Professional Documents
Culture Documents
buoyancy force N
b
s
x
strip length ta'en equal to m ""! . " for modelling m
d depth of the fins m
(tefan,BoltDmann constant 4Js.m
!
.<
>
$P improvement factor
* number of fins
c
s'y emissivity
(
kA h+ m
!
%%
CHAPTER 1 INTRODUCTION
*his chapter reviews present day energy usage1 current energy issues and challenges1
prospects for renewable energy and in particular potential for solar energy. $t also
discusses the requirements for studies on solar water heaters +(-./ and specifically
introduces the potential for integrated collector storage solar water heaters +$&(,(-./
in the U< and conte%tualises the current research. $t also summarises the problem
statement and ob0ectives of the present research and provides an outline of the thesis.
1.1 E.+-(/
1.1.1 E.+-(/; +.vi-2.8+.: ,.> 026i+:/
Energy1 by definition1 is the capacity to perform wor'. (olar energy supports virtually
all life on Earth through photosynthesis and drives the climate and weather. *his energy
can be developed by means of a variety of natural and synthetic processes. Plants
capture solar radiation by photosynthesis and convert it to chemical form1 while direct
heating or electrical conversion is used by solar equipment to generate electricity or to
do other useful wor'. Even the energy stored in widely used energy resources li'e
petroleum and other fossil fuels was originally converted from sunlight by
photosynthesis over time.
Energy use and supply plays an essential role in society and human life. Economic
development and the wellbeing of society are closely lin'ed to energy. *he continual
growth in populations and venture for a better economy resulted in a dramatic increase
of energy consumption1 particularly in the last two centuries due to the increase of
available energy sources. -ith the evolution of society1 humans progressively gained
access to larger amounts of energy and increased significantly their energy consumption
primarily in the more developed nations to cover industrial1 transport1 space,heating1
lighting and refrigeration needs. odern societies use more energy every year for
industry1 services1 homes and transport and ta'e for granted easily available energy.
.owever1 the increasing concern of climate change and the effect of burning fossil fuels
as well as global awareness of the limited supply of fossil fuels such as oil1 coal or
natural gas shows that energy does define and constrain our progress. *hus1 to optimise
9
the future of our society the allocation of Earth5s energy resources needs to be
understood and prioritised.
1.1.2 C7--+.: +.+-(/ 06+.,-i2
-orld energy sources can be divided into two main categoriesF non,renewable and
renewable. Non,renewable energy sources define energies which cannot be replenished
in a short time period and that eventually become too e%pensive and too
environmentally damaging to recover. -orldwide1 non,renewable energy sources are
predominant including fossil fuels and uranium for nuclear power. 3ossil fuels range
from very volatile materials li'e natural gas1 to liquid oil1 to non,volatile materials such
as coal accounting for :#L of the world5s primary energy supply as shown in 3igure 9,
9 from ?EN!9 +!""C/ data. ?enewable energy sources can be replenished naturally in a
short period of time and will never run out. *hey include solar1 wind1 geothermal1
biomass energy as well as hydropower and ocean energy.
Fossil fuels
79%
Nuclear
3%
Renewables
18%
Biofuels 0.3%
Power generation 0.8%
Hot water/heating 1.3%
Large hydropower 3%
Traditional bioass
13%
Fi(. 1)1* +,- 2''% +.+-(/ 01,-+ 23 (425,4 3i.,4 +.+-(/ 62.0789:i2.; REN21 (2''<)
1.1.2.1 N2. -+.+A,54+ 027-6+0
(olid fossil fuels were the first energy resource used. &oal remains the world5s most
abundant and fastest,growing fossil fuel in !""B accounting for about !;L of the global
power usage based on ?EN!9 +!""C/ fossil fuel data and specification fuel data stated
by BP +!"":a/ and BP +!"":b/. &oal remains the leading source for electricity
generation as well as the largest worldwide source of carbon dio%ide emissions. @il is
the leading fuel in the transport sector therefore ma'ing it very vulnerable to any
disruption in oil price and supply as stated by Buc'lin +!"";/. @il accounted for ;>.=L
!
of the global power usage according to ?EN!9 +!""C/1 BP +!"":a/ and BP +!"":b/
data. Natural gas has been used for over a century for lighting and heating and is now
considered a very valuable resource) being more efficient and cleaner than other fossil
fuels releasing lower amounts of carbon dio%ide emissions per unit energy released as
shown in *able 9,9.
T,54+ 1)1* E8i00i2. -,:+0 23 +.+-(/ 027-6+0 (DEFRA; 2''<,)
F7+4
:/9+
E8i00i2. -,:+0
C(CO
2
?CW1
Electricity ".>;
Natural 7as ".9#
@il ".!=
&oal ".;"
Petrol ".!>
Nuclear ".""# to "."9>
?enewables "
-orld natural gas consumption accounts for !9.=L of the global power usage. BP
+!"":a/ states that European consumption decreased mainly due to large increases in
contracted prices in the U< and Eastern European countries resulting in large
consumption decline. *he ma0or disadvantage in the use of natural gas is its
transportation which is more complicated and e%pensive than other fossil fuels.
Nuclear energy accounts for ;L of the global power usage as stated by ?EN!9 +!""C/.
*he -orld Energy &ouncil +-E&/ +!"":/ stated that at the current rate of production1
using current reactor technology1 global nuclear reserves are estimated to last for almost
another C= years. $ts technology is sometimes promoted as a sustainable energy source
that reduces carbon emissions and can increase energy security for countries with access
to this technology by decreasing dependence on fossil fuel sources. .owever1 there are
political1 security and environmental concerns about nuclear reactor safety as well as
radioactive waste disposal and plant decommissioning.
Despite the world5s increasingly heavy dependence on this leading energy source1
depleting fossil fuel resources do not ma'e them reliable choices for the future and
provide impetus for moving the economy towards sustainable energy sources.
;
1.1.2.2 R+.+A,54+ +.+-(/
?enewable energy can be defined as energy flows which are replenished at the same
rate as they are MusedN as stated by (orensen +!""!/. (olar radiation is the principal
source of earth5s renewable energy sources. (ome of the common renewable energy
technologies and their current status in the global energy mi% are reviewed in the
following paragraphs.
N2.)024,- -+.+A,54+0
*idal and geothermal energy do not depend on solar radiation. *idal power traps water
in a basin activating turbines generating electricity as it is released through the tidal
barrage. *idal energy power generation is still in an early stage of development and
accounts for only "."9L of the world5s energy consisting of ".; 7- reported by
?EN!9 +!""C/. .owever1 based on a recent report from the (ustainable Development
&ommission +(D7/ Appleyard +!"":/ stated that U<5s tidal resources could provide
9"L of the U<5s electricity in the near future.
7eothermal energy meaning M.eat from the EarthN is the energy generated by the heat
contained within the Earth. 7eothermal energy is used for power and for heating.
*rapping geothermal energy for power generation can be achieved using hydrothermal
reservoirs1 hot dry roc'1 geopressure brines and magma1 each with engineering
challenges and constraints ma'ing them not always practical or economically feasible.
2ow temperature geothermal resources by heat e%traction from the near sub,surface of
the Earth have been widely used in the past and provide energy for space1 water heating1
district heating1 greenhouse heating or warming of fish ponds in aquaculture. *he
relatively constant temperature of the top 9= metres of the EarthHs surface can be used to
heat or cool buildings indirectly through the use of heat pumps. *he report published by
?EN!9 +!""C/ states by the end of !"": worldwide use for electricity had reached
9" 7-1 with an additional ;; 7- used directly for heating with half being geothermal
heat pump installations and accounted for ".!L of the global energy usage.
S24,- +.+-(/* I.>i-+6: 70+0
(olar energy can be converted indirectly to useful energy by other energy forms. (olar
radiation drives a hydrologic cycle by causing evaporation1 precipitation and surface
run,off. .ydropower is defined as the energy of moving water and has been e%ploited
>
for many years for irrigation purposes or watermills. *he largest hydropower in use
today generates electricity by transforming potential energy of water stored at an
elevation into 'inetic energy by the rotation of a turbine rotating the motor to produce
electricity. Based on ?EN!9 +!""C/ report hydroelectricity generates about 9BL of the
world5s electricity in !""B consisting of ::" 7- of large hydro plants and :; 7- of
small hydropower installation.
(olar radiation results in differences of temperature in the atmosphere and oceans in
such manner that the convective currents produce winds1 ocean currents and waves.
-ind power has been commonly used for centuries by windmills for pumping water or
crushing corn. -ind power can be produced on a large scale such as on land,based or
offshore wind farms connected to electrical grids or by individual wind turbines
providing electricity to isolated rural areas. -ind is now one of the most advanced
renewable energies due to its mass production1 improvement in quality1 reliability and
cost effectiveness. ?EN!9 +!""C/ reported that wind power capacity accounts for :>
7- or ".;L of global energy usage.
-ave power uses ocean surface motion caused by winds and is mainly used for
electricity generation. ost companies and infrastructures are concentrated along the
U< coastline according to &ameron +!"":/. A tremendous amount of energy is
available in ocean waves though initial technical problems associated with this
technology have delayed its development.
Biomass1 and especially the fuels derived from biomass named Biofuels1 is another
indirect manifestation of solar energy. Biofuels have recently been sub0ect to increasing
attention as interest in sustainable fuel sources grows. *raditional biomass accounts for
about #L of the 9;L of global biomass energy usage as stated by ?EN!9 +!""C/.
Biomass can be classified as a renewable resource if a sustainable balance is therefore
maintained between carbon emitted and absorbed. *he combustion of biomass fuel
emits &@
!
to the atmosphere1 however emission are no more than the amount it
absorbed during its lifetime growth. Biomass is used in power and heating with an
estimated capacity of >= 7- in !""B in Europe with two third used for heating.
S24,- +.+-(/* Di-+6: 70+0
(olar energy is often seen as the fuel of the future and represents the energy generated
directly from the sun. Applications vary from the residential1 commercial1 industrial1
=
agricultural and transportation sectors. (olar energy can be used in two main ways to
produce heat and electricity.
(olar electricity generation has been developed primarily through photovoltaic +PO/ and
solar thermal power generation. PO is generally used for small and medium,siDed
applications and can be grid,connected or autonomous. ?EN!9 +!""C/ reported that PO
connected grids represent the fastest growing global power generation technology
reaching an estimated cumulative installed capacity of :.C 7- at end of !"":. ?EN!9
+!""C/ estimated the cumulative e%isting solar PO at 9".=7- by the end of !"":
accounting for about ".";L of the global energy usage.
&oncentrating solar thermal power +&(P/ generation is more commonly built for large,
scale electricity generation. *his system uses direct solar radiation concentrated by
lenses or mirrors and trac'ing systems to provide high temperature heat for generating
electricity or directly generating electricity by using PO. *oday1 total installed &(P
capacity is estimated by ?EN!9 +!""C/ to be ".>7- accounting for about ".""!L of
global energy usage. .owever1 the -orld Energy &ouncil +!"">/ reported that on a
long term scenario1 the contribution of &(P could reach B;"7- by !">". *hese
technologies however have limited use in cloudy locations as they require direct solar
radiation.
(olar thermal applications are the most widely used solar energy technology and include
solar coo'ing1 deto%ification1 desalination or solar thermal systems. (olar distillation1
pasteurisation and desalination purify water using solar radiation by means of different
processes. (olar drying and solar ponds are different thermal methods to provide
process heat to dry agricultural products1 clothes or to reach high temperature for
chemical reactions and melting of metals. (olar coo'ers capture sunlight which is
converted to heat retained for coo'ing. (olar coo'ers are used for coo'ing1 drying or
even pasteurising water and mil'. $t can be a real solution for problems such as fuel
poverty1 impure mil' and drin'ing water and health problems caused by indoor air
pollution from combustion of hydrocarbon faced by developing countries often with
high solar energy potential.
3inally1 solar thermal collector applications are the most widely used solar energy
technology for domestic hot water and space heating using only sunlight to heat water.
Different designs and techniques are available depending on the application and
temperature required. 7laDed solar water heaters +(-./ such as flat plate collectors1
batch systems or evacuated tube and air collectors typically used for space heating are
the most common types while unglaDed collectors are generally used for heating
B
swimming pools. (-. systems are described in more detail in &hapter !. (-.
systems are efficient and reliable technologies compared to other solar technologies and
are gaining ground in a few countries. ?EN!9 +!""C/ reported that (-. e%isting
capacities accounted for 9"=7-*h in !""B or 9.;L of the global energy usage and was
estimated to reach 9!C7-*h in !"":) accounting for a total installed collector area in
use around the world of 9C; million square meters.
1.2 E.+-(/ i007+0 ,.> 61,44+.(+0
1.2.1 I894i6,:i2.0 0+: 5/ :1+ (425,4 +.+-(/ 06+.,-i2
*he global energy demand is increasing worldwide and will continue to rise as
developing nations reach developed status and developed nations maintain their
modernisation trends. 3igure 9,! from BP +!"":a/ reveals that the energy consumption
per capita is significantly higher in developed states than in less developed and
developing countries.
Fi(. 1)2* 2''% C2.0789:i2. 9+- 6,9i:, i. :2..+0 2i4 +=7iv,4+.:; BP (2''&)
$EA +!""B/ reports that energy demand is pro0ected to grow on average by 9.BL a year
thus an increase of 0ust over one,half between !""B and !";". @ver :"L of this increase
would come from developing countries such as $ndia and &hina which alone would
account for ;"L. *oday5s worldwide energy use is C"L fossil fuels which is to remain
the dominant source of energy until !";" in the ?eference (cenario reported by $EA
+!""B/.
:
" A 9.=
9.= A ;."
;." A >.=
>.= A =."
P".=
.owever1 the current global use of fossil and nuclear fuels has many consequences
including the depletion of natural resources1 threat to the world energy security and
global climate change caused by emissions of greenhouse gases from fossil fuel
combustion. *hese consequences have many environmental1 economical and political
impacts over the world.
1.2.1.1 N2.)-+.+A,54+ -+027-6+0 ,.> :1+i- 4i,5i4i:/
3ossil fuels stoc's are rapidly depleting around the globe. At current consumption rates1
BP +!"":/ and -E& +!"":/ state that proven world reserves should last between >" to
9=" years depending on the fuel as shown in 3igure 9,;.
147
85
63.3
40.5
0 20 40 60 80 100 120 140 160
Coal
Nuclear
Natural as
!il
R"# ratio
Fi(. 1)3* E.> 23 /+,- 2''% F200i4 37+4 -+0+-v+0 :2 9-2>76:i2. (R?P) -,:i20; >,:, 3-28 BP (2''&) @
WEC (2''&)
Pressure on e%isting reserves is increasing daily due to increased demand. onbiot
+!"">/ reported that the world currently consumes si% barrels of oil for every new barrel
discovered. Energy e%perts suggest that oil production will probably pea' sometime
between !""> and !"9" as stated by Asif et al. +!"":/1 followed by natural gas. *he
general decline of fossil fuel production will cause a global energy gap which could
result in serious international1 economic and political crises and conflicts detailed
further in section 9.!.9.;.
*he price of oil has been rising because demand is growing faster than a finite supply.
Based on data from Asif and uneer +!"":/ the cost of crude oil per barrel increased by
="L between !""> and !""=. $EA +!""B/ states that rising oil and gas demand could
C
accentuate the consuming countries vulnerability to severe supply disruption following
a price shoc'. Nuclear energy could be a possible route to reduce the energy gap
however environmental and political concerns associated with this technology do not
ma'e it a favourable approach especially with the current e%istence and growth of green
energies.
@ther concerns recently raised are the vulnerability of current energy infrastructures to
adverse weather conditions. Depletion of water could result in serious problems as vast
amounts of water are required for fuel processing and cooling in fossil fuel1 nuclear and
geothermal power plant as stated by 7leic' +9##>/. et @ffice +!"";/ and 4owit Q
EspinoDa +!""B/ outlined the !""; and !""B heat wave consequences on the operations
of several power plants. Plants were put at ris' and were shut down due to lac' of water
to cool the condensers. @ther components such as gas and oil pipelines or transmission
line could be affected by e%treme weather.
*his energy scene is facing another ma0or challenge and is raising serious
environmental concerns associated with fossil fuel consumption and production.
1.2.1.2 E.vi-2.8+.:,4 62.6+-.0
Energy production1 distribution and consumption raised serious environmental concerns
over the last century. &limate change is defined by the UN +9##!/ as Ma change in
climate that is attributed either directly or indirectly to human activity that alters the
composition of the global atmosphere and which is in addition to natural climate
variability observed over comparable time periods.N.
&limate change is caused by an increase of greenhouse gases by natural processes or by
human activities. 3or the last 9=" years1 atmospheric concentrations of greenhouse
gases have been steadily increasing partly due to the industrialisation revolution of
human activities. &arbon Dio%ide +&@
!
/ is one of the main greenhouse gases which is
primarily produced from burning fossil fuels and deforestation. @ther gases such as
methane and nitrous o%ide predominantly produced from agricultural activities and
changes in land use or halocarbons and sulphur he%afluoride released by industrial
processes contribute to climate change. aplecroft net 2td +!"":b/ reported that world
carbon dio%ide emissions in !"!= are pro0ected to reach ;C.C billion tons1 e%ceeding
9##" levels by C9L.
