Maker Pro
Maker Pro

PV cell spectral sensitivity?

M

Mark Fergerson

Jan 1, 1970
0
I was sent here from sci.electronics design, so if I'm
asking a question that's in a FAQ someplace, please point me
there.

I've been trying to find out what part of the spectrum
your common silicon cells scavenged from calculators best
convert to current, but the best I've been able to find
after Googling myself silly is the fact that the bandgap for
silicon puts their sensitivity peak in the near-infrared.

Some claim that such cells are antireflection-coated
which would explain the different colors they come in, but
others claim that's just a moisture barrier and the most
efficient ones are dark blue, which makes sense if they like
NIR best.

Why do I care? I want to hide some cells behind fake
gemstones for a project. If the cells do prefer NIR, it's
going to make it interesting to find fake stones that will
pass enough NIR to be worthwhile. Not impossible, it just
means I get to do a lot of testing.

Is there anyone/a site that can set me straight, or am I
doomed to the tedium of finding out for myself?

TIA

Mark L. Fergerson
 
G

George Ghio

Jan 1, 1970
0
Mark Fergerson said:
I was sent here from sci.electronics design, so if I'm
asking a question that's in a FAQ someplace, please point me
there.

I've been trying to find out what part of the spectrum
your common silicon cells scavenged from calculators best
convert to current, but the best I've been able to find
after Googling myself silly is the fact that the bandgap for
silicon puts their sensitivity peak in the near-infrared.

Some claim that such cells are antireflection-coated
which would explain the different colors they come in, but
others claim that's just a moisture barrier and the most
efficient ones are dark blue, which makes sense if they like
NIR best.

Why do I care? I want to hide some cells behind fake
gemstones for a project. If the cells do prefer NIR, it's
going to make it interesting to find fake stones that will
pass enough NIR to be worthwhile. Not impossible, it just
means I get to do a lot of testing.

Is there anyone/a site that can set me straight, or am I
doomed to the tedium of finding out for myself?

TIA

Mark L. Fergerson

Would you like a chart (graph) of this? PDF.

George
 
M

Mark Fergerson

Jan 1, 1970
0
George said:
Would you like a chart (graph) of this? PDF.

Actual numbers and everything? Yes, please! Your document
may not exactly apply to my cells, but a starting place will
be a huge help.

Mark L. Fergerson
 
M

Mark Fergerson

Jan 1, 1970
0
asking a question that's in a FAQ someplace, please point me
there.your common silicon cells scavenged from calculators best
convert to current, but the best I've been able to find
after Googling myself silly is the fact that the bandgap for
silicon puts their sensitivity peak in the near-infrared.
That depends on the crystalline structure of the silicon the solar cells
are made from. If you're harvesting them from calculators, they are
almost certainly amorphous silicon (as opposed to
crystalline silicon).

That's the reverse of my original impression. However,
upon close inspection they don't have the same obvious
crystalline appearance as do old-fashioned large-grain
polycrystalline cells I have on hand. I was unsure whether
modern cells might be single crystals that wouldn't show the
obvious grain structure the old ones do.
They might also not be silicon at all, but rather cadmium telluride.

Dang.
which would explain the different colors they come in, but
others claim that's just a moisture barrier and the most
efficient ones are dark blue, which makes sense if they like
NIR best.
How many different colors are you seeing? Amorphous silicon cells should
look dark blue or black -- and yes, antireflection coating is involved
-- while cadmium telluride cells typically look dark red or
rust-colored.

Black, nearly violet, dark blue, leaf green, a strange
greenish-brown, and the rust/dark red you say is CdTe.

Oh, dear. This is getting more and more complicated.
As for your primary question, the spectral response of amorphous silicon
solar cells typically peaks near the high end of the visible range, at
600 nm or so. Crystalline silicon cells peak deep into the near IR,
around 950 nm, which is what I suspect you found in your
googling.

Yes, the few numbers I could find were for crystalline
Si. Can you point me to a comprehensive data source on the web?
Cadmium telluride cells should have a peak roughly similar to that of
amorphous silicon.

That's a relief. I'm getting similar voltage/current
numbers from "unhidden" cells of various colors, so I'll
stop worrying about their chemical composition.
gemstones for a project. If the cells do prefer NIR, it's
going to make it interesting to find fake stones that will
pass enough NIR to be worthwhile. Not impossible, it just
means I get to do a lot of testing.
As long as you're using calculator cells, you should be fine with stones
that are relatively transparent in the visible region.

One thing you should be aware of is that the wavelength of peak response
does not necessarily mean that you will get a disproportionate amount of
your power from that wavelength. For example, sunlight's peak intensity
is around 550 nm and even for crystalline silicon cells -- the ones
whose sensitivity is in the near IR -- most of the power produced by
these cells comes from the visible regime. In other words, don't block
short wavelengths on the theory that the cells aren't that efficient
there anyway -- if that's where most of your incoming photons are, then
that's where you're going to get the most power.

Right; I'm aware that the sensitivity of the cells are
curves as opposed to narrow peaks. I'd about settled on
amethyst because I have plenty and it's relatively easy to
work, but it does absorb green strongly (I'm interested in
running some low-power circuits from sunlight and
incandescent room lighting) and there'll be a lot of light
going to waste if I do that.

The idea is to hide the fact that the cells are there
while allowing them to do their job. I may just go with
straight quartz and rough it up a bit on the backside into a
diffuser.
In short, I would suggest at least a little testing under the types of
light sources you expect your project to be illuminated by (which may
not be like sunlight at all). You will quickly get a feel for which
stones work best, and it's also the only way to account for strange
effects that might result from the refractive index of the gemstone,
faceting, etc.

I expected to do quite a bit of testing, but I wanted to
have a starting point to avoid time wasted slabbing the
wrong materials.

Thanks for your help.

Mark L. Fergerson
 
G

George Ghio

Jan 1, 1970
0
Hi Mark

Sorry for the break but the server has been down.

If you would like a graph of spectral sensitivity Where would you like
it sent.

George
 
M

Mark Fergerson

Jan 1, 1970
0
George said:
Hi Mark

Sorry for the break but the server has been down.

Stuff happens.
If you would like a graph of spectral sensitivity Where would you like
it sent.

mfergerson(one)(at)cox.net

Decode as appropriate. Spambot harvesters are getting
smarter. :>(

Thanks again

Mark L. Fergerson
 
J

John Beardmore

Jan 1, 1970
0
Mark Fergerson said:
George Ghio wrote:

Stuff happens.


mfergerson(one)(at)cox.net

Decode as appropriate. Spambot harvesters are getting smarter. :>(

If possible, could you also send a copy to wookie(at)wookie.demon.co.uk
please ?


Thanks, J/.
 
G

George Ghio

Jan 1, 1970
0
John Beardmore said:
If possible, could you also send a copy to wookie(at)wookie.demon.co.uk
please ?


Thanks, J/.

Not a problem

George
 
Top