This slab was neither prepared properly nor was the thinset applied properly. Now all 27,000 feet are being replaced.

The inspection and consulting business has continued to shed new light on methods used for bonding tile in a trade I thought I knew very well after nearly 40 years in the business. I don’t think anything other than the basic principles of tile setting remains the same over that many years — not the tile we install, the bonding materials we use, nor the substrates we cover.

Since the beginning of tile installation, the job of the tile installer has been to firmly support the tile over a suitable substrate. As tile grows increasingly larger and conditions of installation more complex, many look for a magic bullet to make life simpler. Because many have no formal education in the trade or have not had the good fortune of working with technically proficient (not production oriented) tile setters, many installers lack a complete understanding of the basic principles of tile setting and the bonding process. It gets even more complicated when the high tech thinset mortars entered the equation. The days of sand and cement were simple. All tile was installed over wet sand and cement with cement paste. The big arguments of the day were do you dust the cement over the mortar or make a paste to apply before you firmly beat the tile and cement into the semi loose mortar? The argument for walls was how long to soak the tile and what do we use for spacers — match sticks, wood shims or rope?

Those systems (minus the matchsticks) worked well for thousands of years. There are still some 2,000-year-old structures with the tile still firmly attached. If you care to only go back a few centuries, then there are tens of thousands of tiled structures still performing today. As cement provides only a mechanical rather than adhesive bond, setting directly to anything other than a wet cementitious surface was not an option until thinset was born.

Prior to the 1960s a certain degree of skill was required to set a floor over wet sand and cement, even more so to set walls over fresh mortar. Four-year apprenticeships were the primary means of training tile installers. With a new class of bonding materials called thinset, it was now possible to stick to any clean and cured masonry surface as long as the substrate was made moist prior to the application of the bonding material.

A scant 50 years later, we have undoubtedly hundreds of bonding materials for ceramic tile. There is a bonding adhesive for virtually any type of installation environment over nearly any substrate. There is a vast perception that tile setting has never been easier. As any good tile setter is sure to tell you, nothing is further from the truth. People in the trades have become overly complacent about the needs to understand the basics of installation at all levels of the industry. With the array of products available for bonding, and tile sizes reaching that of a half sheet of plywood, life has gotten considerably more complex. Then there are the fast-paced and ill conditions one must install under. It may be a homeowner with a party for 20 guests just five days away from the start of the job or the owner of a new restaurant setting the grand opening date months in advance that can’t be altered. I personally have been in those and many similar situations.

Then there are the competitive circumstances that you got the job under to begin with which left no room for overtime or specialty products to accommodate the customer’s newly mentioned and previously unknown needs or wishes. Yep, I been there and done that too. The incentives to pound the job out by any means can certainly be nearly overwhelming. But, like lending money to good friends, succumbing to that temptation often results in unhappy customers and financial loss by having the job fall apart, sometimes before it is even paid for.

We really need to consider the basic requirements of the installation if we are to have happy customers and at least partially full wallets and/or pocketbooks. Regardless of what the customer wants or needs, the most basic requirements of a tile installation do not change — proper environmental conditions, a suitable clean surface and appropriate bonding material to firmly support the tile.

Now this is the way tile is supposed to look when you pull it back to check coverage.

Let’s take a short look at the suitable and clean surface. A substrate to adequately support a tile installation must do just that, support the tile at the intended service level. Most structures, either wood or concrete, are typically engineered at code-compliant levels and tile is often not considered or an afterthought. Concrete is often steel troweled to a smooth surface and wood structures have overspanned floor joist and insufficient panel support. If tile, natural stone or some other type of rigid flooring product is selected, the deflection requirements for the structure can change substantially, even more so for very large tile. Some of my more recent experiences point out that this is a growing area of concern. In spite of all the bravado some tile installers may express, factually, few are engineers, the rest should not be expected to posses such knowledge. When in doubt about any product or structure, question the appropriate party and don’t make assumptions. You often find that architects and engineers are far removed from the product selection whims of owners and designers. As such, it can be quite costly when the structural support and needs of the selected materials are ignored as they often are.

Clean and suitable substrate takes on a whole new meaning when it comes to bonding materials. There are few if any floors that will not require some type of preparation. Any slab that is shiny is a problem. The majority of the bond issues I have reviewed for various claims have had some very shiny slabs. A good part of the bonding value in thinset mortar comes from the fact that it is installed over a rough absorbent surface, which has a substantial amount of mechanical bonding ability. How big of a difference can it make? In testing for claims involving several large projects this year, the bond strength was half of the advertised value when used over shiny concrete and tested in laboratory conditions.

Another indicator of the thinset bonding ability is a water drop test. If a few drops of water on a slab are not absorbed relatively quick you can be assured the bond of the tile will be compromised. My experience says in general the bigger the tile, the smaller the grout joint, the greater likelihood of problems.

Speaking of big tile, it is increasingly rare to find a floor that is flat enough for any size over an 8”x8”. With 12”x12” and larger tile, that often means some floor flattening is in order. Any educated tile professional can tell you building up a floor with thinset mortar is never a good idea. Standard thinset mortars lose a substantial amount of bonding strength when applied in a thickness in excess of 3/8”. While no specific product standard currently exists, products called medium bed mortars can be used up to a3/4” thickness. Using either product is a relatively expensive means of floor flattening when compared with the cost of a more appropriate material.

Last, but most important, is thinset coverage between the tile and the substrate. While not an encouragement to find out, you can avoid a host of other negative issues with good coverage. A short story: looked at a job with 18” tile in a car dealership, shiny slab, not a single movement joint in 125’ but did use a 12” strip of membrane over the control joints which was useless without a movement joint. The notched trowel was too small to get coverage on the back of the tile, and was followed by five heaping globs of mortar on the back of each tile. Floor was flat, real flat. Looked beautiful. They completed the job just as the area had a cold snap of historic proportions. As soon as it warmed up, tile was popping off the floor everywhere. Why? Shiny slab, no movement joints, and with five spots of mortar, no bond strength to restrain the floor from the inevitable movement. I asked the installer why the dots. He said there was no other way to get it flat. I would beg to differ. It would certainly be more costly in terms of time and setting material, but nowhere near as costly as it is going to be now that he has to replace the whole floor he hadn’t invoiced yet.

As large tile has become the norm, the tile setter’s life has gotten increasingly difficult, as the cost of time and material increases in getting the product properly installed. In the current competitive environment, it’s also accurate to say that getting a fair price for labor is a challenge. But I have never in my nearly 40 years in the trade ever seen so many jobs, and some really big jobs, go bad. In my opinion, the overwhelming reason for this, based solely on my observation, is poor coverage on large tile with tight grout joints and no movement accommodation, all of which are dismissed as important values in lieu of speed and cost savings. No job is ever perfect. Good coverage with the right thinset can be a great asset in keeping the tile where it belongs when other sacrifices have to be made. If you must make a compromise, don’t let it be good coverage with the right thinset and movement joints.

Perhaps one of these days, I’ll be able to write an article and use some of my “dot” or “spot” pictures. I have a whole file folder devoted to spot bonding saved in numerical order going from one dot through 12. It is a sad statement that I have accumulated all of them from looking at jobs that went bad in the past two years. The two primary installation shortcomings that will provide me work as long as I am willing and able to work will be poor coverage and lack of movement accommodation.

Save rolling the dice for the casino and make your very best effort to provide your customer with a good sound tile installation so you can afford to take that casino trip even if you do have to settle for a local one.