Photovoltaic Thermal (PV/T) System:

Effect of Active Cooling

TEO HAN GUAN

(B.Sc Eng. (NCKU))

A THESIS SUBMITTED

FOR THE DEGREE OF MASTER OF ENGINEERING

DEPARTMENT OF MECHANICAL ENGINEERING

NATIONAL UNIVERSITY OF SINGAPORE

2010
Acknowledgements

I am deeply grateful to my supervisors, Assistant Professor Lee Poh Seng and

Associate Professor Hawlader M.N.A, for giving me the guidance, insight,

encouragement, and independence to pursue my research. Their advice, contributions

and invaluable comments inspired me.

I would also like to thank Mr. Yeo Khee Ho, Mr. Chew Yew Lin, Mr. Anwar

Sadat and Mrs. Roslina Bte Abdullah who helped me with the setting up of my

experimental rig. Also, I would like to express my gratitude to my colleagues, Yan Lin,

Wai Soong, Kim Seng, Jayaprakash, Hwang Sheng, Yong Jiun, Sivanand, Karthik B,

Karthik S, Aung Myat, and Satyanarayana for their kind help and valuable advice.

I would like to extend my deepest gratitude to my parents, sisters and brother for

their encouragement and support for the entire duration of this project. Besides, I am

also grateful to my friends, especially Miss Lim Sze Huey, who supported and

encouraged me to come to Singapore for further study.

Lastly, I offer my regards and blessing to all of those who love, care and

supported me during the completion of the project.

i
 Table of Contents

Acknowledgements i

Table of Contents ii

Summary v

List of Tables vii

List of Figures vii

Nomenclature xii

Chapter 1 Introduction 1

1.1 Energy today 1

1.2 Solar Energy 3

1.2.1 Fundamental 3

1.2.2 Solar Thermal Collector 3

1.2.3 Solar Photovoltaic 4

1.2.4 Photovoltaic Thermal (PV/T) System 5

1.3 Objectives 5

1.4 Scope 6

Chapter 2 Literature Review 7

2.1 Water cooled PV/T 7

2.2 Air cooled PV/T 15

Chapter 3 Design of Manifold 26

3.1 Simulation of Different Configurations 26

3.2 Manifold Design of Experiment 29

ii
Chapter 4 Experimental Set-up 32
4.1 Description of the PV/T system 32

4.2 Experimental Components 35

4.2.1 Solar Cells 35

4.2.2 Maximum Power Point Tracker 37

4.2.3 Battery Bank 39

4.2.4 Active Cooling Device-Dc Blower and AC Blower 40

4.2.5 Solar Lamp 42

4.3 Experimental Measurements 42

4.3.1 Data Logger and 20 Channels Multiplexer 42

4.3.2 Pyranometer 43

4.3.3 T-type Thermocouple 44

4.3.3.1 Ambient Temperature 44

4.3.3.2 Temperature Difference Across the PV Panel 46

4.3.3.3 Inlet and Outlet Air Temperature 47

4.3.4 Anemometer 47

4.3.5 Shunt Resistor 48

4.4 Experimental Procedures 49

Chapter 5 Mathematical Formulation 51

5.1 Description of the numerical simulation model 52

5.2 Assumptions of the numerical simulation model 52

5.3 The analysis of heat transfer on Photovoltaic cell 52

5.4 Meteorological data of Singapore 63

iii
Chapter 6 Results and Discussion 66

6.1 Thermal performance 67

6.2 Electrical performance 78

6.3 Comparison of experimental and simulated results 99

Chapter 7 Conclusion 104

Chapter 8 Recommendation 106

References 109

Appendices 117

Appendix A Manufacturer’s Specifications 117

Appendix B Calibration of T-type thermocouple 121

Appendix C Derivation of the result 126

Appendix D Process log of Simulation 127

iv
 Summary

This thesis discusses aspects of a photovoltaic/thermal (PV/T) system which has

been designed to produce both electricity and hot air concurrently. Experiments were

conducted under outdoor conditions to determine the influence of the temperature of

the PV cell on the PV conversion efficiency. At higher operating temperatures of the

PV module, the conversion efficiency of the module can be drastically reduced due to

the significant reduction in the open circuit voltage of the photovoltaic cell. For this

reason, the payback period of the PV system is extended and the lifespan of the PV

module may also be shortened. In order to resolve this problem, several different

cooling techniques can be utilised to more effectively dissipate the heat from PV

module. In this work, forced convective air cooling is utilised to reduce the operating

temperature of the PV module.

It was found that without active air cooling, the temperature of the PV module

was high and solar cells could only achieve a conversion efficiency of only 8 to 9%.

However, when the PV module was operated under active air cooling condition, the

temperature dropped significantly, leading to an increase in the efficiency of solar cells

to between 12 and 14%. The heat which was extracted from the PV module by the

cooling air can contribute to the overall energy output of the system. Hence, the overall

v
system efficiency is no longer only limited by PV conversion efficiency but also

include the thermal efficiency which ranges between 45 to 55 %.

A variable speed blower is studying the effect of flow rate on the electrical

efficiency of PV module. The results showed that the optimum flow rate of this system

is around 0.055 kg/s. The flow field analysis of a parallel array of ducts with

inlet/outlet manifold was simulated using the commercial computational fluid dynamic

(CFD) package – Fluent. The simulation results showed that with a properly designed

manifold, a uniform flow distribution can be obtained. Uniform flow field can evenly

dissipate the heat from the PV module and reduce the occurrence of hotspots. A

mathematical model has been developed to investigate the heat transfer performance of

the PV module under actual meteorological conditions. The absorptivities and

transmittivities of the cover glass, Ethylene-vinyl acetate (EVA), silicon cell are also

considered in the numerical simulation. The simulation results showed good agreement

with the experimental results.

vi
List of Tables

Table 5.1 Thermal Properties of the Material 62

Table 6.1 Thermal efficiency for different flow rate 76

List of Figures

Figure 1.1 World energy demand 1

Figure 2.1 Water PV/T collector 7

Figure 2.2 Water and air mixed-type PV/T collectors 8

Figure 2.3 Hybrid PV/T system schematic 11

Figure 2.4 Monthly changes of available energy gain by exergetic evaluation on

Electrical 12

Figure 2.5 Monthly changes of available energy gain by exergetic evaluation on

Thermal 12

Figure 2.6 (a) Cross-sectional view of unglazed PV/thermal air (i) with tedlar (Model I)
(ii) without tedlar (Model II). (b) Cross-sectional view of glazed
PV/T air (i) with tedlar (Model III), (ii) without tedlar (Model IV) 17

Figure 2.7 (a) Hourly variation of electrical efficiency for a, b, c, d type weather
conditions considering glass to glass PV module with duct 18

Figure 2.8 Daily average of electrical efficiency for a, b, c, d type weather conditions
considering glass to glass PV module with duct. 18

Figure 2.9 Schematics of the double pass photovoltaic thermal solar collector 19

Figure 2.10 Schematic representation of the reflector assembly for a collector unit 20

Figure 2.11 PV module electrical efficiency as function of its operating temperature

for the typical and the combined with diffuse reflector mode 25
vii
Figure 3.1 FLUENT results 27

Figure 3.2 FLUENT results (cont’d) 28

Figure 3.3 3D model of parallel array air duct. Red arrows show the direction of air
flow 29

Figure 3.4 Engineering sketch drawing 30

Figure 4.1 Photograph of the outdoor transient testing set up 32

Figure 4.2 Schematic diagram of the experimental set-up 34

Figure 4.3 Polycrystalline Silicon Photovoltaic Cell 35

Figure 4.4 Structure of Photovoltaic Panel 35

Figure 4.5 The working wavelength of different type of solar cells 37

Figure 4.6 IV Curve and the maximum power point 38

Figure 4.7 MPPT Solar Charger Controller 39

Figure 4.8 Deep Cycle Gel Battery 40

Figure 4.9 DC Blower 41

Figure 4.10 AC Blower 41

Figure 4.11 Solar lamp 42

Figure 4.12 Hewlett-Packard data logger 43

Figure 4.13 20-channel relay multiplexer 43

Figure 4.14 Eppley pyranometer 44

Figure 4.15 T-type Thermocouple 44

viii
Figure 4.16 T type thermocouple miniature connector 45

Figure 4.17 Location which put the thermo probe 45

Figure 4.18 The arrangement of T-type thermocouple 46

Figure 4.19 Inlet thermal Probe 47

Figure 4.20 Outlet Thermal Probe 47

Figure 4.21 Anemometer 48

Figure 4.22 Voltage of the PV panel can be measured directly by connecting to

datalogger. 49

Figure 5.1 Diagram of principal reflections, absorptions and transmissions for a silicon
PV cell imbedded in EVA 53

Figure 5.2 Friction Factor under different Air Flow velocity 59

Figure 5.3 Heat Transfer Coefficient under different Air flow velocity 60

Figure 5.4 Solar Irradiation of Simulation 64

Figure 5.5 Ambient temperature of Simulation 65

Figure 6.1 Irradiation and Average Panel Temperature for the whole day under cooling
condition (23 September 2009) 67

Figure 6.2 Irradiation and Average Panel Temperature for the whole day without
cooling condition (28 September 2009) 68

Figure 6.3 Module Temperature as a function of solar irradiation 69

Figure 6.4 Temperature profile at centre the duct PV module. 69

Figure 6.5 Temperature profile at side of duct. 70

Figure 6.6 Top view of velocity contour of manifold design 71

ix
Figure 6.7 Cross section view of velocity contour of manifold design 71

Figure 6.8 Top view of the pressure contour of manifold design 72

Figure 6.9 Temperature profile of the front glass of module 73

Figure 6.10 Temperature profile of inlet and outlet flow 74

Figure 6.11 Variation of temperature difference (To-Ti) with incident radiation for
flow rate 0.0389 kg/s and 0.0932 kg/s 75

Figure 6.12 Thermal efficiency as a function of (Ti-Ta)/G 76

Figure 6.13 Influence of flow rate on thermal efficiency 77

Figure 6.14 Electrical efficiency as a function of PV temperature at irradiation at

1000W/m2 78

Figure 6.15 Electrical efficiency as a function of PV temperature at irradiation at

250W/m2 79

Figure 6.16 A comparison between theoretical and experimental results 81

Figure 6.17 Influence of flow rate on electrical efficiency 82

Figure 6.18 Influence of temperature difference (To-Ti) on electrical efficiency for

different flow rate 84

Figure 6.19 PV electrical power output under different solar radiation 89

Figure 6.20 Solar radiation of the entire day and the corresponded PV current due to
the solar radiation (23 September 2009) 89

Figure 6.21 Solar irradiation and the PV Voltage for the entire day (23 September
2009) 90

Figure 6.22 Solar radiation and the PV Voltage for the entire day (8 June 2009) 91

x
Figure 6.23 Solar radiation of the entire day and the corresponded PV current due to
the solar radiation (8 June 2009) 91

Figure 6.24 Battery and blower voltage of partially discharged battery bank (23
September 2009) 92

Figure 6.25 Battery and blower voltage of fully charged battery bank (8 June 2009) 93

Figure 6.26 PV current generated by module in case: (a) partially discharged battery
and (b) fully charged battery 94

Figure 6.27 Electrical Efficiency of fully charged and partially discharged at the
similar meteorological condition 95

Figure 6.28 Input solar radiation and thermal and electrical energy production over
five days 96

Figure 6.29 Electrical and thermal energy and the total energy gain over the five days
96
Figure 6.30 A comparison of thermal and electrical efficiency over 5 days. 97

Figure 6.31 A comparison of simulation and experiment in the temperature profile of

the back of PV module 100

Figure 6.32 A comparison of simulation and experiment in the temperature profile of

the front of PV module 100

Figure 6.33 Temperature contour of the PV cell at 1:30 pm (highest solar radiation on
that day) 101

Figure 6.34 Temperature gradient of the PV module at 1:30 pm 102

Figure 8.1 Spectral absorption coefficient of pure water (solid line) and of pure sea
water (dotted line) as a function of wavelength. 107

Figure 8.2: Transparent water passage in front of the PV panel to pre-filter the solar
irradiation before it strikes the solar cell. 108

xi
Nomenclature

A area of the PV module m2

Axs flow cross-sectional area m2

c light speed m/s

cp specific heat capacity under constant pressure J/kg·k

D distance from sun to the earth m

Ec net energy absorbed by the cell W/m2

Ece electrical energy produced by photovoltaic cell W/m2

Ect thermal energy released by photovoltaic cell W/m2

ET rate of solar energy absorbed by Tedlar W/m2

F Radiant Flux Density W/m2

FF Fill factor --

G solar irradiation W/m2

Ho Solar Constant W/m2

Hsun radiation intensity W/m2

h Planck’s constant J·s

hc convection coefficient of air W/m2·℃

hg convective heat transfer coefficient of the glass W/m2·℃

I current A

IL light generated current A

I0 dark saturation current of diode A

IMP maximum current A

Isc short circuit current A

K Boltzmann’s constant J/K

k thermal conductivity W/K·m

xii
m mass flow rate kg/s

Nu Nusselt number --

n ideality factor --

P wetted perimeter m

p cell packing factor --

Pel electrical power ouput W

Pr Prandtl number --

q electron charge C

qc the heat which convected away by the air flow W/m2·℃

Re Reynolds number --

Rsun radius of Sun m

T temperature ℃

Ta ambient temperature ℃

(Ta-6℃) sky temperature ℃

Tb temperature of backsheet ℃

Tc cell temperature ℃

Tg glass temperature ℃

Ti inlet temperature of the air flow ℃

To outlet temperature of air flow ℃

V voltage V

VMP maximum voltage V

Voc open circuit voltage V

v wind speed m/s

um mean fluid velocity m/s

λ wavelength of incident ray μm

xiii
Greek Letters

σ Stefan-Boltzman constant Wm-2K-4

ψB radiated energy Wm-2

αc cell absorptivity --

τg fraction transmitted through the front glass --

ηo nominal electrical efficiency under standard condition --

ηe cell electrical efficiency --

β temperature coefficient of silicon cell C-1

αT absorptivity of the Tedlar --

εg emittance of the glass --

αg absorptivity of glass --

θ module inclination to the horizontal --

v kinematic viscosity m2s-1

μ dynamic viscosity kgm-1s-1

α thermal diffusivity Jm-3k-1

ηtotal total efficiency --

ηth thermal efficiency --

Subscripts and superscripts

a ambient

b backsheet

c cell

e electrical

g glass

h hydraulic

xiv
loss losses

m mean

MP maximum power

oc open circuit

pv Photovoltaic

s sun

sc short circuit

SH shunt

T Tedlar

th thermal

xv
 CHAPTER 1
INTRODUCTION

1.1 Energy Today

Energy is currently an important issue all over the world. The demand for fossil

fuel has grown steadily due to increased industrial activities in developing and

developed countries. It is estimated that the world energy demand will increase by

45% between 2006 and 2030, and the rate of increase will be 1.6% per year [1]. Fig

1.1 shows the estimated world primary energy demand from 1980 to 2030.

Figure 1.1 World energy demands [1]

In general, fossil fuels such as oil, natural gas and coal can be considered as

primary sources of energy, especially, oil is the dominant fuel of the world. The

1
increase of the energy demand may be met by utilizing fossil fuel resources but the

amount of greenhouse gas emissions in the atmosphere will reach a dangerous level.

