Hindawi Publishing Corporation

Advances in Mechanical Engineering

Volume 2011, Article ID 103502, 10 pages
doi:10.1155/2011/103502

Research Article
Effects of Delta-Shaped Obstacles on the Thermal
Performance of Solar Air Heater

Adisu Bekele, Manish Mishra, and Sushanta Dutta

Department of Mechanical and Industrial Engineering, Indian Institute of Technology Roorkee, Roorkee 247667, India

Correspondence should be addressed to Adisu Bekele, addisu2005@yahoo.com

Received 5 March 2011; Revised 3 August 2011; Accepted 18 August 2011

Academic Editor: Hyung Hee Cho

Copyright © 2011 Adisu Bekele et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

An experimental investigation has been carried out to study the effect of delta-shaped obstacles mounted on the absorber surface
of an air heater duct with an aspect ratio 6 : 1 resembling the conditions close to solar air heaters. This study encompassed the
Reynolds number (Re) ranging from 3400 to 27600, longitudinal pitch of the obstacle (Pl /e) varied from 3/2 to 11/2, and relative
obstacle height (e/H) varied from 0.25 to 0.75. The relative obstacle transverse pitch (Pt /b) = 7/3 and the angle of attack of
flow on obstacle = 90◦ are kept constant during the whole experimentation. By comparing the heat transfer data obtained from
the obstacle-mounted duct with that of smooth duct under similar geometrical and flow conditions, the obstacle-mounted duct
enhances the heat transfer by 3.6-times at Re = 7276.82, Pl /e = 3/2, and e/H = 0.75.

