Limits of photovoltaic panel concentration?

[John Symons]

About 20 years ago, ARCO Solar built an 8 MW photovoltaic site in Carissa
Plains, CA. They used rows of standard photovoltaic panels mounted between
angled flat mirrors. The mirrors were used to reflect more sunlight onto
the panel for a solar concentration of 2 (twice normal). The system was
mounted on a two-axis tracker to keep it perpendicular to the sun at all
times (like a heliostat). There were about 1000 of these systems installed
at the site, I believe. I heard the EVA used in the old panels discolored
at the higher temperature and caused problems.

Have any flat panel, low-concentration systems been built recently? Is the
EVA used in current photovoltaic panels much improved over the chemistry
used 20 years ago, or would similar problems occur?

There are several flat panel trackers on the market, but I haven't seen any
that use flat mirrors to increase output. Is this just a warranty issue or
are there real problems with a system like this?

Thanks for the reply!

John Symons

 

[SJC]

About 20 years ago, ARCO Solar built an 8 MW photovoltaic site in Carissa
Plains, CA. They used rows of standard photovoltaic panels mounted between
angled flat mirrors. The mirrors were used to reflect more sunlight onto
the panel for a solar concentration of 2 (twice normal). The system was
mounted on a two-axis tracker to keep it perpendicular to the sun at all
times (like a heliostat). There were about 1000 of these systems installed
at the site, I believe. I heard the EVA used in the old panels discolored
at the higher temperature and caused problems.

Have any flat panel, low-concentration systems been built recently? Is the
EVA used in current photovoltaic panels much improved over the chemistry
used 20 years ago, or would similar problems occur?

There are several flat panel trackers on the market, but I haven't seen any
that use flat mirrors to increase output. Is this just a warranty issue or
are there real problems with a system like this?

Thanks for the reply!

John Symons

I still think that most panel manufacturers recommend that you do not use mirrors.
It has to do with the EVA and also the increased heat. It voids the warranty.

 

[Roland Mösl]

About 20 years ago, ARCO Solar built an 8 MW photovoltaic site in Carissa
Plains, CA. They used rows of standard photovoltaic panels mounted between
angled flat mirrors. The mirrors were used to reflect more sunlight onto
the panel for a solar concentration of 2 (twice normal). The system was
mounted on a two-axis tracker to keep it perpendicular to the sun at all
times (like a heliostat). There were about 1000 of these systems installed
at the site, I believe. I heard the EVA used in the old panels discolored
at the higher temperature and caused problems.

Have any flat panel, low-concentration systems been built recently? Is the
EVA used in current photovoltaic panels much improved over the chemistry
used 20 years ago, or would similar problems occur?

There are several flat panel trackers on the market, but I haven't seen any
that use flat mirrors to increase output. Is this just a warranty issue or
are there real problems with a system like this?

Let's assume to concentrate 2 times the solar energy.

This means 2000 instead of 1000 W per squaremeter.

Both side of the photovoltaic can put away heat
With nearly no wind with about 7 W per squaremeter.

The normal panel becomes 70 degree Celsius more
hot than the air temperature.

The concentrator panel becomes 140 degree Celsius more hot

70 degree Celsius means at cristallin silicon with 0,48% less
output per degree more 33,6% less efficiency.

So the 1:2 concentrator does not produce the double energy.
Because of the worse efficiency at higher temperatures,
it is only 32,8% more.

When there is a strong wind, the temperature difference
is maybe only 40 degree Celsius, 19,2% less efficiency,
bringing by concentration 61,6% more.

 

[[email protected]]

As I understand it, the makers would have to use a silicone material,
maybe slightly different solder connections and the panels would need
some little additional cooling to make them work well in concentrators.

Silicone? Some say EVA browns at high temps, avoidable with more heat
sinking, eg a thin poly film water duct over a horizontal panel face,
under a reflective wall, and solder joints and ion migration may be OK
at 120 F in 2-3 suns.

Nick

 

[[email protected]]

Let's assume to concentrate 2 times the solar energy.

This means 2000 instead of 1000 W per squaremeter.

Both side of the photovoltaic can put away heat
With nearly no wind with about 7 W per squaremeter.

Each side of an outdoor surface might move 2 Btu/h-F-ft^2
in still air, ie 11 W/m^2C, or more, with radiation.