#
&limate change is already happening and its first signs can be witnessed around the
globe. $P&& +!"":/ reported measurements recording an increase of ".:> R ".9C& in
mean global surface temperature from 9#"B to !""= against the baseline of 9>& from
(ims +!"">/ and predicted a rise by 9.> to =.CS& during the !9st century. As a result1
mean sea level rose by 9" to !"cm with an average rate increase of about ;.9 R ".:mm
per year from 9##; to !"";. $P&& +!"":/ reported that since 9##; thermal e%pansion of
the oceans has contributed to =:L of the sea level rise while melting glaciers and ice
caps contributed for about !CL and polar ice sheets contributed for the remaining 9=L.
Asif et al. +!"":/ stated that sea level is pro0ected to rise by about ="cm during the !9st
century with a range of 9= to #=cm. *hese pro0ections have been bac'ed up by recent
satellite data from $P&& +!"":/ showing an annual Artic sea ice reduction of !.: R ".BL
per decade since 9#:C1 with larger decreases in summer of :.> R !.>L per decade.
ountain glaciers and snow cover on average have declined in both hemispheres with
snow cover decline of 9"L since the late 9#B"s in the mid and high latitudes of the
northern .emisphere. aplecroft net 2td +!"":a/ showed that there was a general
increase in precipitation in many regions of the world with a decline only observed over
Northern .emisphere sub,tropical regions while UNEP and UN3&&& +!""9/ described
scenarios of worsening intensity in parts of Africa and Asia. (ome e%treme weather
events have changed in frequency and intensity over the last =" years such as heat
waves1 heavy precipitation events1 intense tropical cyclone activities and incidence of
e%treme high sea,level ma'ing climate change one of the main challenges of today.
1.2.1.3 E62.28i6; 1+,4:1 ,.> 924i:i6,4 i89,6:0
*he impacts of depleting fossil fuels and climatic changes have been a ma0or source of
dispute and are all inter,lin'ed. &limate change impacts will lead to economic losses for
various sectors around the world. According to the (tern ?eview +!""B/ if action is not
ta'en to curb carbon emissions1 climate change could cost between = and !" percent of
the annual global gross domestic product. *hese increases in costs arise from e%treme
weather1 including floods1 droughts and storms.
.ealth impacts as a direct result of climate change are also immense. *he -orld .ealth
@rganisation +!""C/ stated that climate change led to people dying every year from its
side,effects through deaths in heat waves1 and in natural disasters such as floods1 as well
as influencing patterns of life,threatening vector,borne diseases such as malaria and an
increase in food insecurity1 water shortage and other multiple stresses. According to the
9"
$P&& +!"":/ people living in poverty would be worst affected by the effects of climate
change.
@ther stresses such as global economic and political instability are sub0ect to attention.
&onflicts and possible shortages of energy supply have already started to have important
implications. @E&D +!""C/ showed potential water related tensions emerging between
nations that share common freshwater reserves. Access to water1 its allocation and use
are becoming increasingly critical concerns that may have profound consequences for
political and social stability. -ith worldwide demand for energy increasing every day1
dependency on oil imports for most developed countries and therefore energy insecurity
is also becoming an increasing problem. *his results in profound economic and political
implications such as the 9#=; coup in $ran organised by the U( and Britain1 followed by
the (ueD crisis to the 7ulf -ar in 9##9 and the $raq war of !"";. any wars have been
conducted and are still fought all over the world to ensure corporate control over oil.
*ensions are increasing on new discovery of oil. Disputes between countries over oil
reserves represent another potential concern. *he recent e%ample of the ?ussian
e%pedition aiming at strengthening ?ussia5s claim of the oil and gas wealth beneath the
Arctic @cean by planting their country5s flag on the seabed shows the potential e%tent of
the conflict. &ountries bordering the Artic including ?ussia1 the U(1 &anada and
Denmar'1 have launched competing claims to the region.
-ith worldwide demand for energy increasing every day1 the development of new1
clean1 renewable energy sources is critical to EarthHs environment. -orldwide and in the
United <ingdom +U</1 wor' is under way on a variety of potential answers to the
global energy challenge. Energy policy and the potential of renewable energy for the
environment are discussed.
1.2.2 E.+-(/ 924i6/ ,.> 9-209+6:0 23 -+.+A,54+ +.+-(/
1.2.2.1 K/2:2 P-2:2624 ,.> :1+ +.+-(/ 924i6/ 23 UK
Political considerations over the security of supplies1 environmental concerns related to
global warming and sustainability are ma0or political issues and the sub0ect of
international debate and regulation.
99
As a result international treaties such as the United Nations 3ramewor' &onvention on
&limate &hange +UN3&&/ +!""C/ were formed at the United Nations &onference on
Environment and Development. *he treaty aimed to achieve the stabilisation of
greenhouse gas concentrations in the atmosphere at a level that would prevent
dangerous interference with the climate system. Parties have been meeting annually in
&onferences of the Parties +&@P/ to assess progress in dealing with climate change and
to establish legally binding obligations for developed countries to reduce their
greenhouse gas emissions. As a result1 the <yoto Protocol was adopted by &@P,; on
December 991 9##: and finally implemented on 3ebruary 9B1 !""= during the &@P,99.
*his international agreement legally binds most industrialised countries to reduce
emissions of gases contributing to climate change by an average of =.!L below 9##"
levels between the years !""C,!"9!1 defined as the first emissions budget period. *he
<yoto Protocol was the first step towards a truly global emission reduction regime.
.owever1 the commitment period of the <yoto Protocol ends in !"9! and a new
international framewor' needs to be negotiated and ratified to deliver the strict emission
reductions. *he recent &@P,9; held in Bali achieved agreement on a timeline
negotiation on the post,!"9! framewor' as a successor to the <yoto Protocol.
Negotiations will be held in Poland during December !""C. *o date1 9:C nations have
ratified the treaty1 only the U(A and <aDa'hstan have signed but not ratified the <yoto
Protocol. Under the international <yoto Protocol and European Union agreements the
U< must reduce its baseline greenhouse gases emissions by 9!.=L by !"9!.
3urther steps in this direction were added by the U< 7overnment which has set its own
international and domestic energy strategy to reduce carbon emissions in the Energy
-hite paper +!"":/1 the &limate &hange Bill +!"":/ and more recently in the &limate
&hange Act !""C. *he Energy -hite PaperF Meeting the Energy &hallengeN addresses
a commitment to reduce carbon dio%ide emissions1 with consequent constraints to its
energy policy. A mandatory B"L reduction in &@
!
emissions below the 9##" baseline
by !"="1 with an intermediate target of between !BL and ;!L by !"!"1 was published
within the &limate &hange Bill by DE3?A +!"":/. A review of the &limate &hange Bill
published by DE3?A +!""Cb/ resulted in the recently transition to the &limate &hange
Act !""C committing the U< to cut emissions by C"L by !"=". Along with this new
reduction1 the Energy -hite Paper outlines the importance to ensure secure1 clean and
affordable energy. *he (cottish E%ecutive +!"";/ agreed a more stringent overall target
compared to U< with its own energy policy aspiring to meet >"L of its energy through
9!
renewables by !"!". 3urther steps in support of this aspiration were proposed by the
(cottish 7overnment +!"": Q !""C/ such as the introduction of the ?enewables
@bligation (cotland +?@(/ providing an incentive to supply higher levels of renewable
energy and encourage developers to bring forward new renewable energy schemes and
the &limate &hange +(cotland/ Bill setting a C"L beneath 9##" baseline by !"=" target1
with an intermediate target of ="L by !";".
1.2.2.2 R24+ ,.> 9-209+6:0 23 -+.+A,54+ +.+-(/
Ensuring a secure1 clean and affordable supply of energy is essential towards the
sustainable development of a modern society. ?enewable energies are potential
solutions to provide energy with Dero or almost Dero emissions therefore reducing local
and global atmospheric emissions. (olar energy is the most abundant energy resource
on earth and is available for use. According to the -orld Energy &ouncil +!"":/ the
annual solar energy intercepted by the surface of the Earth is appro%imately 9C"P-1
therefore1 more than :=1""" times the world5s total annual primary energy consumption
of 9>*-h. @ther ma0or renewables were estimated at B9*- such as wind1 biomass1
hydro and geothermal. *here are therefore plentiful sources of energy that can meet the
present world energy demand with appropriate technology. An increasing share of
renewable energy in the energy portfolio would offer a certain degree of
manoeuvrability against sudden fossil fuel energy price increases and therefore create a
secure long,term sustainable energy supply.
,'enewable so$rces of ener!y ha-e a considerable potential for increasin! sec$rity of
s$pply in .$rope/ De-elopin! their $se0 howe-er0 will depend on extremely s$bstantial
political and economic efforts/ 123 &n the medi$m term0 renewables are the only so$rce
of ener!y in which the .$ropean #nion has a certain amo$nt of room for manoe$-re
aimed at increasin! s$pply in the c$rrent circ$mstances/ We cannot afford to ne!lect
this form of ener!y/4 7reen paper on the security of energy supply +!""9/.
-ith the recent energy policies and protocols described in section 9.!.!.9 and based on
energy policy goals of the Energy -hite Paper1 the role and prospects for renewables
appear to be good. ?enewables can enhance diversity in energy supply mar'ets and
secure reliable and sustainable energy supplies. *hey can therefore reduce occurrence of
energy poverty increasing with energy pricing and reduce atmospheric emissions.
9;
?enewables can also promote competitive new mar'ets in the U< and beyond to meet
specific needs for energy services.
According to aplecroft 2td +!""B/1 while emissions of developed countries are li'ely
to stabilise1 emissions from developing countries continue to rise steadily and are
e%pected to equal those of developed countries. *hese developing countries often
possess large amounts of renewable energy potential especially for solar applications.
*he 'nowledge transfer of such technologies to help developing countries reduce
emissions as their economies grow is vitally important to their economic development.
?enewables such as solar1 wind1 hydropower and biofuels are potential candidates to
meet global energy requirements in a sustainable way. 3igure 9,> shows a comparison
of world wide and EU installed capacities of three intermittent renewables) namely
photovoltaic +PO/1 solar thermal and wind power. *hese data include &hina which
dominates the solar thermal world mar'et with over B"L of the global installed
capacity.
93700
128000
56300
16176
7800
4500
0 20000 40000 60000 80000 100000 120000 140000
$in% #ower
&olar '(er)al
#(oto*oltaic
+$
!"
#orld
Fi(. 1)#* +,- 2''& W2-4> ,.> EU i.0:,44+> 6,9,6i:i+0 23 024,- PV; 024,- :1+-8,4 ,.> Ai.> 92A+- B
D,:, 3-28 :1+ REN21 (2''&) ,.> EC (2''&)
*he published comparison of PO1 solar thermal and wind capacities loo's different if
European data are considered. -ind power is the leader of the three technologies and is
already a mar'et reality. (olar thermal is a strong second place with many applications
for the direct use of solar thermal energy in space heating1 and cooling1 water heating1
crop drying and solar coo'ing. $t is most used for water heating hot water being a
9>
necessity in today5s lifestyle. $n some countries1 residential and commercial water
heaters consume a substantial portion of the average utility bill and are dependent
particularly upon the combustion of fossil fuels.
1.3 P-209+6: 23 SWH i. UK ,.> S62:4,.>
1.3.1 E.+-(/ ,.> E.vi-2.8+.:,4 S6+.+ i. S62:4,.>
An understanding of current energy supply and demand of the U< and (cotland is vital
to develop an appreciation of the opportunities of (-.. (cotland consumes 9:=*-h of
delivered energy consuming #.9L of U< energy with only C.=L of the U< population
showing greater energy consumption per capita than the average for the U< in total.
*he main end,use sectors can be divided in fourF domestic1 transport1 industry and
services. *he domestic sector is the largest consumer with an energy split of ;>L. A
more detailed picture of the energy used in the domestic sector show that 9".9L of the
total consumption in the U< is used by (cotland1 despite the fact that it represents C.=L
of the U< population.
*he colder climate in (cotland is a ma0or factor e%plaining this figure. (cotland
typically has 9"L more degree,days than the U< average consequently affecting the
requirements for space heating. *he main fuels used in the domestic space and water
heating account for over C"L of a householdHs energy consumption according to the
(cottish E%ecutive +!""B/ are gas1 along with coal and oil.
*he domestic sector consumes =B*-h in total. $f the C"L accounting for space and
water heating were to be replaced by renewable energy1 then >>.C*-h of energy would
be carbon neutral which would result in a decrease in &@
!
emission of 9!.9>7t&@
!
Jyear
based on the (cottish E%ecutive +!""B/ data.
*here is vast scope in addressing the energy requirement1 particularly for hot water from
renewable sources.
1.3.2 S24,- 8,-C+: ,.> 9-209+6:0
&ertain parameters play a 'ey role in the behaviour of (-. performance1 such as
geographic location1 solar potential1 types of (-. and others which could e%plain why
9=
they are more developed in some areas. .owever1 their potential are nowadays valuable
in any location.
*here was a solid growth of >>.;L in the European solar thermal mar'et in !""B
corresponding to more than ; million m
!
of newly installed surface of solar thermal
collectors operating in Europe. *he greatest growth occurred in the 3rench mar'et with
a growth rate of over C;.9L1 while the mar'et is still dominated by 7ermany with a
growth of !:L despite the consecutive decrease of investment subventions granted in
the framewor' of the 7erman mar'et stimulation program as stated by Eur@bserv5E?
+!"":/. *he European mar'et is dominated by 7ermany1 7reece1 Austria and 3rance
with over ::L of the mar'et represented by these countries. *he EU has reached an
average capacity in operation of ;".C '-thJ91""" inhabitants. *his indicator ma'es it
possible to directly compare different countries involvement in developing solar thermal
applications as shown in 3igure 9,=. National values range from =99 in &yprus and !>"
in Austria to less than 9 '-thJ91""" inhabitants in some Baltic countries. *he U<
position in the mar'et is towards the lower end with only !=!19B" m
!
of solar thermal
collectors in operation in !""B which accounts for only 9.!L of the mar'et. *his
indicator shows that the margin of progression is still very large in 7ermany and that
this potential has barely been tapped in 3rance1 (pain and the U<. 3igure 9,= also
shows that despite the higher available solar radiation of (pain1 3rance and the U< there
has been lower involvements in developing solar thermal compared with other
European countries with lower solar radiation availability.
9B
0
100
$00
300
%00
&00
'00
(00
800
0 1 $ 3 % &
,*ailabilit- of solar ener.- on an (ori/ontal 0lane1 2$(")
2
"%a-
&
$
3
i
n
s
t
a
l
l
e
%
c
a
0
a
c
i
t
i
e
s
1
)
2
"
1
0
0
0
c
a
0
i
t
a
s
"nited )ingdo
*erany
*ree+e
,ran+e
-pain
.ustria
/enar0
-weden
1alta
2yprus
Fi(. 1)"* E7-29+,. 024,- 8,-C+:; D,:, 3-28 E7-O50+-vDER (2''&) ,.> E7-29+,. C288i00i2. (EC)
(2''&).
&yprus is the European leader when it comes to ability to capture solar energy with over
#"L of the buildings equipped with solar collectors1 which is twice as much as Austria
and 7reece. .owever if the whole EU was at the same level per capita as Austria today1
the annual EU mar'et would be over 9:" millions m
!
1 with a capacity in operation of
about 99# 7-th. *his capacity would effectively substitute large amounts of
conventional fuels such as oil and gas or electricity and put Europe ahead of &hina in
terms of the global mar'et. $n their report ?EN!9 +!"":/ declared that the global mar'et
is dominated by &hina with annual domestic sales of well over 9" millions m
!
with the
EU holding only a small fraction of the world mar'et as second.
*his section shows that the U< position in the mar'et is towards the lower end and that
the U< potential for developing solar thermal applications in U< has barely been
tapped.
1.3.3 N++> 32- ,332->,54+ SWH ,.> >,:, 82>+44i.( (9-209+6: 23 024,- A,:+-
1+,:i.( 32- S62:4,.>)
$n a period of rapidly growing deployment of sustainable energy sources the
e%ploitation of solar energy systems is imperative. (olar water heating systems have
been well developed in (cotland with a large choice of systems available to suit many
applications. *he higher annual heating loads faced by (cotland allows longer and
9:
&#
+,
R
C4
,5
&$
FR
6N
57
8R
greater savings using solar thermal power ma'ing this type of technology more effective
in northern countries. *he viability of such systems has been proven as the number of
commercial manufacturers increased and the number of installers accredited by the
(cottish &ommunity and .ouseholder ?enewables $nitiative , (&.?$ +!""C/
registration lists increased from ;; in arch !"": to B: in 4uly !""C or =9L more
installers of (-..
*he Energy (aving *rust +!""=/ estimated commercial prices for an installed (-. in
the range of T!""" to T;""" depending on the type of collector and the capacity of the
system required1 with an average lifespan of !" years. According to the (cottish (olar
Energy 7roup , ((E7 +!""C/ about half of the hot water energy demand over the year
could be provided with an average collection efficiency of about >"L for a !,=m
!
collector based on the fairly cloudy (cottish climate getting 9"""'-hJm
!
. *here is then
a niche mar'et for lower cost (-. with a lifespan and annual thermal efficiency equal
or superior to current designs on the mar'et.
