As the amount of ceramic tile purchase in the United States continues to grow dramatically each year, so do the problems experienced by installers. Columnist Dave Gobis explains the top problems associated with ceramic tile installation and offers tips on how to avoid them.

This shows a common action, troweling the floor with thinset and then "5 Spotting" the tile. Unfortunately it shows the appropriate reaction to lack of coverage on floor tile, bond loss.

You may recall last year's article in the FCI Trouble Shooting guide was Version 6. Seems a never ending challenge to try and solve problems so here is our latest version. It has been a busy and insightful year for me since our last Trouble Shooting edition. The type of complaints received remains consistent but perhaps their order has changed. Having recently done more field work than in the years past, one thing has become much clearer. For new construction in today's construction environments, pricing has gotten very competitive and time to do the job very scarce. Jobsite conditions are quite often less than ideal both technically and environmentally. This is really not a new development but it does seem increasingly prevalent. Despite this cheap, fast and nowmentality, top notch performance is expected. In giving the past year's problem make-up careful thought, it would be my opinion that many, but not an overwhelming majority, of the job problems we hear about are due to the current fast paced construction practices. It is also my opinion that the jobs that fail in this category are based on the choice to accept to those less than desirable construction practices as a condition of the job. This is a business practice known as risk management, it is a choice. In my days of dire financial need I accepted many of these types of projects. Never once, despite my hopes to the contrary, was I disappointed in my prophetic abilities, my perception was usually correct, they were better left for someone else. I came to look upon these jobs as loans, knowing that I would receive my check, but there would be an interest payment due at a later date, i.e. a callback. As the years progressed we weeded out these high risk accounts with ones that were more profitable. Nothing wrong with half the work for the same profit.

Exterior decks are always a critical application in waterproofing. The proper waterproof membrane was installed on this deck above and occupied space. But, when backer board was nailed into it, it was not very waterproof and resulted in substantial damage.

By far the most common type of failure is simply caused by a lack of understanding or knowledge. The editor ofFCIrecently informed me that I had offended a reader by saying most failures were caused by not reading the instructions. The first 28 years of my career it well could have been me making that call; I would go so far as to say I would be ready to poke someone in the nose over it twenty years ago. The last seven years have taught me differently. We simply do not receive calls about industry methods that don't work and rarely calls about products that don't perform. The overwhelming majority of calls, guesstimating around 80 percent are simply failure to follow instructions. This would include not only tile related practices but knowledge of those practices required by other trades for a satisfactory tile installation. There is nothing we do today, and not a single product we use, nor structure that we install in that is the same as 20 years ago. A recent seminar we did for Home Builders amply drove this point home. Our host related his phone started ringing the very next morning with callers saying "they did not know that." I might add some were nationally recognized home builders. Knowledge is power, power is money, enough said or I will go into one of my educational rants, which should be expected considering who pays me.

Here is this year's list in the order of calls received.
  1. Loose or Unbonded Tile
  2. Cracked Tile or Grout
  3. Workmanship
  4. Improper Use of Materials and Methods
  5. Moisture Issues

A close look is required but you can see a faint whitening of the stone in this picture. Further investigation disclosed the crawl space below did not have an effective vapor membrane and the OSB subfloor edges swelled in a 4'x8' pattern.

1. Loose Tile

The leading contender by far is delaminating or debonding tile. By this we mean tile that has come lose from the substrate (bonding surface). Sometimes the tile may sound hollow, have a grinding or crunching sound, be "tented" or possibly even come flying off the floor. Flying tile especially occurs on concrete slabs. The cause of these failures is primarily due to a lack of control or soft joints in the tilework and around the perimeter of the area combined with insufficient or poorly chosen bonding material. One reason this failure occurs is that ceramic tile expands and contracts with moisture and temperature at a different rate than a concrete slab. Generally speaking, concrete slabs tend to shrink as they cure. This shrinkage can take place for many years. If there is no place to allow a release of the tension created by this "differential expansion," the tile will come loose. Only the grout will be holding it in place and only for so long. When it goes it can really cut loose. We had one case where the quarry tile exploded off the floor with such force that it flew up in the air and stuck in the 9' ceiling. All structures have movement by design and all materials move at different rates. Buildings also move with seismic shifts, settling, heavy winds, material changes over time, and other factors. It cannot be stressed enough that control joints are required in a tile job of any size, even a 5'x5' bathroom. This can also be caused by excessive deflection covered in the next topic.

This is an exterior deck where roofing shingles were used for flashing. Great on roofs maybe, but roofs drain water; this deck held water. Once again, substantial damage occurred to the structure.

2. Cracked Tile and Grout

The second most common problem is cracked tile and/or grout. Frequently cracked grout is a precursor to cracked tile, although not always as there are many causes of cracked grout. Tile cracks because of movement of the substrate or cracks in a slab telegraphing through the tile. If a concrete slab cracks and a ceramic tile is adhered to the slab, the tile will crack too. Slabs may crack at any time but as years go by the likelihood of cracking decreases. The slab should be cured for at least two weeks or longer depending on ambient conditions. A slab that has cracked should not be tiled directly but may be if preventative measures are taken. Crack suppression techniques may include the use of anti-fracture membranes or unbonded mortar beds. Regardless of the approach, neither eliminates the requirement for control joints as mentioned in number one above.

