Wet UFH in 250mm insulated reinforced raft

[Smallholdertoo]

We are due to build a small very well insulated 1.5 storey timber frame house later this summer. Using the Jeremy Harris spreadsheet, the house will have a thermal demand of under under 2kw when it is -10c outside . U values of 0.09, 0.13  and 0.10 for the roof, walls and floor respectively.

 

The house has passed Part O for overheating and will have MVHR.

 

We are now definitely going to have an85m2 insulated reinforced raft.  I want to have ufh in the raft and in line with what others have done on this site it will have 3-4 loops, a singe zone and one thermostat. We will have a small ashp perhaps 3.5-5kw. I expect water temperatures in the ufh will be under 25c. The ashp will also provide our hot water - a 250l unvented cylinder.

 

I keep on being given the advice that a 250mm slab will be very slow to heat or cool e.g. from my designer "My concern with a heated slab of 250mm thick is that the response times will be so slow that you will find that you are unable to adjust internal temperatures when the house heats up to the required temperature and then continues to heat when you don't want it because of the thermal mass."

 

I am aware that the raft will be slow to warm up and cool down but am I right to believe that with the weather compensation a single thermostat and a low water temperature I am unlikely to have the issues suggested re overshoot?

 

It would be great to hear from those on the forum that have thick rafts, ufh in raft and a well insulated house .

 

Thanks in advance

 

 

 

 

 

 

 

 

[Nick Laslett]

Very quick reply as it is past bedtime. It sounds like you have it all covered and well understood. 
 

The ASHP with weather compensation handles the temp variance. It will take awhile to dial in the best compensation curve. So in the early days the house might not be at quite the right temp. 
 

The heat changes are very gradual, but that is actually its strength. A well insulated, air tight house is nothing like traditional houses. In my build you don’t even think the heating is on, the floor is cool to touch. But somehow the house at 20° does not feel cold, whereas the 90’s built rental house feels freezing even when the room temp is 24°. 
 

Remember, even though you have MVHR you can still just open windows to quickly change the room temp. It will have very little impact on the floor or ASHP in the short term. You are not losing control of the heating, you don’t need to worry about the response time. 
 

Just one extra thought, 250mm raft sounds very thick, my ring beam is 250mm but the main raft is 150mm. I have a very heavy ICF house, concrete stairs and concrete first floor, I would have thought you only needed 100mm raft for a timber frame house. 

[JohnMo]

6 minutes ago, Smallholdertoo said:

expect water temperatures in the ufh will be under 25c.

Unlikely as an ASHP normally will supply a minimum of 25 Deg in heating mode. Then at deltaT 4, plus restart hysterisis for the heat pump, you need the floor temp to drop below 20. If that's the case your floor would provide zero heat to the house. Tried that with mine (25 Deg flow) it ran fine, but never restarted, floor just never cooled enough to allow ASHP to restart.

 

Thermostat you need a hysterisis of 0.1 or less if you can one. It has to be on/off not TPI.

 

I have 100mm concrete with UFH pipes at the bottom - I am also 300mm pipe centres, response is slow. But if you are prepared it's something you can use to an advantage. But don't expect setbacks to work, they won't.  

 

WC works fine.

 

If you use a tou tariff, get a bigger heat pump and batch charge the floor storage heater mode. I am at a out 3kW at -9 and have a 6kW ASHP, will run without stopping buffering into the floor if I want it to.

 

If your heat loss is 2kW, you need 48kWh of heat, if you only have 7 hours to heat you need 7kW of heat input, plus it may need defrost or modulate down to manage dT, so even 8kW would be fine.

[Nickfromwales]

I did one that had a massively over-engineered slab (I think I talked the buffoon in charge down from nearly 300mm with a screed for the UFH down to 230 with the heating pipes inboard of the raft). Bonkers. This was a 1.5 storey ICF build, timber floors.

 

The client has a 7kW heat-pump running UFH with 120mm c's; the idea of the tighter laid pipe loops is so that they can use the additional water volume and additional cross sectional area of pipe > concrete (surface contact area) to quickly chunk heat into the slab during pockets of cheap electricity, but more importantly to allow cooling to be done at a higher temp (to stay WELL away from the dewpoint).

 

Is in and working well, but it's a huge thing to get to a set temp. Once it's there, it's very comfortable, just needs the stat to have the aforementioned tight hysteresis to prevent unwanted under / over shoot.

 

Seriously consider the extra volume of pipe & water as it can reduce or negate the buffer tank  

[Smallholdertoo]

Thanks for the comments and the support.

 

The SE is yet to confirm the thickness though he has mentioned 250mm in the past. If it does turn out to be 250mm any thoughts how to persuade him to let it be thinner?

