This setup is apparently working fine for the past 10 years, its set to give a UFH temp of 30C from a oil fired boiler running with a 65C flow temp. I havn’t seen it in action yet but just can’t figure out how it actually works, the manifold pump is pumping downwards but I can’t see how any cold (UFH manifold return) water is returned to the boiler.
Quite simply, the TRV controls the ‘new’ hot water into the UFH loop when the water in the UFH loop drops below the set temp (30°C). If it opens, heat is introduced from the supply.
The boiler cannot be supplying it directly. It needs some form of buffer. Possibly a rad system or a tank.
1 Like
First the manifold, for any water to recirculate back to the boiler then there must be no connection between the left hand side of the hot (mixed) bottom maniflod and the cold outlet to its left, if not then no problem IMO, its my daughters system but I havn’t had a look since its quite a distance from me. Its fed from a 20kW, 20 year old, non condensing Firebird Oil Boiler and runs for 12/14 hrs per day with the rads only on for maybe 2 hrs/day, it certainly does not have any buffering but cycles ~ 1 to 1.5 minuts on and ~ 6 minutes off, on UFH only, the manifold inlet temp is 30C and I would think the return is 25C with a total 3 loop flow of 7.5LPM for a UFH output of 2.6kW, boiler flowtemp is 65C, this means that only 7.5*(30-25)/(65-25), 0.94LPM is returned to the boiler, the remaining 6.56LPM of “cold” water is just recirculated with just 0.94LPM from the boiler mixing with it. Oil fired boilers have a inherent “advantage” over Gas fired boilers under these conditions since they have a inbuilt buffer, the boiler HEX itself which contains 23L on this boiler vs maybe < 2.0L on a gas boiler, the boiler stat hysteresis is around 10C, so after burner cut out at 65C, in falling to its cut in at 55C, will emit around 0.27kWh, enough to supply the UFH demand of 2.6kW for ~ 6 minutes or so, it will then take around 1 minute or so to replace that 0.27kWh when the boiler refires. I know some have a big thing about boiler cycling but that boiler and my own 20 year old (same), havn’t required a single replacement item apart from a few service items like a few burner nozzles and door gaskets.
There’s another pump in play somewhere, presumably - between the boiler and the UFH manifold - and some controls which send the boiler a ‘call for heat’?
Yes its supplied from the boiler’s circ pump, the Wilo manifold pump is pumping downwards into the flow manifold presumably because the arrow on the pump body is pointing downwards but I will check all this out when next I visit.
pump
Exactly what I would suspect without a buffer of some sort.
Sorry, I’m not following this.
If the return from the UFH was the same temperature as the set (on the TRV) flow temperature, then the TRV will close so the pump just circulates the water round the UFH loop.
As the UFH return temp drops, the TRV will open to allow hot water in to mix with the UFH return water to bring the flow up to the required temperature.
What problem are you trying to solve?
This probe here is what is supplying the UFH flow temp to the TRV so that the TRV can decide to open or not.
Thanks Brian, I have attached my “masterpiece” of a very conventional UFH set up which is quite logical and easy to understand. The UFH loops (shown as one) emit 2.6kW with a flowrate of 7.45LPM at a dT of 5C, 30C in and 25C out of the loops, the TMV will then supply 0.93LPM at 65C which mixed with a UFH return of 6.52LPM at 25C will give the required UFH inlet temp of 30C, (0.93*65)+(6.52**25) = (7.45**30), so the boiler must have a flow and return of 0.93LPM.
Can you or someone just draw a schematic showing how the subject UFH setup achieves this, its probably quite obvious but I just can’t figure it out just now.
The TRV opens and closes based on the temperature the probe reads. If below the set temperature, open a bit more, if higher than the set temp close the TRV.
Or am I missing something?
Your diagram is missing the temperature probe which is the control input to the TRV. The TRV is not like a normal Radiator TRV which works off ambient temperature, this one works off the probe temperature (I highlighted above).
Should have shown the capillary if one is used but most just sense the water temperature in the TMV which is why I omitted it.
Won’t post again now until I see the UFH set up in question and take some readings.
I reckon it’s like this:
In general, the ‘UFH’ water will be re-circulated around the three UFH loops but the Thermostatic valve will allow a bit of hot boiler flow to mix-in to maintain a target UFH flow temperature as sensed by the capillary tube.
Yes, that sounds good, even though its not the UFH return water, I thought the bypass might be flowing downwards but probably not as the UFH manifold pump is closer to the TMV. The TMV is a Ivar FM 750 Flow Mix valve.
Brian @borpin was also on the right track, and obviously it doesn’t have to be the cold UFH outlet that returns to the boiler.
UFH with TMV Bypass.pdf (143.5 KB)
UFH with No TMV Bypass.pdf (133.5 KB)
I’m not getting any sensible search hits for “Ivar FM 750” but I found this document which seems to describe something very similar to what’s in your photos: https://amativeikals.lv/content/visiem_MR_0.pdf
If the valve is behaving the same as the MR 01 described in that document, it aligns with my schematic (with the “bypass” flow vertically upwards to the valve).
This is the info that I have.
Ivar Flow Mix.pdf (807.8 KB)
I’m afraid I’m more or less back to my misgivings about how this will work because whatever the UFH output is and its dT, then, because its the mixed flow (UFH inlet) thats going back to the boiler, as per my schematic (again) there will never be a condition where any inward
UFH with No TMV Bypass.pdf (133.5 KB)
bypass is required so how will that TMV behave under these conditions, should the bypass be blanked off or what?, if not will it pull some of the mixed flow water in, leading to unstable temperature control, there may be some throttling/shut off on that FM 750 Flow Mix valve?.
As Brian highlighted, the thermostatic mixing valve will be responding to the temperature sensed by the (remote) Probe - i.e. the Output from the Pump, which is the Input to the UFH loops. The valve will therefore act to maintain a constant 30C going into the UFH loops, mixing boiler water with by-pass water as required - which sounds like exactly what you want: a constant, moderate flow temperature going into the floor slab.
Yes, interesting to say the least, boiler water at 65C is mixing with UFH inlet water already at its required temperature of 30C, then this mixture is further mixed with the UFH return at 25C, these three mixes are then maintained to 30C by the probe/TMV.
Another way of looking at it is that the probe is sensing how much the UFH flow water has been cooled by dilution with the UFH return (after everything has been thoroughly mixed by the pump) and it’s signalling the thermostatic valve to supply the right amount of boiler water to bring the flow to the target temperature.
The same manifold design will be used where there are individual actuator heads on the three UFH zones, with those zones being cycled on and off based on room thermostats or suchlike, making all the flows and temperatures much more ‘dynamic’. My hunch is that Ivar optimised the manifold design for that more dynamic scenario, making it seem overly-complex for your simpler case.
1 Like
Ah yes, I see what you mean. However, I think it works OK. I think the bypass is there in case all the actuators are closed but the pump continues to run and the boiler pump is off. There is a risk that the bypass has very low restriction so the flow goes round that instead of the UFH loops, but I think at the bottom of the bypass is a valve to prevent that.
Do the loops show some flow on the flow gauges (they also control the flow rate)? They do stick sometimes, but winding in and out again fixes that. You can ‘pop’ off the red collar to adjust them.
As long as you can see flow on the flow meters, all is well.
1 Like