The WG 3 – Mitigation of Climate Change [2] indicated that over the last three

decades, greenhouse gases emissions have increased by an average 1.6% per year with

carbon dioxide (CO2) emissions from the use of fossil fuels growing at a rate of 1.9%

per year. According to the fourth assessment report from 2007 Inter-governmental

Panel on Climate Change [3], the increases of sea level are consistent with global

warming. Furthermore, global average sea level rose at an average of 1.8 mm per year

over 1961 to 2003 and at an average of about 3.1 mm per year 1993 to 2003. The rise

of the sea level is attributed to the melting of snow and ice in the Arctic Sea due to the

global warming effect.

Renewable energies including solar energy, wind power, hydropower, biofuel,

geothermal energy are suggested to provide a solution to resolve the global warming

problem and alleviate the potential of energy crisis. The demand of fossil fuels will be

reduced when the renewable energies become popular in the energy market.

Furthermore, potential climate change will be mitigated when the renewable energies

replace fossil fuels in the future. Solar energy is one of the most promising energy

sources with solar radiation reaching the earth’s surface at a rate approximately 80,000

TW and this figure is more than 10,000 times the present consumption of energy in the

2
world.

1.2 Solar Energy

1.2.1 Fundamental

The surface temperature of the sun is 5778 K. The core temperatures of sun reach

over 15 million K and the energy of the sun comes from nuclear fusion reaction from

H to He that take place deep inside the sun’s core [4]. The sun can be considered as a

blackbody radiator at the surface temperature. According to the Planck’s radiation law

for blackbody, the solar constant is approximately to 1368W/m2. The solar constant is

defined as the incoming power that Sun would deposits per unit area that is directly

exposed to sunlight. To harness the great amount of solar energy, solar thermal

collectors and solar photovoltaic cells have been used to convert the solar energy into

heat and electricity for various applications.

1.2.2 Solar Thermal Collector

Solar thermal energy can be interpreted as direct conversion of the energy from

solar radiation to useful thermal energy. The heat is generated by the absorption of

sun’s ray through a dark coated material, called absorber. The absorber actually is a

system of pipes filled up with a heat transfer medium, and the medium flows to the

collector to collect the heat from sun’s ray and goes back to the hot water store. In

some systems, the heat exchanger is used to extract heat from the water-glycol mixture

3
that is circulated in a closed circuit; is called an indirect system. Other systems, in

which pure water is used as the heat transfer medium, are called drainback systems.

Flat plat collectors are utilised chiefly for domestic hot water heating for showering,

washing and some industrial applications. The efficiency of a solar collector drops

drastically at high temperatures due to heat losses from the large surface area of

collector. Evacuated tube collectors are always utilized for applications at high

temperature and they are very efficient since they contain several rows of glass tube

and the air in the glass also removed from it to reduce the heat loss through the

convection effect. Therefore, the collector can operate at high efficiency and high

temperature.

1.2.3 Solar Photovoltaic cell

The Photovoltaic cell is made of semiconductor materials and used to convert

sunlight into direct-current electricity. When light with wavelengths less than 1100nm

strikes a PV crystalline cell, electron hole pairs are created in the cell, the electric field

sends the electrons from p-type to n-type silicon and the holes from n-type to p-type.

Disruption of electrical neutrality occurs during the photovoltaic effect. An external

load is needed to restore the equilibrium. The external load will provide a current path

which allows electrons to flow from n-type to p-type silicon; electron recombines with

the hole when it reaches the p-type silicon. The photocurrent is generated when the

4
electron passes through the external load.

The electrical efficiency of the PV cell is significantly affected by the operating

temperature. The electrical efficiency of PV cell linearly decreases when the operating

temperature increases, which is an advantage of the PVT system.

1.2.4 Photovoltaic Thermal (PV/T) System

A photovoltaic/thermal hybrid system (or PVT system) is a combination of

photovoltaic and solar thermal system. The PVT system can produce both electricity

and heat simultaneously. The PVT system refers to a system that extracts heat from the

panel with using heat transfer fluid, usually water or air and sometimes both. There are

several reasons which motivate the development of the PV/T system. One of the main

reasons is that PV/T system can provide higher efficiency than individual PV and

thermal collector system. With increased the efficiency, the payback period of the

system can also be shortened.

1.3 Objectives

The objectives of this study are

1. To design a manifold to ensure uniform flow distribution.

2. To investigate how active air cooling affects photovoltaic module

performance.

5
 3. To find the optimum flow rate for the PV/T system under operation.

1.4 Scope

Chapter 1 gives an introduction and brief discussion of the importance and

potential of solar energy and some solar technologies. Chapter 2 is a literature review

of water and air cooled PV/T and includes a summary of the state of the art of the

water and air cooled PV/T. The characteristics of the PV/T will also be investigated in

detail. Since the PV/T-air cooled system is discussed in this project, the detailed

description of the manifold design, which used in current experiment is provided in

Chapter 3. In Chapter 4, the components and functions of the system are described

clearly. Chapter 5 presents the mathematical formulation of the heat transfer on

Photovoltaic cell. The results of experiment and simulation are discussed in Chapter 6.

The experimental data is categorized into 3 parts, thermal performance, electrical

performance and the comparison of experimental and simulation result. Chapter 7

provides conclusion to the entire study and discusses the overall performance of the

experiment. Chapter 8 provides some ideas, which may significantly improve the

overall performance of the PV/T system.

6
 CHAPTER 2
LITERATURE REVIEW

Nowadays, for the PV/T system applications, production of the electricity

becomes more important. Therefore, it is necessary to keep the operating temperature

of the PV module as low as possible to ensure that its conversion efficiency is

maintained within an acceptable range. In these couple of years, PV/T-air and

PV/T-water systems have been widely investigated and different kinds of

configurations developed to test the overall performance of the combined system.

Numerical simulation of PV/T systems has also provided more detailed information on

the performance of the system.

2.1 Water cooled PV/T

Figure 2.1 Water PV/T collector [15].

7
 Figure 2.2 Water and air mixed-type PV/T collectors [5].

Figure 2.1 shows the common configurations of current PV/T systems in use

today nowadays. Water and air are the most common media utilised rather than

refrigerants since the overall cost of the entire system increases due to the capital cost

and maintenance cost for the refrigerant loop. Generally, water is the most effective

fluid to collect the heat from the PV panel and absorber due to its high heat capacity

and thermal conductivity.

Basically, water type PV/T can be categorized according to the water flow pattern

as shown in Fig 3.2. The parameters involved in the design are the sheet, tube, free

flow, channel and absorber types [6]. Numerical analysis is more preferable in

investigating the preliminary studies since it can provide an optimum model before

fabricating the prototype. The first mathematical model of PV/T collector was

8
published by Florschuetz [7]. He modified the Hottel-Willier [8] analytical model for

flat plat thermal collector in order to apply the equations to PV/T collectors. Some

parameters (such as heat removal factor and collector efficiency factor) of the

Hottel-Willier model are still available to be utilised in the PV/T collectors.

A dynamical model and three steady state models have been investigated by

Zondag [9]. He also carried out a prototype experiment to validate the simulated result

generated by his model. All models show good agreement with the experiment within

5% accuracy. Sandnes and Reskstad [10] have developed a polymer solar collector

which combines with crystalline silicon PV cell in a hybrid generating unit. This model

was developed by modifying the Hottel and Willier model for flat plate thermal

collector. Their experiments show that attaching PV cells onto an absorbing surface

reduces the solar energy absorbed by about 10%. This is because that the absorptivity

of PV cell is lower compared to the black absorber.

Zakherchenko [11] showed the importance of having good thermal contact

between the solar cells and thermal absorber. Their study indicates that some

commercial PVT modules should not be used directly in PVT system. Huang et al. [12]

investigated the performance between the integrated photovoltaic and thermal solar

system IPVTS and conventional solar water heater. A corrugated polycarbonate panel

was used to make the solar PV/T collector and the characteristic daily thermal

9
efficiency and primary-energy saving of the collector is 38% and 60%. A hybrid

photovoltaic/thermal water-heating system with natural circulation was constructed by

Jie Ji et al. [13]. Their experiment results showed that the characteristic daily

primary-energy saving could reach up to 65% for this system. The simulated result also

showed that the higher the packing factor and glazing transmissivity, the better is the

overall system performance. Wei He et al [14] indicated that a good thermal-contact

between the absorber and the PV module can significantly increase both thermal and

electrical efficiency of the system. Fin performance of the heat exchanger is also a

crucial factor to boost the overall efficiency.

Chow [15] presented an explicit dynamic model for operation of PV/T collector

since it is not suitable to use a steady state model to predict the working temperatures

of the PV module and the heat removal fluid was also under fluctuating irradiance or

intermittent fluid flow. For that reason, the transient case can more accurately predict

the outcome of experiments. That model was developed based on the control-volume

finite difference approach. The proposed model can provide a detailed analysis of the

transient energy flow through different types of collector components and the

instantaneous energy output can also be monitored.

A simulation of PV/T system was carried out by using the well known TRNSYS

program by Kalogirou [16]. They used the typical meteorological data of Cyprus and

10
the optimized water flow rate via simulation. The system consists of a series of PV

panels, a battery bank, a hot water storage cylinder, a pump, a differential thermostat

and an inverter (Fig 2.3).

Figure 2.3 Hybrid PV/T system schematic diagram [19]

Fujiwa and Tani [17] used exergy analysis to evaluate the experimental

performance of a designed PV/T system since exergy can be used to qualitatively

compare the thermal and electrical energy based on the same standard.

11
 Figure 2.4 Monthly changes of available energy gain by exergetic
evaluation on electrical [17]

Figure 2.4 shows that the coverless PV/T collector produces the highest electrical

exergy and Figure 2.5 shows that thermal exergy of the coverless PV/T was the lowest

amongst the system considered. The latter may be due to heat losses from the top of

the device.

Figure 2.5 Monthly changes of available energy gain by exergetic

evaluation on thermal [17]

12
 The thermal exergy with monthly changes is presented in Fig 2.5.Flow rate affects

the performance of PV/T system since the increase of water velocity in the tube will

result in increasing the heat transfer coefficient. This will help enhance the cooling on

the PV panel or collector.

Bergene and Lovvik [18] investigated the relation between the geometric

parameter W/D and the performance of the PV/T system. They reported thermal

efficiency increases by a factor of 0.1 and the flow fate increases from 0.001 to

0.0075kg/s. Chow [15] also found that when flow rate in the tube increases from 0.002

to 0.016 kg/s, the electrical and thermal efficiencies also increase. Garg and Agarwal

[19] utilised the finite difference method to investigate PV/T system with different

solar cell areas and flow rate. The system comprised of a storage tank, pump,

differential controller and PV modules. The optimum flow rate of this experiment was

0.03kg/s, for maximum thermal efficiency. It was shown that the electrical efficiency

decreased at this flow rate and was minimum when the insolation was maximum (as

the temperature of absorber is maximum).

Nishikawa et al. [20] utilised the refrigerant R22 as the liquid of PV/T collector

which function as the evaporator of a heat pump. A high COP was observed when the

system is efficiently cooled. Ito et al [21] showed that the COP of heat pump is very

low when it is under low irradiance. This is because the flat plat collector is not

13
optimised to extract the energy from the surrounding. In order to solve this problem, a

3.24 m2 multiple-fin evaporator was placed in parallel with a 2.45m2 PV/T absorber

and this resulted in an increase of COP from 2 to 3 under low irradiance conditions.

However, the COP of the system can attain a value of 6 when it is under high

irradiance.

Zondag et al [22] and Jong [23] have conducted a series of comparison between

different types of PV/T design and different types of thermal systems. Those

experiments generally investigated the covered and uncovered PV/T and thermal

system with and without heat pump. The studies indicated that an uncovered PV/T

shows improved efficiency for the case which the PV/T is utilised for low-temperature

ground storage integrated with a heat pump. The high thermal efficiency of the system

is because that the inflow air is always kept in low temperature. However, the net

electrical efficiency of the system turns into negative because of the energy

consumption of the heat pumps.

Currently, PV/T systems are always installed for residential use. In order to

investigate the actual condition of the residential building, the PV/T systems were

installed on the roof top of a residential building. Ji et al [24] installed a 40m2 PV/T

collector on a facade of the residential building in Hong Kong in order to investigate

the difference between the thin film and crystalline silicon PV cell. Under the same

14
meteorological condition, it was found that the thermal efficiency of the thin film is

48% which that of the crystalline silicon is 43%. They also proposed that the system

can also be utilized for pre-heating of hot water for residents in that building.

Furthermore, the systems can also provide cooling for the building with absorption of

heat by the wall of building reduced during the operation of PVT system. It was

concluded that the hybrid system has potential to be widely advocated in a sub-tropical

city such as Hong Kong.

Elswijk et al [25] also claimed that PVT collector arrays installed on multi-family

buildings could save about 38% in area. This is very vital due to the availability of the

roof top space per house. The disadvantage of this system is that the shading angle of

PVT collector must be smaller than the conventional solar thermal collector because of

the shading effect.

2.2 Air cooled PV/T

The first PV/T air facility was built in 1973 at the University of Delaware. This

PV/T air facility was called as ‘Solar House’ and the air collectors were integrated in

the roof top and façade of the house. Besides, one fourth of the collectors were

embedded with CdS/Cu2S cell to generate electrical energy. After the pioneering work

of University of Delaware, some laboratories such as the MIT Lincoln laboratory,

15
Sandia laboratory and Brown University also started developing the PVT air collectors.

The performance of PVT air collectors fabricated by ARCO and Spectrolab [26] was

insufficient but this first generation technology has become a motivation to boost the

development of second generation technology. The effect of thermal gradient on

electrical efficiency of PV panel was investigated by the Sandia [27].

In 1994, the French Company Cythelia [28] developed a PV-air collector, called

the Capthel collector. An unglazed PVT collector with air and liquid heat extraction

was developed and commercialised in Israel [29]. A new type PVT-air collector was

developed by the German Company Grammer Solar and the Danish company Aidt

Miljo [30,31]. This type of PVT air collector is only covered with a small PV cell

which used to drive the fan. The function of this system was utilised for

dehumidification purposes in vacation cottages.

Both experimental and numerical simulation were implemented by Tiwari [32] to

evaluate the overall performance of PV-T air collector. In this study, different kind of

configurations of PVT air collector (like unglazed, glazed, with and without tedlar)

which shown in Fig 2.6 were used to investigate the electrical and thermal

performance. It was shown that the glazed PVT air collector without tedlar provides

the best performance.

16
Figure 2.6. (a) Cross-sectional view of unglazed PV/thermal air (i) with tedlar (Model
I), (ii) without tedlar (Model II). (b) Cross-sectional view of glazed
PV/thermal air (i) with tedlar (Model III), (ii) without tedlar (Model IV).

A PVT-air collector was investigated by Garg and Adhikari [33] using a computer

simulation model. It was concluded that the thermal efficiency of the absorber without

solar cell is higher than that when the absorber is covered with the solar cell. This is

because that some of the incidence irradiance is converted into electrical energy.

