Energy Conversion and Management 78 (2014) 641651

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Performance analysis of photovoltaic thermal (PVT) water collectors

Ahmad Fudholi a,, Kamaruzzaman Sopian a, Mohammad H. Yazdi a, Mohd Hadz Ruslan a,
Adnan Ibrahim b, Hussein A. Kazem c
a

Solar Energy Research Institute, Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia
Universiti Kuala Lumpur Institute of Product Design and Manufacturing (UniKL IPROM), 56100 Cheras, Kuala Lumpur, Malaysia
c
Faculty of Engineering-Sohar University, PO Box 44, Sohar PCI 311, Oman
b

a r t i c l e

i n f o

Article history:
Received 21 June 2013
Accepted 11 November 2013
Available online 18 December 2013
Keywords:
Electrical performance
Thermal performance
Photovoltaic thermal (PVT)
PVT performance
Primary-energy saving efciency

a b s t r a c t
The electrical and thermal performances of photovoltaic thermal (PVT) water collectors were determined
under 500800 W/m2 solar radiation levels. At each solar radiation level, mass ow rates ranging from
0.011 kg/s to 0.041 kg/s were introduced. The PVT collectors were tested with respect to PV efciency,
thermal efciency, and a combination of both (PVT efciency). The results show that the spiral ow
absorber exhibited the highest performance at a solar radiation level of 800 W/m2 and mass ow rate
of 0.041 kg/s. This absorber produced a PVT efciency of 68.4%, a PV efciency of 13.8%, and a thermal
efciency of 54.6%. It also produced a primary-energy saving efciency ranging from 79% to 91% at a mass
ow rate of 0.0110.041 kg/s.
2013 Elsevier Ltd. All rights reserved.

1. Introduction
The idea of combining photovoltaic (PV) and solar thermal collector to provide electrical and heat energy is not new, yet it has
received limited attention. Growing concern about energy sources
and their usage has consequently increased interest in photovoltaic thermal (PVT) solar collectors. PVT solar collectors, which basically combine the functions of a at plate solar collector and a
photovoltaic panel, convert solar radiation directly into both electrical and thermal energies. Research on PVT started during the
mid-1970s focused on PVT collectors, with the primary aim of
increasing PV efciency. Domestic application was regarded as
the main market. Initially the focus was on air- and water-based
glazed collectors. Given these problems, the cost of a complete
PVT system is incredibly high and therefore unaffordable for
industrial and residential owners. One of the most attractive
applications of air- or water-based PVT collectors is the buildingintegrated photovoltaic thermal (BIPVT) system, which has undergone rapid development in recent years. However air-based PVT
systems have undergone more developed. The PVT system has
potential in generating both type of energies because of its higher
reliability and lower environment impact. Generally, a waterbased PVT system consists of a PV module, an absorber collector
in the form of tubes, a transparent glass cover, and an insulated
container. Over the next few years, BIPVT publications are

Corresponding author.
E-mail address: a.fudholi@gmail.com (A. Fudholi).
0196-8904/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.enconman.2013.11.017

expected to increase, and PVT products are expected to undergo rapid growth [13].
Several studies on PVT solar collectors have been conducted.
Fig. 1 shows PVT water collector with glass cover. The purpose of
the transparent cover, rstly to reduce the conduction losses from
the absorber collector through the restraint of the stagnant air
layer between the absorber collector and the glass and secondly
to reduce the radiation losses from the collectors. As shown in
Fig. 3, produced a hybrid PVT systems consist of PV modules made
from polycrystalline and amorphous solar cells with heat extraction unit mounted together using the copper sheet and pipes concept. The application aspects in the industry of PVT systems with
water heat extraction has been studied thoroughly and analyzed
with TRNSYS program. The study includes the industrial process
heat system that operated at two different (load supply) temperatures. The result shows that the electrical production using polycrystalline solar cell is more than when using amorphous solar
cells but in term of solar thermal fraction gives slightly lower results [3].
Theoretically analyses were based on a modied HottelWhillier model, and the results were validated using experimental data
from a prototype PVT collector [4]. The effects of design parameters, such as n efciency, thermal conductivity between the PV
cells and their supporting structure, and lamination method, on
both the electrical and thermal efciencies of the PVT were also
determined. Furthermore, PVT can be prepared using of lower cost
materials, such as precoated color steel, without signicantly
decreasing the efciency. Integration of PVT into rather than onto

642

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

Nomenclature
Ac
b
Cb
Cp
D
Dh
F
F0
FR
GT
h
k
L
l
_
m
N
n
p
Qu
S
T
UL
Ut

frontal area solar collector (m2)

collector width (m)
conductance of the bond between the n and square
tube
specic heat of working uid (J/kg C)
diameter (m)
hydraulic diameter (m)
n efciency factor
collector efciency factor
heat removal efciency factor
solar radiation at NOCT (W/m2)
heat transfer coefcient of uid (W/m2 C)
thermal conductivity (W/m C)
tube length (m)
thickness (m)
mass ow rate (kg/s)
number of glass cover
number of tube
collector perimeter (m)
actual useful heat gain (W)
solar radiation (W/m2)
temperature (C)
overall heat transfer coefcient (W/m2 C)
top loss coefcient (W/m2 C)

a building can also lower the system cost. In one study on a waterbased PVT system, a numerical model of wall-mounted PVT water
collector systems was developed by modifying the HottelWhillier
model, which was originally used for the thermal analysis of atplate solar thermal collectors. Recently, performance analysis
was conducted to analyze the exergy of PVT. The performance
and life cycle cost of PVT systems with PV technology different
from that of a similar PVT system were evaluated. The results show
that the use of PVT systems is generally advantageous over that of
similar PVT systems both from the efciency and economic point
of view. Mono-crystalline silicon PVT systems have higher energy
and exergy efciencies and are suitable for applications that have
higher energy and exergy demands or have limited space for
mounting, such as in multistory buildings [5].
A computer simulation was performed to analyze the system
performance. The combined effects of solar cell packing factor
and water mass ow rate on the electrical and thermal efciencies
were investigated. The simulation results showed that the increase
in working uid mass ow rate is benecial for PV cooling. However, the advantage brought by the increased ow rate diminishes
when the critical ow rate is exceeded, thereby decreasing thermal
efciency. System operation at the optimum mass ow rate can
not only improve the thermal performance of the system but also
meet the PV cooling requirement to achieve higher electrical performance [6]. A centralized PV and hot-water collector wall system

