This shower curb is obviously not waterproof, and has already caused mold growth an adjoining wall.

This tile is not bonded to the slab even though a premium thinset was used.You can see there is no transfer of thinset whatsoever. The problem, curing compound. The pores in the slab have been sealed to prevent water loss and achieve a good cure on the slab. (continued in next photo)
Why version 6? This subject is like software; responses improve as time goes on. I looked at the editorial calendar and this months schedule said tile problems and solutions. Then I looked in my article file and found 5 versions of tile trouble articles I have written in the past few years so my thoughts went to how do I handle this; these same problems just never seem to go away. If anything they seem to increase in number with the growth of the ceramic tile market. When I started in this job 5 years ago Tile Council of America (now Tile Council of North America, TCNA) was receiving about 2000 calls and emails a year, that number has grown by 50 to 75% each year since. Some callers become vary irate when they are not given immediate free advice. TCNA and CTEF have all answered the same calls so many times we have now listed over 100 answers to frequently asked questions on the website Callers are now referred to the website to see if their initial questions can be answered there. If not, TCNA has a relatively new wholly owned subsidiary, TCA Team, who can provide further assistance and if necessary, an onsite inspection. TCNA members do want to provide assistance to anyone wishing more information, preferably before the job goes in. Every single manufacturer member provides technical assistance via 800 number or email willingly on request.

Unfortunately, it leaves nothing for the thinset to grab on even though latex was used. By the way, this was a special UV dissipating curing compound.
However, unfortunately, many don't call till after problems have occurred. They have been doing it this way for 20 or 30 years and never had a problem. If I had a dollar for every time I heard that I wouldn't be working at all, I would be laying under a palm tree on a tropical beach sipping a beverage.

The TCA Team was created out of necessity as calls were consuming vast amounts of time for 4 or 5 people at TCNA and CTEF who had other jobs, answering calls about tile problems not being one of them. They provide a fulltime staff which can answer questions, consult, provide quality control and field inspections or if need be, forensic failure analysis. The pictures in this article are from a few of 100's of inspections they have performed in the past year. Unfortunately, the problems remain the same as always and even more unfortunately are still 97 to 98 percent installer error. I wish I could change that, but facts are facts. I will skip the editorial on why you should continually educate yourself regardless of how many years you have been in the business. Suffice it to say, if you are doing the same thing now that you were 30 years ago, you have a problem. Everything has changed but you.

You can see the base is firmly, and poorly I might add, grouted into the wall.
Surface Preparation

This is a very large project that is having complete bond failure. Initial complaint was bad thinset. Possible? Sure. Likely? No. Most cement-based thinsets have a shelf life of a year, some only 6 months. Mr. I been doing it for 30 years would say no lumps in the bag, use it. An educated Mr. 30 Years would realize that some of the formulations in use today can go bad without a single lump anywhere, is just plain won't bond. The reason for this is highly technical and way beyond the scope of this article, but I am working on it. Regardless of thinset used if you have any curing compound, coatings such as paint overspray, dust and dirt, or any foreign substance on the substrate to which you chose to bond, you are at risk.

In the real world, finding a slab with no curing compound is very rare. Concrete industry standards recommend none be used if tile is to be applied. That raises the labor cost to the concrete contractor. Even if he were willing it could be challenging to locate a specific area where tile was to be installed on a 2,000 much less 20,000 s/f pour. The most reasonable alternative is to grind the areas to receive tile after the pour before the tile goes in. This is not a perfect world scenario, this is real world. If you feel like rolling the dice you could do water drop test. Place a few drops of water on the slab, if it absorbs immediately, your good to go. If it takes a few minutes to absorb either use a latex/polymer or if the tile required one, use a better grade. Ten minutes go by and the water is still beaded on the surface means it is time to leave and come back when the surface has been prepared. It will not bond if the water cannot be absorbed. It may be possible at this point to increase your income by preparing the slab. If you fix it for nothing, because you're a good guy and really want the job or account, be prepared to continue donating free labor whenever the problem occurs in the future.

You can see a picture of a nice hump forming right in the middle of the floor. The initial complaint was also; you guessed it, bad thinset. The true cause, lack of movement accommodation.
We are hired to install tile, not fix concrete, however, for a fee we might be able to help.

Movement Joints

Without question, the single largest caused of floor failures are caused by lack of movement accommodation joints. There is typically another reason associated with this failure, but not always. Tile grows when exposed to heat or moisture. When it is exposed to both it can really move. The industry recommendation for floors is a minimum of 1/4-inch movement accommodation at all wall perimeters and anyplace the tile abuts another rigid surface. There are many additional recommendations such as increasing joint size for temperature range in exterior environments and placing joints at changes of direction contained in both the TCA Handbook and under the American National Standards for the Installation of Ceramic Tile. One argument made fairly often, again by Mr. 30 Years is there is no way grouting against drywall will ever cause a floor to shear or "tent". Wrong again, it takes about 600# of uniform pressure to dent drywall.

