Solar panel/mirror question

How much energy can mirros focus onto a solar panel before it begins to
melt? It's an SSPS based question. Could large light-weight mirrors and
small solar arrays replace the traditional massive structures of the NASA
1980s design study? This is something I haven't found in all my googling.

Anyone here know?

John

In article ,
ANTIcarrot wrote:
How much energy can mirros focus onto a solar panel before it begins to
melt?

A more fundamental problem than melting is that solar-cell efficiency
drops off as the cells get hot, although this is less important for
advanced cell materials than it was for silicon.

Modest amounts of concentration are workable -- Deep Space 1 used lens
concentrators -- and my (dim) recollection is that concentration ratios of
up to 100:1 have been explored in studies. Still, there are limits; even
the best cells die at hundreds, not thousands, of degrees.

...Could large light-weight mirrors and
small solar arrays replace the traditional massive structures of the NASA
1980s design study?

The tradeoffs are complex, particularly if you are thinking about using
extraterrestrial materials (in which case you can forget about the exotic
advanced cell materials -- you're pretty much restricted to silicon, which
is not very heat-tolerant).

Even at 100:1, though, the actual arrays are in not "small" in any
realistic sense of the word.
--
"Think outside the box -- the box isn't our friend." | Henry Spencer
-- George Herbert |

Henry Spencer wrote:

The tradeoffs are complex, particularly if you are thinking about using
extraterrestrial materials (in which case you can forget about the exotic
advanced cell materials -- you're pretty much restricted to silicon, which
is not very heat-tolerant).

That's not entirely clear. Advanced materials can also be direct bandgap semiconductors,
which absorb light over a much shorter distance than does crystalline silicon.
So, you don't need much actual semiconductor. What you do need is mundane stuff
like something on which to deposit that material, perhaps a protective glass
cover, wires, heat sinks, and so on. You could make all that from ET materials
and just ship up the unobtainium from Earth.

Paul

Henry Spencer wrote:
In article ,
ANTIcarrot wrote:
How much energy can mirros focus onto a solar panel before it begins to
melt?

A more fundamental problem than melting is that solar-cell efficiency
drops off as the cells get hot, although this is less important for
advanced cell materials than it was for silicon.

Standard solar cells are some 0.3mm thick.
To create a 1C temperature differential over this (silicon has a thermal
conductivity of 150W/K/m) will take some 500KW/m^2.
I think that it will take around 20C to get significant strains in the
solar cell between the front and the back, so a figure of some 10MW/m^2
might not be unreasonable.

However...
Silicon cells are not just silicon, but require a network of metallisation
over it to carry the current.

But this isn't the primary problem.
The primary problem is what do you do with the heat?
If you paint the back of the solar cells black (in IR) then point them
at the sun, in deep space, they will heat to around 0C (plus or minus
20C).
But, if you concentrate the sunlight to say 4 times, then the temperature
rises to some 100C, at 10 times some 324C.

At some point, you'r cells are going to, if not melt, degrade efficiancy
significantly.

To avoid this, you have to actively cool the cells.
But then you run into the problem that the area you need for the thermal
radiators is pretty closely the area of the solar cells, unless you
use active heat-pumps.

You may be able to do quite a little bit better if you filter the light
at the mirror, and reduce the IR/blue, as these heat the cell without
producing (proportionally) as much electricity.

But then, it's likely that the filter will not quite be free in some
way.

Paul F. Dietz wrote:
Henry Spencer wrote:

The tradeoffs are complex, particularly if you are thinking about using
extraterrestrial materials (in which case you can forget about the exotic
advanced cell materials -- you're pretty much restricted to silicon, which
is not very heat-tolerant).

That's not entirely clear. Advanced materials can also be direct bandgap semiconductors,
which absorb light over a much shorter distance than does crystalline silicon.
So, you don't need much actual semiconductor. What you do need is mundane stuff
like something on which to deposit that material, perhaps a protective glass
cover, wires, heat sinks, and so on. You could make all that from ET materials

Also, the actual amount of silicon/... needed may be rather small.
Standard solar cells are some .3mm thick, and get 18% or so efficiancy.
Assuming that these can be made as cells which can cope with 4* concentration.
This is some 800W/m^2.
The weight is 600g/m^2, so this is 1333W/Kg.
Say $5000/Kg launch costs, so you'r talking about somewhere around $3.75/W.

Coincidentally, this is about the price (of unconcentrated) solar cells
on earth.
$7.50 will buy you around 100Kwh on earth.

If the off-earth resources are free, and the cells last 10 years,
it breaks even with current earthly prices.

A more fundamental problem than melting is that solar-cell efficiency
drops off as the cells get hot, although this is less important for
advanced cell materials than it was for silicon.
So it's a no-win situation, a solar focussing mirror gives the array more
power output with attendant heat gain which needs a radiator to space, which
adds nearly the weight which a larger array would weigh in the first place
:-).
Solar energy is mainly heat energy. [80%?] Heat is what we need to better
utilize for outer space power, not so much the visible part of solar energy. Hi
tech steam engines?
^
//^\\
~~~ near space elevator ~~~~
~~~members.aol.com/beanstalkr/~~~

"Allen Meece" wrote in message
...

Solar energy is mainly heat energy. [80%?] Heat is what we need to
better
utilize for outer space power, not so much the visible part of solar
energy. Hi
tech steam engines?

