Things I would do different next time

I would run the 15 mm ID silicone tube out of the panel and into the roof space before I reduced down to 6 mm ID tube. This stops the worry that the brass reducers would burst through freezing.

Although I think this would be unlikely, and there have been no problems so far, with the panel been frozen many many times.

I would definitely arrange for the panel temp probe to be changed without dismantling the panel. Some way to change the probe from outside would be a good idea.

I would like to use a different pump. The Charles Austen pump is no longer available to the DIY customer. Due to pressure from the solar thermal trade. So I would like to think I could get a replacement pump easily should I need to in the future. There are also no spare parts for the C/A pump. Now using a different PUMP

The two gate/isolating valves I have used are not very good. When messing about with things I notice that they let a small amount of water pass when closed. So I will have to change these with lever type ball valves.

I paid far too much for the Polycarbonate sheet. Do a Google search for "polycarbonate sheet cut to size". You should be able to get a sheet this size for about £35.00 delivered.

Would not bother trying to charge the battery or running the pump with a P/V panel. They are too expensive and unreliable. It is more cost effective to charge the battery from the mains. About £4 a year. P/V is all part of the zero carbon con.

20-02-14 When I was experimenting and setting up this system I remember reading somewhere, that it was important that the cold feed connection to the pump should be close as possible to the cylinder. I could not see the point of this at the time so ignored it as my cylinder was in an akward place to get to. The cold water connection to the bathroom. WC and hot taps were getting full of hot water on good pumping days. So the toilet and cold taps would give a pulse of hot water. About 3 liters. This was wasteful and annoying. Also the thin plastic membrane that acts as a valve in the sistern failed, I think due to the systern filling with hot water. So today I moved the cold water connection so it is no 1 next to the cylinder inlet. This should overcome the hot pulse problem.

ball hotballcold

Also At the same time I changed the two gate valves that can be used to completely isolate the system. I noticed in the past that although these valves were screwed closed tight they would let water past. Lime scale and other debris would stop the valves from sealing. I now have lever ball valves instead.

gate valve

You can see from the picture above the alarming build up of lime scale. Its no wonder that the valve would not close.The lime scale appears to build up on the actual gate of the valve. These valves are alright for regulating flow, but not for shutting off completely. The rest of the system seems surprisingly free from lime scale especially the silicone tube.

To achieve this task today it was necessary to completely drain the cylinder. This was a real pain. It was a dull day so I never had to worry about the panel getting too hot. There was dull sunshine about 3PM when I started to refill the system. But the top of the cylinder soon became almost too hot to touch. With free hot water.