 

Its probably worth mentioning that in readiness for the build we have installed 11kw of PVs and 20kw of battery storage on a new shed. This will provide the power to the new house. would this affect your choice of ASHP

[kandgmitchell]

47 minutes ago, Smallholdertoo said:

If it does turn out to be 250mm any thoughts how to persuade him to let it be thinner?

One issue I had with our raft which was 250mm deep was getting the correct cover to the steel plus the thickness of the main bars, plus the thickness of the strirups which fit around the main bars, then the space between the upper and lower main bars, the bars and strirup depth again and finally the bottom cover. It's surprising how soon all that uses up the depth of the perimeter insulation on site that you bought compared to the SE's drawing! 

[Nickfromwales]

5 hours ago, Smallholdertoo said:

Thanks for the comments and the support.

 

The SE is yet to confirm the thickness though he has mentioned 250mm in the past. If it does turn out to be 250mm any thoughts how to persuade him to let it be thinner?

 

Its probably worth mentioning that in readiness for the build we have installed 11kw of PVs and 20kw of battery storage on a new shed. This will provide the power to the new house. would this affect your choice of ASHP

Best to check if that’s the overall thickness, but yes, I’d persuade them with scores of examples of how this has been done elsewhere, successfully; the caveat is that I don’t know exactly the ground conditions at this time, so that will be key to the development of a foundation design and any further ‘argument’ here from me.

 

If the ground is difficult or you’re going over made up ground / an infilled old swimming pool etc, then the maths all change really quickly!

 

4 hours ago, kandgmitchell said:

One issue I had with our raft which was 250mm deep was getting the correct cover to the steel plus the thickness of the main bars, plus the thickness of the strirups which fit around the main bars, then the space between the upper and lower main bars, the bars and strirup depth again and finally the bottom cover. It's surprising how soon all that uses up the depth of the perimeter insulation on site that you bought compared to the SE's drawing! 

Yup. On the project with the thick slab I mentioned I had to work with said buffoon and his SE to change the pins / stirrups etc, eventually settling on an agreement for the client to bear the cost of cranking the pins on site. 
 

The reason for this is as you say, adding all this together, and then having to preserve adequate concrete cover over the top of the UFH pipes. 
 

On a computer screen it all looks simple and straightforward, but when you’ve done a good few of these you soon realise that overlapping mesh sheets, bars etc soon leaves you way too high up (near to TOC). 
 

I pushed on that one as the SE wanted the giant slab and then the UFH in another 75mm of dry screed on top of that!?! 😯. I said “no” to that, but the builder was clearly a bit of a bully and tried to get me thrown off the job in retaliation. The clients wife pulled me to one side and asked if she should sack the architect and SE, lol, I said no we’ll work with them and I’ll just keep pushing back where appropriate. Saved the client around £10k in unnecessary concrete volume / screed costs etc, plus having to dig further down to get TOC back to the original house datum (infill project so ridge height couldn’t change by even an inch).

 

House got done, builder went on his merry way the second he’d fulfilled his foundation and frame / roof program and framing, and has since gone bust spectacularly.

 

Too many people out there who can’t admit when they’re out of their depth, or, most annoyingly, are simply too set in their ways or too stubborn to ask for support / look at solutions; mostly because they see downtime as a loss and want the easiest route to the next payment.
 

Better to do what I do, eg explain to the client that there are possibilities to explore for a better outcome / to reduce costs and time, but the time to discuss it needs to be paid for. Most agree and see the value, but some are just too worried about upsetting the apple cart and just choose to ‘accept it’. 🤷‍♂️

[saveasteading]

1 hour ago, Nickfromwales said:

giant slab and then the UFH in another 75mm of dry screed on top of that!?!

My thoughts.

questions first.

Where is the floor insulation?

Is the ground levelish and good bearing capacity?

 

Assuming the ground is good, I would lay harcore and 150mm slab. 175 maybe .

Crack control mesh is all that is needed.

This suffices in warehouses and commercial vehicle garages so is ok for your sofa.

Lay PIR on that with ufh on top, then screed in 60mm poured or hand laid.

 

If you need more capacity for internal loadbearing walls then it can either have a thickening in the slab, or a trench footing, as the ground dictates.

 

I think rafts are specified too often. I really don't like structural slabs on top of insulation.

 

Now some other points.

 

85m3 isn't much and I will stab at it being about 8m x 10m.

I'd have to check but this is close to not needing any contraction joints for a finished surface.

But with a screed over it that is irrelevant because the slab will  crack well before the rest of the work, and will not reflect through the screed.

