function of air temperature increment per unit incident flux for different air velocities
in the Type III air heater.

~o ~
~c ~ L :,~m ~ "~ L :Sin
•.c '~ .c E <',,

70 <"," ~ ~ . . . . _" ~ - - . ~ "~ ".4 - -~ ~*~

60 - 2",/s
' ;

",I"
50[ - ///

AO I ~ I [ I ~ } I I I I

v~ s",/s ~ ~ ~ f ,. o-~ .~ •'c ~."

_ , - . .
~o ,' i ,'~-'~"

z.O I I I I I ~I I I I I I
0.0 0.01 0.02 0.03 (3.0z- 0.05 0.0 0-01 0 02 0.03 0.04 0.05
/ To- T~)JI (°C =Zw)
Fig. 7. Efficiency as a function of air temperature increment per unit incident flux for different air velocities
in the Type IV air heater.
602 C. CHOUDHURYand H. P. GARG

L =&m [-
L = 8m
a.~ % )
.°I
70 ~ cn -~

3rn/s~
4m/s -
.#,,
,.~

f I i ~ /

'°I
5O

.,,,,,J
~o J I I I ~ I I I I I

50
"~,~ L =2m L =6m
E~
47
.tZ
70 Cn

60

50
lm/sL.../L~_.~ //
I
t,O I ~ I I I ,,,J I I t I I
0.0 0,0~ 0.02 0.~ 0.0~ 0.05 "0.0 0.01 0.02 0-03 0.0/. 0.05
( To- To)ll (°C m 2 l W )

Fig. 8. Efficiency as a f u n c t i o n o f a i r ' t e m p e r a t u r e i n c r e m e n t per unit incident flux for different air velocities
in a T y p e V air heater.

75
rn = 100 k g ! h m" ~ ~ ~ /.mls

65 ~ ~ 465

L=2m ~ ~ ~ ~ ~Imls
4mlS 60

~rn
55 I I i I I .L. I I I I I 55 v

\& ~ ~~ V ~ : 5mls r~ : 150 k g / h m z~

65 ---Lx~V = 5 m l s -% ~ - m /s 70
• &ml$

60 65

:2m'~
55 60
6rn~Imls 6m
8m

50 [ I [ I l I I I I i I 55
0 ~ 8 ~2 16 20 2 6 10 1~ 18 22 26
d(cm)

Fig. 9. Efficiency as a function o f air channel d e p t h for different air c h a n n e l lengths o f a Type I air heater.
 Design analysis of solar air heaters 603

rn = 2 0 0 k g l h m 2
75 75
~'~C~ ~ ~ -~V = 5 r n l s

70 70

65 65

60 6O

Bm
55 I I I I I I I I 1 I I 55~

65 7O

60 65

55 60

8m
50 I I I J I I I I I I I 55
0 8 12 16 20 2 6 I0 1~ 18 22 26
d(crn)

Fig. 10. Efficiency as a function of air channel depth for different air channel lengths of a Type II air heater.

"/5

r~ = 200 k g l h m 2

~-\'~--.~_ _ v= Sm/s 70

6s

60
8m L = 2 r n ~ ~ 1 mls
Z.rn
~- 5O I I I i I I I I I I I 55
~-'X,- x,,v : s rn/s 2

~V=Sm/$ ~ = 50kglh m2
65

55

SO -- Z,m - - 6 ~ r n ~ 1 mls 6m 55
8m

Z.5 I I I I I I I I I I I 50
L 8 12 16 20 2 6 10 14 18 22
d(cm)

Fig. 11. Efficiency as a function of air channel depth for different air channel lengths of a Type III air
heater.
604 C. CHOUDHURY and H. P. GARG

70 75

"\~ r~ = 200 kglh m2

65 70

65
2mls

60
L: 8m

55
50 8 ml m/s L=2m ~ lm/s
&m

~5 I I I i t I i 50

V 5m/s r~ : 1 5 0 k g l h m 2
~'-~ n~ : 5 0 k g / h m 2
65

55 ~ ~~ , ~~ _ ~ 50

,o .
\ ,,

t,5 I I I I 1 I I 6m I 50
4 8 12 16 20 2 G 10 1~ 18 22 26
d(cm )

Fig. 12. Efficiency as a function of air channel depth for different air channel lengths of a Type IV air
heater.
65
= lOOkg/h m 2

\,~'~"~,-~V = 5mls ~-- r~= 200 kg Ih m 2

60 65

55

50

45 ,m - _ ,0

i t t 1 i L I I, I , I ~s

,~V: 5m15
55

5O 2mls ~ ~ 2mls

/,5
= ~
lmls ~g . . . . ~ Is
'0 I I 8m I I I I I I I I 6ml 5
z. 8 12 16 20 2 6 I0 It, 18 22 26
d(cm)

Fig. 13. Efficiency as a function of air channel depth for different air channel lengths of a Type V air heater.
 Design analysis of solar air heaters 605

100- n~ = 100kglh m2
L=2m
~"" ~m~6rn 8m5 m/s

5 . . . . _ . . . . . . 2rnls Orals

°o 2 t. 6 B I0 12 1J-

200k 2m
JL-l,,~m. ~ = so kg,h m=

'OOIttfl'
" "

/. rn/s
~ so

30 " ~ 3m/s

o 2 /- 6 8 1o
d(cm)
Fig. 14. Pressure drop as a function of air channel depth for different air channel lengths.

