This is not "How to make PV Solar Cells".
It is possible to home-make Copper Oxide and other kinds of materials
but that is a whole nother story which I may do in the future.

I may be a little bit ambitious to try to show you how I made PV
Solarpanels out of various types of cells I collected and how and
where I obtained them rather inexpensively, and some of the differences
in the various kinds, but most of all, how to work with them to get
free electricity under the light of the sun and other sources of light.

In essence, this involves ways to connect cells, which may produce
more or less than one volt, and not only try to increase power output
but also decrease the load, that is, efficiently conserve the energy
whether it is meager or significant.
For example, even the weakest solar panels can run watches,
calculators, radios, charge batteries, and if a computer were
specifically designed to, it would be as solar-powerable as a calculator.

Here are some pictures of Solar Panels which I have constructed.

Step 1: Supplies and Sources

What you may be able to use to build a useful solar panel:

"Broken" solar cells. They are very cheap and they work, they are
just randomly shaped. They are usually crystalline silicon ones,
which ALWAYS (ha!) look broken even when they are not.

Surplus solar cells. Amorphous silicon printed on glass (check) are
excellent, usually producing more than a volt, and much sturdier
than the thin ones that break in bulk quantities. If these break, we
can fix them, usually.

Indium Copper Selenide Cells. These are "new" and are conveniently
sold as glass tiles with easy to solder tabs.

Any of the above, sold as cells prepared for assembly into panels;
in other words, complete and solder - ready or with wires and tabs.
(I will explain how to prepare inferior quality cells in this instructable.)

Miscellaneous items:
Wire Glue - There is already another instructable for using wire glue
on Broken solar cells. (link)

Brass extrusions, bracket |_| shaped - Convenient for connecting to glass cells.

Solder
Soldering Iron - low wattage
Small flat-head screwdriver
Thin (around 20 AWG or less) stranded copper wire

Lamp cord or Speaker Wire
Alligator clips

Deep Picture Frames or Shadow Boxes (Enclosure)
-look for imported frames at the El Cheapo store and pray a machine made them

Acrylic/Lexan/Plexiglas/Etc clear polymer sheets
Router or Dremel to cut out the middle of one out of three sheets
RTV (Silicone Glue) - or :
High Temperature Hot Melt Glue (Caution-you don't want the sun to melt it!)

Rectifier Diode such as 1N4001 or 1N4004

Voltage doubler or multiplier circuits (you can make) to increase voltage output.
-examples: ICL7660, MAX1044, MAX232, etc.

Wide Sticky Tape
Double Sticky Foam Tape

Rechargeable Nickel Batteries
Gel Cells or Car Battery (you got one, might as well use it until it's useless)
-Li Ion not recommended because they are harder to charge

Analog volt meter (only because it doesn't need batteries like a digital one)

AC Inverter - if you are charging a powerful battery and would occasionally
run some mains-powered appliance. Some UPS's can be easily modified to
be inverters, if they can be turned on after a power failure.

Sources:
Broken Solar Cells:
Herbach and Rademan
Silicon Solar
Electronic Goldmine

Glass (Amorphous) Solar Cells:
Electronic Goldmine
Note:Other links here may also supply Glass Solar Cells

Indium Copper Selenide Cells:
All Electronics
Edmund Scientific
Electronic Goldmine

Other sources:
Cheap weather damaged solar powered outdoor night lights
-(common failures are circuit corrosion and defective batteries, not the solar cells)
Defective solar calculators, solar charged flashlights, etc.

Perhaps a little off topic:
For a reasonably good deal on Complete and Useful Solar Panels I recommend
"Solar Car Battery Chargers" that are about 1 or 2 watts and between $20 and $30,
whenever an opportunity to get some arises. But those are what I am trying to show
how to Make an approximate equivalent of.

Step 2: How to Use "broken" Cells

They are the crystalline ones that Always look broken, but if they really are,
then they have not been fully prepared for use. It is an extra challenge to
solder wires onto them but this is how I do it:

Look for the wide line on the pieces, and sort out ones that only have
thin lines. The thin line ones might be useful with Wire Glue but are
too hard to solder.

Then sort the pieces with wide lines by how big they are. They will all be
about 0.55 volts but the larger pieces make more current than the smaller
pieces and it's nice to have a panel with consistent current, especially
the one you make with the biggest pieces.

Let's save the big pieces until we learn to do the small pieces.

Strip apart a short length of stranded wire and put the now loose strands
in a small box just so you can find them and so they don't wander into another
project and cause a short circuit.
ACTUALLY another option may be to use wire-wrap wire instead of bare strands,
if you don't mind stripping the end of each piece.

The broken cells have a very thin conductive layer on the blue side and
a very rough thicker one on the other. It will be more challenging to solder
onto them than on perfect cells but this is how.
First the blue side...

