JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY 4, 043116 (2012)

Experimental investigation of various designs of solar flat

plate collectors: Application for the drying of green chili
Adnane Labed,1 Noureddine Moummi,1 and Adel Benchabane2,a)
1
Laboratoire de G
enie M
ecanique, Universit
e de Biskra, B.P. 145 R.P. 07000 Biskra,
Algeria
2
Laboratoire de G
enie Energ
etique et Mat
eriaux, LGEM, Universit e de Biskra,
B.P. 145 R.P. 07000 Biskra, Algeria
(Received 15 February 2012; accepted 20 July 2012; published online 8 August 2012)

This paper presents an experimental study by comparing between the thermal

performance of three types of solar air flat plate collectors, FPCs: (i) without
obstacles, (ii) with rectangular obstacles, and (iii) with trapezoidal obstacles in the
air flow duct. In order to determine the best performing model, we have proceeded
to reversing the flow direction, in each collector, and comparing the six obtained
models under outdoor conditions. All collectors were designed, constructed, and
tested in the University of Biskra (Algeria) in a stand facing south at an inclination
angle equal to the local latitude. Thus, we have proceeded to the application of
the best system for the drying of the green chili. We have sought to determine the
moisture content and loss of mass for the forced convection hot air drying of the
product and their temperature dependence. In comparison with the recent literature,
at different air mass flow, the highest efficiencies (77%) were obtained from the
FPC with trapezoidal obstacles, when the air was blown down at air flow rate
0.043 kg/s. In addition, this study has allowed us to show that (i) for a same
geometry, the highest efficiencies were always obtained when the air was blown
down in the solar air FPC and (ii) the use of obstacles, in the air flow duct of the
FPCs, is an efficient method to improve their performances, especially when the air
is blown down. The obstacles ensure a good air flow under the absorber plate,
create the turbulence, and reduce the dead zones in the collector. V C 2012 American

Institute of Physics. [http://dx.doi.org/10.1063/1.4742337]

I. INTRODUCTION
There are basically two types of solar collectors: non-concentrating or stationary and con-
centrating. A non-concentrating collector has the same area for intercepting and for absorbing
solar radiation, whereas a sun-tracking concentrating solar collector usually has concave reflect-
ing surfaces to intercept and focus the sun’s beam radiation.1 Among stationary solar heaters,
air flat plate collectors (FPCs) have been widely used for energy conservation and management
in an increasing number of installations.2 They are quite attractive for low-temperature solar
energy technology which requires air temperatures below 100  C. In fact, solar air FPCs are
extensively used over years, because they are relatively simple with a minimal use of materials,
easy to operate, and have low capital costs.1,3 Solar air FPCs are used for space heating, dehy-
dration of industrial products, and drying of agricultural crops such as vegetables, fruits, grains,
spices, medicinal plants, lumber, tobacco, fish, etc.1–6
Compared to the conventional energy sources, the investigation of the solar air FPC per-
formances allows to judge if solar energy is preferable as an energy source for one of the appli-
cations mentioned above. In addition, this energy possesses a thermal conversion mode which

a)
Author to whom all correspondence should be addressed. Electronic mail: adel.benchabane@gmail.com. Tel./Fax:
þ213 (0) 33 73 39 89.

1941-7012/2012/4(4)/043116/15/$30.00 4, 043116-1 C 2012 American Institute of Physics

