Professional Documents
Culture Documents
Ge 200038727
Ge 200038727
Ge 200038727
To cite this article: S. B. Riffat , X. Zhao , R. Boukhanouf & P. S. Doherty (2005) Theoretical and
Experimental Investigation of a Novel Hybrid Heat-Pipe Solar Collector, International Journal of Green
Energy, 1:4, 515-542, DOI: 10.1081/GE-200038727
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
INTERNATIONAL JOURNAL OF GREEN ENERGY
Vol. 1, No. 4, pp. 515–542, 2004
ABSTRACT
This article describes a novel flat plate heat-pipe solar collector, namely,
the hybrid heat-pipe solar collector. An analytical model has been developed
to calculate the collector efficiency as well as simulate the heat transfer processes
occurring in the collector. The effects of heat pipes/absorber, top cover, flue gas
channel geometry, and flue gas temperature and flow rate, on the collector
efficiency were investigated based on three modes of operation, i.e., solar only
operation, solar/exhaust gas combined, and solar, exhaust gas and boiler
combined. Experimental testing of the collector was also carried out for each
of these modes of operation under real climatic conditions. The results were used
to estimate the efficiency of the collector and determine the relation between
the efficiency and general external parameter. The modeling and experimental
results were compared and a correlation factor was used to modify the theoretical
predictions. It was found that the efficiency of the collector was increased
by about 20–30% compared to a conventional flat-plate heat pipe solar collector.
Key Words: Heat pipes; Solar collector; Flue gas; Efficiency; Simulation;
Testing.
515
1. INTRODUCTION
Solar collectors transform solar radiation into thermal energy. There are several
types of solar collector available for practical applications, including evacuated
tubes, flat plate solar collectors, and parabolic dish collectors.
Heat pipes are devices of high thermal conductance, which transfer thermal
energy by two-phase circulation of fluid, and can easily be integrated into most types
of solar collector. The basic difference in thermal performance between a heat-pipe
solar collector and a conventional one lies in the heat-transfer processes from
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
the absorber tube wall to the energy-transporting fluid. In the case with a heat pipe,
the process is evaporation–condensation–convection, while for conventional solar
collectors, heat transfer occurs only in the absorber plate. Thus, solar collectors with
heat pipes have a lower thermal mass, resulting in a reduction of start-up time.
A feature that makes heat pipes attractive for use in solar collectors is their
ability to operate like a thermal-diode, i.e., the flow of the heat is in one direction
only. This minimizes heat loss from the transporting fluid, e.g., water, when incident
radiation is low. Furthermore, when the maximum design temperature of the
collector is reached, additional heat transfer can be prevented. This would prevent
over-heating of the circulating fluid, a common problem in many applications
of solar collectors (Bienert and Wolf, 1976; DeVriers et al., 1980).
Integration of heat pipes also provides the advantage of high heat transfer in the
latent form. Heat pipes are self-contained devices and can be used for collection and
transfer of energy in solar collectors. Solar collectors used for heating of liquids may
incorporate the evaporator sections of the heat pipes as part of the flat plate
collector, and the condenser section inserted in the storage tank containing the liquid
to be heated. The system therefore needs no piping or fittings between the storage
tank and the collector, and this reduces both the cost and the heat loss of the system.
If the working fluid of the heat pipes is chosen to have a low freezing point,
the collector may offer freeze protection (Ismail and Abogderah, 1998).
One of the first studies of heat pipes in solar applications was carried out
by Bienert and Wolf (1976). In this case the evaporator end of a heat pipe was
inserted in a flat-plate collector, and the condenser protruded into a water manifold
attached to the upper end of the collector. Their results were neither conclusive nor
optimistic. Since then numerous studies have been carried out, including theoretical
analysis and calculation (Franken, 1979; Hussein et al., 1999; Lu and Guo, 1984;
Ortabasi and Fehlner, 1979), experimental testing (Bairamov and Toiliev, 1981;
El-nasr et al., 1993; Ramsey, 1986; Ribot and McConnell, 1983; Riffat et al., 2000;
Terpstra and Van Veen, 1987; Zanardi, 1989), combined investigation through both
theoretical analysis and experimental trials (Ernst, 1981; Ismail and Abogderah,
1991, 1992a, b, 1998), as well as a number of applications in practice (Balzar et al.,
1996; Bong et al., 1993; Chun et al., 1999; Mathioulakis and Blessiotis, 2002). It was
found that most of the previous works involved with investigation of thermal
performance of different types of heat pipe solar collectors by using analytical,
numerical, or experimental methods. One of the major aims was to design suitable
structures or system layouts, as well as establish favorite operating conditions to
enable a high efficiency to be achieved. Of existing collector designs, evacuated tube
and flat plate collectors are most widely used and the former is usually found to be
Novel Hybrid Heat-Pipe Solar Collector 517
most efficient for high temperature applications. Flat plate heat-pipe solar collectors,
on the other hand, have their own set of advantages, including simpler structure,
lower cost, easier manufacture, and simple operation.
