solar cell power output

> That it takes more power.
>
> -Carl

Carl,

I've been researching it and I believe that it is still
true. Had to make some guestimates from the info I
found but am pretty sure it takes more power to make a
solar cell than the cell produces in its lifetime. Of
course solar cells are much more cost effective and
environmentally sensitive than batteries for low-power
applications.

In any case, I today sent a letter to the DOE asking
their consultants to provide me with the latest data
and I will forward their response to the list. Here is
what I wrote to the DOE experts:

--

Hi!

I have three questions. First, I would like to know
whether or not solar cells actually produces a net
surplus of energy during their 20-25 yr lifetime. In
other words, is the amount of energy generated by the
cell more than the amount of energy it takes for the
entire manufacturing process, from mining rare-earth
elements to doping the substrate, lighting the
manufacturing facility, baking off the impurities,
employees commuting to work at the factory, creating
the batteries needed, manufacturing the power inverter,
power used to deliver and install -- the total cost.
The general idea here is I am wondering if it is
technically feasible to *replace* power generators with
solar, as opposed to using solar as a efficient
battery.

Second, is there any information available estimating
the amount of lead-acid batteries that would be needed
to take a normal house off the grid for those 20 years.
In otherwords, if everyone in the US kept using as much
energy as they do now, but generated it from solar --
would the amount of batteries be substantial and would
it be a problem to safely and cost-effectively dispose
of them?

Third, what is the amount of pollution and hazardous
waste (if any) created by typical current solar
semiconductor manufacturing facilities.

Thank you very much,

Jeff Scott

> I've been researching it and I believe that it is still
> true. Had to make some guestimates from the info I
> found but am pretty sure it takes more power to make a
> solar cell than the cell produces in its lifetime.

Any URLs or other references?

> Of course solar cells are much more cost effective and
> environmentally sensitive than batteries for low-power
> applications.

Batteries take a lot of energy to produce too, have a
shorter life span, and don't have any way of recharging
themselves from otherwise wasted energy. I don't think
you can compare a solar cell to a conventional battery
like this, Jeff, but I could be wrong.

What you can compare to a battery is this zinc oxide
"fuel cell": http://www.metallicpower.com/

-Carl

> Any URLs or other references?

Sure - I searched google on 'solar cell efficiency',
and then 'solar cell manufacturing cost' and started
reading.

>> Of course solar cells are much more cost effective and
>> environmentally sensitive than batteries for low-power
>> applications.
>
> Batteries take a lot of energy to produce too, have a
> shorter life span, and don't have any way of recharging
> themselves from otherwise wasted energy. I don't think
> you can compare a solar cell to a conventional battery
> like this, Jeff, but I could be wrong.

I don't see the problem. I am saying that solar cells
are better than batteries in that batteries are more
environmentally destructive and consume vastly much
more power to make than they give off in their
lifetime, wheras solar appears to consume a little more
than it gives off in its lifetime (currently). Thus
solar makes great sense as a replacement for batteries
for low power applications where batteries would be
used like electric fence changers and calculators and
such. But as a replacement for fossil fuels, I don't
think it makes sense since you burn more fossil fuels
making a solar cell than you get out of a solar cell
during its lifetime. It's a geed bet that that may
change eventually since solar technology is making
advances -- not giant leaps, but advances.

- Jeff

>I don't see the problem. I am saying that solar cells
>are better than batteries in that batteries are more
>environmentally destructive and consume vastly much
>more power to make than they give off in their
>lifetime, wheras solar appears to consume a little more
>than it gives off in its lifetime (currently). Thus
>solar makes great sense as a replacement for batteries
>for low power applications where batteries would be
>used like electric fence changers and calculators and
>such.

That's like saying that fossil fuels make a great
replacement for batteries because they're more
efficient at storing energy.

>But as a replacement for fossil fuels, I don't think it
>makes sense since you burn more fossil fuels making a
>solar cell than you get out of a solar cell during its
>lifetime. It's a geed bet that that may change eventually
>since solar technology is making advances -- not giant
>leaps, but advances.

The Siemens SP150 costs $870 at realgoods.com. I pay
over 12 cents per kWhr for fossil-fuel juice. Let's
assume the entire cost of the Siemens unit goes to buying
the electricity to make it. Let's assume Siemens pays
what I do for power when they go to manufacture something.
Then the SP150 needs to make more than 7250 kWhr in its
lifetime for me to be right.

Siemens rates the SP150 at 150W (I believe, during "peak"
hours), and guarantees it will come within 80% of this
for 25 years. Let's assume it only ever makes 80% -- 120W.
Let's say I can expect in Berkeley only 2 hours per day of
peak sun. That means in one month the SP150 can pay for
itself, if I'm doing this right.

-Carl

>>I don't see the problem. I am saying that solar cells
>>are better than batteries in that batteries are more
>>environmentally destructive and consume vastly much
>>more power to make than they give off in their
>>lifetime, wheras solar appears to consume a little more
>>than it gives off in its lifetime (currently). Thus
>>solar makes great sense as a replacement for batteries
>>for low power applications where batteries would be
>>used like electric fence changers and calculators and
>>such.
>
> That's like saying that fossil fuels make a great
> replacement for batteries because they're more
> efficient at storing energy.

