Applied Thermal Engineering 221 (2023) 119673

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

Research Paper

Thermo-hydraulic performance of solar air heater with finned absorber

plate forming multiple rectangular air flow passages in parallel under
laminar flow conditions
Rajendra Karwa
Department of Mechanical Engineering, Jai Narain Vyas University, Jodhpur 342011, India

A R T I C L E I N F O A B S T R A C T

Keywords: The paper presents results of an analysis carried out using the governing equations to compare the heat transfer
Solar air heater (collector) performance of a uniformly heated flat plate with longitudinal fins, which form multiple rectangular air flow
Finned absorber plate passages in parallel below the plate and cover whole of the passage surface, with that of the low aspect ratio
Multiple rectangular parallel flow passages
rectangular duct typical of conventional flat plate solar air heater at equal pumping power under laminar flow
Laminar flow
conditions. The heat transfer rate at equal pumping power for the plate with 1 mm thick steel fins at pitch of 25
Heat transfer enhancement
Thermal efficiency mm and 40 mm is found to be about 2.7 and 2.5 times, respectively, over the duct without fins. The heat transfer
study is followed by thermo-hydraulic performance study of the solar air heater with absorber plate having
longitudinal fins forming multiple rectangular air flow passages in parallel. The finned solar air heater with low
air mass flow rates has been found to provide 28–31 % better thermal efficiency than the conventional solar air
heater at equal pumping power. The results of the study can be utilized for the development of enhanced per­
formance finned solar air heater operating with low air mass rates to deliver high temperature air.

out for transitional to early turbulent flow conditions corresponding to

1. Introduction air flow rate of 0.01–0.06 kg/s per unit area of the absorber plate of the
solar air heater delivering heated air at low to moderate temperatures;
The conventional flat plate solar air heater (SAH) consists of a typically, a maximum of 30 ◦ C rise in air temperature is reported [5]. For
smooth surfaced absorber plate with rectangular air flow duct. The some applications, such as space heating, requiring air at higher tem­
absorber plate is subjected to uniform heat flux of the solar radiation. peratures, air flow rates lower than 0.01 kg/(s m2) are to be used. At
The thermal efficiency of such solar air heater is low because of a low these air flow rates, the flow is in the laminar regime and the heat
value of the heat transfer coefficient between the absorber plate and the transfer coefficient is quite low leading to very low thermal efficiency of
air flowing through the duct, which leads to a high operating tempera­ the solar air heater; typically, at mass flow rate of 0.005 kg/(s m2), the
ture of the absorber plate and hence greater heat loss from the air heater. thermal efficiency is reported to be 25–30 % for solar insolation of 1000
One of the most promising methods used for thermal performance W/m2 and air temperature rise of 50–60 ◦ C [12].
improvement is the employment of rib roughness of various shapes The artificial roughness elements employed for heat transfer
(circular, rectangular or chamfered) and in different arrangements with enhancement under transitional to early turbulent flow conditions are
respect to the flow direction (transverse continuous, angled continuous, ineffective in heat transfer enhancement under laminar flow condition
transverse broken, V-shaped or discrete V-shaped) on the air flow side of because the flow remains in hydraulically smooth regime defined by
the absorber plate of solar air heater or on one uniformly heated wall of a [13].
rectangular duct with remaining three smooth walls insulated to In asymmetrically heated rectangular ducts, the air temperature
enhance the heat transfer coefficient from absorber plate or the heated decreases from the heated absorber plate to the lower insulated side of
wall by creating near wall turbulence [1–8]. Recently the researchers the solar air heater ducts with transitional to early turbulent flow con­
[9–11] have carried out studies on equilateral triangular ducts with ditions. Looking to this fact, the researchers [14–16], to mention a few,
artificial roughness and uniform heat flux at one of the walls of the duct found that thin perforated baffles extending deeper into the flow are
keeping the remaining two walls adiabatic. They have recommended thermo-hydraulically very effective in enhancing heat transfer.
such ducts for the solar air heaters. All these studies have been carried Maheshwari et al. [14] employed half-perforated baffles transverse to

E-mail addresses: [email protected], [email protected].

https://doi.org/10.1016/j.applthermaleng.2022.119673
Received 23 May 2022; Received in revised form 10 September 2022; Accepted 14 November 2022
Available online 28 November 2022
1359-4311/© 2022 Elsevier Ltd. All rights reserved.
R. Karwa Applied Thermal Engineering 221 (2023) 119673

Nomenclature SAH solar air heater

Ta ambient temperature, K
A absorber plate area = WL, m2 Tgi glass inner surface temperature, K
Ac cross sectional area of duct/passage, m2 Tgo glass outer surface temperature, K
Acf cross sectional area of fin, m2 Ti inlet air temperature, K
Ae area of side of solar air heater, m2 Tm mean air temperature = (To + Ti)/2, K
Af effective fin surface area for heat transfer, m2 Tmpg mean of the plate and glass temperatures = (Tp + Tgi)/2, K
Ap total heat transfer area for the finned plate, m2 To outlet air temperature, K
C a factor = fapp/Re Tp mean plate temperature, K
cp specific heat of air, W/(kg K) Ts sky temperature, K
dh hydraulic diameter of duct/passage, m UL overall loss coefficient, W/(m2 K)
E performance parameter = (hp Ap)/(hA) Vw wind velocity, m/s
f Fanning friction factor w width of rectangular passage, m
fapp apparent friction factor W width of the collector, m
FR heat removal factor x+ dimensionless axial distance for friction factor = x/(dh Re)
F′ efficiency factor x* dimensionless axial distance for heat transfer = x/(dh Re
G mass flow rate per unit area of plate = M/A, kg/(s m2) Pr)
Gr Grashof number
h heat transfer coefficient, W/(m2 K) Greek symbols
H air flow duct height, m α* duct aspect ratio
Hp height of rectangular passage, m δ absorber plate and fin thickness, m
hw wind heat transfer coefficient, W/(m2 K) δg glass cover thickness, m
I solar radiation on the collector plane, W/m2 δi insulation thickness, m
k thermal conductivity of air, W/(m K) β collector slope, deg
kf thermal conductivity of fin material, W/(m K) δpg gap between absorber plate and glass cover, m
kg thermal conductivity of glass, W/(m K) Δp pressure drop, Pa
L length of collector, m ΔT air temperature rise = To – Ti, K
Lc corrected length of fin, m ΔT/I temperature rise parameter = (To - Ti)/I, K.m2/W
Lfin length of fin, m εg emissivity of glass surface
M mass flow rate, kg/s εp emissivity of absorber plate surface
mfin fin parameter, 1/m ηfin fin efficiency
ṁ mass velocity of the air, kg/(s m2) ηth thermal efficiency of collector
n number of parallel flow passages μ dynamic viscosity, Pa s
nf number of fins νmpg kinematic viscosity of air at temperature Tmpg, m2/s
Nu Nusselt number ρ density of air, kg/m3
p fin pitch, m τα transmittance-absorptance product
P pumping power, W Subscripts
P′ fin perimeter, m b back of collector
Pr Prandtl number = μcp/k g glass
Q useful heat gain, W H1 a thermal boundary condition
Qb back loss, W m mean
Qe edge loss, W p absorber plate
QL total heat loss, W p parallel rectangular passages
Qt top loss, W pg between absorber plate and glass cover
Ra Rayleigh number R rectangular duct of conventional air heater
Re Reynolds number

the flow, Chamoli and Thakur [15] studied the effect of height of fins with baffles, and Yang et al. [21] and Benseddik et al. [22] studied
perforated baffles in V down arrangement, and Habet et al. [16] studied the influence of the use of offset strip fins mounted in staggered fashion
the effect of partially tilted baffles at different angles and perforated in on the absorber plate parallel to the airflow and found that the heat
various ratios mounted in a staggered arrangement on the uniformly transfer enhancement is also accompanied with heavy pressure drop
heated top and bottom surfaces of a rectangular channel with flow in penalty. Yeh and Ho [23] studied theoretically the influence of external
turbulent regime (Re = 12000 to 42000). However, the baffled are also recycle on the efficiency of solar air heater with longitudinal fins. The
not effective in the laminar flow regime. external recycle involves mixing of the heated outlet air with the inlet
In yet another strategy to enhance thermal efficiency of the solar air air, which increases the air velocity in the duct. The maximum tem­
heaters, the researchers [17–24] employed fins on the absorber plate. perature rise of the air is reported to be only15◦ C at solar radiation in­
Naphon [17] studied double-pass rectangular duct solar air heater with tensity I = 830 W/m2. Saravanan et al. [24] employed C-shaped copper
longitudinal fins in both channels with temperature rise < 25 ◦ C while baffles in staggered fashion placed transverse to the air flow direction,
Fudholi et al. [18] used fins in the second channel only. Jalil et al. [19] which also served as fins because the baffles were brazed to the absorber
studied double-pass solar air heater with wavy fins oriented in the flow plate. Heat transfer enhancement at equal pumping power is reported to
direction. The pressure drop in the finned duct increases due to the be 1.36 at the lowest Reynolds (Re = 3000) of their study.
increased wetted perimeter; the greater the number and height of the Jalil et al. [25] studied experimentally and numerically the thermal
fins greater is the pressure drop. Ho et al. [20] employed longitudinal performance of solar air heater with air flow across the longitudinal axis

