Труды Всемирного конгресса Международного общества солнечной энергии

Подождите немного. Документ загружается.

Proceedings of ISES Solar World Congress 2007: Solar Energy and Human Settlement

652

sheet. There are also losses to the edges. Here they are

counted as a part of the top respectively the backside losses.

There is also a useful heat transfer, q

, which goes from the

absorber plate to the heat transfer media. The mean path to

go is from the middle of one flange of a fin (i.e. ¼ of the

tube spacing) + the boundary between absorber and tube +

the mean cross section way in the tube + the convection

layer from tube to the heat transfer media.

The heat transfer via convection and conduction from

absorber to glass (qcpg ) is the one which is affected by

changing gas. The q

cpg

can be expressed as in (2).

***( )qNukSTT=− (2 )

As can be seen from formula 2, a heat flow is dependent of

the conductivity k, shape factor S and temperature

difference in the same way as a pure conduction driven heat

transfer. The only difference between conduction and

convection is the Nusselt number, Nu. The conductivity and

the Nusselt number will be affected when changing gas.

Most of the formulas describing the different heat transfers

in the different heat transfer medias are standard formulas.

Holman has done an almost complete compilation of

suitable heat transfer formulas for setting up a model like

the model used for calculations in this article (Holman

2002). The relation between Nusselt and Rayleigh numbers

in inclined air layers is thoroughly investigated by [1] and

[2]. The depenancy of the Rayleigh number and the slope

on the Nu number is shown in (3)

1708(sin 1.8 1708 cos

11.441 *1 1

cos cos 5830

Ra Ra

ββ

⎡⎤

⎡⎤⎡⎤

⎛⎞

=+ − − + −

⎢⎥

⎜⎟

⎢⎥⎢⎥

⎝⎠

⎣⎦⎣⎦

⎢⎥

⎣⎦

( 3 )

Note: The + exponent means that only positive values of

the terms in the square brackets are to be used.

( 3 ) is valid in a range of the slope between 15 to 60°.

2.2 Calculations

All calculations were made in Matlab, a suitable tool for

practising numerical methods. At given conditions the

unknown factors were possible finding by iterations.

When simulating unlinearities on the absorber, the

calculation between absorber and glass was done several

times with different l

. The resulting q

was a mean value

of these calculations. That approach was also used when

simulating a variation of temperature on the absorber but

then the T

varied. The T

varies depending on distance out

on the fin and the warming of the heat transfer media and

the variation is not linear. A simple way to compensate for

this was to weight the hotter parts a little bit different to the

colder ones which were done in this model. The approach

was a gamma correction used for interpolate the

temperatures between the ends as in (4).

yx= ( 4 )

x=0 corresponds to the coldest point and x=1 to the hottest.

The temperature is calculated as in (5)

T(x) T(0)*y T(1)*(1-y)=+ ( 5 )

The y-values between x=0 and x=1 can be varied so a curve

representing the temperatures in between can get a more

representative temperature distribution. In reality the mid

points are warmer than what a linear interpolation

calculates. In this calculations a γ=0.5 was used.

Material data, i.e. gas properties was lineared in Matlab.

Input data comes from measurement series[3-6]. Physical

dimensions and other typical solar collector properties are

taken from a model made by Schüco, called Premium. It

was state of the art flat plate solar collectors among the

ones sold by companies until 2006 in Sweden. The reason

why this model was chosen is that if the gas filled

collectors cannot achieve better performance than the

already existing good ones there is no good reason to

develop any with this new technology unless it is possible

to reduce the cost of manufacturing. Another reason was

that the physical sizes were known and that it was possible

to verify the model with a known collector.

