Underfloor heating mixer - inbuilt loop stat - any point?

[Tim Watts]

I've got a couple of these to be fitted next week:

(because 2 units is simpler to control than a 2 way manifold and lets me set different loop temperatures)

Couple of things:

1) The 3 port valve - I presume this works like a shower valve:

a) Is there any risk of the boiler pump pushing water through the UFH loop even if this pump is not running? ie do I need a separate valve to shut the loop off?

b) When the loop is up to temperature, from what I can see, no hot water should circulate on the boiler side? This is important because I don't want to have a short circuit shunt on my system - I'd have to install a balancing valve too to counter that.

2) Is there any point in the inbuilt thermostat in the hot leg. From what the instructions say, it's a NO and closes when the temperature on the boiler circuit gets to 43C, then turns the pump on. This implies some let by (Q1a above). But also, it seems largely pointless.

My controls, when finally wired will turn this pump on when there's a call for heat - and will also demand the boiler come on. I'm really not sure I see much point in the inbuilt stat, it's going to save a tiny bit of pump run time.

The only point will be, is if I just supply permanent power to this circuit and use a water valve to govern supply from the radiator circuit***

*** This actually makes the control system simper as I have ZWave and can just use a couple of radiator radio controlled valves on the feed pipes.

Many of these mixer assemblies don't have the stat - this was simply the one I picked that had a sane price for what it is (many are double the price for the same sort of thing).

Cheers, Tim

[John Rumm]

Presumably the flow from the CH goes into the "hot" side, and the cold side is connected to the return from the UHF, thence the CH return?

If so, then there will be a pressure differential between the hot and cold sides, so in theory you could get some flow, although I expect in practice, the flow resistance through a complete UFH loop will be significantly more than the other rads in the system. So any flow would be minimal (especially when the UHF pump in the circuit is dormant)

No sure I follow exactly... the main CH circuit is presumably controlled elsewhere, and so would continue to flow regardless. Presumably the presence of the TMV on the UHF circuit will prevent the loop from ever getting fully "up to temperature" etc, since it will seek to limit the flow temp into the loop - it will do that by blending ever more loop return water into its flow, thus reducing the amount of primary CH water that can flow into the loop. So I would expect to see the flow resistance of the loop as a whole appear to effectively increase as it gets warmer.

I suppose the logic is that the screed with the UFH will have a much higher thermal mass than the rads. So when the main CH cycles off its water temp will fall to below that of the screed. When it restarts, having the stat on the UFH will keep this now cooler water out of the UFH (hence cooling it) until the boiler gets the primary water back up to temperature.

I guess some let by is desirable in the case of the stat to ensure that it can actually "see" the CH flow temperature. You can imagine that in cases where the UFH feed from the main CH is on the end of a fairly long branch from the main CH flow it would always appear cold without at least some flow.

[Tim Watts]

Yes - that's right.

Yes.

That was my theory :)

That does make sense.

Hmm.

I think I have a plan...

I think I am going to put an inline TRV body (15mm-15mm) in the boiler feed to this and pop a Heat Genius battery TRV head on it for starters.

And wire permanent power to the mixer pump/stat combo.

That will be a similar setup to my Myson fan blower in the conservatory:

That works pretty well - the Myson is powered up by a ZWave 13A socket switch on the left, but strictly I could leave it powered all the time as the fan can be set to only run when it sees hot water.

The ZWave rad valve is a Danfoss Living Connect and the only "fault" is they are progressive by default - by the HeatGenius system has a supported hack where you can force one to open fully (as in loose and wobbly on the TRV body).

In practise that gives enough flow for a 3kW load at delta-T 45C

The conservatory UFH is a nominal 2kW so this should work.

Using radio does simplify the electrical wiring.

But I'll make the pipework easy to mod in case I want to take it out and control the pump or change it to a wired zone valve (with the extra wiring and controls that needs).

I had to muck around with the Myson for a couple of weeks trying different control methods until I found a nice one. I suspect this may go the same way.

Hard wired with zone valves would be a more solid solution, but it *may* be, if the let by is minimal, I might go back to controlling the pumps directly - in which case I'll wire it into my controls properly next year when it's warmer.

For now, the pumps will literally be plugged in (but as there's no electrical interconnection with the boiler circuits, that's not really a problem - at least not from a safety POV).