2
R. Karwa Applied Thermal Engineering 221 (2023) 119673

of the air heater in zig zag fashion along straight or wavy fins. Shetty (a). In Fig. 1(b) is shown a flat absorber plate having longitudinal fins
et al. [26] carried out experimental investigation on a solar air heater welded to it, which when assembled forms n number of rectangular air
with flow perpendicular to plane of the perforated circular absorber flow passages in parallel below the plate. The fins form the vertical side
plate for thermal performance augmentation. Both of these schemes walls of the passages and also cover the bottom face of each rectangular
involve very high pressure drop and pumping power requirements. passage and thus the area for the heat transfer is significantly increased.
Sevik and Abusk [27] carried out experimental study on solar air As compared to the simple rectangular duct of the conventional solar
heater with eight 102 mm diameter circular cross-section semi-flexible air heater, the multiple rectangular flow passages increase the wetted
aluminium foil (90 µm thick) ducts arranged in parallel in the rectan­ perimeter for the flow and hence the pressure drop and pumping power
gular duct of the solar air heater. As compared to the flat plate solar air requirement increase. The equal pumping power condition can be ach­
heater, 18.82 % increase in the maximum thermal efficiency is reported ieved by suitably increasing the height Hp of the multiple rectangular
when operating at air mass flow rate providing about 20–40 ◦ C rise in air passages, which has been determined as explained in the next section.
temperature. Zukowski [28] demonstrated that the use of air jets from
rectangular section nozzles impinging on the absorber plate surface to
enhance heat transfer by destroying the laminar sublayer can improve 2.1. Equal pumping power condition
the efficiency of a single glass cover solar air heater from 16 % up to 24
% when the air heater delivered heated air with 5–18 ◦ C temperature The condition of the equal pumping power for the conventional air
rise. heater rectangular duct and the multiple rectangular parallel flow pas­
Since the heat transfer coefficient in the laminar flow regime is very sages under laminar flow conditions can be found by equating the
low, the fins would be very effective in the enhancement of heat transfer pumping power equations of the two schemes for the same mass flow
rate but would be accompanied with increased pressure drop. Further, in rate of the air.
the studies conducted so far, the air flow duct surface opposite to the
finned absorber plate is adiabatic and does not contribute to the heat 2.1.1. Pumping power for rectangular duct of conventional flat plate solar
transfer. Looking to these facts, the objective of the present study was to air heater
present a scheme of fins on the absorber plate such that whole of the The equation of pumping power is
surface of the airflow passage contribute to the heat transfer and this ( )
M
enhancement in heat transfer is achieved with no pumping power pen­ PR = Δp (1)
ρ
alty. The analysis was also carried out to compare the thermal effi­
ciencies of the finned absorber plate and conventional solar air heaters where M is the mass flow rate of air through the duct and Δp is
followed by study of the effects of design and ambient parameters on the pressure drop.
finned solar air heater performance. The analysis has also covered the For laminar flow in a rectangular duct, the friction factor is a func­
effect of the arrangement of the finned solar air heater modules in an tion of the Reynolds number Re, the flow development condition and the
array. duct aspect ratio α*, which is defined as the ratio of the short side to long
side of the rectangular section duct. The friction factor in the hydrody­
2. Uniformly heated flat plate with fins forming multiple namic entrance region, termed as apparent friction factor fapp, is higher
rectangular flow passages in parallel below the plate than that for the fully developed case. Hence, in general, the pressure
drop Δp = 4fappL (m)˙ 2 /(2ρdh). This gives
The conventional rectangular duct solar air heater is shown in Fig. 1 ( )( ) ( )( )
M 4fapp Lṁ2 2ML fapp
PR = = ṁ2 (2)
ρ 2ρdh ρ 2 dh
Solar radiation I Glass
Absorber where ṁ is the mass velocity of the air in the duct.
plate cover
pg For cross-section area Ac = WH of the rectangular section duct, refer
Fig. 1(a), the mass velocity is
( )
W H M
ṁ = (3)
WH
Back insulation Air flow duct W >> H Substitution gives
(a) ( )( )( )2
2ML fapp M
PR = (4)
Finned Solar radiation I ρ2 dh WH
absorber plate Weld
Hollands and Shewen [29] recommended duct height H ≤ 10 mm
w Hp depending on the collector length for mass flow rate of 0.0025–0.02 kg/s
per m2 of the absorber plate. Karwa and Chauhan [5] considered duct
Fins height of 5–20 mm for flow rate of 0.01–0.06 kg/(s m2). Since the width
(i)
W of a solar air heater duct is of the order of 1 m, the geometry of interest
from analysis point of view can be considered as a parallel plate duct for
W = nw H ≤ 10 mm. For a parallel plate duct, the hydraulic diameter dh = 2H.
Using this, Eq. (4) gives
( )( 3 )
L M
Parallel flow passages PR = f (5)
ρ2 W 2 H 3 app
(ii)
Based on the solutions for laminar boundary layer development over
(b)
a flat plate and the fully developed flow in circular and noncircular
Fig. 1. (a) Rectangular cross-section duct of the conventional solar air heater, ducts, the following correlation has been given by Shah [30] for calcu­
(b) (i) the absorber plate with longitudinal fins, (ii) assembled view (fins lating the apparent friction factor fapp in the hydrodynamic entrance
forming multiple rectangular air flow passages in parallel). region in a rectangular duct including a parallel plate duct.

3
R. Karwa Applied Thermal Engineering 221 (2023) 119673

0.5
K(∞)
+ fRe − 3.44(x+ )− ( )( )( )2
fapp Re = 3.44(x+ )− 0.5
+ 4x+
(6) 2ML fapp m
1+C ′
(x+ )− 2 Pp = (15)
ρ2 dhp wHp
where x , defined as x/(dh Re), is the dimensionless axial length and
+
For the rectangular flow passage, let fapp = Cp/Re. Then
refers to the hydrodynamically developing flow. The values of factors K
( )( )( )2
(∞) and Ć in Eq. (6) are given in Table 1 [30]. Factor (f.Re) refers to the 2ML Cp /Re m
Pp = (16)
fully developed flow condition and its analytical correlation is [31] ρ2 dhp wHp

fRe=24[1− 1.3553α* +1.9467(α* )2 − 1.7012(α* )3 +0.9564(α* )4 − 0.2537(α* )5 The Reynolds number for a rectangular passage is
(7) (
ṁdhp
) ( )(
1 m
)
Re = = dhp (17)
Shah and London [in 32] have reviewed and compared several μ μ wHp
analytical and experimental investigations of hydrodynamically devel­
This transforms the pumping power equation to
oping flow in rectangular ducts. They concluded that the friction factor
results from Eq. (6) are in good agreement with the experimental data ( )( )( )2
2ML μwHp Cp m
[32]. Pp =
ρ2 mdhp 2 wHp
For a parallel plate duct, α*= 0. The error between the predicted
values of fapp.Re from Eq. (6) and the corresponding numerical pre­ ( )( )( )
2ML μCp m
dictions of Curr et al. [33] is reported to be of the order of ± 2 % [30]. = (18)
Presenting fapp = CR/Re for the parallel plate duct under discussion, ρ2 dhp 2 wHp
Eq. (5) gives The mass flow rate m of air through the single passage is M/n. Using
( )( 3 )( )
L M CR this relation and simplifying, the above equation transforms to
PR = (8)
ρ2 W 2 H 3 Re ( 2 )( )( )
2M L μC p 1
Pp = (19)
The Reynolds number equation for flow through a parallel plate duct ρ 2
dhp 2
nwHp
is
( ) ( )( ) ( )( ) The condition of equal pumping power is now obtained by equating
Re =
ṁdh
=
1 M
2H = 2
1 M
(9) Eqs. (11) and (19), which gives
μ μ WH μ W ( )( )
CR 4Cp 1
This gives = 2
(20)
WH 3 dhp nwHp
( )( 3 )
L M C
PR = ( )R (10) For the rectangular passage of width w and height Hp, the hydraulic
ρ2 W 2 H 3 2 1 ( M )
μ W diameter is
( ) ( )
Simplification of the above equation gives 4Ac 2wHp
dhp = = (21)
( ) P w+Hp
CR ( μ ) M 2
PR = L 2 (11)
2 ρ WH 3
Substitution and simplification give
( ) ( )
CR Cp 1
2.1.2. Pumping power for multiple rectangular air flow passages in parallel 3
= 2
(w+Hp )2 (22)
WH 2
w Hp nwHp
For n parallel rectangular passages each of width w and height Hp, as
shown in Fig. 1(b), the pumping power equation is The relations for estimating thermo-physical properties of the air are
(m) [34]
Pp = n Δp (12) ( )0.86
ρ Tm
k = 0.0257 (23a)
where m is mass flow rate through a single passage. 293
Using the total mass flow rate through the n parallel passages, M =
( )0.0155
nm and pressure drop Δp = (4fappL ṁ2 )/(2ρdhp), the pumping power is cp = 1006
Tm
(23b)
( )( ) ( )( ) 293
M 4fapp Lṁ2 2ML fapp
Pp = = ṁ2 (13) ( )0.735
ρ 2ρdhp ρ 2 dhp Tm
μ = 1.81 × 10− 5
(23c)
293
The mass velocity of air ṁ through a passage of cross-section area
wHp is (
293
)
( ) ρ = 1.204 (23d)
m Tm
ṁ = (14)
wHp where Tm is the mean temperature of air at which the properties are
Substitution in Eq. (13) gives to be determined.
A small value of passage width w gives a smaller length of the fin, Lfin
≈ Hp + w/2, and hence the fin efficiency would be high giving a greater
effective surface area of the fins for heat transfer. Thus, a low value of w
Table 1
is desirable from the heat transfer enhancement consideration. How­
Values of K(∞), f.Re, and C’ ́ to be used in Eq. (6) [30].
ever, too small value of width w may pose some difficulty in the welding
α* f.Re K(∞) C’ Error of the fins to the absorber plate.
1.0 (square duct) 14.227 1.43 0.00029 ±2.3 % To determine the fin efficiency and heat transfer enhancement due to
0.5 15.548 1.28 0.00021 ±1.9 % the employment of the fins, the pitch of the fins p (=w + δ) has been
0.2 19.071 0.931 0.000076 %
±1.7
considered to be 25 mm and 40 mm, where δ is the thickness of the fin.
0.0 (parallel plate duct) 24 0.674 0.000029 ±2.4 %