The normal panel becomes 70 degree Celsius more
hot than the air temperature.

If 85% of 750 W/m^2 C AM2 sun becomes heat in a T (C) panel in 20 C air,
and it loses heat by convection and radiation...?

The concentrator panel becomes 140 degree Celsius more hot

.... 160 C on a 20 C day? Maybe cooler, with radiation loss...

70 degree Celsius means at cristallin silicon with 0,48% less
output per degree more 33,6% less efficiency.

....70x0.0048 = 0.336, and 1-0.336 = 0.664, so it looks like this cell
would produce 66.4% of the electrical energy it would produce at 20 C.

So the 1:2 concentrator does not produce the double energy.
Because of the worse efficiency at higher temperatures,
it is only 32,8% more.

.... 140x0.0048 = 0.672, but 2(1-0.672) = 0.656, so the 2:1 PV might
produce less electrical power than the 1:1, using these calcs...

But if it's cooled to 60 C with a thin polyethylene film water bag on top
that absorbs 6% of the solar power, it might make 2x0.94(1-40x0.0048) = 1.52
times more power than a 1:1 PV at 20 C, or 1.52/0.664 = 2.3X more power
than a 1:1 PV at 90 C, and also make hot water for showers.

Nick

 

[daestrom]

Each side of an outdoor surface might move 2 Btu/h-F-ft^2
in still air, ie 11 W/m^2C, or more, with radiation.


If 85% of 750 W/m^2 C AM2 sun becomes heat in a T (C) panel in 20 C air,
and it loses heat by convection and radiation...?


... 160 C on a 20 C day? Maybe cooler, with radiation loss...

Indeed. If the backside coating can be something with an emissivity of 0.9,
at 160 C (413K) facing a 20 C (293K) ground, radiant losses could be as much
as 1108 watts/m^2.

When combined with free convection of a modest 7 W/m^2-K on both front and
back surfaces, 2000 W/m^2 and 20 C ambient should stabilize the temperature
below 111C (neglecting reflection or PV conversions, which would make the
temperature lower). In a fairly nice wind with 25 W/m^2-K, the equilibrium
temperature would be even lower, at ~56 C.

Of course, radiant cooling only works on the backside of the unit, since the
front is facing a hotter object (hopefully ;-)

...70x0.0048 = 0.336, and 1-0.336 = 0.664, so it looks like this cell
would produce 66.4% of the electrical energy it would produce at 20 C.


... 140x0.0048 = 0.672, but 2(1-0.672) = 0.656, so the 2:1 PV might
produce less electrical power than the 1:1, using these calcs...

But if it's cooled to 60 C with a thin polyethylene film water bag on top
that absorbs 6% of the solar power, it might make 2x0.94(1-40x0.0048) =
1.52
times more power than a 1:1 PV at 20 C, or 1.52/0.664 = 2.3X more power
than a 1:1 PV at 90 C, and also make hot water for showers.

This points out that keeping the panel cool is an excellent way to improve
performance. But water on the face would have to be kept thin to avoid too
much absorption of the incoming light. In areas where the heat can be put
to use, forced cooling could be used to boost PV output as well as
supplement other heating supplies.

daestrom

 

[[email protected]]

Of course, radiant cooling only works on the backside of the unit, since the
front is facing a hotter object (hopefully ;-)

On a clear day, a horizontal panel might see something like
0.0025x10000K+0.9975x300K = 324 K...

This points out that keeping the panel cool is an excellent way to improve
performance. But water on the face would have to be kept thin to avoid too
much absorption of the incoming light...

I'm not sure that matters much. I measured 94% of the free air power output
under 2" of water in a poly film duct. Water-poly and poly-glass interfaces
have little loss, esp if there's no air in the poly-glass sandwich.

Nick

 

[Steve Spence]

About 20 years ago, ARCO Solar built an 8 MW photovoltaic site in Carissa
Plains, CA. They used rows of standard photovoltaic panels mounted between
angled flat mirrors. The mirrors were used to reflect more sunlight onto
the panel for a solar concentration of 2 (twice normal). The system was
mounted on a two-axis tracker to keep it perpendicular to the sun at all
times (like a heliostat). There were about 1000 of these systems installed
at the site, I believe. I heard the EVA used in the old panels discolored
at the higher temperature and caused problems.

