UFH Mixer valve way out of spec

[Tim Watts]

I've just fitted 2 single zone underfloor heating mixers to my system:

So far, the UFH itself is working very well.

But I am extremely displeased as to the performance of the mixer valves.

They are supposed to be set to around 35C outlet when knob is on minimum. As you can see, 57C when the boiler is racked up to max and producing 78C.

I wouldn't run the boiler that high, but the point of these valves is to blend down. And it's not. This is the conservatory UFH which can sink about 2kW on a good day in theory.

Valves don't seem to be stuck - if I jack it to max, there's a big rush of water, which stops when I put it back to min.

All I can assume it it's letting by through the hot port even when it's hotter than the set temperature. Return temp is 5-8C cooler on the UFH circuit so I'm confused.

Anyone had any experience of this sort of thing?

I will be contacting the reseller too. But just to cover other angles...

It *may* be possible to swap in a better quality of 3 port valve if I can find one with the same dimensions. The pump spigot is an adaptor so I think they are all the same (quite large) thread - possibly a standard of some sort.

Tim

[Tim Watts]

On 15/11/18 13:28, Tim Watts wrote: ...

I did just speak to the supplier when I found their phone number.

They said the mixer valve might be happier with a lower return temperature. I had about 6-8C as the pump was on it's max setting (long UFH circuit, other side of the house).

I've set the pump to its lowest pressure setting to try to aim for a suggested 15C difference of UFH flow/return.

We'll see what that does...

If that works, it's going to leave me with a problem on the shower UFH which is right next to the unit with 3m2 of floor - I'll never get that sort of drop.

I might have to look at some other means of regulation - perhaps a pipe stat on the output in series with the zone valve on the feed and modulate the feed completely off when the output goes beyond an acceptable limit.

It's crude and I can see it wearing out the zone valve...

Other option might be to throttle the feed on the isolator valve in an attempt to balance the system...

Tricky. I assumed 3 port mixers just worked - did not envisage this much trouble dropping the water temperature!

[John Rumm]

What is the temperature on the cold inlet of the blending valve though?

If the return temp from the UFH is above your target mix temperature there is not much the valve can do to lower the output temp... Even if it were taking almost all its input for the mix from its cold leg - that would still be over the set temp.

That would suggest you also need to throttle the feed rate through the UFH so as to achieve enough temperature drop.

[Tim Watts]

Hi John,

Similar on the shower (tiny zone) and was 6-8C lower on the conservatory.

True. But i assumed it would shut off the hot input completely and basically just loop the UFH water until it dropped to target, then leak a little bit of hot feed in to maintain that.

It seems the valve is imperfect in that regards.

I can certainly do that on the shower which has <10m of UFH pipe as it's only 3m2. I may or may not need to on the conservatory as the resistance is high anyway - but it's an option.

Do you think that's a better approach than throtting the boiler feed massively?

I am going to include a pipe stat on the output too and wire that in series with the zone valve on the feed - that way at least I can put a guaranteed ceiling on the UFH feed temp. Might wear the zone valve out if it comes into play all the time, but it would be a good failsafe at least.

This is proving far trickier than I imagined. I knew the shower would be awkward, but the conservatory is a decent size of circuit.

Perhaps I am also being too conservative with my flow temps - I wanted to go gentle, almost run the circuit most of the time with water that was roughly where I wanted the floor temperature to be.

You'll remember I have two soil stacks in the shower room and didn't want to overheat the floor and risk it moving too much. In actual fact the movement between the stacks is only a theoretical 0.7mm for a termparature swing of 20C and I cannot see any evidence of expansion over the whole length of the room (theoretical 1.6mm) in that the silicone beads all look relaxed and unmoved.

In terms if utility, it's working well - the warm floor dries quickly and the conservatory is now fully heated (in fact too hot!) from the UFH alone. Which is not the plan. The plan is to run a lot lower and boot on the air blower when needed.

[Tim Watts]

Got to go to a school evening, but I think it might be time for abit of O-Level physics (water flow, temperature drop, power) to figure out a good flow rate for the floor.

And then to see if that is going to annoy the pump. Had the strangest call to Grundfos just now:

Me: "Does the UPM3 AUTO L have a minimum flow rate requirement?"

Them: "It's a custom pump, we can't answer that".

