Solar panel project

[Jeff7]

I've already got a thread in the Solar forum, but it seems more relevant to this section.

My first prototype failed; I've yet to find a source of Devcon Metal Welder. Toolwatch.com has it, but their online ordering system isn't fully programmed yet. I placed an order by phone, which was tedious; it seems like their company just formed yesterday. I would have shopped elsewhere by now, but I can't find ANYONE who sells Devcon's Metal Welder, and Devcon doesn't seem to sell their stuff themselves. I e-mailed Toolwatch to inquire as to my order, if it's even been processed. No reply to that yet.

I'm hopeful that the Metal Welder will work. The epoxy on the first prototype failed the heating test - the polycarbonate expanded outward, cracking the inflexible 2-ton epoxy I had used. It's got a tensile elasticity rating of 1%. The Metal Welder's TE is 50-75%. So it should survive heating and cooling much better. And it's supposed to adhere better to aluminum and polycarb than the 2-ton epoxy.

Anyway, I've finally done some pricing of materials. I'm planning to build 100W panels, which will measure 59x28" - the cells shall be arranged in 4 columns, with 18 rows. I figure on using aluminum 5052 as a backing - good thermal, mechanical, and weathering properties. These aren't going to be cheap panels though. Prices just for the front and back panels, polycarbonate and aluminum respectively, will total $120.65 per panel. Add to that the cost of 2 rectangular tubes on the back for structural support and mounting, aluminum framing around the perimeter, and the Metal Welder. Plus the cells themselves.

But then, professional panels in the 100W range cost about $500, so this is still a pretty good deal. Not quite as low-cost as a lot of the other home-made panels here, but I'm also building my panels to withstand whatever punishment the weather will dole out, and to continue doing so for many years.

A few things I haven't liked about some of the other designs I've considered:

• Even sealed wood may warp after repeated heating and cooling. And if any of the coating cracks, moisture may get in and accelerate the deterioration of the wood. This is why I looked for something else, and settled on aluminum 5052. Information I found says that it has good corrosion resistance - no coating needed.
  
• Silicone sealant has, in my experience, been somewhat prone to peeling. Using just a single layer of it around two panes of inflexible glass may cause it to break its relatively weak bond to the glass.
  
  

This is why I'm looking at Devcon Metal Welder. Excellent adhesion to both polycarbonate and aluminum, and some flexibility.

- The tempered glass that the big companies use has a few layers of plastic protecting it.

This is why I'm using polycarbonate - flexible, and impact resistant.

[Tom in NH]

Nice project! What does just the aluminum backing run? How does it compare to the price of half inch plywood? Did I miss somewhere where you said how thick the aluminum backing is? Boy you sure know how to give punishing treatments for testing!

The condensation you showed in your picture, it is my guess that it is simply condensation of the ambient humidity when you sealed the panel up. You could probably reduce this by building the panels in the winter time in the cold when the air can hold less absolute humidity.

The longer my open design panels http://www.fieldlines.com/story/2005/1/5/51211/79555

have been operating, the more convinced I am that ventilated panels are the way to go. In the morning there may be a spot or two of condensation on a panel, but within an hour or so of daylight, the condensation is gone. --Tom

[Jeff7]

The aluminum is kind of expensive - I bought it from Onlinemetals.com. But it's pretty much guaranteed not to corrode or decay, unlike plywood.

It was $12.60 plus shipping for a 14x9 inch piece, so the stuff's not cheap.

The aluminum backing is 0.125" thick; I don't recall ever mentioning that. Oops.

The condensation was likely formed after the epoxy cracked, allowing the helium to make a speedy exit. The panel was sealed in a dehumidified basement, in winter, so I think the air is pretty dry. And it was probably very dry inside the panel too, due to the helium used.