A shape factor also had to be constructed, which accounted

for how much more the edge affected the heat loss when l

varied in the calculations. The Schüco Premium has no side

3 SOLAR COLLECTOR TECHNOLOGIES AND SYSTEMS

653

wall insulation and the distance between absorber and side

wall was estimated to be 10 mm. The shape factor was

calculated with a FEM-model assembled in Matlab. It all

ended up with a shape factor which can be seen in (6)

( 1 s2*l s3*l )

=++

( 6 )

The factors of a, b and c in the polynomial in (6) were

found to be: s1 = 0.995 +/- 0.005, s2=1.64 +/- 0.1 and

s3=21 +/- 2 in the range of lpg =0.001 .. 0.1 m. This makes

a maximum error of 2.5 % from the polynomial, but the

heat transfer around the edge is not uniform, so the error is

probably bigger.

2.3 Results

The model was validated with test results from the

reference collector as well as a model set up in

CodePro.exe [7], [8]. The performance were expressed as

the c-factors for the efficiency formula which can be seen

in ( 7 ).

()()

01 2

Tw Ta Tw Ta

cc c

−−

=− −

( 7 )

The c1 and c2 factors are the most interesting factors here,

since they are temperature dependant. The different c1 and

c2 values can be seen in Table 1.

TABLE 1: VALIDATION OF MODEL FOR REFERENCE

COLLECTOR

Value Test CodePro Model

c1 3.518 3.845 3.97

c2 0.012 0.00352 0.0042

ΔU from test, T

=10°C 7% 10%

ΔU from test, T

=50°C -3% 2%

ΔU from test,

=100°C

-12% -7%

The model is almost correct at Tw=50°C and

underestimates the losses on high temperatures and

overestimates them on low temperatures.

In the following the influence of gas filling on the overall

heat loss is studied. It is assumed that the collector has the

same physical dimensions and properties as a Schüco

Premium, except l

, that the l

varies 2 mm, mean T

50°C, highflow system with 0.05m

/(h*m

), slope = 45°, a

variation of l

of 2mm. The overall loss coefficient’s

dependency of the l

for different gas fillings can be seen

in Fig. 1.

Fig. 1: Losses vs distance l

at Tw=50 °C and β = 45 °.

The result is that the overall loss coefficient can be lowered

with about 20% changing to the best alternative, here

Xenon (Xe). All studied gases have the same appearance.

There is a minimum on every curve at a distance of 25..50

mm, depending of gas ( and of course the edge insulation),

which here will be called normal distance minimum. It is

also a minimum on a very shorter distance, somewhere

between 3 and 9 mm, depending of gas, which here will be

called short distance minimum. Between the two

minimums on each curve there is a hump near the short

distance minimum, called local maximum.

Traditionally the normal distance minimum is used, or at

least something near the minimum. The usage of the

normal distance minimum in an enclosed collector will

cause some problems. It will be a large amount of gas and

the variations of temperature, which is normal for a solar

collector in use, will cause either volume variations or

pressure variations or a mixture of them in the enclosed gas,

due to the common gas law. Therefore it is interesting

analyze the short distance minimum.

Proceedings of ISES Solar World Congress 2007: Solar Energy and Human Settlement

654

The short distance minimum (from 3 – 8 mm depending on

gas) is an interesting point of the curves in Fig. 1. It appears

on the longest distance where Nu still is 1 or very near 1.

The short distance minimum has the wanted performance

and a much less volume of gas is needed which causes

smaller stresses in the material compared to the normal

distance minimum. The fly in the ointment is that the

distance is more critical.

A sensibility test was done which can be seen in Table 2.

TABLE 2: VALIDATION OF MODEL FOR REFERENCE

COLLECTOR

Air CO

Ar Kr Xe Ref.

