TECHNICAL PAPER # 30
UNDERSTANDING SOLAR
CONCENTRATORS
By
George M. Kaplan
Technical Reviewers
Dr. Thomas E. Bowman
Dr. Maurice Raiford
Jesse Ribot
Illustrated By
Rick Jali
Published By
VITA
1600 Wilson Boulevard, Suite 500
Arlington, Virginia 22209 USA
Tel: 703/276-1800 * Fax:
703/243-1865
Internet: pr-info@vita.org
Understanding Solar Concentrators
ISBN: 0-86619-239-5
[C]
1985, Volunteers in Technical Assistance
PREFACE
This paper is one of a series published by Volunteers in
Technical
Assistance to provide an introduction to specific
state-of-the-art
technologies of interest to people in developing countries.
The papers are intended to be used as guidelines to help
people choose technologies that are suitable to their
situations.
They are not intended to provide construction or
implementation
details. People are
urged to contact VITA or a similar organization
for further information and technical assistance if they
find that a particular technology seems to meet their needs.
The papers in the series were written, reviewed, and
illustrated
almost entirely by VITA Volunteer technical experts on a
purely
voluntary basis.
Some 500 volunteers were involved in the production
of the first 100 titles issued, contributing approximately
5,000 hours of their time.
VITA staff included Maria Giannuzzi
as editor, Suzanne Brooks handling typesetting and layout,
and
Margaret Crouch as project manager.
The author of this paper, VITA Volunteer George M. Kaplan,
is the
president of KAPL Associates, a consulting firm specializing
in
program and project management, research and development,
planning,
evaluation, energy, and environment.
The reviewers are also
VITA volunteers. Dr.
Thomas E. Bowman is Professor and Head
of
the Mechanical Engineering Department at the Florida
Institute of
Technology in Melbourne, Florida. Dr. Maurice Raiford is a
solar
energy consultant in Greensboro, North Carolina.
Jesse Ribot is
an energy analyst and consultant, and has assisted in the
preparation
of the VITA/USAID Djibouti National Energy Assessment.
VITA is a private, nonprofit organization that supports
people
working on technical problems in developing countries.
VITA offers
information and assistance aimed at helping individuals and
groups to select and implement technologies appropriate to
their
situations. VITA
maintains an international Inquiry Service, a
specialized documentation center, and a computerized roster
of
volunteer technical consultants; manages long-term field
projects;
and publishes a variety of technical manuals and papers.
UNDERSTANDING SOLAR CONCENTRATORS
by
VITA Volunteer George M. Kaplan
I. INTRODUCTION
Although solar energy research, development, and systems
experiments
were conducted in the late 1800s and early 1900s, it was
the sharp increase in the price of oil in 1974 precipitated
by
the Middle-Eastern oil embargo the previous year that
escalated
national and international investment in solar energy.
In the
United States and other industrial countries, the
technological
tools and advancements produced during World War II, the
post-war
rebuilding and prosperity, the U.S. nuclear power and space
programs, and other technological achievements were applied to
solar energy research and development.
The result was that research,
which had been limited to backyard tinkerers and small
specialized companies, was spread to universities, national
laboratories,
and industry. The
federal solar budget rose from less
than $1 million in early 1970s to over $1 billion in the
early
1980s; the budget is now about $200 million, with about $50
million for solar thermal technology.
Solar thermal technology is concerned principally with the
utilization
of solar energy by converting it to heat.
In the concentrating
type of solar collector, solar energy is collected and
concentrated so that higher temperatures can be obtained;
the
limit is the surface temperature of the sun.
However, construction
materials impose a lower, more practical limit for
temperature
capability.
Similarly, overall efficiency of energy collection,
concentration, and retention, as it relates to energy cost,
imposes a practical limit on temperature capability.
If solar energy were very highly concentrated into a tiny
volume,
the result would approach a miniature sun.
If the same energy
were distributed along a thin line, the line would be cooler
than
the miniature sun, but still hot.
If distributed on a large
surface, the surface would be less hot than the line.
There are
solar concentrators that focus sunlight into a point or a
line.
There are also non-focusing concentrators.
Each type has preferred
temperature-dependent applications.
The amount of energy per unit area that can be collected
annually
by a concentrator depends on the positioning of the
concentrator
relative to the sun.
Some types of collectors perform adequately
(cost effectively) if left in a fixed position.
These collectors
generally have limited temperature capability, and provide
little
or no concentration of the incident sunlight.
Most concentrators
would collect so little energy in a fixed position that they
must
be provided with the capability to daily track the sun from
morning (east) to sunset (west) to be cost-effective.
Some concentrators
can only be cost effective by tracking both the sun's
daily path and the sun's annual inclination (which causes
the sun
to appear to move in declination by 47 [degrees] over the
year). Thus,
concentrators may be non-tracking, single-axis tracking
(which
tracks east to west), or two-axis tracking (which tracks
both
east to west and north to south).
Two-axis tracking provides the
maximum solar energy collection but is not cost effective
for
most applications or collector designs.
The U.S. national solar energy research program has led the
world
both in investment and breadth of program.