(olar water heaters also have the benefit of reducing impacts on the environment by
reducing carbon dio%ide emissions by ".>,".:= tonnes per year depending on the fuel
replaced according to the Energy (aving *rust +!""=/. .owever1 as energy efficiency
and better design of collectors can reduce energy use) the energy e%pended during the
manufacturing process1 also called Membodied energyN1 ta'es a more significant part of
the real energy savings and paybac' time. $t is therefore very important to consider the
whole energy life,cycle of the (-. as it allows for a better assessment of its
environmental impact.
&ertain parameters play a 'ey role in the success of (-. technologies such as their
monetary and energy pay bac' period. .owever1 their potential use or application are
valuable at any location and in any application as buying coal1 oil or gas fuelled
substitutes do not offer the customer any financial reward. $n order to encourage the
installation of renewable devices1 a ;"L grant is available for solar thermal
technologies by the (&.?$ +!""C/ thereby increasing the (-. return, on,investment.
*he result of many years of global research on solar water heating systems has outlined
the promising approach of integrated collector storage solar water heaters +$&(,(-./
in cold climates. $&(,(-. are the simplest type of (-. on the mar'et and a popular
9C
choice when choosing a design as they possess several advantages. *hese systems
incorporate the solar collector and thermal storage tan' in a single unit. *heir simplicity
of construction and installation ma'e them affordable with low costs. @ther advantages
such as their aesthetic value1 low maintenance1 ease of integration in roof structures and
good overall performance in overcast s'y conditions ma'e them suitable for northern
hemisphere applications where this research is ta'en. *his calls for a need to estimate
the potential of $&(,(-. for the (cottish climate.
1.# T1+ P-+0+.: R+0+,-61 P-2M+6:
1.#.1 P-2M+6: ,i8
&nte!rated collector stora!e solar water heaters5 *o study and analyse the performance
of a newly developed $&(,(-. for (cottish weather conditions1 optimise its
performance1 model its laboratory and field performance with its environmental impacts
and analyse its integration into buildings and benefits of such a heating system1 for the
primary purpose of proposing a feasible $&(,(-. prototype.
1.#.2 O5M+6:iv+0
*he ob0ectives of the pro0ect can thus be summarised as followsF
9. *o study the possible improvement of $&(,(-. under (cottish climatic
conditions through the literature review) considering the large potential mar'et
for solar thermal systems.
!. 3ollowing the analysis underta'en in +9/1 to test in laboratory conditions the
newly developed $&(,(-. in order to accurately compare its performance with
a previous study carried out for $&(,(-..
;. *o carry field tests of the same $&(,(-. to evaluate its performance under real
(cottish weather conditions.
>. *o develop a thermal macro,model able to compare the temperature variation in
different $&(,(-. materials) including internal temperature and e%ternal
weather conditions for a given aspect ratio and to evaluate the performance of
this $&(,(-. for laboratory and field conditions.
=. *o model the $&(,(-. using 3luent B.! &3D +&omputational 3luid Dynamics/
software to estimate an optimum fin spacing to ma'e further improvements on
the current $&(,(-..
9#
B. *o determine the monetary and energy paybac' considering the whole life
energy of the $&(,(-. using the thermal model previously developed to
determine its feasibility.
:. *o show the integration of $&(,(-. into buildings and its potential benefits for
builders and households.
1.#.3 O7:4i.+ 23 :1+ :1+0i0
&hapter 9 gives the introduction to the thesis1 starting from a very broad sub0ect of
energy it narrows down to renewable energy then to solar energy and then further to
solar water heater and more particularly to $&(,(-.. $t also provides the bac'ground1
statement and ob0ectives of the present research problem.
&hapter ! deals with the literature review covering the following topicsF +a/ solar hot
water systems1 thermal collectors and solar water heater parameters1 +b/ modelling the
solar water heater through a macro,model +thermal model/ and micro,model +&3D/1 +c/
life cycle assessment +2&A/1 weather conditions in (cotland and building integrated
(-..
&hapter ; deals with laboratory and field e%periments. *he methodology used is
introduced followed by the test regime1 measurement and data collection. $t also gives
laboratory e%perimental set,up and results with a comparison with previous wor'.
E%perimental set,up for the field tests and results of e%periments are also provided.
3inally an account of uncertainties and propagation of errors through both e%perimental
set,ups is given.
&hapter > presents the modelling procedure of the $&(,(-.. *he thermal model results
and e%perimental data are inter,compared for laboratory and field tests by means of
various statistical indicators while the micro model is used to determine optimum fin
spacing.
&hapter = performs a 2&A of different types of $&(,(-. for use in (cottish conditions
including the e%perimentally tested system using the previously developed macro
model. Using modelled bul' water temperature in the collector1 performance1 monetary1
energy and carbon savings comparison of three different $&(,(-.s was provided by
changing their material and thic'ness.
!"
&hapter B demonstrates the potential for $&(,(-. to be integrated into buildings and
the different considerations to be ta'en into account when choosing such (-.. Benefits
in installing such heaters and possible installations scheme into buildings are also
provided.
&hapter : draws up important conclusions from each aspect of the presented wor' and
also discusses the potential for future wor'.
1." C2.647>i.( -+8,-C0
*his chapter provides a detailed introduction to energy1 its various resources1 associated
issues and challenges and its current and future status. (ignificance of renewable energy
resources was established and various such forms discussed. (olar energy was
addressed in detail as this forms the basis of the present pro0ect. (-. and underlying
mar'et prospects for U< and (cotland were briefly discussed. *he need for new (-.
technologies was also emphasised. *his was followed by stating the research problem1
its aims and ob0ectives and finally the structure of the thesis.
CHAPTER 2 REVIEW OF THE RELEVANT LITERATURE
$n the conte%t of the present research pro0ect1 investigating solar water heaters has three
core aspects. *he first is studying the prospects and diversity of solar water heaters. $t
also includes a guidance to improve the configuration of solar water heaters and
e%perimental procedures. *he second aspect deals with the modelling methodology
!9
itself. *he third involves solar water heating in (cotland1 commercial designs and
weather considerations.
*his chapter is accordingly sub,divided into three sections. (ection !.9 reviews the
different type of solar water heaters1 their construction and the different influencing
characteristics. (ection !.! provides a comprehensive yet summarised coverage of
different methods of modelling solar water heaters. (ection !.; reviews the integration
of solar water heaters into buildings under (cottish weather considerations.
2.1 S24,- A,:+- 1+,:+-0
2.1.1 S24,- 12: A,:+- 0/0:+80
(-. systems are generally very simple using only sunlight to heat water. A wor'ing
fluid is brought into contact with a dar' surface e%posed to sunlight which causes the
temperature of the fluid to rise. *his fluid may be the water being heated directly1 also
called a direct system1 or it may be a heat transfer fluid such as a glycolJwater mi%ture
that is passed through some form of heat e%changer called an indirect system. *hese
systems are used for domestic hot water or space heating. *hey can be classified into
three main categoriesF +a/ Active systems which use pumps to circulate water or a heat
transfer fluid1 +b/ Passive systems +or *hermosyphon systems/ which circulate water or
a heat transfer fluid by natural convection and +c/ batch systems using a tan' directly
heated by sunlight.
Active systems are also called forced circulation systems and can be direct or indirect.
A system is called direct when the pump is used to circulate drin'ing water between the
hot water storage tan' and the collector. $n an indirect system the pump is used to
circulate a heat transfer fluid1 either water or an antifreeDe mi%ture1 between the
collector and a heat e%changer. .eat is transferred by the heat transfer fluid from the
collector to the drin'ing water through the heat e%changer in the storage tan'. *he use
of a heat e%changer reduces significantly the system efficiencies.
Passive systems simply circulate water or a heat transfer fluid by natural convection
between a collector and an elevated storage tan' +above the collector/. *he principle is
simple1 as the fluid heats up its density decreases. *he fluid becomes lighter and rises to
the top of the collector where it is drawn to the storage tan'. *he fluid which has cooled
down at the foot of the storage tan' then flows bac' to the collector. *his system can be
!!
operated as a direct or indirect system1 whereby a heat e%changer can allow the use of
antifreeDe to prevent frost damage during colder winter months. (ince no pumps or
controls are necessary1 system installation costs are reduced.
3inally1 batch systems have combined collection and storage functions. *hey are the
simplest type of (-. with flow of water throughout the system being set off whenever
there is water demand. .ot water is drawn to the top of the collector and replaced by
cold water at the bottom. $&( systems can supply preheated water directly to the end use
appliance or more commonly to a conventional water heater tan'. Depending on the
system1 there is no requirement for pumps or moving parts. .owever1 the system is
prone to freeDing1 therefore draining water in the collector and pipe is advised during
cold periods. *hese systems could also be integrated in to active or passive systems
depending on the design of the house.
Active and passive (-. systems remain more common than batch (-. systems. *o
date1 a small amount of wor' has given adequate consideration to the integration of
batch systems1 especially within domestic buildings. Batch (-. are normally used for
low temperature applications such as domestic or light industry due to the low capital
outlay. *his ma'es it possible to achieve realistic capital cost paybac's during the
lifetime of the systems. *he three common types of built,in,storage solar collectors are
described in section !.9.;.9 below.
2.1.2 S24,- :1+-8,4 6244+6:2-0
*here are three common designs of solar water heaters. $n order of decreasing cost they
areF concentrating collectors1 evacuated tube collector and flat,plate collectors. *he
choice of collector is determined by the heating requirements and the environmental
conditions in which it is employed.
&oncentrating collectors are mostly used in high temperature solar applications such as
steam production for the generation of electricity. *hey trac' the sun during the daytime
and use reflectors to concentrate the sun5s energy onto the absorber. *hese systems
however are most appropriate for applications where direct solar radiation is high. *hey
are best suited to climates that have a high percentage of clear s'y days and are
therefore not so common in (cotland
!;
Evacuated tube collector systems are principally used for domestic water heating
purposes as suggested by orrison et al. +!"">/. *he solar absorbing surface is placed
in the centre of a vacuum tube which minimises convection losses thereby increasing
efficiency. .owever1 improved efficiency is reflected in their price. *hey are most
suited to e%tremely cold ambient temperatures or in situations of consistently low,light.
*hey are also used in industrial applications1 where high water temperatures or steam
need to be generated where they become more cost effective.
3inally1 flat,plate collectors are used e%tensively for domestic hot water heating
applications due to performance advantages in relation to cost. *hey collect both direct
and diffuse radiation. *heir simplicity in construction reduces initial cost and
maintenance of the system. Although flat plate collectors are outperformed by
evacuated tubes collectors they are generally a more cost effective solution in most
climates and for the ma0ority of domestic hot water services. *here are variants of flat
plate collectors such as serpentine tubes1 parallel tubes or integrated collector storage
solar water heaters +$&(,(-./. *he simplicity in construction and low cost of $&(,
(-. ma'e them an attractive and popular choice when choosing a (-.. A more
detailed picture of these systems is of interest and is presented in the following section.
1.1.1.1 B-i+3 1i0:2-/ 23 ICS)SWH
$&( are one of the oldest solar water heater designs still in use. *he first record of $&(,
(-. systems was in the United (tates of America +U(A/ in the late 9C""s where
sufficient hot water for showering by the late afternoon on clear days was achieved.
$&(,(-. was the first patented and commercially manufactured (-. in 9C#9 under
the trade name U&lima% (olar -ater .eater5 designed by <emp +9C#9/. *he first
detailed study on $&( tan' systems was carried out in 9#;B by Broo's +9#;B/ at the
University of &alifornia Agricultural E%perimental (tation1 U(A. Although $&(,(-.
were a growing mar'et1 the discovery of natural gas and oil postponed investigations of
$&(,(-. until 9#:". After the oil crisis of @ctober 9#:;1 a renewed worldwide interest
in solar energy was revitalised including a renewed interest in $&(,(-.. *he first
developments in solar energy were underta'en by passive solar enthusiastic individuals1
0oined later by universities and research institutions. *oday solar heating and $&(,(-.
is a very popular area of research worldwide encouraged by global initiatives to
promote renewable energy technologies to limit greenhouse gas emissions.
!>
A long review of technical developments of $&(,(-. to benchmar' the performance
of all heaters using common performance indices was carried out by (myth et al.
+!""B/. -orld,wide popular $&(,(-. were reviewed as well as the regional1
environmental and economical factors that have influenced solar heaters designs. Based
on this study1 a review of the 'ey components of $&(,(-. are presented in the
following section
1.1.1.2 K+/ 62892.+.: 23 ICS)SWH
*here are many $&(,(-. designs but generally all consist ofF +a/ a storage tan'1 +b/ a
heat transport fluid flowing through the collector to remove heat from the absorber1 +c/ a
flat plate absorber which intercepts and absorbs the solar energy1 +d/ a heat insulating
bac'ing with baffle and insulation1 +e/ a transparent cover+s/ that allows solar energy to
pass through but reduces heat loss from the absorber.
S:2-,(+ :,.C
(iDe and shape of the storage tan' have a significant effect on the performance of the
(-.. (everal designs of $&(,(-. have been studied over the years. *he first
manufactured commercially U&lima% (olar -ater .eater5 designed by <emp +9C#9/
was composed of four small blac' painted cylindrical storage tan's interconnected and
mounted in a glaDed wooden bo% insulated with felt paper. 3igure !,9 shows the
advertisement of this (-. in 9C#! with a purchase price of V!=.
!=
Fi(. 2)%* A>v+-:i0+8+.: 32- :1+ 64i8,x 024,- A,:+- 1+,:+- i. 1<$2; B7::i ,.> P+-4i. (1$<')
$n 9#":1 .as'ell +9#":/ patented a shallow rectangular tan' design increasing the
surface area allowing a faster warm,up in the morning and hotter water on cloudy days.
*anishita +9#:"/ indicates that in 9#>: Wamamoto designed the first 4apanese
commercial (-. made of a stainless,steel piping. *he same authors states that later in
9#="s the closed membrane1 'nown as plastic bag $&(,(-.1 was successfully
developed and commercialised in 4apan. $t consisted of a large rectangular
polyvinylchloride bag1 with a blac' membrane as a bottom surface and a transparent
membrane on top. Another type of storage tan' was tested by &hinnappa and
7nanalingam +9#:;/ using square coil pipes.
7arg +9#:=/ conducted the first year long performance study of a rectangular design
$&(,(-. followed by uneer and .awas +9#C>/ and uneer +9#C=/ who proposed a
simple rectangular design $&(,(-.. uneer and .awas +9#C>/ compared $&(,(-.
and thermosyphon collector showing that the simple and low cost $&(,(-. narrowly
outperformed the thermosyphon collector. $n his later study1 uneer +9#C=/ studied the
effect of the collector depth on performances. ?esults showed that an increase of !cm in
depth enhanced the storage volume which made the system operate at lower
temperatures and so the losses are limited resulting in an increase of efficiency of CL.
(everal studies carried by Ecevit et al. +9#C#/1 Pra'ash et al. +9##!/ and <aushi' et al.
+9##>/ suggested that a triangular storage tan' can perform better due to increased
!B
natural convection. (o'olov and Oa%man +9#C;/ showed an efficiency of =;L for their
triangular $&(,(-. using a baffle plate to separate the absorbing plate from the storage
volume. Ecevit et al. +9#C#/ studied the effect of different volumes in triangular
collectors showing good overall performance in all designs. A side by side comparative
e%perimental study of a rectangular and triangular shaped $&(,(-. was carried out in
9##> by (oponronnarit et al. +9##>/ under identical conditions. *he thermal efficiencies
of the triangular and rectangular were found to be B;L and =#L respectively. ohamad
+9##:/ introduced a simple thermal diode to prevent reverse circulation at night time in
his triangular $&(,(-.. *he storage tan' efficiencies improved by 9=.;L using the
diode. $n the same year1 an innovative pyramid shape was developed for Egyptian
weather conditions by Abdel,?ehim +9##:/ providing temperature range of >",B" &.
$n his subsequent studies uneer et al. +!""Ba/ launched a new (-. design with the
introduction of fins. *he report details side,by,side testing of a system using a novel
rectangular finned $&( against a plain rectangular non,finned $&( system installed at the
Din *e%tile ills1 Bhai Pheru1 2ahore1 Pa'istan. .ot water demand for this study was
C"S& for the dyeing process. *heir research concluded that on clear days the non,finned
design achieved ma%imum temperatures in the region of :"S& at best1 whilst the finned
heater regularly achieved water temperatures higher than :"S&1 sometimes reaching
temperature in the region of C"S&. *he solar fraction of the finned heater was found to
be :;L compared to B;L for the non,finned system. Using the same design1 7rassie et
al. +!""B/ performed laboratory e%periments in order to optimise the collector design
under (cottish weather conditions.
A recent study developed by &ruD et al. +!""!/ introduced a trapeDoidal storage tan'
design for editerranean climates inducing a thermal stratification in the water store.
*he research indicated that ;"L to :"L of the daily load could be covered by the $&(,
(-..
H+,: :-,.092-: 347i>
*he collector fluid in an $&( system is potable water. -ater is the cheapest1 most
readily available and thermally efficient fluid but does freeDe and can cause corrosion.
A drain,bac' system can empty the collector to a bac',up hot water cylinder at times of
low temperatures and negative heat gains. Using softened water will help e%tend
operational lifespan.
