Think of the floor as a very big radiator. It's cooler because it has a bigger surface area.

Output is a combination of dT between room and floor surface temperature and surface area. As it gets colder outside you need more heat from the emitter, the only variable you have is surface temperature, to get this to rise you increase flow temperature. 

Output of the floor is also proportional to room temperature and floor temperature. If the floor is 21 and room is 21 there is zero output. As the floor temp increase in relation to the room temperature, output increase. If the floor is cooler than the room heat from the room is absorbed by the floor.

How do people find switching from DHW to space heating in respect of the flow and return temps?  For example, going from a 52C flow (for 47C DHW temp) to a, say, 25C flow for UFH - assume the outdoor unit should stay off whilst the circulation pumps move the water round until its temperature goes below the flow setpoint (of 25C - lower limit of HP thermo diff adjust) then the compressor starts?  Is that what happens in practice or is there an amount of cycling until the FTC works out what's happened?

This may seem like a daft question but can I please check - one should always aim for their systems to have a return temp of <target temp> + dT + hysterisis delta?

Asking to avoid wasting days raising the flow temp 1C at a time to determine lowest value as OAT is changing...

Posted by: @johnmo

↑

Watching the discussion on dT. An observation.

As you add heat to the floor the return temperature will rise, the heat pump should let this occur and add more heat to the flow, keeping your dT stable. Once dT and target temp are at point where your settings say is correct, the heat pump should modulate down to keep target and dT stable. If your heat required is below min modulation, the dT will get smaller, and the flow temperature increase, until flow hits your upper limit. Mine can come down to about 3 dT.

This is exactly what I observe with my oversized Samsung when I run it in always on mode via the room stat calling for heat.

If I set the WC curve to start at 30C, my radiators are unable to emit the heat produced at that low flow temperature (they can only emit ~2.5kW and the heat pump is generating ~4.5kW), so the return water temps continue to rise, the unit modulates down to minimum output and eventually the system settles (reaches equilibrium) at around 35C flow temp and 32C return (dT of 3C), whereby the mid point of flow and return (33.5C) correlates exactly with the temp at which the radiators are capable of emitting the amount of heat the heat pump is producing. Effectively the heat pump is telling me that for my setup, it cannot run a flow temp below 35C as what it is connected to cannot emit the generated heat (the heat has to go somewhere else the return temp will just continue to rise*). So the upper (ambient) end of my WC curve is fixed at 35C flow temp due to my radiators, and I have the lower end of the WC curve (-2C ambient temp) set to 40C flow temp (initial estimate based on heat loss figures, which I can tweak as the first winter progresses).

* The heat has to go somewhere else the return temp will continue to rise has two considerations - firstly the emitter needs to be capable of emitting the heat into it's surroundings (radiator and/or UFH sizing), and secondly the environment needs to be capable of accepting the heat output (room heat loss). If neither condition is met, return and hence flow temps will continue to rise until conditions are met.

Posted by: @rhh2348

↑

How do people find switching from DHW to space heating in respect of the flow and return temps?  For example, going from a 52C flow (for 47C DHW temp) to a, say, 25C flow for UFH - assume the outdoor unit should stay off whilst the circulation pumps move the water round until its temperature goes below the flow setpoint (of 25C - lower limit of HP thermo diff adjust) then the compressor starts?  Is that what happens in practice or is there an amount of cycling until the FTC works out what's happened?

@rhh2348 My system has a 50L volumiser fitted in the main 28mm return pipework, which at the end of the DWH cycle will also contain 50L of hot (55C) water, so this effect is even more amplified in my system.

It's simply not an issue for me (I have rads, not UFH). When the system switches back to heating, I get a quick slug of hot water hit the radiators (you can feel the 55C water hit the radiator), and it's completely gone within ~1-2mins - dissipated into the system and absorbed by the room. Within a couple mins the flow/return temps are back down to target. My heat pump compressor still runs during this time but the heat produced appears easily absorbed by the room.

Looking at my radiators spreadsheet, I see that at a mid point of 50C, my radiators are easily capable of emitting over 12kW of heat output, and during this switch over period, the heat pump compressor is likely running quite low, and maybe only putting out 4-5kW heat, hence my observations above that the sudden injection of hot water into the radiators is very quickly dissipated into the house.

Without a volumiser in the DHW circuit, the volume of hot water entering the heating loop would be so small that I doubt it would be significant.

I don't have any measurements to back this up, only my own observations, but within a couple mins at most the radiators feel like they are back to their normal 35C.

I do not know if there are any special considerations for UFH and how that may react to a sudden injection of hot water, or if there are mixing values fitted to prevent such from happening.

I thought volumisers were in the heating loop and not the dhw?

Our ufh floor covering has an absolute maximum temp of 29C for the glue on the tiles so a slug of hot water would be bad news, I suspect it would blow the glu at the input nearest to the manifold. All managed by a mixer to reduce the temperature and not too inefficient since the ufh is <10% of the heat load.