V
043116-2 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

necessities a simple technology adapted to the site and to particular region for each application.
Indeed, many researchers have been conducted regarding flow and heat transfer characteristics
in the solar air FPCs.4,7–11
Choudhury et al.7 have conducted a comparative study between one-pass corrugated, air
heaters with different widths of the air channel and for different specific mass flow rates of air.
Hegazy4 analyzed a criterion for determining the optimum flow-channel depth of conventional
solar air FPC. Thus, some approaches have been developed to treat the absorber itself.8
Furthermore, it is established that the introduction of different geometries of artificial
roughness, in the dynamic air vein of the FPC, increases the transfer rate and favors the
creation of turbulences near the absorber plate. These techniques can improve the thermal
performance either by increasing the heat-transfer area with the use of FPC having aluminum
cans, corrugated/finned absorber surfaces, or by enhancing the absorber to air convective heat
transfer coefficient with the use of roughened absorber surfaces. Indeed, the use of artificial
roughness in different forms and shapes is known us the most effective and economic way of
improving the performance of a solar air heater.8,9,12–18
Among the studies on the performance improvements of solar air FPCs for the drying of
agricultural products, Abene et al.19 analyzed the performances of solar air FPCs with various
obstacles geometries and proceeded to the application of the best tow system for the drying of
grape. The efficiencies of these six collectors range from 42% to 82% when the air flow rate
equal to 50 m3/hm2 and the simple one (without obstacles) has the minimum efficiency (42%).
Abene et al.19 concluded that the most efficient is the collector supplied with waisted delta
lengthways WDL1 obstacles. The study of Ahmed-Zaid et al.20 deals the drying of yellow on-
ion. Authors showed that the use of the obstacles in the collector duct improves likewise the
performance of the drying unit. They concluded that the reduction of the transverse space and
the longitudinal space induces good results.
Recently, Varun et al.10 carried out a review of roughness geometry in solar air heaters. These
authors discussed different roughness geometries used in solar air heaters and explained the concept
of artificial roughness, effects of various roughness parameters on the flow pattern, and also briefly
discussed the roughness geometries used in heat exchangers other than solar air heaters.
In our university, several theoretical and experimental investigations were carried out on the
enhancement of thermal performances of solar air FPCs designed for drying applications.21–26 In
all these studies, the authors were used different form of obstacles mounted under the absorber
plate on the air channel duct.
In a previous study,26 we investigated the performance of single and double-pass solar air
heaters through the use of various fin geometries of four types of solar FPCs: model (N) with-
out obstacles, model (A) with rectangular obstacles and model (B-1) with a trapezoidal
obstacles in the air flow duct. Thus, we have proceeded to the comparison of the best system
with model (B-2), a double pass flat plate collector having the same type of obstacles in order
to determine the best performing model. Through the experiments undertaken, it was found
that, at the air mass flow of 0.028 kg s1, the highest efficiencies were obtained from the double
pass solar air FPC with trapezoidal obstacles. The lowest efficiencies values were obtained
from the collector without obstacles.
In the present study, we conduct a comparative study between three types of solar air FPCs: (i)
without and with tow geometries of artificial roughness and (ii) we reverse the flow direction in each
collector. These six models were compared in order to determine the best performing model for the
drying uses. As application, we present also a comparison between the drying results of green chili
using the solar air FPCs without obstacles and with the best model. In a forthcoming publication, we
will be numerically investigated the different models of solar air FPCs.

II. EXPERIMENTAL
A. Experimental setup
Figure 1 shows the experimental device of one of the three studied solar air FPCs when
the air was blown-down in the collector. We show that the absorber plate was placed behind
043116-3 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

FIG. 1. Experimental device.

the transparent glass cover with a layer of static air separating it from the cover of the
collector.
As reported in our previous investigation,26 all the solar air FPCs were designed and con-
structed in the University of Biskra with the same design and dimensions. Therefore, all the
collector components have the same size: thickness of the single cover glass (5 mm), height of
the air gap between the cover and the absorber plat (25 mm), height of the air duct (25 mm),
dimensions of the absorber (1.96 m  0.9 m with the thickness of 0.4 mm), and thickness of the
rear insulation (40 mm). Only the configuration of fins is different in the studied FPCs.
The materials used in the fabrication of all FPCs components are the same. The absorbers
were made of galvanized steel with nonselective black coating. The heated air flows between
the inner surface of the absorber plate and the back plate with, or without, obstacles. The rear
insulation is provided by a polystyrene sheet (30 mm of thickness), which is sandwiched
between two plywood sheets.
Since this investigation is conducted to evaluate the thermal performances of three types of
solar air FPCs: (i) without obstacles, (ii) with rectangular obstacles, and (iii) with the trapezoi-
dal obstacles; we show in Figure 2 a schematic view of fin obstacles in dynamic air vein, using
the two studied types of obstacles: rectangular and the trapezoidal forms, indexed, respectively,
as A and B forms. The fin obstacles are oriented parallel to the fluid flow and are soldered to
the upper side of the back plate.
In order to determine the best performing model, we have proceeded to reversing the flow
direction, in each solar air FPC, and comparing the six obtained models, namely N-I, N-II, A-I,
A-II, B-I, and B-II (Figure 3). The indices (I) and (II) indicate, respectively, that the air was

FIG. 2. Schematic view of fin obstacles in dynamic air vein, using two types of obstacles: (a) rectangular and (b) trapezoi-
dal forms.
043116-4 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

FIG. 3. Schematic presentation of the six models: (N) without obstacles, (A) with rectangular obstacles and (B) with trape-
zoidal obstacles. The indices (I) and (II) indicate, respectively, that the heating air was blown up or blown down in the
FPC.

blown up or blown down in the FPC. (N), (A), and (B) indicate, respectively, that the solar air
FPCs is: without obstacles, with rectangular obstacles or with trapezoidal obstacles.