Because the potential
U.S. market is large, the U.S. national program was aimed at
the
domestic market and was not intended specifically for
export.
Thus, the U.S. experience is primarily applicable to the
U.S. and
may not be relevant to other countries without modification.
For U.S. applications, for example, mirror-type
concentrators are
more cost effective than lens-type concentrators for small,
intermediate,
and large systems for heat generation and use.
Tracking
systems appear most effective for high-temperature
applications.
However, the effectiveness in the U.S. may be due to
sophisticated technology, availability of skilled
maintenance
personnel and spare parts, an excellent supporting
infrastructure,
rather than an inherent advantage of mirrors or tracking
systems. In a less
industrialized environment, lens concentrators
may prove more appropriate.
Although the terms "collector" and "concentrator"
are used interchangeably
in this paper, the terms are distinctive.
A collector
may not concentrate solar radiation, while concentrators are
considered collectors.
No distinction will be made in this paper
unless necessary.
HISTORY OF SOLAR CONCENTRATORS
The concept of concentrating solar rays to heat a target
area has
been known for at least 4,000 years.
In the clay tablet period
of Mesopotamia, polished gold vessels were reputedly used to
ignite altar fires.
Archimedes is said to have saved Syracuse
from invasion by burning the Roman fleet with concentrated
solar
rays reflected from polished metal.
Experiments to verify the story of Archimedes were performed
in
the seventeenth century with polished metal plates.
Glass lenses
were first used to smelt iron, copper, mercury, and other
materials
from their ores in the seventeenth century.
The eighteenth
century brought solar furnaces and solar ovens.
Advancing tech-in
the nineteenth century produced steam engines and hot
air engines operated with solar energy.
Numerous solar engines
and solar furnaces were constructed early in the twentieth
century.
Experimentation continued into the 1930s before languishing
as inexpensive fossil fuels, particularly natural gas,
became
widely available.
The U.S. solar energy program was initiated in 1970 as part
of
the Research Applied to National Needs (RANN) program of the
U.S.
National Science Foundation.
This program expanded enormously as
a result of the oil embargo of 1974 and the price rise of
oil and
other fossil fuels.
As the program goals changed from research
and development and later to commercialization, program
responsibility
shifted to other federal agencies.
The program is now
part of the U.S. Department of Energy; the focus is again on
long-term high-cost, high-risk research and development
unlikely
to be undertaken by industry; responsibility for
commercialization
has been shifted back to industry.
NEEDS SERVED BY THE TECHNOLOGY
Solar concentrators provide high energy density solar radiation
to a target receiver, thus raising the temperature of the
target.
Depending on the degree of concentration, the optical
properties
(solar absorption and radiation) of the target surface, and
the
target's cooling rate, the following may occur:
o
the target will melt (high concentration);
o
the target will reach an equilibrium
temperature with
natural
cooling (modest concentration); or
o
the target will reach an equilibrium
temperature with a
forced
(circulating) coolant (intermediate concentration).
The first instance is that of a solar furnace.
The second may be
considered a solar cooker or solar oven.
In the third instance,
the heated coolant is used directly as, for example, hot
water or
steam in home or industrial applications, or indirectly, as
a
vapor (steam) to generate electricity.
In the case of electricity
production, common energy conversion devices provide an
intermediate
step--shaft rotation--between the heated fluid and
conversion
to electricity.
If the target of the concentrated sunlight is a photovoltaic
cell, or an array of cells, electricity will be produced
directly.
The degree of solar concentration, cell conversion
efficiency,
the design of the cell assembly, and the cell material will
determine if natural circulation or forced circulation
cooling is
necessary for efficient operation of the cell.
Currently, the
cost/unit area of a concentrator is less than the cost/unit
cell
area. As a result,
concentrators are used to reduce cell area.
Should the cell area become less expensive than the
concentrator
area, concentrators would not be utilized.
This paper deals principally with concentrators for thermal
applications
rather than for applications with photovoltaic cells.
Emphasis is placed on applications in less developed
countries.
II. OPERATING PRINCIPLES
SUNLIGHT
Before discussing concentrators, a few words about the sun
are in
order. Beyond the
earth's atmosphere the intensity of sunlight
is about 1,350 watts per square meter (429 British thermal
units
[Btu] per hour per square foot).
Passage through the atmosphere
depletes the intensity due to absorption by the various
gases and
vapors in the air and by scattering from these gases and
vapors
and from particles of dust and ice also in the air.
Thus, sunlight
reaching the earth is a mixture of direct (unscattered) and
diffuse (scattered) radiation.
At sea level the intensity is
reduced to approximately 1,000 watts/square meter (295
Btu/hour/
square foot) on a bright clear day.
The intensity is further reduced
on overcast days.
Most concentrators utilize direct radiation only.
These concentrators
work well on bright clear days, poorly on hazy days, and
not at all on drab gray days when the sunlight intensity is
reduced and the light consists principally of diffuse
radiation.
Another limiting factor is that the sun is not a point but
has a
diameter equivalent to about one-half degree of arc.
Concentrator
design must consider this arc.
