Is it best to have UFH within the Insulated slab or in a screed layer above

[yellowbert]

I know this has been answered for similar questions but wanted to ask it in a slightly different way as I can’t find an exact answer to my query.

 

We were initially expecting to have a suspended floor, but Structural Engineers have now confirmed they are happy for a ground bearing solution after receiving our soil test results.

 

The build-up that has been suggested by the architects is, from bottom up:

·         Approx 300mm EPS Insulation (approx. U = 0.1)

·         Reinforced Concrete (either as a standard slab or possibly a raft foundation)

·         Thin layer of insulation

·         Unbonded Screed with UFH pipes embedded

(Base layer & DPM omitted in description to focus on thermal aspects)

 

My questions are:

1.      Should we be embedding the UFH pipes within the unbonded screed as in the build-up above, or should we do away with the screed layer and place the UFH withing the concrete as many on this forum have done?

2.      Does that answer change depending on whether a standard concrete slab is used with a strip foundation vs a raft foundation.

 

We are talking to the architect again next week so will hopefully have a better understanding of why they have suggested this build-up and it would be great to get your thoughts so we have some info going into that meeting.

For info - the build is part renovation and part extension aiming for very low heat loss EnerPhit standards with new insulated ground floor everywhere.

 

Thanks everyone

[JohnMo]

How thick is the reinforced concrete and how thick is the proposed screed?

[yellowbert]

Thanks @JohnMo

 

Up until about 5 mins ago - we didnt know - but structural engineers have just given us the concrete thickness as below.

 

Concrete Thickness:

If using a Concrete Slab = 150mm

If using a Raft Foundation = 300mm

 

In theory both options are open - I think the architect would like to go with the raft as they can reduce the thermal bridges to almost zero, but as it will almost definately cost many times more than the standard concrete slab, its likely that we will probably be forced to have the standard concrete slab.

 

In terms of the screed layer - I'm not sure how thick that would be, but I'm guessing as this will be non structural it would be in the region of 50mm to 75mm max. The architect mentioned the screed could possibly be seperated from the slab/raft with another layer of insulation although they haven't yet indicated what thickness that would be if used.

 

[Nick Laslett]

@yellowbert, thank you for using tags in your topic. 👍
 

There are always a lot of site and structure specific requirements that need the structural engineer’s sign-off. 
 

There are many different solutions to foundations, UFH, etc, no one right way. 

 

When I was first on BuildHub researching my build, I came across many posts advocating insulated raft foundations. *Raft/Slab these terms do appear to mean different things to a structural engineer, lets just park that for this post. 

 

What I liked about this approach was how simple it was. As a layman it was easy for me to understand how it worked and how it was constructed. 
 

I felt that an insulated raft foundation with embedded UFH gave you a finished floor in a lot of quick steps. It removed the need for a separate insulation step and screed step. It reduced the moisture going into the build, which a screed step introduces. The larger the mass containing the UFH, the more gradual the heating curve, whereas a thin screed layer will heat and cool quickly. An ASHP is more efficient giving a low heat for long periods, which suite UFH in a larger mass. 
 

Removing the screed and second insulation layer should give you more room height. 
 

One advantage of the screed layer is that it can put right any issues in the level of the raft. 
 

Here is a useful thread discussing insulated raft foundations, with contributions from lots of members that have an insulated raft. Just to confuse things even more, insulated rafts are also called passive slabs. 

 

 

Edited June 19 by Nick Laslett

[JohnMo]

Both the raft and slab support the low and slow heating and they can do batch charge or storage heater mode well. Response will be slow due to the number tonnes of concrete.

 

I see zero advantage to adding a thin layer of insulation then adding screed on top. For me the thin screed will allow heat to leach into the concrete below and it would never get back to the house.

 

With a little care a strip foundation can be made thermal bridge free also. It's also easier (certainly where I am) to get a contractor for strip foundations, but you do have to watch over them.

Edited June 19 by JohnMo

[Nick Laslett]

1 hour ago, yellowbert said:

For info - the build is part renovation and part extension aiming for very low heat loss EnerPhit standards with new insulated ground floor everywhere.

Just re-reading your first post, and this quoted section jumped out at me. My experience is with one new build. Your project will need to take into account the existing foundation and the new foundations and their relationship. 