Fig. 1. PVT water collector with glass cover [3].

v
W

a
h
e

s
g
r

wind velocity (m/s)

tube spacing (m)
absorptance
collector tilt
emittance
transmittance
efciency
Stefans Boltzmann constant (W/m2C4)

Subscripts
a
ambient
abs
absorber thickness
c
cell

inlet uid
g
glass
i
inlet
o
outlet
p
plate
pm
mean plate
PV
photovoltaic
PVT
photovoltaic thermal
r
reference
t
tube
w
wind

mounted at vertical facades was experimentally studied [7]. The

results showed that the thermal efciency was 38.9% at reduced
(zero) temperature and electrical efciency was 8.56% during late
summer. A dynamic simulation model of a PVT and water heating
system was developed. This modeling approach was validated by
comparison with experimental data [8]. The results showed that
the electrical performance is affected by on-site shading. Moreover,
the output from the model showed high agreement with the experimental observations.
A computer simulation of a water-based PVT solar collector
system using energy models was developed. Higher economical
advantages relative to that of a conventional PV system were obtained. The annual average thermal and cell conversion efciencies
of a specic PVT system, which was mounted on a vertical wall of a
fully air-conditioned building with collectors equipped with a atbox-type thermal absorber and polycrystalline silicon cell, were
37.5% and 9.39%, respectively, compared with the normal building
faade [9]. A computational uid dynamic (CFD) model for a novel
PVT collector was developed and experimentally validated [10].
The results indicated that PV cell efciency can be increased to
5.3% and the outlet water temperature of the collector is suitable
for domestic hot-water use. The effect of ow distribution on the
PV performance of a PVT water collector was also investigated
[11]. The results showed that parameters such as the manifoldto-riser pipe ratio, array geometry, manifold ow direction, and
mass ow rate affect the ow distribution, which, in turn, affects
PV conversion.
Innovative applications of PVT collector were performed recently [1218]. PVT applications are cost-effective solar energy
applications. However, additional studies must still be conducted,
particularly on the design of new thermal absorber collectors.
Alternative designs of PVT solar collectors are presented in this paper. A prototype of this new absorber was constructed. To date,
studies on water-based PVT collectors have been few. Therefore,
further experimental and analytical research should be performed
to improve the electrical and thermal performance of water-based
PVT solar collectors using new absorber collector designs.

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

643

2. Materials and methods

The fabricated PVT water collector was tested at the Solar
Energy Laboratory, Faculty Science and Technology, Universiti Kebangsaan Malaysia. The solar simulator at this laboratory consists
of 23 units of halogen tungsten lamps (Brillanta), with an output
power of 500 W each. The lamps are arranged in nine columns
with each column.
Indoor testing is important because in the laboratory, solar radiation can be controlled, and accurate measurements can be obtained. The control parameters for indoor tests include the input
and output temperatures, ambient temperature, wind velocity at
the collector surface, useful current and voltage, and water ow
rate inside the tube. These parameters were matched to the standards set for all PVT absorber collectors.
2.1. New design of absorbers
Fig. 2 shows three PVT water collector designs. The parameters
of these fabricated design congurations are listed in Table 1. The
rst collector (Fig. 2a) is a web ow absorber, the second collector
(Fig. 2b) is direct ow absorber, and the third collector (Fig. 2c) is
spiral ow absorber. Fig. 3 shows the example of a complete PVT
water collector design with spiral ow absorber. The absorber collectors consisted of either rectangular or round hollow stainlesssteel tubes. The tubes were connected using a tungsten inert gas
welding method. Nipples were welded to the rear surface for manifold attachment. Silicone was used to seal the ends and to glue the
top absorber sheet into place. The absorber collector, which consists of a single unilateral channel for water to ow, was inserted
underneath the standard PV module 1 m high, 0.65 m long, and
0.3 m thick. The thermal insulator was packed beneath the absorber collector to prevent heat from escaping further and to provide
more uniform temperatures throughout the system. A standard PV
module, represented as a at-plate single glazing sheet of polycrystalline silicon, was laminated and bonded using a high-temperature silicone adhesive and sealant. Once sealed and watertight,
the absorber collectors were attached to be the bottom side of

Fig. 3. PVT water collector design with spiral ow absorber.

the PV module and then encapsulated in a polyvinyl resin to form

a complete PVT system 815 mm  628 mm  30 mm in size. The
absorber collectors were designed in the form of a continuous coil
or congured tube and consist of at least one inlet and outlet to allow the medium (water) to enter and exit a coil, respectively. The
inlet and outlet nipples were arranged to allow the medium
(water) to ow in and out and cover the entire PV module. The conguration is as follows: a low-temperature (cold) medium (water)
enters the coil, ows in and out, and leaves the absorber collector
as hot water, which can be consumed or stored for later use. In this
manner, solar energy can be fully utilized. Normally, in a PVT system, the hot-water storage tank is located near the ground level,
whereas the solar module is mounted on the roof. In this PVT system, the hot-water storage was located as near as possible to the
collector to maintain the water pressure from the pump to the collector and vice versa. The hot-water storage was connected to the
at-plate collector through sets of pipes. The water was circulated
by a pump. The pipes were assumed to be well-insulated to prevent heat loss. Energy was absorbed only by the collector unit.
The energy absorbed by the collector was used in heating the
water. Only the water storage tank was assumed to lose energy.
The inlet water temperature of the collector was assumed to be
the same as the mean water temperature in the storage tank.

Fig. 2. (a) Web ow absorber, (b) direct ow absorber and (c) spiral ow absorber.