4
R. Karwa Applied Thermal Engineering 221 (2023) 119673

For these values of fin pitch, the number of parallel flow passages n 16
(=W/p) would be 40 and 25, respectively. 15
Karwa [35] recently presented results of a study of heat transfer
14
performance of multiple equilateral triangular parallel air flow passages
13
below a uniformly heated flat plate having inclined fins covering the
side walls of the passages for transition to early turbulent regimes cor­ 12
responding to mass flow rate of 0.01–0.06 kg/s per m2 of the plate. The 11
reported fin efficiency values are about 0.71–0.90 and 0.76–0.93 for 10
steel fins of 30 mm length and thickness 1 mm and 1.5 mm, respectively;

Num
9
the higher efficiency values correspond to the lowest flow rate of 0.01
kg/(s m2). At mass flow rates ≤ 0.01 kg/(s m2), steel fins of 1 mm
8 Eq. (27)
thickness can be expected to provide fin efficiency of about 90 %. 7
Solar air heater module of length L = 2 m and width 1 m has been 6 * = 0.25
considered, which are the normally accepted values for the module from 5.331
5 * = 0.5
the constraints of available sizes of plywood and glass sheets, and the 4.123
4
ease of installation and handling. It has been mentioned earlier that a Eq. (26)
3
small duct height H of the order of 5–10 mm is recommended for the
solar air heaters with low mass flow rates under consideration here. 2
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
In order to determine the height of the rectangular passage Hp to
satisfy the condition of the equal pumping power, the values of various x* = x/(dh Re Pr)
geometric and operational parameters considered are: fin thickness δ =
1 mm, fin pitch p = 25 mm and 40 mm giving passage width w (=p – δ) of Fig. 2. Mean Nusselt number Num versus dimensionless distance x*.
24 mm and 39 mm, respectively, flow rate through a passage m = M/n,
mean air temperature in the duct/passage for the estimate of the thermo-
physical properties of the air Tm = 303 K for inlet air temperature Ti =
Weld
278 K and air temperature rise of about 50 ◦ C. w
Starting with trial values of CR = 25 and Cp = 16, and subsequent
estimates of CR and Cp from Eq. (6), trial and error solution of Eq. (22)
Hp
gives the rectangular passage height Hp ≈ 11.85 mm for w = 24 mm and
Hp ≈ 11.1 mm for w = 39 mm for rectangular duct height H = 10 mm. Fins Weld
Uniformly heated flat plate
2.2. Heat transfer coefficient correlations
w
The Nusselt number Nu = 5.385 for the fully developed flow in
parallel plate duct with uniform heat flux on one of the walls and other Hp
wall insulated [32]. For the simultaneously developing laminar flow in
parallel plate duct with uniform heat flux on one of the walls and other
wall insulated, Holland and Shewen [29] used in their analysis the
following relation of the mean Nusselt number deduced from Kays and Fin length Lfin = Hp + (w 1)/2
Perkins [36].
( ) Fig. 3. Control volume for fin analysis.
H
Num = 5.385 + 0.148Re for Re < 2550 (24)
L number NumH1 values from [35,37,38] are plotted as function of the
where the second term on the right-hand side of the equation takes dimensionless axial distance x* in Fig. 2, where the term x*, expressed as
account of the effect of the developing flow. x/(dh Re Pr), refers to simultaneously developing flows.
For the fully developed laminar flow in a rectangular duct with all For α* = 0.5, the plotted data are correlated as
four walls transferring heat (termed as H1 boundary condition, which
Num = − 18.326(x* )3 + 31.162(x* )2 − 15.766(x* ) + 7.0533
refers to constant axial wall heat flux with constant peripheral wall (26a)
for 0.1 ≤ x* ≤ 0.4
temperature corresponding closely to the boundary condition for the
rectangular passage under the present study), the analytical correlation
for Nusselt number is [31]

NuH1 = 8.235[1 − 2.0421α* + 3.0853(α* )2 − 2.4765(α* )3 + 1.0578(α* )4 − 0.1861(α* )5 (25)

Num = 4.123(x* )− 0.11

It is to note that analytical correlations given by Eqs. (7) and (25) for (26b)
for 0.4 < x* ≤ 1.0
fully developed friction factor and Nusselt number, respectively, have
been verified by many researchers [31]. For α* = 0.25, the data are correlated as
For simultaneously developing flows in rectangular ducts of aspect
Num = − 13.534(x* )3 + 23.14(x* )2 − 11.895(x* ) + 7.6216
ratio α* of 0.5 and 0.25 with H1 boundary condition, the mean Nusselt (27a)
for 0.1 ≤ x* ≤ 0.4

5
R. Karwa Applied Thermal Engineering 221 (2023) 119673

5.0
Num = 5.331(x* )− 0.073
(27b)
for 0.4 < x* ≤ 1.0 4.5 w = 24 mm, Hp = 11.85 mm
- - - - w = 39 mm, Hp = 11.1 mm
3. Fin analysis and heat transfer enhancement 4.0

Steel fins of 1 mm thickness have been considered here. The fins can 3.5
be easily welded to the flat steel sheet. A gap of 1 mm has been provided

E = (hp Ap) /(hA)

3.0
between the ends of the adjacent fins to felicitate manufacturing. This
gap also provides an easy access for cleaning of the air passages, which 2.5
can be easily approached by simply unscrewing and moving the finned
plate upwards, refer Fig. 1. 2.0
For the fin analysis, the vertical part of the fins has been assumed to
be split into two parts each of thickness δ/2 as shown in Fig. 3. Thus, for 1.5
the passage under consideration bounded by the control volume shown
in the figure, the heat is transferred only from one side of the fins in the 1.0
passage (from the fin surface in contact with the air flow). However, the
0.5
lower (bottom part) of the fin is of thickness δ, which will provide
somewhat greater fin efficiency than the estimate being made here. 0.0
The effective surface area of a fin for heat transfer is a product of the 0.004 0.005 0.006 0.007 0.008 0.009 0.01 0.011
fin efficiency ηfin and its surface area. For thin short fins, the fin effi­
G, kg/(s m2)
ciency is given by [34]
Fig. 4. Heat transfer performance parameter E versus the mass flow rate per
tanh(mfin Lc )
ηfin = (28) unit area of the plate G.
mfin Lc

