Author's personal copy

Available online at www.sciencedirect.com

Solar Energy 90 (2013) 43–50

www.elsevier.com/locate/solener

Cost effective design of compound parabolic collector

for steam generation
Ajitkumar S. Gudekar, Atul S. Jadhav, Sudhir V. Panse, Jyeshtharaj B. Joshi,
Aniruddha B. Pandit ⇑
Institute of Chemical Technology, Matunga, Mumbai 400 019, India

Received 2 January 2012; received in revised form 28 November 2012; accepted 27 December 2012

Communicated by: Associate Editor Brian Norton

Abstract

In this paper we present a working model of Compound Parabolic Collector (CPC) system for the application of process steam gen-
eration. It is easy for fabrication, operation and has a lower cost compared to other available concentrating solar collector systems with
further possibility of lowering the cost. An experimental demonstration unit having an aperture area of nearly 30 m2 was set up and
tested for steam generation. The performance analysis of the system shows potential of improving thermal efficiency up to 71%. By virtue
of its geometry, the proposed CPC system requires much lesser mirror area compared to conventional CPC design and require single tilt
adjustment per day for a daily 6 h operation.
Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Solar energy; Compound Parabolic Collector (CPC); Solar thermal system; Solar steam generation

1. Introduction ious applications like – (1) Unit Operations in Chemical

Industry, (2) Textile industry, (3) Polymer and paint indus-
Though electricity is the highest quality of energy and try, etc.
should be given priority, energy consumption in the form For a developed country like US, industrial sector con-
of direct heat also forms a major mode of energy consump- sumes about 40% of country’s commercial energy. Of the
tion. It is the most widely known form of energy. It is used total energy used by industry, a major portion, (approxi-
in various applications, from cooking and space heating to mately 45–65%) is used for direct thermal applications in
an extremely wide range of industrial applications. While the preparation and treatment of goods, as listed above,
at domestic levels it is used at lower temperatures, indus- and is known as Industrial Process Heat (IPH). The ther-
trial use requires much higher temperatures. Storage and mal energy for IPH, in general is below the temperature
transfer is an important consideration in the usage of heat. of 300 °C. the percentage of IPH demand utilized in the
Steam has been widely used over the years as a medium of temperature range of 92–204 °C is 37.2% of the total
heat transfer mainly due to the advantage of its high latent IPH. The largest share of the total IPH demand is currently
heat content. It is interesting to note that steam is just met by steam (Thomas, 1995).
another form of water, an entity most familiar to the man-
kind next to air (to be specific, oxygen). In industry, steam
is used as an economical and easy mode of heating for var- 1.1. A case of developing country like India

Considering India’s energy consumption pattern, indus-

⇑ Corresponding author. Tel.: +91 22 3361111; fax: +91 22 33611020. trial share is at 27.1% (Earth Trends, 2003). India’s pri-
E-mail address: [email protected] (A.B. Pandit). mary energy demand in 2005 was 537 Mtoe (Million

0038-092X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.solener.2012.12.020
 Author's personal copy

44 A.S. Gudekar et al. / Solar Energy 90 (2013) 43–50

Nomenclature

A aperture area of system, m2 e emissivity

Cp heat capacity of water, J kg1 K1 k latent heat of vaporization of water, kCal kg1
dT temperature rise for water in time ‘t’, K r Stefan–Boltzmann constant, W m2 K4
h heat transfer coefficient, W m2 K1
m steam mass flow rate, kg h1 Subscripts
mw mass (water + equivalent of piping etc.), kg A aperture
Qc1 rate of heat loss by convection, W g glass tube
Qc2 rate of heat loss by convection, W w water
Qrad rate of heat loss by radiation, W
S beam solar insolation, W m2 Abbreviations
t time, s CPC Compound Parabolic Collector
IMD Indian Meteorological Department
Greek symbols IPH Industrial Process Heat
h angle, ° Mtoe Million tonnes of oil equivalent
g efficiency, % OD outer diameter