Edited June 19 by Nick Laslett

[yellowbert]

18 hours ago, Nick Laslett said:

Just to confuse things even more, insulated rafts are also called passive slabs

@Nick Laslett - Thanks Nick - Yes, I think the architects are favouring this option - although a standard concrete slab ground floor with strip foundation has also been approved by the SE - so essentially both options are still on the table. Thanks for the link regarding the passive slab - thats super useful.

 

17 hours ago, Nick Laslett said:

Your project will need to take into account the existing foundation and the new foundations and their relationship

This in particular would definately be very important - the existing house has only brick footings with no concrete foundation at all. If we go for the strip foundation and slab - then the new foundation would be at the same approximate depth as the existing footings which is good - whereas the Raft would end up being deeper.

 

18 hours ago, JohnMo said:

I see zero advantage to adding a thin layer of insulation then adding screed on top. For me the thin screed will allow heat to leach into the concrete below and it would never get back to the house.

 

@JohnMo - Yes, from what I've read so far, this was exactly my thoughts - so it will be useful to understand what the architects were intending with this - We have a meeting next week so I'll hopefully be able to update things then 👍 If they suggest keeping a screed layer to ensure having more control over FFL, I guess it would be fine to have the UFH within the concrete slab then the screed layer on top without the intermediate layer of insulation?

[JohnMo]

Yes but you are then spending needless money etc. getting screed done. Just bring concrete up to FFL - done. Make sure UFH don't go under where internal walls are, take all pipes through doorways. Then you won't get any nasty surprises.

 

Architect will take lots of rubbish about heat up times when buried in deep floors, reaction time, which is only relevant when running against a thermostat I suspect.

[jack]

2 hours ago, yellowbert said:

so it will be useful to understand what the architects were intending with this

 

Most likely they're proposing this because that's how it's generally been done. Stuctural slab (no insulation underneath), insulation (and historically not much of it), then a screed on top. This works well in a poorly insulated house, because you probably can't afford to run the heating all the time given the energy losses. You therefore want to be able to heat the top surface of the floor quickly when the heating is on. This is the same paradigm as using radiators in poorly insulated houses - blast with heat for a couple of hours in the morning and the evening. You get fast heat-up of rooms, followed by fast cool-down as the heat escapes and is absorbed into the walls etc.

 

With decent underslab insulation, the slab surface stays at similar temperature throughout the day, irrespective of whether the heating is on or off. In winter, our slab temp varies by a fraction of a degree over any 24 hour period. If I turned all the heating off in the middle of winter, we'd probably lose a degree a day.

 

In short, with good insulation levels, using the structural slab as a heat buffer makes a lot of sense.

 

2 hours ago, yellowbert said:

If they suggest keeping a screed layer to ensure having more control over FFL, I guess it would be fine to have the UFH within the concrete slab then the screed layer on top without the intermediate layer of insulation?

 

I have polished concrete floors with this exact arrangement - UFH pipes in structural raft slab, and a ~65 mm screed over the original slab, with just a slip membrane (basically thick plastic sheet) between them. No issues at all. The extra concrete might even help with buffering. 

 

[Nickfromwales]

2 hours ago, JohnMo said:

Architect will take lots of rubbish about heat up times when buried in deep floors, reaction time, which is only relevant when running against a thermostat I suspect.

Yup. Have dealt with many like this over the years, and there's only a few who actually understand this. Clients often find OK architects vs great ones, but it's very refreshing when one of these unicorns is discovered.

 

2 hours ago, JohnMo said:

Just bring concrete up to FFL - done. Make sure UFH don't go under where internal walls are, take all pipes through doorways. Then you won't get any nasty surprises.

Exactly what we did for a client in Oxford, with 100mm constructional slab which had 200mm PIR over B&B founds. That slab was polished concrete with iirc 2 layers of anti-crack mesh. No fibres as I believe these cause probs when polishing (again iirc).

[saveasteading]

52 minutes ago, Nickfromwales said:

No fibres as I believe these cause probs when polishing 

Polishing for an industrial finish: no problem.

For an architectural finish: no. You will see plastic fibres.

 

[yellowbert]

On 20/06/2025 at 11:49, JohnMo said:

Yes but you are then spending needless money etc. getting screed done. Just bring concrete up to FFL - done.