where mfin = √(hP΄/kf Acf) is a fin parameter. P΄ is fin perimeter, Acf been evaluated by a performance parameter defined as [39]
is fin area of cross-section and kf is the thermal conductivity of the fin
h p Ap
material. For a thin rectangular cross-section plate fin of thickness δ/2 E= (32)
hA
transferring heat from one side, Acf /P΄ = δ/2 because P΄ = L. Hence, mfin
= √(2 h/kf δ). The term Lc in Eq. (28) is the corrected fin length and, for where the product hA refers to the conventional solar air heater
a short fin, it is given by [34] having rectangular duct with flat absorber plate without fins for which
A = WL.
Lc = Lfin + ΔL (29)
Fig. 4 shows the plot of the performance parameter E = (hpAp)/(hA)
where ΔL = Acf /P΄ = δ/2 and Lfin = Hp + (w – gap)/2. as function of the mass flow rate of air per unit area of plate G. The value
The parameters considered for heat transfer enhancement study are of the parameter is found to be 2.7–2.75 and 2.47–2.57 for fin pitch p =
given in Table 2. For each rectangular flow passage with fins at pitch p, 25 mm and 40 mm, respectively. This clearly shows the significant
the width of the rectangular passage w = p – δ and the number of parallel advantage of the use of the finned absorber plate in the laminar flow
flow passages n equals W/p for collector width W. Each passage consists condition and hence the further analysis will concentrate on the solar air
of two fins hence the total number of fins nf is 2n. Since one side of each heater with finned absorber plate. The lower value of enhancement
fin in the passage is transferring the heat as shown in the figure, the when the fin pitch is increased is mainly due to the lower fin efficiency of
effective fin surface area for heat transfer is 0.835–0.845 than the fin efficiency of 0.90–0.91 for p = 25 mm because
of the increased fin length of 30.1 mm in the case of p = 40 mm from
Af = nf (ηfin Lc )L (30) about 23.35 mm for p = 25 mm. The fin efficiency of about 90 % when 1
The total heat transfer area for the plate with fins is mm thick steel fins are used with p = 25 mm can be considered as quite
satisfactory.
Ap = nwL + Af (31) The enhancement in the heat transfer rate with the employment of
The heat transfer rate from the finned plate is proportional to the finned absorber plate will lead to a significant increase in the thermal
product of the heat transfer coefficient hp and the total heat transfer area efficiency of the solar air heater with finned absorber plate. In the next
Ap. The heat transfer enhancement due to the employment of fins has section is presented the thermal analysis of the finned plate solar air
heater and the performance has also been compared with the conven­
tional solar air heater.
Table 2
Parameters for heat transfer enhancement study.
4. Thermo-hydraulic performance analysis of the solar air
Collector width, W 1m
heaters
Fin pitch, p 25 mm and 40 mm
Absorber plate and fin thickness, δ 1 mm
Rectangular passage width, w = p – δ 24 mm and 39 mm The thermal performance of the solar air heaters has been evaluated
Rectangular passage height for equal 11.85 mm and 11.1 mm using the governing heat transfer equations. The useful heat gain has
pumping power, Hp been determined from net of the incident solar radiation absorbed by the
Passage height to width ratio, Hp/w 0.33, 0.5
absorber plate after transmission through the glass cover and the esti­
Fin length, Lfin = Hp + (w – gap)/2 11.85 + 23/2 = 23.35 mm
and 11.1 + 38/2 = 30.1 mm mate of the heat loss from the collector, and equating the same to the
Number of fins, nf (2 in each passage) 80 and 50 convective heat transfer from the absorber plate to the air using
Absorber plate and fins material Steel appropriate heat transfer correlations for the geometry of the air flow
Thermal conductivity of fin material, kf 30 W/(m K)
ducts.
Air flow rate per unit area of plate, G 0.005–0.01 kg/(s m2)
Reynolds number, Re 1080–2200 for the rectangular duct
The thermal performance analysis method employed here is similar
740–1750 for the multiple rectangular to those used by [5,12,14,40] for the performance prediction of solar air
passages in parallel heaters with transitional to early turbulent flow having rib roughness on

6
R. Karwa Applied Thermal Engineering 221 (2023) 119673

Solar radiation I

Finned absorber plate Glass cover

Insulation n parallel flow

Back
passages
(a)

Ta
Absorber Top loss, Qt
Glass cover
plate I( )
Tg Tp
pg

Air in Tm Q Air out

i
Back insulation Back loss, Qb Edge loss, e

Longitudinal section Side view

(b)

Fig. 5. (a) Finned solar air heater with multiple rectangular air flow passages (b) the heat balance on the solar air heater.

the air flow side of the absorber plate or perforated baffles attached to absorber plate at mean temperature Tp to the inner surface of the glass
the absorber plate. cover at temperature Tgi by radiation and convection. Hence,
In Fig. 1(a), the cross-section of the conventional rectangular duct ( )( 1 )− 1
1 ( )]
solar air heater has been shown. Fig. 5(a) shows the schematic diagram Qtpg = A[σ Tp4 − Tgi4 + − 1 + hpg Tp − Tgi (36a)
of the solar air heater with finned absorber plate and a single glass cover. εp εg
Referring Fig. 5(b), the heat balance on the air heater gives the distri­ where εp and εg are emissivity values of the plate and glass surfaces,
bution of incident solar radiation I on the collector plane into the useful respectively, and hpg is the convection heat transfer coefficient between
heat gain or heat collection rate Q and various heat losses. The useful absorber plate and glass cover.
heat gain or heat collection rate can be expressed as From the inner surface of the glass cover, heat flows to the outer
Q = AI(τα) − QL (33) surface by conduction and is expressed by
( )/
where A is the area of the absorber plate receiving the solar radia­ Qtg = kg A Tgi − Tgo δg (36b)
tion, (τα) is the transmittance-absorptance product of the glass cover- where kg is the thermal conductivity of the glass, δg is glass cover
absorber plate combination and QL is the total heat loss from the col­ thickness and Tgo is temperature of the outer surface of the glass cover.
lector, which is a sum of top Qt, back Qb, and edge loss Qe as depicted in From the outer surface of the glass cover, the heat is rejected by
Fig. 5(b). radiation to the sky at temperature Ts and by convection to the sur­
The collected heat Q is transferred to the air flowing through the air rounding hence
heater duct. Hence, ( ) ( )]
Q = GAcp (To − Ti ) (34)
4
Qtgo = A[σεg Tgo − Ts4 + hw Tgo − Ta (36c)

where G is mass flow rate of air per unit area of the absorber plate, where hw is termed as wind heat transfer coefficient, which is a
and Ti and To are the air inlet and outlet temperatures, respectively. The function of the wind velocity.
total mass flow rate M = GA. In the equilibrium, Qtpg = Qtg = Qtgo = Qt.
Estimating the Nusselt number from the appropriate heat transfer The sky temperature Ts is a function of many parameters. For clear
coefficient correlations, Eqs. (24)-(27), the heat transfer coefficient h sky condition assumed in the present analysis, Swinbank’s formula [41]
between the absorber plate and the air flowing through the collector given in Table 3 is widely used.
duct is determined. The heat collection rate from the convective heat The convective heat transfer coefficient between the absorber plate
transfer consideration is and glass cover hpg has been estimated using the following three-region
( ) correlation of Buchberg et al. [42], which has been shown by them to
Qh = hA Tp − Tm (35) confirm well to the experimental data:
where Tm = (Ti + To)/2 is the mean temperature of air in the air ( )+
1708
heater duct or passage. (37a)
′
Nu = 1 + 1.446 1 − ′ for 1708 ≤ Ra ≤ 5900
Ra
(the + bracket goes to zero when negative).
4.1. Top loss
Nu = 0.229(Ra’ )0.252 for 5900 < Ra’ ≤ 9.23 × 104 (37b)
The top loss Qt from the air heater has been calculated from iterative
solution of basic heat transfer equations. The heat transfers from the

7
R. Karwa Applied Thermal Engineering 221 (2023) 119673

Table 3 the air heater rejecting heat to the surroundings.