Dubey et al [34] reported the efficiency of different configurations of PVT-air collector

(Case A-Glass to glass PV module with duct, Case B-Glass to glass PV module

without duct, Case C-Glass to tedlar PV module with duct, Case D-Glass to tedlar PV

module without duct). It was indicated that case A can give the highest efficiency

among the all four cases. The annual average efficiency of case A and B is 10.41% and

9.75%, respectively. The daily average electrical efficiency of the four cases are

17
presented in Figs 2.7 and 2.8

Figure. 2.7. (a) Hourly variation of electrical efficiency for a, b, c, d type weather
conditions considering glass to glass PV module with duct.

Figure 2.8 Daily average of electrical efficiency for a, b, c, d type weather conditions
considering glass to glass PV module with duct.

Hegazy [35] did a comparative study of the performance on four types of PVT

solar air collectors. Their results showed that the air flow on both sides of the absorber

in a single pass demands the least fan power.

18
Figure 2.9 Schematic diagram of a double pass photovoltaic thermal solar collector

Sopian [36] developed a double pass PVT air collector (Fig 2.9) for solar drying

application. Solar cells were put between the glass cover and absorber plate. The air

first enters the channel created by the glass cover and photovoltaic panel and next it

enters the channel created between the photovoltaic panel and absorber. This

configuration can greatly reduce the heat loss and increase its thermal efficiency. The

thermal efficiency of this system can reach to 60%. Some simpler modifications were

utilised to enhance the thermal performance of the air duct and this will help enhance

heat extraction from the PVT air collector.

Prasad and Saini [37] reported that the heat transfer mechanism of the solar

collector can be enhanced by artificially increasing the roughness of absorber plate and

wall of the channel, leading to higher thermal efficiency. However, high roughness of

wall and absorber will induce a higher friction factor and therefore a higher pumping

19
power is needed. Han et al [38] and Gupta et al [39] showed that several types of ribs

in the air channel can provide better performance in heat extraction but it is also

accompanied by a significant increase in friction losses. Some modifications like using

the pins, matrices, porous materials and perforated plates were suggested to improve

the heat extraction in the air channel. However, most of them are not practical to

significantly enhance the overall system performance. Garg and Datta [40] suggested

several practical modifications to enhance the heat transfer in air duct. In the study,

both experiment and numerical simulation were undertaken and the agreement

between the theoretical predictions and experimental results has been satisfactorily.

Figure 2.10 Schematic representation of the reflector assembly for a collector unit.

Garg et al. [41] presented a study of a PVT air hybrid system, this system

20
comprised a plane booster and a flat plat collector mounted with photovoltaic cells

(Fig 2.10). It was concluded that the electrical efficiency of photovoltaic cell will

linearly decrease with increase of the absorber temperature. The results also indicated

that the minimum area of photovoltaic cell needed to operate a pump at a given flow

rate is a function of time. The plane boosters were utilised to reflect the extra incident

rays to the photovoltaic cell in order to increase the intensity of sunlight on the

photovoltaic module.

Optimization the absorber geometry for solar air heating collector has been

investigated by Pottler [42]. It was reported that the optimized distance between the

fins is about 5 to 10 mm. The thermal efficiency of the collector can attain to 77% with

optimized geometry. As the pressure drop increases drastically with decreasing fin

spacing, this factor should also be considered in the design.

Naphon [43] carried out a study on the performance and entropy generation of the

double pass solar air heater with longitudinal fins. The study showed that the thermal

efficiency increases with increasing flow rate as the heat transfer is proportional to the

mass flow rate. The number and height of fins will also increase the heat transfer rate

due to the increase of heat transfer area. Hence, the thermal efficiency is proportional

to the number and height of fins. However, the entropy generation was found to

decrease with increasing height of fins. This is because the outlet temperature increases

21
with increase of height of fins.

Tonui and Tripanagnostopoulos [44] also reported an improvement of heat

extraction achieved by modifying the channels of PV/T air system in low cost. Three

different configurations of air ducts (simple air channel, thin aluminum sheet and

rectangular fin) were investigated by experiment and numerical simulation. Some

parameters (channel length, channel depth and mass flow rate) were used to study the

effect on electrical and thermal efficiency. From the result of experiment and

simulation, a good agreement has been presented and air duct with fins were shown

more effective in enhancing the heat transfer from the wall of the channels to air flow.

Sopian et al [45] presented a steady state simulation of the single and double pass

combined photovoltaic thermal air collector. The simulations indicated that the double

pass photovoltaic thermal collector has superior performance during the operation. The

difference of thermal efficiency for single and double pass combined photovoltaic

thermal collector is about 10%. The air flow in the double pass combined thermal

collector can absorb more thermal energy than that in the single pass. Therefore, the

thermal efficiency of the double pass is higher than that of the single pass. Due to the

large amount of heat absorbed by the air flow, the temperature of the photovoltaic

module decreased significantly and this causes the electrical efficiency of the double

pass was higher than single pass as well.

22
 Garg and Adhikari [46] developed a simulation model to investigate the

performance of single glass and double glass hybrid photovoltaic thermal air heating

collector. The thermal performance of the double-glass configuration was found to be

better than single glass for a normal black paint absorber. This is because the extra

layer of glass can reduce the radiative losses from absorber to glass cover.

For a selective absorber, the thermal efficiency of single-glass is higher than that

of double-glass as the effective transmittance-absorptance product also decreases when

the sun ray passes through the double-glass. The parametric studies showed that the

overall system efficiency increases with the increase in cell density, collector length

and mass flow rate. However, the increase of duct depth will incur the decrease in

system efficiency.

Joshi et al [47] carried out an evaluation of a hybrid photovoltaic thermal system.

Two types of PV module (glass to tedlar and glass to glass) were utilized to investigate

the performance under the climate of New Delhi. The results showed that the overall

performance of hybrid thermal collector with PV module glass-to-glass is better than

glass-to-tedlar. Parametric studies also indicated that thermal efficiency decreases with

the increase of length of the duct. It is because the thermal energy which can be

extracted at the back of PV module decreases. The highest thermal efficiency obtained

from the experiment was 46.28%. Thermal efficiency also increases with air velocity.

23
However, as the air velocity exceeds a certain level, thermal efficiency remains at a

constant level. This could be explained as the time of contact of air with module

reduces and therefore decreases the heat removal from the back of PV module.

Tripanagnostopoulos et al [48] presented a hybrid PV/T experimental model to

investigate the temperature effect on PV electrical efficiency. A booster diffuse

reflector was also utilized to enhance the electrical and thermal performance of the

system. It was found that PV electrical efficiency decreases at the rate of 0.1%/℃.

However, with the diffuse reflector, the electrical efficiency decreased at the rate of

0.0957%/℃ and 0.0814%/℃ for concentration factor at 1.3 and 1.5 respectively. In

this study, a comparison between water cooled and air cooled PVT were presented.

The PV module with thermal insulation leads to high temperature and incurs an

electrical efficiency drop (ηel/insul=0.113), and water cooled PV and air cooled PV with

ηel/water=0.128, ηel/air=0.126, respectively.

24
 Figure 2.11 The PV module electrical efficiency as function of its operating
temperature for the typical PV and combined PV with diffuse reflector mode

Tripanagnostopoulos [49] also showed that the electrical efficiency of PV module

increases by 2% with using the diffuse reflector and without incurring significant

penalty in temperature rise. The decreasing rate of temperature effect in electrical

efficiency was also found to be 0.1%/℃.

25
 CHAPTER 3

DESIGN OF MANIFOLD

A parallel arrangement of air ducts underneath the PV panels is used to create the

passage for the air to pass through. Fins are incorporated in the duct to increase the

heat transfer rate from the PV panel to the moving fluid. Non-uniform air flow usually

causes in recirculation and therefore the hot air will be trapped in the channels. This

causes the panel temperature to be unevenly distributed. A hot spot will result in the

panel and the electrical efficiency decreased due to the uneven temperature distribution

of panel. A series of simulations with different configurations of air duct is done using

the computational fluid dynamics (CFD) software, FLUENT. The results of the

simulation are presented in the Figs 3.1 and 3.2.

3.1 Simulation of Different Configurations

From the results, it can be seen that the re-circulated flows are significant in the C

shaped and S shaped ducts. These flows will result in the fluid flowing back towards

the inlet instead of the outlet, and this reverse flow will be heated up. The flow will

cause the PV module which attached atop of manifold having an uneven temperature

profile.

26
 C-shaped duct C-shaped duct with vanes at inlet

Duct with 1 inlet and 2 outlets V-shaped

Figure 3.1. FLUENT results

27
 S-shaped S-shaped duct with a protruding vane

S-shaped duct with vanes at inlet C-shaped duct with a L-shaped

Figure 3.2 FLUENT results (cont’d)

28
 In the V shaped duct, the simulations indicate no reverse flow and most of the

fluid flows through the centre channels resulting in the uneven heating of the PV

modules. The latter will also seriously affect the overall performance of the PV

module.

3.2 Manifold Design of Experiment

Following the simulation results of the V shaped design, a 90° change in flow

direction has been applied in current design. This will alleviate the focusing of fluid in

the centre channels and recirculation flow can also be avoided in this design.

Figure 3.3 3D model of parallel array air duct. Red arrows show the direction of air
flow

29
Figure 3.4. Engineering sketch drawing

30
 The manifold design shown in Fig 4.6 has a uniform flow field in the simulation.

The simulation results are presented in Chapter 6. The engineering sketch drawings of

the manifold design are presented in the Fig 3.4. All the needed dimensions and sizes

are clearly shown in the sketches.

The entire air duct was made of galvanized steel. Galvanized steel is the steel

coated with zinc which protects the steel from corrosion. As the experimental set up

was always locked on the roof top, it was important that this material could prevent the

air duct from being corroded by daily exposure to the elements. The modules are

incorporated in the air duct and fixed with screws. The design of air duct is also

allowed for inserting the thermocouple in the centre and both sides of the ducts.

During the experiment, gaskets were used to seal the gap between panel and panel

in order to reduce the leakage of air. Isometric views of this configuration are

presented in the Appendix. The entire air duct is put on a stainless steel rack. The

height of this rack is around 1.5 meter, to avoid shading of the PV module during

operation by other experimental set-ups on the roof top.

Summarising, this design permits the flow to enter the ducts uniformly, hence

obviating the potential hotspot problem.

31
 CHAPTER 4
EXPERIMENTAL SET-UP

4.1 Description of the PV/T system

A test set up was designed to investigate the thermal and electrical performances

of the Photovoltaic thermal system. This system was built on the roof top of EA

building at the National University of Singapore. The photograph of the set-up is

shown in Fig 4.1. A schematic diagram of the complete experimental set-up is shown

in Fig 4.2

Figure 4.1 Photograph of the outdoor transient testing set up

32
 The current experiment is designed to investigate how the temperature affects the

electrical efficiency and power output during the operation. Four polycrystalline solar

panels were used in the experiment to generate the electricity. The electricity which

generated by the solar panels will be stored in four deep cycle gel batteries. A direct

current blower connected to the batteries, is used to extract surrounding air to cool the

panels. During the operation, a maximum power point tracker (MPPT) was used to

modulate the power output from solar panel to be the maximum value. Another

alternating current (AC) blower is also used in this experiment because it can function

as variable speed blower, so that the flow rate can be controlled by adjusting the knob

of the controller.

Solar irradiation was measured by the pyranometer, which was put at the same

level as the solar panels. In this experiment, the air speed was measured by the

anemometer and the temperature of air and PV module was obtained by using the

T-type thermocouple directly connected to the datalogger. The voltage and current of

the solar panels were directly recorded by the datalogger.

The experiments normally operated from 8:30 am to 5:00 pm. In the experiment,

PV current, PV voltage, temperature of modules, temperature of inlet and outlet, wind

speed and irradiation of sunlight were collected during the operation of system.

33
Figure 4.2 Schematic diagram of the experimental set-up

34
4.2 Experimental Components

4.2.1 Solar cells

Figure 4.3 Polycrystalline Silicon Photovoltaic Cell

Neste polycrystalline solar cells are used in this experiment. The photovoltaic

module consists of 36 cells, the open circuit voltage and short circuit current are 22.5

V and 3.47 A. At 1000 Watt/m2 and 25℃, the maximum power output of single

module can reach 56.7 Watt/m2.

Figure 4.4 Structure of Photovoltaic Panel

35
 The structure of crystalline silicon solar cells is presented in Fig 4.4. EVA is a

kind of copolymer of ethylene and vinyl acetate. The polymer encapsulant which used

in PV modules serves to provide the functions like structural support, electrical

insulation, physical isolation/protection and thermal conduction for the solar cell

circuit [50].

The backsheet of photovoltaic module normally is a kind of material, called

Tedlar. The function of Tedlar is to prevent the ingress of water of water vapour. It is a

kind of polymer material, called Polyvinyl fluoride. Tedlar will also provide the

functions like UV resistance, mechanical properties, strength and durability, resistance

of weathering and electrical insulation. All of these functions will help PV panel to

sustain at least 20 years and above. Part of the backsheet is normally made as a

laminated film composite and the most common structure is the trilayer structure of

Tedlar/Polyester/Tedlar, also called TPT. This kind of structure can enhance the

functions of abovementioned.

Fig 4.5 shows that the wavelength of polycrystalline solar cell working range is

between 350 nm to 1200 nm [51]. Besides the transmission issues, the reflection of the

front surface of PV panel should be low as well. A low iron glass is most usually used

in the PV industry because it is of low cost, strong, stable, highly transparent,

impervious to water and gases and the front contact glass also has

36
 Figure 4.5 The working wavelength of different type of solar cells

self-cleaning properties after raining. The specification of the Neste polycrystalline

silicon solar panel used in this experiment is shown in the Appendix A.

4.2.2 Maximum Power Point Tracker (MPPT)

A maximum power point tracker is utilised to maximize the power output of PV

module and it is also a high efficiency DC to DC converter. Using a conventional

charge controller to charge a discharged battery, it connects the PV modules to the

battery directly, forcing the PV module to work at the battery voltage. Generally

speaking, this voltage is always not the ideal voltage for the maximum power output of

PV module. However, the Maximum Power Point tracker (MPPT) is not simply the

bridge between the module and battery.

The MPPT controller is able to calculate the voltage at which modules can

37
produce the maximum power output and the working voltage of PV module is at

maximum power output voltage rather than battery voltage. If the whole system wiring

is assumed to be 100% efficient, the battery charge current would be VMODULE /VBATTERY

x IMODULE. The current which stored to the battery with using the MPPT controller rather

than conventional converter will increase significantly. Fig 4.6, shows that there is

always a single operating point which will produce the maximum power output of the

PV cell. This point is called the maximum power point.

Figure 4.6 IV Curve and the maximum power point

The maximum power point tracker is used to seek this point in order to maximize

the power output of the panels under the different irradiation. The power from the solar

panel passes through the Maximum Power Tracker (MPPT), which modulates to the

best level that the module can produce and converts it to get maximum current from

the deep cycle battery. Fig 4.7 indicates the MPPT being used in this experiment.