1. Introduction rate with an undesirable rise of friction: (1) interruption

The thermal efficiency of solar air heaters (SAHs) has been
found to be generally poor because of the inherent poor
thermal conductivity of air. In order to make the solar air
heaters economically viable, their thermal efficiency needs to
be improved by enhancing the heat transfer coefficient. There
are different factors affecting the SAH efficiency, for example,
collector length, collector depth, type of absorber plate, glass
cover plate, wind speed, and so forth.
The shape factor of the absorber plate is the most impor-
tant parameter in the design for any type of SAH. Increasing
the absorber plate shape area will increase the heat transfer to
the flowing air. On the other hand, it will increase the pres-
sure drop in the collector; this increases the required power
consumption to pump the air flow crossing the collector [1].
The main applications of SAHs are space heating and
drying. The SAHs occupy an important place among solar
heating systems because of minimal use of materials.
The conventional solar air heaters are generally improved
by means of various augmentation techniques with emphasis
on many types of surface enhancements. Augmented surfaces
can create one or more combinations of the following con-
ditions that are favorable for the increase in heat transfer
of boundary layer development and increasing turbulence
intensity, (2) increase in heat transfer area, and (3) generating
of vortex and/or secondary flows [2].
In order to increase the thermal performance of solar air
heater various designs, with different shapes and dimensions
of the air flow passage in flat plate-type solar air collectors
has been developed in literature [3–9].
Another method of improving the thermal performance
of solar air heater is by introduction of obstacles in the dy-
namic air vein of the collector which creates a turbulent flow
and hence enhances the thermal energy exchange by convec-
tion and the maximal extraction of the absorber energy can
be obtained of course at the cost of higher increased pressure
drop. Ozgen et al. [10] studied the thermal performance
of double-flow SAHs in which aluminum cans of different
arrangements are used as obstacle. This method substantially
improves the collector efficiency by increasing the fluid ve-
locity and enhancing the heat-transfer coefficient between
the absorber plate and air.
Esen [11] compared the experimental results between
SAH supplied with obstacles and those without obstacles for
the case of the double-flow SAHs, and thus the influences
of various parameters, such as the obstacles types, the mass
2 Advances in Mechanical Engineering
V Source
Bevel shape Variac
entrance MM X
P P
530 1 2
Air in
DAQs X 310
800 1200
1: Inlet section 2: Test section 3: Outlet section 4: Transition section
6: Flexible pipe 7: Blower 8: Motor
Figure 1: Schematic diagram of the experimental setup.
flow rate of air, and the level of absorber plates in duct,
on the energetic and exergetic efficiencies of the SAHs were
examined.
Abene et al. [12] carried out experimental study, which
consists of a solar energy simulation, with an aim to improve
the efficiency-temperature rise couple of the flat plate solar
collector by considering several types of obstacles: ogival
transverse (OT), ogival inclined folded (OIF1), waisted tube
(WT), waisted delta lengthways (WDL1), and waisted ogival
lengthways (WOL1) disposed in rows in the dynamic air
vein of the flat plate collector and have proceeded to the
application of the best two systems (WDL1) and transverse-
longitudinal obstacles (TL) for drying an agricultural prod-
uct.
By comparing with the collector without obstacles (WO),
the thermal transfers, and, consequently, the output tem-
perature (TO ) and the collector efficiency (η) are clearly
improved. The heat quantities and drying times obtained in
the case of WDL1 are very interesting compared with the
collector WO.
The choice of the geometrical shapes of the obstacles to
be used has to satisfy some criterion as the form and the dis-
position of the obstacles affecting the air flow during its tra-
jectory. The obstacles ensure a good air flow over the absorb-
er plate, create the turbulence, and reduce the dead zones in
the collector.
As it has been observed from the literature, a few re-
searches were carried out to study the effect of obstacles
(of various geometries) on heat transfer and friction factor
characteristics of a solar air heater. Even though it is the
goal of every researcher to increase the efficiency of solar
air heater by enhancing the heat transfer coefficient between
the absorber plate and the air, the geometrical characteristics
and arrangement of the obstacles greatly affect the result. So
detail parametric study is always necessary in order to find
the optimum obstacles parameter.
2. Experimental Program
2.1. Experimental Setup. An experimental setup has been
designed to conduct experimental investigation on the heat
Power
A Stepdown voltage 7 Motor
controller
transformer Power
6
Source
3 4 5 8
Motor
76.2
500
Air out
5: Thermal mass flow meter
MM: micro- DAS: data acquisition P: pressure
manometer system tapings
transfer and fluid flow characteristics of a rectangular duct
with obstacles attached on the heated surface. Because of the
variation of the solar energy intensity at different times and
positions repeatability of the work is not possible at outdoor
condition. So an indoor setup is selected to control the heat
flux supplied to the system at desired level all the time so that
all variables can be studied at the same heat flux supplied and
repeatability is also possible.
A schematic diagram of the indoor experimental setup
and its sectional view are shown in the Figures 1 and 2,
respectively. Air is sucked through the rectangular duct by
means of a blower driven by a 3-phase, 440 V, 3.73 KW,
2880 rpm AC motor and delivered back to the atmosphere
making an open loop system.
The rectangular duct consists of an entrance section
(800 mm × 300 mm × 50 mm), a test section (1200 mm ×
300 mm × 50 mm), and an exit section (500 mm × 300 mm
× 50 mm). It may be noted that ASHRAE standard 93–77
[13]
√ recommends√a minimum entry and exit length to be
5 WH and 2.5 WH, respectively, where W is the width
and H is the height of the air channel. (i.e., 612.37 mm for
the entrance length and 306.19 mm for the exit length). Thus
the selected lengths satisfy the ASHRAE’s standard. In the
exit sections four equally spaced baffles are provided for the
purpose of mixing the delivered air (the mixing zone with
baffles is provided before exit of air for proper mixing of
heated air to get the uniform bulk mean temperature of air
at the exit). The test section consists of rectangular duct of
which the top surface is made of 0.5 mm thick GI sheet on
which obstacles are attached on the flow passage side and the
other three sides have smooth surface.
The absorber plate is heated by an electric heater made
of 0.05 mm thick stainless steel foil (SS foil) which is used as