Coverage on this 1x1 is pretty good most tile is firmly embedded, 100 percent would be nice but, not bad. But, lo and behold, it is shearing of the wall after 32 years of service. What happened? Lack of movement accommodation which was especially important in this installation as it is a wall in a gang shower, the contributing factor to failure.
Grout provides a nice uniform load surface when placed between the tile and the wall. And what are the pounds of force required to shear most thinsets from the surface? How about 200 pounds in the case of premium unmodified thinset, slightly more, 250 to 300 pounds for a mid range latex/polymer, or 500 pounds plus for a premium. So not only will the grout packed in next to the drywall have enough force to shear in case of movement (a given in all materials) it has enough in reserve to do it several times over in most instances.

So you're thinking well OK, maybe. "But I have a block wall with wire reinforced mortar and 1-by-1 porcelain mosaic. The wall is solid, reinforced, interior, low growth tile so I don't need no joints, it isn't going anywhere." Says Mr. 30 Years. Sorry, wrong again. The industry recommendation on walls is 1/8-inch movement at abutting surfaces and as determined appropriate in the field of installation by a qualified professional. All materials have a growth rate and they are all at different rates. While that may be one fine solid wall, it is still going to move.

Here is an example of some creative engineering on the part of a tile installer. Do you really think you can use fiberglass tape reinforcement to stop tens of thousands of pounds from moving?
Bridging Control Joints

A favorite term I heard from concrete guys in my contracting days was once it is cracked, it is cracked, the joint is dead and isn't going anywhere. In the real world, all you count on is it is going to move again. With rare exception, very rare, you will always find movement in a concrete separation or crack. Control joints are to control shrinkage and cause the cracks to occur in predetermined areas. There are concrete industry guidelines as to when and where they placement. A qualified person must make this determination, if your reading this article, your not it. All concrete needs control joints, there is no such thing as only in commercial construction, we don't do that on residential jobs. If there were no joints on a slab of any size it would be prudent to use a crack suppression membrane over the entire surface. If there are joints the tile must stop at that joint and a premade movement joint placed or appropriately constructed with caulk and a backer rod.

Another neither right or wrong approach used is to follow the joints with a saw once the tile has been installed. Unfortunately, in this case the installer was a little bit off.
There are some membranes that will allow for relocation of the sealant joint and a few that can eliminate it. This does not apply to all membranes meeting industry standards under loadbearing membranes. There is currently no standard for crack suppression. Consequently, this is where some, including manufacturers, get extremely creative. Uses of roofing felt, sheet vinyl, scribing paper, or scrim reinforced kraft paper glued or unglued to concrete slabs have a long history of failure in both field and strip ( crack chasing) applications over both entire floors and joints. Many of these products lack the performance features and criteria that would allow effective control of concrete fractures without transmission through the finished tile surface. If the slab has excessive moisture or alkalinity failure is imminent with these materials. There are also some even more creative ideas like duct tape and stucco mesh will actually make little difference.

Grout Joints and Shading

Maybe we will finish this months article with another product that Mr. 30 Years says fails all the time, that brand X's is no good, never the right color, and gets effloresce due to poor quality of materials used in manufacturer, grout. Why pray tell does one of the most important part of the installation get left with low man on the totem pole? To find a decent grout job today is almost a rarity, it is very frustrating. We demonstrate various cleaning techniques weekly. It is actually much less work to do it right than wrong, a fact proven over and over again.

Here is what started as a grout efflorescence complaint; it was a big job too. Your first indication that grout is not the problem can be found looking at the sealant joint; what is that sitting there? Turns out to be an acidic cleaning chemical that was being used for regular maintenance and was never rinsed. Now neither the tile nor the grout can be maintained by any reasonable means; it is time for a tear out.
Mix it to the right consistency, which is on the stiff side, not pourable. Wipe the tile with a damp sponge, clean clear water is better than pigmented water if it is going to stick somewhere. Pack it into the joints in a perpendicular passing over the area a minimum of three times. Let the joint stiffen which takes 15 to 30 minutes depending on conditions, longer or less is possible. DO NOT use a fan to accelerate the process, grout needs water to set properly, but that is all it needs water for. Force drying will make it soft. Always employ the driest method possible for clean up. Years ago we used burlap or cheesecloth and large particle sawdust. Sponges were the domains of wall grout, sometimes. With latex formulations came sponges. If you play with it a little there is no easier way to clean grout than with burlap and sawdust. This is a technique that requires development and does not work all the time. Sometimes there is no way around the sponge. If so use as little water as possible in the cleaning process to avoid discoloration and possibly effloresce.

I have been working with several setting material manufacturers lately. You just can't truly appreciate how much these guys care about their products and performance. They are constantly helping out people who fail to follow instructions to the limits of their abilities. I spoke with one who recently had a slow month, 7,000 calls for product information and assistance. It keeps 4 people busy 8 hours a day, 5 days a week. Now is it really that complicated? Some of the questions just don't make any sense at all. Can you use brand Y primer for brand X self leveler or can I use brand Y self-leveler and brand X thinset. In theory possibly, in practice, do you really think brand Y tests brand X's product for compatibility with their own product? Not likely. If they do it is clearly stated in the instructions. Most tech calls are for reading instructions printed and readily available. As products evolve the instructions often change to reflect the new process. There are several things you can count on with tile work, something is always changing, everything is always moving, never enough time to put it in, always enough time to fix it. And to Mr. 30 Years and his buddy Mr. 20 Years (are they related?), I would really appreciate if you would read the instructions before you call, preferably before you do the job. Then we can talk about some of the finer points of the installation and what could have happened.