38
 Figure 4.7 MPPT Solar Charger Controller

4.2.3 Battery Bank

Deep cycle gel batteries are very common in PV systems. It is designed to

produce a consistent voltage when the battery discharges. Fig 4.8 shows the battery

that was used in the experiment. During the discharging, the deep cycle battery can

discharge down to around 20% of its charge capacity without deteriorating its

performance. However, this kind of discharging cannot be applied to other types of

batteries which are not designed for “deep cycle discharge”. As it will deteriorate the

performance of the battery and also reduce the lifespan of the battery. In the

experiment, 4 deep cycle gel batteries are connected in both series and parallel

arrangements. Two 36 W solar lamp and a 24 W blower are utilised to discharge the

battery to ensure that the battery is not fully charged. This is because once the batteries

are fully charged, PV modules cannot properly produce the current from irradiation.

39
Therefore, battery voltage must always be monitored in the experiment. Solar Lamps

are used to discharge the deep cycle gel batteries during the night. The specification of

deep cycle gel battery is shown in Appendix A.

Figure 4.8 Deep Cycle Gel Battery

4.2.4 Active Cooling Device-DC Blower and AC Blower

In this experiment, a direct current (DC) blower and an alternating current (AC)

blower are used to cool the PV modules. The power of the DC blower is supplied by

the battery bank so that the rating of blower may be matched with the battery and PV

modules. Besides, the size, weight and pressure drop must also be taken into account

in choosing type of blower. A Sanyo Denki fan (Fig 4.9) which met the criteria is

utilized in this experiment. The flow rate of this blower is 8.2m3/min, the relevant

information is presented in Appendix A.

40
 Figure 4.9 DC Blower

Figure 4.10 AC Blower

In order to investigate the effect of different flow rates of air passing through the duct,

a variable speed AC fan (Fig 4.10) is needed in this experiment. The flow rate of this

fan is between 2.09m3/min and 7.11 m3/min.

41
4.2.5 Solar Lamp

A solar lamp is used to discharge the battery during the night. Basically, the

lamps are used in street and pedestrian lighting and this type of lamp is known as

Energy Saving Compact Fluorescent Lamps (CFL), 4 pin, Single U with lamp wattage

of 36 W and rated light output of 2900 ± 5% lumens from Trilux-Lenze. A water tight

canopy made of injection molded plastic is used to enclose the lamp to ensure that

water will not permeate into the light. The ballast is only customized with 24 V lamp

bulbs. The relevant information of solar lamp is attached in the Appendix A.

Figure 4.11 Solar lamp

4.3 Experimental Measurements

4.3.1 Data Logger and 20 channels multiplexer

A Hewlett-Packard data logger was used to record the readings at 1 minute

interval. There are 20 channels for each multiplexer. In this experiment, first 15

42
channels are used to measure the temperature of PV modules, inlet temperature, outlet

temperature and ambient temperature. The other 5 channels are used to capture the

data in DC voltage mode. There are PV voltage, shunt resistor voltage, Battery voltage,

Blower voltage and the Pyranometer voltage. Fig 4.12 and 4.13 are the data logger and

multiplexer which being used in the experiment.

Figure 4.12 Hewlett-Packard data logger

Figure 4.13 20-channel relay multiplexer

4.3.2 Pyranometer

Global radiation was measured with an Eppley pyranometer (Fig 4.14) (measures

global radiation). This pyranometer is a World Meteorological Organization First Class

Radiometer and it is designed for the measurement of sun and sky radiation. The

43
hemispheres of the pyranometer are made of clear WG295 glass. Hence, this

instrument can measure the radiation in the spectral range 285 to 2800 nm. The

response time of this instrument is 1s, therefore, the 1 minute interval which is used in

our experiment is perfectly fine to get a stable value from the pyranometer. The

instrument measures solar intensity in the range 0 to 2800 Watt/m2.

Figure 4.14 Eppley pyranometer

4.3.3 T-type Thermocouple

4.3.3.1 Ambient Temperature

Figure 4.15 T-type Thermocouple

44
 Ambient temperature was measured using a T-type thermocouple. T-type

thermocouples are suitable for measurements in the range at -200 to 350 °C. T-type

thermocouples (Fig 4.16) comprise 2 wires, copper and costantan (copper-nickel alloy).

A plastic connector (Fig 4.16) was needed in measuring the ambient temperature.

Figure 4.16 T type thermocouple Figure 4.17 Location which put the thermo probe
miniature connector

The temperature probe was mounted beneath the duct (as shown on Fig 4.17),

which is shaded from direct sunlight and rain but allows the circulation of air. A master

thermometer was used to calibrate the thermometers in this experiment and the

temperature range is 20 to 80 °C. The tolerance of T-type thermocouple is 0.5°C. The

calibration curves are attached in Appendix C.

45
4.3.3.2 Temperature difference across the PV Panel

To measure the temperature of the PV module, T type thermocouples were

attached on the front and back of modules. There are 12 pieces of T-type

thermocouples attached to the panels. The locations of the thermocouples are shown in

Fig 4.18.

Figure 4.18 The arrangement of T-type thermocouple

Before doing the calibration of thermocouples, the constantan and copper wire of

the T-type thermocouples need to be spot welded. The two wires then have a common

joint, called the bead. During measurements, the bead must be well attached to the

surface for more accurate temperature readings.

46
4.3.3.3 Inlet and Outlet air temperature

Inlet and outlet temperature are used to calculate the thermal efficiency of the

system. The inlet and outlet temperature were measured with the T-type thermocouple

mounted in a probe inserted to the flexible hose.

Figure 4.19 Inlet thermo probe Figure 4.20 Outlet thermo probe

4.3.4 Anemometer

The wind speed was measured using an anemometer manufacture by Geneq Inc.

This equipment was used to measure the wind speed for hourly during the experiment.

A picture of the anemometer is given in Fig 4.21

47
 Figure 4.21 Anemometer

The volumetric flow of the system was also measured using an anemometer. As the

cross section area of the flexible hose is fixed, in measuring the air flow speed, the

flow rate can be obtained through the equation-Flow rate=air speed × cross section

area.

4.3.5 Shunt Resistor

To measure the current of the PV modules, a shunt resistor (Fig 4.22) with 0.006

ohm was used in the experiment. As the current produced by the PV modules already

exceeds the working range of data logger, a shunt resistor was placed in series with the

modules. The voltage across the shunt resistor was recorded by the datalogger and the

current output of panels can be calculated by using the Ohm’s law shown in Eq (4.1):

IPV=VSH ÷ RSH (4.1)

48
 Figure 4.22 Voltage of the PV panel can be measured directly by connecting to
datalogger.

The electric circuit diagram of this experiment is presented in the Appendix A.

4.4 Experimental Procedures

The performance of PVT system was monitored from Jun 2009 to November 2009.

The following is the entire experiment procedure:

(1) Switch on the mains in Thermal Process Lab 2 and to supply the power to Data

logger and AC Blower.

(2) Place the data logger inside the steel case and put the Pyranometer on the same

level with the PV modules.

(3) Set the channels of data logger into temperature and DC voltage and also set the

interval of time to capture the data.

(4) Set the speed of the blower to the flow rate which wants to be investigated.

(5) Check that the gaskets of the system are well pasted to ensure that the leakage is

very minor during the experiment.

49
(6) Check that all the thermocouples and the thermal probes are well attached.

(7) Switch on the mains of the MPPT controller

(8) Press the scan button on the data logger to start logging the data.

50
 CHAPTER 5
MATHEMATICAL FORMULATIONS

5.1 Description of the numerical simulation model

The analytical model of hybrid PV/T panel is based on the work of Raghuraman

and Hendrie [52]. This simulation can be simply described as the heat transfer of

photovoltaic modules under the solar irradiation. The cooling mechanism is utilised to

reduce the temperature of the PV modules. The front glass surfaces of the photovoltaic

modules are exposed to the surroundings and therefore radiation and convection need

to be considered in the heat transfer analysis of the module. There are several layers of

material in the PV panel (Fig 4.4). The Fourier conduction law can be implemented in

analyzing the conduction heat transfer in between these layers. The back of

photovoltaic panels is attached to the cooling duct. For that reason, forced convection

is the main mechanism of heat transfer at the back of modules. This is a transient

simulation and the solar irradiation and ambient temperature will be varied from time

to time. The solar irradiation and ambient temperature is based on the experimental

data obtained on 23 September 2009.

51
5.2 Assumptions

In order to simplify the simulation model, the following assumptions are made:

(1) Edge and back heat losses of the collector are neglected in the simulation studies.

(2) The heat transfer in the collector is envisioned as two dimensional heat transfer

process.

(3) Only the single cell is simulated during the process.

(4) Inter-reflections of insolation between the various surfaces are neglected.

(5) The leakage of air from the collector is negligible.

(6) The capacity effect of glass cover and enclosed air is also neglected.

(7) An average wind speed was used to estimate the convection coefficient of

collector.

(8) The data of solar irradiation and ambient temperature from the experiment were

used as the input of numerical simulation.

(9) The ohmic losses in the solar cell are negligible.

5.3 The analysis of heat transfer on Photovoltaic Cell

The net energy absorbed by the cell is:

Ec = pα cτ g G (t ) (5.1)

52
where G (t) is the solar irradiation incident on the glass cover, p is the cell packing

factor which defined as the ratio of area of solar cell to the area of blank absorber, αc is

cell absorptivity to sunlight, τg is the fraction transmitted through the front glass and

low iron glass was used in the experiment, τg =.0.95.

For the wavelength less than 1.1μm the absorption length is less than the

thickness of typical cell (ie: 260μm), hence the absorption process is completed before

the radiation reaches the rear surface. Absorptivity of the silicon solar cell can be

computed through Fig 5.1

αc=0.926×0.8+0.073×0.2

=0.7554

Figure 5.1 Diagram of principal reflections, absorptions and transmissions for a silicon
PV cell imbedded in EVA [53]

53
The PV module used in this experiment is of gridded metal type; EVA absorptivity,

0.073

Polycrystalline silicon PV modules are used in the experiment. According to the

Cox and Raghuraman [53] report, the insolation of wavelength above 1.1μm is

transmitted through the silicon cell without any absorption and this is absorbed by the

backsheet of PV module. The insolation absorbed by the solar cell can be converted

into electrical and thermal energy and the equations are shown respectively below,

Ece = η e pτ g I (t ) (5.2)

Ect = (1 − ηe / α c ) pα cτ g G (t ) (5.3)

where Ece is electrical energy produced by photovoltaic cell, Ect is thermal energy

released by photovoltaic cell, ηe is the cell electrical efficiency and this parameter is

functioned of the cell temperature.

ηe = ηo [1 − β (Tc − To )] (5.4)

Vm p I m p
ηo = (5.5)
GA

54
where ηo is the nominal electrical efficiency under standard condition, A is the area of

the PV module, G is the irradiation and it is defined as 1000W/m2 for standard

condition, Vmp is the PV voltage at maximum power point and Imp is the PV current at

maximum power point. All the relevant data can be obtained from the specification of

PV module which put in the Appendix., To is the temperature of standard condition,

25℃, Tc is the cell temperature, β is the temperature coefficient of silicon cell,

β=0.0045℃-1.

ET = τ g (1 − P )α T I (t ) (5.6)

where ET is the rate of solar energy absorbed by Tedlar (Backsheet) after transmission

from EVA, αT is the absorptivity of the Tedlar.

Energy conservation laws are applied into the components of the collector and the

equations are shown below:

(1 − η e / α c ) pα cτ g G (t ) + τ gα T G (t )(1 − p ) = Eloss + qc (5.7)

Eloss is the energy losses from the front glass to environment through the forced and

free convection and radiation.

55
E = hg [Tg − Ta (t )] + ε gσ Tg 4 − α gσ [Ta (t ) − 6]4 (5.8)

where Tg is the glass temperature, Ta is the ambient temperature, εg is the emittance of

the glass, αg is the absorptivity of glass and (Ta-6℃) is assumed to be the sky

temperature [54]. The solar collector was exposed to the ambient so that the heat loss

is transferred by the top glass cover to the surrounding due to the combination of free

and forced convection. Free convection is due to the air near the collector surface,

which gets heated up producing the natural buoyancy force on the air. Forced

convection is caused by the wind. Therefore, hg is the convective heat transfer

coefficient of the glass to the environment and an empirical correlation from Stultz and

Wen [55] report is used,

hg = 1.247([Tg − Ta (t )]cos θ )1/ 3 + 2.658V (5.9)

where hg is the convection coefficient of surface of collector, θ is the module

inclination to the horizontal, V is the wind speed and assumed to be a constant speed,

V=0.4m/s.

56
qc = hc (Tb − Tave ) (5.10)

The energy balance of the air flow in the duct:

• dTair
m cp dx = qc (5.11)
dx

Where qc is the heat which convected away by the air flow in the channel, Tb is the

temperature of backsheet and hc is the convection coefficient of air in the channel. Tave

is the average temperature of inlet and outlet flow.

To investigate the heat transfer of an internal flow within a duct, the flow

condition (laminar or turbulent) is important to know and this information can be

obtained through the equation below:

um D
Re D = (5.12)
ν

where um is the mean fluid velocity over the duct cross section, v is the kinematics

viscosity of fluid and D is the hydraulic diameter. In a fully developed flow, to achieve

a turbulence, the Reynolds number must somewhere between ReD=2300 and 10000.

The smoothness of wall surface is also a factor to the affect the Reynolds number. For

the hydraulic diameter, it is defined as

57
 4 Axs
Dh = (5.13)
P

where Axs and P are the flow cross-sectional area and wetted perimeter.

The Nusselt number Nu is a dimensionless measure to determine the convective heat

transfer coefficient from the inside surface of a duct. It can be physically interpreted as

the dimensionless temperature gradient at surface.

hc D
NuD = (5.14)
k

where hc is the heat transfer coefficient for convection, k is the thermal conductivity of

the fluid, and D is the hydraulic diameter of duct. For fully developed turbulent flow,

the Nusselt number is much more complicated to determine. Therefore, empirical

correlation is always utilised to calculate the Nusselt number.

A correlation, which is widely utilised and is attributed to Petukhov [56], is valid for

0.5＜Pr＜2000 and 104＜ReD＜5×106

( f / 8) Re D Pr
NuD = (5.15)
1.07 + 12.7( f / 8)1/ 2 (Pr 2 / 3 − 1)

where Pr is the Prandtl number, which can be physically described as the ratio of the

58
momentum and thermal diffusivities, and shown below:

Cpμ
Pr = (5.16)
k

Where cp is specific heat capacity under constant pressure, μ is dynamic viscosity. f is

the friction factor, which can be obtained through checking the Moody diagram or

equation below:

f = (0.790 ln Re D − 1.64)−2 (5.17)

This correlation is valid for 3000≦ReD≦5×106

Figure 5.2 Friction Factor under different Air Flow velocity

59
 Friction factor is a function of Reynolds number, as expressed in Eq (5.17) and

the variation of friction factor with air flow velocity is shown in Fig 5.2. It shows that

the friction factor is observed to decrease with increasing air flow rate. There is a

significant drop in the range of 0 to 2 m/s.

From Equations 5.10 and 5.11, hc, convection coefficient of air flow in the duct can be

written as follows:

k ( f / 8)U m Pr
hc = (5.18)
ν [1.07 + 12.7( f / 8)1/ 2 (Pr 2 / 3 − 1)]

Figure 5.3 Heat Transfer Coefficient under different Air flow velocity

The heat transfer coefficient in the cooling duct which is a function of the air flow

velocity is plotted in Fig 5.3. With increasing air flow velocity, the heat transfer in the

cooling duct will be enhanced as well as can be observed in Fig 5.3.

