Making a solar panel

[Dinominant]

Recently I purchased some scrap solar cells. I'm attempting to connect the scrap cells together. Soldering works, it makes a low resistance connection, but it seems the solar cells and the solder expand at different rates when heated/cooled, and the solder literally falls off, flux helps, but I'd like a more reliable connection.

I tried some conductive ink, and that worked for connecting the negative traces on the surface of the cells, but conductive ink is expensive and my pen dried up after I drew the first line.

My latest attempt was to use some aluminum foil strips, taped to the surface of the cells. The tape doesn't affect the performance because it is transparent, and it holds the cells together (the panel is somewhat flexible; bonus). But the connection between the cells and the aluminum foil is crappy unless I apply pressure.

The surface of the cells is of course a silicon crystal; the texture is similar to the ceramic on old CPU's.

A quick search on wikipedia gave me these numbers:

Silicon melting point: 1414 C
Aluminum melting point: 661 C

Using a lighter to melt some of the aluminum foil strips, I found that it didn't ball up like solder does, I think my best option would be to melt (bond) the aluminum foil to the surface of the panels. As I type I have a small panel with aluminum foil in a small toaster oven, as close to the element as I could get it, with the temperature cranked to 220 C, but I don't think that will bond the foil to the silicon...

These cells are incredibly cheap because they are scrap cells, a 4"x4" panel would produce 1W of power in a nice sunny day, which translates to about 2W per $1 spent, not including the price of a roll of aluminum foil + picture frame.

Any ideas on how I can get the foil bonded to the surface of the silicon cells, without spending too much money?

Nathan

 

[PottedMeat]

Rear window defroster repair glue or conductive epoxy maybe? Should be able to pick it up at an auto parts store.

 

[Analog]

Aluminum does not take solder. Uneven heating of the Si will make it crack, so I'd stay away from that. Short of the conductive ink, a mechanical attachment may be the best bet. If you were in my lab, I'd suggest PVD Al deposition in the vacuum chamber....

 

[silverpig]

If you put too much Al into the silicon my guess is you'll change the band structure of the silicon somewhat, perhaps even enough to change the performance of the cell, probably for the worse.

You can in fact solder to silicon cells; I've done several myself. The trick is to find the right iron tip and strips.

The iron tips you want are the chisel ends, probably 1/8" wide.

The strips you want are I believe tin that come in a long ribbon that are pre-coated with solder. Apply the flux to the cell, lay the strip down on top of the flux, hold the iron down on top of the strip at one end and once it melts, move it at a steady pace down the strip. I helped build a several hundred cell array this way.

 

[Dinominant]

I can solder to the traces on the silicon, but not the silicon itself. When I solder to the thick connecting trace on the surface of the cell, the connection is good. But because these are scrap cells, I need to make my own connecting trace on the surface. I wasted a large amount of solder on a piece too small for any use, and I could hear the solder cracking as it contracted, later when I picked up the cell, the solder trace just fell off; I can only imagine what would happen during fall/spring when the temperature can shift from -20C to 20C in 2 days with 100Km/Hour Chinook winds.

I made a small prototype 4"x4" panel by drawing solder traces on the surface, it worked, but it would require repair work if I even though of bringing it outside to test it. It was extremely fragile.

I had initially planed on using the packing tape + aluminum foil with a stuffed picture frame to apply pressure, but it turned out, the required pressure only happened near the edges, while the center of the panel would dome out creating air gaps. I could layer more sheets of paper near the center of the panel, but the strain on the glass + the temperature shifts could be disastrous. Large picture frames were defiantly not made for that.

I'm going to buy some conductive epoxy and see how that pans out. Of everything I've tried, the aluminum foil + packing tape has worked the best while being the most durable. Its performance suffers when the temperature rises in full sunlight so I would get low performance all year; unless I apply pressure with a clear plate of glass/plastic.

One thing that has surprised me is the majority of the cost that goes into constructing a solar panel out of scrap cells seems to be in everything but the cells. I may have to cut my losses and live with the below average performance, but I'm sure there is a solution...

 

[Sahakiel]

Foil strip + paper clip.

 

[NeoPTLD]

Been there done that.
Solution: metal report cover. That black clip thing with two metal clips attached to it. Search google images for metal, clip, office.

Sand down so steel is exposed and simply clip down.

If that's not elegant enough, use conductive adhesive together with nickel strips used for making battery packs. Use aforementioned clip to hold the strip in place while adhesive is curing.

 

[heyheybooboo]

Would copper foil tape used for stained glass help? The adhesive backing is quite stout and it most definately would take solder.

A quick Google found:
5/32 in. wide strip
10 mil. Thick
25 ft. Length
Price: $ 3.95

 

[DrPizza]

how about the stuff they use for making good electrical connections (it's like a gel) when connecting aluminum wires in breaker boxes?