heating element and a uniform heat flux of 800 W/m2 is sup-
plied by using a variac and step-down transformer. All sides
of rectangular duct are insulated with a 50 mm thick thermo-
cole. The piping section up to the thermal mass flow meter is
insulated by 192 mm thick foamed polygene and foam sheet.
A GI sheet of 1200 mm × 400 mm × 0.5 mm is used as an
absorber plate (heated plate; Figure 3) on which delta shaped
obstacles (Figure 4) of the same material and thickness are
Advances in Mechanical Engineering
50
Thermocole
Wooden plank
Sunmica
Absorber plate
Asbestos sheet
Figure 2: Cross-sectional details of test section (Section X-X).
Pl
Pt
Figure 3: Photographic view of plate with obstacles mounted.
attached on the lower side and the upper side is exposed
to the heating element (uniform heat flux). The effective
area (heat transfer area) of the absorber plate is 1200 mm ×
300 mm = 0.36 × 106 mm2 . The upper side of the absorber
plate is painted dull black to increase the absorptivity of the
plate.
The air from the atmosphere enters the inlet section of
the rectangular duct through a bell mouth entrance. The uni-
form fully developed air passing through the inlet section of
the rectangular duct enters the test section where it is heated
by a power supplied through the heater assembly and then
passes through the exit section of the rectangular duct to
3
530
257
50
300
SS foil
Wooden ply
Glass wool
Cover (wooden ply)
Channel
Sense of air flow is
perpendicular to the
e plane
b
Figure 4: Photographic view of delta-shape obstacle.
the circular pipe and finally discharged to the atmosphere
through the centrifugal blower. A thermal mass flow meter
is provided at the circular pipe line for the measurement of
air mass flow rate.
The temperature of the air stream and the absorber plate
at different position of the test section are measured by K-
type thermocouples using National Instruments data acqui-
sition system of chassis model NI cDAQ-9172. Analog-type
micromanometer is used to measure the pressure drop across
the test section. The rate of flow of air in the duct is controlled
by programmable blower motor speed controller.
The test was conducted at a constant heat input and by
varying air flow rate such that the Reynolds number is main-
tained in the range of 3,400–27,600 and required readings
were recorded. All readings are noted under steady-state
4 Advances in Mechanical Engineering
Table 1: Details of obstacle geometry and arrangement.
Angle of Relative Relative obstacle
Plate
incidence obstacle height longitudinal pitch
number
[◦ ] (e/H) (Pl /e)
Plate 1 11/2
Plate 2 0.75 7/2
Plate 3 3/2
Plate 4 11/2
Plate 5 90 0.50 7/2
Plate 6 3/2
Plate 7 11/2
Plate 8 0.25 7/2
Plate 9 3/2
condition. The steady state for each run was found to reach
within 1.5 to 2 hrs.
The variation of geometry and arrangement of obstacles
are presented in Table 1. The ranges of parameters for this
study have been decided on the basis of practical considera-
tions of the system and operating conditions of the solar air
heater. The range of relative obstacle height is selected on the
basis of percentage of air flow blockage in the channel with
minimum 25% blockage and maximum 75% blockage. The
range of obstacle transverse and longitudinal pitch is selected
on the basis of consideration of flow separation and reattach-
ment length. And similarly the angle of twisting the obstacle
or angle of incidence is selected on the basis of creating
turbulence. The range of Reynolds number is selected on the
basis of the air flow rates reported in the literature [14, 15]
for solar air heater and on the capacity of the blower used.
2.1.1. Instrumentation
(i) Temperature Measurement and Thermocouple Locations.
K-type (chromel (Ni-Cr alloy)/alumel (Ni-Al alloy)) ther-
mocouples of 0.315 mm wires diameter are used to measure
the air and the heated plate temperatures at different loca-
tions. Before installing the thermocouples in place, they were
calibrated under similar environmental conditions using a
temperature calibrator for an accuracy of ±0.1◦ C. Twelve
thermocouples are provided along the axial centre line of the
absorber plate while twenty-four thermocouples were used
to measure the spanwise variation of absorber plate temper-
ature as shown in the Figure 5(a). Overall a total of thirty-
six thermocouples are used to measure the temperature
variation in the heated plate. After passing over the heated
plate in the test section, the air enters to the 500 mm long
exit section, at the end of which a baffled mixing chamber
exists. It is expected that air will leave at nearly uniform
temperature after passing through the mixing chamber.
After the mixing section, five thermocouples are arranged
spanwise to measure the outlet air temperature as shown in
Figure 5(b) and one thermocouple at inlet section to measure
the inlet air temperature as shown in Figure 5(c).
(ii) Air Flow Measurement. The actual mass flow rate
through the pipe is measured by using insertion-type ther-
Relative obstacle
Duct height, H Duct aspect ratio Base of obstacle
transversal pitch
(mm) (AR) b (mm)
(Pt /b)
7/3 50 6:1 30
mal mass flow meter Model MMFC-100 2, operating tem-
perature 0–120◦ C, and maximum flow range 9 kg/min,
which first calibrated using a standard micromanometer with
the help of pitot tube.
2.2. Experimental Procedure. Before starting the experiment,
all the thermocouples were checked carefully so that they
indicate the room temperature, the thermal mass flow meter
for the measurement of mass flow rate of air through the
duct, and the micromanometer for pressure drop measuring
indicates zero. The test runs were conducted under steady-
state conditions to collect relevant heat transfer and flow
friction data. The steady-state condition was assumed to have
been reached when the temperature at any location does not
change for about 10 min. The first steady-state for each set
of run was observed to arrive after about 1.5–2 hrs. When
a change in the operating conditions is made, it takes 15–
25 min to reach steady state. Seven values of flow rates (and
hence Reynolds number) are used for each set of experiment
at a fixed heat flux of the test. After each change of flow rate,
the system was allowed to attain a steady state before the data
were recorded.
The following parameters were measured during the
experiments:
(i) temperature of air at inlet and outlet,
(ii) temperature of the absorber plate,
(iii) mass flow rate of air,
(iv) pressure drop across the test section,
(v) voltage and current supplied to the heating section.
2.3. Data Reduction. Steady-state values of the plate and air
temperatures in the duct, at various locations for a given
heat flux and mass flow rate of air, are used to determine the
values of performance parameters.
(a) Mean Air and Plate Temperatures. The mean air temper-
ature T f = (Tai + Tao )/2 is a simple arithmetic mean of the
measured values of air temperature at the inlet and exit of the
test section.
Advances in Mechanical Engineering 5