I'm worried about exposing the cells to the elements. After prolonged exposure to moisture (read: submersion), the silkscreening starts to peel off of the cells, and they get rather dingy looking overall. That happened after I did a second submersion test of the panel, this time after the condensation and epoxy cracks had formed. It filled up about halfway with water, and I left it like that.

I just get the idea that the cells themselves aren't meant to withstand exposure to the elements - why else would the manufacturers seal up their panels so thoroughly, when a simple frame with a sheet of glass would work just as well?

When and if I ever find out where the heck I can buy the @*%#!^& Metal Welder, maybe I can get this project back underway. What's the point of making this wonderful adhesive, when you make it impossible for us ordinary folk to buy it? It isn't extraordinarily toxic - the Material Safety Data Sheet makes it sound a good bit tamer than Plastic Welder. And that Plastic Welder is some pretty damn nasty stuff. Awful, awful stink to it. It does an excellent job of bonding to plastic (except polypropylene and polyethylene), but you have to use it outside.

[BeenzMeenzWind]

If you want to build sealed panels and avoid condensation chuck a load of those little silica gel packets in there before you seal it up. They're always in the boxes for new shoes and all sorts of other stuff here in UK, I imagine they are in US, too. If you tack them to the inside of the frame with epoxy, they won't get in the way of the cells and should soak up most or all of the water vapour. Maybe even create a very tiny drop in internal pressure, too, as mass is removed from the air inside, which has got to be good.

I've helped my friend build a couple of 'em from Ebay-cells and we did that every time. Seemed to work fine. He also likes to create a partial vacuum in the panel by grafting a little glass tube on to the corner of the glass, sucking out some of the air with a little vacuum pump and then heating and twisting the nipple off to seal it. He reckons that you get a tiny bit more power out the cells, as the lower air pressure inside is harder for the sun to heat up. Not sure whether that's the case, but his panels seem to work OK, so...

[Jeff7]

I've seen dessicant packs mentioned, but they have a problem - they don't remove the condensation from the panel. They merely store it temporarily. I'd just worry about them releasing the water when they're baked in the sun.

Unfortunate update on the Metal Welder. Seems that it's a professional product, available only to large companies using fair amounts of it on a regular basis. Wonder why that is? Guess I'll have to look at other options. Maybe their Flex Welder or Composite Welder will suffice, assuming it's possible to obtain those products.

[BeenzMeenzWind]

It depends on how hot the interior of the panel gets, I suppose. I know you normally heat them to 200 degrees to 'flush' them, but is that just to be on the safe side and make sure all the water's gone?

With regard to aluminium welding / adhesive stuff. Have you looked at TechnoWeld?

http://techno-weld.co.uk/home.html

Not sure whether it's the same kinda stuff you want, but I've used it before for making up ally manifolds and it works pretty well.

Might be useful for something else, if not for this?

[Jeff7]

I estimate that the inside of the panel will get pretty damn hot. Sitting in the sun, indoors, the cells reach 60C. Now seal them up, and set them in the sun on a hot summer day. You'd get burns if you would touch the panel.

Update too - I discovered that Toolwatch.com has a forum, so I gave my ordering issues a little publicity. 41 minutes after posting, their admin posts, wanting to help. I e-mailed, and they called me back. So, I should find out soon if they'll be following through on this. Still no idea though when their online ordering system will be ready.

Was interesting to see how this one thread sparked such a quick response, whereas my phone calls and other e-mails went unanswered.

So hopefully, I've got some Metal Welder and a dispenser gun on the way. All I can say is, this stuff had damn well better work. I've got a few design revisions in mind too for my Version 1.1 prototype.

[Jeff7]

And that Techno Weld, while it looks like it might be handy for other projects, has one problem - since you've got to get it fairly hot, I'd be wary of what the heat source would likely do to polycarbonate.

[Jeff7]

Update - finally got the Metal Welder and the applicator gun. I've just got to wait for the applicator tips to arrive - ordered them elsewhere. That should be later today. Prototype v1.1 might be underway soon.