tonnes of oil equivalent) and with the annual growth rate et al. (2012) for PV/Thermal hybrid collector) also offer a
of 3.2%, it was predicted to be 590 Mtoe in the year 2008 good option for applications in the temperature range up
by International Energy Agency (IEA, 2007). With the to 200 °C. The complexities involved and associated costs
share of IPH in industrial energy consumption as indicated in manufacturing evacuated absorber and additionally,
above, approximate steam requirement (below 200 °C) in making provisions for maintaining the vacuum with special
India can be calculated as to be of the order of means would be an important factor to be considered while
70,000 Tons/h. Looking at the available solar energy tech- commercialization. The research area which needed greater
nologies PTC seems to be the most suitable for this appli- investigation was the nature of the reflector to generate
cation (Thomas, 1995). Based on the report of Sargent and concentration. The standard parabolic reflectors have very
Lundy (2003) for PTC technology (Capital cost estimation small acceptance angle. A Compound Parabolic Collector
– Rs.15.13 Crore/MW in 2020) and the work by Eck and (CPC) has comparatively a larger acceptance angle. But
Zarza (2006), the cost of PTC technology for steam gener- CPC suffers from limitations because of multiple reflections
ation works out to be Rs.12,500/m2 aperture area and the and unwieldy size. In this work, modified CPC reflector
cost of most widely employed Scheffler dish technology curves were designed, fabricated and tested to overcome
(especially for cooking applications in India) is in the range these limitations. The main objective of the present work
of 18,000 Rs/m2. Thus, for India, there is a need for devel- was to develop various experimental models of the CPCs
oping a moderate temperature (6200 °C) and a low cost for steam generation in moderately low temperature range.
solar energy based technology for the annual energy It is to be noted here that, to understand the working of
requirements at levels of about 32.7 Mtoe. CPC systems, various models of CPC (including conven-
In this context, relatively stationary non-evacuated CPC tional, truncated CPC) were designed, fabricated and
solar collectors could be of great interest for thermal tested for water heating/steam generation. It was the sys-
energy supply of industrial processes heat if they were cost tem described below which gave the encouraging results
effective compared to parabolic trough and flat plate collec- for steam generation with reduced total system costs.
tors. Tripanagnostopoulos et al. (1999) fabricated an asym-
metric CPC collector with two separate absorbers in order 1.2. Basics of compound parabolic collectors
to absorb and trap maximum solar radiation. At the same
time they made the system cost effective compared to flat CPC are non-imaging concentrators and their potential
plate collector using low cost material, but with lower con- as collectors of solar energy was pointed out by Winston
centration and working fluid temperature less than 100 °C. and Hinterberger (1975). The basic shape of CPC is shown
Azhari and Khonkar (1996) tried to improve the efficiency schematically in Fig. 1A and its theory and working prin-
of the CPC system using modified absorber. They modified ciples can be found in the literature (Rabl, 1980). As seen
absorber by introducing two cavities in the appropriate from the figure, CPC is made of two halves of parabola
location for radiation trapping. Buttinger et al. (2010) with closely located focal points and their axes inclined
designed a CPC collector which encloses the collector to each other, such that rays incident within the angle
and absorber in an evacuated enclosure, thus increasing between the two axes (acceptance angle of the CPC) are
system performance but at higher cost. CPC designs with reflected with single or multiple reflections towards the
evacuated tube absorber (similar to one presented by Jiang region between the two focal points and get concentrated
 Author's personal copy

A.S. Gudekar et al. / Solar Energy 90 (2013) 43–50 45

(A) Axis of CPC

d1

Axis of ‘A’ Axis of ‘B’

Parabola A Parabola B

Truncated ‘A’ Truncated ‘B’

FA FB

d2

(B)
Parabola A Parabola B

Cylindrical
absorber

Fig. 1. (A) The geometry of conventional compound parabolic concentrator (CPC). d1: Aperture of CPC. hA: Acceptance angle of CPC. FA: Focus of
Parabola A. FB: Focus of Parabola B. d2: Receiver opening. (B) Schematic of CPC with cylindrical absorber.