GENERIC TYPES AND USE
Although the discussion that follows deals with
concentrators as
entities, concentrators are only a portion of an energy
collection
system. To be useful
the concentrated rays must be directed
to a target called a receiver, which converts the rays into
another form of energy, heat.
The concentrator and receiver must
be matched for optimum performance.
Frequently, the receiver is
expected to impart heat to a fluid in order that the heat be
utilized or dissipated.
When the main purpose of the concentrator
is to obtain heat effectively, then the combination of
concentrator
and receiver must be carefully designed to reduce stray
loss of energy from either the concentrator or receiver.
There are many ways to characterize concentrators.
These include:
o
Means of concentration--reflection or
refraction
o
Point, line, or non-focusing
o
Fixed or tracking concentrator
o
Fixed or tracking receiver
Means of Concentration
Concentration of light is achieved with mirrors (reflection)
or
with transparent lens (refraction).
Cameras and small telescopes
use lenses; large telescopes use mirrors.
A mirror reflects
incoming light so that the angle of the reflected ray is
equal to
the angle of the incident ray (Figure 1).
This relation also
25p05a.gif (486x486)
holds when the mirror is tilted (Figure 2).
A single flat mirror
25p05b.gif (486x486)
does not concentrate but concentration can be obtained by
superimposing
the reflections of many mirrors.
Alternately, concentration
can be achieved by bending the mirror into a pre-determined
shape and relying on the optical properties of the resulting
curved surface.
The lens relies on bending (refracting) incoming light so as
to
converge to a common focus (Figure 3).
As the size of the lens
25p06a.gif (353x353)
increases, lens thickness also increases.
A Fresnel lens (Figure 4)
25p06b.gif (393x393)
maintains the optical characteristics of the standard lens
by
retaining the same curvature piecewise.
This permits a significant
reduction in the thickness and weight of the lens with only
a modest performance penalty.
Each method of concentration has drawbacks.
The mirror requires
a clean smooth reflecting surface: clean since dust
particles
could scatter light away from the receiver or the light
could be
partly absorbed by a thin dirty film; smooth because contour
error can also result in missing the receiver.
The reflecting
material may be placed on the surface of the mirror (first
surface,
Figure 5), or behind a transparent surface (second surface,
25p07a.gif (393x393)
Figure 6). Silver is
the preferred reflector material with
25p07b.gif (393x393)
aluminum second.
Silver is very susceptible to degradation by
moisture and airborne contaminants.
Available protective coatings
have not proven effective for silver in first surface
application.
Aluminum is more durable but less reflective.
Second-surface mirrors have some energy loss due to
absorption of
light by the transparent surface, usually glass or plastic,
as
the light is incident and as it is reflected through the
material.
Low-iron glass is preferred over high-iron glass because
of reduced absorption of light.
If plastic is used, it must be
stabilized against degradation by the ultraviolet light of
the
sun.
Because of the greater thickness of the lens, the degree of
energy absorption is higher than that of the second surface
mirror. The Fresnel
lens, which can be made much thinner than a
standard lens, has less energy loss due to energy absorption
than
the standard lens.
The lens surface must also be clean and smooth for the same
reasons as for the mirror.
Fresnel lens performance is enhanced
when the vertical portion has little or no slope error.
Plastics
can be formed to produce Fresnel lens of higher quality and
less
cost than with glass.
However, plastic lenses tend to deteriorate
under ultraviolet light and must be stabilized.
Point, Line, or Non-Focusing
One criterion for selection of a specific concentrator is
the
degree of concentration and hence temperature that is to be
achieved. As a rule,
concentrating energy onto a point produces
high to very high temperature; and onto a line, moderate to
high
temperature.
Non-focusing concentrators produce low to moderate
temperature.
Point. The parabolic
dish reflector (Figure 7) utilizes the
25p08.gif (393x393)
optical properties of the parabolic curved surface to
concentrate
direct light to the focal point.
The dish geometry is
familiar being used for automobile headlights, searchlights,
radar, and to receive transmissions from broadcast
satellites.
Standard circular and Fresnel lenses are also point focus
concentrators.
The Fresnel lens has been utilized in conjunction with
photovoltaic cells in several test installations in the
United
States and abroad.
The overlapping images from many flat mirrors can be
considered
the equivalent of point focusing.
The focal shape is not a point
but rather the finite image of the sun further broadened by
the
characteristics of the reflector material and various errors
in
manufacture and in the precision of image overlap.
Figure 8
25p09a.gif (393x393)
illustrates the central receiver concept wherein heliostats
(flat
or slightly curved mirrors mounted on tracking devices)
redirect
the sun's rays toward a receiver atop a tower.
A 10-megawatt
electrical generating plant employing this principle has
been
successfully operated in California since 1982.
Line. The parabolic
trough (Figure 9) is an example of line focus
25p09b.gif (393x393)
optics. The
incident direct radiation is reflected from the
trough to the focal line the length of the trough.
To maximize
energy collection the trough is designed to track the
sun. The
trough may be oriented with the focal line running east-west,
north-south, or north-south with simultaneous tilt toward
the sun
(polar mount).