Table 1
The parameters congurations of PVT solar collectors.
Absorber type

Unit

Web ow (Fig. 2a)

Absorber material: round hollow tubes of stainless steel

Absorber collector module: 1 channel each of size 12.7 mm  1 mm  1000 mm (length) and 640 mm (width)
Method of joining: welding
Inlet/outlet no: 2

Direct ow (Fig. 2b)

Absorber material: rectangular hollow tubes of stainless steel

Absorber collector module: 19 channel each of size 12.7 mm  12.7 mm  1 mm  1000 mm (length) and 640 mm (width)
Method of joining: welding
Inlet/outlet no: 4

Spiral ow (Fig. 2c)

Absorber material: rectangular hollow tubes of stainless steel

Absorber collector module: 1 channel each of size 12.7 mm  12.7 mm  1 mm  700 mm (length) and 640 mm (width)
Method of joining: welding
Inlet/outlet no: 4

644

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

The temperature of the water entering the tank was assumed to be

the same as the outlet temperature of the collector.
2.2. Method
The three PVT water collectors were tested in a laboratory
atmosphere at temperatures and solar radiation levels similar to
those used during the PV module testing. The solar collector was
tested using a simulator based on the testing procedures of ISO
9806-1: 1994 and the ANSI/ASHRAE Standard 93: 2003. The testing
procedures include the following features:
(i) The testing procedure can be used to evaluate the performance of PVT solar collectors.
(ii) The thermal performance of the collectors can be determined partly by obtaining the instantaneous efciencies of
different combinations of incident solar radiation, ambient
temperature, and inlet uid temperature.
(iii) The test measures the rate of the incident solar radiation
falling onto the solar collector as well as the rate of energy
addition to the transfer uid as it passes through the
collector.
(iv) The testing should be conducted under steady-state or
quasi-steady-state conditions.
The PVT collectors were exposed to solar radiation of 500, 600,
700 and 800 W/m2. When the load was applied to the collectors,
the changes in current and voltage were recorded for each solar
radiation level. During this testing, mass ow rate of 0.011,
0.024, 0.032 kg/s, and 0.041 kg/s were set. Any changes in the collector due to these mass ow rates were recorded. Data were collected and stored every minute using a data-acquisition system,
and these data were subsequently used to calculate the PV efciency and thermal efciency of the collector. The water inlet
and outlet during testing were controlled and connected back to
the storage tank to form a close-loop system. The complete measuring setup for the PVT collector is shown in Fig. 4.
3. Analysis of the PVT solar collectors
The performance of the PVT collectors can be expressed by a
combination of efciency expressions [19] consisting of thermal
efciency (gth) and electrical efciency (gPV). These efciencies

Table 2
PVT solar collector characteristics.
Description

Symbol

Value

Unit

Ambient temperature
Collector area
Number of glass cover
Emittance of glass
Emittance of plate
Collector tilt
Fluid thermal conductivity
Specic heat of working uid
Back insulation conductivity
Back insulation thickness
Insulation conductivity
Edge insulation thickness
Absorber conductivity
Absorber thickness
Fin conductivity
Fin thickness
Heat transfer coefcient from cell to absorber
Heat transfer inside tube
Transmittance
Absorptance

Ta
Ac
N
eg
ep
h
kf
Cp
kb
lb
ke
le
kabs
labs
kf
d
hca
h

20
0.65
1
0.88
0.95
14
0.613
4180
0.045
0.05
0.045
0.025
51
0.002
84
0.0005
45
333
0.88
0.95

C
m2

W/m C
J/kg C
W/m C
m
W/m C
m
W/m C
m
W/m C
m
W/m C
W/m C

s
a

usually include the ratio of the useful thermal gain and electrical
gain of the system to the incident solar irradiation on the collector
gap within a specic time or period. The analytical parameters of
the PVT collector are presented in Table 2. The total efciencies,
known as total efciency or PVT efciency (gPVT), are used to evaluate the overall performance of the system [1921]:

gPVT gth gPV

Electrical energy (gf) is a high-grade form of energy gain, the

primary energy-saving efciency is proposed as another performance indicator of the energy-grade difference between electricity
and thermal energy and is given by [1921]:

gf

gPV
gth
gp

where gp is the electric-power generation efciency of a conventional power plant; its value can be taken as 38%. The evaluation
indicator of the primary energy-saving efciency also considers
the quality and quantity of the energy that the PVT system converts
into solar energy.

Fig. 4. Schematic diagram of IPVT water collector.

645

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

3.1. Analysis of at plate PVT collector

Ue

The thermal performance of PVT is affected by a number of system design parameters and operating conditions. In this study, the
system was analyzed using various congurations of solar radiation, ambient temperature, and ow rate conditions. The collector
was assumed to be represented as a at-plate collector with a single glazing sheet. Based on this assumption, the thermal and PV
performance of the PVT unit was evaluated by the deriving the efciency parameters based on the HottelWhillier equations [22].
The thermal efciency of a conventional at-plate solar collector
is the ratio of the useful thermal energy (Qu) to the overall incident
solar radiation (S) and can be expressed as:

Q
gth u
S

_ p T o  T i
Q u mC

The difference between the absorber solar radiation and thermal heat losses is determined using the HottelWhillier equation
[23]:

Q u Ac F R GT saPV  U L T i  T a 

where Ac is the collector area, Ta is the ambient temperature, Ti is

the inlet temperature, UL is the overall collector heat loss, (sa)PV
is the PV thermal efciency, GT is the solar radiation at NOCT (radiation level 800 W/m2, wind velocity 1 m/s, and ambient temperature 26 C), and FR is the heat removal efciency factor introduced
[22,23]. This factor is expressed as follows:




_ p
Ac U L F 0
mC
FR
1  exp 
_ p
Ac U L
mC

where F0 is the collector efciency factor, which is calculated using

"
F

8
>
<
>
:

12
91
>
1=
h
ie
T pm T a
hw >
;
N

C
T pm

Nf

rT pm T a T 2pm T 2a
1

ep 0:00591Nhw

2Nf 10:133ep

eg

N

13

where

C 5201  0:000051b2

14

f 1 0:089hw  0:1166hw ep 1 0:07866N

15



100
e 0:43 1 
T pm

16

The useful collected heat absorbed by the at-plate solar collector can be given as the combined results of the average mass ow
_ heat capacity of owing medium (Cp) and temperature
rate m,
difference at the collector inlet (Ti) and outlet (To) and can be expressed as:

Ut

ke pl
L e Ac

1
UL

U L Dh W  Dh F

1
1

C b 2a bhfi

T pm T i

Q=Ac
1  F R
F R UL

17

where p is the collector perimeter, N is the number of glass covers, r

is the StefanBoltzmann constant, ep is the plate emittance, eg is the
glass emittance, b is the collector tilt, Tpm is the mean plate temperature, and hw is the wind heat-transfer coefcient.
The heat transfer coefcients such as the forced convection (hw)
can be calculated using Eq. (18) [25], whereas the natural heat
transfers (hnat) can be calculated using Eq. (19) [25], as follows:

hw 2:8 3:0v

18

hnat 1:78T pm  T a

19

A combination of the natural and forced convection heat transfer (Eqs. (18) and (19)) determines the overall convection heat
transfer (hc) and possibly the overall top loss heat-transfer coefcient for the collector [26].

hc

q
3
3
hw hnat

20

where a is the duct width, b is the duct height, Cb is the conductance

of the bond between the n and square tube, h is the heat-transfer
coefcient of the uid, Dh is the hydraulic diameter, and F is the n
efciency factor given by

The useful heat gain produced by the PVT collector can then be
calculated using Eqs. (320). By rearranging Eq. (3), the thermal
efciency of the collector is expressed as [27]



h
tanh M WD
2
F q
h
M WD
2

gth F R saPV  F R U L
8

2ab
a b

The coefcient M in Eq. (8) considers both the thermal conductivity of the absorber and the PV cell. M is calculated using [11,24]

s
UL
M
kabs labs kPV lPV

10

where kabs is the absorber thermal conductivity, labs is the absorber thickness, kPV is the PVT conductivity, and lPV is the PV collector
thickness. The overall loss coefcient (UL) of the collector is the
sum of the edge (Ue) and top (Ut) loss coefcients and can be expressed as

UL Ue Ut

21

3.2. Principle of the PV module

where

Dh



Ti  Ta
GT

11

The electrical efciency of the PV module (gPV), which is a function of module temperature, is given by [3,28]

gPV gr 1  cT c  T r

22

where gr is the reference efciency of the PV module (gr = 0.12), c is

a temperature coefcient (c = 0.0045 C), Tc is the cell temperature,
and Tr is the reference temperature.
4. Results and observations
The performance and efciency of the PVT collectors are determined by their electrical and thermal characteristics. The analyses
of the PVT collectors are segregated into three sections, namely PV
efciency, thermal efciency, and combination of both.

646

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

4.1. Effects of the PV module temperature on the PVT collector PV

efciency
Figs. 58 and simplied in Tables 35 show the analysis results
of the PVT collectors after exposure to 500800 W/m2 of solar radiation at 0.0110.041 kg/s mass ow rates. The results show that
the PV efciencies of the collectors signicantly changed under
various PV temperature and mass ow rates. Among the basic
PVT absorber collectors, the web system at 500 W/m2 exhibited a
decreased PV efciency of 0.35%, from 11.42% at 47.88 C to
11.07% at 50.24 C. At 600 W/m2, the total PV efciency increased
to 0.39% but showed an average decrease from 11.94% at
48.57 C to 11.55% at 51.65 C. At 700 W/m2, the total PV efciency
increased by 0.42% from 12.13% at 49.65 C to 11.71% at 52.85 C.
When the solar radiation further increased to 800 W/m2, the total
PV efciency increased by 0.46% from 12.37% at 50.03 C to 11.91%
at 53.54 C. Meanwhile, the change trend of direct ow is nearly
similar to that of web ow, with the total PV efciency increasing
by 0.37% from 11.78% at 47.28 C to 11.41% at 50.15 C. When the
solar radiation was increased to 600 W/m2, the efciency further
increased by 0.42, from 12.18% at 48.5 C to 11.76% at 51.75 C.
At 700 W/m2, the total efciency increased by 0.45% from 12.38%
at 49.22 C to 11.93% at 52.95 C. Further increase in the solar radiation to 800 W/m2 resulted in a 0.5% increase in the total PV efciency, from 12.69% at 49.89 C to 12.19% at 53.64 C. On the
other hand, the spiral ow exhibited a 0.53% increase in the total
PV efciency, from 12.52% at 46.24 C to 11.99% at 50.86 C. When
the solar radiation was further increased to 600 W/m2, the total PV
efciency increased by 0.58% from 13.07% at 46.84 C to 12.49% at
51.95 C. At 700 W/m2, the total PV efciency increased by 0.63%
from 13.47% at 47.64 C to 12.84% at 52.83 C. Finally, at 800 W/
m2, the PV efciency increased by 0.63%. This inclination is resulting from the average efciency, which dropped due to the cooling
effect of the collector. The data collected shows when the temperature increased, mass ow rate dropped, the efciency dropped
simultaneously. As result, the collector efciency dropped from
13.81% at 48.03 C to 13.01% at 53.35 C.

Fig. 6. Changes in PV efciency with the mean PV temperature of the PVT absorber
collectors under 600 W/m2 of solar radiation.

Fig. 7. Changes in PV efciency with the mean PV temperature of the PVT absorber
collectors under 700 W/m2 of solar radiation.

4.2. Effect of mass ow rate on the PVT collectors

The mass ow rate through the collectors and into the designated channels indirectly affects PV module cooling. The effects
of the mass ow rate on the absorber collectors are shown in
Figs. 911. The mass ow rates used in this analysis (0.011
0.041 kg/s) were later applied under various solar radiation levels.
The results show that increasing the mass ow rate simultaneously
decreased the PV temperature of the PVT collectors at all solar radiation levels. At the same mass ow rate, the PV temperatures increased in the solar radiation level. Figs. 59 show that from
0.011 kg/s to 0.041 kg/s mass ow rates and under 500 W/m2 solar

Fig. 5. Changes in PV efciency with the mean PV temperature of the PVT absorber
collectors under 500 W/m2 of solar radiation.

