I haven't time to write much today, so here's a post that's been sitting around in the
drafts folder since the end of April:
Last time I wrote about the solar panels, they were installed, just about finished, and I was waiting for sunshine. That arrived about a week later and I got very excited to see how the panels would perform. They (at least, the booster panel, where the sensor is) got hot very effectively, but the temperature drop between panel and tank was so great that I started to suspect that the water wasn't circulating at all. I wondered whether there was something wrong with the valve I'd replaced - maybe I'd wired it up the wrong way round? I stood on a ladder and, reaching into the top of the cupboard, played with the connections, accidentally touched them together, and with a flash and a bang the whole thing stopped working.
I was sure I'd blown up the pump, but a visiting friend poked around with a multimeter and tracked down the fault to the relay. It was still switching, which had me thinking it was fine, but he explained that I could have burnt out the contacts within the switch. Thanks, Adrian! In fact, when I prepared to replace the relay, I found the fault was even simpler than that. The solder had melted on one of the connections and a wire had come loose, so that was easy to fix.
So much for all that, what was the problem with the system? If the panels were getting hot, why was no heat reaching the house? I concluded eventually that the water was circulating and it was a matter of losing all the heat en route: not enough insulation. Since I was dubious about my boxing in of the pipes across the conservatory, that seemed the most likely place for heat loss. On the other hand, it would also be the most difficult to improve, so I did further testing first. I added a long wire to the third temperature sensor and moved it to each end of the conservatory pipes in turn. To my surprise, they didn't seem to be losing much heat at all across that distance.
Where next, then? I took the sensor out to the booster panel and tested the temperature as the pipe left the box: Almost 20 degrees lower than the temperature inside the panel. What? How about just inside the box, then? Just the same. How about in exactly the same place as the main sensor? Maybe this sensor isn't working properly. No, it is working - same reading for the two sensors in the same place. OK, I really was losing 20 degrees between the point where the pipe left the collector (black painted aluminium sheet) and the point where it left the box. That was about 18 inches! Also, the pipe ran through sheep's wool insulation, and I'd expect the space in the box to be fairly warm, anyway. Apparently not. I added lots of insulation to that short piece of pipe.
On testing, that reduced the heat loss from about 20 degrees to about 10. Good, but still room for improvement. Now knowing how much impact a short stretch of pipe can have, I turned my attention to where the pipes go under the conservatory roof.
The draft post ends here. I stuffed a load of insulation around the pipes under the roof. It didn't make much difference. That's probably all you need to know about that. There's still work to be done to improve the insulation, but I'm not sure where. That's a job for another day, preferably a sunny one so I can test things.