U (W/(m

*K) 4.56 4.62 4.04 3.76 3.56 4.68

Ideal l

(mm) 8.5 5.6 7.8 5.2 3.9 40

δl

+10% -0.2% -0.3% -0.3% -0.3% -0.3% 0.0%

+ 10% -1.1% -0.7% -1.1% -0.6% -0.4% 0.1%

δl

+10%

+ 10%

-1.1% -0.8% -1.2% -0.7% -0.5% 0.1%

+ 5K 1.9% 1.8% 1.9% 1.9% 1.9% 2.0%

Slope -10° -1.3% -1.4% -1.2% -1.1% -1.1% -0.9%

+ 5K

slope -10°

0.7% 0.4% 0.7% 0.8% 0.8% 1.1%

+10% 1.5% 1.5% 1.5% 1.6% 1.6% 1.4%

Wind +10% -0.3% -0.3% -0.3% -0.3% -0.3% -0.3%

The changed parameters are in the left column and the

changes of U from the starting point are in the columns to

the right. The varied parameters are: + 10% change of

unlinearity of the absorber plate (δl

), +10% deviation

from the l

, a combination of δl

and l

where both were

10% larger, +5K change of heat transfer media

temperature, , a change in the slope by -10°, a combination

of Tw and slope (+5K, -10°), +10% change in wind speed,

and a change in plate thickness by 10%.

The most spectacular observation is that a rise in

temperature will lower the U-value. At the same time as the

upward losses grows the losses due to the fin efficiency

decreases. Totally the losses seem not increase as much as

the temperature changes which leads to a lower U-value,

but remember: this special effect is depending on the

thickness of absorber and the irradiation and the model also

underestimated the c2 factor. The thickness of the absorber

will also give us a better U-value when only changing that.

Variations of l

and δl

got an expected result, and it is not

any superposition relationship between them. If L

grows,

the U-value will reach the local maximum as most, and

then it will decrease again.

The slope and wind also have some influences, but not

much more than on the reference.

Fig. 2 shows the result in an ordinary collector efficiency

diagram for a better understanding of the influence of

changing gas in a collector with no expansion. The curve

for the reference collector applies to 40 mm, while all

others curves apply for the short distance minima, i.e. for a

much thinner box. CO

and thin air have about the same

performance as the reference collector. All the inert gas

filled collectors achieve better performance, and the

succession order is Xe, Kr and Ar (with the best first).

Fig. 2: Efficiency curve.

3. DISCUSSION

Models are models and the truth is out there with

experimenting tests. Nevertheless a model can compare

different alternatives and find a succession order and it can

also find out some promising aspects. The sources of errors

are several in the model, as it usually is in convection

theory. Nevertheless, all gases are treated in the same way

and it is the differences between the gases this article wants

to focus on.

3 SOLAR COLLECTOR TECHNOLOGIES AND SYSTEMS

655

As can be seen, short distances can be useful even if you

don’t hit the absolute minimum. Moderate deviations from

the optimal l

only gives moderate more losses of

performance. It is also interesting to see that a “thin”

collector with inert gas still gives lower losses compared to

a common “normal thick” air filled solar collector.

4. NOMENCLATURE

A Area (m

)

c0..c2 factors describing losses in efficiency equation

G Irradiance (W/m

)

k Heat conductivity (W/(m*K))

l Length (m)

Nu Nusselt number (-)

q Heat transfer per area (W/m

)

Ra Rayleigh number(-)

S Conduction shape factor (m)

s Factor in edge shape polynomial

T Temperature (K)

U Heat transfer per area and temp. diff (W/(m

* K)

x Relative position between nodes (0..1)

y Weight factor (0..1)

γ Gamma interpolation correction factor(-)

Δ,δ Difference

η Efficiency (-)

Subscripts

a Ambience

b Backside

c Convection

g Cover Glass

p Absorber plate

z z-axis (height)

r Radiation

w Heat transfer media

5. REFERENCES

(1) K.G.T. Hollands, T.E. Unni, G.D. Raithby, and K. L.:

'Free convective heat transfer across incllined air

layers'," J. Heat Transf.-Trans. ASME"

, 1976, 2, (May),

pp. 189-193

(2) S.M. EISherbiny, G.D. Raithby, and K.G.T. Hollands:

'Heat transfer by natural convection across vertical and

inclined air layers'," J. Heat Transf.-Trans. ASME"