B. Experimental measurements
All the solar air FPCs were tested in the University of Biskra under similar environmental
conditions in spring with relatively clear sky. Biskra is located in the East of Algeria with lati-
tude of 34 480 N, longitude of 5 440 E, and altitude of 85 m. The collectors were placed in a
stand facing south at an inclination angle equal to the local latitude.
In all studied collectors, the air outlet and inlet cross-sections are equipped by divergent
channel duct, and the test facility permit to vary the mass flow rate of the air. In summary, the
above experimental setup was instrumented for the measurement of the solar radiation, wind
velocity, pressure drop, temperature of the atmosphere air, inlet and outlet air temperatures, sur-
face temperature of absorber plate, relative humidity, the sample weight, and the air mass flow
rate.
To carry out these experiments, 12 insulated thermocouples have been used for the mean
temperature measurement. Five thermocouples were provided in the air channel along the direc-
tion of flow. Inlet and outlet air temperatures were measured by two thermocouples and the am-
bient temperature was measured by a mercury thermometer. Thus, four thermocouples were
positioned on the top surface of the absorber plates at identical positions along the axial center
line to obtain the average temperature of the absorber. In addition, a thermocouple was posi-
tioned in the drying chamber on the same level of the product to dry.
The test data were measured at an average interval of 10 min, so the temperatures, solar
radiation intensity, pressure drop, the air flow rate and wind velocity, relative humidity, and
weight of drying product are, respectively, measured with the k-type thermocouples with accu-
racy 0.01  C, Kipp and Zonen pyranometer CM 11 with 1% accuracy, pressure transducer accu-
racy (Kimo CP301) with 61 Pa and 0.5% of reading, a Kimo type anemometer with hot wire
(VT300) with 63% of reading and 610 m3 for the flow rate measurement and 63% of reading
and 60.1 m/s accuracy, Q.C.58 hygrometer with error ranges of 3% for RH (relative humidity
of ambient air), and laboratory balance with 1% accuracy and maximum weight of 2 kg were
used.

C. Experimental and uncertainty analysis

Before the presentation of these six configurations in Sec. II B, we present the expressions
used for the calculation of global heat loss, useful energy, and efficiency of the solar collectors.
The useful energy gain had been established as follows:3
043116-5 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

Qu ¼ m_ cP ðTf o  Tf i Þ: (1)

And the following heat balance expresses the thermal performance of a collector under steady-
state condition
 
Qu ¼ AC  FR IG ðsv aabs Þ  UL ðTf i  Ta Þ : (2)

A measure of collector performance is the collector efficiency, defined as the ratio of useful
heat gain over any time period to the incident solar radiation over the same period we can,
thus, define efficiency as

QU
g¼ : (3)
IG A C

Solving Eqs. (1)–(3), and g can be obtained as

m_ cP ðTf o  Tf i Þ
g¼ : (4)
IG :AC

The collector efficiency can also be related to the inlet fluid temperature using the following
equation:

ðTf i  Ta Þ
g ¼ FR ðsv aabs Þ  FR UL : (5)
IG

Here, FR ðsv aabs Þ and FR UL are two major parameters that constitute the simplest practical col-
lector model. FR ðsv aabs Þ is a parameter that indicate how energy is absorbed and FR UL indicate
how energy is lost. The thermal energy is lost from the collector to the surroundings by con-
duction, convection, and infrared radiation. UL is equal to the sum of energy loss through the
top (Ut), bottom (Ub), and edge (Ue) of the collectors as follows:22

UL ¼ Ut þ Ub þ Ue : (6)

The drying characteristics of chilies such as moisture content and drying rate were determined
by using Eqs. (7) and (8), respectively.
The moisture content, Mdb, on dry basis was calculated by using

ðMt  Md Þ
Mdb ¼  100%: (7)
Md

The drying rate, DR, should be proportional to the difference in moisture content between ma-
terial to be dried and the equilibrium moisture content.27 The concept of thin layer drying was
assumed for the experiments as reported by

dM
DR ¼ ¼ kðMt  Me Þ: (8)
dt

Based on the analysis of the errors in the experimental measurements through the used instru-
ments, the uncertainties in experimental measurement and results are often used to refer to pos-
sible values that may include errors. The result R of an experiment is assumed to be calculated
from a set of measurements, and it is given as a function of the independent variables X1,
X2, …, Xn (Ref. 28)
R ¼ RðX1 ; X2 ; X3 ; :::; XN Þ; (9)

where in, X1, X2, X3 are measured variables.