!:
A502-5+- 94,:+
*he absorber plate material is a main parameter in an $&(,(-. as it has the principal
function to absorb solar radiation and transfer heat to the storage fluid. As mentioned by
(myth et al. +!""B/1 wood was one of the first materials used1 later systems using plastic
were developed and then replaced by new resistant material such as copper1 aluminium1
stainless steel and polymers with painted absorber surfaces. *he use of polymer
absorber has been studies by Oan Nie'er' et al. +9##B/ for parallel tubes collectors in
(outh Africa. Polymers1 compared to metals have a lower thermal conductivity and may
have high temperature limitations1 but they are relatively ine%pensive and can offer
good freeDe protection. *ransfer of absorbed solar energy to the storage fluid depends
on the collector materials thermal conductivity. aterials such as steel1 aluminium or
copper are more frequently used with copper and aluminium as the most effective.
Another parameter has to be loo'ed at when choosing the collector material. $t must be
of satisfactory strength and durability1 lend itself to manufacturing constraints and be of
an appropriate cost. &apral,Aluminium,2td +!""B/ stated that in most surroundings1
aluminium has a very high resistance to corrosion. Aluminium is covered by a thin
effective coating of impermeable o%ide which protects from further o%idation. .owever1
this material should be used by itself as in most combinations with other metals
galvanic corrosion occurs. (everal studies carried out by (tic'ney and Nagy +9#C"/1
Burton and Eweig +9#C9/1 Bainbridge +9#C9/ and *iller and -ochatD +9#C!/ and
recently *ripanagnostopoulos and Wianoulis +9##!/ suggested the use of an optional
selective coating for absorber plate. &oatings can be applied such as electroplated blac'
nic'el1 blac' chrome1 and copper o%ide or anodiDed aluminium. All studies showed that
coatings enhanced the heat transfer and protected the absorber plate. *o ma%imiDe
efficiency1 their absorptivity should be at least ".# and emissivity less than ".9.
I.074,:i2.
Back heat losses
Early systems were simply e%posed tan's of water left out to warm in the sun. *hese
systems suffered substantially from ambient heat losses1 particularly at night,time and
during non,collection periods. -ide theoretical and e%perimental studies on an $&(
(-. undertoo' by 7arg and ?ani +9#C!/ in $ndia proposed a baffle plate to reduce
night time losses. &ollector performance improvement was found during night time in
addition to during the day. *hey also observed that by using an insulation cover during
cooling hours1 the collector performance could be improved by :"L. <aushi' et al.
!C
+9##=/ carried out a comprehensive evaluation of a triangular $&( system with a baffle.
*hey concluded that the thic'ness and the material of the baffle had little impact on the
system performance and that the inclusion of the baffle had a significant benefit on the
system performance during non collection periods. Smyth et al. (2003) and
*ripanagnostopoulos and (ouliotis +!"">/ proved that collector performance
improvement is closely lin'ed with collector bo% efficacy. *he baffle should be
weather,proof1 guard the absorber1 insulate from the environment1 reduce convection
and conduction losses and provide a sturdy housing to hold the components. *he
collector bo% should also be designed to allow for roof integration. 7as'ets and seals
are further important parameters in maintaining a weather,tight seal between
components whilst allowing thermal e%pansion of the components. -ithout them1
moisture would fog the glaDing thereby reduce transmittance and possibly damage the
insulation and the absorber coating.
uneer et al. +!""Ba/ and 7rassie et al. +!""B/ e%plained that the insulation of the
storage tan' was essential to protect from conduction and convection heat losses. Both
authors proposed an easy solution using "."=m of glass wool insulation on all sides of
the collector housed in a wooden bo%. $nsulation is usually between !=mm and 9=" mm
and ideally would have reflective foil to reflect thermal radiation bac' to the absorber.
&ommon insulation materials are high,temperature resistant fibreglass batting1 semi,
rigid board or mineral wool. $nsulation should have a low thermal e%pansion coefficient
and should not melt or outgas at collector stagnation temperatures.
Top heat losses
Another way to decrease heat losses is the addition of a lid on top of the (-..
oveable insulated lid were introduced by Baer +9#:=/ resulting in a significant
improvement in the heat retaining properties of the $&(,(-.. Early attempts by
c&rac'en +9#:C/ using materials such as *ransparent $nsulating aterial +*$/ in
order to minimise heat losses caused decreases in the system efficiency. .owever1
recent studies of ?eddy and <aushi'a +9###/ and .enderson et al. +!""B/ showed that
the use of new *$ has led to greater collection performances.
T-,.09,-+.: 62v+-
!#
Bishop +9#C;/1 Duffie and Bec'man +9##9/ and (myth et al. +!""B/ established that the
use of one or more transparent covers reduced thermal losses by radiation and
convection by restricting air movement. 3urthermore1 covers protect the absorber and
selective coating +if applicable/ from harm1 dirt1 rain and other environmental factors.
7lass or plastic sheets are the most common materials used for covers. *he same
authors indicated that the most important property required of the transparent cover
material is high transmittance of solar radiation1 as any loss in transmittance will lead to
a direct reduction in collection efficiency. 7lass is considered more appropriate than
plastic as it is more durable and does not degrade in sunlight or at high collector
temperatures. *he only disadvantages of glass material are the more e%pensive costs and
the possibility of impact damage. Duffie and Bec'man +9##9/ suggested the use of
additional covers to reduce convective losses. .owever1 increase in glaDing would result
in higher transmission losses. Bainbridge +9#C9/ showed that using selective coating
glaDing could result in enhancing greenhouse effect and maintain high transmission.
7laDing with very high solar transmittance1 such as A? coatings used in PO application
by 7ombert et al. +9##C/1 could result in cost effective alternative for solar thermal
application.
(tudies conducted by uneer et al. +!""Ba/1 .enderson et al. +!""B/ and 7rassie et al.
+!""B/ using a >mm thic' glass sheet to cover the top of the bo% collector and
maintaining an air gap of ;=mm established good collector performance.
1.1.1.3 A>v,.:,(+0 23 ICS)SWH
?ectangular $&(,(-. have several advantages over other systems. *he direct contact
of the wor'ing fluid with the absorber plate increases the system efficiency by
eliminating the need of a heat e%changer used in most heating systems. *he absence of
conduits and connecting pipes to the storage tan' results in eliminating the bul' heat
losses appearing in other type of systems. @ther advantages of $&(,(-. are the
effective use of direct and diffuse radiation as well as the absence of solar trac'ing
requirements and minimal maintenance as well as their low cost and good overall
performance. *hey can be designed such that they seamlessly integrate in roof
structures giving them an additional aesthetical and architectural value.
;"
2.1.3 S24,- A,:+- 1+,:+- 9,-,8+:+-0
1.1.1.# O9:i878 ,.(4+ 23 i.64i.,:i2.
Different studies show different optimum angle of inclinations for Edinburgh city.
Duffie and Bec'man +9#:>/ stated an optimal angle to be ".# times the latitude of the
location. *his asserts the collector tilt can ta'e values from "S at the equator to #"S1 in
Polar regions and would be ="S in Edinburgh. ?eiss and Bainbridge +9#:C/ in their
research showed that systems in higher northern latitudes are often optimised for winter
collection and are therefore tilted = to 9" more than the angle of latitude. Another
research carried out by uneer et al. +!""Bb/ based on !: year irradiance data for
Edinburgh established that higher irradiation occurred at ;"S inclination for a south
facing wall. .owever1 irradiation is not the only factor influencing a solar water heater.
A more recent study conducted by .enderson et al. +!"":/ showed that an $&(,(-.
would perform better in Edinburgh at ="S compared to ;=S because of its higher Nusselt
number value. *he study covered the behaviour of the water cavity collector for
different inclination angles. *he effects were evaluated using two modes of analysis in
con0unction) &omputational 3luid Dynamic +&3D/ software and e%perimental testing.
*he authors found another advantage of having an angle of ="S compared to ;=S would
be the reduction of heat losses for higher inclination angles. 3or regions with lower
latitudes1 increasing the tilt angle initially improves the thermal efficiency. .owever1
any successive increase in tilt would reduce the efficiency as the solar gain value would
become very low. Based on these past studies1 an inclination angle of >= south was
implemented.
1.1.1." S:-,:i3i6,:i2.
*hermal stratification is an important parameter in the efficient operation of solar
collector systems. (everal authors XOan <oppen et al. +9#:#/1 3urbo and i''elsen
+9#C:/ and Oan Ber'el +9##B/Y highlighted the importance of stratification in detached
vertical storage tan's of solar water heating systems. *he merits of stratification were
then described by .ollands and 2ightstone +9#C#/ who demonstrated that (-. systems
operating with thermally stratified storage tan's have substantially improved
performance1 particularly at low flow rates. &omparisons between stratified and mi%ed
tan's were carried out by Oan Ber'el +9##B/ and &ristofari et al. +!"";/. Oan Ber'el
+9##B/ demonstrated that mi%ed water tan's performed disappointingly in terms of
output and collection efficiency compared to a well stratified tan'. &ristofari et al.
;9
+!"";/ studied the performance of a classical solar water heating installation with a flat
plate thermal collector and water tan' storage for storage in low flow conditions
installed at a editerranean site1 A0accio1 &orsica1 3rance. ?esults showed that a
stratified tan' has higher performance than a fully mi%ed tan' with monthly mean
temperatures varying from 9#.;& to B#.:& for a fully mi%ed tan' and from !".9& to
:9.!& for a high thermal stratified tan' illustrated in 3igure !,!.
Fi(. 2)&* M2.:14/ 8+,. :+89+-,:7-+ 23 :1+ 0:2-,(+ 347i> 32- 3744 8ix+> :,.C ,.> 0:-,:i3i+> :,.C;
C-i0:23,-i +: ,4. (2''3).
*hermal stratification in $&(,(-. has been observed in much research but poorly
analysed. @nly four studies on stratification in $&(,(-. were reported by uneer
+9#C=/1 Burns et al. +9#C=/1 (myth et al. +9###/ and .enderson et al. +!""B/. *he
uneer +9#C=/ study involved change of vessel depth and showed no significant
influence of performance with change of depth. Burns et al. +9#C=/ evaluated an
insulated single glaDed rectangular wooden bo% containing a !="l blac' painted drum at
a tilt angle of >" in (an Diego1 &alifornia1 U(A. .is research concluded that thermal
stratification was very important in the efficient operation of $&(,(-. systems1
however1 it was suggested that these systems should only be used for preheating. *his
research is limited1 however as tests were only carried out for eight days in April and
thus cannot predict the hot water production for a year period. (myth et al. +9###/
developed a heat retaining $&(,vessel. .is research showed that reduced vessel wall
conduction and thus convective motion and fluid mi%ing1 improved thermal
stratification. 3inally1 the last research in date of thermal stratification in $&(,(-. was
;!
performed by .enderson et al. +!""B/ who developed one of the first studies on thermal
stratification for a rectangular $&(,(-. at various inclinations and constant heat flu%
values. *he authors showed that stratification played an essential role in defining the
heat gain characteristics when the storage tan' was in direct contact with the heating
surface. E%periments and &3D analysis were performed and showed that the degree of
stratification in the $&(,(-. depends upon the angle of inclination1 the e%posure time
and the incident insolation. *he same author posited that stratification in solar water
heaters can be roughly divided into two classesF +a/ static mode1 +b/ dynamic mode. *he
static mode is a limited study of temperature for fully or partially charged heaters1 while
a dynamic mode deals with charging and discharging cycle.
(everal methods are suggested in the literature to maintain and promote stratification in
storage tan's by utiliDing sleeves partitions1 baffles1 porous inta'e manifolds and low
flow rates. A water storage tan' would naturally remain stratified until water is supplied
to the load. As the hot water leaves the tan' and fresh cold water is infused1 mi%ing
occurs and as a result of flow circulation leads to de,stratification. *he location of the
outlet and inlet pipes should then be designed in accordance to the stratification studies
to not destroy stratification into the water tan'. 7ertDos and &aouris +!""C/ investigated
the influence of ten different inlet and outlet port placement on the velocity field. .is
research concluded that the most suitable arrangement consisted in the inlet and outlet
ports placed sideways and in opposite position in order to not destroy stratification in
the $&(,(-..
odelling stratification has been sub0ect to much research. (everal multi,node stratified
tan' models have been developed and compiled by Duffie and Bec'man +9##9/. ost
recent research on stratification was carried using the transient simulation pac'age
*?N(W( to simulate thermal stratification in water storage tan's such as &ruic'shan'
and .arrison +!""B/ and 4ordan and 3urbo +!""=/. .owever1 e%isting models are only
applicable to vertical storage tan's as a separate component of the collector. *o the
'nowledge of the author1 no stratification model has been presently suggested for
inclined $&(,(-..
*he literature showed thermal benefit of high stratification and emphasised the use of
low flow rates as being essential for the good performance of a collector1 however
improved mi%ing in the tan' results in higher rates of heat transmission. *hen
;;
stratification is important but good mi%ing in some case would result in better
performances for the (-.. (tratification is significant when trying to optimise the
collector for draw off as it helps to provide temperatures closer to those required for
domestic hot water systems. (tratification is critical to evaluating the actual heat content
inside the heater1 which is vital for estimating the efficiency. *herefore1 further research
is required in modelling stratification for $&(,(-..
1.1.1.% D-,A)233 61,-,6:+-i0:i60
*he word MDraw,offN refers to the collection of hot water from the collector. *he study
of draw off patterns from an $&(,(-. is important in order to evaluate the practical
performance and thus the effective output of the collector. Draw,off characteristics can
be organised in four sections. 3irst1 hot water temperature to be achieved and hot water
consumption1 representing the total water drawn,off1 need to be identified for the U<.
*hen the flow rate of water draw,off varies and is of primary importance. 3inally1 the
time of the day the water is drawn,off1 thus the draw off pattern1 needs to be identified.
*he hot water consumption within a domestic property is dependent upon the property
siDe1 the number of occupants1 household income1 lifestyle habits and the type and
number of end,use appliances. *he primary domestic uses of hot water includes
bathingJshowering1 hand washing1 clothes washing and dish washing. uneer +!""9/
indicates that the appro%imate temperature requirements for domestic hot water are
>"S& for bathsJshowers and hand basins1 ="S& for dishwashing by hand and up to C"S&
for clothes and dish washing machines. &ourtney +9#::/ reports typical delivered hot
water temperature in the region of ="S& to B"S& which are then diluted with cold water
or electrically heated at the end,use appliance to achieve the desired temperature.
?eliable data on domestic hot water consumption is diverse and some of the available
information is dated. -hittle +9#:C/ indicates demand for hot water consumption can be
anywhere between ;B lJdayJperson to =B lJdayJperson at a delivery temperature of ==S&.
.owever1 in a 9##C Energy 3ollow Up (urvey conducted on behalf of D*$ +!""=/ and
DE3?A an average ;# litres per day for washing machines1 baths and showers with a
further 9" litres for assumptions about the use of hot water at sin's and basins was
suggested per person per day. According to EU?@PA +!""B/ an average European uses
;B litres of B"S& hot water per day. A recent survey conducted by c2ennan +!""B/
e%amined the water consumption through a typical day for (cottish households. *he
survey was carried out for ;! properties monitoring hourly activity for domestic hot
;>
water appliances over a one,wee' period. *he average number of people living in a
household was !.>> persons1 comparable to the U<,&ensus +!""9/ average of !.;B
persons per household in !""9. *he hot water consumption too' account of all
instantaneous heaters such as electric power showers1 water heated internally by
domestic appliances i.e. washing machines and dishwashers1 and hot water drawn from
boilers or hot water tan's that is not used for space heating. *he average wee'day hot
water consumption for all properties surveyed was >#.# lJdayJperson based on a delivery
temperature of ==S&.
As stated before1 hot water requirement varies between households. *he flow rate and
pattern of water drawn are important parameters to the performance of the heater.
.ollands and 2ightstone +9#C#/ suggested that low flow rates from the storage tan' can
improve the system performance. 3anney and <lein +9#C:/ also investigated the
influence of the collector and heat e%changer flow rates on the performance of the
(-.. 2owering the collector fluid flow rate for direct systems resulted in
improvements to the thermal performance of the heater. *his was attributed to better
stratification within the solar tan' and a significant reduction in mi%ing1 which occurred
between the solar and au%iliary heated portions in the single tan'. Until the 9#C"s flow
rates were commonly around "."9= 'gJm
!
.s as stated by Duffie and Bec'man +9##9/.
2ower rates are recently being used such as ".""! to ".""B 'gJm
!
.s in (weden where
flow rates are given per m
!
of collector area.
$t is important to review the different draw,off patterns1 utilised in the past in order to
determine the optimum draw off profile for the present research. -hen water is drawn
off the corresponding quantity of cold water enters the $&( at the bottom1 thereby
decreasing the temperature1 which ta'es time for the store temperature to recover.
