2'x1' tile popping and grount falling out

[bc brick john]

Hi folks just
got home to nite after going to view a install that was put in approx two yrs ago the tile has started to lift away from the thinset
looks like no back butter at all.
loose tile are almost clean
with not much signs of kiln release either
the floor is solid as there are mulitple layers of plywood down( 3+") and its above a warm enclosed basement
the tile install has several heat mats and the tile is down a hallway mudroom bathrrom and entance and kitchen area so quite a large 700- 1000 sq

look's like the heat mats may have been turned on too soon????
but why ? does the cilent have broken loose tiles grout coming out all over this job not just one area
looks like a whole tear out is needed
just dont know if its worth trying to save the heat mat cable

i'll try get a couple of pic's

[jwmezzanotte]

Did it look something like the pictures in the first post of my old thread?http://www.johnbridge.com/vbulletin/...ad.php?t=95819

In this case it was partly deflection, and I think partly the thinset itself. I used to work for the guy that installed this at one time. I can recall following his instructions to mix thin set. He was using Kerabond/keralastic with half keralastic half water if I remember correctly to float cables and install tile. I know he never back buttered his tiles either. Actually very few do around here.

I know that the owner is an impatient man. I have a feeling he also cranked up the heat way too early. He is still after me to

"Can I just pay you to "pffhhtt" (makes slapping sounds with hands for emphasis) shtickem back in there with some sikaflex or something?"

The installer (my former employer) is no longer with us. Too bad, he was a great guy and easily the hardest working man I ever met.

[Higher Standard Tile]

Any movement joints in the field or allowance for perimeter expansion?

Low quality thinset used , poor coverage?

Floor framing?

[bc brick john]

Hi john

looks very similar grout has popped and tiles have come loose
almost clean sure if i started taking it up they will be not much to take off the backs
looks like no back buttering at all and the thin set could have most likely skinned up before the guy set them

the confusing part is the floor is almost 3" of plywood so not much deflection?
and as stated it not just in one area but all over in differing places

there are hairline cracks in the thin set where the tile's have become totally lift-able
going to check to see if the thin set is bonded to the floor
if that's the case then look's likely installer let it skin up before setting
and most tiles have no bond to the floor/thin set

[tilelayer]

Maybe they used a dry-set mortar? That would cause this, no?

[mmullen]

No. A properly used dry set mortar would not cause this.

The 1950's house I live in has two tile baths and slate floors in two rooms, both installed with unmodified mortar. As far as I can see, they are as strong as ever, probably stronger.

[Higher Standard Tile]

Trying to stick a porcelain tile directly to plywood with a dryset is a bad idea.

[pgc555]

Am I reading that there is 3+" of plywood?????? Could be movement between the layers of ply.
Why would ANYBODY need that much underlayment. Please tell me I'm reading this wrong. I of course agree that those size tiles needed to be back buttered but that ALONE would not of been the reason.

[tilelayer]

Mike they are probably installed over mud, a totally different install all together.

[Dave Gobis]

Along the line of movement joints, thermal expansion/contraction coupled with minimal deformation thinset? What size were the grout joints and what type of grout?

[bc brick john]

Hi Dave thanks for stopping by on this i am in good hands now....lol

std custom cement grout laid up at 1/8th to 3/16ths width

the house was a reno almost a complete rebuild so there is no new home warranty which sucks for the client

there was differing levels of floors between new and old hence in some of the areas there is multiple layers of plywood
it looks like also there was slc pour put down so hence i am going to go back and check to see if the thin set is bonded down too the ply
hopefully i can get back there sometime this weekend.

no mud its installed over ply and there are several heat mats as well (3-4)

[Dave Gobis]

So with a heated floor from an installation standpoint you likely have wood shrinkage (take the RH) with tile thermal expansion. Then you consider no matter how thick the floor, all wood moves. Then install big tile (with 25,000PSI lateral strength) straddling the floor joist and use a cement grout that is 1,500 to 3,000 PSI. Install all this over multiple types of substrates and support. What could go wrong with that? You would have to have a real Cadillac thinset and a lot of joints to make that work out well. I think you will find the issues are cumulative, not singular.

[bc brick john]

Hi guys the pic's are not the best poor camera phone
but you can see just how clean the tiles are

went over to the job to nite and seams the thinset is bonded to the floor or at least to the SLC and within the heat mat
but there are lots of hairline cracks within the body of the thinset

prolite or a custom modified thinset was used

[Swedes Tile]

Thinset did not bond to the tile at all, either the tile was dirty or it's just a typical skinning over failure.

[MacTile]

I would guess that whatever was used as the substrate under the heat mat was not properly bonded to the plywood underneath. (Thinset and nailed or screwed if it is a CBU,) or if it was a self levelling underlayment, there exists too much deflection for a normal prime and pour SLC application. That substrate is continuously moving up and down and has been since installed. The clean tile back is evidence since that would be the weakest bond until allowed to fully set up.
Just my opinion. Would like to know what was used to bond the tile substrate to the plywood though.