043116-6 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

Let dR be the uncertainty in the result and dX1 , dX2 , …, dXn be the uncertainties in the in-
dependent variables. If the uncertainties in the independent variables are all given with same
odds, then uncertainty in the result having these odds is calculated by Kalogirou.1
" 2  2  2 #12
@R @R @R
dR ¼ dX1 þ dX2 þ :::: þ dXn : (10)
@X1 @X2 @Xn

The independent parameters measured in the experiments reported here are collector inlet tem-
perature Tfi, collector outlet temperature Tfo, ambient temperature Ta, mass flow rate, and solar
irradiation.
If Ac and cp are considered constants in Eq. (4), it can be written as

g ¼ f ðTf o ; Tf i ; IG ; mÞ:
_ (11)

The total uncertainty equation for collector efficiency, g, (TfiTa)/IG, moisture content and dry-
ing rate can be written as
" 2  2  2  2 #12
@g @g @g @g
dg ¼ dm_ þ dTf o þ dTf i þ dIG ; (12)
@ m_ @Tf o @Tf i @IG

" 2  2  2 #12
@F @F @F
dF ¼ dTf o þ dTf i þ dIG ; (13)
@Tf o @Tf i @IG

where F ¼ ðTf i  Ta Þ=IG .

" 2  2 #12
@Mdb @Mdb
dMdb ¼ dMt þ dMd ; (14)
@Mt @Md

" 2  2  2 #12
@DR @DR @DR
dDR ¼ dMt þ dMd þ dt : (15)
@Mt @Md @t

Calculations show that the total uncertainty in calculating efficiency, g, F ¼ ðTf i  Ta Þ=IG ,
moisture content Mdb, and drying rate DR are almost in the order of 1%, 1%, 2.8%, and 3%,
respectively.

III. RESULTS AND DISCUSSIONS

A. Measurements at a fixed airflow
To show the influence of these offset plate fins and flow direction on the enhancement of
thermal performance, these experiments were conducted in outdoor conditions in the spring of
2009. All FPCs were tested in similar environmental conditions as indicated above and
explained in our previous work.26 ASHRAE standard requires that, for the collector efficiency
test, the solar insulation must be above 630 W/m2.14 Instantaneous efficiencies are determined
from Eq. (4) for the averaged pairs and are plotted as a function of (Tfi  Tfa)/IG. The thermal
efficiency curves for the six collectors are plotted in Fig. 4 and the characteristic parameters
obtained from these curves for each collector are presented in Table I. The scatter of the data
around the straight line is mainly attributed to the pump temperature, angle of incidence varia-
tions, wind speed, and the dependence of UL on the plate temperature. Also, the variations of
the relative proportions of beam diffuse and ground reflective components of solar radiation are
participating in the data scattering. Thus, scatters in the data are to be expected.29 However, it
043116-7 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

FIG. 4. Efficiencies versus (Tfi–Ta)/IG for different collector models.

is acceptable that collectors are characterized by the line intercept FR ðsv aabs Þ and the slope
FR UL . From Figure 4, it is evident that the efficiency decreases as the temperature parameter
increases. That means, at higher reduced temperature parameters, the overall loss is lower. The
maximum efficiencies for the six models N-I, N-II, A-I, A-II, B-I, and B-II are determined as
55%, 60.7%, 66.3%, 73.5%, 71%, and 73.7%, respectively, at m_ ¼ 0:028 kg=s.
It is also found that collector B-II has the highest efficiency, followed by collectors A-II,
B-I, A-I, N-II, and at least N-I. For a same geometry, the highest efficiencies were obtained
when the air was blown down in the solar air FPC. In this case, the flow direction influences
the pressure loss and consequently the electrical consumption of the air pump. For the collec-
tors studied here, this pressure loss remains always minim compared to the profit brought to the
efficiency of the FPC. As an example: the pressure loss of model A-II is higher than that of
model A-I with the maximum difference surrounding 3 Pa, which allows writing

DPAII ¼ 1:07 DPAI : (16)

In Figs. 5(a) and 5(b), the average values of ambient temperatures, solar radiation, temperature
differences (TfoTfi) between outlet and inlet air, and efficiencies for different collector models
measured in different days are presented. Fig. 5(a) presents the means ambient temperatures
and solar radiations of different test days depending on the experiment time (9:00–16:00).
The recorded values of temperature differences (TfoTfi) for each collector model are given in
Fig. 5(b).
Since the airflow rate was kept constant during the experimental time, these parameters
directly could give us a sign about the efficiencies of the collectors which can also be seen in
this figure. Moreover, these differences signal the amount of useful heat taken from the solar
radiation. There is a maximum difference between the inlet and outlet air temperature of model

TABLE I. Characteristic parameters of collectors based on the ASHRAE 93-2003 standard.