(everal studies have been carried out on draw,off profiles in the U<. An investigation
by &ourtney +9#::/ showed that if the period of pea' water demand occurred in the
evening rather than the morning1 annual heat output would increase by appro%imately
".= 74 for a > m
!
solar collector delivering 9:" lJday. Evening pea' demands ensure
solar contributions are utilised and there are no significant overnight losses that may
occur with a poorly insulated hot water heating system1 however lower tan'
temperatures achieved through daytime draw,offs can allow collectors to operate more
effectively. &ruD et al. +!""!/ utilised the British (tandard domestic heating load profile
as the model for the simulation of water demand in his trapeDoidal $&(,(-.. Different
;=
draw,off profiles can be found for the U<. *ypical demand profile data from U<,$(E(
supplying ;""lJday are represented in 3igure !,; compared to a more recent study
carried out by c2ennan +!""B/ for ;! Edinburgh households. .is study showed hot
water draw,off profile for a > m
!
$&( system supplying !"" lJday for a property
occupied by four persons1 assuming =" litres per day per person. $n both hot water
draw,off profiles1 significant periods of hot water demand are between Cam and #am
and :pm to Cpm.
0
&
10
1&
$0
$&
30
3&
%0
%&
&0
1 $ 3 % & ' ( 8 3 10 11 1$ 13 1% 1& 1' 1( 18 13 $0 $1 $$ $3 $%
3our of %a-
9
i
t
r
e
s
1+Lennan deand profile
")45-!- deand profile
Fi(. 2)<* H2: A,:+- >-,A)233 9-23i4+0 32- 0:7>i+0 23 M6L+..,. (2''%) ,.> UK)ISES 32- 2''4?>,/
,.> 3''4?>,/ -+09+6:iv+ 42,>0.
Another more recent study for Edinburgh was carried out by 4unaidi et al. +!""C/ for an
$&(,(-. connected to a water tan'. 3our draw off patterns were studied shown in
table !,9 below.
T,54+ 2)2* D+:,i40 23 >-,A)233 9,::+-.0 70+> 5/ H7.,i>i +: ,4. (2''<)
Draw,off
pattern
9
st
Draw,off
Oolume J time
!
nd
Draw,off
Oolume J time
;
rd
Draw,off
Oolume J time
>
th
Draw,off
Oolume J time
A !=l "#F"" !=l 9"F"" !=l 9;F"" !=l 9:F""
B !=l "CF"" !=l "#F"" !=l 9!F"" !=l 9:F""
& ="l 9"F"" ="l 9#F""
D ="l 9!F"" ="l 9#F""
*he research suggested that solution A and B could achieve a temperature of >"& in
the tan's. (cheme B was suggested as the best in terms of total energy e%traction.
4unaidi et al. +!""C/ show the benefit in having four draw offs instead of two and
therefore classified the four draw,off scheme from best to worstF B1 A1 D1 &. .owever
this draw,off profiles negate the advantage of the integrated tan' within an $&(,(-..
;B
*herefore1 more realistic draw,off scheme such as c2ennan +!""B/ are generally
preferred.
1.1.1.& P+-32-8,.6+ 61,-,6:+-i0,:i2. 23 ,. ICS)SWH
*he various types of $&(,(-. developed over time have impelled the need for a
common criterion to measure the effectiveness of (-.. As a result1 organisations such
as the European (olar &ollector and (ystems *esting 7roup1 E(&(*7 +9#C#/1
developed a standard testing procedure for all forms of (-. used in the European
&ommunity +E&/. $n the U(A1 the A(.?AE (tandard,#= +9#C9/ also developed
standards procedures in order to assess short term testing performance of (-.. *he
widely used test procedure proposed by the E(&(*7 based on a combination of test
methods showed a realistic $&(,(-. performance representation allowing a direct
comparison with other e%perimentally investigated $&(,(-.. As stated by (myth et al.
+!""B/1 the collection efficiency is similar to the .ottel,-hillier,Bliss equation
developed by .ottel and -hillier +9#==/ and Bliss +9#=#/ for distributed type solar
collectors1 but applies to diurnal system performance. &ollection efficiency is plotted
against values of
a-e
& T J
where
a-e
&
is the average insolation on the collector aperture
and T is the difference between the water temperature of the heating period and
ambient temperature over the same period. *his performance method is applied to flat
plate collector wor'ing in steady state mode. $t can be applied to $&(,(-.) however it
is not representative of the actual system performance as it is not a steady state situation.
Direct comparison of measured system performances is relatively straight forward)
however1 predicting the performance of an $&(,(-. at another location or with a
different installation configuration is more challenging. *herefore a review of the
different methods to predict $&(,(-. performances was carried out in the following
section.
2.2 M2>+44i.( :1+ 024,- A,:+- 1+,:+-
2.2.1 M,6-2 82>+4 B T1+-8,4 82>+4
As stated above1 predicting the performance of an $&(,(-. is a challenging tas'. A
model able to predict water temperature in an $&(,(-. would be a useful tool to allow
any user to predict the performance of the installed (-.1 depending on the orientation
;:
and location. Builders could use such a tool to demonstrate the performance of the $&(,
(-. and the benefit of such installation.
*he first study to propose and use computer simulation to predict $&(,(-.
performance was conducted by 7arg et al. +9#:!/ using a single node system. Burns et
al. +9#C=/ developed an analytical model to predict the performance of a stratified $&(,
(-.. 7ood agreement between theoretical and e%perimental results was reported. 2ater
studies by &ruD et al. +!""!/ investigated the energy saving potential provided by a
trapeDoidal,shaped solar collector. *his was evaluated e%perimentally and by
computation1 using a thermal networ' analysis model based on 2uiD +9#C=/. *he
thermal networ' showing the equivalent heat flow circuit collector1 insulation and losses
is showed in 3igure !,>. A full scale e%perimental test rig was employed to validate
their modelling. Despite good agreement of simulation and e%perimental results1 the
authors failed to account for two factors. 3irst1 the heat capacity of the system was
neglected. (econd1 tests were carried out in a laboratory which is unrepresentative of
field conditions. *hus1 there is a need to develop a more accurate device to predict
performance for $&(,(-.. 4unaidi +!"":/ developed a similar program for fluid flow
and heat transfer generating bul' water temperatures for a rectangular shaped $&(,(-.
using specific parameter such as a constant heat flu% for the day. *he e%perimental and
computed temperatures were found with a deviation of 9"L. *hese models are limited
as they can not be applied to real weather conditions. *here is therefore a need to
develop an improved model able to predict the bul' water temperature inside the
collector for any weather condition and account for different $&(,(-. design
parameters. *o achieve such a program1 heat transfer mechanisms in the collector need
to be reviewed and assessed.
;C
Fi(. 2)$* E=7iv,4+.: 1+,: 342A 6i-67i: 32- C-7E +: ,4 (2''2) 6244+6:2-; i.074,:i2. ,.> 4200+0
*he three heat transfer mechanisms1 conduction1 radiation and convection1 occur inside
the (-. and result in heat losses or gains. An $&(,(-. is an assembly of two natural
convection cavities composed of an air cavity1 represented by the space between the
absorber plate and the glass cover1 sitting on top of a water cavity. &onvective
behaviour of both cavities are different for a given angle of inclination as they operate at
different ?ayleigh numbers and most li'ely have dissimilar geometrical dimensions as
their aspect ratio. Both cavities are insulated by a batch system where conduction is
occurring. A representation of an $&(,(-. heat transfer mechanism is represented in
3igure !,=.
;#
Fi(. 2)1'* H+,: :-,.03+- 8+61,.i08 i. ,. ICS)SWH
*he different mechanisms emerging from the system are detailed belowF
, &onvection from two parallel inclined plates at different temperatures +air cavity/
, &onvection from an inclined flat plate +water cavity/
, &onvection from cover to the environment
, &onvection from bo% to the environment
, ?adiation from plate to cover
, ?adiation from cover to s'y
, &onduction through bo%
, .eat capacitance of the system
A review of widely used equations for natural convection and radiation are detailed
below.
1.1.1.< N,:7-,4 62.v+6:i2. 3-28 :A2 9,-,44+4 i.64i.+> 94,:+0 B ,i- 6,vi:/
Natural convection in air cavities has been studied by many researchers and is a well,
established phenomenon. .ollands et al. +9#:B/ presented new e%perimental
measurements on free convective heat transfer rates through inclined air layers of high
aspect ratio1 heated from below. *he ?ayleigh number ranged from subcritical to 9"
=
)
the range of the angle of inclination MN was between " and :" degrees. Although it
was anticipated that the results might be identical to the results for the horiDontal layer if
one replaced M?aN by M 'a(os N1 significant departures from the introduction of the
angle were observed. As a result of their study1 .ollands et al. +9#:B/ recommended a
relationship giving the Nusselt number as a function of 'a(os e%pressed below. *his
e%pression has been since widely accepted.
>"
1
1
]
1
,
_
+
1
]
1
1
]
1
+ 9
=C;"
cos
cos
/ C . 9 +sin 9:"C
9
cos
9:"C
9 >> . 9 9
; J 9
B . 9
L
L L
a
'a
'a 'a
*$ +!.9/
*he notation K implies that1 if the quantity in brac'ets is negative1 it must be set equal to
Dero. ?estriction in usage of this relation are that .J2 Z 9! and "[\ ]
('
1 where
('
1 is the critical tilt angle which varies with the aspect ratios as represented in *able
!,!.
T,54+ 2)3* C-i:i6,4 ,.(4+ v+-070 ,09+6: -,:i2; H244,.>0 +: ,4. (1$&%)
.J2 9 ; B 9! 9!
('
!= =; B" B: :"
*he ?ayleigh number is e%pressed belowF
;
/. .+ . L T T !
'a
c p
+!.!/
Equation +9/ has been found to give e%cellent results when applied inside the regime
described above. *he angle depends on the orientation of the hot surface as shown in
3igure !,B.
Fi(. 2)11* O-i+.:,:i2. ,.(4+ 32- ,i- 6,vi:/ -+(-+00i2.
1.1.1.$ N,:7-,4 62.v+6:i2. 3-28 i.64i.+> 34,: 94,:+0 B A,:+- 6,vi:/
any studies have been carried out on inclined heated plates X3u0i and $mura +9#:!/1
&hurchill and &hu +9#:=/1 3ussey and -arneford +9#:C/1 &ruD et al. +!""!/Y providing
a good breadth of conditions with appropriate regressions. 3u0i and $mura +9#:!/
studied natural heat transfer convection from a plate with arbitrary inclination. *he heat
>9
9C"
"
&ool surface
.ot surface
was transferred from one side surface of two plates. *he authors focused on natural
convection from an isolated1 vertical plate1 and e%amined the influence of the tilt angle
over the range of
C# " and
99 =
9" 9" < < 'a(os . 3or downward facing
heating plate the authors suggested the following regressionF
( )
> J 9
=B . " 'a(os *$ +!.;/
*he value of the e%ponent
( ) > J 9
suggest a laminar flow up to values of the parameter
'a(os of
99
9" . 2ater studies from &ruD et al. +9###/ investigated buoyancy,driven
convective heat e%change in a trapeDoidal,shaped solar water heater. Using results from
3u0i and $mura +9#:!/1 &ruD et al. +9###/ developed a new formula of the Nusselt
number for inclined water cavities with an estimated uncertainty of :L e%pressed
belowF
( )
B J 9 # J 9
> J 9
H
=B . "
,
_
,
_
#
H
#
D
'a(os *$ +!.>/
-here D is the +front,to,bac'/ bottom depth of cavity floor1 .5 is the inclined height of
the heater) characteristic length and U is the +front,to,bac'/ top depth of cavity floor
with ?ayleigh number of equationF
!
;
H
/. .+ . . Pr
,
_
H T T !
'a
p
+!.=/
*his e%pression was found in good agreement with the correlation of 3u0i and $mura
+9#:!/1 when
>= 1 indicating that the geometric parameters +DJU and .5JU/ have
only a small influence in comparison with the modified ?ayleigh number1 'a(os .
*he parameter .5 represents the inclined height of the heater1 characteristic length. *he
geometric parameters of &ruD et al. +9###/ are shown in 3igure !,:.
>!
Fi(. 2)12* G+28+:-i6 9,-,8+:+- 23 C-7E +: ,4. (1$$$) :-,9+E2i>,4)01,9+> 024,-?+.+-(/ 0:2-+
1.1.1.1' R,>i,:i2. 3-28 94,:+ :2 62v+-
Duffie and Bec'man +9#:>/ suggested a standard equation for heat e%change radiation
between two parallel plates such as the absorber plate and the glass cover of a collector.
B
c p
c p r
T T
%
/ +
> >
1
+!.B/
where
B
are the
temperature and emissivity of the absorber plate1
c
T
and
c
/ .+
> >
1
+!.:/
where
s$r
are the
temperature and emissivity of the s'y.
>;
*he s'y5s temperature is assumed to be equal to the ambient temperature for laboratory
conditions. .owever1 for field tests the s'y acts as a blac' body during a clear night s'y.
*he s'y emissivity is therefore an important parameter for inclusion in models in order
to simulate real weather conditions. any correlations have been reported in literature
for calculating the coefficient of emissivity inside the atmospheric transparency window
such as $dso +9#C9/1 Berdahl and artin +9#C>/ or &ucumo et al. +!""B/. Equations for
s'y emissivity are generally valid only for clear s'y conditions) as in equation +C/ below
developed by &ucumo et al. +!""B/ for s'y emissivity in the atmospheric transparency
window1
(S
.
> . :;=#> C . BC"
/ 9 + 9":#=!
9
!
+
+
a a
(S
T T
+!.C/
-ith
!
9""
BBBC . "
9""
B""9 . " :>: . "
,
_
+
dp dp
T T
+!.#/
-here
(S
'
)
1
<
C
.luiniu6,inned
-tainless4steel6,inned
-tainless4steel6"nfinned
Fi(. 3)#2* W,:+- :+89+-,:7-+ 9-23i4+0 23 3i..+> ,478i.i78; 3i..+> ,.> 7.)3i..+> 0:,i.4+00)0:++4
ICS)SWH0 32- +i(1: 127-0 23 +x9207-+ :i8+ ,: 2''W.
$n order to support the results identified above the overall water temperature of the un,
finned stainless,steel collector was plotted as a function of the finned stainless,steel
$&(,(-. in 3igure ;,!B.1 and as a function of the finned aluminium $&(,(-. in
3igure ;,!:. *hese scatter plots have the very specific purpose of demonstrating how
much the MfinN affects water temperature achieved in the collector. $t is apparent from
3igure ;,!B and ;,!: that the finned heaters outperform the un,finned by achieving a
higher temperature rise
m
T
illustrated in 3igure ;,!B and ;,!: by the deviation from
the Mone to oneN straight line relationship. 3igure ;,!B and ;,!: also shows that over
time1 data points converge) approaching to the Mone to oneN straight line relationship.
7radually1 un,finned and finned water temperature in the collector tends to reach the
same value) approaching an equilibrium temperature observed for the 9""-Jm
!
heat
flu% profile.
::
0
10
$0
30
%0
&0
0 10 $0 30 %0 &0
Finne% stainless=steel '
)
1 <C
5
n
f
i
n
n
e
%
s
t
a
i
n
l
e
s
s
=
s
t
e
e
l
'
)
1
<
C
%00#
$00#
100#
Fi(. 3)#3* ICS)SWH A,:+- :+89+-,:7-+ i.6-+,0+* 7.)3i..+> v+-070 3i..+> 0:,i.4+00)0:++4
0
10
$0
30
%0
&0
0 10 $0 30 %0 &0
Finne% alu)iniu) '
)
1 <C
5
n
f
i
n
n
e
%
s
t
a
i
n
l
e
s
s
=
s
t
e
e
l
'
)
1
<
C
%00#
$00#
100#
Fi(. 3)##* ICS)SWH A,:+- :+89+-,:7-+ i.6-+,0+* 7.)3i..+> 0:,i.4+00)0:++4 v+-070 3i..+> ,478i.i78
:C
*hese findings suggest that incorporating fins in the collector results in an increase in
heat transfer1 and that using aluminium generates higher overall water temperatures as
shown in 3igure ;,!B and ;,!:. $n a real world application water will be regularly
drawn from the $&(,(-.1 not allowing enough time for the water to achieve
equilibrium temperature1 early differences in temperature rise is critical for the good
performance of the $&(,(-..
1.1.1.22 E33i6i+.6/ i89-2v+8+.:
*a'ing into account all the parameters discussed in section ;.!.;.91 efficiencies were
calculated by the following equation and then corrected. *he collector efficiency is the
ratio of the bul' heat collected to the heat available as shown in equation ;.!.
&
"
c
+;.!/
-here
c
"
is the heat collected at each control volume +-Jm
!
/
& is total irradiation from the heating pad +-Jm
!
/
Also e%pressed as
j j
j i
i
i i p
j
P
T V C
wi
.
. N . .O
;
22
1
-ith
/ + F
9 +
>
6 6 6
t t
+;.;/
@verall efficiencyF
j j
n
j
j i
i
i i p
n
j
P
T V C
wi
.
. N . .O
1
;
22
1 1
+;.>/
-here
i
)
is the control volume associated of the thermocouple MiN +m
;
/1 6
+
is
the heat flu% applied to the system +-/1
wi
p
C
is the specific heat capacity of water
associated with thermocouple temperature +4J'g.</1
i
is the water density associated
with the thermocouple temperature +'gJm
;
/1 6 i
T
1
+;.:/
As discussed by E(&(*7 +9#C#/ equation +;.B/ is useful because it can be used to plot
an informative efficiency curve1 which predicts collector performance for any operating
point.