50
1+ 4+ 7+ 10+ 13+ 16+ 19+ 22+ 25+ 28+ 31+ 34+

100
300
+ + + + + + + + + + + +
2 5 8 11 14 17 20 23 26 29 32 35

100
+ + + + + + + + + + + +
3 6 9 12 15 18 21 24 27 30 33 36
50 100 100 100 100 100 100 100 100 100 100 100 50

1200
(a)

50

50
+ + + + + +
37 38 39 40 41 42

30 60 60 60 60 30

300 300
(b) (c)

Figure 5: Location of thermocouples (a) on the absorber plate, (b) on the outlet section, and (c) on the inlet section.

The mean plate temperature can be found from weighted where ṁ is directly measured by using the thermal mass flow
average
 of the temperature of all plate thermocouples. T p = meter;
( ni=1 T pi )/n, where n is number of thermocouples on the
Qu
absorber plate. h=  , (5)
Ap Tp − T f
(b) Velocity of Air through Duct. The velocity of air flowing where A p is the effective heat transfer area of the absorber
through the duct is calculated from the knowledge of mass plate.
flow rate and area of flow as
ṁ (f) Nusselt Number and Stanton Number. The convective
V= . (1) heat transfer coefficient is then used to obtain Nusselt
ρWH
number, Nu, as
(c) Equivalent Hydraulic Diameter. The hydraulic diameter hDh
Nu = ,
of the rectangular section of the duct is defined as k
(6)
Nu
4(W · H) St = ,
Dh = . (2) Re ·Pr
2(W + H)
where k is thermal conductivity of air and Pr is Prandtl
(d) Reynolds Number. The Reynolds number of air flow in number.
the duct is calculated from
(g) Friction Factor. The friction factor is determined from
V Dh ρV Dh the measured values of pressure drop, (ΔP) across the test
Re = = , (3)
ν μ section length;

where ν is kinematic viscosity and μ is dynamic viscosity. 2(ΔP) Dh 2(ΔP)ρDh

f = Or f = , (7)
4ρLV 2 4LG2
(e) Heat Transfer Coefficient. Steady-state values of the plate where G = ṁ/(WH) is the mass velocity of air.
and air temperatures in the duct at various locations were From the analysis of uncertainties in the measurements
used to determine the values of useful parameters, namely, by various instruments [16] the maximum uncertainties in
heat supplied to the air “Qu ” and heat transfer coefficient the calculated values of various parameters are given below,
“h” calculated as
Reynolds number, Re = 0.00938 = 0.938%,
Qu = ṁC p (Tao − Tai ), (4) useful heat gain, Qu = 0.0444 = 4.44%,
6 Advances in Mechanical Engineering

80 0.016
Nusselt number (Nu)

70 0.014

Friction factor (f )
60 0.012
50 0.01
40 0.008
30 0.006
20
0.004
10
0.002
0
0
0 5000 10000 15000 20000 25000 30000
0 5000 10000 15000 20000 25000
Reynolds number (Re)
Reynolds number (Re)
Nu-experimental
Nu-Dittus-Boelter equation f -experimental
f -modified Blasius equation
Figure 6: Variation of Nusselt number with Reynolds number for
smooth duct. Figure 7: Variation of friction factor with Reynolds number for
smooth duct.

heat transfer coefficient, h = 0.046 = 4.6%, 100

90
Nusselt number, Nu = 0.046 = 4.6%, 80

Temperature (◦ C)
Stanton number, St = 0.047 = 4.7%, 70
60
friction factor, f = 1.84%. 50
40
2.3.1. Validity Test for Smooth Duct. Prior to the actual data 30
20 Plate 2: Pl /e = 7/2, e/H = 0.75, Re = 6903
collection, the setup is checked by conducting experiments
10
for smooth duct. The Nusselt number and friction factor 0
determined from these experimental data are compared 0 0.2 0.4 0.6 0.8 1 1.2
with the values obtained from the standard correlations like Test section length (m)
Dittus-Boelter equation [17] for the Nusselt number and
Plate temperatures
modified Blasius equation [18] for friction factor as shown Air temperatures
in the Figures 6 and 7, respectively.
Dittus-Boelter equation: Figure 8: Plate and air temperature distributions along the length
of the test duct.
Nus = 0.024Re0.8 Pr0.4 for 104 ≤ Re ≤ 1.24 ∗ 105 . (8)