The lower efficiency of flat plate collectors is mainly due to the heat loss via the
cover surface due to conduction and convection. Standard flat plate collectors have
efficiencies of typically 50% or less (Riffat et al., 2000), while evacuated devices
have efficiencies of about 50–80% (Ramsey, 1986; Ribot and McConnell, 1983;
Zanardi, 1989). It would be desirable to develop a new structure for flat plate
collectors that would overcome heat loss problems and allow a high efficiency to be
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
2. EXPERIMENTAL SET-UP
which would result in higher solar absorption and relatively larger heat losses.
The bottom chamber of the enclosure contained a number of channels to allow
exhaust gas discharged from a boiler to pass through them to improve the efficiency
of the collectors, as well as provide back-up (or compensation) when solar
irradiation is unavailable (or low). Fibre-glass insulation was fitted beneath and
around the channels to reduce heat loss through the chamber. As heat from both the
sun and boiler exhaust gas is used, the unit is termed a ‘‘Hybrid Heat Pipe Solar
Collector.’’ Schematic diagrams showing the structure of the collector are given
in Figs. 1a–c.
The embedded panel included 20 heat pipes, and had a gross absorbing area
of 2.4 m2. The heat pipes had integrated bodies and therefore no physical separation
between the evaporators and condensers. The manifold was connected to the
condensers of the heat pipes using copper rods, which were drilled with co-axial
cylindrical cavities. Heat transfer between the condensers and the manifold was
largely dependent on the conductivity of the rods. The structure is shown
schematically in Fig. 2. Photograph of the collector is shown in Fig. 3.
The prototype collector was mounted outside a laboratory building of the
University of Nottingham, adjacent to a shed housing the boiler that was to be used
as part of heat source of a CHP system. The collector was co-operated with the
boiler to supply heat for the CHP system. It was intended to direct the exhaust flue
gas from the boiler across the flow channels fitted to the backside of the solar
collector to enhance its heat transfer, and provide compensation, or back up, when
Novel Hybrid Heat-Pipe Solar Collector 519
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
solar irradiation was low or unavailable. However, it was not possible to carry this
out in practice for safety reasons. In order to simulate the intended situation, a flue
gas simulator was used to generate hot air equivalent to gas flow. T-type
thermocouple probes were mounted on the inlet and outlet of the collector
to measure water temperature, and a single jet water meter was mounted on the
pipeline connected to the collector to measure water flow rate. A normal T-type
thermocouple was hung in air to measure outdoor temperature, and a pyranometer
was fixed to the absorber surface of the collector to measure instant global solar
radiation. Each of item measurement instrumentation was connected to a Datataker
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
(DT 500) which acquired data from the measured points and transferred these
to a computer. The collector was connected to the main flow system that included
the boiler and CHP test rig. When CHP operation stopped, the boiler also ceased
operation. The layout of the integrated system, including the outdoor and indoor
parts, is shown in Figs. 4a and 4b.
The analytical model focused on the heat transfer processes in the prototype
heat pipe solar collector. In fact, heat transfers exist in three major parts of the
collector prototype, i.e., the top cover, the absorber (evaporator plate), and
the condenser/manifold. These heat transfers finally achieve balance themselves and
are inter-linked by a distributed temperature profile. In the modeling development,
a few assumptions were made in order to simplify the analysis, including:
. A steady state condition has been achieved and heat balance exists in each
component and whole area of the collector prototype.