Collector N-I N-II A-I A-II B-I B-II

FR ðsv aabs Þ 0.55 0.607 0.663 0.735 0.71 0.80

FR UL (Wm2 k1) 18 18.5 17.5 16.7 19 20.2
043116-8 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

FIG. 5. Average values of (a) the ambient temperature, Ta, and the solar radiation, IG, and (b) the temperature difference
(Tfo–Tfi) and the efficiency, g, during the testing days (d1, d2, d3, d4, d5, and d6), related, respectively, to N-I, N-II, A-I,
A-I, B-I, and B-II.

B-II as seen from Fig. 5(b). This difference is also minimum for model N-I. The most efficient
is model B-II and the least one is model N-I. The average efficiency of model B-II and N-I are
66.8% and 44.87%, respectively, for a flow rate of 0.028 kg/s. Furthermore, it is seen from
the investigation results that the effect of the inversing flow direction on the performance
is considerably more (models N-II, A-II, and B-II). The average efficiencies for the six models
N-I, N-II, A-I, A-II, B-I, and B-II are determined as 44.87%, 52.2%, 55.36%, 62.2%, 58.7%,
and 66.8%, respectively, for the same flow rate m_ ¼ 0:028 kg=s.

B. Measurements at different airflow rates

The basic method of measuring collector performance is to expose operating collector to
solar radiation and measure the fluid inlet and outlet temperature and the fluid flow rate. In
addition, radiation on the collector, ambient temperature, and wind speed are also recorded.

FIG. 6. Variation of the FPC efficiencies, g, as a function of the air flow rate, m,
_ for different studied models (N-I, N-II,
A-I,A-II, B-II, and B-II).
043116-9 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

The variation of the efficiency (g) is studied as a function of variation in the mass flow
rate (Fig. 6), and the improvements of thermal performances are important in relation to the
FPC. It can be seen that the collector efficiency increases with increasing air mass flow rate of
air considerably. Furthermore, the efficiencies of the studied models are in decreasing order,
respectively, as follows: B-II, A-II, B-I, A-I, N-II, and N-I model.
To make a comparison of the thermal performances of the models N-II, A-II, and B-II (the
heating air was blown down), three configurations of air heaters reported in the literature were
selected (flat plate, finned, and v-corrugated collectors).14 Fig. 7 shows the thermal efficiency
of models (N-II, A-II, and B-II) and the FPC efficiencies reported by Karim and Hawlader.14
The corresponding air mass flow rate is from 0.01 to 0.055 kg/s. The comparison proved that
the most efficient model was model B-II, followed by v-groove collector, than model A-II,
finned collector was the fourth one, than we have FPC and at least model N-II.
In order to provide a simple comparison between the thermal performances of models
(N-II, A-II, and B-II) tested in this study and the reported ones, three configurations of solar
FPCs were selected from the literature (Fig. 8). Beside the thermal efficiency of N-II, A-II, and
B-II models, Figure 8 depicts the thermal efficiencies reported by the following authors: Ozgen
et al.,13 Karsli,29 and Parker et al.30 We would note here that Parker et al.30 performed the ther-
mal efficiency of a FPC of single air flow and double air flow collector for air mass flow rate
of 0.0238 kg/s. Karsli29 compared four types of air heating flat plate solar collectors: a finned
collector with an angle of 75 , a finned collector with an angle of 70 , a collector with tubes,
and a base collector. All collectors had 0.45 wide, 0.9 m long, and 25 mm flow duct height.
Ozgen et al.13 designed and tested three different absorber plates. In the first type (type I), cans
had been staggered as zigzag on absorber plate, while in type II, they were arranged in order.
Type III is a flat plate (without cans) and experiments had been performed for air mass flow
rates of 0.03 kg/s and 0.05 kg/s. The collector dimensions were 2.14 m and 0.84 m (length/
wide).
The comparison showed that the most efficient model is the model of Karsli29 followed by
the model B-II in the second position, and thereafter model A-II and model of Parker et al.30
The worst performing models are the model of Ozgen et al.13 and the model N-II. We also
note that the slope of the efficiency curve decreases, which means reduction in loss coefficient

FIG. 7. Comparison between the thermal efficiency of models (N-II, A-II, and B-II) and the FPC efficiencies reported by
Karim and Hawlader.14
043116-10 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

FIG. 8. Comparison between the thermal efficiency of models (N-II, A-II, and B-II) and the reported ones.

with an increase of the reduced temperature, and good agreement was found between the ther-
mal efficiency of the FPCs tested in this study with those reported in the literature.