A constant heat loss coefficient would result in a straight line if efficiency is plotted
against the 'nown operating point. 3igure ;,;# illustrates the efficiency curve for the
#"
$&(,(-. e%periment data. Based on equation ;.B1 the heat loss coefficient of the $&(,
(-. was found to be >.= -Jm
!
.<.
0
0.1
0.$
0.3
0.%
0.&
0.'
0.(
0.8
0 0.0$ 0.0% 0.0' 0.08 0.1 0.1$ 0.1% 0.1'
?'w='a@ " A 1 7.)
2
" $
+>.>/
9"=
.eat losses .eat gains
T
A
Air cavity
-ater cavity
T
,
P
4200+0
T
6
T
9
P
U0+374
7lass cover
$nsulation
Absorber plate
where
B
1
+
T
is the
temperature of the absorber plate1
c
T
is the temperature of the glass cover temperature1
pc
#
is the overall U,value between the plate and the cover1 is the (tefan,BoltDmann
constant and pc
h
is the convective heat transfer coefficient between inclined parallel
plates with a slope from the absorber plate to the glass cover. Based on the wor' of
.ollands et al +9#:B/ the following relationship of pc
h
for inclined cavities was
developedF
L
k
'a
'a 'a
h
fa
L
L L
pc
,
_
1
1
]
1
,
_
+
1
]
1
1
]
1
9
=C;"
cos
cos
/ C . 9 +sin 9:"C
9
cos
9:"C
9 >> . 9 9
; J 9
B . 9
+>.=/
-here fa
k
is the thermal conductivity of air at film temperature and L is the
characteristic length. *he notation K implies that1 if the quantity in brac'ets is negative1
it must be set equal to Dero.
Heat loss from the cover to the surroundings
2osses to surroundings will differ whether the models compute laboratory or field
conditions. Both are e%pressed below respectivelyF
2aboratory conditionsF
( ) / + / +
> >
. a c ca Sky c c a c Wind s c
T T # T T T T h % +
+>.B/
3ield conditionsF
( ) ( )
Sky c cs a c ca
s$r
Sky c
a c Wind s c
T T # T T #
T T
T T h % +
/ +
/ .+
> >
+>.:/
9"B
where
Wind
h
is the e%ternal convection from the glass cover to ambient1 Sky
T
is the s'y
temperature1
a
T
is the ambient temperature1
c
is the bul' emissivity of equation
9
9 9
+
Sky c
where Sky
.
> . :;=#> C . BC"
/ 9 + 9":#=!
9
!
+
+
a a
(S
T T
+>.9"/
-ith
!
9""
BBBC . "
9""
B""9 . " :>: . "
,
_
+
dp dp
T T
9":
-here
(S
,
_
,
_
,
_
B J 9 # J 9
> J 9
=B . "
+>.9B/
-here m H 9 and m # D "= . " 1
9
+>.9C/
-here
&ns
'
is the resistance of the insulation material which will vary between
laboratory and field installations1 and
Wind
'
is the resistance occurring at the bo%
surface in contact with the wind.
Heat loss from the fins to the water
*he inclusion of e%tended heat transfer surfaces +fins/ into the design resulted in an
increase in heat transfer to the water. $deally the fin material should have a large thermal
conductivity to minimiDe temperature variations from its base to tip. 3our fins of
straight1 rectangular uniform cross sectional area were used for the design. Each fin was
attached to a base surface of temperature
/ " + T T
p
called plate temperature and
e%tends into fluid temperature
w
T
as described in 3igure >,B.
Fi(. #)%%* S:-,i(1: -+6:,.(74,- 3i.0 23 7.i32-8 6-200 0+6:i2. ,.> +.+-(/ 5,4,.6+ 32- ,. +x:+.>+>
07-3,6+
Based on the wor' of $ncropera and Dewitt +!""9/1 a new model was developed using
an elemental strip method to determine the heat transferred from the fin to the water
assuming an adiabatic tip condition + d x /F
"
d x
o
dx d
. *he fin temperature
distribution along the length
x
was determined by the following formulaF
999
T
p
w o x f
T
md
x d m
T +
cosh
/ + cosh
1
+>.9#/
-here w p o
T T
c
kA h+ m
!
c
A
G cross sectional area +
w t A
c
/
+ G perimeter
x
G strip position
d G depth of the fins
*he fin heat transfer rate f
%
per strip can be e%pressed asF
/ + .
w f f f f
T T A h %
+>.!"/
-here
f
A
fin area
w x *
s
* number of fins
s
x
strip length ta'en equal to m ""! . " for modelling.
*he heat transfer to the water is mainly a convective process. uneer et al +!"";/ used
the equation of &hurchill and &hu +9#:=/ incorporating the following regression
assuming a uniform heat flu%F
( ) [ ]
# J >
9B J #
> J 9
Pr J >;: . " 9
. B: . "
BC . "
+
+
L
'a
*$
+>.!9/
-here
/ + (os d L
with
T
p
!
T
f
"
x
#
/
T
p
!
T
w
#
.luiniu +olle+tor43
.luiniu +olle+tor41.&
-tainless4steel +olle+tor41.&
Fi(. #)%&* Di8+.0i2.4+00 3i. :+89+-,:7-+ v+-070 3i. 4+.(:1
*he total heat transferred through the fins to the water
ins
%
was calculated using the
effective length of the fin. *he quantity
ins
%
1 e%pressed below was calculated by
adding up the fin heat transfer rate f
%
per strip until reaching the limit of effective
length.
L #"
"
x x x ins
A h %
+>.!!/
An improvement factor +&/ was introduced representing the improvement achieved by
using fins in the $&(,(-. over an un,finned collector and is e%pressed in equation >.!;
below.
finned #n
w p
ins
%
%
&
+ 9
+>.!;/
Calculation of the bulk water temperature
Using the equations developed above1 an iterative computer program was developed
using OBA with initial conditions of p
T
and
c
T
set asF
= . " +
w p
T T
99;
,
_
;
a p
p c
T T
T T
Using equations +>.!/ and +>.;/1 e%pressions for p
T
and
c
T
for tG" were developed
below in equations +>.!>/ and +>.!=/. *he glass cover temperature at tG" is a function of
the assumed value of absorber plate temperature.
pc ca
a ca p pc
c
# #
T # T #
T
+
+
+>.!=/
$terations were made until a balance was obtained between heat losses. *he program
finally output the glass cover temperature U
c
T
5 and the absorber plate temperature U
p
T
5. .eat flu%es in the networ' are calculated in the macro and using equation +>.9/ a
transformed equation +>.!B/ can be developed to calculate the water temperature riseF
+
+ +
(
% % %
T
(
%
T T
a w ins w p
w
#sef$l
w w
H
+>.!B/
An interval of one hour was ta'en to display results. After each interval the value of the
cover1 absorber plate and water temperature were calculated and used as the input for
the ne%t time step. *he cycle was repeated for !> hour time increments.
$f the option MN@N fins is activated1 then
x
A
is automatically set to Dero1 then fins are
not ta'en into account into the calculations.
#.1." M2>+44i.( 0:-,:i3i6,:i2. 32- 4,52-,:2-/ 62.>i:i2.0
(tratification in a water tan' is an essential factor for the good performance of an $&(,
(-.. *his section e%plores the modelling of stratification inside an aluminium $&(,
(-. for laboratory conditions.
99>
-ater temperatures within the collector were observed1 in &hapter ;1 to be dependent
on the longitudinal position1 imposed heat flu%1 and e%posure time. Using e%perimental
data1 a polynomial function for the stratification profile at each hour was developed
using heat flu% and longitudinal position in the collector as the main parameters. $n
order to obtain a generalised equation for any heat flu%1 dimensionless temperature1 1
and dimensionless length1 H h h J K 1 were calculated and plotted for each temperature
measurement location for the aluminium $&(,(-. on an hourly interval. *his is shown
in 3igure >,C.
40.08
40.0'
40.0%
40.0$
0.00
0.0$
0.0%
0.0 0.1 0.$ 0.3 0.% 0.& 0.' 0.( 0.8 0.3 1.0
(CD("3
D
?
'
(
=
'
w
b
@
"
'
w
b
100#
$00#
%00#
Fi(. #)%<* Di8+.0i2.4+00 :+89+-,:7-+ 0:-,:i3i6,:i2. ,3:+- < 127-0
Polynomial functions were developed from 3igure >,C for each wattage as defined
belowF
c bh ah + + K K
!
+>.!:/
A single function was then developed incorporating the change of stratification with
position and heat flu%. *he following equations were developed to produce a
temperature stratification profile for an average eight hour day. Polynomial functions
for heat flu% of 9""1 !"" and >""-Jm
!
are e%pressed belowF
"=>" . " K 9CB: . " K 9!9! . "
!
9"" 1 C
+ h h
W hrs
+>.!C/
"CC: . " K ;9!! . " K !":= . "
!
!"" 1 C
+ h h
W hrs
+>.!#/
99=
99"! . " K ;C>C . " K !=;9 . "
!
>"" 1 C
+ h h
W hrs
+>.;"/
A final overall e%pression was then developed for each hour as a function of the heat
flu% input and the longitudinal position in the collector. *he e%pression for the stratified
temperature at the end of eight hours of operation is given asF
c bh ah
hrs
+ + K K
!
C
+>.;9/
-here a1 b and c are the variation of the polynomial coefficient with heat flu% for each
hour. &oefficients were noted and modelled in a more generalised function and are
provided belowF
; ; ! B
9" > . : 9" 9= 9" !
+ " " a
+>.;!/
; ; ! B
9" : . 9 9" 9 . ! 9" ;
+ + " " b
+>.;;/
; ; ! B
9" ; . ; 9" B . " 9" C . "
" " c
+>.;>/
*he same approach was used for each hour.
Using the previously developed macro model for laboratory conditions) polynomial
functions were added to the macro model to compute water temperature stratification
with height. *he initial laboratory macro model was able to generate corresponding bul'
water temperatures in the collector with a given hourly incident solar radiation1 ambient
temperature and inlet water temperature.
*he new model was able to compute the bul' water temperature variation in different
$&(,(-. collectors for a given aspect ratio and to evaluate temperatures at various
points in the collector +temperature stratification/ based on the mean water temperature
given by the macro model and the total time of e%posure.
?esults comparing computed with e%perimental stratification are discussed in section
>.9.:.!.
99B
#.1.% Di(i:,4 0i874,:i2. 342A 61,-:
$nput .eat 3lu% or
(olar radiation1
7+/
&alculation of *
w
from _
Useful
Update time step1
tF i G i I 9
i G 9"K
END
&hec'
i
Update *
p
and *
c
i G i I 9
&alculation of U
ca
and U
pc
$nlet water
temperature1 *
w
$nitialisation of
*
p
and *
c
Ambient
temperature1 *
a
&loud cover1 &&
&alculation of
*
dp
&alculation of
*
('y
and
('y
Air properties at
film temperature1
*
fa
-ind velocity O
w
&over and plate
emissivity1
c
1
p
-ater properties at
film temperature
*
fw
&alculation of
_
3ins
&alculation of
_
Useful
&alculation of
*
p
and *
c
+eq. >.!> A >.!=/
&alculation of
U
pw
&alculation of
U
wa
&hec'
i G 9"
i G 9"
i \ 9"
i \ 9"K
99:
#.1.& C2897:,:i2.,4 ,.> +x9+-i8+.:,4 >,:, 6289,-i02.
1.1.1.31 L,52-,:2-/ -+074:0
R+074:0 32- 3i..+> 6244+6:2-0
3igures >,# and >,9" show the simulated heat transfer rate through the fins and the fin
temperature through the fin length respectively. *he fins were used to enhance heat
transfer between the material and the ad0oining fluid. *he heat transfer rate was
increased by employing different fins configuration as demonstrated in 3igure >,#. *he
thermal conductivity of the fin material has a strong effect on the temperature
distribution along the fins +3igure >,9"/ and therefore influences the degree to which the
heat transfer rate is enhanced as shown in 3igure >,#. $deally the fin material should
have a large thermal conductivity to minimise temperature variations from its base to its
tip.
0.0
0.1
0.$
0.3
0.%
0.&
0.'
0.(
0.8
0 % 8 1$ 1' $0 $% $8 3$ 3' %0 %% %8
Fin len.t( >1 ))
F
i
n
(
e
a
t
t
r
a
n
s
f
e
r
r
a
t
e
1
$
.luiniu +olle+tor43
.luiniu +olle+tor41.&
-tainless4steel +olle+tor41.&
Fi(. #)%$* Si874,:+> 3i. 1+,: :-,.03+- -,:+
99C
&1.30
&1.3&
&1.%0
&1.%&
&1.&0
&1.&&
&1.'0
0 & 10 1& $0 $& 30 3& %0 %& &0
Fin len.t( >1 ))
F
i
n
t
e
)
0
e
r
a
t
u
r
e
1
<
C
.luiniu +olle+tor43
.luiniu +olle+tor41.&
-tainless4steel +olle+tor41.&
0.1
01,9=3
0.1
0,
,9=1.5
0.1
0
1&'=1.5
Fi(. #)&'* Si874,:+> 3i. :+89+-,:7-+ >-29 A1+-+
) ( 1 . ' 1 . '
$ p %
T T
i0 :1+ +x6+00
:+89+-,:7-+ ,61i+v+> ,0078i.( $'K 23 1+,: i0 62.>76:+> :1-27(1 :1+ 3i.0.
3in length effectiveness1 and consequently heat transfer rate are enhanced by choosing a
material with high thermal conductivity) such as aluminium. $t was also enhanced by
increasing the thic'ness of the fin1 increasing at the same time the surface in contact
with water. *he optimum effective length of the fin was defined earlier when #" percent
of the actual $ p %
T T
was achieved. *he optimum effective length for the
aluminium collector fin of ;mm and 9.=mm thic'ness1 and stainless,steel of 9.=mm
thic'ness were found to be ;;.B:1 ;9.=" and 9!.B!mm respectively) equating to an
effective fin utilisation of B:.;L1 B;L and !=.;L1 based on their respective optimum
effective length.
(imilarly1 a higher fin heat transfer coefficient is achieved using the higher conductivity
material and greater material thic'ness. $ndeed1 by increasing the thic'ness a greater
amount of heat was transferred through the fins5 length to achieve improved rates of
heat transfer to the bottom of the collector as shown in 3igure >,99.
99#
%8.0
38.0
1%8.0
138.0
$%8.0
$38.0
3%8.0
338.0
%%8.0
0 % 8 1$ 1' $0 $% $8 3$ 3' %0 %% %8
Fin len.t( >1 ))
F
i
n
(
e
a
t
t
r
a
n
s
f
e
r
c
o
e
f
f
.
1
$
"
)
2
.
7
.luiniu +olle+tor43
.luiniu +olle+tor41.&
-tainless4steel4+olle+tor 1.&
Fi(. #)&1* Si874,:+> 3i. 1+,: :-,.03+- 62+33i6i+.:
?esults showed that the higher the heat flu% applied to the system1 the higher the fin
heat transfer rate achieved. .owever1 the fin heat transfer rate decreases along the
length until reaching an asymptotic value corresponding to the heat flu% applied. 3igure
>,9! reveals another typical trend showing that with time the fin heat transfer rate
decreases. *his is because with time the collector water temperature increases thus
reducing heat transfer from the fins as the water approaches a temperature equilibrium
with the e%ternal heat flu% applied to the system
0.0
0.&
1.0
1.&
$.0
$.&
0 % 8 1$ 1' $0 $% $8 3$ 3' %0 %% %8
Fin len.t( >1 ))
F
i
n
(
e
a
t
t
r
a
s
n
f
e
r
r
a
t
e
1
$
Fi(. #)&2* Si874,:+> 3i. 1+,: :-,.03+- -,:+ Ai:1 :i8+ B A478i.i78 388 SWH
9!"
1 (r
5(rs 10(rs 15(rs 20(rs
A fin efficiency f
f
,
%
.luiniu +olle+tor43
.luiniu +olle+tor41.&
-tainless4steel +olle+tor41.&
Fi(. #)&3* Si874,:+> 3i. +33i6i+.6/
*he improvement factor +$3/ decreases with time while the water temperature increases.
$ndeed
ins
%
decreases while w p
%
increases with time and are related to the water
temperature and the $3. *hese results are interrelated. A comparison between different
types of fins was made for an input temperature of !!.9& and a heat flu% of 9""-Jm
!
applied to the system. ?esults are presented in *able >.9 below.
9!9
T,54+ #)<* Ov+-,44 i89-2v+8+.: 3,6:2- 32- (iv+. 9,-,8+:+-0
aterial 3in thic'ness1 mm @verall $3
Aluminium ; 9.9>
Aluminium 9.= 9.9"
(tainless,steel 9.= 9.">
Unfinned , 9.""
By increasing the thic'ness of the fins the overall $3 improved by ;.=L1 while by
changing the material to a more conductive material the $3 improved by =.=L. *he
improvement factor measuring the fin performance showed that1 overall1 a higher
performance was achieved for the aluminium $&(,(-. with a ;mm fin thic'ness.
C289,-i02. 23 :1+ >i(i:,4 0i874,:i2. ,.> 4,52-,:2-/ -+074:0
A comparison was underta'en between the computational results from the model and
the actual e%perimental data. (imulated and computed efficiency of the aluminium $&(,
(-. achieved with time for >""-Jm
!
was plotted in 3igure >,9> and show a good
empirical correlation.