Range and values of various parameters. The outlet and mean temperatures of the air have been calculated
Parameter Range from
Transmittance-absorptance product, τα 0.8 (fixed) for single glass cover To = Ti + Q/(Mcp ) (42)
Gap between the absorber plate and glass 40 mm [42]
cover, δpg
Tm = (Ti + To )/2 (43)
Collector slope, β 45◦ (winter operation)1
Back insulation Foamed polystyrene or glass wool The thermal efficiency ηth of the air heater is defined as the ratio of
Thermal conductivity of insulation, ki 0.037 W/(m K)
the useful heat gain Q and the incident solar radiation I on the collector
Insulation thickness, δi 60 mm
Glass cover thickness, δg 3.25 mm plane, i.e.,
Thermal conductivity of glass, kg 0.78 W/(m K)
ηth = Q/(IA) (44)
Long wave emissivity: Glass cover, εg 0.88
Absorber plate, εp 0.95 (flat black-paint)
Niles et al. [47] have used the following equation to calculate the
Solar insolation, I 500–1000 W/m2
Wind heat transfer coefficient, hw 5 W/(m2 K) (no wind) and 15 W/
outlet air temperature when the solar air heater operates in open loop
(m2 K) mode (i.e., Ti = Ta):
Mass flow rate per unit area of absorber plate, 0.005–0.01 kg/(s m2) ( )
G To = Ti + I(τα)FR / Gcp (45)
Sky temperature, Ts = 0.0552 Ta1.5 [41]
Absorber plate and fin material Steel Parameter FR in Eq. (45) is termed as heat removal factor and is given
Length of the collector in array, L 1–3 m by
Duct/passage height, ( )[ ( ′ )]
Parallel plate duct, H 10 mm Gcp F UL
FR = 1 − exp − (46)
Multiple parallel rectangular passages, Hp 12, 7 and 6 mm UL Gcp
Width of collector module in array, W 1 and 2 m
Ambient Temperature2, Ta 273, 278 and 283 K (0◦ , 5◦ and 10 where F΄ is termed as efficiency factor and UL is termed as overall loss
◦
C) coefficient of the collector. The efficiency factor F΄ is given by
Flow Reynolds number, Re 750–1500 (L = 2 m)
370–750 (L = 1 m)
[ ]− 1
UL
(47)
′
1150–2400 (L = 3 m) F = 1+
h
1
The latitude of the Western Rajasthan is 23.3 to 30.12 N, i.e., the mean is
27.21◦ . Optimum slope for winter operation is about 27.21 + 15 = 42.15◦ . For the mean absorber plate temperature Tp and ambient tempera­
2
The northern part of Western Rajasthan experiences a minimum tempera­ ture Ta, the overall loss coefficient UL is defined as
ture of 0 ◦ C in the winter [43]. [ ( )]
UL = QL / A Tp − Ta (48)

The equation of the mean air temperature in the terms of parameters

Nu = 0.157(Ra’ )0.285 for 9.23 × 104 < Ra’ ≤ 106 (37c) FR and F΄ is [48]
where Ra΄ (=Ra cos β) is the Rayleigh number for the inclined air
(49)
′
Tm = Ti + (Q/A)(1 − FR /F )/(UL FR )
layers. The Rayleigh number Ra for the internal natural convection heat
transfer between parallel plates is given by Equations (45) and (49) have been used to cross-check their values
( ) calculated from Eqs. (42) and (43), respectively.
g Tp − Tgi δpg 3
Ra = GrPr = Pr (38) Equation (6) along with Eq. (7) of friction factor has been utilized for
Tmpg νmpg 2 calculating pumping power from Eqs. (11) and (19) and pressure loss
where δpg is the gap between the absorber plate and glass cover. from the calculated pumping power using Eq. (1). The thermo-physical
As stated earlier, the wind heat transfer coefficient hw at the outer properties of the air have been taken at the corresponding mean tem­
surface of the glass cover in Eq. (36c) is a function of the wind velocity perature from Eq. (23).
Vw. In present study, wind heat transfer coefficient values considered are Equations (6) and (7), (23)-(27) and (33)-(49) have been solved by
5 W/(m2 K) for no wind condition and 15 W/(m2 K) corresponding to the following an iterative process presented in Fig. 6 for the computation of
average wind velocity of 2–3 m/s in the winter of Western Rajasthan the useful heat gain or heat collection rate Q, thermal efficiency ηth,
[43], which has been estimated using the following correlation proposed pressure loss Δp and pumping power P. For the heat collection estimate,
by McAdams [44]. the iteration was terminated when the successive values of the estimates
of heat collection rate Q given by Eq. (33) and the heat collection rate Qh
hw = 5.6214 + 3.912Vw for Vw < 4.88 m/s (39) from the heat transfer consideration given by Eq. (35) differed by<0.4
This correlation has been widely used in modeling and simulation in W, and when the estimate of heat loss from the absorber plate to the
spite of its shortcomings [45]. glass cover and glass cover to the ambient, i.e., Qtpg and Qtgo from Eqs.
(36a) and (36c), respectively, differed by<0.05 W.
Sensitivity analysis shows that 1 % change in the heat transfer co­
4.2. Back and edge losses efficient between the absorber plate and air affects the thermal effi­
ciency prediction by 0.15 %. An error of 1 % in the estimate of heat
The heat losses from the back surface of the air heater Qb and from transfer coefficient between the glass cover and absorber plate hcpg has
the edge Qe have been estimated from the following empirical equations an effect of the order of 0.1 % on the predicted thermal efficiency.
of Klein [46]. The nonlinear mathematical model used here has been validated by
( ) Karwa et al. [12] against the data from an experimental study of a
A Tp − Ta
Qb = δi 1 (40) smooth duct solar air heater operating in transitional to early turbulent
ki
+ hw flow published in an earlier work of the author [49].

Qe = 0.5Ae (Tp − Ta ) (41)

4.3. Range and values of various parameters
where δi is the thickness of the back insulation, ki is the thermal
conductivity of the insulating material and Ae is the area of the side of The range and values of various parameters for the present study are

8
R. Karwa Applied Thermal Engineering 221 (2023) 119673

Start 80

Read Ta, I, hw, G, p, g, , , pg, g, i, ki, kg, W, p, H, Hp, L, - - - - Conventional solar air heater
70
INITIALIZE Tm = Ta + tm

th (%)
60
INITIALIZE Tp = Tm + tp

Thermal efficiency,
COMPUTE AIR PROPERTIES AT Tm
50

COMPUTE Qe, Qb
40
INITIALIZE Tgi = Ta + 1.0

COMPUTE Qtpg, Qtgo 30

No (Qtpg - Qtgo)
Tgi = Tgi + 0.01 20
Accuracy

Yes 10
Qt = Qtpg 0.03 0.04 0.05 0.06 0.07 0.08
T/I, K.m2/W
COMPUTE QL, Q
Fig. 7. Thermal efficiency of conventional and finned solar air heater with
COMPUTE Re, x*, Nu, h, multiple rectangular passages in parallel [L = 2 m, W = 1 m, H = 10 mm, Hp =
12.0 mm, p = 25 mm, δ = 1 mm, εp = 0.95, β = 45◦ , steel absorber plate and
fin, Ap, E, Qh
fins, Ta = 278 K, hw = 15 W/(m2 K) and I = 800 W/m2, Re = 750–1500 (finned
SAH), Re = 1080–2200 (conventional SAH)].

No ABS(Q Qh)
Tp =Tp + 0.001
Accuracy
10
COMPUTE To, Tm
Yes 9
COMPUTE x+, f, p, P, th, T,
T/I, UL, F', FR, Tp, To, Tm 8

7
Pressure drop, p (Pa)

Print G, M, I, Ta, hw, p, L, W, H, Hp, p,

Re, Nu, E, Q, UL, To, Tp, Tm, th, f, p, P 6

5
Stop
4
Fig. 6. Flow chart for iterative solution of governing equations.
3
given in Table 3. The collector slope β is fixed at 45◦ , which is the near
optimum inclination for the winter operation in the Western Rajasthan. 2
As mentioned earlier, the collector module considered is 2 m in length H = 10 mm)
1
and 1 m in width W, which will be termed as 2 × 1 module in the present - - - - Finned solar air heater (H = 12.0 mm)
discussion. For greater solar air heater area requirements, the collector 0
modules are connected in series and parallel [50], which has been dis­ 0.004 0.005 0.006 0.007 0.008 0.009 0.01 0.011
cussed later on. Inlet air is at the ambient temperature for open loop
G, kg/(s m2)
operation. The air mass flow rate per unit area of the plate G is selected
to give an air temperature rise ΔT ≥ 30 ◦ C when the solar insolation is at Fig. 8. Pressure drop Δp versus flow rate G for conventional and finned solar
its lowest of about 500 W/m2 in morning and evening hours. air heaters.

5. Results and discussion comparison, performance plots of the conventional solar air heater with
single rectangular air flow duct below the absorber plate are given in the
Heat transfer enhancement due to the fins has been presented in figures. The direction of the increasing Reynolds number is also marked
Fig. 4 as plot of the performance parameter E = (hp Ap)/(hA) versus the in the Fig. 7. The Reynolds numbers for these plots range from 750-1500
mass flow rate of air per unit area of plate G and discussed. and 1080–2200 for the finned and conventional solar air heaters,
respectively.
5.1. Thermo-hydraulic performance It can be seen from the plots in Fig. 7 that the finned absorber solar
air heater has 28–31 % better thermal efficiency than the conventional
The results of the thermo-hydraulic performance analysis have been solar air heater, which is quite significant. This enhancement is the
presented as plots of thermal efficiency ηth versus the temperature rise result of the enhanced heat transfer rate due to the fins on the absorber
parameter ΔT/I [= (To - Ti)/I] in Fig. 7 for air mass flow rate G = plate leading to a higher heat collection rate.
0.005–0.01 kg/(s m2), and pressure drop Δp versus G in Fig. 8. For direct The performance plot of Fig. 7 can be used to determine the mass