Another major cause of cracked tile is excess deflection in the substrate. In the case of wood structures, floor joists or trusses most be rated for the load they carry and properly secured. Ceramic tile industry specifications call for deflection less than L/360. What this means that in any given span the deflection must not exceed 1/360th of the span. The most deflection doesnotoccur over the joists butmidwaybetween joists. If a span is 10 feet for example, then the deflection should not exceed 10-by-12"/360 or 1/3 of an inch. In the case of the area between the joists, that amount of movement can be as little as .040 to .060 depending on support spacing. This is usually engineered into the construction by selection of the proper sized and spaced framing members, subfloor, and underlayment and cannot be measured in the field.

One classic call the TCA received was about a large home in the Aspen ski area. This 6,000 square foot home had marble and ceramic tile through most of the floor area. The homeowner called and reported that tile was cracking in all rooms in the house. It had been tiled within the past year. We asked if he had talked to the tile setter to see what he felt was the issue. The homeowner said that the tile setter said it "was normal for tiles to crack and in fact the tile in his own house was cracked." Who would hire a tile setter that had cracked tile in his or her own house? This problem was traced to inadequate floor thickness, which includes the sub-floor and underlayment thickness in total. TCA Handbook for Ceramic Tile Installation in conjunction with the ANSI standards spells out the thickness required and the installation techniques specified. These recommendations are not covered by building codes. Of course the manufacturers' own specifications prevail over generic standards.

Drywall was used over an OSB subfloor. That might have even lasted a while except for the swollen edges of the OSB left in place under the panel and lack of coverage under the tile.

3. Workmanship

Third on our list of oldies but goodies is workmanship. This is a very broad category but includes visual results including grout work, tile spacing, layout, cutting, lippage, and other items that result in a less than desirable job. We find this category tough to deal with, as there is subjectivity to the homeowner's expectations. Sometimes the consumer demands too much such as very narrow grout joints or no caulk in the control joints. Designers specify wall washer lighting that highlight every variation in the tile surface. We literally have heard of consumers who have gotten down on their hands and knees and inspected floors with a magnifying glass. ANSI standards allow tile to be inspected at a normal floor viewing distance, about 6 feet. To inspect tile at closer range is unfair and uncalled for. Yet there are still many examples of truly unsatisfactory tile work out there. A customer has a reasonable expectation that the cuts should be balanced and the tile laying flat. Beauty is certainly in the eye of the beholder, but there is a lot of ugly tile work out there to go along with some true beauties.

Another exterior project with low grout joints and poor drainage resulting in efflorescence. With these conditions, any other moisture preventive measure would have been ineffective.

4. Improper Materials

Next on the list of the fabulous five, is the use of improper materials. Materials that are suitable for some uses may not work in others. Examples include the use of roofing felt as a moisture barrier, as an anti-fracture membrane, or a shower pan liner. Improper placement of certain types of backerboards is also bad. Bluntly put, there are materials designed for specific purposes. Proper shower pan membranes, high temperature membranes for steam rooms, floor backerboards, vapor barriers, crack suppression membranes, sound reduction materials and other must be selected and used properly. When one does, the manufacturer will stand behind their products, if not it's take your own chances. We can only assume that this consistent problem comes from a desire to use cheaper materials and to reduce the types in use. But this will create a potential failure and is false economy indeed. We have seen a very high number of complaints involving liquid waterproofing materials, mastic, and "pre-mixed thinset and grout" as of late.

A famous "Blue Dot" photo. If you been around awhile, you're bound to get these occasionally. In this one the installer used a standard thinset to build up the stone level throughout the job. The shrinkage caused the stone to fracture. It was very unfortunate because it was a beautiful job otherwise.

5. Moisture

Last on our list are moisture and the attendant problems when it encroaches into wall and floor systems. Incorrectly installed shower pans are a major component of this category. We are asked to define this method by phone, fax, and email weekly. This is not practical as it takes time to train an experienced tile setter as to how to properly install a shower pan, the mortar bed, and the tile. Some of the causes of failure are: puncturing the membrane during installation, failure to slope a shower membrane to drain, installing porous backerboards into wet areas (such as mortar beds and tub lips), failure to caulk joints that move (such as the junction of a tub with the tile work), use of the wrong type of caulk, and other lack of attention to detail that result in gaps in the system.

It is always a surprise to callers when we tell them that tile work is not waterproof. Tile itself generally is but water will get through grout no matter what type and how well it's installed. Either the top of the substrate or underlying areas need to be adequately waterproofed in wet areas. This is a key to preventing moisture from entering the building cavity. Just a drop of water a day can lead to dry rot, mildew, and other undesirable results. This typically takes quite some time before the leak becomes noticeable, at which point the damage is already substantial. Given attorneys seemingly field day with mold lawsuits, and exclusions that most insurance companies have made to policies, these types of failures can be quite costly.

Want to venture a guess at what would cause a crack exactly 3 feet above the tub? Backer board on both walls and floors should have the joints taped.

These are the top five but there are many more types of problems that the we are often called upon to resolve. TCNA members do want to provide assistance to anyone wishing more information, preferablybeforethe job goes in. Every single manufacturer member provides technical assistance via 800 number or email willingly on request. However, unfortunately, many don't call till after problems have occurred. Vast amounts of money are spent every year to provide a solid support network among manufacturers of tile and related products. They await your calls and want to hear your concerns and suggestions.