250mm insulation under UFH slab

[epsilonGreedy]

5 minutes ago, Ian said:

No benefit

I used 300mm of EPS 70 under our ground floor slab. It's been down 4 years with no issues. The slab was 100mm concrete with a D49 mesh with UFH

 

 

The D49 in your 100mm floor indicates there is a concern about the support offered by 300mm of EPS. Since the starting point for the OP is 70mm UFH slab what is right for that situation?

[Bitpipe]

It should not be difficult to calculate the point loading in this environment. You then chose an EPS grade that just exceeds that.

 

https://insulationcart.com/knowledge/about-products/expanded-polystyrene-eps/108

 

 

[Ian]

1 hour ago, epsilonGreedy said:

 

The D49 in your 100mm floor indicates there is a concern about the support offered by 300mm of EPS. Since the starting point for the OP is 70mm UFH slab what is right for that situation?

mesh is cheap and it was more of a guarantee that we wouldn't have issues with the UFH as it was about the bearing capacity of the concrete or the EPS 70. (my slab has no screed on top)

Edited March 2, 2021 by Ian

[Nickfromwales]

2 hours ago, epsilonGreedy said:

This thread highlights the varying degrees of faith people have in the strength of concrete screed sitting on different thicknesses and different relative spongy behavior of insulation sheet types.

Experience you mean?

[Nickfromwales]

57 minutes ago, Ian said:

mesh is cheap

Probably the same price as a clip rail system to hold the UFH pipes, so that box ticked plus a lot of insurance.

2 hours ago, epsilonGreedy said:

Since the starting point for the OP is 70mm UFH slab what is right for that situation?

As prescribed. Also, from my limited knowledge of slabs and concrete, IIRC you need a minimum of 40mm cementitious coverage over bright steel to stop atmospheric moisture from penetrating the screed / slab and allowing the steel to start corroding. Maybe galv or stainless is employed in those circumstances, but i'm not 100% sure.

Got a load of floors made up this way that have been in for north of a decade, and I've had the same phone number for longer  Not a dicky bird    

[epsilonGreedy]

29 minutes ago, Nickfromwales said:

Experience you mean?

 

 

I did a catch up on this whole thread this morning and concluded there was a range of advice being offered. My suspicion is that the next UK wide building scandal is cooking here as the industry explores ever thinner flow screeds with UFH embedded.

 

The thing that keeps me awake at night is the thought that PIR does not rebound from compression unlike the more elastic and bouncy XPS and EPS. Is it possible that PIR will crush and settle a bit on an irregular base leaving 50mm of flow screed hanging over a small void? Then add in the inherent fault lines caused by UFH pipes and I begin to worry that these thin-crust screeds will start behaving like tectonic plates riding on a mantel of bouncy raft of insulation.

[Nickfromwales]

Just now, epsilonGreedy said:

 

I did a catch up on this whole thread this morning and concluded there was a range of advice being offered. My suspicion is that the next UK wide building scandal is cooking here as the industry explores ever thinner flow screeds with UFH embedded.

 

The thing that keeps me awake at night is the thought that PIR does not rebound from compression unlike the more elastic and bouncy XPS and EPS. Is it possible that PIR will crush and settle a bit on an irregular base leaving 50mm of flow screed hanging over a small void? Then add in the inherent fault lines caused by UFH pipes and I begin to worry that these thin-crust screeds will start behaving like tectonic plates riding on a mantel of bouncy raft of insulation.

An academic approach is fine, but my approach is of being on sites over the last 25 years and seeing these in, working, still working today, and all the ones that failed, and why. That comment of your suspicions is loosely scaremongering and without evidence also, ( IMPO ).

PIR does not crush with a raft of screed over the top, as there is virtually zero distinguishable point loading.

 

Order of events is usually;

• whack and blind ( with sharp stone ) the sub floor until compacted and level
• fit a secondary DPM atop the blinding layer
• install 25mm EPS as a sacrificial layer, employed to absorb any remaining points / adverse undulations / cheeky bit of hardcore that sneaked through when you were having tea and biscuits  
• fit the primary DPM atop the 25mm of EPS, ergo it is safe from being penetrated from said foreign items in the subfloor
• carry on building the EPS layer up to 100mm or whatever in thin layers so it conforms
• fit final PIR layer(s) as required
• fit perimeter expansion skirting / insulation 
• install screed / concrete / other to suit

The weight of the screed going down will hold all the insulation layers against each other, and is essentially the worst case for compression in the lifespan of the floor. It will sit there quite happily...........................unless someone buys a grand piano, sticks it in the corner of the room, and then invites the Tomi-Tukamoto Sumo-wrestling team, and their family, from both sides, to sit on it.

[epsilonGreedy]

4 minutes ago, Nickfromwales said:

An academic approach is fine, but my approach is of being on sites over the last 25 years and seeing these in, working, still working today, and all the ones that failed, and why.

 

 

Building materials and techniques do not remain constant as the shift from traditional screeding to flow-screeds demonstrates. Some character opened an account here last year and told us he was a pro builder who was going to offer advice on flow screeding. He was recommending 35mm or 40mm flowscreed with UFH embedded! There is a natural bravado in the industry to continually push limits until things start failing.

 

Take batch mortar that is sold on performance criteria rather than mix proportions. The end result is new build houses being demolished when the mortar is discovered to be ridiculously weak.

 

11 minutes ago, Nickfromwales said:

PIR does not crush with a raft of screed over the top, as there is virtually zero distinguishable point loading.

 

 

But does it crush and so subside at the bottom on an irregular surface as being discussed in this thread? The answer is maybe hence your recommendation to start with the sacrificial EPS layer.

 

Yes the passiv slabbers are happy with their 300mm of bouncy insulation riding on a hardcore base because they top that with a substantial 100mm reinforced concrete/UFH raft. What concerns me about this thread is that different building techniques are being mixed i.e. rough-base, thick insulation raft capped with the thin-crust screed without reinforcement.

[Nickfromwales]

2 minutes ago, epsilonGreedy said:

What concerns me about this thread is that different building techniques are being mixed i.e. rough-base, thick insulation raft capped with the thin-crust screed without reinforcement.

And what can reassure you is, the FACT that I have done many like this, and its all good. Same with anything, the job will be as good as the execution and attention to detail. If that slides, so will the things you put on the floor. Very much relies on the builder having more brain cells than a house plant.

[MortarThePoint]

On 02/03/2021 at 13:35, epsilonGreedy said:

Building materials and techniques do not remain constant as the shift from traditional screeding to flow-screeds demonstrates. Some character opened an account here last year and told us he was a pro builder who was going to offer advice on flow screeding. He was recommending 35mm or 40mm flowscreed with UFH embedded! There is a natural bravado in the industry to continually push limits until things start failing.

 

I'm looking at a total liquid screed thickness of 40mm, but there is no insulation between that and 50mm of concrete so I expect that is OK. Would you still be nervous?

[joe90]

I agree with @Nickfromwales, I have a 70mm screed with UFH pipes In it over 250mm EPS on a ground slab throughout my whole ground floor, I expected cracks in the doorways but no, no cracks at all anywhere . ? the only thing I would consider is a thin sand layer over your base concrete to give a flat surface fir the insulation to rest on ?.

[dpmiller]

The gypsum screed we went with is extremely stiff and I'd best describe it as being like a council 3x2 paver. Ever tried breaking one of those?

[Dave Jones]

On 01/03/2021 at 08:11, Ian said:

but it's not a screed.

Screeds are laid on top of a concrete floor (not 250mm of insulation)

 

nah. 75mm semi dry screed with fibres and joins wont crack. Put it with a laser is best floor you can have.