1982, 104, (February), pp. 96-102

(3) D.R. Lide: "CRC Handbook of chemistry and physics:

a ready-reference book of chemical and physical data"

CRC Press, Boca Raton, 1994, 75 edn, 0-8493-0596-9

(4) http://www.airliquide.com/, accessed 2006-03-29 2006

(5) J.P. Holman: "Heat Transfer" McGraw-Hill College,

Boston, USA, 2002, 9 edn, 0-07-240655-0

(6) H. Alvarez: "Energiteknik" Studentlitteratur, Lund,

Sweden, 1990, 91-44-31471-X

(7) H. Quicklund, and G. Johansson, "Rapport Energide-

klaration för solfångare med AR-glas (Report energy

declaration for solar collector with AR-glass)",

SP Sveriges Provnings och Forskningsinstitut AB,

2004-03-18, pp 5

(8) S.A. Klein,"Software Engineering Equation Solver",

2005-08-15, v7.454

A PERFORMANCE AND RELIABILITY STUDY OF A NOVEL ICPC

SOLAR COLLECTOR INSTALLATION

William S. Duff, Jirachote Daosukho

Department of Mechanical Engineering

Colorado State University

Fort Collins CO 80524

duff@engr.colostate.edu

ABSTRACT

A novel integral compound parabolic concentrator

evacuated solar collector (ICPC) array has been in

continuous operation at a demonstration project in

Sacramento California since 1998. This paper reports on

recent progress in an ongoing reliability and degradation

study of the changes in performance of the array over eight

years of operation. The study addresses the impact of both

optical and thermal factors on array performance. Recent

progress in the study includes 1) development of an

animated graphical ray tracing simulation tool to

investigate changes to the incidence angle modifier of the

ICPC and 2) the design of a device to measure the loss of

reflectivity of the internal mirrored surfaces of the ICPC.

The paper presentation will include the design of the

simulation, animations of rays striking at various angles,

the summary of the incidence angle evaluation, the design

of the laser device and the plans for generating the tube

reflectance map.

1. BACKGROUND

1.1 Development of the Novel ICPC

Research on CPC solar collectors has been going on for

more than twenty years. See Garrison [1] and Snail et al [2].

In the early 1990s a new ICPC evacuated collector design

was developed. The new ICPC design allows a relatively

simple manufacturing approach and solves many of the

operational problems of previous ICPC designs. The design

and the fabrication approaches are described in Duff et al [3]

and Winston et al [4].

1.2 ICPC Performance Testing at Sandia

Prior to the start of the 1998 Sacramento demonstration,

individual fourteen tube modules were tested on Sandia

National Laboratory’s two-axis tracking (AZTRAK)

platform. See Winston et al [3]. These tests showed losses

from the new ICPC collector operating at 2E chiller

temperatures on a 35C summer day of only 120w/m

2 of

collector aperture.

1.3 Sacramento Demonstration

A 100 m

336 Novel ICPC evacuated tube solar collector

array has been in continuous operation at a demonstration

project in Sacramento California since 1998. The evacuated

collector tubes are based on a novel ICPC design that was

developed by researchers at the University of Chicago and

Colorado State University in 1993. The evacuated collector

tubes were hand-fabricated from NEG Sun Tube

components by a Chicago area manufacturer of glass

vacuum products.

From 1998 through 2002 demonstration project ICPC solar

3 SOLAR COLLECTOR TECHNOLOGIES AND SYSTEMS

657

collectors supplied heated pressurized 150C water to a

double effect (2E) absorption chiller. The ICPC collector

design operates nearly as efficiently at 2E chiller

temperatures (150C) as do more conventional collectors at

lower temperatures. This new collector made it possible to

produce cooling with a 2E chiller using a collector field

that is about half the size of that required for a single effect

(1E) absorption chiller with the same cooling output. The

nontracking ICPC evacuated solar collector array provided

daily solar collection efficiencies (based on the total solar

energy falling on the collector) approaching fifty percent at

the 140C to 160C collector operating temperature range.