I have a set of Carrizo Gold Quadlams. Still as good as the day they
were made.

 

[Douglas Siebert]

Let's assume to concentrate 2 times the solar energy.

This means 2000 instead of 1000 W per squaremeter.

Both side of the photovoltaic can put away heat
With nearly no wind with about 7 W per squaremeter.

The normal panel becomes 70 degree Celsius more
hot than the air temperature.

The concentrator panel becomes 140 degree Celsius more hot

70 degree Celsius means at cristallin silicon with 0,48% less
output per degree more 33,6% less efficiency.

So the 1:2 concentrator does not produce the double energy.
Because of the worse efficiency at higher temperatures,
it is only 32,8% more.

When there is a strong wind, the temperature difference
is maybe only 40 degree Celsius, 19,2% less efficiency,
bringing by concentration 61,6% more.

I've read about PV losing efficiency due to higher temperatures, but I had
no idea it was that big of a difference! Why aren't panels built to allow
some way of removing the excess heat if it robs efficiency to such a large
degree? Given the difference between no wind and a "strong wind", it sure
sounds like it would be worth the electricity to run some fans underneath
the panels to take away that heat and insure there is always a strong wind
no matter what the weather!

Better yet, build some small capillaries into the structure of the modules
and run a water/gylcol mix through them to take all that heat away, which
can then be run through a heat exchanger to serve a useful purpose, such as
providing hot water for the home or to heat a swimming pool. If you have
more heat than you can use, just run it through a radiator with lots of
surface area and a few fans. Or for a really cheap option, just use that
heat to boil off cold water into steam.

 

[John Symons]

No no no! This is not a heliostat. A heliostat is
generally a flat, or nearly flat, mirror reflection light
onto a stationary target.

Sorry about the confustion. I meant to say that the hardware was very
similar to a heliostat, not the direction of pointing. The 24" diameter
pedestal tube, the Winsmith gear drive, the 12" diameter torque tubes, the
truss structures (to create the flat surface), the drive motors, and the
control hardware were all previously designed for use as a heliostat. Since
heliostats do not point directly at the sun, the control system software
must have been modified.

My understanding is that the encapsulant has been improved
to handle a bit higher temperatures with longer lifetimes.

The encapsulant is not deteriorated by higher concentration
factors. It's temperature that is the problem.

True, but without active cooling, the higher concentration will result in a
temperature increase. I have no information on the allowable temperatures
of the new EVA. Have you heard anything about it?

The Sunpower PV panels with a flux of 1000 watts/m^2 can convert about 20%
of that energy into electricity. That leaves 800 watts to be convected and
radiated from both sides of the panel. Assuming 3/4 of the thermal load is
taken off the front surface and the emissivity of the panel is 95%, the
temperature of the surface should raise about 49 degrees C above ambient.
The convective heat transfer coefficient on the front surface should be
about 5.33 W/m^2C without wind. I was suprised that the PV surface was that
hot. Does that sound reasonable?

With a net concentration of 1.75, there should be 1750 watts/m^2 arriving on
the panel (a little conservative since 1/3 of the indirect solar flux will
be blocked by the mirror.). I expect the overall conversion efficiency will
be reduced to 18% or so with the higher silicon temperature (will need to
verify this assumption later). That should leave a total of 1475 watts of
thermal energy. Assuming the same 75% heat transfer from the front surface,
the surface temperature should rise about 78 degrees C over ambient and the
new convection heat transfer coefficient should be about 5.81 W/m^2C in
perfectly still air. That's about 29 degrees C higher than the
non-concentrated case. .

The ambient temperature in the desert can get as high as 49 degrees C (120
F), so I assume the PV panels were designed to be reliable at this
temperature. Luckily, I don't live in the desert and my maximum ambient
temperature is about 38 degrees C. This is still about 18 degrees C hotter
than the panels were probably designed to operate. Is that a problem for
the new EVA?

How about this one. See:
http://www.redrok.com/electron.htm#poulek
http://www.solar-trackers.com/concentrator2.htm

Are any of these designs still under patent protection or have they fallen
into the public domain?

And some others.


I have asked many PV manufactures about their warrantees.
They don't actually say that concentrating is prohibited.
They do say that over heating the panel is grounds for
violating the warrantee.

Since temperature is the problem any way to help cool
the panels is a plus. Heat sinks and water cooling has
been effective.

