I’m generally considered a reasonably smart guy. I can blunder my way through some fairly complex technical issues, write coherently, and I read stuff generally considered challenging. I’ve started and ran several successful companies, I know how to fix things. But I am capable of doing some stunningly stupid stuff. And the most recent screwup is a great example.
I have a legacy solar installation at my home in Maui that I’m using for my solar wildfire mitigation system. It doesn’t put out a lot of power, but it’s installed, has relatively new Enphase microcontrollers, 12 year old 250 watt panels, and it’s adequate. The most it’s ever done (other than some crazy, inaccurate erroneous data) is about 7Kw of power, about 220W per panel for 50Kwh per day when the system was new in the middle of summer. So I didn’t expect much. There’s a scrub tree shading one end, but not during the middle of the day (I thought) just in the morning. I really only looked at it in the late afternoon.
Recently a viewer of one of my videos commented on what a wide band of production I was getting from my panels. I had lots of theories about why. The panels leak DC to the frames (not much, but still…). The Microinverters are weird, etc., etc.. The lowest output panels were on the opposite end from the scrub tree, so that couldn’t be the problem. Right?
But a few mornings ago I realized how much shading there really was. Four of the panels on one end had a little shade when the wind blew hard. This is Maui North Shore–the wind pretty much always blows. So I pulled up my big boy pants, eased my way down the ridiculous grade to the ground mount, chainsaw in hand, and cut the bugger down. I’ve cut it down before, as the mutilated stump attested, but this time I took it off flush to the ground.
The next day I looked at the output for the day, and discovered the “weak” panels were now the same as all the rest. Some nitwit (me) did an end for end flip when I commissioned the new inverters and recorded their position in the interface.
