CTEF Tile Talk: Remodeling with Tile
September 27, 2006
We are thinking about remodeling; can we tile our...? That is the question we in the tile business want to hear in remodeling work. The statement is near and dear to the hearts of us all who sell construction products and services. Residential sales of ceramic tile have driven the market higher every year since World War II. Ceramic tile continues its higher percentage of growth against other floor covering products in the United States, and most feel we have a long way to go. With our current housing boom fading somewhat, remodeling continues to hold a substantial (or some would say equal) share of that residential market. Time has shown that when one segment goes down, the other goes up. The skill set for sales and installation of products in renovation projects differ from those needed for new construction. In new construction things are relatively cut and dried as you can visibly see how the floor or wall was constructed and easily select the appropriate materials for the job based on those observations. In remodeling projects, this is not the case. X-ray vision would be most helpful in analyzing the make-up of the structure and successive layers of flooring. Unfortunately, we on planet Earth do not have that option; we have to do it the old fashioned way: search and discover. Equally unfortunate, many do not search and discover, hence the start of profits slipping away due to unforeseen problems.
Successful and profitable renovation work requires substantially more knowledge and skill than typical new home construction. Consummate estimators, contractors, and installers up to the challenge are often difficult to find. You will find most prefer to avoid dealing with the inevitable unknown factors of renovation. Those with such fears should be allowed to steer clear of such projects so that they may be enjoyed by those of us with NO FEAR! Remodeling is where the money is for contractors and installers. Historically in my company, our margins for remodeling projects were double that of new construction. Remodeling projects typically use high-end products and are more material intensive than new construction. They require an estimator with above-average construction knowledge, and they must be familiar with local construction practices. Someone once told me you make all your money when you get the sale or job, the challenge is to keep it, there is no place these words are more important to remember than in remodeling.
We will explore some of the more common areas of concern in this article. Manufacturers have come up with a plethora of products to allow successful installation of ceramic tile under a multitude of circumstances however; first and foremost, the structure needs the ability to support the tile installation. In my days of estimating it never ceased to amaze me that very few of my competitors ever looked at what was supporting the floor or examined the structure of the current floor system. That amazement has continued in my current position. One typical type of call we receive time and time again is from an installer where the tile has cracked on a job. When asked what the span of the joist, spacing of the joist and floor panel thickness is you get this big silent pause on the other end of the phone. Usually followed by a "how am I expected to know that?" or an "I don't know, couldn't tell." There is not a floor out there whose make-up can't be determined by some means. Some are more work than other to be sure.
The most basic requirement of any tile installation is support. If we have a concrete structure, we are fortunate; there is no question of support if we are slab on grade. However, in most areas of the country we use wood construction including second stories in slab on grade construction. All wood framed structures are not automatically adequate to support ceramic tile. Building code requires that the floor joist support structure meet L/360, which coincides with the tile industry recommendation. However, from a tile industry perspective, L/360 applies to the entire floor, not just the joist, as we discussed in a previous article. There is no rating system in the wood industry in-place that rates deflection of the wood panels between the floor joists as we do in the tile industry. Installation recommendations relative to panel thickness come from the tile industry using the Robinson Floor Tester (ASTM C-627). All panel thickness recommendations are a result of this test. The proper method of installation needs to be determined by the supporting structure. Understanding few of us are qualified as engineers there are a few generalizations that can be made by the average person. The overwhelming preference of all manufacturers is a structure that has 16-inch centers. This provides a good base for subfloor panels. Most new homes today use engineered floor joist. The largest manufacturer of engineered floor joist also recommends 16-inch centers for tile areas. If the spacing between the floor supports is 19.2 or 24 inches, there are numerous methods that can be employed successfully listed in the Tile Council of America handbook. While it is always important that instructions of the manufacturer of specific products used in the installation be followed, it becomes critical at 24-inch centers. Floors with such wide spacing leave very little if any margin for installer error. In general, if you have a 2-by-10 floor joist, the maximum unsupported length can be 16 feet; for a 2-by-12, the maximum length is 18 feet. Engineered floor joist are much tougher to judge; it takes an engineer to make the calculation, hence the term. There has been little study in determining what the maximum length of an engineered joist can be. What study has been done resulted in a recommendation of 16-inch centers and a maximum length of 27 feet.
For discovery of subfloor and underlayment(s) thickness, the best place to go is the heat vents; gas lines and refrigerator lines work well too. If worse comes to worst you may even have to drill a hole to determine total thickness, but having to go that far is extremely rare. The minimum thickness for a subfloor panel to receive a backerboard is 5/8-inch for _-inch boards and _-inch for 5/16-inch or _-inch boards under industry standards. Membrane systems typically require _-inch subfloor on 16-inch centers but requirements vary by manufacturer for 19.2- and 24-inch centers. Subfloor panels should be tongue and groove. If they are not, which is not uncommon in some areas, the seams must be blocked between the joists. Quite often in remodeling we discover multiple layers of underlayment and flooring material. While it is obvious that we would remove any carpet, what do we do about layers of vinyl or hardwood? What if there is pressboard on the floor or two layers of _-inch luan and sheet vinyl? This is where experience comes into play. There are always going to be judgment calls to make on renovation projects and these can be tough areas to call. The world is full of opinions on going over existing floor materials. Different systems come with different risks. We have now reached a point in the sale where the manufacturer wants to sell a product, the distributor or store wants to sell it for them, the installer wants a job, and the customer wants a floor. The decisions made here will have a great impact on cost. Removal of existing underlayment panels of floor covering materials is going to affect cost, there may be environmental concerns, and the final price is totally dependent on the direction the installation takes from this point. There is no clear right and wrong in many instances, it truly becomes your perception of risk. So let's take a look at some "what if'" situations.