Each orientation has its own seasonal and yearly collection
characteristics.
No one orientation is universally preferred (i.e.,
is more cost-effective).
The standard and Fresnel lenses can be fabricated in linear
form
(Figure 10) with the same cross section as the circular lens
but
25p10.gif (534x534)
now producing a focal line instead of a focal point.
Plastic
linear Fresnel lenses of good quality can easily be produced
by
extrusion.
The hemispherical bowl (Figure 11) is another example of
linear
25p11a.gif (540x540)
focal optics. Unlike the trough or lens, two-axis tracking
is
mandatory. The
hemispherical bowl is always fixed, and the receiver
does the tracking.
The focal line falls on the line connecting
the center of the sphere with the sun.
The focal line is
restricted to the lower half of the radius by the optical
properties
of the bowl. Because
some rays reach the focal line with
only one reflection and others require multiple reflections,
the
intensity is not uniform along the length of the focal line.
Figure 12 shows a 65-foot (19.7-meter) diameter experimental
bowl
25p11b.gif (600x600)
that has operated successfully in Texas for many years.
Annual
energy collection is lower than for other collector optics
and
there appears to be no compensating advantages, except that
it is
much easier for a small receiver to track the sun's image
than it
is for a larger and much heavier concentrator.
Non-Focusing. The hemispherical trough (Figure 13) and the
flat
25p12a.gif (393x486)
plate collector with booster mirrors are examples of
concentrators
that are non-focusing.
Non-focusing concentrators do not
focus sunlight into a specific geometrical shape, but
reflect
sunlight onto a receiver, thus increasing the total amount
of
sunlight received.
The category of non-focusing concentrators
also includes concentrators in which the focus is of poor
quality.
The cylindrical collector (Figure 14), a variation of the
25p12b.gif (437x437)
hemispherical trough, is of interest because the entire
cylinder
may be fabricated with inexpensive, inflatable plastic.
A simple method of achieving a modest increase in
concentration
on a large area is to use booster mirrors in conjunction
with a
flat plate collector (Figure 15).
Before noon the mirrors face
25p13a.gif (437x540)
east; after noon they face west.
The energy collection advantage
of boosters for a flat plate collector is shown in Figure
16.
25p13b.gif (437x437)
Fixed or Tracking Concentrators
Maximum energy collection on a daily or annual basis
requires
tracking of the sun (or the sun's reflected image) since
concentrators,
particularly those capable of high concentration, utilize
only direct radiation.
Thus a parabolic dish, when pointed
at the sun, has reflected rays passing through the focus. As
the
sun moves, some of the reflected rays will miss the focus
and, in
time, all will miss the focus.
The dish must be moved to maintain
the reflected rays at the focus.
The central receiver,
parabolic dish, parabolic trough, standard lens, and Fresnel
lens
are examples of tracking concentrator systems.
The hemispherical bowl likewise must continuously track the
sun.
Large bowls are too unwieldly to move.
Thus, the receiver is
moved continuously instead. It tracks the focal line of the
sphere (the reflected image of the sun) throughout the day.
Like the hemispherical bowl, the Russell concentrator is
fixed
and the receiver must track the sun's image (Figure
17). This
25p14.gif (393x486)
concentrator consists of long narrow mirrors whose centers
all
fall on the perimeter of a circle.
The mirrors are oriented so
that all reflected images focus on a point on the same
perimeter.
As the sun moves the focus moves along the perimeter.
The Winston collector is usually considered a non-tracking
concentrator.
Its energy collection can be increased by tracking. As
a trough-type collector (Figure 18), it consists of a
parabolic
25p15.gif (486x486)
surface whose axis is horizontal and whose focal point is
close
to the surface. The collector is frequently found as a
paraboloid
in shape but can also be in trough form.
The collector accepts
both direct and diffuse radiation.
The acceptance angle (angle
of acceptance of sunlight) depends on the height of the
parabola.
The shorter the height, the greater the acceptance angle and
the
period of daily operation, but the less the concentration
and
maximum temperature capability.
The collector has been utilized
as a highly effective fixed collector, which reaches higher
temperature than a typical flat plate collector.
Fixed or Tracking Receivers
The central receiver and parabolic trough have fixed
receivers,
due to the optical characteristics of the systems.
The parabolic
dish receiver is usually positioned at the focus so as to
move
with the dish as the dish tracks the sun.
Neither the bowl nor
the Russell collector track the sun, hence their receivers
must
track the sun's image.
The Winston collector, the cylindrical
collector, and the flat plate collector with booster mirrors
are
normally utilized in fixed position and with fixed
receivers. The
flat plate is, of course, both the collector and the
receiver.
Other Fixed Concentrators
There are many ingenious concentrators that work quite well
and
can be cost effective in some applications.
The cusp collector
(Figure 19), whose surface geometry is the locus of the
position
25p16a.gif (486x486)
of the end of a string as it is unwrapped from a pipe can
provide
a modest concentration suitable for hot water.
A conical collector
(Figure 20) can be substituted for the Winston paraboloid,
25p16b.gif (540x540)
gaining simplicity of manufacture with some performance
penalty.