Some others have been mentioning use of space materials. I've heard it said
that the silicon-cells approach is a light-weight approach driven by launch
costs, and that if space materials are being used, solar thermal might be
more the way to go. I guess the maintenance costs might go up, but the
greater efficiency might pay for it.

--


Regards,
Mike Combs
----------------------------------------------------------------------
We should ask, critically and with appeal to the numbers, whether the
best site for a growing advancing industrial society is Earth, the
Moon, Mars, some other planet, or somewhere else entirely.
Surprisingly, the answer will be inescapable - the best site is
"somewhere else entirely."

Gerard O'Neill - "The High Frontier"

ANTIcarrot wrote:
How much energy can mirros focus onto a solar panel before it begins
to melt?

Henry Spencer replied:
A more fundamental problem than melting is that solar-cell efficiency
drops off as the cells get hot, although this is less important for
advanced cell materials than it was for silicon.

Henry has it right here (footnote 1). The answer depends significantly
on the technology. Silicon cells lose about 0.45% of their power per
degree C; GaAs cells about 0.35, and advanced cells (e.g., triple
junction) something like 0.25% per degree C. As you concentrate light
on them, you get them hot, and they work less well.

You can mitigate this *some* by clever thermal design, reflecting or
transmitting the part of the radiation that you can't use, and making
cells that radiate well in the infrared. But that only gets you a
little bit more concentration, maybe as much as a factor of two, but not
more.

A while back, when I was looking at laser power beaming, I calculated
that you reach the maximum output for GaAs cells at about ten suns input--
above that, you actually get *less* power out when you put more light on
the cells. (And at that ten sun level, they were performing at about
half their one-sun efficiency). [see footnote 2]

Modest amounts of concentration are workable -- Deep Space 1 used lens
concentrators -- and my (dim) recollection is that concentration
ratios of up to 100:1 have been explored in studies.

Ah, that depends on how you do it.

If you have huge lenses focussing on solar arrays, the cells get hot.
DS-1, however, had small lenses focussing on quite small cells (
significantly smaller than the standard space cell size). The cells
were mounted on heat spreaders, so the radiating area was much larger
than the cells. In terms of thermal performance, the input power
divided by radiating area was about one sun, so the performance didn't
suffer.


Still, there are
limits; even the best cells die at hundreds, not thousands, of degrees.

[ANTIcarrot continued]
...Could large light-weight mirrors and small solar arrays replace the
traditional
massive structures of the NASA 1980s design study [of solar power
satellites]?

It's been proposed in many different different variants. The recent
proposal of 15-meter Fresnel lenses and small solar arrays (something
seen in a lot of recent artist's conceptions) turns out not to be a
really a workable idea

.unless, perhaps, you have a program to develop high-temmperature
solar cells ... like the research program at NASA Glenn

The tradeoffs are complex, particularly if you are thinking about
using extraterrestrial materials (in which case you can forget about
the exotic advanced cell materials -- you're pretty much restricted to
silicon, which is not very heat-tolerant).

Right. You might be able to gain a bit, but not an order of magnitude.

Even at 100:1, though, the actual arrays are in not "small" in any
realistic sense of the word.

---------------
Footnote 1:
As usual.

Footnote 2:
The calculations for power output as a function of incident
concentration for sunlight were never really published, -- I ought to
write them up, but it's not way high on my list of things to do. The
ones for laser light were in an old issue of AIAA Journal of Power and
Propulsion, I think, and may have been in an old SPRAT proceedings as
well. Figure 2 of this publication:
http://powerweb.grc.nasa.gov/pvsee/p...AA92_3213.html
shows a summary graph. Note efficiency of GaAs cells under laser light
is roughly double the efficiency under sunlight, so the power output
moves down by a factor of two, although the location of the peak at
about 13000 W/m2 incident sunlight stays about the same.

--
Geoffrey A. Landis
Photovoltaic scientist, Mars guy, and part-time SF writer
http://www.sff.net/people/geoffrey.landis

Geoffrey A. Landis wrote:
The answer depends significantly on the technology. Silicon cells
lose about 0.45% of their power per degree C; GaAs cells about 0.35,
and advanced cells (e.g., triple junction) something like 0.25% per
degree C. As you concentrate light on them, you get them hot, and
they work less well.

Why not use *vaccum* cells? Light hits the cathode, and knocks
electrons out of it with enough momentum to reach the anode against a
moderate voltage gradient. Should work at any temperature, so long as
the components aren't white hot.

Perhaps the best "material" for space manufacture is the stuff there's
no shortage of in space -- vacuum.
--
Keith F. Lynch - http://keithlynch.net/
Please see http://keithlynch.net/email.html before emailing me.

In article ,
Keith F. Lynch wrote:
Why not use *vaccum* cells? Light hits the cathode, and knocks
electrons out of it with enough momentum to reach the anode against a
moderate voltage gradient. Should work at any temperature, so long as
the components aren't white hot.

According to a rather old reference... The working temperature is quite
limited because the best cathode and anode materials are complex alloys
involving alkali metals, which melt or decompose at modest temperatures.
Moreover, under good conditions the attainable efficiency is about 2%,
with an increase by perhaps a factor of two thought conceivable -- but by
no means certain -- with extensive development.
--
"Think outside the box -- the box isn't our friend." | Henry Spencer
-- George Herbert |