I have been proposing the use of heliostats, yes real
heliostats, that reflect light onto stationary PV panels
mounted vertically and facing north. 1 to 4 heliostats
reflect light onto the panels. A temperature sensor
along with the heliostat controller adjust the number
of heliostats aiming at the panels.

Do you expect difficulty keeping all of the PV panels evenly concentrated
with the heliostats because the reflected shape will change so much
througout the day? In other words, a heliostat with a square mirror will
reflect a stretched parallelogram during early and late hours, right? Can
you keep the entire PV area uniformly covered? Is this an issue or not a
big deal?

Of course, cold weather allows more concentration
without exceeding the maximum temperature.

Then Minnesota winters should allow higher concentration .

Thanks for the reply!
John Symons

Duane

--
Home of the $35 Solar Tracker Receiver
http://www.redrok.com/led3xassm.htm [*]
Powered by \ \ \ //|
Thermonuclear Solar Energy from the Sun / |
Energy (the SUN) \ \ \ / / |
Red Rock Energy \ \ / / |
Duane C. Johnson Designer \ \ / \ / |
1825 Florence St Heliostat,Control,& Mounts |
White Bear Lake, Minnesota === \ / \ |
USA 55110-3364 === \ |
(651)426-4766 use Courier New Font \ |
[email protected] (my email: address) \ |
http://www.redrok.com (Web site) ===

 

[Roland Mösl]

When there is a strong wind, the temperature difference

I've read about PV losing efficiency due to higher temperatures, but I had
no idea it was that big of a difference! Why aren't panels built to allow
some way of removing the excess heat if it robs efficiency to such a large
degree?

In the next days will a translation about a new invention be translated
to English.

It's an Austrian solar pioneer already 30 years in the solar business.

His invention is a photovoltaic with thin cupper pipes on the back side.

The heat of the photovoltaic is brought in deep drills
or earth collectors or direct to a heat pump when heat is required.

So the photovoltaic becomes cool and produces more electric power.
The underground with the earth collecotor or deep drill becomes
warmer increasing the efficiency of the heat pump.

Better yet, build some small capillaries into the structure of the modules
and run a water/gylcol mix through them to take all that heat away, which
can then be run through a heat exchanger to serve a useful purpose, such as
providing hot water for the home or to heat a swimming pool.

To have best efficiency, the water glycol mix is only heated to
25 to 30 degree Celsius. Enough for the swimming pool, not enough
for domestic warm water. But a heat pumpis very efficent to raise
from this to 50 degree Celsius

If you have
more heat than you can use, just run it through a radiator with lots of
surface area and a few fans.

Much more efficent and more silent is an earth collector,
as long as it's not so far south.

Or for a really cheap option, just use that
heat to boil off cold water into steam.

Target is to cool down the PV surface to 30 .. 40 degree Celsius.

 

[John Beardmore]

Have any flat panel, low-concentration systems been built recently? Is the
EVA

What's EVA ?


Cheers, J/.

 

[SJC]

Have any flat panel, low-concentration systems been built recently? Is the
EVA

What's EVA ?


Cheers, J/.
--
John Beardmore

If you do a search for EVA+PV+solar, one of the many links you will find is:

http://www.eere.energy.gov/solar/fp_modules.html

Where it says: "Ethyl vinyl acetate-or EVA" - thin transparent acetate sheet...see through encapsulant
Encapsulant-An encapsulant helps to hold together the top surface, PV cells, and rear surface of the PV module. The encapsulant must
be stable at high temperatures and high levels of ultraviolet radiation. It must also be optically transparent and have a low
thermal resistance. Ethyl vinyl acetate-or EVA-is the most commonly used encapsulant. Thin sheets of EVA are inserted between the
solar cells and the top and rear surfaces. Heating this "sandwich" causes the EVA to polymerize, thus bonding the module into one
piece.

 

[John Beardmore]

Where it says: "Ethyl vinyl acetate...

Thanks !


Cheers, J/.

 

[Mike Reed]

John,

Here are some links to more information beyond the threads already posted.

3X mirror modules:
http://jxcrystals.com/SolarPV.htm

Scientist suggests large-scale CPV program in US:
http://www.photon-magazine.com/media/media_2005-09_ww_med_feat_Fraas.htm

Regards,

Mike Reed