C. Improvement of drying time

Chilies are a potential cash crop in Algeria and dried chili is a popular spices food in this
country. Huge amount of chilies are consumed in every part of Algeria. For this reason, the
forced convection solar drier appears to have potential for adoption and application in the
tropics and subtropics.
Improvement in drying time allows a reduction in the operating costs. We have selected a
system with forced convection, composed of an enclosure with one rack, a solar air flat plate
collector, and a fan. The drying chamber is a cylindrical enclosure made of hard plastic of
8 mm thickness, 50 cm in diameter, and 80 cm height, positioned meadows of the collectors and
supported by a metal frame.
The primary aim of the experiment work is to investigate the chili drying of ability under
forced-ventilation condition with model B-II (case II), while the drying with model N-II (case
I) also tested merely for comparison purpose.
Field tests of the dryer for drying of chili also were carried out in the spring 2010, and the
typical results are shown in Figs. 9–11. In all experiments, we have considered the initial prod-
uct mass to be 200 g, with the initial moisture content (dry basis) of 7.17 Kgwatter/kg. The prod-
uct is sliced into discs of average thickness of 5 mm, and the final moisture content must be
equal to 0.05 kgwater/kg (db). We present a comparison between results from our experiments
with the collector without obstacles (model N-II) and with our best configuration (model B-II),
with the air flow rate of 0.0250 kg/s.
043116-11 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

FIG. 9. Evolution of green chili loss of mass (DM) and temperature of the product as a function of the drying time, in the
case of models B-II and N-II during the testing days (13/04, 16/05/2010), respectively, (m_ ¼ 0.0250 kg/s).

Figure 9 presents the product temperature (TPr) and the loss of mass (DM) evolution as a
function of the drying time in the case of N-II and B-II models. After 7 h of drying with model
A-2, the resulting loss of mass in product was 79.61%, when it takes 4 h 20 min to get a similar
value of loss of mass. The drying time is reduced by 2 h 40 min, corresponding to a relative
reduction of 38%.
Figure 10 shows a typical experimental evolution of green chili moisture content and dry-
ing rate as a function of the drying time, in the cases of models B-II and N-II with
m_ ¼ 0.0250 kg/s. Drying is started at 09:00 with an initial moisture content of 7.17 kgwater/kg
(db) and continued until 16:00 (7 h of drying time). The final moisture content of the samples

FIG. 10. Evolution of green chili moisture content and drying rate (dry basis) as a function of the drying time, in the cases
of models B-II and N-II during the testing days (13/04, 16/05/2010), respectively, (m_ ¼ 0.0250 kg/s).
043116-12 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

FIG. 11. Evolution of green chili moisture content (dry basis) as a function of the drying time, with 0.0250, 0.0179, and
0.0107 kg/s, respectively, in the case of model B-II.