The ICPC array has recently been operating at the lower

temperatures to drive a single effect absorption chiller. The

ICPC array has provided daily solar collection efficiencies

approaching fifty-five percent at the 80°C to 100°C collector

operating temperature range.

1.3.1 Array Layout and Absorber Orientation the New

ICPC evacuated tubes were fabricated with two absorber

orientations, one with a vertical absorber fin and one with a

horizontal fin. A cross-section of the collector tube

illustrating the two orientations is shown in Fig. 1.

Fig. 1: Novel ICPC Design Showing Vertical and

Horizontal Fin Orientations.

The collector array is made up of three banks. The flow

pattern through the 112 evacuated tubes in each bank is

parallel and the three banks are plumbed in parallel. The

north bank consists of all horizontal fin tubes, the middle

bank consists of all vertical fin evacuated tubes and the

south bank includes an even mixture of the two types.

1.4 Evacuated Tube Reliability

After a year of operation several distinct patterns in the

development of cracks in the evacuated tubes emerged.

One of these involved the production sequence or,

equivalently, the fin orientation and the other, the end of

the tube where the crack occurred.

Vertical and horizontal tube absorber orientations were

produced in the first and second halves of the ICPC tube

production run respectively. One year after installation 1.2

percent of the vertical fin orientation tubes and 9.8 percent

of the horizontal tubes had developed cracks. This strongly

suggests that there were distinct differences in the longevity

of the vertically finned tubes versus that of the horizontally

finned tubes (or, equivalently, of the first half of the

production run versus the second half). Statistically, if one

assumes that the entire production run is characterized by

the overall fraction of cracked tubes of 0.05865 then the

likelihood that the first half of the production run came

from such a process is less than 0.3 percent. Moreover,

after six years of operation only 3.6 percent of the

vertically finned tubes had developed cracks, whereas the

horizontally finned tubes continued to develop cracks at a

much higher rate.

Since the evacuated tubes were essentially hand built, this

3.6 percent failure rate is about what one would expect. The

end caps of each end of the evacuated tubes were identical,

each consisting of a dish shaped piece of glass and a metal

cap bonded to the glass. At the top end a metal tubulation

was brazed to the metal cap to provide flow of heated fluid.

At the bottom end a metal tubulation was brazed to the

metal cap to provide a means to evacuate the tube. Thus,

only the top end was subject to both thermal stress (the

155C fluid) and mechanical stress (partial support of the fin

and heat transport tube). One might expect the failure rates

due to cracking to be higher at the top end of the tube than

at the bottom. In fact the opposite occurred. Out of 19

cracked tubes after one year, 7 were cracked at their tops

and 12 at their bottoms. Statistically, if one assumes that

the true proportion of cracks at the top to be 60 percent,

then there is only a 0.1 percent chance that one would

observe seven or fewer cracks out of 19 at the top end.

1.5 Preliminary Degradation Study

A preliminary study of possible degradation modes and

magnitudes conducted in 2006 was reported in Duff [5, 6].

This study investigated various collector optical and loss

Proceedings of ISES Solar World Congress 2007: Solar Energy and Human Settlement

658

mechanisms to model collector performance and per-

formance degradation over a range of operating temperatures.

Performance changes due to vacuum loss and other failure

modes in a number of the individual evacuated collector

tubes are explicitly treated in the modeling process.

The modeling links the energy production of the array to

the condition of the array over time to provide a time

varying estimate of performance change due to evacuated

tube failures and the degradation of material properties.

Performance over time is then compared to the actual

performance of the collector array. The modelled

performance provided a statistically based accounting for

the observed change in array performance.

2. OPTICAL PERFORMANCE MODELING AND

EXPERIMENTATION

2.1 Graphical Ray Tracing

In the simulation discrete rays from the sun impacting the

collector tube have been simulated across an incident angle

range of 15 degrees to 165 degrees. A ray striking the

collector at a given angle and in given location is monitored

as to how it reacts to various surfaces of the collector. The

number of rays hitting the absorber directly or indirectly

via reflections has been recorded. On average, each

reflection reduces the number of incident rays in proportion

to the reflectance of the surface.