Modified Blasius equation:

fs = 0.085Re−0.25 for 4000 < Re < 105 .
The average absolute percentage deviation of the present
experimental Nusselt number is 2.77% from the value pre-
dicted by (8), and the average absolute percentage deviation
of the present experimental friction factor is 8.77% from the
value predicted by (9) as shown in the Figures 6 and 7. Thus
there is a good agreement between the predicted values and
the experimental values of the Nusselt number and friction
factor. This ensures the accuracy of the experimental data
obtained from the present setup.
3. Results and Discussion
Figure 8 shows the variation of temperature of plate and air
along the length of the test duct. The longitudinal distribu-
tion of air temperature has been represented as straight line
connecting the measured mean values at the inlet and exit
of the test section. This pattern of temperature variation is
similar to that of [14, 15].
The effects of flow and obstacle parameters on the heat
transfer and friction characteristics for flow of air in a rec-
tangular duct are observed. Results have also been compared
with those of the smooth duct under similar flow and ther-
(9) mal boundary conditions to determine the enhancement in
the heat transfer coefficient and friction factor.
When the fluid strikes with the obstacles some of the fluid
will flow back causing the flow to separate near the obstacles
and as the flow passes the obstacles some of the fluid will
circulate back causing secondary flow to occur, and the fluid
will reattach again after some flow length next to the obstacle.
This causes vortex to be created next to the obstacles that will
increase the turbulence intensity near the obstacle and allow
more fluid to come and stay in contact with the heated plate
(absorber plate) such that heat transfer to the fluid will be
increased.
Figures 9(a) and 9(b) show the variation of Nusselt
number for obstacled plate and smooth plate ducts as a
function of Reynolds number for different values of relative
obstacle height (e/H) and relative obstacle longitudinal pitch
(Pl /e), respectively. Further Figures 10(a) and 10(b) show the
variation of friction factor for obstacled plate and smooth
plate ducts as a function of Reynolds number for different
values of relative obstacle height (e/H) and relative obstacle
longitudinal pitch (Pl /e), respectively.
The values of Nusselt number were found to increase
with increasing Reynolds number in all cases whereas
Advances in Mechanical Engineering 7

160 300
140 250 e/H = 0.75
Pl /e = 11/2
120
200
100

Nu
Nu

80 150
60 100
40
50
20
0 0
3000 8000 13000 18000 23000 28000 3000 8000 13000 18000 23000 28000
Re Re
(i) (i)
180 250
160
e/H = 0.5
140 Pl /e = 7/2 200
120
150
100

Nu
Nu

80 100
60
40 50
20
0 0
3000 8000 13000 18000 23000 28000 3000 8000 13000 18000 23000 28000
Re Re
(ii) (ii)
300 180
160
250 e/H = 0.25
Pl /e = 3/2 140
200 120
100
Nu
Nu

150
80
100 60
40
50
20
0 0
3000 8000 13000 18000 23000 28000 3000 8000 13000 18000 23000 28000
Re Re
Smooth plate e/H = 0.5 Pl /e = 7/2
Smooth plate
e/H = 0.75 e/H = 0.25 Pl /e = 11/2 Pl /e = 3/2
(iii) (iii)

(a) (b)

Figure 9: (a) Variation of Nusselt numbers with Reynolds number for different values of e/H and for (i) Pl /e = 11/2, (ii) Pl /e = 7/2, and
(iii) Pl /e = 3/2. (b) Variation of Nusselt number with Reynolds number for different values of Pl /e and for (i) e/H = 0.75, (ii) e/H = 0.50,
and (iii) e/H = 0.25.

