[PLUG-TALK] solar energy

Keith Lofstrom keithl at kl-ic.com
Fri Dec 8 20:53:51 PST 2006

On 12/6/06, Edward Terry <eterry at openboxbuilder.com> wrote:
>Speaking of solar energy, the DOE has announced that Boeing-Spectrolab
>has created a solar cell with 40.7% efficiency.

On Wed, Dec 06, 2006 at 11:53:04PM -0800, Bill Barry wrote:
> That's amazing!
> Here's a couple of papers  I found following up on this
> http://www.boeing.com/news/releases/2006/q4/061206b_nr.html
> http://www.nrel.gov/docs/fy05osti/38689.pdf
> These seem to be 3 junction solar cells with different bandgaps in each
> region for absorbing different parts of the spectrum. Pretty impressive.

Stacked cells is inefficient.  A friend of mine, William Mook of Ohio,
came up with a better way to do this.  He uses three different cells
with different bandgaps, as above, but rather than stack them he
puts light on them with dichroic Fresnel mirrors, which focus the
red and green and blue light on the different cells.  By doing this,
he gets even higher efficiencies.  Given the high efficiency, there
is much less heat loss, which means he can focus more light on each
cell.  The dichroic mirrors are not too difficult, he can stamp them
out of flat sheets of plastic, sort of like the surface of a CD but with
less resolution and more error tolerance. He can stamp out a 4 foot
mirror with the requisite accuracy, shining full daylight into three
6 inch cells with passive fluid cooling.

Yes, the mirrors do have to be aimed, and rigid against wind, but this
can be done with ganged mechanics that moves very slowly.   A really
cool idea for high efficiency solar farms, and possibly practical for
building rooftops (though logistics gets inefficient at that level).
Boeing is no doubt developing for their cells for satellite power,
and these would be fine for that.


Keith Lofstrom          keithl at keithl.com         Voice (503)-520-1993
KLIC --- Keith Lofstrom Integrated Circuits --- "Your Ideas in Silicon"
Design Contracting in Bipolar and CMOS - Analog, Digital, and Scan ICs

More information about the PLUG-talk mailing list