*he comparison of the water temperatures achieved e%perimentally and numerically
was plotted in 3igure >,9= for the three $&(,(-.s. 7ood correlations were observed
for all the $&(,(-. types.
0
10
$0
30
%0
&0
'0
(0
80
0 $ % ' 8 10 1$ 1% 1' 18 $0 $$
'i)e1 (rs
1
%
%00#6!8periental
%00#6-iulation
Fi(. #)&#* ICS)SWH +33i6i+.6/ Ai:1 :i8+ 32- #''W?82 1+,: 347x ,994i+>
9!!
0
&
10
1&
$0
$&
30
3&
%0
0 $ % ' 8 10 1$ 1% 1'
'i)e1 (rs
'
)
1
<
C
,inned .uiniul6!8periental
,inned .luiniu6-iulation
,inned -tainless4steel6!8periental
,inned -tainless4steel6-iulation
"nfinned -tainless4steel6!8periental
"nfinned -tainless4steel6-iulation
Fi(. #)&"* C2897:,:i2.,4 ,.> +x9+-i8+.:,4 6289,-i02. 32- 2"'W?8
2
1+,: 347x
&omputed data were plotted as a function of e%perimental data in 3igure >,9B to
e%amine the validity of the simulation. $t was demonstrated that e%perimental and
computer simulation results are in good agreement as they follow a 9F9 relationship with
a 9"L over prediction in the modelled value.
0
$0
%0
'0
80
0 $0 %0 '0 80
' )easure%1 <C
'
c
o
)
0
u
t
e
%
1
<
C
Fi(. #)&%* C2897:+> v0 Ex9+-i8+.:,4 A,:+- :+89+-,:7-+0
9!;
.owever1 a proper validation of the results by statistical methods and e%treme range
tests was carried out in order to improve confidence in the findings.
'tatistical measures used for evaluation of the model
$n order to evaluate the performance of each set of models1 the following statistical
indicators were employedF
, (lope of the best,fit line
, &oefficient of determination +?
!
/
, ean bias error +BE/
, ?oot mean square error +?(E/
, ('ewness
, <urtosis
The slope of the best:fit line between the computed and measured variable is desirably
equal to one. (lope values e%ceeding one indicate over,estimation) while slope values
less than one indicate under,estimation of the computed variable.
The coefficient of determination0 '
;
0 is the ratio of e%plained variation to the total
variation. $t lies between Dero and one. A high value of ?
!
indicates a lower une%plained
variation and is desirable. ?
!
is often used to 0udge the adequacy of a regression model
but it should not be the only criterion for choosing a particular model.
( )
( )
!
!
!
< 9
< (
= =
= =
'
+>.;B/
-here
(
=
G computed value
<
=
G mean value
9
=
G measured value
n
G number of points
The mean bias error0 <B.1 provides a measure of the trend of the model1 whether it has
a tendency to under,predict or over,predict its modelled values. Positive values of BE
indicate under estimation while negative values imply over estimation by the proposed
model. A BE nearest to Dero is desired. $t is given by the following equationF
9!>
n
= =
<B.
n
i
( 9
9
/ +
+>.;:/
-here
n
G number of points
The root mean s%$are error0 '<S.1 gives a value of the level of scatter that the model
produces by providing a comparison of the actual deviation between the predicted and
the measured values. (ince it is a measure of the absolute deviation1 ?(E is always
positive. A lower absolute value of ?(E indicates a better model.
! J 9
9
!
/ +
,
_
n
= =
'<S.
n
i
( 9
+>.;C/
Skewness represents a measure of the lac' of symmetry in a distribution.
A distribution is symmetric or normal if it loo's the same to the left and right of the
centre point1 resulting in s'ewness statistic of Dero value for perfect symmetry. *his
would indicate a robust model. A positively s'ewed distribution tails off to the high end
of the scale while negative s'ew tails off the low end of the scale.
7$rtosis is defined as a measure of the degree of pea'edness in the distribution1 relative
to its width. *he 'urtosis statistic will be Dero for a normal distribution1 positive for
pea'ed distributions and negative for flat distributions. A pea'ed distribution of the
errors is highly desirable. A high positive value of 'urtosis suggests that there are fewer
outliers in the estimation.
*he si% statistical indicators detailed above were used in the research to quantitatively
evaluate the performance of the model. *hese were calculated using a macro within
OBA and are summarised in *able >,!. E%treme range tests were also used to evaluate
the reliability of the laboratory macro model to predict bul' water temperature in the
$&(,(-..
T,54+ #)$* S:,:i0:i6,4 i.>i6,:2-0 v,47+0 B T2:,4 D,:,
&tatistical in%icator Ealues
9!=
-lope 1.01
;
$
0.33'
1B! <2= 40.30
;1-! <2= 1.00
-0ewness <2= 40.(3
)urtosis 3.$1
*he slope of the best fit line between computed and measured temperature was found to
be equal to 9."9 suggesting a slight over estimation of the computed variable. *he high
value of ?
!
indicates a low une%plained variation showing a good capability of the
regression model. *he BE value of ,".;" shows that the model has a tendency to over
predict its computed values. *he low ?(E value of 9." shows an acceptable actual
deviation. *he low negative s'ewed distribution tends to the lower end of the scale.
.owever1 its low value of , ".:# indicates a robust model. *he positive 'urtosis of ;.!9
indicates a pea'ed distribution of the errors suggesting that there are low outliers in the
estimation.
3inally e%treme range tests were underta'en to show the credibility of the model. 3or a
period of eight hours at constant heat flu% of =" and B""-Jm
!
good close fitting curves
of computed and e%perimental data were achieved with very low BE and ?(E
values of ,".>; and ".=; respectively suggesting a robust model.
(tatistical and e%treme range results validated the accuracy of the model to compute
bul' water temperature.
1.1.1.32 S:-,:i3i6,:i2. ) 82>+4 -+074:0
(esults
*his section e%plores the validity of the model predicting stratification inside the $&(,
(-. by comparing computed results with e%perimental data. *he simulation was
carried out for a wide range of heat inputs +="1 9""1 9="1 !""1 !="1 ;""1 ;="1 >"" and
B""-/ for a period of eight hours simulating a typical day of operation. &omputed
water temperatures at any given height were plotted as a function of e%perimental data
to e%amine the validity of the simulation. $t was demonstrated that e%perimental and
computer simulation results were in good agreement as they followed a 9F9 relationship
shown in 3igure >,9:.
9!B
0
10
$0
30
%0
&0
'0
(0
80
0 10 $0 30 %0 &0 '0 (0 80
' )easure%1 <C
'
c
o
)
0
u
t
e
%
1
<
C
Fi(. #)&&* C2897:+> v0 Ex9+-i8+.:,4 0:-,:i3i6,:i2. ,3:+- < 127-0 23 29+-,:i2. 32- ,44 1+,: i.97:0
$t is observed in 3igure >,9C that variation between e%perimental and computed results
data occurs during e%tended +!!hr/ model validation and e%perimental testing regime
displayed graphically by a deviation from the uniform 9F9 relationship. $t suggests that
as temperatures increase towards boiling1 bul' fluid movement is brea'ing down the
stratification. Based on the assumption that water temperatures to be achieved in a
(cottish climate are unli'ely to e%ceed :"& and that a typical solar heating day would
be of eight hours or less ma'e this model suitable for (cottish weather conditions.
9!:
0
10
$0
30
%0
&0
'0
(0
80
30
100
0 10 $0 30 %0 &0 '0 (0 80 30 100
' )easure%1 <C
'
c
o
)
0
u
t
e
%
1
<
C
Fi(. #)&<* C2897:+> v0 Ex9+-i8+.:,4 0:-,:i3i6,:i2. ,3:+- 22 127-0 32- ,44 1+,: i.97:0
*he stratification model was developed using e%perimental data for 9""1 !"" and
>""-Jm
!
. A detailed picture of the stratification model results for a heat input of
!""-Jm
!
applied to the system for up to eight hours is showed in 3igure >,9#. Each
wattage followed a 9F9 relationship showing good agreement between e%perimental and
computed data.
9!C
0
10
$0
30
%0
&0
0 10 $0 30 %0 &0
' )easure%1 <C
'
c
o
)
0
u
t
e
%
1
<
C
Fi(. #)&$* 2''W?8
2
>,:, ) C2897:+> v0 Ex9+-i8+.:,4 0:-,:i3i6,:i2. ,3:+- < 127-0 23 29+-,:i2.
'tatistical measures used for evaluation of the model
*he previously detailed statistical indicators were used to evaluate the performance of
the model to predict temperature stratification in the $&(,(-. and are summarised in
*able >,;.
T,54+ #)1'* S:,:i0:i6,4 i.>i6,:2-0 v,47+0 B T2:,4 D,:,
-tatisti+al indi+ator >alues
-lope 1.00
;
$
0.33
1B! <2= 40.%(
;1-! <2= 0.83
-0ewness <2= 40.1(
)urtosis 3.33
*he slope of the best fit line was found to be equal to one suggesting no over or under
estimation of the computed variable. *he high value of the coefficient of determination
+?
!
/ indicates a low une%plained variation and thus good correlation. *he BE value of
,".>: shows that the model has a tendency to over predict its computed values by nearly
half of one degree &elsius. *he low ?(E value of ".C; suggests a satisfactory actual
deviation. *he low s'ewness negative value of ,".9: is close enough to Dero to indicate
a robust model. *he positive 'urtosis value of ;.;# indicates a pea'ed distribution of the
errors suggestive of fewer outliers in the estimation.
9!#
Different heat inputs outside the normal e%perimental range were then simulated in
order to validate the model. ?esults for ="1 9="1 !="1 ;=" and B""-Jm
!
were plotted for
a typical day of eight hours in 3igure >,!". *he figure confirmed a good agreement
between e%perimental and computed data following a 9F9 relationship with an BE and
?(E of 9 and 9.> respectively and a slope and coefficient of determination of ".#:
suggesting an under estimation of the computed variable1 validating the stratification
method used.
0
10
$0
30
%0
&0
'0
(0
0 10 $0 30 %0 &0 '0 (0
' )easure%1 <C
'
c
o
)
0
u
t
e
%
1
<
C
&0#
1&0#
$&0#
3&0#
'00#
Fi(. #)<'* V,4i>,:i2. 23 :1+ 82>+4 ) C2897:+> v0 Ex9+-i8+.:,4 0:-,:i3i6,:i2. ,3:+- < 127-0 23
29+-,:i2.
1.1.1.33 Fi+4> ) 82>+4 -+074:0
*his section e%plores the validity of the model to generate and predict water bul'
temperature inside the $&(,(-. for field weather conditions. (imulations were
underta'en for an aluminium collector of ;mm thic'ness tilted at an inclination angle of
>= in order to compare computed results with e%perimental results underta'en on the
roof of Napier University as described in &hapter ;.
9;"
(esults
$n accordance with laboratory computed data1 the role of e%tended heat transfer fins was
demonstrated and showed good improvement by achieving higher bul' water
temperatures.
A first analysis carried out for a period of five days +99
th
, 9=
th
4uly !"":/ e%plores the
validity of the model to predict bul' water temperature inside the $&(,(-. by
comparing computed results with e%perimental data. ?esults are plotted in 3igure >,!9.
0
10
$0
30
%0
&0
'0
(0
1 ( 13 13 1 ( 13 13 1 ( 13 13 1 ( 13 13 1 ( 13 13
'i)e1 (rs
'
e
)
0
e
r
a
t
u
r
e
1
<
C
0
100
$00
300
%00
&00
'00
(00
800
300
1000
&
o
l
a
r
r
a
%
i
a
t
i
o
n
1
$
"
)
2
'w=8>0eri)ental 'w=Co)0ute%
'0=8>0eri)ental '0=Co)0ute%
'a &olar ra%iation
Fi(. #)<1* C2897:+> ,.> +x9+-i8+.:,4 >,:, 32- , 3iv+ >,/ 9+-i2> i. H74/ 2''&
3igure >,!9 shows that similar to e%perimental field results1 computed results show that
each day is characterised by a heating and a cooling profile which are dictated by solar
radiation. *he higher the incident solar radiation received by the absorber plate1 the
higher the water temperature computed. $t was observed that gradual heating and
cooling profiles due to the thermal mass of the system were e%perienced due to the
consideration of thermal mass in the macro model. .owever1 a lower cooling rate of the
modelled values was observed each day. *his could be e%plained by the propagation of
errors and uncertainty by the use of regression analysis and assumed weather data)
resulting in a decrease in accuracy of the model.
$t was also observed in 3igure >,!9 that e%perimental water temperature was followed
very closely by predicted water temperature although it showed a slight over,estimation
9;9
by the proposed model. $n order to e%amine the validity of the simulation computed data
were plotted as a function of e%perimental data in 3igure >,!!.
0
10
$0
30
%0
&0
'0
(0
0 10 $0 30 %0 &0 '0 (0
' )easure%1 <C
'
c
o
)
0
u
t
e
%
1
<
C
Fi(. #)<2* C2897:+> v0 Ex9+-i8+.:,4 32- , 3iv+ >,/ 9+-i2> i. H74/ 2''&
$t was demonstrated that e%perimental and computer simulation results were in good
agreement as they followed a 9F9 relationship shown in 3igure >,!!. (tatistical
indicators were then used to evaluate the performance of the model for those five days
and are resumed in *able >,>.
T,54+ #)11* S:,:i0:i6,4 i.>i6,:2-0 v,47+0
-tatisti+al indi+ator >alues
-lope 1.03
;
$
0.3(
1B! <2= 41.$(
;1-! <2= $.13
-0ewness <2= 0.8'
)urtosis 0.%$
*he slope of the best fit line suggests a slight over estimation of the computed variable.
*he high value of ?
!
indicates a low une%plained variation showing a satisfactory
accuracy for the regression model. *he BE value of ,9.!: shows that the model has a
tendency to over predict its computed values. *he ?(E value of !.9# shows an
acceptable actual deviation. *he low positive s'ewed distribution trends to the high end
9;!
of the scale and indicates a robust model. *he positive low 'urtosis indicates a pea'ed
distribution of the errors suggesting that there are low outliers in the estimation.
A study of the effect of wind velocity on the bul' water temperature was carried out and
is illustrated by 3igure >,!; where O"1 O!1 O> and OB represents the wind velocity
values of "mJs1 !mJs1 >mJs and BmJs respectively. *he higher the wind velocity1 the
higher the heat losses resulting in lower bul' water temperature showing the influence
of wind speed on overall performance. *hus1 having accurate wind speeds would result
in more accurate bul' water temperatures. *he more data input1 the more accurate
becomes the model.
$0
$&
30
3&
%0
%&
&0
&&
'0
'&
(0
0 8 1' 0 8 1' 0 8 1' 0 8 1' 0 8 1'
'i)e1 (rs
;
u
l
2
w
a
t
e
r
t
e
)
0
e
r
a
t
u
r
e
1
<
C
>0 >$
>% >'
Fi(. #)<3* Si874,:+> +33+6: 23 Ai.> v+426i:/ 2. A,:+- :+89+-,:7-+ 32- , 3iv+ >,/ 9+-i2> i. H74/ 2''&
A1+-+ V'; V2; V# ,.> V% -+9-+0+.:0 :1+ Ai.> v+426i:/ v,47+0 23 '8?0; 28?0; #8?0 ,.> %8?0
-+09+6:iv+4/
A second simulation was carried out over a period of =" days +:
th
4une , !:
th
4uly !"":/
in order to determine the validity of the model to predict bul' water temperature for an
e%tended period. *otal computed and e%perimental data are plotted in 3igure >,;> at the
end of the chapter.
&omputed data were then plotted as a function of e%perimental data to e%amine the
validity of the simulation. $t was demonstrated that e%perimental and computer
simulation results were in good agreement as shown in 3igure >,!>.
9;;
0
10
$0
30
%0
&0
'0
(0
0 10 $0 30 %0 &0 '0 (0
' )easure%1 <C
'
c
o
)
0
u
t
e
%
1
<
C
Fi(. #)<#* C2897:+> v0 Ex9+-i8+.:,4 32- , "' >,/ 9+-i2> i. H74/
*he same statistical indicators were then used to evaluate the performance of the model
to predict bul' water temperature in the $&(,(-. for the e%tended =" day test and are
given in *able >,=.
T,54+ #)12* S:,:i0:i6,4 i.>i6,:2-0 v,47+0
-tatisti+al indi+ator >alues
-lope 1.0%
;
$
0.3(
1B! <2= 41.%3
;1-! <2= $.$$
-0ewness <2= 0.$1
)urtosis 0.1'
*he slope of the best fit line suggests an over estimation of the computed variable. *he
high value of ?
!
indicates a low une%plained variation showing a good adequacy of the
regression model. *he BE value of ,9.># shows that the model has a tendency to over
predict its computed values by nearly 9.=S&. *he ?(E value of !.!! shows an
acceptable actual deviation. *he low positive s'ewed distribution tails off to the high
end of the scale while the positive low 'urtosis indicates a pea'ed distribution of the
errors suggesting that there are low outliers in the estimation.
9;>
?esults validated the field macro model as a robust model to predict bul' water
temperature in the $&(,(-. under different weather conditions.