the friction factor decreases with increase of Reynolds
number as expected. It can also be seen from Figures (9(a):
(i), (ii), (iii)) and Figures (10(a): (i), (ii), (iii)) that the values
of both the Nusselt number as well as the friction factor
are getting higher as relative obstacle height (e/H) increase
from 0.25 to 0.75 for all relative obstacles longitudinal pitch
(Pl /e) and Reynolds number. “This is due to the fact that an
increase in obstacles height increases the air flow blockage
and results in higher turbulence as well as higher the heat
transfer coefficient.”
Likewise Figures (9(b): (i), (ii), (iii)) and Figures (10(b):
(i), (ii), (iii)) show that the Nusselt numbers as well as
the friction factors are getting higher as relative obstacles
longitudinal pitch (Pl /e) decreases from 11/2 to 3/2 for all
relative obstacle height (e/H) and Reynolds number because
as the relative obstacles longitudinal pitch (Pl /e) decrease
there are more numbers of rows of obstacles for the same
absorber plate length such that the number of obstacles
increases creating more turbulence that will increases the
heat transfer coefficient and simultaneously the friction fac-
tor also increases because of the increase in air flow blockage.
In all the cases, the rate of increase in both heat transfer
and friction factor is very high for Pl /e = 3/2 compared to
that for Pl /e = 7/2 and for Pl /e = 11/2.
8 Advances in Mechanical Engineering

0.045 0.08
0.04 0.07
0.035 Pl /e = 11/2 e/H = 0.75
0.06
0.03 0.05
0.025
f f 0.04
0.02
0.015 0.03
0.01 0.02
0.005 0.01
0 0
3000 8000 13000 18000 23000 28000 3000 8000 13000 18000 23000 28000
Re Re
(i) (i)
0.05 0.04
0.045 0.035
0.04 Pl /e = 7/2 e/H = 0.5
0.03
0.035
0.03 0.025
f 0.025 f 0.02
0.02 0.015
0.015
0.01
0.01
0.005 0.005
0 0
3000 8000 13000 18000 23000 28000 3000 8000 13000 18000 23000 28000
Re Re
(ii) (ii)
0.08 0.03
0.07 0.025
Pl /e = 3/2 e/H = 0.25
0.06
0.05 0.02
f 0.04 f 0.015
0.03 0.01
0.02
0.01 0.005
0 0
3000 8000 13000 18000 23000 28000 3000 8000 13000 18000 23000 28000
Re Re
Smooth plate e/H = 0.5 Smooth plate Pl /e = 7/2
e/H = 0.75 e/H = 0.25 Pl /e = 11/2 Pl /e = 3/2
(iii) (iii)
(a) (b)

Figure 10: (a) Variation of friction factor with Reynolds number for different values of e/H and for (i) Pl /e = 11/2, (ii) Pl /e = 7/2, and (iii)
Pl /e = 3/2. (b) Variation of friction factor with Reynolds number for different values of Pl /e and for (i) e/H = 0.75, (ii) e/H = 0.50, and (iii)
e/H = 0.25.

As shown in Figures 11 and 13 for e/H = 0.50 and 0.75,
the enhancement in heat transfer and thermohydraulic per-
formance parameter initially increases with Rynolds number
and then decreases with further increase in Reynolds number
whereas for e/H = 0.25 the behavior is just the reverse. This
happens because with the large increase in Reynolds number
(i.e., an increase in air velocity) greater back flow in the
upstream of the obstacles will occur that pushes the air away
from the heated surface near the obstacles causing a long
reattachment length in the downstream of the obstacles that
can reduce the effective heat transfer area of the heated plate.
“It has been observed that as the relative obstacle height, e/H
decreases to 0.25 the enhancement in heat transfer increases
with Reynolds number; however, magnitude wise it is lower
than that obtained at relative obstacle height of 0.50 and
0.75.”
The maximum enhancement in heat transfer and pres-
sure drop penalty is obtained for Pl /e = 3/2 and e/H = 0.75
for whole range of Reynolds number whereas the thermohy-
draulic performance curve (Figure 13) shows the maximum
overall thermohydraulic performance is obtained for Pl /e =
3/2 and e/H = 0.50.
Further, from Figures 11 and 12 it has been observed that
as the relative obstacle longitudinal pitch, (Pl /e) decreases
from 11/2 to 3/2 the heat transfer enhancement as well as
friction penalty increase for all relative obstacles height. This
is because of the fact that as the relative longitudinal pitch of
the obstacle decreases there are more numbers of obstacles
Advances in Mechanical Engineering 9

4 2.5
3.5
2
3

(Sto /Sts )/( fo / fs )1.3

Nuo/Nus

2.5
1.5
2
1.5 1
1
0.5
0.5
0
0
3000 8000 13000 18000 23000 28000
3000 8000 13000 18000 23000 28000
Re
Re
Pl /e = 11/2, e/H = 0.75 Pl /e = 7/2, e/H = 0.25 Pl /e = 11/2, e/H = 0.75 Pl /e = 7/2, e/H = 0.25
Pl /e = 11/2, e/H = 0.5 Pl /e = 3/2, e/H = 0.75 Pl /e = 11/2, e/H = 0.5 Pl /e = 3/2, e/H = 0.75
Pl /e = 11/2, e/H = 0.25 Pl /e = 3/2, e/H = 0.5 Pl /e = 11/2, e/H = 0.25 Pl /e = 3/2, e/H = 0.5
Pl /e = 7/2, e/H = 0.75 Pl /e = 3/2, e/H = 0.25 Pl /e = 7/2, e/H = 0.75 Pl /e = 3/2, e/H = 0.25
Pl /e = 7/2, e/H = 0.5 Pl /e = 7/2, e/H = 0.5

Figure 11: Enhancement ratio as a function of Reynolds number. Figure 13: Variation of thermohydraulic performance parameter
with Reynolds number.