9
R. Karwa Applied Thermal Engineering 221 (2023) 119673

flow rate at which the solar air heater must be operated to achieve the equation as
desired air temperature rise. Alternatively, the temperature rise of the ( )( )( ) ( )
air can be determined from the plot when the air heater is operating at a Δp =
2ML μC p 1
∝
ML
(50)
particular mass flow rate. This has been explained below. ρ dhp 2 nwHp dhp 2 nwHp
Let the desired temperature rise ΔT is 50 ◦ C when solar insolation I is
The mass flow rate M = GWL and W = np = n (w + δ) ≈ nw. Using
800 W/m2. The temperature rise parameter (ΔT/I) works out to be
these relations, Eq. (50) transforms to
0.0625. A vertical line on the graph from ΔT/I = 0.0625 cuts the plot of
( )
the finned air heater at G = 0.007 kg/(s m2) with thermal efficiency of GL2
about 44 %. If the solar air heater is operating at G = 0.006 kg/(s m2), a Δp∝ (51)
dhp 2 Hp
vertical line from the point of intersection of line of G = 0.006 with the
plot gives ΔT/I ≈ 0.068, which gives the air temperature of about 54.5 The Nusselt number is a function of dimensionless axial distance x* =
◦
C and thermal efficiency of about 41 %. L/(dh Re Pr) in case of a developing laminar flow.
Analysis showed that aluminium fins provide 0.34–0.51 % better Using m = M/np, M = GWL, np ≈ W/w, Eq. (17) of the Reynolds
thermal efficiency of the finned solar air heater than the air heater with number transforms to
steel fins of the same thickness because of the higher thermal conduc­ ( )( )
1 GLdhp
tivity of aluminium fins; thermal conductivity of aluminium is 200 W/ Re = (52)
(m K) as compared to 30 W/(m K) for the steel. Copper fins (kf = 400)
μ Hp
have been found to provide 0.38–0.56 % better performance, i.e., the Thus, for the given G, w and Hp, Reynolds number Re is proportional
thermal performance advantage over the aluminium fins is marginal to the collector length L. Using this deduction, x* can be seen to acquire
only. Steel is a much cheaper material than aluminium and copper. In a constant value, i.e., the Nusselt number does not change with the
India, the ratio of their price is about 1:3:9. Since the density of change in the length of the collector for the given air flow passage
aluminium is about one third of the steel, the cost of the aluminium height.
sheet will be nearly the same for their equal thickness. However, the Equation (51) shows that the pressure drop is proportional to G, L2
fabrication cost of the finned absorber plate of aluminium is somewhat and inversely proportional to dhp2 and Hp. So, it is evident that in order to
higher. It is also to note that the fins must be structurally stable and satisfy the condition of the equal pumping power, the air flow passage
hence the aluminium fins may be required to be made of thicker sheets, height must be increased for a collector of greater length and vice versa.
which will further increase the cost of the aluminium absorber plate. The For most of the applications, a greater area of the solar energy
density of the copper is about 14 % higher than that of the steel and the collection is required than that of a single module of 2 m2. To achieve
price per kg of copper is nearly-nine times while the performance this, the modules are often arranged in a collector array in series, par­
advantage is only marginal. Hence, copper for absorber plate and fins allel, or combined series and parallel mode. In Fig. 9(a) and 9(b), two
has not been considered. basic arrangements of flow through a single module 2 × 1 and 1 × 2 are
The pressure drop for the finned air heater with passage height Hp = shown giving collector length in flow direction of 2 m and 1 m,
12 mm is about 2.9–5.1 % lower than that for the conventional solar air respectively. In Fig. 9(c) is shown a simple collector array consisting of
heater duct as seen in Fig. 8 because of the use of 12 mm height instead three 1 × 2 modules in series giving effective collector length of 3 m in
of 11.85 mm determined earlier for equal pumping power. the flow direction. As discussed above, to satisfy the condition of equal
The thermal efficiency plot in Fig. 7 refers to fixed values of design pressure drop, the passage height Hp has been modified from 12 mm for
and ambient parameters. The effect of variations of these parameters is 2 m long module to 7 mm and 15.5 mm for collector length of 1 m and 3
being discussed below. m, respectively. The thermal efficiencies of the arrangements in Fig. 9
are depicted in Fig. 10.
5.2. Effect of variation of design and ambient parameters Since the heat transfer coefficient h = Nu k/dh, it increases with the
decrease in the height of the passage and vice versa. This strong effect of
The solar insolation I varies from about 500 to 1000 W/m2; mini­ the reduction in Hp from 12 mm to 6 mm for fixed collector length of 2 m
mum in the morning and evening to the maximum at the solar noon. The can be seen in Fig. 10. The pressure drop Δp can be seen in Fig. 10(a) to
wind velocity variation has been considered to be from no wind con­ increase by about 6 times when the passage height is reduced to half
dition to the mean velocity of 2–3 m/s and, in the Western Rajasthan, the because of the increased velocity of the air through the collector air flow
ambient temperature Ta lies between about 0◦ and 10 ◦ C in the winter. passage while the gain in thermal efficiency due to the increased heat
For large solar collector area requirement for greater heat duty, the transfer coefficient is 6.8–7.1 % as seen in Fig. 10(b). Thus, the decision
collector modules are arranged in series and parallel in the collector regarding the choice of the passage height is strongly related to the
array. The individual collector module 2 × 1 in the array may also be allowable pressure drop. Fig. 10(b) also presents the thermal efficiency
arranged as 1 × 2, i.e., 1 m in flow direction and 2 m transverse to the of solar air heaters of 1 m, 2 m and 3 m length depicted in Fig. 9 with
flow. In order to investigate the effect of variation of these parameters, passage heights of 7 mm, 12 mm and 15.5 mm, respectively, to fulfil the
one parameter has been varied at a time from the values of these pa­ condition of equal pressure drop. The thermal efficiency of the solar
rameters in Fig. 7. collector with L = 1 m, p = 25 mm and Hp = 7 mm is 5.6–5.9 % higher
while the thermal efficiency of the collector with L = 3 m, p = 40 mm
5.2.1. Effect of variation of design parameters Hp and L and Hp = 15.5 mm, which is basically series arrangement of 1 × 2
The thermal efficiency of the solar air heater is a strong function of
the allowable pressure drop Δp and the mass flow rate per unit area of
the absorber plate G. Hollands and Shewen [29] have discussed the
considerations for fixing the pressure drop. Leakage of air from the air 2m 1m
heater and connecting ductwork is an important consideration which 1m 3m
2m
places an upper bound on Δp. In the case of arrays, a flow imbalance
2m
across the parallel passages may occur. Hollands and Shewen have
(a) (b) (c)
mentioned the need of a sufficient pressure drop to permit the inlet and
outlet manifolds for the flow passages to maintain a uniform flow across Fig. 9. (a) A 2 × 1 module (L = 2 m, W = 1 m), Hp = 12 mm, (b) A 1 × 2
the parallel passages; this consideration places a lower bound on Δp. module (L = 1 m, W = 2 m), Hp = 7 mm, (c) 3 modules of 1 × 2 in series giving
Since pumping power Pp = (M/ρ)Δp, Eq. (19) gives the pressure drop L = 3 m, W = 2 m, Hp = 15.5 mm (parallel flow passages are not shown).

10
R. Karwa Applied Thermal Engineering 221 (2023) 119673

60 70
L = 2 m, H = 6 mm 2
hw = 5 W/(m K)
- - - - L = 2 m, H = 12 mm 2
50 - - - - hw = 15 W/(m K)
60

th (%)
Pressure drop p, Pa

40

Thermal efficiency,
50
30
0.01
0.009 G
20
40 0.008
0.007
10 0.006
0.005
30
0
0.004 0.005 0.006 0.007 0.008 0.009 0.01 0.011
G, kg/(s m2)
20
0.04 0.05 0.06 0.07 0.08 0.09
T/I, K.m2/W
80
L = 2 m, p = 25 mm, H = 6 mm Fig. 11. Effect of wind heat transfer coefficient hw on thermal efficiency of
. finned solar air heater (L = 2 m, Hp = 12 mm, p = 25 mm).
70 L = 1 m, p = 25 mm, H = 7 mm Equal
(%)

L = 2 m, p = 25 mm, H = 12 mm pressure
well as with the month. The efficiency increases by a maximum of 0.75
......... L = 3 m, p = 40 mm, H = 15.5 mm drop
th

60 % when the solar insolation decreases to 500 W/m2 while the reduction
Thermal efficiency,

in the thermal efficiency with the increase in the insolation to 1000 W/

m2 is found to be maximum of 0.94 % at the lowest flow rate of 0.005
50
kg/(s m2).
It is to note that in the case of applications requiring nearly constant
40 outlet air temperature from the collector, the operation point will shift
to the right in Fig. 7 when the insolation will decrease and the thermal
efficiency can be seen to be significantly reduced. For example, let the
30 finned solar air heater is operating at G = 0.01 kg/(s m2) when solar
insolation is 800 W/m2. The value of the parameter ΔT/I from Fig. 7 is
20 0.0495 and the thermal efficiency ηth is 49.5 %. If the solar insolation
0.04 0.05 0.06 0.07 0.08 0.09 decreases to 600 W/m2, the value of the parameter ΔT/I will be 0.066
for the fixed value of ΔT and the operating point will shift to the right at
T/I, K.m2/W
flow rate of about 0.0065 kg/(s m2) giving thermal efficiency of about
43 %, i.e., down from 49.5 % by about 15 %.