An animated graphical ray tracing simulation tool has been

designed to investigate changes to the optical performance

of the ICPC. Factors incorporated are the transmittance of

the glass tube, the reflectivity of the reflective surface, the

gap between reflective surface and the absorber fin, and the

absorptance of the absorber fin. The sun ray is simulated as

discrete uniform rays crossing from the incident angle of 15

degrees to 165 degrees. The number of rays hitting the

absorber plate, the heat transfer tube, and reflector are

recorded. The collector efficiency graph shows an amount

of energy absorbed during a typical daytime period.

Figure 2 shows the optical efficiency based on a surface

reflectance of 0.94, a gap between reflective surface and

the absorber fin of 4 mm and an absorptance of 1. For

example, refer to Figure 3 which shows a ray striking the

collector at an incident angle of 44 degrees. A ray striking

the collector is followed to determine how it reacts to

surfaces of the collector. The first gap loss (green rays) is

also detected at an incident angle of 44 degrees. This is

depicted in the optical efficiency responses seen in Figure 2.

Rays incident from 65 degree to 115 degrees are examined.

Fig. 2: Optical Efficiency with 0.94 Reflectivity at Incident

Angles of 15 to 165.

Fig. 3: Collection of Rays Striking the Collector at 44

°.

To illustrate that gap losses cause of the optical efficiency

drop at 44 degree incident angle, another simulation is run

in which there is no gap and therefore no loss. Figure 4

3 SOLAR COLLECTOR TECHNOLOGIES AND SYSTEMS

659

shows the collector efficiency with no gap loss. The graph

depicts rounded distribution with no jump in efficiency at

any incident angle.

To understand the nature of gap loss, the gap loss is plotted

over the incident angles, as in Figure 5. The gap losses are

separated into roughly two levels. Figure 6 illustrates why

the optical efficiency changes to the flattened nature of the

optical response around 90 degrees.

Fig. 4: Optical Efficiency from Incident Angle of 15 to 165

With No Gap Losses.

Fig. 5: Percentage of Rays Lost over the Incident Angle

Range versus No Gap Losses.

In figure 7 the reflectivity is changed to 0.7 to achieve an

optical efficiency curve that has its shape altered to a more

dished appearance for rays close to the 90 degrees incident

angle. In figure 6 most of the rays are reflected to the

absorber and just a few hit the heat transport tube directly.

Fig. 6: Collection of Rays Striking at 90°.

Fig. 7: Optical Efficiency with 0.7 Reflectivity at Incident

Angles of 15 to 165.

2.2 Reflectivity Measurement

The device shown in Figures 8 and 9 consists of a laser and

detector mounted on a support structure that can be

positioned around the various tubes of the ICPC array to

measure transmittance and reflectance losses. Figure 13

shows the addition of a tube on the sensor end of the device

that will reduce the extraneous radiation that reaches the

Proceedings of ISES Solar World Congress 2007: Solar Energy and Human Settlement

660

sensor. Using this device, a map of reflector performance

for the ICPC array will be generated in the summer of

2007.

Fig. 8: Photograph of the Laser and Sensor Assembly.

Fig. 9: Diagram of the Laser and Sensor Assembly.

3. CONCLUSIONS

A detailed simulation for characterizing the optical performance

of ICPC evacuated tubes has been described and its results

illustrated. An instrument for examining the transmittance

and reflectance of the ICPC has been developed. Detailed

thermal modeling of the evacuated collector tubes,

manifolding and collector array will be the next analysis

step.

