Hi All
I completed the installation of a solar thermal system last year and filled it with glycol over the weekend. It's had water in it up until now.
It's a basic system. Panel on the roof. Cylinder and pump station in the basement. The pipe run (15mm copper) between the two is long. About 12 m so that's a 24m round trip. I have thermostats on the panel and in the base of the cylinder ( near the coil ). The pump computer does it's job by switching on the pump when the panel is hotter than the store. It was only ever expected to pre-heat the 270L tank in the day and let the boiler finish it off in the evening.

So .. to the problem.

I am convinced that in the evening I am actually seeing the panel being heated up by heat transfered from the store to the panel. The panel temp remains weirdly high even when it's dark and cool outside. Because the panel is still sending a high temperature i.e 45c and this is still higher than the bottom of the tank the pump switches on.

The pipe from the panel is cold and the pipe from the cylinder ( to the panel ) is warm to the touch. Why is the panel sending 45c and the fluid is coming down cold ?

I tested my theory last night.

I manually disabled the pump. Panel temp starts dropping i.e to 38 quite quickly. I then re-enabled the pump which switches on ( 38c still higher than cylinder ) and after a few minutes the panel temp begins to rise !! this was 8pm in the evening and it was dark and cool outside. Day had been very sunny.

So any ideas from the experts ? I am thinking that either the panel thermometer is broken or i've missed some critical setting in pump station computer. Thermometer seems to work ok during the rest of the day.

Any advice would be great as I think i'm currently heating the sky !

Are you missing a non-return valve? Sounds like it is thrmosyphoning.

Edit: This (''The pipe from the panel is cold and the pipe from the cylinder ( to the panel ) is warm to the touch. Why is the panel sending 45c and the fluid is coming down cold ?'') seemed to suggest thermosyphoning, but then your 'test' suggests no reverse flow until the pump was running. Sounds like sensor cables muxed ip? excuse me, I'm just going away to think that one through.....

Tony, that's the same as I thought, but he says the panel temp did not increase when he disabled the pump. When he connected the pump back in, the pump ran, and the temp increased.

Wertert, could the pump be connected so it's pumping backwards? That would allow what you suggest is happening. It would work (if a bit strangely) in the day (taking hot out of the 'cold in' side, and feeding the bottom of the coil), but at night it would pull water from the 'cold in' side of the panel, through the coil, heating up as it went, and push (as you describe) hot water up to the panel. Bet that's it.

Obviously I don't know your panel, but I suspect (assuming it's evac tube) that it does not matter too much which way the fluid flows. What matters is which way the pump is pushing.

That's was quick. So many comments.

I think the pump is ok as the pump station as a little flow gauge build it. This seems to sit at about 1-3L depending on the speed of the pump.

I can certainly fit a non return valve but again I think the pump station has one built in.

I checked the wiring of the sensor cables at the weekend and it seemed ok.

I should add that the system seems to function ok during the day i.e starts off cold heats up gradually during the day. It does a good job of heating up the tank ( around the solar coil anyway )

Nick - i'll have another check of the over flow and connection. Look into the backwards pump idea.

Thanks again guys,

J

Tony wrote:

''What happens is that the pump turns on because the panel gets warm because thermosyphon heats it up, I have seen this problem several times before.''


That figures Tony, but in most cases would the pump not turn off again v shortly afterwards? Assuming the thermosyphoning is going 'the wrong way' from the top of the coil, heating up the panel so it switches the pump on, won't the pump then push back against the thermosyphon, taking out the hot water, cooling the panel and thus switching off the pump?

Wertert, did you notice te pump stop again soon after?

Is/are the pumpstation and/or the cylinder coil connections plumbed in the right way round? You shouldn't get thermosyphoning unless you've got movement of fluid all the way around the circuit, under normal circumstances that should stop at the pump, when it's off. Heat transference up the pipework should only extend a short way from the cylinder coil.

Hmmmmm.....

More thoughts later!

Interesting post....

http://www.navitron.org.uk/forum/index.php?topic=12798.0

I have a Resol Flowcon pumpstation.

What are you worrying about? You had a problem; now you have an answer!! :) Glad you're sorted.

Nick

wertert,

Just look at it this way, you have solved the problem at no cost and have provided "hindsite" for many on this forum who may have made the same mistake and now wont!!!!

This is just another way in which this forum helps people by learning from others mistakes.

What used to happen at Beatles concerts?

Rgds

Damon

Posted By: wertertSteamyTea - what is hysteresis ?!

In this context, the difference between the temperature at which the pump turns on and the temperature at which it turns off. They're not set to the same temperature otherwise the pump would be turning on and off all the time.