QUESTION

Slate debonding on Patio - I had 18inch slate tiles installed over a concrete patio that was approximately six weeks old. The contractor is noticing that many of the tile thinset is not adhering to the crack isolation membrane that was applied. The thinset is adhering to the back of the tiles, just not the membrane. Do you know of any situations or occurrences where this happened. Are rapid or fast cure thinset more probable from not adhering to the membrane. I do not know the brand of membrane but I do know that it was blue and that they used a roller to paint it on and then it dried to almost a spongy/fabric material. I apologize about so much detail but I am very confused and troubled why this is happening.

ANSWER

ANSWER - Generally when there is a problem there are compounding factors that lead to the problem.  Without intrusively investigating your situation then it would be difficult to determine what has caused the debonding of your tiles.  Here are some general points to consider:
• They don’t make one-sided thin-set.  If it bonds to one side and not the other then there is something acting as a bond breaker.
• Need to understand what type of membrane was used and what type of thin-set was used to bond to the membrane to see if they are compatible.  The thin-set used should be what the membrane manufacturer recommends to be used.  Normally it requires a modified thin-set that meets ANSI A118.4.
• If the membrane is compatible with the thin-set, then is there some sort of contaminate on the membrane that is acting as a bond breaker?  E.g. dirt, sealer, etc.
• If the membrane doesn't have a contaminate is there evidence that the thin-set had skinned over before the tile was applied to the membrane?  Was thin-set also applied to the membrane directly to key the thin-set into the membrane?  Yes, rapid setting thin-sets are more prone to skinning over because there is less open time and faster setting time.
• Are there expansion joints (open or filled with an appropriate sealant – not hard grouted in or butted up to a restraining surface) at the perimeters and transitions of the floor and within every 8’ to 12’ in each direction?   The lack of expansion joints puts stress on the tile assembly during normal movements caused by moisture, temperature and dynamic structural movements.  If the tile isn't bonded well then the stress could contribute to the debonding.
• It is assumed the substrate is a slab on grade or otherwise the deflection of it should not exceed L/720 over wood subfloors or L/360 over concrete subfloors.  If the tile isn’t bonded well then the excessive movement could contribute to the debonding.

If it becomes necessary to have a professional investigation performed for your situation, you can consider retaining CTaSC services to do so.  Visit www.CTaSC.com for more information.