#.1.< E--2- ,.,4/0i0
$nvestigating uncertainties and propagation of errors in the modelling procedure is
necessary to determine the quality of the data presented. Any sources of errors may be
categorised under the following two main categoriesF modelling uncertainties and
statistical disadvantages. A brief analysis of both categories is detailed below.
1.1.1.3# M2>+44i.( +--2-0
*he model was coded using numerous regressions that have a band of error. *he use of
a series of regressions results in the propagation and intensification in uncertainty of the
final result.
@ther uncertainties associated with assumptions made for the different type of model
and data used also intensify the uncertainty of the results. Assumption and data
uncertainties were analysed and are highlighted belowF
, Uncertainties of weather data recorded for simulation
, Uncertainty in material properties for calculation
, Uncertainty associated with the solar model of &lar'e et al +!"":/ used to incline
horiDontal solar radiation
, Uncertainty of estimation of the total enthalpy of the water when highly stratified
, Uncertainties due to air lea'age from the air cavity
, Uncertainty in calculation due to lateral stratification
1.1.1.3" Li8i:,:i2.0 23 0:,:i0:i6,4 +--2- i.>i6+0
Each statistical indicator has associated disadvantages which are important to address1
some of which are described belowF
, -ithin a dataset overestimation of one observation can over,compensate for another
+underestimate/1 ma'ing it difficult for BE to represent the true nature of the
accuracy of the model.
, ?(E fails to give a complete assessment of a model as only a few large errors can
increase its value substantially.
9;=
, ?
!
does not always measure the appropriateness of the model1 since it can be
artificially inflated by adding another parameter or higher order polynomial terms in
the regression equation.
, ('ewness does not account for the magnitude of errors) it only gives information on
the normality of their distribution.
, <urtosis can be very sensitive to outliers1 particularly if the data set involves
measured values. *hus1 only a few errors in the tails of the distribution1 resulting as a
deviation from erroneous or irrelevant observations1 can affect its value.
#.1.$ Di06700i2. 23 :1+-8,4 82>+40
1.1.1.3% L,52-,:2-/ 82>+4
A macro,model for fluid flow and heat transfer was developed using Oisual Basic for
Applications +OBA/. *he role of e%tended heat transfer fins was modelled and showed
good improvement by achieving higher bul' water temperatures. Effective fin heat
transfer rates were enhanced by the choice of a material of high conductivity such as
aluminium. *he effective use of the fin was found to be B:.;L and B;L for the
aluminium of ;mm and 9.=mm thic'ness and !=.;L for the 9.=mm thic' stainless,steel.
*he fin efficiency f
+=.!/
-here
L(A
.
is the energy input during the $&(,(-. life cycle +'-h/1
S
.
is the
annual useful energy saved by using the $&(,(-. +'-hJyear/ and
#S.
.
is the annual
energy employed during the use of the $&(,(-. +'-hJyear/.
*he use of a passive collector system or a solar driven pump was therefore assumed for
#S.
.
to be null. .ence EPB* were calculated for A2,;1 A2,9.= and (*,9.=. ?esults
are presented in *able =,C.
T,54+ ")2'* EPBT
C244+6:2- :/9+
L(A
.( (CW1)
S
.( (CW1?/+,-)
EPBT (/+,-0)
ST)1." 9B>B 9">9 9.B
AL)1." 9B:" 99": 9.=
AL)3 !C#: 99"# !.B
*he energy paybac' times related to the studied $&(,(-.s were found to be lower than
three years. *his value shows the great energy potential of such technology. A common
thought would be that the aluminium paybac' period should be significantly longer
because of its more intensive production process resulting in embodied energy value per
'ilogram of material being three times the stainless,steel one. .owever1 the recycling
property of aluminium decreases its embodied energy1 bringing down the energy
paybac' time considerably. 3uture scenario of C=L recycled and 9=L raw aluminium
material would lower the embodied energy of aluminium production therefore reducing
the EPB* of A2,9.= to 9" months.
9BB
".".# C,-52. >i2xi>+ +8i00i2. 9,/5,6C :i8+ B ECPBT
*he E&PB* was calculated based on the previously calculated annual emission savings
of &@
!
1
S
.(
1 in section =.=.!.!. *he E&PB* can be defined as the time necessary for
an $&(,(-. to equal the &@
!
emissions released during the production and use of the
installation itself as stated by Ardente et al +!""=b/. $t summarises the total impacts
during the life cycle and the emission savings. *he E&PB* is e%pressed in equation =.;.
#S. S
L(A
.( .(
.(
.(+BT
+=.;/
-here
L(A
.(
is the total &@
!
emissions related to the production of a ;m
!
$&(,(-.
during its life cycle +'g&@!/ represented in *able =,:1
S
.(
is the annual &@
!
emission
saving +'g&@
!
Jyear/ represented in 3igure =,B and
#S.
.(
is the annual &@
!
emissions
related to the $&(,(-. use +'g&@
!
Jyear/.
S
.(
represents the emissions saved by
using the $&(,(-. compared to the au%iliary system used in the household to deliver
hot water and therefore depends on the type of au%iliary heater employed.
*he use of a passive collector system or a solar driven pump was assumed therefore
#S.
.(
was set to null. E&PB* for A2,;1 A2,9.= and (*,9.= were calculated and
results are presented in *able =,# and 3igure =,:.
T,54+ ")21* ECPBT
C244+6:2- :/9+
L(A
.( (C(CO2)
S
.( (C(CO2?/+,-)
ECPBT (/+,-0)
G,0 E4+6:-i6i:/ G,0 E4+6:-i6i:/
ST)1." B"> 9#C >>C ;.9 9.;
AL)1." ;;! !9" >:B 9.B ".:
AL)3 =:C !99 >:: !.: 9.!
9B:
1.$
0.(
1.3
$.(
1.'
3.1
0.0
0.&
1.0
1.&
$.0
$.&
3.0
3.&
*as !le+tri+ity
Fuel
8
C
#
;
'
1
-
e
a
r
s
.luiniu +olle+tor 4 3
.luiniu +olle+tor 4 1.&
-tainless4steel +olle+tor 4 1.&
Fi(. ")1'1* ICS)SWH ECPBT* C,:+(2-i0,:i2. Ai:1 -+09+6: :2 37+4 :/9+
*he carbon dio%ide paybac' period depends on the au%iliary system used by the
household. 2i'e the energy paybac',time1 the &@
!
paybac',time results were generally
lower than three years.
*he per unit &@
!
evolution of electricity is much higher than natural gas resulting in
shorter environmental paybac' period for households using electricity as an au%iliary
heating system. *his is due to electricity having a high environmental impact due to the
fuel mi% used in its generation. E&PB* ranges from 9.B1 !.: to ;.9 years for the A2,9.=1
A2,; to (*,9.= respectively using gas while using electricity would result in a decrease
of E&PB* to C months1 9.! and 9.; years respectively) corresponding to a =BL
difference in E&PB* depending on the fuel type.
A considerable variation in the paybac' time of collectors of identical dimensions1 Al,
9.= and (*,9.=1 was observed. *his is e%plained by the recyclable property of
aluminium diminishing its embodied carbon and by its higher thermal conductivity
resulting in higher energy gain. A paybac' period based on future scenarios of C=L
recycled and 9=L raw aluminium material would result in lowering the E&PB* to 9"
and = months when gas and electricity au%iliary heating systems are used.
9BC
"."." C2.6470i2.0 ,.> -+6288+.>,:i2.0
An 2&A was carried out for each $&(,(-. type in order to establish the $&(,(-.
with the smallest carbon footprint. $t was found that despite the high amount of energy
to produce aluminium1 the MlightN and recyclable properties of the material resulted in
similar embodied energy when compared with stainless,steel. *his resulted in a smaller
environmental impact due to lower carbon emissions released to the atmosphere during
the production process.
Energy and carbon dio%ide savings associated by using those $&(,(-.s were modelled
and showed that although the A2,9.= $&(,(-. was out,performed by ".9L by the A2,
; $&(,(-.1 the fact that it only uses half the amount of material in producion ma'es it
more viable. *he carbon dio%ide savings showed that depending on the $&(,(-.
installed there is a potential for saving BL more 'g&@
!
per year by using an Al,9.=
compare with a (*,9.= and only ".=L by increasing the thic'ness of a same $&(,(-.
material.
*he EPB* of the $&(,(-.s was found to be consistently lower than three years for all
$&(,(-.s with the lowest EPB* value of 9 year : months achieved by the A2,9.=
$&(,(-.. *his value shows the great energy potential of such a $&(,(-. design. *he
calculations of the carbon dio%ide emission paybac' times confirm the higher potential
of using an A2,9.= $&(,(-. by decreasing the E&PB* by >!L and >CL when
compared with A2,; and (*,9.= $&(,(-.s respectively.
All results point to the A2,9.= $&(,(-. as the best alternative to operate in (cottish
weather conditions. $t would therefore be recommended to use an A2,9.= $&(,(-. for
further investigations due to its energy and environmental benefits. $ncreasing the
recyclable percentage in the A2,9.= could significantly reduce the environmental
impact of the collector. &opper and brass could be realistic alternatives to use in the
manufacture of the $&(,(-. in terms of embodied energy1 malleability and thermal
conductivity. .owever1 the cost of these materials being so much higher than that of
stainless,steel or aluminium would not ma'e them economically viable. Economics
plays a central role in any decision to purchase a solar water heater system. *he
customer is unli'ely to buy an $&(,(-. if they 'now that its only benefits are to the
environment. *herefore1 a monetary analysis was developed to consider the economics
of each system to ensure that such $&(,(-.s will be profitable.
9B#
".% M2.+:,-/ ,.,4/0i0 (MA)
*he monetary analysis +A/ can be defined as an economic evaluation technique to
determine the total cost of owning and operating a product with time. *he value of a
A lies in the ability to compare the cost of product alternatives and to determine
which one provides the best value per (terling pound spent. $n order to achieve an
appropriate A1 an insight of the collector material costs is provided1 followed by the
actual cost of the $&(,(-.s and a monetary paybac' time +PB*/ assessment.
".%.1 C244+6:2- 8,:+-i,4 620:0
$t is difficult to find a definite figure for the price of aluminium and stainless,steel as
they commonly fluctuate. Average figures obtained for the last year from ay !"": to
ay !""C would be T!#"" per tonne of stainless,steel and T9>>= per tonne of
aluminium. *he large price difference can have several possible e%planations. *he main
factor of the cost difference between the two metals is the added cost of machining
stainless,steel compared with aluminium) which is ductile and more easily machined.
-holesale metal suppliers EP( and 2E could guarantee costs up to three to four
months but not further than that period which reflects the frequency with which prices
vary. A comparison of the price trend of aluminium and stainless,steel over the last
three years is shown in 3igure =,C.
1000
$000
3000
%000
&000
'000
1
a
r
4
0
&
?
u
n
4
0
&
-
e
p
4
0
&
/
e
+
4
0
&
1
a
r
4
0
'
?
u
n
4
0
'
-
e
p
4
0
'
/
e
+
4
0
'
1
a
r
4
0
(
?
u
n
4
0
(
-
e
p
4
0
(
/
e
+
4
0
(
1
a
r
4
0
8
6ate
,
l
u
)
i
n
i
u
)
#
r
i
c
e
1
F
5
&
"
t
o
n
n
e
.luiniu -tainless4steel
Fi(. ")1'2* C20: 6289,-i02.
9:"
(tainless,steel has been the more e%pensive of the two materials but both materials have
been increasing at the same rate until recently. @ver the course of the last three years the
price of stainless,steel has risen beyond levels that anyone inside or outside the
stainless,steel industry could have forecasted. *here are a number of contributing
factors that have lead to these increased costs.
> : (hinese stainless:steel cons$mption
*here has been a recent and important increase in &hinese consumption of stainless,
steel which has caused a global price increase in spite of a consistent consumption for
the rest of the world. &hina has e%perienced phenomenal growth in the last five years
resulting in being the world5s largest stainless,steel consumer and has not yet reached
its pea'.
; : 'aw material shorta!e
(crap metal comes from waste stainless,steel items. oll +!""B/ announced a decline
from ;>.CL in !""> to ;!L by !"9" of the worldwide share of scrap in stainless,steel
production suggesting that scrap is being consumed at a far greater rate than it is being
created. *his current scrap shortage results in a notable increase in scrap prices. $ron ore
is another material which is currently in short supply due to the significant lac' of
investment in iron ore mines anywhere in the world e%cept for &hina and (outh
America. Another main reason is the shortage of nic'el ore supply1 a ma0or raw material
for stainless,steel production resulting in a high nic'el price. *hese raw material
shortages result in rising stainless,steel production costs.
; : Stainless:steel prod$ction
According to the $nternational (tainless,steel 3orum , $((3 +!"":/ global stainless,steel
production decreased by !.#L in !"":. *his happened across most regions and
countries e%cept for &hina and the Asian region accounting for B"L of the worldHs
stainless,steel production. A strong increase in stainless,steel production occurred in the
first half of !"": followed by a collapsed in the second half of the year after nic'el
prices dropped resulting in stainless,steel being withdrawn from the first half of !"":
stoc's.
? : Shippin!
9:9
Another item influencing stainless,steel prices is a shortage of shipping capacity. @ne of
the factors behind the growth in international trade over the past decade is a reduced
shipping cost but now the trade has grown to such an e%tent that there is a shortage of
shipping available. *he higher cost of shipping is hampering the movement of raw
materials around the globe.
@ : .ner!y costs
Energy costs are another factor increasing stainless,steel production costs. Production
cost increases due to higher energy input costs +i.e. Electricity is an output from many
energy resources and is used to power most equipment and facilities/. Energy mar'ets
are e%periencing high demand and limited supplies1 resulting in volatile mar'et and high
prices. (ince energy is a 'ey input in the production process1 sharp increases per unit in
goods occur as stated by OelaDqueD +!""B/. (tainless,steel being an energy intensive
process uses significant energy input for their production resulting in higher costs.
".%.2 ICS)SWH0 620:0
*he price of installing a domestic flat plate collector solar water heating system in
the U< today is anywhere from T!1""" to T;1""" which would include delivery1
installation and associated equipments costs as stated by E(* +!"":/. *he ma0ority of
solar water heating systems installed would come with a = to9" year warranty and
require very little maintenance. *he costs of the materials used in the $&(,(-.s
investigated for this pro0ect are detailed in *able =,9". Prices were updated to most
recent material cost data collected in ay !""C.
T,54+ ")22* R,A 8,:+-i,4 620:0 9+- 0=7,-+ 8+:+- 23 6244+6:2-
C244+6:2- :/9+ C244+6:2- 620: (Q) M,:+-i,4 620: (Q) ICS)SWH 620: (Q)
S:,i.4+00)0:++4
1."88
;=! 9"= >=:
A478i.i78
1."88
C9 9"= 9CB
A478i.i78
388
9B9 9"= !BB
*he difference in price of the two materials is a reflection of the recent surge in
global stainless,steel prices. Aluminium prices were ta'en from 2ondon etal
E%change +2E/ while stainless,steel ;">b prices were ta'en from EP( $nternational
2td. -hile the cost of materials may not be very high1 the additional labour and
9:!
delivery costs1 along with other required parts such as a pump1 controller1 valves and
piping would raise the price of parts up to the region of T91""".
*o encourage the public to become involved in the effort to reduce carbon emissions1
government grants for homes that install these heating systems are made available. *he
U< government 2ow &arbon Buildings Programme +2&BP/ offers a T>"" grant for
every professionally installed solar water heating system as stated by the D*$ +!"":/.
Another grant available in (cotland from the (cottish &ommunity .ouseholder
?enewables $nitiative +(&.?$/ provides ;"L of the installed cost up to a value of
T>1""" as stated by the Energy (aving *rusts +!"":/.
".%.3 M2.+:,-/ 9,/5,6C :i8+ (MPBT) ,.,4/0i0
*he monetary paybac' time +PB*/ can be defined as the time necessary for the $&(,
(-. to achieve the monetary equivalent to its capital cost and is e%pressed in equation
=.>.
< S
((
< <
<
<+BT
+
+=.>/
((
<
is the capital cost associated with the $&(,(-. life cycle +T/1
S
<
is the annual
monetary saved by using the $&(,(-. +TJyear/ and
<
<
is the annual associated cost
of using $&(,(-.s such as operational and maintenance costs +TJyear/.
7ull $ndustries +!""C/ also suggested that $&(,(-.s are virtually maintenance free
although the glaDing needs to be cleaned and the unit to be drained once a year1 as is
recommended for au%iliary water heaters. $t was then assumed for this study that annual
associated costs could be neglected1 therefore setting
<
<
to null.
*he potential annual monetary savings1
S
<
1 of a ;m
!
$&(,(-.s installation vary
considerably between homes using au%iliary gas fuelled or electric heating systems. *he
roller,coaster profile of energy prices in the U< will have a recognised impact on the
study findings. Data of 9!.9pJ'-h for electricity and ;.BpJ'-h including OA* for gas
prices refers to domestic standard prices of (cottish Power and British 7as prices
respectively in (eptember !""C. Based on this reference prices1 savings achieved by
9:;
using each $&(,(-.s were calculated. Annual savings1
S
<
1 were calculated from
equation =.=.
+ S S
. <
+=.=/
-here
S
.
is the useful annual energy saved by using the $&(,(-. +'-hJyear/ and
+