Fig. 8. Changes in PV efciency with the mean PV temperature of the PVT absorber
collectors under 800 W/m2 of solar radiation.

radiation, the temperature decreased from 50.20 C to 47.76 C and

PV efciency simultaneously increased from 11.07% to 11.42%.
When solar radiation was increased to 800 W/m2, temperature
dropped from 53.5 C to 50 C, whereas PV efciency increased
from 11.91% to 12.37%. For the direct ow absorber as (Fig. 10)
and referring to Figs. 58, at solar radiation of 500 W/m2 and when
mass ow rate increased from 0.011 kg/s to 0.041 kg/s, PV temperature dropped from 50.11 C to 47.18 C, and PV efciency increased from 11.41% to 11.78%. The same result was obtained
when the solar radiation increased to 800 W/m2: temperature decreased from 53.6 C to 49.8 C, whereas PV efciency increased
from 12.19% to 12.69%.
For the spiral ow absorber collector (Figs. 58 and 11), PV temperature decreased from 50.76 C to 46.2 C, and PV efciency increased from 11.99% to 12.52% at a solar radiation of 500 W/m2

647

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

Table 3
Results of PV efciency (gPV) and PV temperature (TPV) for web ow under various mass ow rates and solar radiations.
500 W/m2

600 W/m2

700 W/m2

800 W/m2

_ (kg/s)
m

TPV (C)

gPV (%)

TPV (C)

gPV (%)

TPV (C)

gPV (%)

TPV (C)

gPV (%)

0.011
0.013
0.016
0.020
0.024
0.027
0.029
0.032
0.035
0.038
0.041

50.24
49.93
49.65
49.26
48.83
48.58
48.46
48.27
48.09
47.90
47.88

11.07
11.12
11.18
11.22
11.27
11.30
11.32
11.35
11.38
11.40
11.42

51.65
51.04
50.63
50.26
49.76
49.59
49.27
49.09
48.87
48.76
48.57

11.55
11.62
11.69
11.75
11.80
11.84
11.87
11.90
11.92
11.93
11.94

52.85
52.33
51.95
51.37
50.89
50.69
50.45
50.22
50.01
49.82
49.65

11.71
11.78
11.82
11.93
11.98
12.01
12.02
12.04
12.07
12.10
12.13

53.54
53.03
52.65
52.06
51.69
51.17
51.05
50.84
50.55
50.14
50.03

11.91
12.00
12.07
12.15
12.20
12.22
12.26
12.28
12.32
12.35
12.37

Table 4
Results of PV efciency (gPV) and PV temperature (TPV) for direct ow under various mass ow rates and solar radiations.
500 W/m2

600 W/m2

700 W/m2

800 W/m2

_ (kg/s)
m

TPV (C)

gPV (%)

TPV (C)

gPV (%)

TPV (C)

gPV (%)

TPV (C)

gPV (%)

0.011
0.013
0.016
0.020
0.024
0.027
0.029
0.032
0.035
0.038
0.041

50.15
50.04
49.56
48.84
48.53
48.17
48.35
48.06
47.85
47.67
47.28

11.41
11.45
11.50
11.56
11.60
11.62
11.65
11.68
11.70
11.74
11.78

51.75
51.34
50.86
50.47
49.95
49.63
49.46
49.38
48.99
48.76
48.50

11.76
11.84
11.90
11.96
12.01
12.05
12.09
12.12
12.14
12.15
12.18

52.95
52.54
51.93
51.36
50.74
50.57
50.15
49.94
49.63
49.47
49.22

11.93
11.99
12.04
12.13
12.18
12.23
12.25
12.26
12.31
12.35
12.38

53.64
53.25
52.86
52.13
51.56
51.04
51.03
50.66
50.23
50.06
49.89

12.19
12.24
12.30
12.41
12.49
12.52
12.55
12.59
12.62
12.66
12.69

Table 5
Results of PV efciency (gPV) and PV temperature (TPV) for spiral ow under various mass ow rates and solar radiations.
500 W/m2

600 W/m2

700 W/m2

800 W/m2

_ (kg/s)
m

TPV (C)

gPV (%)

TPV (C)

gPV (%)

TPV (C)

gPV (%)

TPV (C)

gPV (%)

0.011
0.013
0.016
0.020
0.024
0.027
0.029
0.032
0.035
0.038
0.041

50.86
49.94
49.33
48.72
48.05
47.82
47.40
47.07
46.74
46.43
46.24

11.99
12.06
12.15
12.22
12.27
12.32
12.36
12.40
12.44
12.49
12.52

51.95
51.13
50.44
49.53
48.76
48.43
48.06
47.82
47.58
47.15
46.84

12.49
12.53
12.66
12.75
12.84
12.90
12.92
12.94
13.00
13.05
13.07

52.83
52.15
51.23
50.26
49.73
49.21
48.86
48.48
48.28
47.86
47.64

12.84
12.90
13.00
13.14
13.22
13.29
13.31
13.36
13.42
13.44
13.47

53.35
52.73
52.05
50.83
50.25
49.73
49.26
48.97
48.68
48.34
48.03

13.01
13.20
13.33
13.44
13.56
13.63
13.66
13.71
13.75
13.79
13.81

Fig. 9. Changes in PV temperature of web ow absorber with the mass ow rates

under different solar radiation levels.

with the mass ow rate increased from 0.011 kg/s to 0.014 kg/s. At
800 W/m2 solar radiation level, PV temperature was 53.3 C and

Fig. 10. Changes in PV temperature of the direct ow absorber with the mass ow
rate under different solar radiation levels.

decreased to 48 C, whereas PV efciency increased from 13.01%

to 13.81%.

648

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

63.34 C to 49.54 C, whereas those for the direct ow and spiral

ow absorbers dropped from 62.57 C to 50.32 C and 65 C to
52.4 C, respectively.

