Grout coming out and crackling noise under some tiles - Tile Advice

[Sachin Parab]

Removing laitence from the floor is not just a case of a quick skim over or using a scrapper.
What method did he use and what equipment?
This year I ripped out and refixed an anhydride floor, which I know 100% was sanded by the contractor.
The reason I know is the client made a point of asking what they were doing.
The Builder then continued to install the floor with his Tiler.
9 months later it failed, on inspection, laitence was still present, it failed.
So we had to start from scratch, including re sanding of the floor.

The moisture reading using the hygrometer, how long were they left in place before readings taken?

I think a hand held sander was used. Is there a specific equipment he should have used.

The hygrometer readings were taken in 72 hours.

How do we check if laitance has been completely removed.

Thanks

 

[Old Mod]

You really do need to remove a tile to see what’s beneath.
The cracking sound is where the tile has de bonded from the floor.
It’s important now to determine what has de bonded. The tile from adhesive, or adhesive from floor.
That will narrow down your search for an answer.
The floor need sanding back to expose the aggregates to be 100% sure it’s been removed.
First two images have laitence still remaining on the failed floor, bearing in mind some has already been removed and some was stuck to the tile.


Enlarging the last image you can see the dark spots, that is the aggregates.
Can be done with a hand grinder but that’s quite an area for a small machine.
More typically a floor sander would be used.
(They look like an industrial cleaners buffing machine) but you would fit abrasive discs instead, around 60# grit.
And you need dust extraction.
You can see the quantity produced.
And priming is critical when using a cementitious adhesive on an anhydride floor.
2 coats is not necessarily enough.
And they should be applied perpendicular to each other.

 

[Sachin Parab]

You really do need to remove a tile to see what’s beneath.
The cracking sound is where the tile has de bonded from the floor.
It’s important now to determine what has de bonded. The tile from adhesive, or adhesive from floor.
That will narrow down your search for an answer.
The floor need sanding back to expose the aggregates to be 100% sure it’s been removed.
First two images have laitence still remaining on the failed floor, bearing in mind some has already been removed and some was stuck to the tile.
View attachment 94599 View attachment 94600 View attachment 94601 View attachment 94602

Enlarging the last image you can see the dark spots, that is the aggregates.
Can be done with a hand grinder but that’s quite an area for a small machine.
More typically a floor sander would be used.
(They look like an industrial cleaners buffing machine) but you would fit abrasive discs instead, around 60# grit.
And you need dust extraction.
You can see the quantity produced.
And priming is critical when using a cementitious adhesive on an anhydride floor.
2 coats is not necessarily enough.
And they should be applied perpendicular to each other.

View attachment 94603

Thanks a lot for this. I will speak to the builder accordingly.

 

[Qwerty]

What is his previous experience with anhydrite screeds & tiling to them?

 

[Sachin Parab]

What is his previous experience with anhydrite screeds & tiling to them?

He claims to have done these before and used the same screed and it has worked well

 

[Chalker]

He claims to have done these before and used the same screed and it has worked well

Maybe he hasn't done a floor this big before.
Tiling a large floor has more complications. The main one on underfloor heating is expansion. If there are large areas with form of expansion, like doorways and large single rooms the floor will move and the tiles won't. This can debond the tiles.
As said above the screed should be dry, then it should be abraded to remove the latience. Then vacuumed and primed. 2 coats is a general rule, but it should be primed until it stops soaking in.
Then a flexible adhesive suitable for the screed in question. With good priming, a cement based adhesive can be used. But far better to use an gypsum based adhesive, as that's what the floor is.depending on the size of the are a decoupling membrane should be used. Expansion strips should be installed to help large rooms and in doorways, as different rooms expand at different rates. Tiles should then be fixed with a notched trowel and tiles back buttered.

Not knowing the sizes of the room,I would guess this is an expansion related problem. I would bet your going to remove a tile and agin find the adhesive is stuck to the tile and not the floor.

To sort this, it needs two things.
1. Fitting as above with close attention to movement joints and uncoupling.
2. Get a tiler with experience of large gypsum based floors.

 

[Rookery]

or 3, avoid gypsum floors like the plague they are.

 

[Flintstone]

You have said that a couple of times in different threads. Have you had a bad experience your self with anhydrite floors? If you do what your supposed to do and prepare them correctly, they are a dream to tile due to there flatness.

 

[Rookery]

They're a complete pain in the arse. I've never had a failure myself but have re-done a few where the people before didn't know what to do and so there often were failures. Its always the same things too, too much moisture/laitance not removed/removed wrong time/inadequate priming/etc/etc and they're still not always flat. What's wrong with a sand & cement screed, you know where you are with one of those and if you want it flat, lay it flat. It's interesting that the regular firms I use such as Mapei/BAL/Weber here don't sell a gypsum adhesive. Do they know more than me? Well probably but I'm not going to tile onto anhydrite again. Too many variables. If they want to tile onto anhydrite they can get some other idiot to do it, I don't need it.

 

[Blunt Tool]

They're a complete pain in the arse. I've never had a failure myself but have re-done a few where the people before didn't know what to do and so there often were failures. Its always the same things too, too much moisture/laitance not removed/removed wrong time/inadequate priming/etc/etc and they're still not always flat. What's wrong with a sand & cement screed, you know where you are with one of those and if you want it flat, lay it flat. It's interesting that the regular firms I use such as Mapei/BAL/Weber here don't sell a gypsum adhesive. Do they know more than me? Well probably but I'm not going to tile onto anhydrite again. Too many variables. If they want to tile onto anhydrite they can get some other idiot to do it, I don't need it.

You really don’t like it do you

 

[Rookery]

No. And there's another 2 threads today which reinforces my point.

 

[Flintstone]

There only a pain if you don’t know how to deal with them.

 

[Ajax123]

Havent read through the entjre thread but was curious...If the builder used lafarge anhydrite screed why has an investigation been started by cemex?

Problem sounds relatively straight forward... screed not dry and no moisture test done 3 weeks is not long enough for drying a screed properly. Im guessing prep was not done correctly either.

 

[Chalker]

We've had it me this year on a massive new build. Laid in February and took until September to fully dry, with the heat in for a month.

 

[Rookery]

There only a pain if you don’t know how to deal with them.

Like on this floor which has been done twice by someone who allegedly knew what he was doing and it's still failed.