obtained by model B-II is of 0.05 kg water/kg (db), while it is of 0.66 kgwater/kg (db) obtained
by model A-II after the same drying time (7 h).
There is a relationship between drying rate and moisture content of the product during
drying.
As a result, the drying rate of green chili with B-II dryer was found to be higher than that
of model A-II in the first hours of drying (9:00–12:00), and it was lower in the last hours of
drying time (9:20–16:00). This is due to the fact that the sample in B-II dryer loses the majority
of its wet mass in the first hours of drying time and the moisture content of the sample in A-II
dryer was found to be higher than that of B-II dryer, what influences directly the rate transfer
of moisture to the ambient air.
To study the influence of the air flow rate on the drying time, we have used model B-2
with three air flow rate 0.0107, 0.0179, and 0.0250 kg/s. Figure 11 illustrate how the moisture
content variation is affected during the drying time. As expected, there is an inverse relation-
ship between the moisture content and drying time. Drying was started with initial moisture
content of 7.17 kgwater/kg (db) and continued, until a final moisture content of the samples
ranged from 0.05 to 0.3 kgwater/kg (db) attained after 7 h.
Time taken to reach particular moisture content from the initial moisture content decreases
with the increase in the mass flow rate of the drying air. To lead to a value of moisture content
of 0.306 Kgwater/Kg(db), the drying time with the first air flow rate (0.0107 kg/s) is longer (7 h
00 min) but is equal to 6 h 00 min and 4 h 40 min with the second, and the third flow rate
(0.0179 and 0.0250 kg/s), which correspond to a relatives reductions in drying time of 14.28%
and 33.33%, respectively (Fig. 11).
Figure 12 shows a comparison between the drying results of green chili of the present
study and the experimental study of Mohanraj and Chandrasekar31 and Hossain and Bala.32
It should be noted that there is a difference between the solar collector sections ins these
references.31,32 The collector used in Mohanraj and Chandrasekar’s study31 includes storage
material under the absorber plate; whereas the collector used by Hossain and Bala32 com-
posed of a transparent plastic covered flat-plate collector and a drying tunnel connected in
series.
We observe in Figure 12 that all drying rates follow a similar pattern. As well, the drying
rates obtained in this study (model N-II and B-II) are higher than those given by the referen-
ces.31,32 These observations may be due to the difference between the initial weights of green
chili, which is 200 g for the present study, 40 and 80 kg, respectively, for Mohanraj and
043116-13 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

FIG. 12. Comparison of drying of green chilies of present study with other investigators.

Chandrasekar31 and Hossain and Bala.32 In addition, these authors have dried green chili with-
out cut it, while it is cut into slices in the present study.

IV. CONCLUSION
An experimental study was conducted to evaluate the thermal performances of three types
of solar air FPC for drying applications: (N) without obstacles, (A) with rectangular obstacles,
and (B) with trapezoidal obstacles. In order to determine the best performing model, under
environmental conditions, we have proceeded to reversing the flow direction, in each collector,
and comparing the six obtained models, namely N-I, N-II, A-I, A-II, B-I, and B-II. It is worth
remembering here that the indices (I) and (II) indicate, respectively, that the air was blown up
or blown down in the FPC under a similar environmental conditions, in a stand facing south at
an inclination angle equal to the Biskra latitude. In each air heater, the absorber plate was
placed behind the transparent cover with a layer of static air separating it from the cover. The
heated air flows between the inner surface of the absorber plate and the back plate with, or
without, obstacles. Through the experiments undertaken, we measured: the solar radiation, wind
velocity, temperature of the atmosphere air, inlet and outlet air temperatures, surface tempera-
ture of absorber plate, and the air mass flow rate.
It was found that, at the air mass flow of 0.028 kg s1, the highest efficiencies were
obtained from the solar air FPC with trapezoidal obstacles, when the air was blown down. The
lowest efficiencies values were obtained from the collector without obstacles when the air was
blown up. However, for a best evaluation of the efficiency improvement obtained with the
trapezoidal obstacles, we compared our results with other results from the literature obtained
under similar conditions. Indeed, in comparison with the selected experimental investigation of
Karim and Hawlader,14 at different air mass flow (from 0.005 to 0.045 kg s1), the highest effi-
ciencies were obtained also from the FPC with trapezoidal obstacles, when the air was blown
down. In addition, this study has allowed us to show that the use of obstacles, in the air flow
duct of the FPCs, is an efficient method to improve their performances and to reduce the prod-
uct drying time, especially when the air was blown down. In this case, the obstacles ensure a
good air flow under the absorber plate, create the turbulence, and reduce the dead zones in the
collector.
In addition, an increase in the air flow rate leads to results which are again clearly better.
However, the target drying air temperature must take account of constraints (quality, flavor,
color and nutritional value) imposed by the finished product.
043116-14 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

ACKNOWLEDGMENTS
The authors would like to acknowledge the technical assistance of O. Kaddour and M.-T.
Baissi.