in that region. Thus, CPCs can accept incoming radiation where hA is the acceptance angle of the CPC. The actual
over a relatively wide range of angles. By using multiple concentration ratio is usually lower than the ideal one.
internal reflections, any radiation that is entering the aper- Rabl (1980) gives tilt requirements of CPCs with different
ture, within the collector acceptance angle, finds its way to acceptance angles along with daily collection time.
the absorber surface located at the bottom of the collector.
As the upper part of a CPC (dotted line as indicated in 1.3. Modified design of CPC
Fig. 1A) contribute little due to a steep angle of incidence
to the radiation reaching the absorber, they are usually The design of CPC (as shown in Fig. 1) has two disad-
truncated thus forming a shorter version of the CPC (Rabl, vantages: (i) its height increases rapidly with aperture,
1976). In addition to the flat absorber (receiver) design dis- making the structure unwieldy to handle and (ii) a sizable
cussed above, CPCs with other absorber geometries can percentage of radiation incident within the acceptance
also be designed. One such design is a cylindrical absorber angle suffers multiple reflections before reaching the recei-
as shown in Fig. 1B. Cylindrical shape has the advantage of ver, resulting into a drop in its optical efficiency.
utilizing full surface for energy absorption unlike flat A modified version of CPC was designed, overcoming
absorber where the back side has to be insulated properly disadvantages of conventional CPC. Here, like conven-
to prevent/minimize heat loss. tional design, the axes of the two half parabolas were
A CPC concentrator is mostly orientated with its long inclined to each other, defining the angle of acceptance
axis along the east–west direction and for a location in and their foci were separated by a small distance; but,
northern hemisphere; its aperture is tilted towards south unlike conventional design, the two foci were very close
for most of the time of the year, such that the sun rays to the plane of the aperture. To overcome disadvantages
are incident on CPC aperture within the acceptance angle. of conventional CPC, segments of two parabolic curves
The tilt of the CPC may have to be adjusted periodically (forming CPC) above the focal point were removed and
when the incident solar radiation moves outside the accep- those below focal point were selected. As shown in
tance angle of the CPC. Fig. 2, the receiver pipe is located near the aperture, and
The ideal concentration ratio of a CPC is related to the its size was selected such that all rays incident through
acceptance angle by the angle of acceptance are captured by it after reflection.
The incident rays change from one extreme (Ray 1, Ray
1
CR ¼ ð1Þ 2) to other (Ray10 , Ray 2’) during the non-tracking period
sinð12 hA Þ for CPC. In the first extreme position (Ray1), they are par-
 Author's personal copy

46 A.S. Gudekar et al. / Solar Energy 90 (2013) 43–50

Ray 1’ Ray 1 Ray 2’ Ray 2

960 mm

Receiver

1050 mm

Parabola 2

Parabola 1

240 mm

530 mm

Fig. 2. Schematics of single unit of new CPC system.

allel to axis of one of the parabola and hence are reflected 10% extra mirror area over aperture. This is much less
to the focus of respective parabola on the receiver pipe (in compared to conventional CPC designs. The acceptance
this case, top of receiver pipe: Fig. 2). At this time, the angle of the CPC was 6° for which tilt adjustment is
reflected rays from other arm of CPC are focused on the required every day. Decreasing the acceptance angle
bottom of the receiver pipe (Ray 2). It is to be noted that, resulted in an increase in the concentration ratio with
due to the intersection of the reflected rays with receiver increased aperture requiring reduced receiver pipe size.
pipe wall, on their way to the focus, the reflected rays are The receiver pipe size was 1.500 (outer diameter of
actually distributed on the receiver surface. The position 48.3 mm) with concentration ratio (CR) of 6.3 (upper limit,
of the reflected rays with respect to the receiver pipe is as per Eq. (1), CR = 19). The length of each unit was kept
exactly reversed when the incident rays come from other at 1 m so as to conveniently fix the mirror strips. The exact
extreme (Ray 10 and Ray 20 ). length of the CPC fame was 1050 mm to allow fixing from
Further, when the rays are shifting from one extreme both ends giving clear length of 1 m. To reduce the end
position (Ray1 and Ray2) to the other extreme position losses, it is necessary to connect as many units in series
(Ray10 and Ray20 ), they are not getting focused on either as possible. The CPC frames were fabricated from MS with
of the focal points. The dimensions of the assembly how- semicircular ring at the bottom such that absorber remains
ever are such that, under this condition, the rays reflected fixed and the CPC unit can rotate around it. In order to
from the end points of the two halves of the compound reduce the cost by minimizing the material requirement,
parabolic collector strike the receiver pipe at some interme- the CPC units were supported from the receiver pipe itself
diate points. From the ‘Edge Ray Principle’ it follows with intermediate support for the pipe. With the require-
therefore, that all the rays reflected from the intermediate ment of the perfectly straight receiver pipe, maximum of
points of the reflector also strike the receiver, during the 3 units of above design could be supported from the pipe.
transition from one extreme of the acceptance angle to Thus, the final configuration was three CPC units attached
the other. in series to form a single larger unit with supports for pipe
Thus, a CPC with acceptance angle of 6°, requiring tilt after every 3 m length. 30 units were arranged in two rows
adjustments once a day for a daily operation of 6 h, was as shown in photograph (Fig. 3). The total land area used
designed aiming at steam temperature up to 150 °C. was approximately 60 m2 (20 m  3 m). This arrangement
ensured maximum effective utilization of the piping (mini-
2. Experimental set up and procedure mum pipe length requirement) and reduced the auxiliary
length of pipe that requires to be insulated. The piping at
As explained above, the CPC system was designed in two ends, around the tank and pump was insulated by glass
such a way that the focus was at the level of the aperture wool with aluminum cladding. All the CPC units were fixed
as shown schematically in Fig. 2. It had an aperture area on the receiver pipe with 3 layers of asbestos cloth (total
of 0.96 m2 per unit and system parameters are listed in thickness  20 mm) between them to prevent the conduc-
Table 1. The curve length was 53 cm (on one side) requiring tive heat loss. Each CPC unit was fabricated using angles
 Author's personal copy