. The absorber has uniform temperature distribution over its surface area,
such that a uniform heat input is exerted onto it.
. The top cover has uniform temperature distribution over its surface area,
such that a uniform solar irradiation strikes on its outer surface.
. There is no heat loss on the edge area of the absorber where the top cover and
the bottom chamber are covered, due to a good insulation applied.
. The multi-flue gas channels, which are adjacent to each other, are treated
as a continuous, straight channel.
For a given collector area and total solar irradiation, the heat striking the top
cover surface is partly absorbed by the cover, partly transmitted through the cover
and reaching the absorber, and the remaining is reflected to the atmosphere. This
heat transfer process can be expressed as:
Qinc ¼ Qabs þ Qref þ Qtra ð1Þ
Novel Hybrid Heat-Pipe Solar Collector 521
Solar
collector
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
(a)
CHP
system
(b)
Figure 4. Layout of the integrated hybrid solar collector/CHP system. (a) Outdoor part. (b)
Indoor part.
522 Riffat et al.
The absorbed heat will be dispersed to the surroundings or (and) absorber by ways
of convection, conduction, or radiation, to achieve a heat balance relative to the top
cover. This balance may be expressed as:
Qabs ¼ Qtcab Qtca ð2Þ
Heat dissipation to the surroundings occurs mainly by the combined effect
of conduction and convection. However, heat transfer between the top cover and
absorber may be complex. If the absorber chamber were perfectly evacuated, heat
transfer between the top cover and absorber would only be induced by radiation.
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
If the chamber were not evacuated, heat transfer between the top cover and the
absorber will be a combined effect of conduction, convection, and radiation.
Heat reaching the absorber plate comes mainly from two sources: the sun and
the exhaust gas. Part of the heat will be transferred to the working fluid through the
flat plate and heat pipe wall, which causes the liquid to vaporise, and this is therefore
termed the effective heat input. The remainder will be dispersed to the environment
through the top cover and bottom casings, resulting in heat losses due to conduction,
convection, and radiation. If Qtcab is negative, the absorber has an upward heat
loss. If Qtcab is positive, the absorber has a heat addition from the top cover.
There are temperature differentials in the absorber area, which result in heat
transmission from the flat plate to the pipes, or from one part to another part of the
area. However, these differentials are small as both the plate and pipes are made
of copper—an excellent heat conductor. In order to simplify the thermal analysis,
the differentials are considered to be negligible and thus the absorber surface
is assumed to be at the same temperature over the whole area.
As the collector is fitted with channels allowing exhaust gas to pass through,
as shown in Fig. 1c, there will be heat transfer between the flue gas, absorber panel,
and surroundings. However, under the situation without gas flow, there will be heat
loss from the downside wall to surroundings, Qdwa. Thus, actual heat obtained
by the absorber (evaporator) plate is then expressed as:
Qab ¼ Qtra Qtcab Qdwa ð3Þ
In normal situations, the exhaust gas passes through a number of adjacent
rectangular channels and heat transfer occurs through the enclosure walls.
To simplify the analysis, the adjacent channels are treated as a continuous, straight
channel. The heat transfer between channel flow, absorber, and enclosure walls may
then be expressed as follows (Yang and Tao, 1998):
hf Aab tf ab þ hf Adw tf dw ¼ mf cp, f ðtf1 tf2 Þ ¼ Qf ab þ Qf dw ð4Þ
where tfab is the log mean temperature difference between the flue gas and
absorber, tfdw is the log mean temperature difference between the flue gas and
downside wall, and hf is the flue gas convective heat transfer coefficient and is largely
dependant on the flow of the flue gas in channels. Since this flow is normally
Novel Hybrid Heat-Pipe Solar Collector 523
If the gas flow is heated by the side-walls, then n ¼ 0.4. Otherwise, n ¼ 0.3.
The formula is most suitable for the situation with medium temperature difference,
i.e., 50 C or less, which matches the conditions for this application. In this case
the mean temperature of the gas flow is taken as the characteristic temperature, and
the inner diameter of the channel is taken as the characteristic length. For the
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
The heat obtained from the absorber, Qob, will be transported to the cooling
fluid passing across the manifold through evaporation and condensation of the
working fluid in the heat pipes. There are several thermal resistances in this process,
namely, the evaporator wall resistance, the equivalent resistance of the working fluid
and wick in the evaporator, the vapor flow resistance, the equivalent resistance of the
working fluid and wick in the condenser, and the condenser wall resistance. These
524 Riffat et al.
resistances can be calculated using the equations given in Dunn and Reay (1982).