Posted by: @judith

↑

Our ufh floor covering has an absolute maximum temp of 29C for the glue on the tiles so a slug of hot water would be bad news,

You need a lot of energy to heat what I assume is a screed floor. A slug of water will have a little or no effect by the time the heat as dissipated vertically and radially out from the pipe, especially if a cooler slug of water comes along after.

Posted by: @judith

↑

I thought volumisers were in the heating loop and not the dhw?

That would be the more logical place to fit them, but ours is installed on the main return pipe back to the heat pump, so affects both DHW and heating circuits. Not ideal.

Hi @rhh2348 I have a question on your underfloor setup.

This may seem like a daft question but can I please check - one should always aim for their systems to have a return temp of <target temp> + dT + hysterisis delta?

Feels like this was way out after some of the replies, however are there any rules of thumb on such matters - for example, given UFH in an area of Xm2 at a depth of 100mm and OAT of Y it will take Z kW/J to raise the temperature 1C/K at a flow temp of A degrees C?

 

Asking to avoid wasting days raising the flow temp 1C at a time to determine lowest value as OAT is changing...

The floor you’re describing above, assuming the 100mm thickness is a concrete screed, would be an ideal efficient emitter. Giving a depth of low output mass storage however spacing of the pipe runs according to heat geek need to be around 100mm to maintain optimal heat transfer. This spacing would presumably need to be part of the above calculation.

i don’t have an answer for the maths but I do have a question. 

What is the daily room/house temperature fall of such an Under Floor heating setup if the heat pump was switched off? (assuming a known ambient and modern construction)

I ask this because I believe a screed floor has a much lower temperature drop say on an overnight setback and may be as low as only 1c overnight. 

I thought that if you are describing your own floor you might have an idea of it’s thermal properties and how it goes on releasing heat.

TIA

Posted by: @rhh2348

↑

How do people find switching from DHW to space heating in respect of the flow and return temps?  For example, going from a 52C flow (for 47C DHW temp) to a, say, 25C flow for UFH - assume the outdoor unit should stay off whilst the circulation pumps move the water round until its temperature goes below the flow setpoint (of 25C - lower limit of HP thermo diff adjust) then the compressor starts?  Is that what happens in practice or is there an amount of cycling until the FTC works out what's happened?

With UFH, no buffers or volumisers. Once you switch from DHW, the circulation pump runs until the return temp is within limits to allow restart. This will take a few minutes to 30-40 mins depending on outside temp and house temp. I see zero cycling, as the compressor restart hysteresis (set by the manufacture) will prevent this.

Posted by: @sunandair

↑

Giving a depth of low output mass storage however spacing of the pipe runs according to heat geek need to be around 100mm to maintain optimal heat transfer. This spacing would presumably need to be part of the above calculation.

My UFH pipes are at 300mm centres and 100mm concrete screed. No issues. 

Posted by: @sunandair

↑

I ask this because I believe a screed floor has a much lower temperature drop say on an overnight setback and may be as low as only 1c overnight. 

I did plenty of experiments with my floor, believing setbacks were the thing to do. When I got back to before lunch to activate the setback. Found I could switch the heating off at 7am and restart at midnight and only get a 1 deg setback at night, the 7 to 8 hrs charging the floor was ok the next day and so on.

But all the above depends on insulation under the floor and in the house and general heat losses.

Posted by: @johnmo

↑

My UFH pipes are at 300mm centres and 100mm concrete screed. No issues. 

....

I did plenty of experiments with my floor, believing setbacks were the thing to do. When I got back to before lunch to activate the setback. Found I could switch the heating off at 7am and restart at midnight and only get a 1 deg setback at night, the 7 to 8 hrs charging the floor was ok the next day and so on.

It’s clear from what you say, above and what we have experienced in our own system that radiator systems behave very differently to screed laid under floor heating systems. Our radiator heat loss is relatively quick but also our reheat in the morning is also relatively quick. (This time of year a 9pm setback might give us a 2 degC drop by 4am but later in the year the same setback period might give a temperature drop of 4 degC.) And, Conversely when the reheat starts on a low and slow heat up it might take only 3 hours to regain a 21degC room temp.

So this would possibly confirm under floor heating, done properly is a more efficient way to use heat pumps with only marginal benefits for operating a setback. Whereas retrofit radiator systems appear to be less efficient but can recoup with energy savings and quicker response times from night time setbacks. 

What do you think @johnmo?

Posted by: @sunandair

↑

...spacing of the pipe runs according to heat geek need to be around 100mm to maintain optimal heat transfer. This spacing would presumably need to be part of the above calculation.

Assume so too.

I think my pipes are at 150mm centres, which I understand is common for screeded underfloor - but I watched the HG video and (mainly!) understood why 100mm is best.

What is the daily room/house temperature fall of such an Under Floor heating setup if the heat pump was switched off? (assuming a known ambient and modern construction)

I ask this because I believe a screed floor has a much lower temperature drop say on an overnight setback and may be as low as only 1c overnight.