the width of the air heater should be fixed. Once the
heater width is chosen, the air flow rate can be fixed
by the requirement of the temperature increment level
in the system. All other parameters being determined,
the selection of the length and depth of the air channel
can be made by the upper limit of the pressure drop
experienced by the flowing air. For example, if the
pressure drop is assumed fixed at a value of 30 Pa and
the air flow rate at 200 kg/h m 2, then for flow channel
lengths of 2, 4, 6 and 8 m, the channel depths should
be chosen close to about 2, 3.6, 5.5 and 7.3 cm (Fig.
15), respectively. This would correspond to an
efficiency of about 61% (Fig. 13) and an air tem-
perature increment of about 0.013 K per unit incident
intensity ( T o - T a ) / I of solar radiation (Fig. 4). If
the required air temperature increment level for a
particular application is higher than this, for example,
in the range of about 0.02-0.025 K m2/W, which
corresponds to a mass flow rate of 100 kg/h m 2 (Figs
4--8), for air channel lengths of 2, 4, 6 and 8 m, the air
channel depths should be, respectively, (assuming a
pressure drop of 30 Pa) 1.5, 2.5, 3.5 and 4.6 cm (Fig.
14) which would correspond to an efficiency of about
59-70% (Figs 9-13). In summary, by adopting a
606 C. CHOUDHURYand H. P. GARG

= 200kglh m 2
50
L=2m
30

20
a.

IO

6 I
0 2 ~ 6 8 IO 12 14 16

r~ = 150kglh m 2
50 L=2m

I
°o 2 z 6 8 1o 12 IA 16
d(cm)

Fig. 15. Pressure drop as a function of air channel depth for different air channel lengths.

theoretical analysis and the curves discussed above, it D equivalent diameter of air channel, m
would be possible to obtain economical and tech- f friction factor
h heat transfer coefficient, W/m 2 K
nically logical air heater designs. I solar radiation incident on the collector, W/m 2
k thermal conductivity of air, W/m K
L length of air heater, m
4. CONCLUSIONS ~t mass flow rate of air, kg/h
Clearly, increasing the air velocity through the air n) specific mass flow rate of air, kg/h m 2
M constant dependent of N
heaters results in higher collection efficiency, but also N number of equivalent diameters
in increased pressure drop. A reasonable com- Nu Nusselt number for any N
promising point can be determined by investigating Nuoo Nusselt number for fully developed thermal flow
the effects of various design and operational par- Pr Prandtl number
Re Reynolds numbr
ameters on the efficiency, the air temperature V velocity of air flowing through the collector, m/s
increment and the pressure drop in the system and by W width of absorber, m
adopting the method of design analysis discussed in ct solar absorptance of corrugated absorber
the article. This process should be of interest to design- fl corrugation factor
ers, engineers and consumers who like to compare the AP pressure drop experienced by the flowing air, Pa
~t efficiency of system
cost-effectiveness of commercial products. /~ dynamic viscosity of air, kg/m s
p density of air, kg/m 3
NOMENCLATURE
Suffix :
C specific heat capacity of air, Wh/kg °C a ambient
d depth of air channel, m f fluid
 Design analysis of solar air heaters 607

o outlet entrance region on turbulent forced convective heat trans-

1 upper plate fer for an asymmetrically heated rectangular duct with
2 middle plate uniform heat flux. Solar Energy 12, 513 (1969).
3 bottom plate 9. H.Y. Wong, Handbook of Essential Formula and Data on
Heat Transfer for Engineers. Longrnan, London (1977).
REFERENCES
1. M. K. Seluck, Solar Air Heaters and Their Applications
in Solar Energy Engineering (Edited by A. A. M. Sayigh).
APPENDIX
Academic Press, New York (1977).
2. G. O. G. Lof, Solar Air Heating Systems in Solar Energy P~ = hl~+h:2~
Conversion (Edited by A. E. Dixon and J. D. Leslie). P2 --" h21 q-h23q-h2f
Pergamon Press, New York (1979). P3 = h~l/Pi
3. J. A. Duffle and W. A. Beckman, Solar Engineering Ther- P4 = P3/h21
mal Process. John Wiley and Sons, New York (1980). P5 = P4hl~
4. N. K. Bansal, R, Chandra and M. A. S. Malik, Solar Air P6 = h23 q-h3a"l-h3r
Heaters, Review of Renewable Energy Sources (Edited by P7 =- h23/(P2 - P3)
M. S. Sodha, S. S. Mathur and M. A. S. Malik). Wiley Ps = PTh23
Eastern, India (1983). P9 -----P7h2r
5. W. W. S. Charters, Some aspects of flow duct design of P~o = P6-P8
solar air heating applications. Solar Energy 13, 283 ( 1971). PI1 = PsPT+h3a
6. K. G. T. HoUands and E. C. Schewen, Optimization of PI2 = Pg+h3f
flow passage geometry for air heating plate-type solar Pi 3 = h2t(S21+ S~IP4 + TaP~)/(P2 - P3)
collectors. J. Solar Energy Engng, Trans. A S M E 103, 323 PI4= Pg+h3f
(1981). Pl5 = h~r/(P2- P3) - (h2f+ h3r)
7. C. Choudhury, S. L. Andersen and J. Rekstad, A solar Pi6 = PI4(Sfl+ SflP'7+ SIIP4P7+ T~,Pll)
air heater for low temperature applications. Solar Energy PL7 = PI2PI4/Plo
40, 77 (1988). Pls = Pta+Pi6
8. H. M. Tan and W. W. S. Charters, Effect of thermal P19 = Pis+Pi7