The total resistance would be the sum of the individual resistances.
For a single heat pipe, heat transportation from the evaporator outer surface
to the condenser outer surface may be written as:
The heat will be transferred to the cooling liquid by heat conduction through
the copper rods and manifold wall, and heat convection between the manifold wall
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
and the cooling liquid. The cooling liquid will be heated when flowing through
the manifold channel, which is tightly fixed to the copper rods and heat pipe
condensers by means of clamps. For an inlet temperature given as t0, a temperature
increase t1, (t1t0), will be achieved after the fluid passes around the first heat pipe
due to heat absorption from the pipe. The fluid temperature increases gradually
along the flow direction due to continuous heat transfer from the parallel-array
of heat pipes. The heat transfer between a single heat pipe and the cooling liquid may
be expressed as:
where hcl is the convective heat transfer coefficient of the cooling fluid, which
is largely dependent on the velocity of fluid passing over the surface, and the
cross-sectional area, as well as the geometry of the flow channel. For the collector
indicated above, the flow of the cooling liquid flow and the manifold geometry are
shown schematically in Fig. 5. The flow may be treated as half of the annular flow.
To solve for the convective heat transfer coefficient, hcon, the channel needs to be
treated as an annular geometry rather than a semi-annular one, and correspondingly,
heat flow from the inner wall needs to be doubled to comply with this treatment.
a1
Manifold a2
Cooling liquid
Figure 5. Schematic diagram showing cooling liquid flow and manifold geometry in the
hybrid heat pipe solar collector—wicked type.
Novel Hybrid Heat-Pipe Solar Collector 525
Heat transfer through the outer walls was negligible as a satisfactory insulation
was provided. For both situations, calculation of hcl could be carried out using
the annular flow model (Incropera Frank and De Witt David, 1997)
hcl Dhy,con
Nucl ¼ ð10Þ
kcon
of the equivalent annular flow channel, respectively. Cooling liquid flow in the
manifold is taken as fully developed laminar flow, which has a Reynolds number
less than 400 due to its very low velocity and the relatively large cross-sectional area.
Hence, Nucl is known as a constant for the given geometry (Frank and De Witt
David, 1997).
For a single heat pipe, given the inlet temperature ti1 and the outlet
temperature ti may be obtained by solving Eq. (11). For the whole condenser/
manifold configuration, the overall heat transfer may be expressed as:
Qcon ¼ Qcon,1 þ Qcon,2 þ þ Qcon,n ¼ Cp,cl mcl ðtn t0 Þ ð12Þ
where Qcon should be equal to Qob according to the principle of energy balance.
Equations (8)–(12) may be used to obtain solutions for the outlet temperature and
the mean temperature of the cooling water, as well as the temperatures in different
parts of the heat pipe panel.
The three heat transfer processes described previously are actually inter-linked,
and hence an iterative numerical procedure is implemented for the solution as stated
below:
. Heat balance of the top cover could be analysed by using Eqs. (1) and (2),
which results in the surface temperature of the top cover.
. Heat balance of the absorber (evaporator) plate could be analysed using
Eqs. (3)–(7), which results in the absorber heat gain, Qab.
. Heat balance of the heat pipes, and condenser/manifold pair could
be analyzed using Eqs. (8)–(12), which results in the heat gain of the
526 Riffat et al.
5. If Qab Qcon > 1 (error allowance), then increase ts by 0.1 C, and return
to step 4 for recalculation.
6. If Qab Qcon < 1 (error allowance), then decrease ts by 0.1 C, and return
to step 4 for a re-calculation.
7. If 1 (Qab Qcon) 1, heat balances in the whole system, as well
as different areas of the system, are achieved.
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
8. Calculating the cooling water temperature at the outlet and different points
along the flow channel, as well as the temperatures at different areas of the
collector.