60
The internal heat transfer mechanism of solar cell is dominated by Fourier conduction

law. The equation can be written as following equation:

∂T
ρC p = ∇(k ∇T ) (5.19)
∂t

A two dimensional simulation is discussed in this study and therefore the above

equation can be transformed to the following equation:

∂ 2T ∂ 2T 1 ∂T
+ = (5.20)
∂ 2 x ∂ 2 y α ∂t

Where α is the thermal diffusivity, α=k/ρcp, T is the temperature. The thermal

properties (thermal conductivity, density and specific heat capacity) of the layers inside

the photovoltaic module are tabulated in table 5.1

61
 Table 5.1 Thermal Properties of the Material

Thermal Specific Density Thermal

Conductivity Heat (Kg/m3) Diffusivity
Material
(W/m-ºK) Capacity
(KJ/kg K)
Tedlar
(Polyvinyl 0.14 1010 1450 9.56E-08
Fluoride)
EVA
(Ethylene-vinyl 0.3836 2220 1080 1.6E-07
acetate)
PET
(Polyethylene 0.24 1000 1455 1.65E-07
terephthalate)

Silicon
148 712 2330 8.92E-05
(Polycrystalline)

PV Glass 1 858 2500 4.66E-07

The thermal efficiency can be computed with the following equation:

•
m c p ∫ (To − Ti )dt
ηth = (5.21)
Ac ∫ G (t )dt

where m is the mass flow rate, cp is the specific heat capacity, To is the outlet

temperature of air flow, Ti is the inlet temperature of the air flow, Ac is the area of

collector.

The electrical efficiency of the PV module can be described as following equation:

62
ηe =
∫ VIdt
(5.22)
A∫ G (t ) dt

The total efficiency of the hybrid PV/T system is:

•
m c p ∫ (To − Ti )dt + ∫ VIdt
ηtotal = ηth + ηe == (5.23)
Ac ∫ G (t )dt

The thermal and electrical efficiencies are presented in Eqs (5.22) and (5.23). It

can be seen that the solar irradiation is a function of time and those parameters which

are affected by the solar irradiation, such as inlet and outlet temperatures, PV voltage

and PV current, are also functions of time. That is the reason to integrate the equation

with time.

5.4 Meteorological data of Simulation

In order to compare the temperature profile of simulation and experiment, the

same meteorological condition must be applied to the simulation programming. A

second order polynomial equation was used to curve the real meteorological data on 23

September 2009. The solar irradiation of the simulation is a function of time and the

equation is shown below:

63
G (t ) = −0.003t 2 + 0.6746t + 580.47 (5.24)

where t is the time and the unit of t is minute.

1100
1000
900
Irradiation (W/m2)

800
700
600
500
400
300
200 2
y = -0.0003x + 0.6746x + 580.47
100
2

7
47

:1

:4

:1

:4

:0

:3

:5

:2

:4

:1

:3

:0

:2

:5

:1
9:

10

10

11

11

12

12

12

13

13

14

14

15

15

15

16
Time

Figure 5.4 Solar irradiation of the simulation

The curve seems to perfectly fit the real data in the morning, and the deviation of the

curve started increasing after 1 pm. Another parameter which is used to approach the

real condition for the simulation is ambient temperature. Fig 5.5 is the curve of the

ambient temperature at that day. A six order polynomial equation was used to represent

the meteorological data on 23 September 2009 and the equation is:

64
 35 y = -8E-19x6 + 7E-15x5 - 2E-11x4 + 4E-08x3 - 3E-05x2 + 0.0109x + 31.271
Ambient Temperature (C)
34

33

32

31

30
2

7
47

:1

:4

:1

:4

:0

:3

:5

:2

:4

:1

:3

:0

:2

:5

:1
9:

10

10

11

11

12

12

12

13

13

14

14

15

15

15

16
Time

Figure 5.5 Ambient temperature of simulation

Ta(t) =-1×10-14t6 +2×10-11t5 - 1×10-8t4 +5×10-06t3 - 0.0009t2 +0.0561t +31.297 (5.25)

The ambient temperature of the simulation can only averagely curve the

meteorological condition on 23 September 2009. The fluctuation of the ambient

temperature could be attributed to the wind blowing unsteadily. However, the

polynomial equation is still acceptable to represent the ambient temperature in the

numerical simulation.

65
 CHAPTER 6
RESULTS AND DISCUSSION

The results of the experiments and simulation are discussed in this chapter. Good

agreement between the experiment and simulation results is found in this study. The

electrical and thermal performances of the system are also clearly presented in this

chapter and the comparison of different meteorological conditions will be provided in

order to investigate the function of the PV/T system. In this study, thermal and

electrical performance will be discussed respectively. In thermal aspect, those

parameters like temperature of Photovoltaic module at different location, inlet and

outlet temperature of air flow, ambient temperature, solar intensity, thermal efficiency,

thermal gain and flow rate are investigated thoroughly in order to figure out the

potential of combining the Photovoltaic module and thermal collector. Apart from

these, battery voltage, Photovoltaic voltage, Photovoltaic current, external load and

electrical efficiency are also needed to be elaborated. This is because that these

parameters can adequately indicate whether the cooling mechanism could improve the

electrical performance of system.

66
6.1 Thermal Performance

1100 50
49
1000
Irradiation (W/m )

48
2

TEmperature (C)
900 47
800 46
45
700 44
Irradiation
600 Average Temperature 43
42
500
41
400 40
9:47 10:30 11:30 12:30 13:30 14:30 15:31 16:21
Time

Figure 6.1 Irradiation and Average Panel Temperature for the whole day under cooling
condition (23 September 2009)

The cases with and without cooling are presented in Figs 6.1 and 6.2. The

maximum solar intensity of both days occurs at around 1:30 pm. The maximum solar

intensities on 23rd September and 28th September were about 1050W/m2 and

1200W/m2, respectively. For the case with cooling, the module temperature was

maintained at 48℃ at maximum solar intensity. However, from Fig 6.2, it can be seen

that the maximum temperature attained by module is 63℃. The temperature profile of

the PV module almost corresponds to the solar intensity and this may be observed

from the Figs 6.1 and 6.2 These two figures provide vital information in illustrating

that how effective of using the cooling mechanism to reduce the temperature of PV

module. These two figures also show that the maximum solar intensity always occurs

67
at solar noon. The local solar noon of Singapore is around 1:00 to 1:30 pm.

1200 68
Irradiation
1000 Average Temperature 63

AverageTemperature (C)
Irradiation (W/m )
2

800 58

600 53

400 48

200 43

0 38
9:46 10:14 10:41 11:50 12:02 12:49 12:55 13:18 13:43 14:25 15:15 15:48 16:42 16:57
Time

Figure 6.2 Irradiation and Average Panel Temperature for the whole day without
cooling condition (28 September 2009)

The Photovoltaic module temperature is linearly proportionate to the irradiation

and it is displayed in the Fig 6.3. Under the cooling mechanism, for every 100W/m2

increment of solar irradiation, the temperature of module increases 1.39℃. However, if

the PV module is not associated with the cooling mechanism, the increase of

temperature will be 1.8℃ for every 100W/m2.

68
 Without Cooling
65
With Cooling

Module Temperature (C)

60
y = 0.018x + 41.752
55

50

45
y = 0.0139x + 34.424
40
200 300 400 500 600 700 800 900 1000 1100 1200
Irradiation (W/m2)

Figure 6.3 Module Temperature as a function of solar irradiation

This also shows that the increase of temperature of PV module without the

cooling mechanism is higher than with the cooling mechanism under the same solar

irradiation. The variation of temperature between the cooling and non-cooling cases

can can be as high as 10℃. The high temperature will seriously affect the electrical

performance of the system and also degrade the Photovoltaic module.

54
T1
52 T2
T3
Temperature (C)

50
T4
48

46

44

42

40
9:47 10:30 11:30 12:30 13:30 14:30 15:30 16:30
Time

Figure 6.4 Temperature profile at centre the duct PV module.

69
 T7
48 T11
47 T12
Temperature (C) 46 T13
45
44
43
42
41
40
39
38
9:47 10:00 10:30 10:59 11:30 12:00 12:30 13:00 13:30 13:52 14:30 15:00 15:30 16:00 16:30

Time

Figure 6.5 Temperature profile at side of duct.

Figs 6.4 and 6.5 provided the temperature profile of back of PV module. The

shapes of the curves are shown consistently to each other. It was observed that the

temperature of the module has shown a variation in the sequence of the location of

thermocouple. The temperature of PV module which near to the inlet of air flow will

always be lower than the temperature of panel which near to the outlet. This could be

attributed to the air flow keep absorbing the heat from the panel and the temperature

difference (Tbacksheet-Tair) between the surface and air flow will become smaller.

Therefore, the sequence of temperature of PV module will be T4>T3>T2>T1.

From Figs 6.4 Fig 6.5, it is may be observed that the temperature of the panel in

the centre channel is higher than that of the side channel. This phenomenon can be

described as non-uniform flow distribution. The flow distribution simulation results are

displayed in Figs 6.6 and 6.7. These figures indicate that the non-uniform flow caused

70
the temperature of PV module in the centre channel increases due to the low flow rate

passing through.

Figure 6.6 Top view of velocity contour of manifold design

Figure 6.7 Cross section view of velocity contour of manifold design

71
The simulation results also showed that the pressure drop in this design is not very

significant and therefore the energy consumption which utilised to drive the blower

can be reduced. Figure 6.8 presents the pressure contour of the manifold design

Figure 6.8 Top view of the pressure contour of manifold design

The scale of Fig 6.8 shows that the pressure drop of this manifold design is less

than 55.3 Pascal. This information is useful for the proper selection of the blower.

Although the flow rate at centre channel is slightly different from the side channel, the

manifold duct design can still provide the Photovoltaic module with a well distributed

flow field.

72
 108(C)
54
109(C)
52 110(C)
50 114(C)
Temperature (C)

48
46
44
42
40
38
9:47 10:00 10:30 10:59 11:30 12:00 12:30 13:00 13:30 13:52 14:30 15:00 15:30 16:00 16:30
Time

Figure 6.9 Temperature profile of the front glass of module

Temperature profiles of the front glass of PV module are presented in Fig 6.9.

From the comparison of Figs 6.5 and 6.9, it is clearly observed that the temperature of

front glass is higher than at the back. This indicates that the convection heat transfer of

front glass is not constant at the back of the PV module. The convection heat transfer

of the front glass is dominated by the forced and free convection. Free convection in

this experiment is insignificant as the buoyancy force involved is very small; therefore,

the heat transfer is also insignificant. However, forced convection at the front glass is

dominated by the wind blowing over the PV module.

Hence, it is not comparable to the forced convection at the back of PV module. In

addition, the wind blow at the front glass of module is very unstable. Thus, the

temperature of the front glass of PV module is higher than the back of PV module.

However, the trend of the temperature profile can still be observed and it still

73
corresponds to the solar intensity even though under the disturbance of unstable heat

transfer mechanism.

44
Inlet (C)
42
Outlet (C)
Temperature (C)

40

38

36

34

32

30
9:47 10:30 11:30 12:30 13:30 14:30 15:30 16:30

Time

Figure 6.10 Temperature profile of inlet and outlet flow

Inlet and outlet temperature of the air flow are also investigated in this study, as

the heat gain and thermal efficiency can be computed by using the difference of inlet

and outlet flow temperature. Fig 6.10 indicates the temperature profile of inlet and

outlet flow over the entire day. The maximum temperature difference between the inlet

and outlet flow can be 8℃ and it happened at 1:30 pm. The minimum temperature

difference between the inlet and outlet flow occurred at 3:30 pm and just has 4℃ in

difference. This might be the reason that the solar irradiation is low at that moment and

the ambient temperature of the surrounding is still high and therefore the temperature

difference of inlet and outlet is limited by those reasons.

74
 Mass Flow= 0.0932 kg/s

Temperature Difference (C)

10
9 Mass Flow= 0.0389kg/s
8
7
6
5
4
3
2
1
0
200 300 400 500 600 700 800 900 1000
2
Irradiation (W/m )

Figure 6.11 Variation of temperature difference (To-Ti) with incident radiation for flow
rate 0.0389 kg/s and 0.0932 kg/s

Temperature differences of the inlet and outlet flow are presented in Fig 6.11. The

temperature difference of inlet and outlet flow will increase 0.55 ℃ for every

increment of 100W/m2 of solar irradiation when the flow rate is 0.0932kg/s. However,

when the flow rate at 0.0389 kg/s, the temperature difference of inlet and outlet flow

will increase 0.89℃ for every increment of 100W/m2 of solar irradiation. This can be

explained that the increase of flow rate will cause the temperature difference of inlet

and outlet flow decreases. As the flow rate is inversely proportional to the temperature

difference at a given heat gain.

75
 0.600

Mass Flow=0.0932kg/s
0.550 Mass Flow=0.0389kg/s
Mass Flow=0.1379kg/s

Thermal Efficiency
0.500

0.450

0.400

0.350
-0.006 -0.004 -0.002 0.000 0.002 0.004 0.006 0.008 0.010 0.012
(Ti-Ta)/G(K．m2W-1)

Figure 6.12 Thermal efficiency as a function of (Ti-Ta)/G

The thermal efficiencies of the experiments are presented in Fig 6.12. The thermal

efficiency of the PV/T air collector at different mass flow rate are tabulated in table 6.1

Table 6.1 Thermal efficiency for different flow rate

Mass Flow Rate (kg/s) Equation

0.0389 ηth = 0.4095-3.7491ΔT/G
0.0932 ηth = 0.5104-4.7574ΔT/G
0.1379 ηth = 0.5342-5.4084ΔT/G

The thermal efficiencies of the experiments are between 40 to 55%. However, Tonui

and Tripanagnostopoulos [57] reported, a PVT air collector with fins can provide the

thermal efficiency at 0.30-6.14ΔT/G. The flow rate which used at that experiment was

very low compared to current study. Therefore, the thermal efficiency of that study can

only attain to 30%. From the result of theoretical models, high values of air flow rate,

76
long PV/T system and small air duct depth, thermal efficiency can up to 55% [48]. For

the model of a finned double-pass photovoltaic-thermal solar air heater, Othman et al

[58] reported, the thermal efficiency of the system can attain to around 45% and 70%

at flow rate 0.027kg/s and 0.181kg/s, respectively. This may be attributed to the air

having sufficient time for good a heat transfer with the PV module when the air moves

from the top of the cell to the bottom. Othman et al [59] performed an experimental

analysis on PVT collector of double pass with flat plat, and achieved the thermal

efficiency of about 58% at 0.1kg/s of air flow rate.

60

50
Thermal Efficiency (%)

40

30
Irradiation=1000W/m2
20 Irradiation=900W/m2

10

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14

Flow rate (kg/s)

Figure 6.13 Influence of flow rate on thermal efficiency

Fig 6.13 showed that, by varying the flow rate, it can be seen that the thermal

efficiency will come to be a constant after the flow rate at around 0.05 kg/s. Sopian et

al [45] showed that the thermal efficiency will come to a constant after the flow rate at

77
0.042 kg/s. Hegazy [35] reported that the daily thermal efficiency of the system will

reach an asymptotic value when the flow rate reaches to 0.045 kg/s., Sopian et al [36]

found that for a solar drying system the thermal efficiency will be maintained at 58%

with the flow rate at 0.05 kg/s. The thermal efficiency of the PVT air system seems to

be maintained at fixed level when the flow rate at 0.04 to 0.05 kg/s, regardless of the

configuration of collector.