Fig. 10. (a) Effect of passage height Hp on pressure drop (L = 2 m), (b) effect of 5.2.4. Effect of the wind heat transfer coefficient
length L and passage height Hp on thermal efficiency. The decrease in the wind heat transfer coefficient from 15 to 5 W/
(m2 K) corresponding to no wind condition has significant effect on the
module of Fig. 9(b), is 5–6.3 % lower. This result clearly establishes the thermal efficiency, which increases by 5.6–8.6 % as can be seen in
performance advantage of 11–12 % of the parallel arrangement of Fig. 11. This increase in the thermal efficiency is because of the reduc­
modules (for example, 1 × 2 modules in parallel) over the series tion in convection heat loss from the solar air heater; the effect is higher
arrangement (three 1 × 2 modules in series). This result is in line with at low flow rate because of the higher operating temperature of the
the observation of Karwa et al. [50] based on the performance evalua­ absorber plate.
tion criterion of equal pumping power for the mass flow rate of 0.005–
0.04 kg/s per unit area of the absorber plate. 5.2.5. Effect of ambient temperature
The minimum ambient temperature in the winter varies between
5.2.2. Effect of fin thickness near 0 ◦ C in the North region to about 10 ◦ C in the South region of the
For the 2 × 1 module with fin pitch p = 25 mm, the increase in the Western Rajasthan. It is an established fact that a decrease in the
thickness of the fin from 1 mm to 1.5 mm has been found to improve the ambient temperature improves the thermal performance and vice versa.
thermal efficiency by 0.35–0.4 % because of the increased fin efficiency. The thermal efficiency has been found to improve by 0.5–0.7 % with the
This is accompanied with increased pressure drop of about 3.7 % decrease in the ambient temperature from 278 K to 273 K and decrease
because of reduction in width of the rectangular passage (w = p – δ) from by 0.5–0.7 % with the increase in the ambient temperature from 278 K
24 mm to 23.5 mm. to 283 K.

5.2.3. Effect of variation of solar insolation 6. Conclusions

The results presented in the previous sections are based on the
average value of solar insolation, which is encountered within 3–4 h Comparative heat transfer performance studies of uniformly heated
near the noon. However, the solar insolation varies during the day as flat plate with longitudinal fins forming multiple parallel flow

11
R. Karwa Applied Thermal Engineering 221 (2023) 119673

rectangular passages below the plate with the conventional flat plate rectangular duct of solar air heater, Renew. Energy 33 (4) (2008) 585–596, https://
doi.org/10.1016/j.renene.2007.03.023.
solar air heater rectangular duct of low aspect ratio have been carried
[5] R. Karwa, K. Chauhan, Performance evaluation of solar air heaters having v-down
out at equal pumping power under laminar flow conditions, which has discrete rib roughness on the absorber plate, Energy 35 (1) (2010) 398–409,
been followed by thermo-hydraulic performance studies of the finned https://doi.org/10.1016/j.energy.2009.10.007.
and conventional solar air heaters. [6] G. Tanda, Performance of solar air heater ducts with different types of ribs on the
absorber plate, Energy 36 (11) (2011) 6651–6660, https://doi.org/10.1016/j.
The important findings of the study are as follows: energy.2011.08.043.
[7] A.P. Singh, Varun, Siddhartha, Heat transfer and friction factor correlations for
1. The high efficiency metallic fins (fin efficiency ≥ 90 %) welded to the multiple arc shape roughness elements on the absorber plate used in solar air
heaters, Exp. Therm. Fluid Sci. 54 (2014) 117–126.
uniformly heated flat plate cover whole of the surface of the rect­ [8] A.K. Patil, J.S. Saini, K. Kumar, Experimental investigation of enhanced heat
angular air flow passages and thus increases the effective surface transfer and pressure drop in a solar air heater duct with discretized broken v-rib
area for heat transfer. The performance parameter E = (hp Ap)/(hA), roughness, J. Solar Energy Eng. 137 (2) (2015) e021013, https://doi.org/10.1115/
1.4028071.
depicting enhancement in the heat transfer rate due to use of fins, for [9] R. Kumar, Varun, A. Kumar, Experimental and computational fluid dynamics study
the uniformly heated flat plate with 1 mm thick steel fins forming on fluid flow and heat transfer in triangular passage solar air heater of different
parallel flow passages is found to be 2.7–2.75 and 2.47–2.57 for fin configurations, J. Solar Energy Eng. 139 (4) (2017) e041013, doi: 10.1115/
1.4036775.
pitch of 25 mm and 40 mm, respectively. [10] K. Nidhul, S. Kumar, A.K. Yadav, S. Anish, Enhanced thermo-hydraulic
2. At equal pressure drop and pumping power, the 2 × 1 solar air heater performance in a v-ribbed triangular duct solar air heater: CFD and exergy analysis,
module with finned absorber plate has been found to provide 28–31 Energy 200 (2020) e117448, https://doi.org/10.1016/j.energy.2020.117448.
[11] R. Misra, J. Singh, S.K. Jain, S. Faujdar, M. Agrawal, A. Mishra, P.K. Goyal,
% better thermal efficiency than the conventional solar air heater,
Prediction of behavior of triangular solar air heater duct using v-down rib with
which is quite significant. This enhancement is the result of the multiple gaps and turbulence promoters as artificial roughness: a CFD analysis, Int.
enhanced rate of heat transfer due to the fins on the absorber plate J. Heat Mass Transfer. 162 (2020) e120376, https://doi.org/10.1016/j.
leading to a higher heat collection rate of the air heater. ijheatmasstransfer.2020.120376.
[12] R. Karwa, N. Karwa, R. Misra, P.C. Agarwal, Effect of flow maldistribution on
3. Effect of design, operating and ambient parameters on the thermal thermal performance of a solar air heater array with subcollectors in parallel,
efficiency of the finned solar air heater has been studied. The length Energy 32 (7) (2007) 1260–1270, https://doi.org/10.1016/j.energy.2006.08.002.
and height of the air flow passage of the solar air heater have strong [13] J. Nikuradse, Laws of flow in rough pipes, National Advisory Committee for
Aeronautics, Washington, Technical Memorandum 1292, November 1950.
effect on pressure drop, pumping power and thermal efficiency. The [14] B.K. Maheshwari, R. Karwa, S.K. Gharai, 2011. Performance study of solar air
study shows that, at equal pumping power, the passage height must heater having absorber plate with half-perforated baffles. ISRN Renew. Energy.
be increased with the increase in the length. (2011) e634025, doi: 10.5402/2011/634025.
[15] S. Chamoli, N.S. Thakur, Effect of roughness height ratio in V down perforated
4. In a collector array, the parallel arrangement of the collector mod­ baffle roughness on thermohydraulic performance of solar air heater: an
ules has been found to provide significantly better thermal perfor­ experimental study, Int. J. Ambient Energy 36 (5) (2015) 242–247, https://doi.
mance as compared to the modules in series at equal pumping power. org/10.1080/01430750.2013.853206.
[16] M.A.E. Habet, S.A. Ahmed, M.A. Saleh, The effect of using staggered and partially
5. The results of the present study can be utilized for the development tilted perforated baffles on heat transfer and flow characteristics in a rectangular
of enhanced thermal performance solar air heater with finned channel, Int. J. Thermal Sci. 174 (2022) e107422, https://doi.org/10.1016/j.
absorber plate operating under laminar flow condition for delivery of ijthermalsci.2021.107422.
[17] P. Naphon, On the performance and entropy generation on double pass solar air
high temperature air. Since the thermal performance enhancement
heater with fins, Renew. Energy 30 (9) (2005) 1345–1357, https://doi.org/
has been shown to be obtained without increased pumping power 10.1016/j.renene.2004.10.014.
penalty, the existing smooth rectangular duct solar air heater can [18] A. Fudholi, K. Sopian, M.H. Ruslan, M.Y. Othman, Performance and cost benefits
also be modified to enhance their thermal performance by using analysis of double-pass solar collector with and without fins, Energy Convers.
Manag. 76 (2013) 8–19, https://doi.org/10.1016/j.enconman.2013.07.015.
finned absorber plate. [19] J.M. Jalil, R.F. Nothim, M.M. Hameed, Effect of wavy fins on thermal performance
of double pass solar air heater, Eng. Tech. J. 39 (9) (2021) 1362–1368, doi:
10.30684/etj.v39i9.1775.
Declaration of Competing Interest
[20] C.D. Ho, H. Chang, R.C. Wang, C.S. Lin, Performance improvement of a double pass
solar air heater with fins and baffles under recycling operation, Appl. Energy 100
The authors declare that they have no known competing financial (2012) 155–163, https://doi.org/10.1016/j.apenergy.2012.03.065.
[21] M. Yang, X. Yang, X. Li, Z. Wanga, P. Wang, Design and optimization of a solar air
interests or personal relationships that could have appeared to influence
heater with offset strip fin absorber plate, Appl. Energy 113 (2014) 1349–1362,
the work reported in this paper. https://doi.org/10.1016/j.apenergy.2013.08.091.
[22] A. Benseddik, A. Azzi, R. Khanniche, A.K. Allaf, Simulation study of solar air
Data availability collector with offset strip fin absorber plate for drying agricultural products in a
semi-arid climate, Int. J. Heat Tech. 36 (2) (2018) 557–568, https://doi.org/
10.18280/ijht.360219.
No data was used for the research described in the article. [23] H.-M. Yeh, C.-D. Ho, Effect of external recycle on the performances of flat-plate
solar air heaters with internal fins attached, Renew. Energy 34 (5) (2009)
1340–1347, https://doi.org/10.1016/j.renene.2008.09.005.
Acknowledgement [24] A. Saravanan, M. Murugan, M.S. Reddy, P.S. Ranjit, P.V. Elumalai, P. Kumar,
S. Rama Sree, Thermo-hydraulic performance of a solar air heater with staggered
The author is grateful to Jai Narain Vyas University, Jodhpur for C-shape finned absorber plate, Int. J. Thermal Sci. 168 (2012) e107068, https://
doi.org/10.1016/j.ijthermalsci.2021.107068.
their support. [25] J.M. Jalil, G.K. Salih, Y.A. Madhi, Performance study of solar air heater with
thermally conducted multi v shaped baffles and ribs, Eng. Tech. J. 36, Part A, No. 8
References (2018) 930–938, https://doi.org/10.30684/etj.36.8A.14.
[26] S.P. Shetty, A. Paineni, M. Kande, N. Madhwesh, N.Y. Sharma, K.V. Karanth,
Experimental investigations on a cross flow solar air heater having perforated
[1] R.P. Saini, J.S. Saini, Heat transfer and friction factor correlations for artificially
circular absorber plate for thermal performance augmentation, Sol. Energy 197
roughened ducts with expanded metal mesh as roughness element, Int. J. Heat
(2020) 254–265, https://doi.org/10.1016/j.solener.2020.01.005.
Mass Transfer. 40 (4) (1997) 973–986, https://doi.org/10.1016/0017-9310(96)
[27] S. Şevik, M. Abuşk, Enhancing the thermal performance of a solar air heater by
00019-1.
using single-pass semi-flexible foil ducts, Appl. Therm. Eng. 179 (2020) e115746.
[2] R. Karwa, Experimental studies of augmented heat transfer and friction in
[28] M. Zukowski, Experimental investigations of thermal and flow characteristics of a
asymmetrically heated rectangular ducts with ribs on the heated wall in transverse,
novel microjet air solar heater, Appl. Energy 142 (2015) 10–20, https://doi.org/
inclined, v-continuous and v-discrete pattern, Int. Commun. Heat Mass Transfer. 30
10.1016/j.apenergy.2014.12.052.
(2) (2003) 241–250, https://doi.org/10.1016/S0735-1933(03)00035-6.
[29] K.G.T. Hollands, E.C. Shewen, Optimization of flow passage geometry for air
[3] A.R. Jaurker, J.S. Saini, B.K. Gandhi, Heat transfer and friction characteristics of
heating plate type solar collectors, J. Solar Energy Eng. 103 (4) (1981) 323–330,
rectangular solar air heater duct using rib-grooved artificial roughness, Sol. Energy
https://doi.org/10.1115/1.3266260.
80 (8) (2006) 895–907, https://doi.org/10.1016/j.solener.2005.08.006.
[4] K.R. Aharwal, B.K. Gandhi, J.S. Saini, Experimental investigation on heat transfer
enhancement due to a gap in an inclined continuous rib arrangement in a