Step 3: Preparing Broken Cells

If you can solder onto the cells then they are higher quality than the ones I have
so you can skip these preparing steps:

On the blue side, scratch the thick line with a very small flat screwdriver with just a little force
so that the cell doesn't break, and the line should turn from white to shiny unless it's already
shiny and ready to solder. Try to make a little shiny circle. We will solder there.
Make the flat edge of the screwdriver completely touch the scratched area so it rubs wide.
Mostly push back and forth so that the rubbing removes the thin oxidation.
After scratching the line, turn the cell and scratch the circle back and forth again.
Maybe turn it once more and scratch it once more.

Now flip the cell over and notice the rough stuff on the back. If there appears to be
two different roughnesses or shades of grey, we are going to scratch in two places.
Again, turn the cell and scratch it in one or two little circles by pushing the edge of the
screwdriver up and down to remove the coating that solder won't stick to.

Now back to the blue side. Try to get a solderball to stick.
If it does not stick, and rosin gunks up the area, scrape it off and try again,
and if it seems hopeless, scrape another part of the wide line on the cell.
I did not have the problem because of practice.

Now try to put a bump of solder in the two places scratched on the bottom of
the cell. I was only able to get one bump to stick. There are areas on the
bottom where solder just won't stick. But if neither spot sticks, try scraping the
rosin off the spots and soldering again, or carefully scratching another spot.
If you have a bump on the blue side, it's good but you can't lay the cell flat now.
The spot that worked was rougher and thicker than the one that didn't, and
that means there's a lot more silver there, and more likely it will solder.

Now that you have two solder bumps, you can attach two thin wires,
either strands from stranded wire, or thin wire-wrap wire.
What about thicker wire? It can pull the lines off the cell and then you can
forget about soldering it. Put it in the "wire glue" bin.

Now that there are two wires on the cell, test it with a meter. The Blue side
of the cell will make up to 0.55 Negative volts, so connect the meter PLUS
to the wire on the silver-gray bottom of the cell. My cell isn't getting much
light but the meter needle is indicating that it is making electricity.

Step 4: "Broken" or "crystalline" Cell Panels

In the last step I mentioned that the Blue side is Negative and
the silver side is Positive.

Now all you have to do is solder your cells in Series to get
more voltage. To do that you only need one more wire for
each additional cell you add.
Remember each cell makes up to half a volt, so consider
a 12 volt panel to have 24 or more cells. A few extra is good.
One reason for that is a diode lowers the voltage just a little bit,
and another is that it's nice to have 12 volts for charging batteries
when it's not the sunniest time of day. A diode is used when the
panel charges batteries, so the batteries don't give any power
back to the panel in the dark. That would be a waste of free power.

Because the cells are so fragile, it would be good to install them
in a deep picture frame (shadow box) with double stick foam tape
or RTV glue. Be careful, this is permanent. You could make it
less permanent with hot-melt glue also.

At this point you don't need to think that the cells are "already broken",
and you will have a well working panel. You could hide the shard-shapes
with fluorescent lighting diffraction plastic over the framed panel if you like.
Perhaps you've seen a shard-cell panel just like that being sold before.

Step 5: Preparing Glass (Amorphous) Cells

I received a surplus glass cell with instructions on how to use copper mesh
to make a connection to the glass cell. The glass cell was pre-scratched in
the area where the mesh and wires were supposed to go. But... even with
the copper mesh, it didn't stick. It was doable, but hard to do, and not very
strong. All the wires pulled off. Some of you may have had success with
using copper mesh soldered to scratched areas of glass cells, but there is an
easier way.

Perhaps you have a broken / damaged glass cell. You may still be able to use it,
unless the damage has made the glass transparent, in which case there is severe
damage to the photovoltaic part of the cell.

One interesting thing about the glass cells. Looking at them, you see lines, just
as you may on "broken" or "crystalline" cells, but those lines are not current-
collecting conductors. They are gaps between areas of the glass cell that each
make about half a volt. So, glass cells can be expected to have 2 lines for every
volt of output. And they can make 6 or 9 or 12 or 20 volts.

So, we want to connect the wires to places with the most amount of lines between them
to get the highest voltage. And out the wires on the silver side, of course.

Scratch the silver (probably aluminum) near the edges and test the voltage and polarity,
for your information. I usually use a red wire for Plus and a black or green for Negative.

Easy connection method:
You need two brass extrusions, carefully cut with a dremel (safety goggles!),
and wires soldered on this side of the extrusions ---> C
The extrusion must have enough space inside it for the glass cell to fit.
The extrusion is then crushed a little (before putting it on the glass)
so that it will bite the glass with some pressure and make contact with the scratched edge.

Slide the crushed extrusion onto the glass. If it's too crushed it won't go on, so pry
it open. If it's not crushed enough it falls off, so crush it more. When it bites, and
there is voltage in the light across the two extrusions, put stickytape or
just a little plastic cement over the extrusion to help it stay there.

The Glass cell is now ready to use.
The long one shown is actually two 9-volt ones on one glass, and is the one
that I put extruded contacts on because the copper mesh wouldn't stick..

Step 6: Preparing Copper Indium Selenide Cells.

These are rather well prepared already.
They have easy to solder tabs, and are marked which end is Negative with a
dash of a black marker.
The ones I got, I mounted in frames and in an acrylic polymer sheet sandwich.
Three in series ... in parallel with three more in series ... makes nice 12 volts.

