Dennis, they are randomly distributed. The more I looked at them, the more I've decided that they are "water" and perhaps "freezing water" damage.
I had ZERO problems for the first 3 weeks of show time. It didn't rain once. Perhaps some dew or frost.
Temperatures ranged from 70F down to 20F. The damage that I see is primarily on the "downhill" side of the pixels where rain water catches in the metal lip. The pictures above were from the right-hand side of my roof and the picture is oriented as you would see it from the street.
The pixels going across the roof show more damage on their downhill side. All it may have taken is a little water intrusion, then a freeze enlarging the gap, then more water. In the presence of 12vdc and with the ground wire right next to it, being powered 24/7, I think galvanic corrosion occurred. Not seeing similar corrosion on the blue Data wire.
I think these pixels would be fine if used vertically to allow immediate drainage. Sustained rain water (relatively conductive) on the surface of a nearly level pixel is not good. If the coating were perhaps twice as thick, or the metal square edge were not so high to hold water, my luck would have been better.
I am talking with Ray Wu about this now. These pixels were absolutely beautiful in action. Pixelnet and my Active Hub on the roof were flawless. But, without some improvement somewhere, I may be asking for similar results next year.
My choices are:
- Removing the 12vdc any time the pixels are wet (I'm not always going to be home)
- Adding my own coating to the pixel surface (better IP68 and water runoff, perhaps CorrosionX?)
- Custom order from Ray with a thicker coating
- Using compressed air to blow the pixels dry after it rains (on a wet roof? not a good idea)
- Finding a different pixel. How have the plastic rectangles been doing?
- I'm open to other suggestions