NOMENCLATURE
AC Collector surface area (m2)
cp Specific heat of air at constant pressure (kJ/kg K)
db Dry basis
DR Drying rate (kg kg1 h1)
FR Collector heat removal factor depending air inlet temperature (dimensionless)
HR Relative humidity of ambient air (%)
IG Global irradiance incident on solar air heater collector (Wm2)
Lc Length of the flat plate collector (m)
lc Width of the flat plate collector (m)
Mdb Moisture content (dry basis) (kg kg1)
m_ Air mass flow rate (kg s1)
QU Useful energy gain of the collector (Wm2)
Ta Ambient temperature ( C)
Tf i Inlet air temperature of the collector ( C)
Tf o Outlet fluid temperature of the collector ( C)
UL Collector overall heat loss coefficient (W/m2  C)
DM Loss of mass of the drying product (%)
DP Pressure loss (Pa)
aabs Absorptance (dimensionless)
b Collector tilt ( ) (dimensionless)
sv Transparent cover transmittance (dimensionless)
g Thermal efficiency (dimensionless)

1
S. A. Kalogirou, Prog. Energy Combust. Sci. 30, 231 (2004).
2
A. Fudholi, K. Sopian, M. H. Ruslan, M. A. Alghoul, and M. Y. Sulaiman, Renewable Sustainable Energy Rev. 14, 1
(2010).
3
J. A. Duffie and D. Beckman, Solar Engineering of Thermal Processes (John Wiley & Sons, Inc., Hoboken, New Jersey,
USA, 2006).
4
A. A. Hegazy, Energy Convers. Manage. 41, 861 (2000).
5
A. A. Hegazy, Energy Convers. Manage. 41, 1361 (2000).
6
S. Kalogirou, Appl. Energy 76, 337 (2003).
7
C. Choudhury, S. Andersen, and J. Rekstad, Sol. Energy 40, 335 (1988).
8
S. Ouard, Ph.D. dissertation, Universite de Valenciennes, France, 1989.
9
T. Koyuncu, Renewable Energy 31, 1073 (2006).
10
Varun, R. P. Saini, and S. K. Singal, Sol. Energy 81, 1340 (2007).
11
T. Liu, W. Lin, W. Gao, C. Luo, M. Li, Q. Zheng, and C. Xia, Int. J. Green Energy 4, 601 (2007).
12
N. Moummi, Ph.D. dissertation, Universite de Valenciennes, France, 1994.
13
F. Ozgen, M. Esen, and H. Esen, Renewable Energy 34, 2391 (2009).
14
M. Karim and M. Hawlader, Energy 31, 452 (2006).
15
S. Youcef-Ali and J. Desmons, Renewable Energy 31, 2063 (2006).
16
M. K. Mittal, Varun, R. P. Saini, and S. K. Singal, Energy 32, 739 (2007).
17
S. Singh, S. Chander, and J. Saini, J. Renewable Sustainable Energy 3, 013108 (2011).
18
S. Singh, S. Chander, and J. Saini, J. Renewable Sustainable Energy 3, 023107 (2011).
19
A. Abene, V. Dubois, M. Le Ray, and A. Ouagued, J. Food Eng. 65, 15 (2004).
20
A. Ahmed-Zaid, H. Messaoudi, A. Abenne, M. Le Ray, J. Desmons, and B. Abed, Int. J. Energy Res. 23, 1083
(1999).
21
K. Aoues, N. Moummi, A. Moummi, M. Zellouf, A. Labed, and E. Achouri, Rev. Energ. Renouvelables 11, 219–227
(2008).
22
A. Labed, N. Moummi, K. Aoues, M. Zellouf, and A. Moummi, Rev. Energ. Renouvelables 12, 551 (2009).
23
N. Moummi, A. Moummi, K. Aoues, C. Mahboub, and S. Youcef Ali, Int. J. Sustainable Energy 29, 142–150
(2010).
24
K. Aoues, N. Moummi, M. Zellouf, and A. Benchabane, Int. J. Ambient Energ. 32, 95 (2011).
25
C. Mahboub, N. Moummi, A. Moummi, and S. Youcef-Ali, Sol. Energy 85, 776 (2011).
26
A. Labed, N. Moummi, A. Benchabane, K. Aoues, and A. Moummi, “Performance investigation of single- and double-
pass solar air heaters through the use of various fin geometries,” Int. J. Sustainable Energy (in press).
043116-15 Labed, Moummi, and Benchabane J. Renewable Sustainable Energy 4, 043116 (2012)

27
O. Ekechukwu, Energy Convers. Manage. 40, 593 (1999).
28
J. P. Holman, Experimental Methods for Engineers, 6th ed. (McGraw-Hill, New York, 1994).
29
S. Karsli, Renewable Energy 32, 1645 (2007).
30
B. F. Parker, M. R. Lindley, D. G. Colliver, and W. E. Murphy, Sol. Energy 51, 467 (1993).
31
M. Mohanraj and P. Chandrasekar, J. Eng. Sci. Technol. 4, 305 (2009).
32
M. Hossain and B. Bala, Sol. Energy 81, 85 (2007).