6.2 Electrical Performance

14 With Cooling

13 Without Cooling
Electrical Efficiency (%)

12

11

10

8
35 40 45 50 55 60 65 70

Temperature (C)

Figure 6.14 Electrical efficiency as a function of PV temperature at irradiation at

1000W/m2

78
 14
With Cooling
13 Without Cooling
Electrical Efficiency (%)
12

11

10

8
30 32 34 36 38 40 42 44 46

Temperature (C)

Figure 6.15 Electrical efficiency as a function of PV temperature at irradiation at

250W/m2

In Figs 6.14 and 6.15 show that the variation of electrical efficiency of the PV

module with the operating temperature. It can be seen that the electrical efficiency of

the PV module is significantly affected by the operating temperature under high

irradiation conditions. However, under low irradiation conditions as shown in Fig 6.15,

cooling mechanism is insignificant in affecting the electrical efficiency of PV module.

It is probably due to the low operating temperature of PV module under low irradiation.

Hence, the cooling mechanism cannot significantly affect the electrical efficiency of

PV module. High irradiation will incur high operating temperature of PV module, this

has been shown at Fig 6.14. Without applying the cooling mechanism on the PV

module, the temperature of PV module can attain to around 68℃ and electrical

efficiency of the PV module is around 8.6%. If the PV module cooled by air, the

79
electrical efficiency of PV module can be boosted to around 13% and the operating

temperature is only 36℃. From the result of experiment, it strongly proved that the

cooling mechanism can greatly help increase the electrical efficiency of PV module by

reducing the operating temperature. Furthermore, the high operating temperature can

also reduce the lifespan of PV module by degrading the material of module. Zondag

and Helden [60] showed that the temperature of PV module can reach 120℃ when the

flow fails in the collector. The thermal cycling tests also showed that the EVA inside

the PV module may have a risk in delamination when operating temperatures of PV

module over 135℃. Fig 6.14 also provides an indicative trend in the relation of

electrical efficiency and operating temperature. A linear equation obtained from the Fig

6.14:

η el = 0.1577 − 0.0009Tpanel (6.1)

The theoretical efficiency of PV module can be obtained from the Eq 5-4

From the theoretical deduction, the electrical efficiency of the module can be written

as the equation below:

ηel = 0.1664 − 0.0007Tpanel (6.2)

80
 15 Experimental Result
Theoretical Result
14 y = -0.07x + 16.64

Electrical Efficiency (%)

13

12

11
y = -0.09x + 15.77
10

8
30 35 40 45 50 55 60 65 70 75

Temperature (C)

Figure 6.16 Comparison between theoretical and experimental results

Based on experimental data as shown in Fig 6.16, showed that the theoretical

electrical efficiency is about 1 to 2% higher than experimental electrical efficiency.

This discrepancy can be attributed to the connection of module to module will incur

the electrical efficiency drop. Tonui et al [57] reported that the linear correlation

between the electrical efficiency and the module temperature which obtained from the

experiment is:

ηel = 0.147 − 0.0008Tpanel (6.3)

Tripanagnostopoulos [49] also presented another experimental result on correlation

between the electrical efficiency and module temperature and the linear equation is

given by:

81
η el = 0.166 − 0.0001Tpanel (6.4)

From the electrical efficiency correlation of the work of Tonui,

Tripanagnostopoulos and author, it can be concluded that the increase of temperature

of PV module can reduce the electrical efficiency. However, the variation of the

constant term of each equation may be attributed to the different model of PV module

and therefore the performances of the PV module are also different to each other. The

comparison of these three models also showed that under the same increment of

temperature the reduction of electrical efficiency of Tripanagnostopoulos’s experiment

is lower than other two cases.

13.00

12.50
Electrical Efficiency (%)

12.00

11.50
Irradiation=1000W/m2
11.00
Irradiation=900W/m2
10.50
Irradiation=500W/m2
10.00
Irradiation=700W/m2
9.50

9.00

8.50
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18

Flow Rate (kg/s)

Figure 6.17 Influence of flow rate on electrical efficiency

82
 This shows that the reduction of electrical efficiency of Tripanagnostopoulos’s

experiment will be lower than those for two other experiments under the same

irradiation. By varying the flow rate, the electrical efficiencies of PV module are also

investigated in this study. The effect of varying the flow rate on PV electrical

efficiency is presented in Fig 6.17 and the trend is similar to shown in Fig 6.13 The

electrical efficiency of the PV module increases with the flow rate until the flow rate

reaches 0.05 to 0.055 kg/s.

The electrical efficiency of PV module will be maintained at a fixed value after

the flow rate at 0.055 kg/s. This could be explained associated with the thermal

efficiency of collector. When the flow rate increases to around 0.05 kg/s, the thermal

efficiency of the collector will also be maintained at certain level. In other words, the

heat which extracted by the flow has come to a saturated level and it can no longer be

increased by increasing the flow rate. Thus, the electrical efficiency of PV module will

also be maintained at a fixed value after the flow rate at 0.05 kg/s. Fig 6.17 also

showed that the electrical efficiency of PV module at low irradiation will be higher

than the high irradiation. It is highly likely that the operating temperature of PV

module is much higher in the condition of high irradiation.

Hegazy [35] reported that four types of PV/T air system were investigated in a

comparative study and the electrical efficiency of those systems will come to

83
maximum value at the flow rate at 0.045kg/s and the electrical efficiency will also

keep constant after this flow rate. The discrepancy between the author and Hegazy’s

study could be attributed to the difference of configuration for both systems. Hegazy

utilized a double pass flow design to investigate the electrical efficiency. Therefore, the

air flow has sufficient time to contact with the module and module can be effectively

cooled. Electrical efficiency of the module will be increased due to the effective

cooling. For that reason, the flow rate of Hegazy’s experiment will be lower than

author to attain the maximum electrical efficiency in experiment.

Mass Flow Rate= 0.0389 kg/s

13.2
Mass Flow Rate= 0.0676 kg/s
13
Electrical Efficiency (%)

12.8 Mass Flow Rate =0.0783kg/s

12.6
12.4
12.2
12
11.8
11.6
11.4
11.2
0 1 2 3 4 5 6 7 8 9 10

Temperature Difference (C)

Figure 6.18 Influence of temperature difference (To-Ti) on electrical efficiency for

different flow rate

The electrical efficiency of PV module decreases when the temperature difference

of the inlet and outlet flow increases. Fig 6.18 shows that the electrical efficiency of

84
the PV module significantly decreases with the increase of temperature difference over

the inlet and outlet flow. This may be explained by the high temperature gradient,

which causes the occurrence of hot spots in the PV module. Therefore, the overall

electrical efficiency was decreased due to the local high temperature spot emerges in

the PV module. The inlet and outlet flow temperature difference should be controlled

in an optimum range to ensure that the electrical efficiency of the PV module can still

be maintained at desired output value.

Temperature of the PV module also affects the PV electrical power output and the

results are also shown in Fig 6.19. Under the solar irradiation at 1000W/m2, there is a

decrease of 0.69% of electrical power output for every Celsius degree increase.

However, a decrease of electrical output power by 0.65%/K has been reported by

Radziemska [61]. The decrease in insolation at 800W/m2, 600W/m2 and 400W/m2 are

0.47%/K, 0.39%/K and 0.34%/K, respectively. The trend could be explained from the

derivation below:

It may be assumed that the relation between the solar irradiation and temperature

of PV module to be linear, as also observed from Fig. 6.3.

Therefore, the solar radiation will be linearly proportional to the temperature of the

module.

Let assume,

85
G = FT − C (6.5)

where G is the solar irradiation, T is the temperature of PV module, F and C are the

coefficient. To find the relation between the PV power output and module temperature

can be started from Eq 5-4

ηe = ηo [1 − β (Tc − To )]

Pel
ηe = (6.6)
AG

Substituting the Eqs 5-4 to 6-6, the equation becomes:

Pel = AGηo [1 − β (T − To )] (6.7)

The details of derivation are presented in Appendix. After arranging the equation, it

can be written in the form:

Pel = ( FTo − FTo β − C ) + ( F + C β )[T − To ] − F β (T − To ) 2 (6.8)

Let

86
T ' = T − To (6.9)

Substituting Eq (6.9) into the equation (6. 8) yields,

F + C β 2 F + C β 2 FTo − C − F β To
P = Aηo [(T '− ) +( ) + ] (6.10)
2F β 2F β Fβ

In order to simplify the Eq (6.10), let:

F + C β 2 FTo − C − F β To
Z =( ) + (6.11)
2F β Fβ

F + Cβ
L= (6.12)
2F β

Substituting Eq (6.11) and Eq (6.12) to Eq (6.10) yields,

Pel = −[T '− L]2 + Z (6.13)

Substituting T1 and T2 to equation 6.13

Pel1 = −[T1 '− L ]2 + Z (6.14)

And

87
Pel 2 = −[T2 '− L]2 + Z (6.15)

Let Eq (6.14) subtract Eq (6.15)

Pel1 − Pel 2 = −[T1 '− L]2 + Z + [T2 '− L]2 − Z (6.16)

Percentage of Electrical power decrease over the temperature difference T1 and T2

Pel1 − Pel 2 (T2 '+ T1 '− 2 L)(T2 '− T1 ')

Pel1 Z − (T1 '− L) 2
= (6.17)
T1 − T2 T1 − T2

Electrical output decrease over temperature increase

Pel1 − Pel 2
Pel1 (T '+ T1 '− 2 L)
=− 2 (6.18)
T1 − T2 Z − (T1 '− L) 2

Thee derivation above, it shows that when temperature T2 is much larger than T1

then the electrical output power percentage decreases significantly for every Celsius

degree. This situation could happen in high irradiation under with and without cooling

as under the high irradiation temperature of PV module can be very high without any

cooling mechanism. However, if the air is circulated at the back of PV module, the

88
temperature of PV module can be reduced. It means that under the same solar

irradiation, the temperature of PV module with and without cooling could be varied a

lot. Eq 6.18 shows good agreement with the experimental data which shown in Fig

6.19. At low solar irradiation, the temperature of T1 and T2 cannot be varied a lot and

therefore the electrical output power percentage is lower than the situation in high

solar irradiation.

Irradiation= 1000W/m2
250
Irradiation= 800W/m2

Irradiation= 600W/m2
PV Power Output (W)

200
Irradiation= 400W/m2
150

100

50

0
30 35 40 45 50 55 60 65
Temperature (C)

Figure 6.19 PV electrical power output under different solar radiation

1100 Irradiation 6.0

1000 PV Current 5.5
900 5.0
800 4.5
Irradiation (W/m2)

PV Current (A)

700 4.0
600 3.5
500 3.0
400 2.5
300 2.0
200 1.5
100 1.0
10:48

11:54

12:50

13:46

14:42

15:38
9:47

Time

Figure 6.20 Solar radiation of the entire day and the corresponded PV current due to
the solar radiation (23 September 2009)
89
 33 1100
PV Voltage
32 Irradiation 1000
32 900

Irradiation (W/m )
2
32
PV Voltage (V)
800
32
700
32
600
31
500
31
31 400

31 300
31 200
9:47
10:05
10:23
10:46
11:14
11:32
11:50
12:08
12:26
12:44
13:02
13:20
13:38
13:56
14:14
14:32
14:50
15:08
15:26
15:44
16:02
16:20
Time

Figure 6.21 Solar irradiation and the PV Voltage for the entire day
(23 September 2009)

Fig 6.20 shows the relation between the irradiation and PV current output. This

figure also provides important information about the stand alone system. Stand alone

systems need to use the battery bank to store the electricity generated by the PV

module during operation. This figure shows that the PV current generated corresponds

to the solar irradiation. Fig 6.21 also shows the relation of irradiation and PV voltage.

Some minor fluctuations are seen in this figure but the overall trend of the PV voltage

is well corresponded to the irradiation. However, when the battery bank of stand-alone

PV systems are fully charged, the PV current and PV voltage of the system do not

correspond to the irradiation. Figs 6.22 and 6.23 display the results when the battery

banks are fully charged. Once the battery bank of the stand alone systems are fully

charged, the voltage of the PV module becomes constant at, around 38 V. This could be

90
because the PV current cannot flow through the external load but the electron-hole pair

continues to generate voltage by the Photovoltaic effect.

1000 39

900 38
Irradiation (W/m )

800 37
2

PV Voltage (V)
700 36

600 35
Irradiation
500 34
PV Voltage
400 33

300 32
9:41
10:02
10:23
10:44
11:05
11:26
11:47
12:08
12:29
12:50
13:11
13:32
13:53
14:14
14:35
14:56
15:17
15:38
15:59
16:20
16:41
17:02
17:23
Time

Figure 6.22 Solar radiation and the PV Voltage for the entire day (8 June 2009)

3
900 Irradiation 3
PV Current
800 3
Irradiation (W/m )
2

PV Current (A)

700 2

2
600
2
500
2
400
2

300 1
9:41
10:03
10:25
10:47
11:09
11:31
11:53
12:15
12:37
12:59
13:21
13:43
14:05
14:27
14:49
15:11
15:33
15:55
16:17
16:39
17:01
17:23

Time

Figure 6.23 Solar radiation of the entire day and the corresponded PV current due to
the solar radiation (8 June 2009)

91
 Thus, the voltage in the PN junction will increase to abnormal level. Fig 6.21

indicated that the PV voltage is around 32 V when the battery banks of the stand alone

system are partially discharged. In other words, the PV current which generated by the

Photovoltaic effect can flow through the external load and the electricity can be stored

by the battery bank. Fig 6.23 provides useful information to the stand alone system, as

it can tell whether the battery bank of the system is fully charged.

26.2
26.1
26.0
25.9
Voltage (V)

25.8
25.7
25.6
25.5
Blower Voltage
25.4 Battery Voltage
25.3
25.2
10:07
10:27
10:52
11:22
11:42
12:02
12:22
12:42
13:02
13:22
13:42
14:02
14:22
14:42
15:02
15:22
15:42
16:02
16:22
9:47

Time

Figure 6.24 Battery and blower voltage of partially discharged battery bank
(23 September 2009)

92
 31

30

30

Voltage (V) 29

29
Battery Voltage
28
Blower Voltage
28

27

27
9:41

9:59

10:17

10:35

10:53

11:11

11:29

11:47

12:05

12:23

12:41

12:59

13:17

13:35

13:53

14:11

14:29

14:47

15:05

15:23

15:41

15:59

16:17

16:35

16:53

17:11
Time

Figure 6.25 Battery and blower voltage of fully charged battery bank (8 June 2009)

Figs 6.24 and 6.25 are the battery and blower voltage in fully charged and

partially discharged conditions, respectively. The voltage of the blower was almost

identical to the battery voltage as the blower was hooked up to the battery directly. In

the partially discharged case, the voltage of battery will vary with the state of battery

storage. The battery voltage will keep constant all the way when it is in fully charged

condition. The battery voltage will be around 25 to 26 V when it is under the partially

discharged condition. However, if the battery bank is always hooked up with the panel

without any discharging mechanism, the battery bank may come to a saturated

condition and the battery voltage will be around 30. 5 V. Gassing problems might

occur if the battery continues to be charged after being fully charged as during the

charging process of a fully charged battery, hydrogen and oxygen are released.