12
R. Karwa Applied Thermal Engineering 221 (2023) 119673

[30] R.K. Shah, Correlation for laminar hydrodynamic entry length solutions for circular [40] R. Karwa, G. Chitoshiya, Performance study of solar air heater having v-down
and noncircular ducts, J. Fluids Eng. 100 (1978) 177–179, https://doi.org/ discrete ribs on absorber plate, Energy 55 (2013) 939–955, https://doi.org/
10.1115/1.3448626. 10.1016/j.energy.2013.03.068.
[31] R.K. Shah, D.P. Sekulic, Fundamentals of Heat Exchanger Design, Wiley, New [41] W.C. Swinbank, Long-wave radiation from clear skies, Quart. J. Roy Meteorol. Soc.
Jersey, 2003. 89 (381) (1963) 339–348, https://doi.org/10.1002/qj.49708938105.
[32] M.A. Ebadian, Z.F. Dong, Forced convection, internal flow in ducts, in: W.M. [42] H. Buchberg, I. Catton, D.K. Edwards, Natural convection in enclosed spaces-a
Rohsenow, J.P. Hartnett, Y.I. Cho (Eds.), Handbook of Heat Transfer, third ed., review of application to solar energy collection, J. Heat Transfer. (1976) 182–188,
McGraw-Hill, New York, 1998, pp. 5.1–5.137. https://doi.org/10.1115/1.3450516.
[33] R.M. Curr, D. Sharma, D.G. Tatchell, Numerical predictions of some three- [43] Climate of Rajasthan, India Meteorological Department, Pune, India. Available
dimensional boundary layers in ducts, Comput. Methods Appl. Mech. Eng. 1 (1972) from: <https://imdpune.gov.in/library/public/Climate%20of%20Rajashtan.pdf,
143–158, doi: 10.1016/0045-7825(72)90001-1. 2010> [accessed March 17, 2022].
[34] R. Karwa, Heat and Mass Transfer, second ed., Springer Nature Singapore, [44] W.H. McAdams, Heat Transmission, McGraw-Hill, New York, 1954.
Singapore, 2020, doi: 10.1007/978-981-15-3988-6. [45] J.A. Palyvos, A survey of wind convection coefficient correlations for building
[35] R. Karwa, Enhanced heat transfer performance of multiple triangular air flow envelope energy systems’ modeling, Appl. Therm. Eng. 28 (8,9) (2008) 801–808,
passages in parallel with inclined fins for flat plate solar air heater, J. Solar Energy https://doi.org/10.1016/j.applthermaleng.2007.12.005.
Eng. 144 (5) (2022) e051003, https://doi.org/10.1115/1.4053976. [46] S.A. Klein, Calculation of flat-plate collector loss coefficients, Sol. Energy 17 (1975)
[36] W.M. Kays, H.C. Perkins, Forced convection, internal flow in ducts, in: W.M. 79–80, https://doi.org/10.1016/0038-092X(75)90020-1.
Rohsenow, J.P. Hartnett (Eds.), Handbook of Heat Transfer, McGraw-Hill, New [47] P.W. Niles, E.J. Carnegie, J.G. Pohl, J.M. Cherne, Design and performance of an air
York, 1973, pp. 7.1–7.193. collector for industrial crop dehydration, Sol. Energy 20 (1978) 19–23, https://doi.
[37] P. Wibulswas, Laminar flow heat transfer in non-circular ducts [Ph.D. thesis], org/10.1016/0038-092X(78)90136-6.
London University, London, 1966. [48] J.A. Duffie, W.A. Beckman, Solar Engineering of Thermal Processes, Wiley, New
[38] R.K. Shah, A.L. London, Rectangular ducts, in: R.K. Shah, A.L. London (Eds.), Jersey, 2013, doi: 10.1002/9781118671603.
Laminar Flow Forced Convection in Ducts, Academic Press, 1978, pp. 196–222, [49] R. Karwa, S.C. Solanki, J.S. Saini, Thermo-hydraulic performance of solar air
doi: 10.1016/C2013-0-06152-X. heaters having integral chamfered rib roughness on absorber plates, Energy 26 (2)
[39] R. Karwa, C. Sharma, N. Karwa, 2013. Performance evaluation criterion at equal (2001) 161–176, https://doi.org/10.1016/S0360-5442(00)00062-1.
pumping power for enhanced performance heat transfer surfaces. J. Solar Energy. [50] R. Karwa, S.N. Garg, A.K. Arya, Thermo-hydraulic performance of a solar air heater
2013, e370823, doi: 10.1155/2013/370823. with n-subcollectors in series and parallel configuration, Energy 27 (2002)
807–812, https://doi.org/10.1016/S0360-5442(02)00016-6.

13