9. Program stops.
The data obtained by running the computer program could be used to evaluate
the thermal performance of a solar collector. For a normal solar collector,
its performance is usually evaluated using efficiency which is defined as the ratio
of the heat taken from the manifold by the cooling liquid to the incident irradiation
striking the collector absorber. However, in the case of the hybrid solar collector,
another efficiency, 1, will also be used for performance evaluation. This efficiency
(1) is defined as the ratio of the heat taken from the manifold by the cooling liquid
to the sum of the incident irradiation striking the collector absorber and the heat
from exhaust gas. In this case, exhaust gas heat means the potentially available
sensible heat contained in the exhaust gas, which is measured relative to the
temperature of the ambient air, and may be expressed as:
Qf ¼ Cp, f f Vf ðtf ,i ta Þ ð13Þ
The efficiency of the hybrid solar collector is also influenced by the channel
geometry and flue gas properties, such as mass flow rate, and inlet and outlet
temperatures. Both and 1 are used to evaluate the performance of the collector.
However, to enable performance comparison between normal collectors and the
hybrid one, is normally used as the major index.
The performance of the hybrid heat pipe solar collector was simulated using
the model developed. The top covers were designed with 2 different options,
i.e., double-glazing or 10 mm polycarbonate sheet, and their solar optical as well as
other thermal parameters are given in Table 1. The characteristic parameters of the
heat pipes are detailed in Table 2.
The simulation was based on three modes of operation, i.e., solar-only
operation, solar/exhaust gas combined operation, and solar, exhaust gas, and
boiler combined operation. The simulation conditions are outlined in Table 3.
For solar-only operation, vs. (tmeanta)/In relations are shown in Fig. 6.
Although the polycarbonate cover has a higher overall heat transfer coefficient
U value (3 W/m2. C) than the double-glazing (1.1 W/m2. C), which may lead to
a larger heat loss to the ambient, the collector with the polycarbonate cover still
performed better than the one with the double-glazing cover. This was due to the
528 Riffat et al.
Table 1. Specifications of the top covers and their solar optical and thermal parameters.
Cover condition
1. Double-glazing cover 0.5 0.21 0.29 0.88 1.2
with a chamber 0.18
un-evacuated
2. 10 mm polycarbonate 0.85 0.07 0.08 0.88 3
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
Tests were carried out for collectors with a double-glazing cover and with
a polycarbonate cover under three modes of operation, i.e., solar-only, solar/exhaust
gas combined as well as solar, gas, and boiler combined. All the tests were carried
out under the real weather conditions in order to provide a realistic evaluation of the
collector performance. For the collector with a polycarbonate cover, tests were
carried out on three separate days in order to examine the reliability and
repeatability of its performance. The collector with the polycarbonate cover was
selected for this trial as it is the one most likely to be accepted for future application.
It was found that the collector system was unsteady during the first hour
operation, as part of the collected heat was used to warm up the collector body.
To observe the collector’s real performance and enable comparison between
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
Air heat resistance of top cover 0.12 Diameter of heat pipe(outer) 0.008 for wicked
inner surface (Rtc,i, m2. C/W) in condenser section (dhp,o, m) 0.02 for wickless
Air heat resistance of top cover outer 0.06 Thermal conductivity of heat pipe 401
surface (Rtc,o, m2. C/W) wall (khp,w, W/m2. C)
Thermal conductivity of insulation 0.005 Thickness of heat pipe wall (hp,w, m) 0.00376
layer (kins, W/m2. C)
Thickness of insulation layer (ins, m) 0.025 Thermal conductivity of liquid film 0.5 for wicked
on heat pipe inner wall (khp,lw, W/m2. C) 0.68 for wickless
Thermal conductivity of downside 177 Diameter of vapor flow cross section 3.752e 03
wall (kdw, W/m2. C) in evaporator (dvap,evap, m)
Thickness of dowside wall (dw, m) 0.002 Diameter of vapor flow cross 3.752e 3 for wicked
section (dvap,con, m) 1.094e 3 for wickless
529
530 Riffat et al.
Case
Solar only 700 9 168 10–25 — —
Solar–gas 720 11 168 9–24 0.035 92
Solar–gas–boiler 400 11 168 88–100 0.035 114
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
70
Efficiency,%
65
60
y = −154.88x + 57.28
R2 = 0.7483
55
50
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
(tmean−ta)/In ° 2
, C.m /W
the simulation and experimental results, test results were recorded one hour after the
system started up. To remain at a stable condition for collector operation, the volume
of the system is made sufficiently large (about 8 m3) compared to the measured flow
rate (2.8 L/m) through the collector. Moreover, to avoid the random error caused
by instant measurement values, such as the variation in the solar irradiation, average
values for the measured parameters in 10 min interval were used in efficiency
calculation. Under these conditions, the operation of the collector system can be
treated as quasi-steady and their testing efficiencies would become comparable with
the simulation results.