Similarly, flat reflectors can substitute for the parabolic
sides
of the Winston trough collector.
Table 1 summarizes the characteristics and potential uses of
the
concentrators described above.
Table 1. Classification of Concentrators
Type Sun's
Tracking
Capability of
Type of
of Lens or
Concen-
Tracking Receiver
Temperature
Typical
Concentrator
Focus Mirror
tration
(yes/no) (yes/no)
([degrees] C)
([degrees] F)
Applications Comments
Parabolic
point mirror
> 1000
yes yes
>2638
>3000
electricity Small-scale
applications
dish
two-axis
heat
Central
point mirror
> 1000
yes no
>2638
>3000
electricity Large-scale
applications
receiver
two-axis
heat
Lens point
lens
> 1000 yes
yes
>2638
>3000
electricity Utilized with
photovoltaic cells
(round)
two-axis
heat
Parabolic
line mirror
100
yes no
538
1000
electricty Can be used for
both small and
trough
one-axis
heat large systems
Fixed mirror
line mirror
100
no yes
538
1000
electricity Can be used for
both small and
moving focus
one-axis
heat
large systems; not economic in
U.S. experience
Lens
line mirror
100
yes yes
538
1000
electricity Little U.S.
experience
(linear)
one-axis
heat
Sphere
line mirror
80
no yes
538
1000
electricity Awkward in large
size
two-axis
Cylinder
line mirror
2
no no
121
250
heat
Cusp
line mirror
1.5-2.5 no
no
121 250
heat
Winston
line mirror
3 - 6
no no
121
250
heat Concentration
decreases as
acceptance
angle increases
Flat plate
with booster
area mirror
> 1
no no
121
250
heat
booster and < 2
ANNUAL ENERGY COLLECTION EFFICIENCY
Collectors that maintain their surfaces facing the sun
(right
angle for most collectors) have the highest annual
collection
efficiency. The
parabolic dish and other two-axis tracking collectors
are examples. The
central receiver, although a two-axis
tracking system, does not direct the heliostat reflectors to
face
the sun but rather maintains an angle to the sun so that the
image is reflected to the receiver.
As expected, its collection
efficiency is lower than the dish.
The parabolic trough is a
single-axis tracking system; thus, the surface is only
occasionally
at a right angle to the sun and has a lower annual
collection
efficiency than the central receiver.
Fixed collectors with tracking receivers such as the bowl
and
Russell collector have even lower collection
efficiency. The
least efficiency is exhibited by Winston and other fixed
collectors
and receivers.
The theoretical annual efficiency of the three principal
concentrating
collectors utilized in the United States is 80 percent
for the dish, 60 percent for the central receiver, and 43
percent
for the parabolic trough on an annual basis.
Collector efficiency
is determined for the period extending from the beginning
of tracking when the sun climbs to 15 degrees above the
horizon
until tracking stops when the sun declines below 15 degrees
at
the end of the day.
The efficiency depends on direct solar radiation
and system optics.
Actual efficiency depends on mirror or lens surface
accuracy,
surface dust and film, energy absorption by lens or
mirror, the
properties of the reflecting, material, pointing accuracy,
effects
of temperature variations on these factors,
weather--including
clouds, dust and haze, and so on.
The efficiency is further
reduced by receiver performance and receiver subsystem
design,
including care given to reduction of heat loss by
conduction,
convection, and radiation.
III. DESIGN
VARIATIONS AND EXPERIENCE
PARABOLIC DISHES
A recent paper on the parabolic dish prepared by the Jet
Propulsion
Laboratory(*) describes nine designs sponsored by the U.S.
(*) V.C. Truscello, "Status of the Parabolic Dish
Concentrator,
Proceedings of the Energy Research and Development Agency
Conference
on Concentrating Solar Collectors, Georgia Institute of
Technology,
September 26-28, 1977
(Washington, D. C.: U. S.
Department
of Energy, undated, circa 1982-1983).
Department of Energy, eight privately-funded U.S. designs,
and
10 dishes developed by other countries.
Although no two dishes
are identical, they fall into four categories:
1.
Rigid reflector.
The reflective surface is attached to
a rigid
curved structure. This is the standard
(radar
type)
structure (Figure 21).
25p20a.gif (437x437)
2.
Pressure-stabilized membrane.
The reflective surface is
attached to
a flexible membrane, which takes the
shape
of a rigid,
curved support structure by creation of a
vacuum
between the membrane and structure. The
intent
is to reduce
cost by reducing weight of materials of
construction
(Figure 22).
25p20b.gif (486x486)
3.
Fresnel lens or Fresnel mirror.
The lens is built up
from several
narrow concentric parts; the mirror is
a series of
concentric reflective surfaces. The intent
is to reduce
cost by simplifying the compound curvature
of the
paraboloid (Figure 23).
4.
Secondary reflector.
A second mirror, which may be
hyperbolic(*) (cassegrain) or elliptic(**) (gregorian),
reflects the
rays from the parabolic reflector to a
receiver
behind the parabola. The intent is to eliminate
the heavy
receiver structural demands on the
dish and
also to provide easy access to the receiver
for
maintenance (Figure 24).