For a "custom" pump, it's certainly pretty ubiquitous! The flow/pressure curves show the curves going down to zero flow, so in the absence of other information, I am going to assume on Constant Pressure Level 1 (1.5m head of water) that it will just regulate its power even if I throttle the UFH circuit down to a very low figure.

Really, a full size pump is a bit massive for a single circuit, but what can you do... At least they are user adjustable these days.

[The Natural Philosopher]

16 years ago I set mine (polyplumb) to 5, and just haven't touched it since.

I'd guess the water goes out at around 50C max.

No screeds have cracked.

[John Rumm]

Does the hydraulic arrangement you have allow the pump to do that? (a quick diagram may be handy)

Its seems sensible to have the feed to the system able to supply whatever it demands, and not use it selectively, rather than limit what it can supply artificially (or at least to anything lower than the theoretical maximum)

Does your boiler use weather compensation or have other mechanisms that will limit the normal flow temp anyway?

(mine will quite often run a flow in the mid 50s anyway - it only gets into the 70s when recharging the hot water or on really cold days (i.e. several degrees below 0)

You will probably find the plastic soil pipe moves more with its own temperature changes anyway...

(I notice that when you run hot water in an upstairs bathroom here, you get a certain amount of clanking from the soil pipe that runs through the building as it expands and contracts)

Yup keep the chill off, but then heat as required...

[Tim Watts]

Sure John - here you go:

(No login needed)

I agree.

No - WB didn't have an option for simple WC (you needed their fancy programmer to do that).

I run on setting 4 which is 55/45C most of the year and jack up manually to 5 or 6 (65/55) in the depths of winter if required.

That's the master plan :)

[Tim Watts]

Have you measured it?

[Andrew]

It's not just the screed that matters. The pipe itself does not like being forced to carry water at a higher temp than it is designed to.

[Andrew]

When I run any hot tap in the bathroom, almost immediately there is a ticking noise above the ceiling where the stack goes through. I guess it is in contact with one the trusses and it is expanding,rubbing and creaking.

[John Rumm]

Make me wonder if it could end up drawing some of its "cold" water from the return of the primary heating system in preference to the output of the UFH loop... Again that could lead to a hot cold input.

It would be handy to see the inlet temps on both sides of that TMV when the system is running.

Oh well, that gives a bit of time (hopefully) to fiddle before you have to worry about really high flow temps ;-)

[tabbypurr]

not possible. Consider the various flow rates.

NT

[Tim Watts]

John Guest is rated to full CH temperatures :)

[Tim Watts]

I wouldn't have thought so as the pump is pushing against the return.

I'm pretty sure the mixer is letting by via the hot inlet - if I operate the hot isolator while it's running at at temperature, there is a sound variation as I alter what should be no flow.

I noticed some potable mixers designed for sinks, like the Honeywell, have an anti scald function that shuts off the hot inlet 100% if the outlet goes 10C above the set limit.

That makes me suspect that some mixers do not fully close off and assume they will always have a certain amount of cool water available.

But I will do a full temperature assessment tomorrow. I have a couple of inline (well, I hope they are 2x15mm compression - looked like it in the picture) lockshield valves.

I'll cut one into the shower circuit which is short and see if throttling UFH helps the mixer to behave.

Indeed - I'm not even wiring this up properly until debugged :)

Pumps are plugged into an extension lead right now...

[tabbypurr]

I'd have been inclined to use an electrical pipe thermostat. They're adjustable, reliable, cheap, don't leak & are easy to replace or make programmable. Make all 4 pipes meet in one spot with only a narrowed opening between primary & UFH loops. Without the ufh pump running not much hot would get into the ufh. With it running rather more would.

NT

[Tim Watts]

Added a rad balancer valve to the UFH loop flow side.

Still refuses to regulate - T1 UFH is 60C - boiler flow is 65-70C at this point.

What I noticed is with the rad valve shut down to 1/4 turn open, the Cold port and boiler return were red hot suggesting that boiler flow water is travelling directly from the Hot inlet of the mixer to the Cold inlet and in fact exiting.

Crank the rad valve open a bit more and I can get enough cool return from the UFH to cool this down but the mixer still refused to regulate even with a flow/ret on the UFH of 50/30C

The only thing that "works" is to shut the red inlet butterfly right down to about 1/8 turn open.

I think the mixer valve is a piece of crap and possibly not suitable for UFH. Just waiting for my ESBE VTA362 to come from ebay and see if I can get that to fit (if the threads are the same and the spacings).

I have told the seller exactly what I think!