A.S. Gudekar et al. / Solar Energy 90 (2013) 43–50 47

Table 1
CPC-design parameters.
Parameter Value
Mounting E–W
Acceptance angle 6o
Aperture area 28.8 m2 (30 units; 0.96 m2 per unit)
Concentration 6.3
ratio
Absorber 1.500 pipe with matt black paint with 2 glass tube
covers
Reflecting Surface Glass mirror strips (15 mm  1000 mm  2 mm
thickness)
Curve length 1.06 m
Mirror area 32 m2

Fig. 3. Photograph of installed CPC system.

to form the main frame to give mechanical strength. The quantity of fresh cold water was taken for each measure-
CPC curve was formed by metal strips of appropriate cross ment. Thermal efficiency of the system was calculated as
section for rigidity. The curved reflector surface was per Eqs. (2) and (3).
obtained by using mirror strips of 15 mm width and For sensible heat gain period (water temperature rise):
1050 mm length. The strips were first fixed on the metal mw C p dT
frame using double sided adhesive tape and then from g¼  100 ð2Þ
SAt
the top with a dummy CPC curve strip. The depth of each
unit was 240 mm with the receiver pipe located at aperture and for latent heat gain period (steam generation):
level (Fig. 2). Two glass tubes (60 mm and 70 mm OD and mk
of nearly 2 mm thickness, separated by a Teflon ring of g¼  100 ð3Þ
0:86SA
thickness 3 mm) were fixed on the receiver pipe to prevent
the convective losses which were found to be substantial where ‘mw’ is the amount of water heated through temper-
after few initial trials with bare receiver pipe. Water tem- ature rise of ‘dT’ in time ‘t’. ‘S’ is the solar beam radiation
perature measurements were done at inlet, outlet of system falling on aperture area ‘A’ and ‘m’ is the rate of steam
and in the tank. Solar intensity was measured with pyra- generation.
nometer supplied by M/s Weathertec, Pune, India (cali-
brated by IMD). Following is the typical procedure 3. Results and discussion
followed during the experiments.
The centrifugal pump was started and the water circu- Following is the discussion of the results obtained with
lated through the receiver pipe of the CPC system. After the new CPC system and the analysis for estimation of heat
noting initial temperatures, all the units were tracked man- losses and the possible measures to improve the system
ually and fixed in their position. The temperatures and the performance.
solar intensity were recorded at regular time intervals dur-
ing the experiment. After the steam generation was started, 3.1. Steam generation efficiency
the generated hot water–steam mixture was fed back to the
tank, which also acted as the separator for the steam. The As noted before, initial trials were carried out without
steam was taken out from the connection provided at the the glass tubes on the receiver pipes. During these trials,
top and its flow rate was measured by condensing directly it was found that substantial time was taken for initial rise
in cold water. To ensure complete condensation, sufficient of temperature to boiling point and also, the steam gener-
 Author's personal copy

48 A.S. Gudekar et al. / Solar Energy 90 (2013) 43–50

ation rate was much lower. For instance, on a clear day 40

(January 24) trial was started at 10 a.m. and steam genera-
35
tion was found to be started at 12:30 p.m. requiring 2 h and
30 min for raising the water temperature to its boiling 30
point. Also, the rate of steam obtained was very low at