The rigid reflector has been the most popular since it
resembles
current radar technology.
The Shenandoah project, a U.S. Department
of Energy demonstration project near Atlanta, Georgia,
deployed
114 7-meter-diameter dishes coated with a reflective film
to produce 399 [degrees] C (750 [degrees] F) steam.
The steam was used to generate
400 kilowatts of electricity and process steam at 9.70
kilograms
per square centimeter
(138 pounds per square inch gauge [psig])
for an adjacent knitwear factory.
After some initial problems,
the system is now operating satisfactorily.
The project is a
joint effort of the U.S. Department of Energy, the local
power
company, and the knit-wear factory.
Its goal was to demonstrate
the viability of rigid-reflector collectors, not to be a
commercial
prototype.
(*) A curve formed by the section of a cone cut
by a plane that makes a greater angle with
the base than the side of the cone makes.
(**) Oval-shaped.
25p19.gif (393x393)
CENTRAL RECEIVERS
The best U.S. example of a central receiver is Solar One, a
joint
project of the U.S. Department of Energy and two Southern
California
utilities. This
10-megawatt electric pilot plant utilizes
1,818 heliostats (or reflectors), each with 41.8 square
meters
(450 square feet) of second-surface glass mirrors.
The heliostats
surround a tower on which the receiver is located.
Most of the
heliostats are located south of the tower.
The plant has exceeded
its specifications and is operating very successfully.
The design
was based on a 100-megawatt plant and then reduced to 10
megawatts.
An optimized 10-megawatt plant would likely have a different
heliostat field configuration.
A 100-megawatt version (Solar 100) with similar technology
is
being considered by the utilities, assuming government
investment
credits are provided.
Without these financial incentives, the
plant would not be economical in the United States due to
falling
oil prices. However,
such a plant may be economical in other
countries with high energy costs.
Heliostats have evolved through a series of designs that
reduced
the initial weight of over 97.6 kilograms/square meter (20
pounds/square foot) to about 39 kilograms/square meter (8
pounds/square
foot). Over 20 heliostat designs have been constructed and
tested. The current
preference is for a second-surface glass
mirror on a glass backing. The U.S. Department of Energy's
Solar
Energy Research Institute is developing a lightweight
reflector
(plastic/silver/plastic), which promises to drastically
reduce
the cost of heliostats.
When developed, the material may be of
interest for use in less-industrialized countries.
Heliostat size is governed by rigidity and wind load
requirements.
Due to the present cost elements of heliostats (which are
influenced by the fact that every heliostat needs its own
tracking
system), in the United States, system designs favor large
heliostats. The
distribution of cost elements may vary in other
countries. While
only larger central receivers are likely to be
economical in the United States, some advanced developing
countries
may be able to utilize the smaller Solar One technology
economically.
LENSES
Circular lenses, whether standard or Fresnel, tend to be
limited
in size, much like the parabolic dish.
Size is also limited by
current fabrication capabilities.
Small glass lenses for cameras
and spotlights are available, as are larger plastic
lenses. But
a 7-meter diameter lens (a size comparable to the Shenandoah
dish) is certainly not widely available either in glass or
plastic.
In large sizes, a glass lens would be very heavy;
plastic,
probably in a Fresnel design, is likely to be the only
practical
lens, if available.
Linear Fresnel lenses may offer the advantage
of being fabricable in both small and large widths and
lengths.
PARABOLIC TROUGHS
A significant number of parabolic troughs have been
designed,
built, and tested, primarily with private funds.
Many types are
available on the market.
Troughs differ in their reflective
materials, structural materials, receiver concepts, etc.
The
attainable temperature reaches about 540 [degrees] C (1000
[degrees] F). The designs
vary with intended temperature application, since surface
error,
tracking error, and receiver losses assume considerable
importance
for a high temperature design.
Troughs have been utilized by many federal demonstration
projects
to provide process heat for industrial applications and to
supply
vapor for suitable small engines (e.g., irrigation pump
devices).
All designs had initial problems, usually with materials and
nonsolar
hardware. After
repair or modification, operation was reliable
and successful. Many
federally-funded projects tended to
be shut down when they ended and rarely restarted because of
lack
of sustained interest by the user.
An excellent source of information
on private trough manufacturers is the Solar Energy
Industries
Association (SEIA) in Washington, D.C.
Troughs may be attractive because of their relative
simplicity.
Because their surface curvature is singular, not compound as
for
dishes, troughs are more easily fabricated.
A second-surface
reflective plastic with adhesive backing can be easily
placed on
the curved substrate.
A simple pipe or tube will serve adequately
as the receiver although various simple techniques, such as
a
glass vacuum jacket around the receiver tube, will enhance
performance.
Single-axis tracking is less complex than two-axis
tracking.
IV. SPECIAL TOPICS
RECEIVERS
The concentrated sunlight must be converted to a useful form
of
energy, usually heat.
If desired, heat can be converted to electricity
by means of an engine and generator.
The receiver should
be designed to minimize heat loss.
