|
Post by badger on Jul 28, 2011 18:35:07 GMT -5
Having bought one of Cye's little pumps I promised that I'd post details of my plans to capture a bit more energy from my SHW system. If nothing else, it should be good for a bit of comment. My 50-tube collector array (chinese heat pipe tubes, 47mm) generally gets the tank up to 55-70°C on sunny-ish days if we have a couple of showers. In high summer and/or if we have a no-shower day or awayday it can certainly get into danger territory, so I need to add a heat dump. Given that on quite a few days the system 'tops-out' at say 4pm and 70°C, there are still a couple of hours where more energy could be captured in another store. My mains-pressure HW system rules out any economic addition of a further connected store, but I do have an old indirect cylinder, some 3-way valves and a second differential controller. The attached pdf sketches show how I can easily dump excess heat to this 2nd cylinder just using the switching in my existing controller (Prozeda Solareg Vision). Rather than leaving this cylinder uninsulated, and being mean/green, I would much rather insulate it very well and be able to sometimes reclaim the heat when the main store drops to a suitably lower temperature (typically after a late night shower). I can do this by use of a 2nd 3-way valve (and a rather odd flow path, I admit, see pics), but the tricky bit will be control. I've picked up a 2nd diff controller (a simple Worcester TDS 10) and sensors, but I have yet to work out how to stop the controllers fighting each other. I also probably need some sort of 'stirrer' in the tank, or the hottest water in the top of the dump tank (above the coil) will not be doing anything useful. Any suggestions for change etc welcomed! Attachments:
|
|
|
Post by cye on Aug 2, 2011 18:01:07 GMT -5
badger. i'm having diffs following the logic of the 2 X 3port valves. Can you confirm i've got this right:
[1] i can see how V2 could switch the solar panel hot output between the two stores. one could use the heat dump trigger on your original controller to throw v2, so when store1 gets too hot you divert heat to store2, but i am guessing instead of using the heat dump trigger you're using the second diff controller to flip v2 when one tank is X degrees above the other. This seems to make sound sense, except that controller number one will need some way of flipping the reading of 'tank temp' between the first and second store depending on which way v2 is 'pointing' - this could be achieved by feeding the two tank sensors (store1 sensor and store 2 sensor) into a changeover relay activated by the same control voltage as that which activates v2. this way controller number one always reads the tank temp of the tank that's in the 'active' solar loop. have i understood this correctly?
[2] now for v1 to move heat between the stores, presumably when sun not shining (?), the pump will need to be to the left of v1, as it's current position will pump water through the panels. not sure how you woud control this either. perhaps manually?
[3] a little brown pump can destratify your second store by circulating from top to bottom.
|
|
|
Post by badger on Aug 4, 2011 15:58:57 GMT -5
Cye; Yes, my issue will be the differential controls, and you are only not clear on exactly how it will work for the same reasons that I'm not! In general I would always expect the heat store to have been largely exhausted by the time the next cycle flips V2 again, but there will be times when this is not so, and a diff control or at least a temp-based override will be needed. With 2 diff controllers in the Prozeda and a separate TDS10 I should be able to do what I need if I can figure out the detail. Your suggestion for a c/o relay looks promising. I couldn't see an easy way of minimising pipework without pumping back through the panel (final page of the pdf), so that is indeed the way it is schemed. The heat-tube panels don't radiate in reverse and the manifold is superbly insulated, so although it looks illogical, there are actually only small losses this way. I had considered using one of you little pumps to de-stratify the store if that is potentially better than a 'mixer'....a good plan.
|
|
|
Post by cye on Aug 5, 2011 14:51:48 GMT -5
Badger, thanks for clarifying, I don't feel quite as bad now - I was scratching my head for a while on that one! Your idea on relying on the insulated vac tube panel is fairly sound too, provided you've checked the waterproofing on the manifold if your's is std chinese vac tube. i went over the end caps on my manifolds with RTV silicon because i was uncomfortable with the provision for waterproofing on the manifold casing. The manufacturer was relying on a good seal between the plastic end cap and the stainless manifold shell, which was questionable in my view.
if you retain a permanent T just south of the pump, then, as your diagram already shows, you do not need V1 at all if you are just using tank 2 as a heat dump. So V2 is needed no matter what.
now, for the trickier bit - how to transfer between the two tanks.
I'm not necessarily recommending the following, just walking through a possible alternative, just to see whether another scenario is plausible:
(1) assume we only need transfer between the two tanks at night, when the main pump is off. (2) assume we have only a single NRV located between the daytime (main) pump and the panel. I.e., no NRVs between the two lower coil connections. (3) assume that we will not pump through the panel at all for tranferring heat between the tanks.
what then may be required to achieve the above night-time heat transfer? (a) do not use the main 'daytime' pump as this always pumps through the panels. (b) dump the v1 3PV altogether. (c) use an additional small pump to shunt water between the bottom of the two coils (there'll be no NRV in the way) (d) use a normally closed 2 port solenoid valve, which, when opened, creates a free flowing shunt between the two hots. this could be located up at V2 to minimise plumbing. in this way, it doesn't matter which way V2 is pointing, it is now irrelevant, as the two coil tops are now connected.
now how do we control these? Simplest thing is a manual switch of course.
However, if you're dead set on electrickery control.......as your second controller is probably going to be monitoring the difference between the two tanks anyway for daytime flipping of V2 and the swapping of the tank sensors of controller #1 , you could use the inverse of controller #2's output to power the extra pump and the solenoid referred to above (and your destratification pump too if you like). use a photocell controlled relay or a timer to disable all of what i've just described [ in 1to 3 and a-d above] during the day, so that daytime is focussed on getting both tanks as hot as possible.
i'm sure there are several other scenarios that may also work...
arthur guinness is calling, so that's me off.....
|
|
|
Post by badger on Aug 8, 2011 18:24:05 GMT -5
About to go caravanning so have read your latest, but not fully absorbed....more on this later, thanks.
|
|