 

Even as a finished surface this cracking would be trivial.  With careful mix control (absolutely don't allow extra water added on site) and fibre additive (£30 or so) any cracking will be microscopic.

Cracks would be tiny and many. The UFH pipes will likely debond at cracks but are also well capable of stretching the 1/10th mm required.

 

I did put UFH into a 175mm slab once, as the  client's specific requirement. (It was a factory and this was the area where people stood to do craftmanship stuff.)

It had anti-crack mesh in the floor, and the UFH pipes were fixed on top of it.  Shrinkage joints were either the old fashioned method of pouring alternate long strips, or crack inducers built in....I can't remember. So there was no cutting into the slab afterwards. More mesh reduces crack width if necessary, which it wasn't in this case.

I guess the shrinkage cracks were about 1mm.

It went well and worked welI. I heard no more about it so it must  have been ok.  I had a further job for them so it's not as if we lost touch either.

It was an uninsulated for various reasons I won't divert into.

 

btw I've done 200,000 m2 of slabs and never more than 175mm  when ground bearing. 

 

So there are choices.  

Even the best SEs don't know all the ins and outs of costings (it isn't just the  slab but the walls, the insulation , UFH , access to site)......That needs a close collaboration between contractors and consultants.

The biggest and best contractors will know, if they have an SE inhouse, or have developed a standard method.  But that will cost you in either overheads or conservative design.

Small contractors may well know, but that is rare, and with some, the less they know, the more they think they know. BUT perhaps they and your SE together can optimise this.

 

If there is a reason against my suggested method, then please say.

 

 

 

 

 

 

[saveasteading]

I may have mixed up 2 current discussions!  take what is relevant.

 

[Nickfromwales]

8 minutes ago, saveasteading said:

My thoughts.

questions first.

Where is the floor insulation?

Is the ground levelish and good bearing capacity?

All to be decided at the outset I imagine? 
 

Insulted raft is always my first choice, after that a strip foundation with passive levels of insulation in the floating infills between strips, then block & beam I guess (as long as it’s got at least 175mm of insulation and the bare b&b has been thoroughly grouted and sealed to prevent air rising up through it.

 

Raft is very much worth the effort afaic, and they can show go down in conjunction with piles on difficult clays / trees nearby etc. 👌.

12 minutes ago, saveasteading said:

I think rafts are specified too often. I really don't like structural slabs on top of insulation.

Odd you think that tbh, I think they’re an excellent choice, and have many more merits. Heating a thinner screed etc, over insulation over a slab, doesn’t allow much heat to be retained so more sporadic heat needs putting in during winter.
 

I much prefer the idea of a massive storage heater running 24/7 ‘long & low’ and the feedback from clients says a lot. The floor seems cold / cool to the touch but the whole house is a constant comfortable 19.5/20.5°C. Much harder to achieve the same with a screed tbh, as these tend to over and under shoot the stat set points.

[saveasteading]

28 minutes ago, Nickfromwales said:

harder to achieve the same with a screed tbh, as these tend to over and under shoot the stat set points.

Now I don't understand that I'm afraid. Sounds a mechanical thing, not to do with the heatsink.

Why should it not be good that a screed  reacts quickly when the weather changes.  It's like 'let's light the fire' evenings..

 

30 minutes ago, Nickfromwales said:

Odd you think that tbh,

It is fundamental I think, that you should not need an exceptionally heavy duty  factory spec floor for a house, then build it on pir or eps.

 

But I am interested. Do you think it isn't that expensive all things considered?

 

I also like that the first slab can be run on and bricks dropped on it, and scuffed, knowing it will be covered.

[Canski]

3 hours ago, Nickfromwales said:

On a computer screen it all looks simple and straightforward, but when you’ve done a good few of these you soon realise that overlapping mesh sheets, bars etc soon leaves you way too high up (near to TOC). 

During my 8 years over in sausage side we used to get this problem on a weekly basis.

Why oh why do they not remedy the problem at the design stage so it can be built as per the design ?

A bit like imperial window sizes in metric builds. 🤯

[Nickfromwales]

2 minutes ago, saveasteading said:

Now I don't understand that I'm afraid. Sounds a mechanical thing, not to do with the heatsink.

Why should it not be good that a screed  reacts quickly when the weather changes.  It's like 'let's light the fire' evenings..

Trying to ‘change the temp’ morning / evening in a decent ‘passiv level’ dwelling just doesn’t happen. A steed can react quickly because it’s heated hotter for shorter periods, but why does it have to react? Why would you tolerate the hysteresis of cool > hot > cool > hot when you can have comfort > comfort > more comfort 🫡. 