Heat loss occurs through
radiation to a cooler object; through convection currents
created
by heating air in contact with the hot receiver surface; and
through conduction from the hot parts of the receiver to
colder
parts and to attached structural members and insulation.
Heat retention by the receiver is enhanced by covering the
receiver
with a selective coating which will absorb virtually all
the concentrated radiation but will reradiate comparatively
little energy.
Furthermore, since the total energy radiated
depends directly on the radiating area, the receiver surface
area
should be minimized.
Convection can be reduced by preventing the
build-up of air currents that remove air heated by the
receiver
and provide the receiver with colder air for continued heat
loss.
A transparent window (glass or plastic depending on
temperature)
can reduce air currents.
The window introduces other heat loss and heat gain effects.
Some energy will be reflected from the front surface and
rear
surface of the window and never reach the receiver.
Additional
energy will be absorbed by the window and not reach the
receiver.
The inner surface of the window may be coated with a heat
mirror
such as tin oxide, which reduces the radiation loss by
reflecting
radiated energy back to the receiver.
Etching of the outer surface
of a glass window reduces the reflection from the surface.
Insulation serves to reduce convection and radiation losses
from
parts of the receiver outside the path of the incoming
radiation.
Conduction loss is reduced by decreasing the cross-section
of
structures in direct contact with the receiver, and using
poor
heat conductors for these structures where possible.
Creating a
vacuum between the window and the receiver will further
reduce
convection and conduction losses.
Figure 25 shows the reflectivity of several mirror
systems. Note
25p24a.gif (540x540)
not only the differences in reflectivity but also that for
some
materials the reflected energy falls within a small solid
angle*
(Figure 26). These
materials allow a small target area for
25p24b.gif (486x486)
receipt of the reflected rays.
If a larger solid angle is required
to enclose the reflection, then a compromise between
target size and loss of reflected rays must be made.
Energy which
is not reflected is converted to heat at the reflecting
surface.
This may require positive cooling efforts to ease or
eliminate
thermal stress.
COST
Concentrator cost represents only one portion of the cost of
a
system. The cost of
the quantity of heat delivered at the required
temperature is the preferred method of determining cost.
For a given system, the cost per million kilowatt-hours, or
kWh
(per million Btu) usually decreases as the total number of
kWh
(Btu) delivered increases, i.e., as system size
increases. Similarly,
the cost per million kWh (per million Btu) is likely to be
less at lower temperatures than at higher temperatures.
In general,
the higher the concentration and complexity, the higher the
cost.
(*) If you have an angle, one side of which is vertical and
the
other side not vertical, and that side is rotated around the
vertical
(maintaining the same angle), the angle created is called
the solid angle.
Cost is frequently represented by purchase price but not
always.
Sellers may reduce selling price to penetrate a market, to
expand
market share, to anticipate future manufacturing economies
and
cost reductions, and to limit or exclude potential competition.
Sellers with a monopoly or a preferred position may sell at
higher than reasonable rates.
Sellers faced with unknown or
indeterminate risks and liabilities for the product will try
to
transfer the risk to the purchaser through higher prices or
other
means.
In the United States, many solar energy systems are cost
effective
only because of federal and state tax policies to aid the
solar energy industry.
These systems cost two to five times more
than competing energy systems.
However, energy costs in many
less-developed countries are several times greater than in
the
United States, and therefore solar systems may be cost
effective
in those countries.
In the United States, the cost of a solar thermal electric
system
utilizing relatively new technology and incorporating
research
and development costs would range from $10 to about $30 per
watt.
The central receiver experiment in California (Solar One)
cost
about $15 per watt; a proposed 100-megawatt plant
incorporating
the lessons of Solar One and the economies of a ten-fold
increase
in size is anticipated to cost about $4 per watt.
Heliostats were
about one-third of the total cost of Solar One, and are
expected
to be about one-half the cost of the large plant.
(A coal-fired
electric plant costs about $1.00-$1.40 per watt of installed
capacity.)
Studies of dish technologies indicate costs ranging to $50
per
watt for the system, with dish costs of one-third to
one-half of
the system cost.
Dish technology is well behind heliostat experience.
Parabolic troughs appear to cost about $538 per square
meter ($50 per square foot) at present with possible
reduction to
about $270 per square meter ($25 per square foot) with a
larger
market. Again, these
costs reflect only one-third to one-half the
system cost.
Of possible interest to developing countries is the class of
collectors using transparent plastic in cylindrical form
with the
reflector film partially located in the lower arc and a
"black"
tube located at the focus.
This type of collector appears to
offer low cost. Some
versions using an evacuated glass tube with
an inner blackened copper tube in "once through"
(straight tube)
or bayonet style are commercially available in the United
States
(Figures 27, 28, and 29).
25p26a0.gif (81x486)
The hemispherical bowl has been tested in Crosbyton, Texas,
by
the U.S. Department
of Energy. The unit, 20 meters in
diameter,
produced high temperatures and high pressure steam suitable
for
modern steam turbines.
The compound curvature is difficult to
build, as is the two-axis tracking required of the receiver.
However, a tracking receiver is simpler than a tracking
concentrator.
The concentrator may be more acceptable in smaller size
and lower concentration (temperature).
