Mystery of the hollow tile ???

[cx]

If I wanted to remove one of the tiles to determine the cause of failure I think I'd wanna cut completely through the ditra all around the tile and try to remove that tile with the Ditra attached first. That doesn't work, then I'd start breaking the tile to see what comes up attached to what.

My opinion; worth price charged.

[Davy]

I would think the thinset under the Ditra is letting go of the slab. The thinset should be locked in the waffles under the tiles. I've had a few bonding issues with Ditra sticking to concrete even when i think I mixed it on the wet side.

If you patch in the area, you might go back with Modified thinset under the Ditra, especially if that's where it's turning loose.

[Brad Denny]

I look at a lot of potential repairs in commercial settings, and get "hollow" questions like this from time to time. My usual spill is to wait to see if there is any tenting or crumbling/crunching sounds before removing if there are no obvious places where expansion/contraction is the cause. That may be easier to live with outside of a home, but if it isn't totally broke I don't like fix it. I will add that even my total brick fireplace assembly has started to sound hollow directly above the lintel due intense wood fires in my three year old home. I realize it is exposed to much more heat than what you have, but I believe there is merit in the possibility it has affected the bond. Be sure to set back pieces for future problems, though.

[ddmoit]

Brad, you got some nerve, discouraging George from tearin' up his tile for the benefit of our curiosity. I wanna know what's going on under there!

[Davy]

Yeah, no more than there is, I would repair them, myself. Alot easier to do now than after grouted.

If you do repair them, be sure and cut around them first like Gueuze said.

[Brad Denny]

Oh, I missed the part about it not being grouted. I read that it had been down for months, so I's figured George woulda done had that part done by now.

[Trask]

Simple to pull a single tile..just cut with a new razor blade all around the tile... I'll bet it'll come out fairly easy.

[George K]

My plan

Next week, I'll SpectraLOCK up to the hollow sounding area. I'll then move the wood stove and check out the sound of the tiles that are under that stainless steel sheet. I'll have to move that stove sooner or later anyway, in order to grout. The wood stove will be put on a dolly and rolled over the hollow sounding tiles in order to get it out of the way. That was how it was placed there originally. The stove weighs 540#. If I crack a tile or two, then I guess that will be the deciding factor to remove them. If the tiles do not crack, and will handle the 130# point loads on each of the dollies 4 wheels, then that will be useful information and argue in favor of leaving well enough alone. Maybe I should even roll it back and forth over the hollow sounding area.

I have just 19 large tiles left over from my color lot to handle any future problems. I'm reluctant to cut through the DITRA at this point. I like the idea that the DITRA membrane is intact, both for Radon and for Moisture. I did not just butt the DITRA strips, I seamed them with Kerdi-band.

[Old World Tile and Marble]

george make sure you put some soft joints in that floor 1300 sqfts a whole lot of rosie

[George K]

I have divided the floor into 1/3 sections with 2 soft joints at the 1/3 field points. It was a challenge with that pattern. The soft joints follow the grout lines and are not straight lines as they travel wall to wall. Also 1/4" soft joints around the perimeter and the lolly columns.

[George K]

The first picture clearly shows that the hollow tiles are not under the stainless steel sheet that the stove rested on, but form an arc around it. Remember that these tiles are not yet grouted.

The second picture is my Mark 101 patented hollow tile tester. Just roll it back and forth over any hollow tile. If the tile does not crack, you are good to go. All my hollow tiles passed that test.

One other observation, that was most curious. On the underside of the steel sheet, there was condensation. The condensation was only immediately above the open grout space, and only about a couple of inches long in the exact center of the sheet.
Remember that the tile is over DITRA that has been seamed with Kerdi-band.


Your comments gentleman?????

[ddmoit]

Looks like you made your own Robinson tester, George.

[George K]

I would eliminate an application problem, either slab to DITRA, or DITRA to tile.
It is unlikely that I would have erred in a arc pattern EXACTLY where a future stove would be placed and nowhere else on the other 98 percent of the floor. The area in question also extends over 2 DITRA strips and that would eliminate the DITRA itself since it would be unlikely that a defect in the sheets would exactly align just so.

Thinset is a possibility. It was the same unmodified Laticrete 377 thinset for both bonds, but the pattern of failure would be difficult to account for, on the basis of just the Thinset itself. Also, more than sufficient time, several months, was allowed for the thinset to cure before the stove was fired up.

Tile is a possibility. The tile is being used indoors. The rating for Water Absorption is <3 percent. It is not porcelain, but ceramic, and I do not think it is rated for outdoor use. Outdoor use would imply the ability to handle a wider thermal range. While there was an indoor temperature change that the tile was subjected to, I doubt if it exceeded 25 degrees going from 60 degrees F. to 85 degrees F.

Things to think about:

The fact that the area was subjected to a thermal change.
The fact that the tile was NOT grouted.
The fact that condensation was found under the steel sheet and corresponded to an open grout space in the center of the steel sheet.

We are dealing with only two bonds. One or both partially failed. I would guess that it is the tile to DITRA bond, rather than the DITRA to concrete bond. The slab itself, because of its' mass, would have attenuated any wide thermal change in the DITRA-to-slab bond.

One additional test I will perform. I will place the steel sheet on another ungrouted area of the floor, not too far from it's original location, and see if I get condensation there. That could be useful information.

So far, since the tiles are NOT loose, but only sound hollow, a tear out would not be fruitful if the same thing is only going to happen again.

There is a lesson to be learned here. It could help others who are contemplating a wood stove installation. A ceramic tile-DITRA-concrete slab, is not that unusual an installation.

[Davy]

Hi George. If the thinset is letting go under the tiles, they would lift up right now. You say they are staying put when you tug on them. The thinset won't come out of the waffles in the Ditra. To me, that only leaves the bond to the slab and the bond to the Ditra on the bottom side. I would use a margin trowel or something similar to stick in a grout joint to pry on a few tiles, checking the tile to thinset bond.

[George K]

I would use a margin trowel or something similar to stick in a grout joint to pry on a few tiles

Tight. Nothing popped up.

I will place the steel sheet on another ungrouted area of the floor, not too far from it's original location, and see if I get condensation

No condensation yet.