Thermal Eff, %
25
3–3.5 kg/h. This indicated high convective heat losses from
the receiver pipes and the need for preventing/reducing this 20
convective heat loss. Hence the set up was dismantled and
15
the receiver pipe was covered with two borosilicate glass
tubes (having 60 and 70 mm OD, 2 mm thickness and 10
3 mm clearance). The results for steam generation on a typ- No Glass Tube
5
ical day (April 12) are shown in Fig. 4. It shows the varia- With Glass Tube
tion of the rate of steam generation with time along with 0
solar intensity. The initial time of heating of water to boil- 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00
Time of day, Hr:min
ing point was found to reduce by more than 25% to 1 h
50 min. The steam generation rate was found to steeply Fig. 5. Comparison of thermal efficiencies for CPC system. ‘no glass tube’
increase with an increase in the irradiation intensity and and ‘with glass tube’ on receiver pipe.
then falls after the maxima obtained around noon-time.
The steam generation started after 1 h 50 min of irradi-
ation with an initial rate of 7 kg/h. The maximum steam tem, resulting into lower heat absorption rate by the recei-
flow rate obtained was around 10 kg/h. The efficiency for ver pipe due to the lower heat transfer coefficient of the
the initial time up to steam generation was calculated con- vapor film. The overall increase in thermal efficiency due
sidering the sensible heat gain by the water and the set up to covering of receiver pipe with glass tubes was found
(thermal inertia). After the steam generation, the efficiency to be 7.4% in absolute terms (46.8% relative rise). Thus,
calculation considered the rate of steam generation and the the use of double glass tubes as cover for receiver pipe
heat of vaporization (latent heat = 540 kCal/kg at 100 °C). resulted in reduction in convective losses and improved
It can be observed from Fig. 4 that the solar intensity was the system performance by slightly less than 50%. There-
maximum at 12:30 pm. The average intensity over the per- fore, the system was further studied for the quantification
iod of the experiment was found to be 700 W/m2. of thermal losses occurring at various stages of energy
The temperature of outlet water was always slightly transfer and the possibility of improvements was exam-
higher than that at the inlet. The time average thermal effi- ined. It was found that apart from material properties such
ciency (over a period of 6 h) of the system was found to be as-mirror reflectivity, optical transmissivity of the glass-
25%. The average steam generation rate was 8 kg/h. Fig. 5 tube and receiver coating absorptivity, the factors contrib-
shows the comparison of thermal efficiencies for the new uting to the reduction in the heat collected are: (a) use of
CPC system without and with glass tube cover on receiver mirror strips (instead of continuous curved mirror), (b)
pipe (Case 1 and Case 2). We can see that the thermal effi- radiative and convective losses from receiver and other
ciency is always higher for Case 2 indicating the reduction hot surfaces and (c) loss of concentrated radiation at recei-
in heat losses except at a time of 11:30 a.m. This observa- ver due to end supports (100 mm length is used for support
tion is due to the fact that, in Case 2, the steam generation per m length of receiver pipe) which are quantified in the
was observed at this time which may have flooded the sys- next session.

3.2. Heat loss analysis and measures for improvement

900 12
The loss of energy at various stages is estimated as
800
10 shown below and the possible measures to reduce it are
700 also listed. Table 2 shows the estimated loss values (and
Solar Intensity, W/m2

Steam Rate, kg/hr

600 8 available energy) at various steps for two cases-the present

500 and that possible after improvements.
6
400
3.2.1. Radiation loss at mirror surface
300 4
Use of mirror strips to form the required curved shape
200 gives rise to losses at each joint between adjacent strips.
Solar Intensity
2
100 Steam Rate The entire area of the CPC is not covered by the mirrors
0 0 and space of 1 strip (15 mm) is lost in the length of
9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 530 mm; amounting to 2.8% of surface loss. In ideal case,
Time of the Day, Hr: min
the curved mirrors could be used and this loss can be elim-
Fig. 4. Variation of solar intensity and steam generation rate. inated completely.
 Author's personal copy