8
7
insulated walls, has been performed. The conditions and geo-
6
metrical configuration correspond to the flow in the actual
5
solar air heater duct.
fo / fs

4
Investigations have been carried out over a wide range
3
of Reynolds number (Re = 3400–27, 600) and results have
2 been compared with those of a smooth duct under similar
1 flow conditions to determine enhancement in heat transfer
0 coefficient and friction factor.
3000 8000 13000 18000 23000 28000 Following conclusions can be drawn:
Re
Pl /e = 11/2, e/H = 0.75 Pl /e = 7/2, e/H = 0.25
(1) the effect of Reynolds number, relative obstacle
Pl /e = 11/2, e/H = 0.5 Pl /e = 3/2, e/H = 0.75 height (e/H), and relative obstacle longitudinal pitch
Pl /e = 11/2, e/H = 0.25 Pl /e = 3/2, e/H = 0.5 (Pl /e) on the heat transfer coefficient and friction
Pl /e = 7/2, e/H = 0.75 Pl /e = 3/2, e/H = 0.25 factor has been studied for a fixed value of relative
Pl /e = 7/2, e/H = 0.5 obstacle transverse pitch (Pt /b) = 7/3.
(2) Thermal performances of such collectors have been
Figure 12: Friction factor ratio as a function of Reynolds number. found superior to that of smooth ones for the same
operating parameters. However, pressure drop in-
creases in such collectors.
on the same length of the heated plate so that the turbulence (3) Obstacle longitudinal pitch strongly affects the flow
intensity as well as air flow blockage increase resulting in pattern and hence the performance of the duct.
an increase in heat transfer enhancement and friction factor (4) The maximum enhancement in Nusselt number com-
penalty, respectively. pared to that of smooth duct is obtained for Pl /e =
The effect of relative obstacles height (e/H) on the 3/2, e/H = 0.75 at Reynolds number Re = 7276.82
heat transfer enhancement is negligible for relative obstacles and is of the order of 3.6.
longitudinal pitch (Pl /e) = 11/2 at higher Reynolds number
whereas for Pl /e = 7/2 and Pl /e = 3/2 the enhancement (5) The maximum thermohydraulic performance is ob-
in heat transfer is increasing with an increase in relative tained for Pl /e = 3/2, e/H = 0.50 at Reynolds num-
obstacles height. The friction factor ratio increases with an ber Re = 9757.38, and its value is of the order of 2.14.
increase in relative obstacles height (e/H) for all relative (6) The obstacle geometry chosen shows an appreciable
obstacles longitudinal pitch (Pl /e). heat transfer enhancement.
(7) In general, it can be said that for all the relative height
4. Conclusion and any arrangement of obstacles, the enhancement
of heat transfer is caused due to the increased turbu-
An experimental study of the flow of air, in a rectangular duct lence and also due to the secondary flow of air pro-
with delta shaped obstacles mounted on one wall subjected duced along the obstacles, which carry the heated air
to uniform heat flux and with remaining three smooth away from the plate surface.
10 Advances in Mechanical Engineering