@JohnMo - Absolutely - and making the required excavation deeper with more cost there too. I've spoken to the architects today and they are happy with using just the structural concrete without a screed. It looks like their initial spec was to provide a more rapid response.

 

On 20/06/2025 at 13:40, jack said:

In short, with good insulation levels, using the structural slab as a heat buffer makes a lot of sense.

@jack - Architects are totally happy with this now - and I feel like this is definately the best solution for us with Insulation and airtightness at Enerphit levels.

 

 

Thanks also @Nickfromwales - Although I suspect we are going to go for a tiled finish mainly due to cost, Polished concrete is still an option if its affordable and great to hear that it works well with this setup.

[yellowbert]

Thanks everyone for all the helpful input. It looks like we are going with a standard solid concrete ground bearing slab with strip foundations as long as the SE doesn't have any additional worries with this setup.

 

It does look like we might need to locally reduce the amount of EPS insulation where we need to run a soil pipe but hopefully this wont be an issue. @Nickfromwales - I notice that there was PIR rather than EPS in the build-up you mentioned above. Are there any additional considerations or precautions needed if using PIR below the slab? This would possibly help us to reduce the thickness and possibly remove the need for running the soil pipe within the insulation layer.

[Nickfromwales]

35 minutes ago, yellowbert said:

Thanks everyone for all the helpful input. It looks like we are going with a standard solid concrete ground bearing slab with strip foundations as long as the SE doesn't have any additional worries with this setup.

 

It does look like we might need to locally reduce the amount of EPS insulation where we need to run a soil pipe but hopefully this wont be an issue. @Nickfromwales - I notice that there was PIR rather than EPS in the build-up you mentioned above. Are there any additional considerations or precautions needed if using PIR below the slab? 

You'll need closer to 300mm of EPS to get the same performance of 200mm PIR, so that's the difference (in a nutshell).

 

You'll need a thin membrane on top of the PIR if you have foil-faced stuff, to stop the concrete reacting with the alu face.

 

53 minutes ago, yellowbert said:

This would possibly help us to reduce the thickness and possibly remove the need for running the soil pipe within the insulation layer.

That bit makes no sense to me sorry? If you need a soil pipe in there then you need the thickness to accommodate it?

[G and J]

Presumably that means the thinner PIR insulation layer means the oversite is higher and thus can accommodate your poo pipes.   We’ve very tight invert levels so I’ve been round and round and round and round this one. 

[JohnMo]

11 hours ago, Nickfromwales said:

You'll need a thin membrane on top of the PIR if you have foil-faced stuff, to stop the concrete reacting with the alu face.

You may also want that with eps, to help eliminate any chance of concrete getting below the insulation and floating it, especially if using a concrete pump.

 

12 hours ago, yellowbert said:

locally reduce the amount of EPS insulation

If you paying to have soil removed and taken away from site, you may be more economical to use PIR as you are digging down 100mm less for the same U value. Plus you have less faff to do with waste pipe heights etc, without compromising insulation.

 

12 hours ago, yellowbert said:

standard solid concrete ground bearing slab with strip foundations

So your next issue is to get that thermal bridge free. Which isn't difficult or expensive with a little thought. 3 areas to look at, external walls to ground, finished floor to ground and internal structural walls to ground.

[yellowbert]

12 hours ago, G and J said:

Presumably that means the thinner PIR insulation layer means the oversite is higher and thus can accommodate your poo pipes.

@G and J - Thats exactly it - The SE had indicated that we were limited by the invert levels. Ideally the poo pipes will go down and then run completely below the insulation so we dont have to compromise on having a thinner layer of insulation over the pipe itself - PIR could definately help us here too.

 

 

On 20/06/2025 at 14:21, Nickfromwales said:

Yup. Have dealt with many like this over the years, and there's only a few who actually understand this. Clients often find OK architects vs great ones, but it's very refreshing when one of these unicorns is discovered.

We have been super lucky with the architect. Originally we had a not so great one, which didnt last very long. Our architect now is excellent and very knowledgable regaring the passivehaus side of things in particular - The advice for quicker response times had come from one of the UFH manufacturers directly - interestingly because they specialised in those kind of quick response systems.

[SteamyTea]

Would the losses of having a small area of reduced insulation make much difference, thermally.

As it is a pipe, there is only a very small area on the upper facing area, and it is mainly full of air, which is a good insulator.

Try modelling it and see what the numbers say.