A.S. Gudekar et al. / Solar Energy 90 (2013) 43–50 49

Table 2
Losses and improvements in CPC system.
No Step/Item Present system After improvements
2
Description Loss (%) Available (W/m ) Loss (%) Available (W/m2)
0 Available solar radiation – 700.0 – 700.0
1 Mirror strip 02.8 680.4 0.0 700.0
2 Reflectivity 15.0 578.3 5.0 665.0
3 Manufacturing errors 05.0 549.4 1.0 658.4
4 Glass tube Transmitivity 15.4 465.0 5.0 625.4
5 Receiver absorption 20.0 372.0 8.0 575.4
6 Radiation loss 15.4 314.7 2.8 559.4
7 Convective loss 30.5 218.6 1.0 553.8
8 Other losses 20.0 174.9 5.0 526.1
Available energy to water 25.0a 174.9 71.2a 498.4
a
Energy available to water as percentage of incident solar radiation (700 W/m2).

3.2.2. Mirror reflectivity Surface area = 0.1517 m2 (per m2 of aperture area).

The glass mirrors used (MODI GUARD) have the Pipe temperature = 120 °C.
reflectivity of 85% (measured using pyranometer). Thus, Inner glass tube temperature = 80 °C.
15% of the radiation is lost at this stage. The mirrors avail-
able for commercial solar reflectors (e.g. developed for The steady state losses can be computed from the radi-
PTC technology) have reflectivity in the range of 95% ation losses from the metal pipe to inner glass tube which
bringing down these losses to only 5%. in turn is losing to outer glass tube and then to the
surrounding.
3.2.3. Manufacturing errors
Qrad ¼ reAðT 4  T 4g Þ
The present CPC units were fabricated by a local skilled
fabricator but without any automation. The CPC curve With radiation constant r = 5.67  108 W/m2/K4, these
was prepared manually by hammering/bending. After losses comes to be 57.3 W for 1 m2 aperture area. This
matching the fabricated curve with the plotted one from amount is 15.4% of the incident solar energy at this stage
x–y data, this fabrication error was found to be around or nearly 8.2% of the incident radiant energy of 700 W/m2.
5%. It can be assumed that with automation these losses With selective commercial solar coatings like Solkote
can be reduced to less than 1%. (Make: Solec, US), these losses can be reduced substan-
tially (Absorptivity = 0.9137 and emissivity = 0.2244);
3.2.4. Transmission through glass tubes hence, radiative = losses come down to 16 W/m2 or to only
The borosilicate glass tube has a transmitivity of 92% 2.8% of the incident energy available at this stage.
for the sunlight. Since two glass tubes were used in series,
the losses are 15.36%. With special type of glass for solar 3.2.7. Convective losses
applications (low iron content) having transmitivity of The convective losses occur from all of the hot surfaces
95%, these losses can also come down to 10% and with like piping, tank, pump and the glass tubes. These were
an evacuated single glass tube receivers, transmission losses estimated roughly as follows.
can come down to less than 5%.
(a) Insulated surfaces
3.2.5. Losses at receiver Qc1 ¼ h  A  dT
These come from the properties of the coating material
and its combination with the receiver pipe material. The The area is estimated from the length of the piping, pump,
absorptivity of the paint used from the local market was tank, etc. and comes to be nearly 10 m2; the heat transfer
only 80% thus loosing 20% of the incident energy in the coefficient, ‘h’ is typically 10 W/m2/K (McAdams, 1954),
form of reflection. Selective coatings have absorptivity in and the temperature difference (insulation at 45 °C and
excess of 91%, which can reduce these losses to about 8%. air) is of the order of 15 °C.
Thus the total loss comes out to be 1500 W for entire
3.2.6. Radiation from receiver CPC system, or 52.1 W/m2 for system aperture area of
For the experimental stet up used in this study (receive 28.8 m2.
pipe without selective coating), the radiation losses can
be estimated as follows: (b) Glass tube surface