4. SELECTED REFERENCES

(1) Garrison, J. D., Optimization of Fixed Solar Thermal

Collectors, Solar Energy, v23, 1979

(2) Snail, J. J., O’Gallagher and R. Winston, A Stationary

Evacuated Collector with Integrated Concentrator,

Solar Energy, v33, 1983

(3) Winston, R, O’Gallagher, J., Mahoney, A. R., Dudley,

E. and Hoffman, R., “Initial Performance

Measurements from a Low Concentration Version of

an Integrated Compound Parabolic Concentrator

(ICPC)”, Proceedings of the 1999 ASES Annual

Conference, Albuquerque NM, June, 1998

(4) Duff, William S., R. Winston, J. O’Gallagher, T.

Henkel, and J. Bergquam, “Update on the New High

Performance ICPC Sacramento Cooling

Demonstration”, American Solar Energy Society

Conference, Madison Wisconsin, June 2000

(5) Duff, William S. and Jirachote Daosukho, “Reliability

and Degradation Study of the Sacramento

Demonstration Novel ICPC Solar Collectors”,

American Solar Energy Society 2006 Congress, Denver,

Colorado, July 2006

(6) Duff, William, Jirachote Daosukho, Klaus Vanoli,

Roland Winston, Joseph O’Gallagher, Tom Henkel and

Jim Bergquam, “Comparisons of the Performance of

Three Different Types of Evacuated Tubular Solar

Collectors”, American Solar Energy Society 2006

Congress, Denver, Colorado, July 2006

HIGH TEMPERATURE CPC COLLECTORS WITH CHINESE

VACUUM TUBE RECEIVERS

R.Winston, G. Diaz, A. Ritchel, A. Tovar, J. Cisneros

University of California,

Merced, CA 95344

R. Gee, M. J. Hale, S. Horne

Solfocus, Inc.

Mountain View, CA 94043

X. Jiang

Beijing Eurocon Solar energy Tech.Co.,Ltd.

NO.30 Xueyuan road,Haidian District,Beijing China

R. Radhakrishnan, J. Zhang

United Technology Research Center

East Hartford, Connecticut 06108

ABSTRACT

At the new University of California in Merced, CA and

SolFocus at Mountain View, CA we are developing high

temperature CPC collectors under the auspices of the

California Energy Commission. By combining external

CPC reflectors with Chinese vacuum tube receivers, high

temperatures are achieved. These receivers are specially

designed and made for efficient operation at 200 degrees

Celsius. The emphasis of the system design is low cost,

high reliability and excellent manufacturability. It is

expected that these collectors will become a useful heat

source for applications ranging from process heat, to

absorption air conditioning, to organic Rankine cycle

engines for electric power.

1. INTRODUCTION

The authors developed a prototype of an External Compound

Parabolic Concentrator (XCPC)

. Theoretical models indicate

that the design may achieve a system efficiency of 50 percent

at a temperature of 200ºC. The collector may have production

cost as low as $150 per square meter. It can be used for

process heat applications, solar absorption cooling, water

desalination and purification, and for power generation using

an Organic Rankine Cycle engine.

2. BODY OF PAPER

The design consists of an assembly of stationary evacuated

tube absorbers with external nonimaging reflectors. The

evacuated glass tubes with a glass-to-metal seal are used to

insulate the metal absorber. The metal absorber is a thin

cylindrical aluminum fin with a selective coating that enables

high solar absorptance and reduces radiative heat loss. The

absorber tubes are placed so that external reflectors

concentrate sunlight onto the evacuated absorber fins.

The XCPC design consists of the following elements:

 Evacuated glass tube with metal absorber

 Manifold

 Reflector

 Frame

2.1 Evacuated Glass Tube

The evacuated glass tube (including the metal absorber) is

made by Beijing Eurocon Solar Energy Tech. Company in

Beijing, China. The tube has the following dimensions:

Diameter of outer tube wall: 65 mm

Diameter of inner tube wall: 61.8 mm

Length: 1,800 mm

Труды Всемирного конгресса Международного общества солнечной энергии - 2007. Том 2

Подождите немного. Документ загружается.