I have been advised that these cells undergo some kind of reaction if first exposed to
full sun with no load for about 15 minutes, and that the result is good. I'm told that the
result generates more output than if they are not treated this way. Just FYI. I didn't
notice the difference between the panel that had pre-sunned cells and another that
didn't.

The cells are glass tiles that appear to be made similar to the Amorphous glass,
but they are more efficient, and produce around 4.5 volts and 100ma each in full sun,
approximately. As they say, your mileage may vary. I have no advice for broken
CIS cells.

It is very easy to connect CIS cells together. Peel back the tabs a little, which point to
each other under the cell, and start to peel back the stickytape that holds it on,
just enough so that you can solder them in series.
And watch the polarity! I goofed it up a couple of times.
No damage done, but I had to do it over.
When soldering, wet the ends of the tabs with solder, then press down quickly
with a popsickle stick or something to flatten them against the bottom of the cells.
The cells go together nicely like tiles.
With moderate carefulness, you don't need to worry much about ruining them yourself,
just don't leave them alone with curious people until your panel is done and safe inside
a solid frame. I've fastened them with both RTV Silicone and double-sticky-foam-tape.
I prefer the Silicone glued result, with the cell tiles grouted against the glass from behind.
(No silicone between the cells and the frame glass)
DSFT (foam tape) is more likely to (it has, in fact) let go of a couple of the cells.

As mentioned before, although I don't know if it's necessary for CIS cells, use a diode
when charging batteries with the panels.

Step 7: Applications for Small Solar Panels

The solar panels I made and pictured generate around 1 or 2 watts generally.

These are the applications I use them for:

Charging batteries.
In the blackout of 2003, those batteries ran our blackout party,
which included black lights, fans (it was a hot day), radio, small TV, and low voltage lights.
And an AC inverter.
(I go to the rechargeable battery recycle bins with a meter and if they are not really dead
then I borrow them until they are. I didn't buy any of these batteries.)

Solar night lights - nowadays a very common thing where I live.

Solar powered fans - although my solar panels run computer fans directly when it's hot,
(The sun makes it hot, and the sun runs the fans!)
I notice that solar charged battery powered fans are MUCH MORE POWERFUL.

Solar Flashlights

Solar powered radios - including my ham radio shack.

ABOUT SOLAR POWERED COMPUTERS
I guess people don't leave their laptops in the sun...
My approach to designing a solar powered computer,
(and my definition of computer is a processor with memory
and a keyboard and a screen that runs not-necessarily-an-operating-system)
is to use very high resistance CMOS chips which use very little electricity,
just like watches and calculators... a computer is also a calculator with lots
of memory, and CMOS memory is a common thing!
At night time, the computer has not used up all it's solar power so it uses
what is stored in the rechargeable battery.
There is simply no demand for the solar powered computers,
nor any obstacle to solar powering a PDA or a laptop with similarly sized panels.

DUTY CYCLES:
In simple theory, if you get eight hours of sun and need one hour of power,
you can get by with one eighth the solar power by saving it up in batteries.
Also, if LED lights should run all night, it's easy to collect more than enough
solar power during the day in batteries with the right sized panel.

Step 8: Getting More Practical Power From Your Panel

It is very easy to get a few solar cells and put them together into a panel,
but sometimes it gets expensive to get enough cells to make a useful voltage.

If you obtained one or two large cells, you may have a whole watt or two,
but only a volt or less, and that's sad. Not too many things run on less than
a volt.

Perhaps you got enough big broken cells to make 6 volts , but wouldn't it be
nice to have 12 volts? Then maybe you could keep a battery charged and
occasionally run an inverter on it.

In the last step I mentioned how time could be used to save up power for
another time when it will be used. And a small panel can make enough
power over a long time to run a big load for a short time.

In this step I am talking about matching the voltage of the panel, whatever
it may be, to the voltage that you find useful. Or generally, matching
supply and demand in a satisfying practical way.

It may be possible to design a 2 volt circuit for a 2 volt panel, but unnecessary.

It is possible, although as far as I know, using obsolete Germanium transistors,
to get any voltage out of a big half-volt cell, and I don't know a modern way,
so I'll leave that idea alone.

But there are many voltage doubler or multiplier circuits that work at slightly
higher voltages, and I see that I've made a few panels around 6 volts which
I'd like to get 12 out of. There is a voltage doubler chip still available called
ICL7660 or MAX1044 that is very convenient to use. So I will use it as an example,
since I'd rather have around a watt at 12 volts than at 6 volts.

There is something else I did that was very obvious in the picture for step 1,
where I had 3 "broken cell" panels around 6 volts and put them in series to get
around 18 volts... and since the cells were large that array has a lot of current.

But if I use just one 6 volt panel and want 12 volts, I use the voltage doubler
and get twice the voltage in exchange for half the current. AC transformers
do the same thing... almost the same power goes out as goes in, but at a
more useful voltage. Some circuits that do this are called "DC to DC converters".