4.4. PVT performance of PVT water collectors

The performance of the PVT collectors can be represented by a
combination of efciency expressions, which consists of electrical
efciency and thermal efciency. The sum of both efciencies,
known as the PVT efciency, is used to evaluate the overall performance of the system. Based on the testing performed on the collectors, both efciencies increased when the mass ow rate increased.
Therefore, the total efciency (PVT efciency) and primary energysaving efciency increased concurrently when the mass ow rate
increased.
Figs. 1417 and Table 6 show PV, thermal, PVT, and primary energy-saving efciencies, respectively, of the web ow, direct ow,
and spiral ows at 0.0110.041 kg/s mass ow rates and 800 W/
m2 solar radiation. The PVT efciency of the web, direct, and spiral
ow absorbers further increased by 7.4%, 8.2%, and 10.4%, respectively, and their primary energy-saving efciency increased by
8.2%, 9%, and 11.7%, respectively. The web ow absorber exhibited
an increase PVT efciency from 53.0% to 60.44% when the mass
ow rate increased from 0.011 kg/s to 0.041 kg/s. The other
absorbers exhibited the same trend. The direct ow absorber
showed an increased PVT efciency from 58.62% to 66.82% when
the mass ow rate increased. The highest PVT efciency obtained
was that of the spiral ow absorber, which exhibited an increase
from 58.01% to 68.42%.
The web ow absorber showed an increase in primary energysaving efciency from 72.45% to 80.62%. Meanwhile, the direct
ow absorber exhibited an increased primary energy-saving

Fig. 11. Changes in PV temperature of the spiral ow absorber with the mass ow
rate under different solar radiation levels.

Fig. 12. Changes in thermal efciency of 800 W/m2 solar radiation with the mass
ow rate for different ow absorbers.

4.3. Thermal performance of PVT collectors

Fig. 12 and Table 6 show the increase in thermal efciency with
the mass ow rates of the PVT collectors under 800 W/m2 of solar
radiation. The web ow absorber exhibited an increase in the thermal efciency from 41.11% to 48.07% as a result of the increased
mass ow rate from 0.011 kg/s to 0.041 kg/s. The other absorbers
exhibited the same trend. The direct ow absorber exhibited an increased thermal efciency from 46.43% to 54.13% when the mass
ow rate increased from 0.011 kg/s to 0.041 kg/s. The highest thermal efciency was produced by the spiral ow absorber, which
exhibited an increase from 58.01% to 68.42% when the mass ow
rate increased from 0.011 kg/s to 0.041 kg/s. Fig. 13 and Table 7
show that the outlet temperature of the collectors decreased with
increased mass ow rates under 800 W/m2 of solar radiation. The
outlet temperature of the web ow absorber dropped from

Fig. 13. Changes in outlet temperature of 800 W/m2 solar radiation with the mass
ow rates for different ow absorbers.

Table 6
Results of efciencies at different mass ow rates under 800 W/m2 of solar radiation.

_ (kg/s)
m

Web ow
Efciencies (%)

Direct ow
Efciencies (%)

Spiral ow
Efciencies (%)

gth

gPV

gPVT

gf

gth

gPV

gPVT

gf

gth

gPV

gPVT

gf

0.011
0.013
0.016
0.020
0.024
0.027
0.029
0.032
0.035
0.038
0.041

41.11
42.66
43.98
46.66
47.03
47.25
47.37
47.52
47.66
47.79
48.07

11.91
12.00
12.07
12.15
12.20
12.22
12.26
12.28
12.32
12.35
12.37

53.02
54.66
56.05
58.81
59.23
59.47
59.63
59.80
59.98
60.14
60.44

72.45
74.24
75.74
78.63
79.14
79.41
79.63
79.84
80.08
80.29
80.62

46.43
48.53
50.01
50.98
52.47
52.67
52.84
52.97
53.08
53.19
54.13

12.19
12.24
12.30
12.41
12.49
12.52
12.55
12.59
12.62
12.66
12.69

58.62
60.77
62.31
63.39
64.96
65.19
65.39
65.56
65.70
65.85
66.82

78.51
80.74
82.38
83.64
85.34
85.62
85.87
86.10
86.29
86.51
87.52

45.00
48.49
49.64
51.05
51.73
51.95
52.07
52.25
52.38
52.51
54.61

13.01
13.20
13.33
13.44
13.56
13.63
13.66
13.71
13.75
13.79
13.81

58.01
61.69
62.97
64.49
65.29
65.58
65.73
65.96
66.13
66.30
68.42

79.24
83.23
84.72
86.42
87.41
87.82
88.02
88.33
88.56
88.80
90.95

649

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

Table 7
Results of temperatures at different mass ow rates under 800 W/m2 of solar radiation.
Web ow

Direct ow

Spiral ow

_ (kg/s)
m

Ti (C)

To (C)

To  Ti (C)

TPV (C)

Ti (C)

To (C)

To  Ti (C)

TPV (C)

Ti (C)

To (C)

To  Ti (C)

TPV (C)

0.011
0.013
0.016
0.020
0.024
0.027
0.029
0.032
0.035
0.038
0.041

51.00
48.32
46.32
41.04
40.93
40.85
40.81
40.79
40.72
40.64
40.20

63.34
60.39
58.25
52.90
52.13
51.36
50.59
50.12
49.85
49.70
49.54

12.34
12.07
11.93
11.86
11.20
10.51
9.78
9.33
9.13
9.06
9.34

53.54
53.03
52.65
52.06
51.69
51.17
51.05
50.84
50.55
50.14
50.03

50.90
47.22
45.00
43.78
41.22
41.12
40.96
40.91
40.88
40.82
39.00

62.57
58.67
56.36
55.08
52.50
52.00
51.50
50.93
50.63
50.48
50.32

11.67
11.45
11.36
11.30
11.28
10.88
10.54
10.02
9.75
9.66
11.32

53.64
53.25
52.86
52.13
51.56
51.04
51.03
50.66
50.23
50.06
49.89

54.00
47.32
45.78
43.63
42.78
42.64
42.58
42.42
42.34
42.25
42.09

65.00
61.00
59.04
56.97
54.99
54.84
54.72
54.54
53.78
53.18
52.40

11.00
13.68
13.26
13.34
12.21
12.20
12.14
12.12
11.44
10.93
10.31

53.35
52.73
52.05
50.83
50.25
49.73
49.26
48.97
48.68
48.34
48.03

Fig. 14. Changes in web ow absorber efciency with mass ow rate under 800 W/
m2 of solar radiation.