The opinions expressed from here on are mine and mine alone. They do not reflect any industry position though they draw on experience from many areas of construction trades. I base these opinions on training, experience, and thousands of floors I have installed as an individual along with skilled company employees over a 28-year period. Please note: I did not say all were successful; they were not. There is a learning curve to success in remodeling and I paid for some of my training with trial and error just like everyone else. The purpose of creating the following conditions is to create a thought process, not make a specific recommendation. So let's take a look at a few scenarios.
#1: The customer wants ceramic tile and has a great support structure but there are two layers of _-inch underlayment with sheet vinyl. What to do now? Well, from the vinyl side of the flooring business, they wouldn't be all that excited to put another layer over that assembly; it's too compressive. If it is too compressive for vinyl, it certainly is the same for tile. Cover with a layer of backerboard? That would make a bad situation worse in my opinion and I would anticipate some cracked tile or grout. Every successive layer adds to the problem; it really needs to come out. You cannot nail all of the compressiveness out of the floor, especially with an air gun, which is commonly used today. Air guns will not pull the floor tight. But, what about the potential for asbestos in the sheet vinyl? A local environmental lab can answer that. If it does contain asbestos the laws of your state apply on how the removal needs to be done. Everyone wants to put his or her head in the sand on this issue but there is no way to avoid it. If it were a single layer of luan with sheet vinyl it is still a bad idea; two layers is way too much risk. Do yourself a favor and let your competitor do the job. If you are contemplating a stone floor, you positively need to go all the way back down to the subfloor.
#2: The residence has _-inch tongue and groove sub-floor with a 5/8-inch plywood underlayment and sheet vinyl. Can you go over the sheet vinyl? Any setting material manufacturer will say certainly with our super top shelf new and improved thinset. The conditions: make sure the floor is firmly bonded (many are not), clean and de-glossed (requires scrubbing the floor), fastened correctly (different requirement for fasteners from sheet vinyl), and of the non-cushioned variety. The last one is really tough as almost all modern floors have a certain amount of cushion. How much is too much? You are entering a very gray area. If the floor has a heavy texture it is probably too much. Ideally, the sheet vinyl should be removed however concerns about the ability to completely remove it and the adhesive along with environmental concerns can make that a challenge. In many cases estimators gravitate to installing a tile underlayment over the sheet vinyl. Depending on the compressive nature of the floor, it's not a bad idea. Too much compressivness will result in cracked joints at the panel edges at a minimum. As for bonding directly to the sheet vinyl product, you should double the initial cure time before you get back on the floor to grout. Bonding porcelain tile to sheet vinyl is a very bad idea in my opinion and certain to fail at some point if the floor is not properly prepared. Never use mastic over sheet vinyl with porcelain tile under any condition.
#3: The residence has hardwood floors; the customer wants tile. As usual, it would be best to remove the floor and get back to basics but depending on how it was installed, it can be some serious work. Going over the floor with a tile underlayment would be what most prefer. This is a tough call; most wood floors tend to cup with seasonal changes, in some areas much worse than others. If the home is acclimated year around that helps some but then the concern is moisture coming from the basement, crawl space, or slab. Covering the wood is certainly going to alter the ability of the wood to breathe. Lot of "what ifs" to this condition. If cupping has been a problem in the past, the floor really should be removed.
#4: It's time to upgrade the basement or slab-on-grade kitchen that currently has vinyl composition tile bonded to the slab with black glue. Oh boy, here we go; this can be trouble with a capital T. First, from an environmental perspective, not all tile or black glue contains asbestos. Many types of glue did have some fiber as recently as 1985; VCT tile stopped a few years before that. If removal is considered, have the tile and adhesive tested by a local lab. Removal requirements vary by state and if this material ends up in a dumpster, be prepared to have the documentation in hand. Other than environmental concerns the caution here is how the adhesive is removed. Liquid adhesive removers are not a good option unless you have the ability to neutralize and flood rinse the floor several times. Adhesive removers attack polymers and it is these same polymers that we would be using in thinset to install the tile. Mechanical removal is preferred. There is power equipment available to remove adhesive down to the opaque layer the manufacturers desire. You cannot sand glue; heat reactivates adhesive and it will gum up whatever you try to sand it with. Once again, while removal may be preferred, it may not be possible for environmental or economic reasons. The preparation is then the same as sheet vinyl type products. The downside on going over vinyl tile in slab on grade applications is you are altering the vapor emission of the installation. It may have been there for 50 years and never exhibited any problems, until it was covered. On wood installations, an underlayment would be desirable. The cleaning prep may well trap moisture in the vinyl installation and then be covered with tile causing the floor to release. Depending on the nature of the tile, backerboard underlayment may fracture the tiles when they are fastened, making for a crunchy walk across the floor when the job is done. The thinset will mute it until it is dry.
We are about out of space for this issue. This is a never-ending subject, and I have tried to touch on the more common issues we deal with on a daily basis. There is no clear right way or wrong way in many instances when faced with sales and installation issues in renovation. If you would like to hear more on renovation or have interest in other installation related subjects for a future issue please submit your ideas or questions to the editor.