Well I don't agree. I think you have to look at the
whole picture and if a device produces less power in
its lifetime than the power used to produce it,
regardless of its nature, it is in effect a battery and
not a generator. But I think we know where we each
other stands here and can move on from this point.

> The Siemens SP150 costs $870 at realgoods.com. I pay
> over 12 cents per kWhr for fossil-fuel juice. Let's
> assume the entire cost of the Siemens unit goes to buying
> the electricity to make it. Let's assume Siemens pays
> what I do for power when they go to manufacture something.
> Then the SP150 needs to make more than 7250 kWhr in its
> lifetime for me to be right.

OK, great - this is similar to the sort of analysis I
am having to do to reconstruct the claims I once read.

> Siemens rates the SP150 at 150W (I believe, during "peak"
> hours), and guarantees it will come within 80% of this
> for 25 years. Let's assume it only ever makes 80% -- 120W.
> Let's say I can expect in Berkeley only 2 hours per day of
> peak sun. That means in one month the SP150 can pay for
> itself, if I'm doing this right.

OK, using these numbers you're off by a factor of 1000.
(kW vs W). (Sanity check here: obviously very few
people pay $870/month for electricity & most people are
also using a lot more than 150W of electricity at any
given time since that is only one 150W lightbulb and
1/3 of a typical 450W computer.)

In 1000 months, it will pay off using these figures.
That's 83 years, which is more than 3 times as long as
the cells are expected to last.

Yes, the actual rated wattage of a cell is near the
optimal output - I have some figures on that I found on
the web which you might find useful.

A 50W panel costs about $200. For the panel only it
seems and the total cost of the setup seems to run 3x
(with the instalation and the inverter and the
batteries) that of the cells themselves, but lets say
$200 is the total cost, which is actualy cheaper than
the Siemans model you mentioned.

A standard 50W panel produces 250W during the summer
under ideal circumstances (noon, no clouds or shadows,
module nearly new). Note here that your Siemens 80%
claim might possibly mean that after 25 years it it
producing 0.8 * 50W = 40W *peak* which is actually 16%
of its original *peak* production (which should be 5
times the rating). Cells do degrade quite a bit over
time using the methods that have been devised for
creating them and the harsh conditions to which they
are exposed in use.

The same standard 50W panel will produce 100W peak
during the winter. So lets say that over the year a
panel will produce 175W peak during the day. Maybe that
then amounts to an average of 50W during an entire 24
hours period - I don't know, but we should find out.
Also we need to find out what the non-peak production
is. Does it produce 1/10 the power if its cloudy? or
1/2? Don't know. Anyway, as I have said I think that
panels are *close* to producing more than they take to
make but not quite there, so I expect the final numbers
will be close.

Next round?

- Jeff

>>That's like saying that fossil fuels make a great
>>replacement for batteries because they're more
>>efficient at storing energy.
>
>Well I don't agree. I think you have to look at the
>whole picture and if a device produces less power in
>its lifetime than the power used to produce it,
>regardless of its nature, it is in effect a battery and
>not a generator. But I think we know where we each
>other stands here and can move on from this point.

Every device produces less usable power in its lifetime
than the power used to produce and run it.

>>Siemens rates the SP150 at 150W (I believe, during "peak"
>>hours), and guarantees it will come within 80% of this
>>for 25 years. Let's assume it only ever makes 80% -- 120W.
>>Let's say I can expect in Berkeley only 2 hours per day of
>>peak sun. That means in one month the SP150 can pay for
>>itself, if I'm doing this right.
>
>OK, using these numbers you're off by a factor of 1000.
>(kW vs W). (Sanity check here: obviously very few
>people pay $870/month for electricity & most people are
>also using a lot more than 150W of electricity at any
>given time since that is only one 150W lightbulb and
>1/3 of a typical 450W computer.)
>
>In 1000 months, it will pay off using these figures.
>That's 83 years, which is more than 3 times as long as
>the cells are expected to last.

Oops; silly factors of 1000.

I can make it pay for itself if we give realgoods the
normal retail markup of 100%, and assume I get 3 hours
of peak sun a day (which is quite reasonable, from what
I've read). And that's still assuming Siemens uses
electric heat to cook the thing, pays retail for
electricity, and has no other expenses in making it.

>A standard 50W panel produces 250W during the summer
>under ideal circumstances (noon, no clouds or shadows,
>module nearly new).

Oh, really?

>of its original *peak* production (which should be 5
>times the rating).

Oh, I've been using the rating as the peak production.
If I get three hours of peak sun a day at 600Wp/hr
(.8 * 150 * 5), and assume once again that realgoods
makes no money on the transaction, in pays off in 11
years.

>Cells do degrade quite a bit over time using the methods
>that have been devised for creating them and the harsh
>conditions to which they are exposed in use.

They guarantee 80% of rated output for 25 years. I'm
assuming it never makes more than that from day 1. And
I'm assuming it's pitch black except for my 3 hours of
peak sun every day.

-Carl