Pipe OD = 48.3 mm. The outer glass tubes were of 70 mm OD and 1 m

Length = 1 m. length; each for 1 m2 aperture area.
 Author's personal copy

50 A.S. Gudekar et al. / Solar Energy 90 (2013) 43–50

The glass surface temperature was 50 °C and with same been carried out and it was found that more refinements
value for ‘h’, in the design and scale up would further enhance the sys-
The losses are tem performance and will enable steam generation at tem-
Qc2 ¼ 44 W peratures at which significant fraction of process heat can
be used.
Thus, total convective losses are around 96.1 W/m2
accounting for 30.5% of the incident energy available at Acknowledgement
this stage or 13.7% of the initial incident energy. Ideally,
with evacuated tubes and proper insulation, these losses One of the authors, Ajitkumar S. Gudekar, is grateful to
can be reduced substantially (of the order of 1%). the Department of Atomic Energy, Government of India,
for providing the research fellowship during the course of
3.2.8. Other losses work.
For supporting of the CPC units on the absorber pipes,
clamps of 100 mm length were provided. Two adjacent References
units shared one clamp totaling 16 clamps for one row of
15 units. Thus, though direct end losses were avoided com- Azhari, A., Khonkar, H., 1996. A thermal comparison performance of
bining three CPC units to form a group, concentrated radi- CPC with modified (duel-cavity) and non-modified absorber. Renew-
ation falling on these clamps was lost. This loss can be able Energy 9, 584–588.
Buttinger, F., Beikircher, T., Proll, M., Scholkopf, W., 2010. Development
estimated to be 10.6% of the concentrated radiation. of a new flat stationary evacuated CPC Collector for process heat
Another loss, not considered previously is the conductive applications. Solar Energy 84, 1166–1174.
loss through joints. These along with clamp losses are Earth Trends, 2003. Energy and Resources – India. <http://earthtr-
roughly estimated to be of the order of 20%. ends.wri.org/pdf_library/country_profiles/ene_cou_356.pdf>
Thus, it can be seen from Table 2 that, the final available (accessed on 30.12.11).
Eck, M., Zarza, E., 2006. Saturated steam process with direct steam
energy to water is 175 W/m2 which is 25% of incident generating parabolic troughs. Solar Energy 80, 1424–1433.
700 W/m2. Also, as seen from table, the system perfor- IEA, 2007. World Energy Outlook 2007 – China and India Insights.
mance can be improved to 71% with energy made available International Energy Agency, Paris, France.
through these suggested improvements to 498 W/m2. Jiang, L., Lan, C., Kim, Y., Ma, Y., Winston, R., 2012. An evacuated PV/
thermal hybrid collector with the tube/XCPC design. In: Proceedings
Thus, with better materials (solar-special mirror and
of WREF 2012 (World Renewable Energy Forum), Denver, Colorado.
glass tubes, selective receiver coating) the losses related to <https://ases.conference-services.net/resources/252/2859/pdf/
basic properties can be minimized. Further, the scale up SOLAR2012_0072_full%20paper.pdf>.
of the system with appropriate design modifications McAdams, W.H., 1954. Heat Transmission, third ed. McGraw-Hill Book
(increasing the aperture diameter and unit length) will also Co., NY.
help in lowering the various heat losses discussed above. Rabl, A., 1976. Optical and thermal properties of compound parabolic
concentrators. Solar Energy 18, 497–511.
Thus with proper improvements, the proposed new design Rabl, A., 1980. Concentrating collectors. In: Dickinson, W.C., Cherem-
has the capability of harnessing solar energy with relatively isinoff, P.N. (Eds.), Solar Energy Technology Handbook Part A –
high thermal efficiency. Engineering Fundamentals. Marcel Dekker Inc., New York, pp. 257–
343.
4. Conclusions Sargent and Lundy LLC Consulting Group, 2003. Assessment of
Parabolic Trough and Power Tower Solar Technology Cost and
Performance Forecasts. A Report Prepared for NREL (National
From the study undertaken, it can be concluded that (i) Renewable Energy Laboratory, US Department of Energy Labora-
modified CPC design is easy to fabricate and worked rea- tory) Golden, Colorado 80401-3393. <http://www.osti.gov/bridge>.
sonably well for steam generation at atmospheric pressure Thomas, A., 1995. Solar steam generating systems using parabolic trough
(ii) the limitations of conventional CPC systems are quan- concentrators. Energy Conversion and Management 37, 215–245.
Tripanagnostopoulos, Y., Yianoulis, P., Papaefthimious, S., Souliotis, M.,
titatively identified and were partially overcome in the new Nousia, T., 1999. Cost effective asymmetric CPC solar collectors.
design which helped in reduction in overall system cost (iii) Renewable Energy 16, 628–631.
newly designed CPC models can substantially reduce the Winston, R., Hinterberger, H., 1975. Principles of cylindrical concentra-
mirror area requirements per unit aperture area compared tors for solar energy. Solar Energy 17, 255–258.
to conventional CPC systems (iv) heat loss analysis has