Fig. 16. Changes in spiral ow absorber efciency with mass ow rate under
800 W/m2 of solar radiation.

conventional solar water heater, which is a PVT system known as

the integrated photovoltaic/thermal system was studied. The results showed that the solar PVT collector made from a corrugated
polycarbonate module produced high thermal efciency. Huang
et al. [29] suggested that further improvements can be achieved
by using proper insulation for the PVT design. They managed to
achieve a PV efciency of 9%, a thermal efciency of 38%, and a total efciency (PVT efciency) of 47%.
A PVT collector system for domestic application in China was
developed [30,31]. In this experiment, an aluminum-alloy at
box with a square or rectangular channel was designed and
constructed. The test results showed the high efciency of the

Fig. 15. Changes in direct ow absorber efciency with mass ow rate under
800 W/m2 of solar radiation.

efciency from 78.51% to 87.52%. The highest primary-energy saving efciency was exhibited by the spiral ow absorber, which
showed an increase from 79.24% to 90.95%.
Figs. 1820 and Table 7 show inlet, outlet, and PV temperatures,
respectively, of the web ow, direct ow, and spiral ows at
0.0110.041 kg/s mass ow rates and 800 W/m2 solar radiation.
The results show that increasing the mass ow rate simultaneously
decreased the temperatures (inlet, outlet and PV) of the PVT
collectors.
4.5. Comparison with other absorber collector designs
A performance evaluation was conducted to investigate the
integration of PV and thermal system. The performance of a

Fig. 17. Comparison of the PVT and primary energy-saving efciencies of the
absorbers at different mass ow rates under 800 W/m2 of solar radiation.

650

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

Table 8
The comparison present study with other absorber collector designs [20,2933].

Fig. 18. Changes in inlet temperature (Ti), outlet temperature (To) and PV
temperature (TPV) of 800 W/m2 solar radiation with the mass ow rate for web
ow absorber.

Performance

Ref.

gPV = 9.5%, gth = 50%, gPVT = 59.5%

gPV = 9%, gth = 38%, gPVT = 47%
gPVT = 65% at 0 C operation temperature
gPV = 11%, gth = 51%, gPVT = 62%
gf = 64.9%
gPV = 9.87%, gth = 40%, gPVT = 49.87%
gPV = 10.15%, gth = 45%, gPVT = 55.15%
gPV = 11.912.4%, gth = 41.148%, gPVT = 5360.4%
gf = 72.480.6% for web ow absorber
gPV = 12.212.7%, gth = 46.454.1%%, gPVT = 58.666.8%
gf = 78.587.5% for direct ow absorber
gPV = 1313.8%, gth = 4554.6%, gPVT = 5868.4%
gf = 79.290.9% for spiral ow absorber

[20]
[29]
[30]
[31]
[32]
[33]
Present study

system. Ji et al. [33] managed to achieve a PV efciency of 10.15%, a

thermal efciency of 45%, and a total efciency of 55.15%. Table 8
shows the summarized comparison results between the proposed
absorber congurations and other absorber collector designs
[20,2933]. According to the literature [20], the PVT water-based
collector can achieve a maximum electrical efciency of around
9.5% and a thermal efciency of approximately 50%.
5. Conclusion

Fig. 19. Changes in inlet temperature (Ti), outlet temperature (To) and PV
temperature (TPV) of 800 W/m2 solar radiation with the mass ow rate for direct
ow absorber.

Fig. 20. Changes in inlet temperature (Ti), outlet temperature (To) and PV
temperature (TPV) of 800 W/m2 solar radiation with the mass ow rate for spiral
ow absorber.

combined system, with a nearly 65% primary energy saving for daily exposure at reduced (zero) temperature operation. Chow et al.
[30,31] managed to achieve a PV efciency of 11%, a thermal efciency of 51%, and a total efciency (PVT efciency) of 62%. A similar experiment was performed using an aluminum-alloy at box
with a square or rectangular channel and polycrystalline silicone
cells, which utilized water as a [32]. The results showed that the
thermal efciency reached 40% when the initial temperature in
the system matched the daily mean ambient temperature. He
et al. [32] managed to achieve a PV efciency of 9.87%, a thermal
efciency of 40%, and a total efciency of 49.87%.
Another experiment on a natural-circulation hybrid PVT waterheating system was conducted [33]. In this experiment, a sensitivity study of the system showed that by combining different
systems, the installation area produce more energy per unit surface
area than a system consisting of one PV module and one hot-water

PVT water collector consisting of a combined PV module and an

absorber collector were investigated. The performances of three
PVT water collectors were determined. The results indicate that a
solar radiation level of 800 W/m2 and a mass ow rate of
0.041 kg/s, the spiral ow absorber produced a PVT efciency of
approximately 65%, a PV efciency of 13%, and a thermal efciency
of 52%. It also exhibited a primary energy-saving efciency of 79
91% at mass ow rates ranging of 0.0110.041 kg/s.
The results show that the efciency of the PV module increase
when the temperature decrease. The decrease in temperature is
not linear with the mass ow rate increase. However, temperature
signicantly decrease after the mass ow rate reaches 0.024 kg/s.
Overall, the efciency of the PVT water collectors increases with
the mass ow rates under various solar radiation levels. This result
is due to the increase in the cooling factor of the PV module cells
when the mass ow rate increases. Therefore, mass ow rate
indirectly contributes to the increase in PVT water collector
temperature.
Acknowledgements
The authors would like to thank the Ministry of Science,
Technology, and Innovation Malaysia for funding this research
(Sciencefund 03-01-02-SF0039) and the Solar Energy Research
Institute (SERI), Universiti Kebangsaan Malaysia for providing the
laboratory facilities and technical support.
References
[1] Touafek K, Haddadi M, Malek A. Design and modeling of a photovoltaic
thermal collector for domestic air heating and electricity production. Energy
Build 2013;59:218.
[2] Daghigh R, Ibrahim A, Jin GL, Ruslan MH, Sopian K. Predicting the performance
of amorphous and crystalline silicon based photovoltaic solar thermal
collectors. Energy Convers Manage 2011;52:17417.
[3] Ibrahim A, Othman MY, Ruslan MH, Mat S, Sopian K. Recent advances in at
plate photovoltaic/thermal (PV/T) solar collectors. Renew Sustain Energy Rev
2011;15:35265.
[4] Anderson TN, Duke M, Morrison GL, Carson JK. Performance of a building
integrated photovoltaic/thermal (BIPVT) solar collector. Sol Energy
2009;83:44555.