Nomenclature [7] K. G. T. Hollands and E. C. Shewen, “Optimization of flow

passage geometry for air-heating, plate-type solar collectors,”
AC : Area of absorber Plate (m2 ) Journal of Solar Energy Engineering, vol. 103, no. 4, pp. 323–
b: Base of obstacles (m) 330, 1981.
Cp: Specific heat of air (J)/(kgK) [8] N. Moummi, S. Y. Ali, A. Moummi, and J. Y. Desmons, “Ener-
Dh : Equivalent hydraulic diameter of the air passage (m) gy analysis of a solar air collector with rows of fins,” Renewable
e: Obstacle height (m) Energy, vol. 29, no. 13, pp. 2053–2064, 2004.
e/H: Relative obstacle height [9] H. M. Yeh, C. D. Ho, and C. Y. Lin, “Effect of collector aspect
f: Friction factor ratio on the collector efficiency of upward type baffled solar
fo : Friction factor in duct having obstacle air heaters,” Energy Conversion and Management, vol. 41, no. 9,
pp. 971–981, 2000.
fs : Friction factor in a smooth duct
h: Convective heat transfer coefficient (W/m2 k) [10] F. Ozgen, M. Esen, and H. Esen, “Experimental investiga-
tion of thermal performance of a double-flow solar air heater
H: Duct height (m) having aluminium cans,” Renewable Energy, vol. 34, no. 11,
k: Thermal conductivity of air (W/mK) pp. 2391–2398, 2009.
L: Test section duct length (m) [11] H. Esen, “Experimental energy and exergy analysis of a dou-
ṁ: Mass flow rate of air (Kg/sec) ble-flow solar air heater having different obstacles on absorber
Nu: Nusselt number plates,” Building and Environment, vol. 43, no. 6, pp. 1046–
Nuo : Nusselt number of obstacle duct 1054, 2008.
Nus : Nusselt number of smooth duct [12] A. Abene, V. Dubois, M. Le Ray, and A. Ouagued, “Study of
Pr: Prandtl number a solar air flat plate collector: use of obstacles and application
ΔP: Pressure drop across the test section (Pa) for the drying of grape,” Journal of Food Engineering, vol. 65,
Pl /e: Relative obstacle longitudinal pitch no. 1, pp. 15–22, 2004.
Pt /b: Relative obstacle transversal pitch [13] ASHRAE STANDARD 93-86. Methods of Testing To Deter-
mine the Thermal Performance of Solar Collectors, 1977.
Re: Reynolds number
St: Stanton number [14] J. C. Han, “Heat transfer and friction in a channel with two
opposite rib roughened walls,” International Journal of Heat
Sto : Stanton number of obstacle duct
Transfer, vol. 106, no. 4, pp. 774–781, 1984.
Sts : Stanton number of smooth duct
[15] R. Karwa, S. C. Solanki, and J. S. Saini, “Heat transfer co-
Tf : Average temperature of air (K) efficient and friction factor correlations for the transitional
Tai : Bulk mean temperature of air at inlet (K) flow regime in rib-roughened rectangular ducts,” International
Tao : Bulk mean temperature of air at outlet (K) Journal of Heat and Mass Transfer, vol. 42, no. 9, pp. 1597–
Tp: Average temperature of plate (K) 1615, 1999.
W: Duct width (m) [16] S. J. Kline and F. A. McClintock, “Describe uncertainties in
ρ: Density of air (Kg/m3 ). single sample experiments,” Mechanical engineering, vol. 75,
pp. 3–8, 1953.
References [17] M. S. Bhatti and R. K. Shah, “Turbulent and transition flow
convective heat transfer,” in Handbook of Single-Phase Convec-
[1] A. E. Kabeel and K. K. Mecarik, “Shape optimization for ab- tive Heat Transfer, S. Kakac, R. K. Shah, and W. Aung, Eds.,
sorber plates of solar air collectors,” Renewable Energy, vol. 13, John Wiley & Sons, New York, NY, USA, 1987.
no. 1, pp. 121–131, 1998. [18] W. M. Kays and H. Perkin, “Forced convection internal flow in
[2] N. Depaiwa, T. Chompookham, and P. Promvonge, “Thermal ducts,” in Handbook of Heat Transfer, W. M. Rohsenow and I.
enhancement in a solar air heater channel using rectangular V. Hartnett, Eds., McGraw-Hill, New York, NY, USA, 1973.
winglet vortex generators,” in Proceedings of the PEA-AIT In-
ternational Conference on Energy and Sustainable Development:
Issues and Strategies, June 2010.
[3] C. Choudhury and H. P. Garg, “Design analysis of corrugated
and flat plate solar air heaters,” Renewable Energy, vol. 1, no. 5-
6, pp. 595–607, 1991.
[4] D. F. Dipprey and R. H. Sabersky, “Heat and momentum
transfer in smooth and rough tubes at various prandtl num-
bers,” International Journal of Heat and Mass Transfer, vol. 6,
no. 5, pp. 329–IN1, 1963.
[5] A. Hachemi, “Thermal performance enhancement of solar air
heaters, by a fan-blown absorber plate with rectangular fins,”
International Journal of Energy Research, vol. 19, no. 7, pp. 567–
578, 1995.
[6] A. A. Hegazy, “Performance of flat plate solar air heaters with
optimum channel geometry for constant/variable flow oper-
ation,” Energy Conversion and Management, vol. 41, no. 4,
pp. 401–417, 2000.