In solar-only operation, no flue gas flowed through the collectors. The heating
processes for the collectors with the double-glazing cover and with the polycarbonate
cover were recorded to show the variation of the solar irradiation, water inlet
and outlet temperatures, and the outdoor temperature during the periods of testing.
The thermal efficiencies of the collectors were calculated using the experimental
Novel Hybrid Heat-Pipe Solar Collector 531
eff1
80
Efficiency , %
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
60
40
y =−306.8x + 45.048
R 2 = 0.9554
20
0
0.004 0.008 0.012 0.016 0.02 0.024
(tmean−ta)/In, °C.m2/W
Efficiency − (tmean−ta)/In relation for the solar, gas, and boiler co-operated case
55
50
45
40
y =−693.82x + 183.26
35 R2 = 0.8979
Efficiency,%
30
eff
25
eff1
20
15
y = −152.47x + 40.273
R2 = 0.8979
10
0
0.195 0.2 0.205 0.21 0.215 0.22 0.225
(tmean−ta)/In, °C.m2/W
Figure 8. (1)(tmeanta)/In relations for solar, gas, and boiler combined operation.
532 Riffat et al.
40
Efficiency, %
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
35
30
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05
(tmean−ta)/In, °C.m2/W
Figure 9. (tmeanta)/In relation for the collector with double-glazing cover (solar only
operation).
55
50
Efficiency, %
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
45
40
test 1
test 2
test 3
35
30
0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045
(tmean−ta)/In, °C.m /W
2
Figure 10. (tmeanta)/In relation for the collector with polycarbonate cover (solar only
operation).
80
70
60
Efficiency, %
50 eff, test 1
eff1, test 1
eff, test 2
40 eff1, test 2
eff, test 3
eff1, test 3
30 P ( ff t t 1)
20
10
0
0.004 0.008 0.012 0.016 0.02 0.024 0.028
Figure 11. (1)(tmeanta)/In relation for the collector with polycarbonate cover (solar–gas
combined operation).
534 Riffat et al.
90
80 eff - test1
eff1- test1
eff - test2
70 eff1- test2
eff - test3
Efficiency, %
60 eff1- test3
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
50
40
30
20
10
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
(tmean−ta)/In, °C.m2/W
Figure 12. (1)(tmeanta)/In relation for the collector with polycarbonate cover (solar–gas
combined operation).
required for a CHP system; the flow rate of water passing across the collector
manifold was only 2.8 L/m. The collector was connected to the return flow pipes
to allow part of the return water passing through the manifold to absorb heat from
the solar insolation and exhaust gas. This part of water was then merged with the
remainder before being delivered to the boiler for further heating. Tests were carried
out at similar methods used previously and results for efficiencies are shown
in Fig. 12. When the boiler was not operating, the value of (tmeanta)/In was lower
(below 0.05) and the efficiencies, and 1, were relatively higher ( in the range
60–90% and 1 10–30%). Once the boiler was started, the value of (tmeanta)/In
increased immediately, and the efficiencies fell accordingly. At steady state
operation, was in the range of 20–25%, and 1 5–10%. This indicates that
combined operation would result in a significant reduction of the collector efficiency
due to the increased inlet water temperature, which in turn led to reduced heat
transfer between the collector absorber and the condenser.
Comparison was carried out between the modeling and experimental results
for the hybrid heat-pipe solar collector, for each of the three operation modes.
The results of these comparisons are summarised in Figs. 13–15, respectively.
For each mode of operation, the testing efficiencies are the average values of
the efficiencies of three separate tests.
