At 05:49 PM 8/29/2008, you wrote:
>FYI,
>
>"NREL Solar Cell Sets World Efficiency Record at 40.8 Percent"
>National Renewable Energy Laboratory
>http://www.nrel.gov/news/press/2008/625.html
>
>: Scientists at the U.S. Department of Energy's National Renewable
>: Energy Laboratory (NREL) have set a world record in solar cell
>: efficiency with a photovoltaic device that converts 40.8 percent of
>: the light that hits it into electricity. This is the highest
>: confirmed efficiency of any photovoltaic device to date.
>
>: The inverted metamorphic triple-junction solar cell was designed,
>: fabricated and independently measured at NREL. The 40.8 percent
>: efficiency was measured under concentrated light of 326 suns. One
>: sun is about the amount of light that typically hits Earth on a
>: sunny day.

This is interesting. Let's take a look at some numbers.

Consider peak sunlight (on the ground) as a kW/m exp 2. So the
output of a sq meter would be around 130 kW. I am not sure how to
account for the reflected light from the cell surface. Ignoring
that, then 60% of 326 kW/m exp 2 will go into heating the
cell. That's about 195 kW/m exp 2.

195,000/0.9 = 5.67 x 10 exp -8 T exp 4, T would be almost 1400 deg K
or over 1100 deg C. Solar cells don't operate that hot so it would
have to be cooled.

Installed in a power sat and kept well below 100 deg C, it would use
almost as much radiator as a steam turbine system.

Fewer moving parts though.

10 GW would require about 77,000 square meters of cell and 23 million
square meters of reflectors or a square close to 5 km on an edge.

15 GW of waste heat at room temperature would need a radiator of
about the same size, a 5 x 6 km radiator.

Hmm

Keith Henson