Floor tile decoupling layer. Several cons , few pros. Other views?

[saveasteading]

We have a lot of floor tiling to do. A recommended tiler says we "really should" have a decoupling membrane.

I can't see any evidence that this isn't just an extra earner or " what most people recommend".

 

The stated reasons online are.

1. Peace of mind. 

For the tiler or me?

2. Waterproof layer.

no need.

3. Allows for settlement of the building. 

It's all on an existing industrial slab, 30 years in use.

4. a) Allows for shrinkage of slab.

Ditto

b) Allows for expansion and shrinkage and drying of screed.

40mm screed is already cured and isn't going to expand or contract .

 

5. Allows for thermal movement due to underfloor heating.

I don't see that as significant. Any cracks are microscopic.

 

6. Ideal for Large Format Tiles – Enhances stability and strength. 

What difference does size make apart from the cost of big tiles?

 

 

7. Esp important with ufh due to heating and cooling cycles.

I don't see 30 degrees as extreme

 

8. Enhances strength.

I'd think the opposite is the case. To snap a tile it is bridged over soft material and point loaded. A tile cutter doesn't work with the tile flat on a solid base. So if i drop a pot I want the tile to stay solid, not deflect.

 

My other cons.

 

A. £7/M2 or more, plus waste, plus labour.

B. We end up with a 10 To 20mm skin of ceramic, barely stuck down or restrained. 

C. The decoupler works both ways, and could encourage vertical movement under loads, eg table legs, and the grand piano.

D. Even if the slab was poor quality, what is that membrane actually doing? Allowing the slab to move but leaving the tiles exactly where they are? I can't see it.

E. The membrane acts as insulation on the wrong side.

 

I've looked hard for any evidence and all I can find is that it is normal. Perhaps in a new build of questionable quality it is wise, and for small areas the cost seems relatively minor.

 

 

I've currently got an old house with poor quality concrete floors. Large expanses of tile have lasted 15 years so far. Lucky again?

And a holiday home abroad with lots of tiles on concrete. The only cracks are where I've dropped a log.

 

I should explain. The project in question has 175 reinforced slab on dpm on stone, 30 years industrial use.

Then 300mm eps and pir, and dpm. Then 40 to 50 poured screed with ufh.

 

But I wouldn't be asking if I would not welcome opposing views or evidence.

Any comments?

 

 

 

 

 

 

 

 

[Nickfromwales]

The only concern is 40mm of screed is as thin as it gets. That is the only reason I would suggest decoupling here. 

[saveasteading]

40 minutes ago, Nickfromwales said:

I would suggest decoupling

If you could expand the logic, or instinct, please?

 

At nil screed of course it would be impossible to tile:  at 100mm you appear to agree with my logic.   I have some hunch that there is a purpose for membrane at 40 /50mm but I can't grab the logic out of my head.

Btw we will have 200m2 of tile (the rest timber laminate) so, at say £10/m2 plus any labour implications, it is a big cost.

We're doing well on tile costs so to then add back 1/3 "for peace of mind" wrankles.

[G and J]

We"ve literally just had 88m2 of 1200 x 600 tiles laid by an very experienced (read older just like us!) on 100mm screed, which was laid early dec 25 and has been heated by the ufh

 

Has been laid on a decouplinging mat as recommended by the supplier, whther we needed to is perhaps questionable as in the tilers view slab "would have cracked by now if it was going to".

 

However having it allows more peace of mind (for all of us) in not having room thresholds, especially given size of tiles, and allowed a slight overlap (75mm) of the expansion strip in the screed, which then facilitated cuts working out to give best aesthetics..

 

In the end our view was they are not coming up again in our lifetime (we hope!) so in the scheme of things.......bit of a slipoery slope cost control wise, but we're at that stage of the build to better understand spend.

[saveasteading]

12 minutes ago, G and J said:

allowed a slight overlap (75mm) of the expansion strip in the screed

That's another thing that some trades-people "just do".

Screed and slabs indoors do not expand. They may benefit from contraction joints/crack inducers but that is for tidy cracks as opposed to random ones.

