Rather techy DHW question...

[PhilTilson]

Sorry if this is a bit long, but...

I am setting up a system which incorporates a recirculating DHW loop so that hot water is immediately available at all taps, but I've come up against a problem that I'm not sure how to solve.

It is perhaps worth pointing out that I am based in Spain, that the whole system is solar-based and that there will be no shortage of hot water in the summer!

The first, most obvious solution was like this:

The TMV makes sure the circulating water doesn't get too hot, but there's the problem. Given that we have a LOT of solar heat in the summer, the tank temperature may well rise well above the TMV set temperature. This is fine when the taps are off, as the TMV will circulate the same water which will cool gradually, but when a tap is turned on, where does the water come from? If the tank temp is too high, the TMV will be pulling water from the cool inlet - but this will have to come from the recirculate inlet of the tank, and may well be above the set temp too!

So I came up with option 2:

This gets round the above problem by allowing cold water into the TMV if the tank water is too hot. But now how does the water recirculate when the taps are closed? The TMV tries to draw water from the cold supply because the tank water is too hot, but can't because there is nowhere for it to go!

I'm sure there must be a simple solution, but I'm at a loss as to what it is!

Phil

 

[MarkBarl]

Not an expert, but would appear that you need to dump the heat into a something like a radiator or similar device to git rid of it. I'm thinking car cooling system. thermostat opens when water reached required temp and is then circulated through radiator to dissipate heat.

 

[Agile]

You are looking at it from the wrong angle!

If you must have the recirculating loop then there is nothing you can do to control the temperature and cannot use a TMV at the cylinder. You would need TMVs at each tap instead. In the UK we would also need a combined temperature and pressure valve on the cylinder.

However, the flaw in your proposals is that its the solar heat which you need to control instead. You have not talked about that but thats what you need to be working on. The cylinder temperature must not be able to rise above 80 C at the most.

Tony

 

[PhilTilson]

You are looking at it from the wrong angle!

If you must have the recirculating loop then there is nothing you can do to control the temperature and cannot use a TMV at the cylinder. You would need TMVs at each tap instead. In the UK we would also need a combined temperature and pressure valve on the cylinder.

However, the flaw in your proposals is that its the solar heat which you need to control instead. You have not talked about that but thats what you need to be working on. The cylinder temperature must not be able to rise above 80 C at the most.

Tony

A very good point, Tony! Certainly I have simplified things to pose the question and I will, indeed, be controlling the heat input. However, I am actually using a 'tank in tank' where the DHW tank is immersed in a bigger tank (1000 litres) which supplies the UFH water. In order to store as much heat as I can for non-sunny periods, I want to get the tank as hot as I (reasonably) can - probably 80+ degrees, as you say (the specified max is 100 degrees!). But this is still far too hot for DHW, which I would like to limit to around 55 degrees. Any other thoughts?

Phil

 

[PhilTilson]

Not an expert, but would appear that you need to dump the heat into a something like a radiator or similar device to git rid of it. I'm thinking car cooling system. thermostat opens when water reached required temp and is then circulated through radiator to dissipate heat.

An interesting idea Mark - but see my reply to Tony! Still, it's worth scratching my head about...

Phil

 

[PhilTilson]

Have had another thought!

What if I put in a bypass valve (not sure about symbol!), as below:

This way, when the tank water gets too hot, the TMV shuts down the hot port, but cannot draw cold from the cold port, so the pump drives water through the bypass valve and circulates that instead until water is drawn from a tap. At that point, cold can once again flow into the TMV and the bypass closes.

As the water in the loop cools, the TMV will also cool (as no water is being drawn through it) and will reopen to let some hot through to warm up the loop again.

Or have I got hold of completely the wrong end of the stick?!

Phil

 

[Doctor Drivel]

Do the above.

Think through the TMV operation (draw it out in colours and and work through he pump flows).. If the stored water is above setpoint of 55C the cold port will be fully open giving a route for circulation by-passing the cylinder. If the cold port is closed and the hot open, the route is via the cylinder. It is worth having a pipe stat set to 50C to 53C on the return just after the pump. When water is hot, no circulation, and no waste of heat.

The loop must be very well lagged.

The water temperature can be over 80C, to about 95C without any problems as the TMV catches high temperatures. You may want to put another pipe stat on the flow from the TMV, as a high limit set to 60 to 65C to be sure.

You may want to put a high limit stat on the cylinder. When V hot water because of high solar it activates a pump and dumps heat via a radiator or towel rail or whatever.

 

[chrishutt]

I haven't thought this one through properly yet but it looks like it does more or less what you want. You'll need a non-return valve in series with the pump of course.

PS. Since I posted this Dr Drivel has copied my diagram and edited it into his post just above mine to give the impression that he thought of it first.

 

[Doctor Drivel]

I haven't thought this one through properly yet but it looks like it does more or less what you want. You'll need a non-return valve in series with the pump of course.

Yep as I described with a few non-return valves.

 

[kevindgas]

unless you return water to cylinder it will cool down in no time so defeating the object of your aim - sorry to throw a spanner in an otherwise interesting debate - as a solar engineer i'm interested in the final solution if there is one

 

[gas4you]

I personally would put the dhw pumped return in a swept T but connect this after the T that goes up to the TMV with the non return valve fitted after the TMV T but obviously before the secondary return swept T.

 

[Doctor Drivel]

unless you return water to cylinder it will cool down in no time so defeating the object of your aim - sorry to throw a spanner in an otherwise interesting debate - as a solar engineer i'm interested in the final solution if there is one

With well lagged pipes and a good pipe stat that would not be the case.

 

[Doctor Drivel]

I personally would put the dhw pumped return in a swept T but connect this after the T that goes up to the TMV with the non return valve fitted after the TMV T but obviously before the secondary return swept T.

I didn't quite get that.

 

[gas4you]

I personally would put the dhw pumped return in a swept T but connect this after the T that goes up to the TMV with the non return valve fitted after the TMV T but obviously before the secondary return swept T.

Although I now see it was not clear, I meant tee the swept T in the pipe between the T on the cold inlet pipe and the cold inlet on the cylinder and the NRV between this T and the swept return T.

 

[chrishutt]

unless you return water to cylinder it will cool down in no time so defeating the object of your aim

Is anyone suggesting anything different?

.. I meant tee the swept T in the pipe between the T on the cold inlet pipe and the cold inlet on the cylinder and the NRV between this T and the swept return T.

Still none too clear but I think you're neglecting the fact that the connection from the cold supply to the cylinder up to the return connection will accommodate flows in both directions, depending on the setting of the TMV.