Homemade Photovoltaics

[Dan M]

I'm sure this is an old topic, but after doing a search on the site for "home made pv", "homemade photovoltaics" etc. I don't see any information.

Is there any straightforward way to make photovoltaic cells in the garage?  I have the sneaking suspicion that it's easy as long as you don't mind power-output-densities in the neighborhood of 40 acres per kilowatt.

Is there a way to do this and actually achieve useable power, or is this a little like asking if I can make a pentium processor in my garage?

One thought I had was to use thermocouples. But I don't know haw practical that would be.

Thanks,

-Dan M

[linesrg]

Dan,

I've been looking at the same subject as PV panels here in the UK are not cheap, more in the £3 per watt range than the $3 you see in the US.

Try doing a Google search on dye sensitized nano-crystalline solar cells (DSC's). You'll need to get hold of conductive glass, titanium dioxide, alcohol, crushed berry juice and some iodine/ triodine based electrolyte in order to try it out.

Try www.solideas.com/solrcell/cellkit.html who do a teaching kit for I believe $40.

I've no idea how the costs stack up for buying the necessary chemicals required or what price the glass.

Additionally you'd need a better way of taking the power of each cell as most of the sites I've visited have overlapping plates to connect crocodile clips to which is hardly practical. I assume some very thin copper strip with a suitable insulator would work.

I hope this is of some use.

Regards

[JW]

Dan M,

 Try this link

 http://www.scitoys.com/scitoys/scitoys/echem/echem2.html

 It shows how to make a copper oxide solar cell in the kitchen. Not very efficient but its something.

-JW

 

[Arno]

Dan M, Lindsay Publications has a booklet for $4.95 "how to build a solar cell that really works"    Arno

[Dan M]

I checked out the Lindsay Publications site and I think I'll order that book.

Also, is anyone familiar with these guys:

www.fuellesspower.com/solar_cell.htm

They offer plans to screen print solar cells and seem pretty optimistic about the output.  My problem is, they also sell perpetual motor plans on the same site, so I'm afraid the whole site might be a scam?!?!?!?

-Dan M

[hvirtane]

Actually I've seen that paper. One of friends bought their plan.

Basically there is almost nothing made by themselves, it is more or less a collection of articles available there and there in the Net.

Basically I think that these plans might even work.

Still now my friend didn't have time to try making one, but his opinion was that all the materials should be available and not that costly.

The US patent on the paper is

jul 23,1996   5,538,903

There is a Japanese patent as well described

in their paper.

The technology in the paper is mainly

screen printing.

- Hannu

[Dan M]

Hannu,

Thanks!

I looked up that patent.  Everything looks fairly straightforward provided you can buy the raw materials reasonably.

Do you recal if the paper you saw required sintering?

The patent states that you need to sinter the CdS layer at 620 degrees C.  That could be a little tough to do in my kitchen or garage (and control the temp).

Thanks Again,

-Dan

[paulh]

Interesting  refs to the copper oxide cell. I got this off one to give you some idea of possible output from a copper oxide cell:

-------------------------------------------------

A note about power    

  The cell produces 50 microamps at 0.25 volts.

This is 0.0000125 watts (12.5 microwatts).

Don't expect to light light bulbs or charge batteries with this device. It can be used as a light detector or light meter, but it would take acres of them to power your house.    

  The 0.0000125 watts (12.5 microwatts) is for a 0.01 square meter cell, or 1.25 milliwatts per square meter. To light a 100 watt light bulb, it would take 80 square meters of cuprous oxide for the sunlit side, and 80 square meters of copper for the dark electrode. To run a 1,000 watt stove, you would need 800 square meters of cuprous oxide, and another 800 square meters of plain copper, or 1,600 square meters all together. If this were to form the roof of a home, each home would be 30 meters long and 30 meters wide, assuming all they needed electricity for was one stove.    

  There are 17,222 square feet in 1,600 square meters. If copper sheeting costs $5 per square foot, the copper alone would cost $86,110.00 USD. Making it one tenth the thickness can bring this down to $8,611.00. Since you are buying in bulk, you might get it for half that, or about $4,300.00.    

  If you used silicon solar panels costing $4 per watt, you could run the same stove for $4,000.00. But the panels would only be about 10 square meters.    

  Or, for about a dollar, you can build a solar stove out of aluminum foil and cardboard. For about $20, you can build a very nice polished aluminum parabolic solar cooker.    

-----------------------------------------

The lindsay plans are for the CuO cell.

However the nanocrystalline type cell has more promise, the conductive glass could be made using a tin (iv) chloride solution sprayed onto a heated glass plate, though I suspect it isn't going to be very expensive to buy.The "sintering" referred to is simply heating the plates to drive off the volatile organic carriers for the titanium dioxide and to make it into a homogenous layer and this could be accomplished using a domestic oven modified to achieve at least 450 degrees centigrade and its my understanding that a very specific temp control is not required.As to the connections between cells, overlapping edges are a really good idea as long as you produce alternate overlaps in the positive/negative conducting layers that way the cells are just in effect "clipped" together in series. I do wonder at the long term durability if these cells though because as a general rule dyes break down in sunlight, eg. your clothes fade!

some light passes through these cells and the dyes can be of different colours so perhaps a nice electricity producing stained glass window?

sorry for such a long post,

Paulh

[thunderhead]

Cadmium salts are very, very poisonous, and, like mercury or some other heavy metals, they are the sort of poison that accumulates in the body.  They also tend to linger in the environmnent.

You might be able to do this in your kitchen, but I wouldn't want to eat in there afterwards. :-(

[thunderhead]

If what you're looking for is some kind of way to cheaply turn sunlight into electricity, the two options that I think have the most mileage are thermocouples and Stirling engines.  A thermocouple is remarkably cheap to make (you can buy copper and constantan wire from your local electronic shop), or a Stirling engine using the heat from a conventional solar panel ought to deliver quite a reasonable amount of power: possibly 100W for every 1kW of thermal energy.

For a copper/constantan thermocouple the voltage at a 30C temperature difference is about 1.2mV.  So if you wanted to charge a 12v battery you'd want at least 12,000 junction pairs to get 14.4v.  Arranging them into a panel by twisting or soldering them together either side of an insulating plate should be possible.  On one side you want to cool them with some sort of circulating water, and on the other side paint them black and glaze them.

For a Stirling engine using air at atmospheric pressure, the power available is going to be something around 10W per litre of working volume per revolution per second.  So if you can get up to 1500rpm and you have a litre of volume, that'll give you about 250W.  But to drive it you'd need 2500W of heat - maybe four to six square metres of panel where I live.

In both designs, it should be possible to use a heat store and double the energy on starry nights.  Both thermocouples and Stirling engines are reversible, and a solar panel staring at the night sky will get cold.

If the designs are cheaper than silicon photovoltaics probably depends on how much your time costs: the materials for the thermocouple version would be dirt cheap, and the materials for a Stirling engine not much more expensive.  Both have the advantage of being not beyond the home constructor, which silicon panels certainly are!