A. Fudholi et al. / Energy Conversion and Management 78 (2014) 641651

[5] Agrawal B, Tiwari GN. Life cycle cost assessment of building integrated
photovoltaic thermal (BIPVT) systems. Energy Build 2010;42:147281.
[6] Ji J, Jun H, Chow TT, Hua Y, Jianping L, Wei H, et al. Effect of uid ow and
packing factor on energy performance of a wall-mounted hybrid photovoltaic/
water heating collector system. Energy Build 2006;38:13807.
[7] Chow TT, He W, Ji J. An experimental study of faade-integrated photovoltaic/
water-heating system. Appl Therm Eng 2007;27:3745.
[8] Chow TT, He W, Chan ALS, Fong KF, Lin Z, Ji J. Computer modeling and
experimental validation of a building-integrated photovoltaic and water
heating system. Appl Therm Eng 2008;28:135664.
[9] Chow TT, Chan ALS, Fong KF, Lin Z, He W, Ji J. Annual performance of buildingintegrated photovoltaic/water-heating system for warm climate application.
Appl Energy 2009;86:68996.
[10] Corbin CD, Zhai ZJ. Experimental and numerical investigation on thermal and
electrical performance of a building integrated photovoltaicthermal collector
system. Energy Build 2010;42:7682.
[11] Ghani F, Duke M, Carson JK. Effect of ow distribution on the photovoltaic/
thermal (BIPV/T) collector. Sol Energy 2012;86:151830.
[12] Kandilli C. Performance analysis of a novel concentrating photovoltaic
combined system. Energy Convers Manage 2013;67:18696.
[13] Ammar MB, Chaabene M, Chtourou Z. Articial neural network based control
for PV/T panel to track optimum thermal and electrical power. Energy Convers
Manage 2013;65:37280.
[14] Gang P, Huide F, Jie J, Tin-tai C, Tao Z. Annual analysis of heat pipe PV/T system
for domestic hot water and electricity production. Energy Convers Manage
2012;56:821.
[15] Calise F, Dentice dAccadia M, Vanoli L. Design and dynamic simulation of a
novel solar trigeneration system based on hybrid photovoltaic/thermal
collector (PVT). Energy Convers Manage 2012;60:21425.
[16] Zhao J, Song Y, Lam WH, Liu W, Liu Y, Zang Y, et al. Solar radiation transfer and
performance analysis of an optimum photovoltaic/thermal system. Energy
Convers Manage 2011;52:134353.
[17] Chemisana D, Ibanez M, Rosell JI. Characterization of a photovoltaicthermal
module for Fresnel linear concentrator. Energy Convers Manage
2011;52:323440.
[18] Li M, Li GL, Ji X, Yin F, Xu L. The performance analysis of the trough
concentrating solar photovoltaic/thermal system. Energy Convers Manage
2011;52:237883.

651

[19] He W, Zhang Y, Ji J. Comparative experiment study on photovoltaic and

thermal solar system under natural circulation of water. Appl Therm Eng
2011;31:336976.
[20] Zhang X, Zhao X, Smith S, Xu J, Yu X. Review of R&D progress and practical
application of the solar photovoltaic/thermal (PV/T) technologies. Renew
Sustain Energy Rev 2012;16:599617.
[21] Radziemska E. Performance analysis of a photovoltaicthermal integrated
system. Int J Photoenergy 2009:16.
[22] Hottel HC. Whillier A. Evaluation of at-plate solar collector performance, In:
The Conference on the Use of Solar Energy: The Scientic Basis
1958;2(1):74104.
[23] Dufe JA, Beckman WA. Solar engineering of thermal processes. 2nd ed. New
York: John Willey and Sons, Inc.; 1991.
[24] Zondag HA, Vries DW, Van Steenhoven AA, Zolingen RJC. The thermal and
electrical yield of a PVthermal collector. Sol Energy 2002;72(2):11328.
[25] Watmuff JH, Charter WWS, Proctor D. Solar and wind induced external
coefcients for solar collector. Comples 1977;2:56.
[26] Eicker U. Solar technologies for buildings. Chichester: John Wiley and Sons;
2003.
[27] Vokas G, Christandonis N, Skittides F. Hybrid photovoltaicthermal system for
domestic heating and cooling a theoretical approach. Sol Energy
2006;80(5):60715.
[28] Tiwari A, Sodha MS. Performance evaluation of hybrid pv/thermal water/air
heating system: a parametric study. Renew Energy 2006;31(15):246074.
[29] Huang BJ, Lin TH, Hung WC, Sun FS. Performance evaluation of solar
photovoltaic/thermal systems. Sol Energy 2001;70(5):4438.
[30] Chow TT, Ji J, He W. Photovoltaic thermal collector system for domestic
application. In: Proceedings of solar word congress ISEC2005-76128 (ISES
2005).
[31] Chow TT, Ji J, He W. Photovoltaicthermal collector system for domestic
application. J Sol Energy Eng 2007;129(2):2059.
[32] He W, Chow TT, Ji J, Lu JP, Pei G, Chan LS. Hybrid photovoltaic and thermal
solar-collector designed for natural circulation of water. Appl Energy
2006;83(3):199210.
[33] Ji J, Lu JP, Chow TT, He W, Pei G. A sensitivity study of a hybrid photovoltaic/
thermal water-heating system with natural circulation. Appl Energy
2007;84(2):22237.