Therefore, to avoid the hydrogen explosion hazard, the battery should be kept in a well

ventilated area.

93
 6.00
5.50 Fully Charged Battery
5.00 Partially Discharged Battery
4.50
PV Current (A)

4.00
3.50
3.00
2.50
2.00
1.50
1.00
500 550 600 650 700 750 800 850 900 950 1000
2
Irradiation (W/m )

Figure 6.26 PV current generated by module in case: (a) partially discharged battery
and (b) fully charged battery

Fig 6.26 indicates that, the current which generated by the PV module is linearly

proportional to the solar radiation when the battery bank was not at the condition of

fully charged. However, if the battery bank was under fully charged condition, the PV

current will keep constant at 1.6 A, regardless of the increase of solar irradiation. For

every 100 W/m2 of solar irradiation increment, the PV current output from the module

will increase 0.56 A. It can be seen that when the solar irradiation reaches 1000 W/m2,

the PV current output can attain to around 5.6 A. By partially discharging the battery

bank can ensure that the PV module working properly and effectively.

94
 Battery Partially Discharged
14
Battery Fully Charged
13

Electrical Efficiency (%)

12
11
10
9
8
7
6
5
4
9:47

10:07

10:27

10:52

11:22

11:42

12:02

12:22

12:42

13:02

13:22

13:42

14:02

14:22

14:42

15:02

15:22

15:42

16:02

16:22
Time

Figure 6.27 Electrical Efficiency of fully charged and partially discharged at the
similar meteorological condition

The electrical efficiencies of stand alone system which in battery bank fully

charged and partially discharged condition are presented in Fig 6.27. The electrical

efficiency of the system at the fully charged battery condition will be lower than that in

the usual condition and this may be observed in Fig 6.27. The electrical efficiencies of

PV module are similar in the initial condition, once the battery bank is fully charged by

the PV module; the electrical efficiency of the PV module will drastically drop. The

electrical efficiency of PV module can decrease to 4.6% in the condition of fully

charged of battery bank. However, if the battery bank of the system is partially

discharged, the electrical efficiency may be able to reach 13.4%. This significant

difference has been found from the experiment and it also states the importance of

discharging battery regularly to ensure that the PV module can always be operated

effectively.

95
 35.00
31.20 31.00
Total Solar Power
30.00
Electrical Energy
25.00 22.70
Thermal Energy
Energy (MJ)
20.00 17.10
16.40

15.00 13.70
10.60 10.90
10.00
6.18
4.37
5.00 3.24 3.26
2.30
1.39 1.11
0.00
1 2 3 4 5
Days

Figure 6.28 Input solar radiation and thermal and electrical energy production over five
days

25.00

Electrical Energy
20.00 Thermal Energy
Total Energy Gain
Energy (MJ)

15.00

10.00

5.00

0.00
1 2 3 4 5
Days

Figure 6.29 Electrical and thermal energy and the total energy gain over the five days

96
 70.00% Thermal Efficiency
Electrical Efficiency
60.00%

50.00%

Efficiency 40.00% 5 5 .1 7 %
5 2 .5 6 %
4 5 .1 1 % 4 7 .9 5 %
4 1 .2 3 %
30.00%

20.00%

10.00%
1 0 .1 5 % 1 0 .3 8 % 1 0 .4 7 % 1 0 .5 2 % 1 0 .1 7 %
0.00%
1 2 3 4 5
Days

Figure 6.30 Comparison of thermal and electrical efficiency over 5 days.

The solar power input over the five days (from 22 September to 26 September) is

displayed in Fig 6.28. The electrical and thermal energy produced by the system are

also given. This provides a good estimation of how much energy can be generated by

using this experimental set-up. Application wise, it can also show that how much

electrical energy can be withdrawn from the PV module to be utilised in household

application. Fig 6.28 also shows the incident solar power for those five days which

means that according to the solar power input information, the energy power output of

thermal and electrical can be estimated.

Fig 6.29 show that a large amount of thermal energy was generated during the

operation of PV system, and this also shows that the thermal energy can be utilised in

other aspects like, drying the food product or using as a heater in temperate zone

country instead of exhausting the hot air to the surrounding. Due to the meteorological

97
condition, the solar radiation at Day 3 was much lower than the rest of days and this

was also reflected on the energy output in thermal and electrical aspect. Fig 6.29 shows

that the peak of total energy output occurs on the second day and fourth days. This can

be attributed to the meteorological conditions on those days. The ambient temperature

of those two days was relatively high and the solar radiation was also very intense.

Therefore, it can be concluded that under the proper function of the system, the output

energy can be generated proportional to the solar power input.

The efficiency of the system shown in Fig 6.30 indicates that the electrical

efficiency seems to be more stable than the thermal efficiency. The average electrical

efficiency range is around 10.1% to 10.9%. However, the thermal efficiencies of the

system are around 40% higher than the electrical efficiency of the system. From the

graph, it can be easily seen that the thermal efficiency fluctuates significantly, unlike

electrical efficiency. The reason could be that the thermal efficiency is a function not

only of solar irradiation but also of the ambient temperature, heat losses to the

surrounding and other meteorological parameters. Due to those factors, the variation of

the thermal efficiency of the system is understandable. The total efficiency of the

system is around 55% to 65%. It can be concluded that the overall efficiency of the

PV/T system is much higher than the PV system. This is also implied that the PV/T

system can adequately harness the solar energy.

98
6.3 Simulation heat transfer on a single cell under the meteorological

condition on 23 September 2009

This simulation was done using a commercial finite element software-COMSOL

MULTIPHYSICS. In this simulation, the meteorological data of 23 September 2009

were used to simulate the operation of PV system under the cooling condition. The

purpose of this simulation is to investigate the temperature profile of PV module under

the solar irradiation at 23 September 2009 and the experimental data are used to verify

the simulation result. A good agreement between the simulation and experimental

results has been shown in the Fig 6.31. Some deviations are presented from 12.00 pm

to 2 pm. This may be attributed to the variation of the irradiation data and it can be

seen at Fig 6. 20. This figure shows the temperature profile of the back of the PV

module. However, Fig 6.32 shows the temperature profile of the front glass of PV

module. The discrepancy of simulation and experimental data are significant in this

figure and this is because that the wind speed of the simulation is assumed to be

constant but in the real meteorological condition the wind speed is varied from time to

time and it is also very difficult to use a polynomial equation to represent. For that

reason, the temperature profile of the front glass of panel is quite different that from

the experiment but the trend according to the solar irradiation is still in agreement.

99
 56
54
52
Temperature (C) 50
48
46
44 Experiment
42 Simulation

40
38
9:47 10:21 10:51 11:19 11:58 12:04 13:20 13:52 15:42 16:02 16:25

Time

Figure 6.31 A comparison of simulation and experiment in the temperature profile of

the back of PV module

54

52
Temperature (C)

50

48

46
Experiment
44 Simulation

42

40
9:47 10:21 10:51 11:19 11:58 12:04 13:20 13:52 15:42 16:02 16:25

Time

Figure 6.32 A comparison of simulation and experiment in the temperature profile of

the front of PV module

100
 Figure 6.33 Temperature contour of the PV cell at 1:30 pm (highest solar radiation on
that day)

Temperature profile of PV module is displayed in Fig 6.33. The maximum

temperature of the module occurrs at the silicon cell. This is attributed to the high

absorption of silicon cell in solar irradiation. Temperature of Tedlar (backsheet) is

higher than front glass of PV module, this can be explained that the tedlar is closer to

the silicon cell compared to the front glass even though the thermal diffusivity of the

glass is higher than Tedlar. In addition, from the analysis in chapter 5, it is showed that

the solar irradiation with the wavelength more than 1.1μm will transmit the silicon

solar cell and absorbed by the Tedlar (backsheet). Therefore, the simulation results

show that the temperature at the back of PV module is higher than that of the front

glass.

101
 Figure 6.34 Temperature gradient of the PV module at 1:30 pm

The temperature gradient over the module is also investigated in this simulation.

The results of simulation are presented in Fig 6. 34. The maximum and minimum

temperature gradient of the PV module occurs in the material of Tedlar and silicon

solar cell, respectively. According to the heat diffusion equation Eq 5-19, the thermal

diffusivity is inversely proportional to the temperature gradient. Table 5-1 shows the

thermal diffusivity of each layer of material inside the PV module. The thermal

diffusivity of silicon solar cell is the highest among the materials and it means that the

material of large thermal diffusivity will respond quickly to change in its thermal

environment. For that reason, the temperature gradient of the silicon solar cell is the

lowest among the materials. However, the thermal diffusivity of Tedlar is the lowest

among the materials, by using the Eq 5-19, it is understandable that the temperature

102
gradient of this material is the highest due to the low thermal diffusivity in the

denominator of the equation.

This simulation has accurately predicted the behaviour of the PV module under

the meteorological condition at 23 September 2009. This model can be used to

simulate different type of meteorological conditions to predict the temperature profile

of PV module. However, the meteorological condition, like solar irradiation, ambient

temperature and wind speed, must be able to use a polynomial equation to represent it

otherwise the simulation result might have a significant discrepancy with the

experimental outcome. This is because that the simulation model is created by using a

transient equation and the ambient temperature and solar irradiation is varied from

time to time, therefore, if those meteorological conditions cannot be represented as a

polynomial equation, the discrepancy between the experiment and simulation would be

very significant.

103
 CHAPTER 7
CONCLUSION

The performance of a PV/T system has been successfully determined in an

experimental study. The result of the heat transfer simulation of the silicon solar cell

was also in good agreement with the experimental results.

The impact of the cooling on the PV module has been thoroughly discussed in the

thesis. The PV module temperature is a function of conversion efficiency, which can

severely adversely affect the electrical performance of the PV system. The electrical

efficiency of PV module at 68℃ is around 8.6%. Therefore, the pay back period of the

overall system needs to be extended and the degradation of the PV materials could also

happen due to the high operating temperature. The effects of cooling and non-cooling

on the PV module operating temperature are clearly presented in this thesis.

By using the active cooling technique, the experimental result has shown the

significant improvement on the electrical efficiency of PV module, the electrical

efficiency of PV module can be maintained at around 13%. The optimum flow rate to

enhance the heat transfer from the PV module to air is also found in this study to be

around 0.05 kg/s. The thermal performance of the PV/T system is also computed and it

shows that large amount of thermal energy is collected instead of being dissipated to

104
the environment or trapped in the PV module. The total energy efficiency of the system

can reach 65 %. In other words, there will be 65 % of solar irradiation converted into

the usable energy through the PV/T system of this study.

Furthermore, the uniform flow field which presented in the experiment also

presented a minor temperature difference over the PV module. It can be observed in

the layout of experimental result of Chapter 6. Reducing the temperature difference

over the different PV module can also help to increase the entire system electrical

efficiency. In short, the flow field of cooling medium is also a factor to get an efficient

PV/T system.

It then seems the advantageous to combine the PV module with the collector. The

high efficiency of the combined system can shorten the payback period of the entire

system. The cost of adding the collector to the PV module is not very significant

compared to the price of PV module itself. Therefore, the PV/T system is worth

developing in the industry.

105
 CHAPTER 8
RECOMMENDATION

Some methods of increasing the performance of the PV/T system are

recommended in this chapter. The output power could then be increased and the

payback period shortened. Therefore, the modification should not significantly

increase the cost of the entire system, otherwise the pay back period might need to be

extended. The methods introduced below are of low cost but has significant impact on

the electrical efficiency.

For the PV cell, the electron in the PV material can only be knocked into higher

energy state by a photon of certain wavelength which corresponds to the band gap of

the PV material. For silicon cell, the photon of wavelength above 1.1 μm is unable to

knock the electron from valence band to conduction band and that part of energy will

be converted into phonon and increases the temperature of PV module. The radiation

of wavelength above 1.1 μm can be addressed as infrared radiation.

However, according to some research, it was found that water can be effectively

used to absorb the thermal energy from the IR band but allows transmission of the

visible spectrum [62] most useful for the PV operation. Fig 8.1 shows that the

absorption coefficient of water in visible band is very low compared to that in the

infrared band. This adequately indicates that the water effectively absorbs energy in

106
the infrared band and this there is too energy to knock the electron from valence band

to conduction band. Therefore the contribution of phonons from the infrared band can

be reduced and temperature of PV module can also be maintained.

Figure 8.1 Spectral absorption coefficient of pure water (solid line) and of pure sea
water (dotted line) as a function of wavelength.

In order to investigate this phenomenon, a design has been proposed as shown in

Fig 8.2. According to that design, the water can absorb the infrared in the sunlight and

let the visible band pass through the water without any interference. The temperature

of water which absorbs the infrared will increase and it can be utilised as household

application. The thermal energy from the infrared can be recovered instead of throwing

to the environment and thereby the total efficiency of the system will increase

significantly.

107
 Figure 8.2 Transparent water passage in front of the PV panel to pre-filter the solar
irradiation before it strikes the solar cell.

Another method to boost the overall efficiency of the PV module involves a plane

reflector to augment the aperture area of the PV module. The plane reflector can be

used to increase the area of receiving sun light of PV module. The sunlight will be

reflected by the reflector to the PV module and the overall power output of the PV

module increases due to the increase of incident solar irradiation to the PV module.

The cost of the plane reflector is less than 5% of the cost of the PV system but the

efficiency of the PV module can significantly be boosted. However, due to the

meteorological condition, the inclination angle of the reflector should be adjusted to

the optimum value. In short, this is a very promising way to enhance the overall

performance of PV system without paying the high cost on the plane reflector itself.

108
 References
1. International Energy Agency, 2008. World Energy Outlook 2008 Factsheet:

Global Energy Trends. International Energy Agency.

2. WG 3-Mitigation of climate

change- http://www.ipcc.ch/, http://www.ipcc.ch/pdf/assessment-report/ar4/wg3/ar

4-wg3-chapter1.pdf

3. IPCC- http://www.ipcc.ch/pdf/assessment-report/ar4/syr/ar4_syr.pdf

4. http://www.nasa.gov/worldbook/sun_worldbook.html

5. K.S. Jairaj, S.P. Singh, K. Srikant. A review of solar dryers developed for grape

drying, Solar Energy, 83, pp.1698-1712. 2009.

6. P.G. Charalambous, G.G. Maidment, S.A. Kalogirou, K. Yiakoumetti.

Photovoltaic Thermal (PV/T) collectore: A review, Applied Thermal Engineering,

27, pp.275-286. 2007.

7. L.W. Florschuetz, Extension of the Hottel-Whillier Model to the Analysis of

Combined Photovoltaic/Thermal Flat Plat Collectors, Solar Energy,

22,pp.361-366. 1979.

8. H.C.Hottel, A. Willier, Evaluation of Flat-plate Solar Collector performance,

Transactions of the Conference on the Use of Solar Energy, Vol. 2, University of

Arizona Press, Tucson, Arizona, 1958.

109
9. H.A.Zondag, D.W.DE Vries, W.G.J. Van Helden, R.J.C. Van Zolingen and

A.A.Van Steenhoven., The Thermal and Electrical Yield of a PV-Thermal

Collector, Solar Energy, 72, pp.113-128. 2002.