Novel Hybrid Heat-Pipe Solar Collector 535
70
50
40
30
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05
(tmean−ta)/In, °C.m /W
2
Figure 13. Comparison of the modeling and testing results for solar only operation.
100
60
40
20
0
0.004 0.008 0.012 0.016 0.02 0.024 0.028
(tmean−ta)/In,°C.m2/W
Figure 14. Comparison of the modeling and testing results for solar and flue gas combined
operation.
536 Riffat et al.
80
Efficiency, %
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
60
40
20
0
0 0.04 0.08 0.12 0.16 0.2 0.24
(tmean−ta)/In,°C.m2/W
Figure 15. Comparison of the modeling and testing results for solar, flue gas, and boiler
combined operation.
For solar-only operation, solar/gas combined operation, or solar, gas, and boiler
combined operation, the theoretical efficiencies are always higher than
the corresponding testing efficiencies. The differences are in the range 10–20%.
The reasons for this were investigated, and it was found that a number of factors
contributed to the discrepancy. First, the heat pipe panel was subject to a small edge
heat loss as it was fixed onto a metal frame that would conduct part of the absorbed
heat away from the panel. Second, the insulation of the chamber and manifold was
not as good as expected, which contributed to an increased heat loss. Finally, several
assumptions were made when developing the model, e.g., an identical temperature
layout was assumed to exist on the absorber area, and a direct straight through
flow channel was assumed to represent the actual multiple return channels for
the simplicity of the analysis. These would all contribute to the difference between
the model and experimental results.
Even so, the modeling results and experimental data showed a very similar trend
for efficiency variation. Overall, efficiencies (either or 1) decreased with increasing
(tmeanta)/In, but rates of variation differed for the three operation modes.
Combined solar/gas and boiler operation showed the highest rate of decline, but
little difference is observed for the other two modes.
Although differences existed between the modeling and experimental results,
the model was able to predict the collector performance if a correlation factor
was provided. This factor was determined by dividing the experimental value by
the modeling value under the same (tmeanta)/In, and was still a function
of (tmeanta)/In, as shown in Figs. 16–18 for the three modes of operation.
Novel Hybrid Heat-Pipe Solar Collector 537
0.8
Correlation factor
0.75
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
0.7
0.65
0.6
y = 245.35x2 − 11.754x + 0.7366
0.55
0.5
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05
(tmean−ta)/In,°C.m2/W
Figure 16. Correlation factor vs. (tmeanta)/In relation for solar only operation.
0.85
cfe,correction factor of eff
0.8
cfe1,correction factor of eff1
Correlation factor
0.75
0.7
0.65
0.5
0 0.005 0.01 0.015 0.02 0.025 0.03
(tmean−ta)/In,°C.m /W
2
Figure 17. Correlation factor vs. (tmeanta)/In relation for solar–gas combined operation.
538 Riffat et al.
0.7
Correlation factor
0.6
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
0.5
0.3
0.1
0
0 0.04 0.08 0.12 0.16 0.2 0.24
(tmean−ta)/In,°C.m2/W
Figure 18. Correlation factor vs. (tmeanta)/In relation for solar, gas, and boiler combined
operation.
5. CONCLUSIONS
A novel flat plate heat-pipe solar collector, namely, a hybrid heat pipe solar
collector, has been designed and constructed. The collector was designed to allow
heat from both solar and boiler exhaust gas to be collected with the aim of improving
the thermal performance of the collector while keeping the capital cost low.
The efficiency of the collector was found 20–30% collectors.
A theoretical model was developed to analyse the heat transfers occurring in the
collector. The heat transfers in different areas of the collectors were investigated, and
these were linked together by a set of heat balance equations. In the model,
two efficiencies were defined to evaluate the thermal performance of the collector.
The efficiencies were found to be affected by collector configuration, including the
heat pipes and absorber panel, the top cover, as well as the external conditions,
including global solar irradiation, ambient temperature, cooling fluid inlet
temperature, and mass flow rate. The external conditions could be described using
a general external parameter (x).
Laboratory tests were carried out to validate the model predictions and provide
experimental results on the thermal performance of the collectors. For the hybrid
heat-pipe solar collector, a set of tests was carried out under real climatic conditions
by connecting it to a CHP/boiler system. Three modes of operation, i.e., solar-only,
combined solar/exhaust gas, and combined solar, exhaust gas, and boiler operation,
were examined.
