Categories
Consumer Tech

Consumer tech: Solar Panels – the things they never tell you

Intro
Since solar panels are a major commitment I did some research first. My criteria (high-efficiency, not manufactured in China, carried by a local installer) was met by Sunpower X-series whose panels generate 345 KWh, which was pretty much at the high end in 2017.

The details
I live in a northern latitude area (41° latitude) with plenty of snowfall. I have a sloped roof.

I just assumed that the snow would melt off the panels at more or less the same rate as off the rest of the roof.

That is not at all the case.

I was working from home one day in the kitchen when I heard someone on the roof. At least that’s what it sounded like. The person seemed to be clearing the snow off my panels, how thoughtful of my installer to send someone to do that. The snow was thundering and avalanching off the roof onto my deck. Eventually I realized the mini-avalanches were real, the person up there on the roof was no more real than Santa Claus.

So yes, the snow slides off those panels in thunderous mini-avalanches. So today after a big snow event, this has been going on this morning, the day after, on and off for hours. Around the panels the roof retains its snow, but the panels themselves have lost all theirs. I see my neighbors’ panels are also cleared so this must be a universal phenomenon.

It’s worth mentioning because it’s a little frightening when you first hear it.

Power consumption vastly overestimated
I suppose this next problem is peculiar to just my installation. Sunpower gives you this nice portal so you see what you’re generating and what you’re using. In my case the generation numbers seem plausible, but the usage numbers are way off.

February bill shows 621 KWh metered, 386 KWh out = 235 KWh billed.

Sunpower shows 1442 KWh used, 531 KWh generated, for a net of -910 KWh.

So we can compare 235 to the 910. The should be about the same yet there is a huge difference.

The usage is almost, but not quite doubled. if we add 235 billed to the 531 generated we’d have 766 used. So usage is overestimated by a factor 1.89. But I doubt it’s a simple formula like that to correct their numbers. During the time of generation – daylight – the usage estimates numbers dip. So I don’t know what they’ve done wrong, and my installer says their support is horrible. It’s been nine months and I’ve just asked for an update. It’s more annoyance than anything.