After a little bit of snow...

...the sun came out again, with a few little clouds there just to make things interesting. The light travels over 92 million miles, and is stopped short in the last few thousand feet.

That's just a little motor there to give the panel a load, with some tape on the end so I know it's spinning. The capacitor is there as a starter, to help get the motor going in low light. The LCD is a thermometer, with the probe inside the panel's backside reinforcements. The sun wasn't on the panel for very long when this picture was taken, so it didn't get a chance to get to its normal winter temperature of around 40oC (104oF)

Some stress testing was done for a day or two to make sure no water found its way in. Then it was time for some ice cubes. These were for the purpose of an informal simulation of a hailstorm. The very first impact exposed a weakness - I guess that's why I put that particular cell on the "reject pile." It also explains why it had slightly lower output than the others I tested. That was the only anomaly with the entire ice-throwing fit. The panel took at least 10 impacts - not just glancing blows either, as it would receive mounted slanted on a rooftop - direct hits to the front and the frame. The ice cracked and eventually shattered after several throws, but the panel seemed unaffected. Maximum output on the voltmeter reads 2.21V, at a maximum of 3.30 amps, quite impressive considering that one of the cells has lost over 10% of its surface area.

I still hope that the yellowing of the epoxy around the corner reinforcement triangles is only cosmetic. Even if that epoxy should fail at some point, the stuff doing the real adhesion work is safe underneath the triangles, protected from UV radiation. Other adhesives inside the panel may be safer, as the polycarbonate has been treated for UV exposure.

Time for a simulation of summertime heat. The panel made it to 60oC (140oF) sitting in the sun while inside. It'll endure much greater heat when it's outside, baking in the sun, while the air temp might exceed 100 degrees. So I set the panel on top of the wood pellet furnace yet again. It reached a temperature of 89oC (192oF) The polycarbonate front bulged out, and the epoxy and caulking got soft, but no seals broke. A good deal of condensation also formed on the front, as it was the coolest surface available. It is my hope that the water was being held in the wooden dowels, which I only used because I didn't have any cheap aluminum rods handy. Once the panel was set in the sun for a few days, the condensation went away, possibly back into the highly absorbent dowels. The final version of the panel shall use aluminum rods around the perimeter.

Update - the panel went through another submersion test, and it started taking on water. I let it fill up about halfway, which took nearly 4 days. Pressing on the polycarbonate forced a little water out of both the lower seam between the poly front and the aluminum frame, as well as through the output wires. I'll use solid wiring in the final version, as the stranded wire seems to act as a wick. As for the tiny leaks along the seam, they are likely because of the inflexible epoxy I used. It is Devcon 2-ton epoxy, and it has a rated tensile elongation of 1%, so it's basically solid. I'm going to use Devcon's Metal Welder, which has a TE rating of up to 75%, which hopefully will be enough.
Also, the polyurethane foam is not going to be used in future revisions. It was either too inflexible, or else it was unable to bond properly to the aluminum, which prevented it from keeping water out of the panel as well.

< < < Page 5

Page 7 > > >