 

If you lit a fire in any one room, that room would instantly overheat and become unbearable, poor analogy in this discussion imo. In a PH you should never be void of heat, therefore you should not need more of it here or there. 
 

I turned the heating off on a Monday, on one project with a big slab like is mentioned here, and the client rang me to complain that they were still sweating their arses off on the Wednesday, (whilst cleaning, sanding and decorating etc), asking why I hadn’t done as they asked? I said go feel the HP pipes, they said “ooh, they’re stone cold!”. Yes says I, and it was turned off Monday evening when I left to go back home. 

Monday saw it at 20.5°, and it was 17.8° on the Wednesday, in the winter. With the MVHR running. They wanted the 15-16° that the house settled to without heating. I said you’ll have to wait another 24 hrs +. 

 

These things just don’t yo-yo heat up and down morning / evening, so having weather comp attempt to manipulate this twice or more a day just seems a waste of effort. Why not just have one happy temp per 24hrs? 

10 minutes ago, saveasteading said:

It is fundamental I think, that you should not need an exceptionally heavy duty  factory spec floor for a house, then build it on pir or eps.

But you ARE saying to have a ‘heavy duty’ slab, (*150-175mm + 60mm of screed = a lot of mass*), just you suggest it should go under the Insulaton 😵‍💫
 

I am saying you don’t need to suffer a buffoon to over spec over size and cost you more than is actually necessary.   

 

11 minutes ago, saveasteading said:

But I am interested. Do you think it isn't that expensive all things considered?

If someone driving this is involved from the get go there’s a bunch of value engineering that can be added in, so with a machine and groundworks contractors around you it’s easy to achieve great things, if someone’s there to instruct them, and then such things can be done far more economically.

 

If it costs more overall, but is a better outcome, then the client has time to decide that when we budget the project out at the concept/ pre-construction phase .

 

You can drive to work in a Ferrari or a fiesta, the client decides who drives what and why.

 

17 minutes ago, saveasteading said:

I also like that the first slab can be run on and bricks dropped on it, and scuffed, knowing it will be covered.

And constructional rafts don’t get covered?

 

Only if the client specifies polished concrete is this ever a consideration or concern afaik.

 

This had only ever been raised as a concern with my ICF guy, who asked how he was supposed to fix the shoring systems for the walls; this was usually fixed down to the slab by drilling and concrete screwing etc, something you can’t do if the UFH pipes are 30 or 40mm under TOC. We resolved this by putting timber down with shorter fixings, and fixing shoring to that.

 

 

Apart from all that we’re 2 peas in a pod mate 🙂

[saveasteading]

13 minutes ago, Nickfromwales said:

we’re 2 peas in a pod mate 🙂

pods have differing peas.  2 varieties of pea perhaps.

 

Forget the fire analogy , I only meant a quick change, not the amount and controllability of heat.

 

18 minutes ago, Nickfromwales said:

constructional rafts don’t get covered?

Not usually in my experience. . Scratches, machine marks, mortar splashes.  But I'm more used to bigger projects than houses, with steels etc. on and off the floor, and mewps running around.

Perhaps different if its very very closely supervised.

 

otoh, with a house, it seems that most builders can't get the slab remotely level, and they are too small for the specialist contractors. so the second chance with the screed comes to the rescue.

 

 

still sweating their arses off on the Wednesday,

Doesn't that show that the heat in the slab was then wasted, and it would take 3 days to warm up again?

 

Bottom line....I have never priced or built a new house with a raft. so it is a hunch. and of course I am interested in he all rounder's view.

But I was an Estimator for years and the hunch is usually about right.

Worth £5k to £10k in my mind. That can be wiped out by consultants fees or a builder with a proven system not wanting to change processes.

[Nickfromwales]

21 minutes ago, saveasteading said:

still sweating their arses off on the Wednesday,

Doesn't that show that the heat in the slab was then wasted, and it would take 3 days to warm up again?

I’m a mushy pea 👍. 
 

That couple were preparing the new house to accept their furniture out of long term storage. They just wanted the house ‘cold’ as they were busying about the place and in coveralls.

 

Nothing wasted there as the 14kWp of solar / batteries > heat pump meant an average cost of heating at < 1.6p/kWh, prob less now that octopus is in the picture, so they could turn it on and off willy-nilly without pooping the bed, but now they’re moved in the heating is on at one constant temperature and comfortable 24/7 without other control or manipulation. 

 

Point being that the heat gets slowly released over a long period of time, with UFH in the insulated raft, which means the heat from the floor is pretty much undetectable.

 

Tres bien. 
 