 

[G and J]

8 minutes ago, saveasteading said:

That's another thing that some trades-people "just do"

Agreed, in this case at a very specific point allowing for the steels/load points but also allowed the screeder (traditional dry screed) to go home for the night!

[saveasteading]

43 minutes ago, G and J said:

allowed the screeder (traditional dry screed) to go home for the night

That's a " day joint". Not a bad idea though to put a factory edge to work to. As compared to a piece of 4x2.

Edited 11 hours ago by saveasteading

[Nickfromwales]

6 hours ago, saveasteading said:

If you could expand the logic, or instinct, please?

 

At nil screed of course it would be impossible to tile:  at 100mm you appear to agree with my logic.   I have some hunch that there is a purpose for membrane at 40 /50mm but I can't grab the logic out of my head.

Btw we will have 200m2 of tile (the rest timber laminate) so, at say £10/m2 plus any labour implications, it is a big cost.

We're doing well on tile costs so to then add back 1/3 "for peace of mind" wrankles.

The thinner the screed, the more likelihood of fractures (slight cracks) which over time could translate to visible failures of the tiles or grout.

 

If it’s 40mm of fibre impregnated, cementitious liquid screed then you’re at the 49% club, if its 40mm of dry screed you’re in the 80-90% club; dry screed does not like going down as thin as 40mm, but you mention ‘poured’ so I’m assuming liquid screed?
 

Then, if it’s gypsum vs cementitious, at 40mm this would have a reasonably high risk of cracks vs fractures imo, so I’d decouple as I can’t take any risks with clients projects. I’m also considering the repeated thinning out where the (?)16mm UFH pipes lay, reducing thickness to 24mm, linear to the whole length of each pipe run. That’s a significant, further compromise when looking at this holistically. 
 

If it’s fibrous, cementitious screed, then I think you’ll get away with having just a good quality flexible tile adhesive, with a min 6mm bed, laid over an appropriate primer (with the tiles being laid onto the primer when it’s wet).

 

BAL do a tile adhesive that has (x)mm of decoupling qualities, like an ‘ultra-flexible’ adhesive.
 

I always Ultra (Insta) products, never had an issue.

 

Are you actually using ceramic tiles? These are quite feeble compared to porcelain, so with ceramic and the very thin screed I’d be saying 100% you should decouple.

 

Points B,C,D, and E are all moot. 

[saveasteading]

28 minutes ago, Nickfromwales said:

thinner the screed, the more likelihood of fractures (slight cracks)

Thanks for this.

I'm not convinced on this argument but as i don't know much about these screeds I need to do homework.

It does not have fibre.  I've gone off fibre anyway as it clumps and is very inconsistent.

It is a gypsum product, pumped.

49% club... I don't understand.

 

28 minutes ago, Nickfromwales said:

BAL do a tile adhesive that has (x)mm of decoupling qualities, like an ‘ultra-flexible’ adhesive.

Brilliant info:  I was going to look into this principle.

Cracks will be about 0.1mm and will refill with dust. Even if they did somehow open or close again, it would need a minuscule movement in the adhesive... and the grout should give way first.

 

28 minutes ago, Nickfromwales said:

I'd decouple as I can’t take any risks with clients projects. 

Quite right. 

In such cases I would often explain the perceived risk with the client, and often they took the saving. It could still be awkward if there was a problem,  whatever the contract says.

But it's for ourselves using the pension I've saved by making such decisions for decades. i. e. Questioning almost everything, especially the standard ways of doing things that nobody can justify. These are usually "trade" things where they say its best but hsve nil justifiication. Most radical experijents have been on ourselves. Some haven't been ideal but no dramas and usually it's fine.

 

Strangely I see that the ufh pipes are mostly visible by having 1mm or so ridge over them. I'm thinking this has been by displacement of the screed after partial setting, and being walked on. But cracking is only visible in one place I've noticed, though if it's like concrete, there will be millions of tiny cracks/crazing.

They didn't put crack inducers at doors and there are no cracks there, which suggests shrinkage is miniscule.

 

I will research further.