The reduction in concentration
will decrease temperature, which increases the number of
materials that can be used for the receiver, and may ease
fabrication
of the sphere.
To compare solar thermal technologies, costs should be
reduced to
common bases such as cost per watt electric or per kWh
(Btu). The
base should distinguish between average and peak capacity;
the
amount of storage incorporated; temperature, if heat is the
desired end product; and the yearly energy delivered.
Other
technologies have their own bases; photovoltaics use cost
per
peak watt, and installed cost per annual kilowatt-hour
produced.
Electricity from wind energy, as well as from other solar
electric
technologies, may have different value to the user depending
on the time of generation.
These considerations should be included
in any evaluation methodology for selection of
cost-effective
systems.
V. COMPARING THE ALTERNATIVES
Simple flat plate collectors are the most widely used and
most
cost-effective solar collectors.
Their primary use is for domestic
and commercial (e.g., hospitals, restaurants, etc.) hot
water
applications; however they may also be used in preheat
systems
for higher temperature applications.
They can achieve a temperature
of about 38 [degrees] C (100 [degrees] F) above the ambient
by capturing sunlight,
converting sunlight to heat, and carefully minimizing
unwanted heat loss from the collector.
Flat plate (usually non-tracking) collectors are the
simplest to
fabricate. Simple,
unsophisticated, functioning collectors can
easily be built with simple tools.
Care must be taken to enhance
solar collection and prevent thermal losses.
Careful use of local
materials to the maximum extent possible can reduce
cost. While
selective absorbers enhance performance and yield higher
temperature,
almost any "black" surface will perform
adequately. Some
simple, low-cost flat plate collectors may be better than
concentrators
for temperatures below 93 [degrees] C (200 [degrees] F),
particularly in
less-industrialized countries.
Expectations of better performance
for flat plate (non-concentrating) collectors over
concentrating
collectors, for the same temperature application, have not
been
verified in practice.
The expectations were based on utilization
of both direct and diffuse radiation by flat plate
collectors and
use of only direct radiation by concentrators.
BIBLIOGRAPHY/SUGGESTED READING LIST
Reports and Conference Proceedings
Dougherty, D.A. Line-Pocus Receiver Heat Losses.
SERI/TR-632-868.
Golden,
Colorado: Solar Energy Research Institute, July 1982.
Murphy, L.M. Technical and Cost Potential for Lightweight,
Stretched-Membrane Heliostat Technology.
SERI/TP-253-2070.
Golden,
Colorado: Solar Energy Research Institute, January
1984.
Scholten, W.B. A Comparison of Energy Delivery Capabilities
of
Solar
Collectors. McLean, Virginia: Science Applications,
Inc., 1983.
Solar Energy Research Institute.
Solar Thermal Technology Annual
Evaluation
Report, Fiscal Year 1983. Golden,
Colorado: Solar
Energy Research
Institute, August 1984.
Truscello, V.C. "Status of the Parabolic Dish
Concentrator."
Proceedings of
the Energy Research and Development Agency
Conference on
Concentrating Solar Collectors. Georgia
Institute
of Technology,
September 26-28, 1977. Washington,
D.C.:
U.S. Department
of Energy, undated (circa 1982-1983).
U.S. Department of Energy.
Solar Parabolic Dish Annual Technology
Evaluation
Report, Fiscal Year 1982.
DOE/JPL1060-63. Washington,
D.C.: U.S.
Department of Energy, September 15, 1983.
U.S. Department of Energy/Sandia Laboratories.
Proceedings of the
Line-Focus solar
Thermal Energy Technology Development Conference,
A Seminar for
Industry (September 9-11, 1980).
Washington,
D.C.: U.S. Department of Energy, September 1980.
Books
Duffie, J.A., and Beckman, W.A. Solar Engineering of Thermal
Processes.
New York, New
York: John Wiley and Sons, 1980.
Kreith, F., and Kreider, J.F. Principles of Solar
Engineering.
Washington, D.
C.: Hemisphere Publishing Corp., 1978.
Lunde, P.J. Solar Thermal Engineering.
New York, New York: John
Wiley and Sons,
1980.
Meinel, A.B., and Meinel, M.P. Applied Solar Energy.
Reading,
Massachusetts:
Addison-Wesley Publishing Co., 1976.
SOURCES OF INFORMATION
Government Printing Office
Many government reports
Washington, D.C. 20402 USA
are available through
this office.
Jet Propulsion Laboratory
4800 Oak Grove Drive
Pasadena, California 91103 USA
National Technical Information
Source of most federal
Service
project reports
5285 Port Royal Road
Springfield, Virginia 22161 USA
Solar Energy Industries Association
List of manufacturers
1717 Massachusses Avenue N.W.
with companies and
Washington, D.C. 20036 USA
systems
Solar Energy Research Institute
Information on all
1617 Cole Boulevard
thermal systems
Golden, Colorado 80401 USA
U.S. Department of Energy
Information on all
Office of Thermal Systems
thermal systems
1000 Independence Avenue,
S.W.
Washington, D.C. 20585 USA
========================================
========================================