Novel Hybrid Heat-Pipe Solar Collector 539
Modeling and experimental results were compared. For the hybrid heat-pipe
solar collector under each of the three operation modes, the theoretical efficiencies
were always higher than the corresponding experimental efficiencies. The differences
were in the range 10–20%, and the reasons for this were mainly due to several
factors, including edge heat loss, unsatisfactory insulation, and the simplified
assumptions made in the model formulation. The model was able to predict
the collector performance if an appropriate correlation factor was provided.
The factor was determined by dividing the experimental result by the corresponding
model prediction under the same conditions, and was still shown as a function
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
of the general external parameter (x). These correlations could be applied to any
circumstance for this type of collector.
NOMENCLATURE
A Area (m2)
Cp Specific heat (J/kg C)
d Diameter (m)
h Convection heat transfer coefficient (W/m2 C)
In Global solar irradiation (W/m2)
k Thermal conductivity (W/m C)
V Volume flow rate (m3/s)
L Length (m)
m Mass flow rate (kg/s)
n Number of the heat pipes included
Nu Nusselt number
Pr Prandtl number
Q Heat (W)
r Radius (m)
R Heat resistance (m2 C/W)
t Temperature ( C)
U Heat transfer coefficient (W/m2 C)
x General external parameter
Thickness (m)
Density of air (kg/m3)
Collector thermal efficiency—term 1
1 Collector thermal efficiency—term 2
Subscripts
0 Manifold inlet
1 Inlet
2 Outlet
a Ambient
ab Absorber
abs Absorption
540 Riffat et al.
ew Enclosure wall
f Flue gas
gas Gas from boiler
hp Heat pipe
hy Hydraulic
i Inside
inc Incident radiation
ins Insulation layer
lim Limit
max Maximum
mean Average value
o Outside
ob Obtained
ref Reflection
sa Still air in the enclosure
tc Top cover
tra Transmission
vap Vapour
ACKNOWLEDGMENTS
The authors would like to acknowledge the financial support provided for this
research by the European Commission, under the Energy, Environmental, and
Sustainable Development Programme.
REFERENCES
Bairamov, R., Toiliev, K. (1981). Heat pipe in solar collectors. In: Ray, D. A., ed.
Advances in Heat Pipe Technology. Oxford: Pergamon Press.
Balzar, A., Stumpf, P., Eckhoff, S., Ackermann, H., Grupp, M. (1996). A solar
cooker using vacuum-tube collectors with integrated heat pipes. Solar Energy
58:63–68.
Bienert, W. B., Wolf, D. A. (1976). Heat Pipe Applied to Flat-Plate Solar Collectors,
Final Report COO/2604-76/1.
Novel Hybrid Heat-Pipe Solar Collector 541
Bong, T. Y., Ng, K. C., Bao, H. (1993). Thermal performance of a flat-plate heat
pipe collector array. Solar Energy 50(6):491–498.
Chun, W., Kang, Y. H., Kwak, H. Y., Lee, Y. S. (1999). An experimental study of
the utilization of heat pipes for solar water heaters. Applied Thermal
Engineering 19:807–817.
DeVriers, H. F. W., Kamminga, W., Franken, J. C. (1980). Fluid circulation control
in conventional and heat pipe planar solar collectors. Solar Energy 24:209–213.
Dunn, P. D., Reay, D. A. (1982). Heat Pipes. 3rd ed. New York: Pergamon Press.
El-nasr, M. A., Nuglisch, J. (1993). Performance of a wickless heat pipe solar
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014
Terpstra, M., Van Veen, J. (1987). Heat pipes: construction and application.
A study of Patents and Patent Applications. New York: Elsevier Applied
Science.
Yagoub, W. K. A. (May 1999). A Review of Coating Materials for a Novel Membrane
Heat Pipe Solar Panel, Report to project Coordinator, University of
Nottingham.
Yang, S., Tao, W. (1998). Heat Transfer. China High Education Press,
pp. 162–165.
Zanardi, M. A. (1989). Numerical and Experimental Analysis of a Cylindrical
Downloaded by [The University of Manchester Library] at 05:27 10 October 2014