There are many ways to skin that poor old cat, so your method would work, of course it would, just I have done loads like that and now a load like this, and I prefer ‘this’, so do the clients and their opinion is where I base my feedback upon.

 

We do have to self-level a lot of slabs, but MBC got the last one within 8mm over 140m2, so almost zero levelling needing doing on that one, just a double bed of tile adhesive vs bed and butter.

 

21 minutes ago, saveasteading said:

Bottom line....I have never priced or built a new house with a raft. so it is a hunch. and of course I am interested in he all rounder's view.

100% get that, and that’s why my replies are so detailed / objectionable. Trust me, it works better, and that’s my 2 cents. As above, other methodology will also skin the cat.

 

21 minutes ago, saveasteading said:

But I was an Estimator for years and the hunch is usually about right.

Worth £5k to £10k in my mind. That can be wiped out by consultants fees or a builder with a proven system not wanting to change processes.

Defo, and about spot on for what I’ve seen. The force is strong with you. 
 

But would you like to spend £10k on pumping water out of your boat, or spend £10k once getting the hole plugged permanently? Either option will cost you £10k. 

[saveasteading]

1 hour ago, Nickfromwales said:

Either option

Neither. I would do option 3.

 

btw, I only mentioned slabs 175th as what  a factory needs. A house doesn't need that.

So shave off another 50mm at 100m2, 5m3 @£150.   for 250mm it is 12m3.  but muck away, shutters, and the mesh spec. Rain or frost damage or delay.

Risk of damage to sticking out, loose ends of ufh pipes?

What was your method for fixing the ufh? tie to the mesh? In the factory I mentioned, the building was enclosed and the slab was poured 'indoors'.

 

 

[Nickfromwales]

2 minutes ago, saveasteading said:

Risk of damage to sticking out, loose ends of ufh pipes?

Just saying what I've been around for 20+ years of installing UFH, and there's never been any issue that I've seen or heard of. On site we tie in rebar to make a frame, add some battens to it, fix all the rising loose ends to that (then test or don't test), and foam around the base at TOC to allow some wiggle room when connecting the manifold afterwards.

 

4 minutes ago, saveasteading said:

tie to the mesh?

Yup, loads on here have done the same tbf. Cheap, simple, effective. Just less than great on the knees and back, so the 1st year apprentice or labourer gets the short straw whilst the old guys watch on from the pipe decoiler lol.

[BotusBuild]

We have a 250mm 135m² insulated raft, with UFH loops at 200mm cc's. A Vaillant 7kW ASHP was switched on 10 days ago and we are already at a comfortable 17.5^oC. UFH loops are tied to the A393 mesh.

Why 250mm? SE specified, has retaining ICF walls with an internal ICF wall. By the time I'd thinned the slab in some areas, I figured it would just be quicker and easier to mesh a single thickness slab.

Good luck with your build

[saveasteading]

2 hours ago, Nickfromwales said:

what I've been around for 20+ years

Your thoughts on crack control if we don't want shutters or slicing of the cured surface. Crack inducers? Or with all that mesh it is microscopic. Clearly no worries about the pipes failing when they stretch that tiny bit.

[Nickfromwales]

51 minutes ago, saveasteading said:

Your thoughts on crack control if we don't want shutters or slicing of the cured surface. Crack inducers? Or with all that mesh it is microscopic. Clearly no worries about the pipes failing when they stretch that tiny bit.

My thoughts are, if you install “anti-crack” mesh into the slab, then it should be doing its job. I do make sure that the mesh passes through door openings, vs stops at the thresholds and then a new mesh starts again the other side with no crossover. That’s kind of in the idiots guide to tying mesh, but most guys I’ve been around who are installing mesh do this without being asked eg have at least one grid crossover at every sheet to sheet junction. 
 

With most good quality flexible tile adhesives you get a couple of mm of decoupling effect from the substrate, specialist adhesives such as BAL offer S1 which had between 2-5mm and S2 >5mm, so I guess getting 1-2mm with regular / popular flexible tile adhesives would be the assumption.

 

Whilst the L shape slab cured a hairline crack appeared, completely expected, and then never changed state. It started prob 150mm inside the outside ring beam, spanned the living room of 5-6m, in the 100mm section, and then disappeared again at the opposite ring beam. I knew the tiles would be fine to traverse this without issue, as it was ‘microscopic’. 
 

Pipes are fine, there’s really just so very little movement, and the heating Durand go hot > cold > hot > cold every day, it just warns once and stays pretty constant throughout the winter heating season.

 

Its all good, as long as the guys executing the slab are robust and give at least a slight feck about what they’re doing.

 

All the steel / ground workers I’ve met to date have been spot on with this, including the MBC crew who do this in their sleep.