10. Bjornar Sandnes and John Resstad. A Photovoltaic/Thermal (PV/T) Collector with

a polymer Absorber Plate. Experimental Study and Analytical Model, Solar

Energy, 72, pp.63-73. 2002.

11. R. Zakharchenko, L. licea-Jimenez, S.A. Perez-Garcia, P. Vorobiev, U.

Dehesa-Carrasco, J.F. Perez-Robles, J. Gonzalez-Hernandez, Yu. Vorobiev.

Photovoltaic solar panel for a hybrid PV/thermal system, Solar Energy Materials

& Solar Cells, 82,pp.253-261. 2004.

12. B.J.Huang, T.H.Lin, W.C.Hung and F.S.Sun. Performance Evaluation if Solar

Photovoltaic/Thermal Systems, Solar Energy, 70, pp.443-448. 2001.

13. Jie Ji, Jian-Ping Lu, Tin-Tai Chow, Wei He, Gang Pei. A sensitivity study of a

hybrid photovoltaic/thermal water-heating system with natural circulation,

Applied Energy, 84,pp.222-237. 2007.

14. Wei He, Tin-tai Chow, Jie Ji, Jianping Lu, Gang Pei, Lok-shun Chun. Hybrid

photovoltaic and thermal solar-collector designed for natural circulation of water,

Applied Energy, 83, pp.199-210. 2006.

15. T.T. Chow, A review on photovoltaic/thermal hybrid solar technology, Applied

110
 Energy, 87, pp.365-379. 2010.

16. S.A. Kalogirou. Use of TRNSYS for modelling and simulation of a hybrid

PV-thermal solar system for Cyprus, Renewable Energy, 23, pp.247-260. 2001.

17. T. Fujiwa, T. Tani. Annual exergy evaluation on photovoltaic thermal hybrid

collector, Solar Energy Materials on Solar Cells, 47, pp.135-148. 1997.

18. T. Bergene, O.M Lovvik, Model calculations on a flat plat solar heat collector

with integrated solar cells, Solar Energy, 55, pp.453-462. 1995.

19. H.P. Garg, R.K. Agarwal, Some aspects of a PV/T collector/ forced circulation flat

plate solar water heater with solar cells, Energy Conversion and Management, 36,

pp.87-99. 1995.

20. Nishikawa M, Sone T, Ito S. A heat pump using solar hybrids panel as evaporator.

In: ISES Solar World Congress, Korea, 1997.

21. Ito S, Miura N, Wang K. Heat Pump using solar collector with photovoltaic

modules on the surface. JSEE, 119, pp.147-51. 1997

22. Zondag HA, Jong MJM, Van Helden WGJ. Development and applications for PV

thermal. In: 17th EPSEC 2001, Munich, 2001.

23. Jong MJM. System studies on combined PV/Thermal panels. In: 11th symposium

thermische solarenergie, Staffelstein, 2001.

24. Ji J, Chow TT, He W. Dynamic performance of hybrid photovoltaic/thermal

111
 collector wall in Hong Kong Building Environ, 38, pp.1327-34. 2003.

25. Elswijk MJ, Jong MJM, Braakman JNC, Lange ETN de, Smit WF. Photovoltaic

/Thermal collectors in large solar thermal system. IN: 19th EPSEC 2004, Paris,

2004

26. Hendrie SD. Photovoltaic/thermal collector development program-final program.

Report, MIT, 1982.

27. Smith DR, Biringer KL, Pritchard DA. Combined photovoltaic thermal collector

testing. In: 13th IEEE, Washington, 1978.

28. Ricaud A, Roubeau P. Capthel, a 66% efficient hybrid solar module and the

‘ecothel’ co-generation solar system. In: First WCPEC, Hawaii, 1994.

29. Elazari A. Multi solar system: solar multi module for electrical and hot water

supply for residentially building. In: 2nd WCPEC, Vienna, Austria; 1998.

30. Bosanac M, Soerensen B, Katic I, Soerensen H, Nielsen B, Badran J.

Photovoltaic/thermal solar collectors and their potential in Denmark. Report EFP

Project 1713/00-0014, 2003.

31. PVT Roadmap-a European guide for the development and market introduction of

PV-Thermal technology. Zondag, HA, Bakker M, Helden WGJ, editor. Report

EU-Project PV-Catapult www.pvtforum.org, 2005.

32. Arvind Tiwari, M.S.Sodha, Parametric study of various hybrid PV/thermal air

112
 collector: Experimental validation of theoretical model, Solar Energy material &

Solar Cells, 91, pp.17-28. 2007.

33. H.P.Garg, R.S.Adhikari, System performance Studies on Photovoltaic/Thermal

(PV/T) Air Heating Collector, Renewable Energy, 16, pp.725-730. 1999.

34. Swapnil Dubey, G.S.Sandhu, G.N.Tiwari, Analytical expression for electrical

efficiency of PV/T hybrid air collector, Applied Energy, 86, pp.697-705. 2009.

35. Adel A. Hegazy, Comparative study of the performances of four

photovoltaic/thermal solar air collectors. Energy Conversion & Management, 41,

pp.861-881. 2000.

36. K. Sopian, H.T.Liu,S.Kakac,T.N.Veziroglu, Performance of a double pass

photovoltaic thermal solar collector suitable for solar drying systems, Energy

Conversion & Management, 41, pp.353-365. 2000.

37. Prasad, B.N., Saini, J.S. Optimal thermohydraulic performance of artificially

roughened solar air heaters. Solar Energy, 47, pp.91-96. 1991.

38. Han, J.C., Park, J.S. Developing heat transfer in rectangular channels with ribs

turbulators, International of Heat and Mass Transfer, 31, pp.183-195. 1988.

39. Gupta, D., Solanki, S.C., Saini, J.S. Heat and fluid flow in rectangular solar air

heater ducts having transverse ribs roughness on absorber plate, Solar Energy, 51,

pp. 31-37. 1993.

113
40. Garg, H.P, Datta, G. Performance studies on a finned-air heater, Solar Energy, 14,

pp.87-92. 1989.

41. Garg, H.P., Agarwal, P.K., Bhargava, A.K. The effect of plane booster reflectors

on the performance of a solar air heater with solar cells suitable for a solar dryer.

Energy Conversion and Management, 32, pp.543-554. 1991.

42. Pottler, K., Sippel, C.M., Beck, A., Fricke, J. Optimized finned absorber

geometries for solar air heating collectors, Solar Energy, 67, pp.35-52. 1999.

43. Naphon, P. On the performance and entropy generation of the double-pass solar

air heater with longitudinal fins, Renewable Energy, 30, pp. 1345-1357. 2005.

44. Tonui, J.K., Tripanagnostopoulos, Y. Air-cooled PV/T solar collectors with low

cost performance improvements. Solar Energy, 81, pp.498-511. 2007.

45. K.Sopian, K.S.Yigit, H. T. Liu, S. Kakac and T. N. Veziroglu, Performance

Analysis of Photovoltaic Thermal Air Heaters, Energy Convers, 37, pp.1657-1670.

1996.

46. H.P.Garg, R.S.Adhikari, Conventional hybrid photovoltaic/thermal (PV/T) air

heating collectors: steady-state simulation, Renewable Energy, 11, pp.363-385.

1997.

47. A.S.Joshi, A.Tiwari, G.N Tiwari, I. Dincer B.V.Reddy, Performance evaluation of

a hybrid photovoltaic thermal (PV/T) (glass-to-glass) system, International Journal

114
 of Thermal Science, 48, pp.154-164. 2009.

48. Y. Tripanagnostopoulos, TH. Nousia, M. Souliotis and P. Yianoulis, Hybrid

Photovoltaic/Thermal Solar Systems, Solar Energy, 72, pp.217-234. 2002.

49. Y. Tripanagnostopoulos, Aspects and improvements of hybrid

photovoltaic/thermal solar energy systems, Solar Energy, 81, pp.1117-1131. 2007.

50. K.J. Lewis, Encapsulant material requirements for photovoltaic modules, in: C.G.

Geblein, D.J. Williams, R.D. Deanin (Eds.), Polymers in Solar energy Utilisation,

ACS, Washington, DC, 1983, pp. 367–385 (Chapter 23).

51. Halden Field. Solar Cell Spectral Response Measurement Errors Related to

Spectral Band Width and Chopped Light Waveform. 26th IEEE Photovoltaic

Specialists Conference, September 1997, Anaheim, California.

52. Raghuraman, P., Hendrie, S.D. Analytical Predictions of Liquid and Air

Photovoltaic/Thermal, Flat-plate collector Performance, American Society of

Mechanical Engineers, (80-WA/Sol-3), 1980.

53. C. H. Cox, III and P. Raghuraman, Design Consideration for Flat Plat

Photovoltaic/Thermal Collectors, Solar Energy, 35, pp.227-241. 1985.

54. J. A. Duffie and W. A. Beckman, Solar Energy Thermal Processes, Wiley, New

York, pp.76, 103, 138-147 (1974).

55. J. W. Stultz and L. C. Wen. Thermal performance testing and analysis of

115
 photovoltaic modules in natural sunlight. LSSA Project Task Report 5101-31, Jet

Propulsion Laboratory, California Institute of Technology, Pasadena, CA (29 July

1977).

56. Petukhov, B.S., in T. F. Irvine and J.P. Hartnett, Eds., Advances in Heat Transfer,

Vol. 6, Academic Press, New York, 1983.

57. Tonui, J.K., Tripanagnostopoulos, Y, Improved PV/T solar collectors with heat

extraction by forced or natural air circulation, Renewable Energy, 32, pp.623-637.

2007.

58. Mohd. Yusof Othmana*, Baharudin Yatima, Kamaruzzaman Sopianb, Mohd.

Nazari Abu Bakara Performance studies on a finned double-pass

photovoltaic-thermal (PV/T) solar collector, Desalination, 209, pp.43–49. 2007.

59. M.Y.H. Othman, K. Sopian, B. Yatim, W.R.W. Daud, Development of advanced

solar assisted drying systems, Renewable Energy, 31, pp.703-709. 2006.

60. H. A. Zondag, W. G. J. van Helden, Stagnation Temperature in PVT Collector,

Energy research Centre of Netherland ECN, Nov 2002.

61. E.Radziemska, The effect of temperature on the power drop in crystalline silicon

solar cells, Renewable energy, 28, pp.1-12. 2003.

62. Mobley C. D., Optical Properties of Water, in Handbook of Optics, Second

Edition, M. Bass, editor, 1994, McGraw-Hill, Inc.

116
 Appendix A

Manufacturer’s Specifications

A.1 Neste Advanced Power System Solar Modules

A.2 FirstPower Technology Deep Cycle Gel Batteries

A.3 Sanyo Denki DC Fan

117
Fig A.1. Specifications of PV modules

118
Fig A.2. Specifications of solar batteries

119
Fig A.3. Specifications of DC fan (model no. 109E2024MH002)

120
 Appendix B

Calibration of T-type thermocouple

Thermocouple 1
Master Thermometer (℃)

80

70

60
y = 0.9946x + 0.8102
50 R2 = 0.9997
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T1 (℃)

Thermocouple 2
80
Master Thermometer

70 y = 0.9946x + 0.6602
60
R2 = 0.9997
50
(℃)

40

30

20

10

0
0 10 20 30 40 50 60 70 80

T2 (℃)

121
 Thermocouple 3

Master Thermometer (℃) 80

70
y = 0.9922x + 0.5936
60

50
R2 = 0.9996
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T3 (℃)

Thermocouple 4

80
Master Thermometer (℃)

70
y = 0.9898x + 0.6075
60

50
R2 = 0.9996
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T4 (℃)

Thermocouple-Inlet
80
Master Thermometer (℃)

70 y = 0.9869x + 0.8089
R2 = 0.9997
60

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80

Inlet Temperature (℃)

122
 Thermocouple-Outlet

Master Thermometer (℃) 80

70
y = 0.9929x + 0.3123
60

50
R2 = 0.9995
40

30

20

10

0
0 10 20 30 40 50 60 70 80

Outlet Temperature (℃)

Thermocouple 7

80
Master Thermometer (℃)

70 y = 0.9844x + 0.5501
R2 = 0.9995
60

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80

T7 (℃)

Thermocouple 8

80
Master Thermometer (℃)

70
y = 0.9873x + 0.4387
60

50
R2 = 0.9996
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T8 (℃)

123
 Thermocouple 9

Master Thermometer (℃) 80

70
y = 0.9887x + 0.3807
60

50
R2 = 0.9994
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T9 (℃)

Thermocouple 10
Master Thermometer (℃)

80

70
y = 0.9799x + 0.6747
60

50 R2 = 0.9996
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T10 (℃)

Thermocouple 11
Master Thermometer (℃)

80

70
y = 0.9881x + 0.5503
60

50
R2 = 0.9994
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T11 (℃)

124
 Thermocouple 12

Master Thermometer (℃) 80

70
y = 0.9902x + 0.3977
60

50 R2 = 0.9993
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T12 (℃)

Thermocouple 13
Master Thermometer (℃)

80

70

60
y = 0.9935x + 0.2027
50 R2 = 0.999
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T13 (℃)

Thermocouple 14
Master Thermometer (℃)

80

70
y = 0.9919x + 0.2999
60

50 R2 = 0.9993
40

30

20

10

0
0 10 20 30 40 50 60 70 80

T14 (℃)

125
 Appendix C

Derivation of the Result

P = Aηo ( FT − C )[1 − β (T − To )]

= Aηo [ FT − C − FT β (T − To ) + C β (T − To )]

= Aηo [ F (T − To ) + FTo − C − F β (T − To ) 2 − F β To + C β (T − To )]

= Aηo [− F β (T − To ) 2 + ( F + C β )(T − To ) + ( FTo − C − F β To )]

(F + Cβ ) ( FTo − C − F β To )
= Aηo [−(T − To ) 2 + (T − To ) + ]
Fβ Fβ

Let T ' = T − To

F + C β 2 F + C β 2 FTo − C − F β To
P = Aηo [(T '− ) +( ) + ]
2F β 2F β Fβ

126
 Appendix D

Progress Log of Simulation

Progress - Solve Problem: Mon Dec 07 00:58:44 CST 2009
fem.sol=femtime(fem, ...
'solcomp',{'T'}, ...
'outcomp',{'T'}, ...
'tlist',[0:1:2331], ...
'tout','tlist');
--------------------------------------------------------------------------------
Number of degrees of freedom solved for: 1675

Symmetric matrices found.

Format not changed since UMFPACK uses unsymmetric storage.

Step Time Res Jac Sol Order Tfail NLfail

1 0.001 3 2 3 1 0 0
2 0.003 5 2 5 2 0 0
3 0.007 7 3 7 1 0 0
4 0.015 9 4 9 1 0 0
5 0.031 11 5 11 1 0 0
6 0.063 13 6 13 1 0 0
7 0.127 15 7 15 1 0 0
8 0.255 17 8 17 1 0 0
9 0.511 19 9 19 1 0 0
1 out
10 1.023 21 10 21 1 0 0
2 out
11 2.047 23 11 23 1 0 0
3 out
4 out

27 2331 out 39 22 39 3 0 0

Time-stepping completed.

127