Thinset: So What's All the Confusion About?
March 29, 2007
Thinset! So just how confusing do you want to make it? There can be nothing more bewildering in the tile industry today than thinset mortar but grout is or soon to be a close second. How bad can it be? In counting the complete thinset product lines of just three major manufacturers there were 84 different varieties. This is a product where proper selection is vital for successful installations. How do you make a choice? Unfortunately, currently we must rely solely on the manufacturer’s representation of performance and suitability of use. Marketing and creating perceived value has become the primary means of selling many of the products produced today rather than independently verified technical performance. There are no independent lab tests to substantiate many of the performance claims. Some products sound too good to be true, and in my opinion, a few are exactly that. So how did this happen, do we need these revolutionary products, and are all the performance claims true? Some of the following explanations may be dry, dull, and boring but it would seem very prudent to understand just what it is you are selling or paying for. Maybe you’re paying too much, maybe your not paying enough. If you don’t read on, you may never know.
Industry standards for thinset, published in the American National Standard for Setting Materials, (ANSI) A118, contains references to primarily two cement based setting materials. ANSI A118-1 the American national standard specifications for dry-set (unmodified) portland cement mortar and ANSI 118-4 American national standard specifications for latex (modified) portland cement mortar, which includes polymer modified products. At the time they were created most installations were done on mortar beds. When used in residential applications, they were still primarily over mortar beds. I could be mistaken but I don’t believe the first backer board was even patented (1968) when the thinset standards were written. Most of the construction at that time used 12” centers in wood construction or concrete slabs to support the mortar beds. In the case of wood structures typically 2-by-12 sawn joists were the standard. Latex came about primarily as a means of enhancing both the bond and flexural value (how much it bends) when in high performance applications. Many tried latex products to bond directly to plywood, (OSB was unheard of), with very mixed results. This resulted in a search for cement based product which could be successfully bonded to plywood and resulted in the only additional thinset designation in the last 40 years, ANSI 118.11 American national standard specifications for EGP (Exterior glue plywood) latex-Portland cement mortar.
There are instances where a product meeting the basic requirement of a modified thinset, ANSI A118-4, would not meet the slightly more rigorous requirements of ANSI A118.11. The performance requirements for all these products are very basic and include;
• Initial and Final Set
• Open Time and Adjustability
• Sag (for wall mortars)
• Shear Strength (wall, mosaic, and quarry)
• Compressive Strength
The values required vary based on the type of tile tested. It would take more space than we have in this article for a complete breakdown. The testing requirements are about 4 pages long and have numerous parameters; besides, it is really boring. When manufacturers design products, they seek to meet or hopefully exceed the basic performance under the standards. There are a few things that are noticeably absent from the list of basic requirements. Currently there are no standards for shrinkage of thinset nor flexibility.
With current construction techniques and tile products there is a much greater demand on thinset mortar performance than the good old days. The reasons are numerous but we will explore a few of the more substantial issues.
Engineered Lumber, a case for increased bond strength and flexible-mortars.
In the good old days back when Mr. “I’ve been doing it for 30 or 40 years” started plying his trade construction of wood homes was a bit different. As mentioned earlier, 12” joist spacing was common. The lumber used often came from old growth forests which is much stronger than the lumber in use today. In the early days, it was all mortar over board floors, then plywood, then backer board over plywood. Currently we are seeing a dramatic increase in the use of membranes which may be the next evolutionary step. Engineered floor joist and OSB started to take off in the early 90s and this brought a new dilemma, long spans and dimensionally unstable material subfloor panels.
With long spans, much greater downward force is placed on the floor system, know as tension. This can also create more compressive force in the installation. Hence the greater need for movement accommodation joints and thinset products with greater deformation or elastic type properties and increased shear bond values.
Large floor tiles are a case for medium beds, porcelain mortar, contact mortars, and underlayments. Big tile presents a set of problems too. In today’s wood structures we need the properties previously mentioned plus we have several additional concerns. Whether wood or concrete I have seen few floors in my life that were flat. When using smaller size tile it will often conform to the irregularities of the substrate. With large tile, it becomes a much more challenging installation. Large size tile does not conform to an undulating structure. Whether wood or concrete, most floors need to be prepared beyond normal industry tolerances of ¼” in 10’ flatness for tile 12-by-12 and over. On may jobs a cementious underlayment is needed to bring the floor into an acceptable tolerance level. Rarely is this included in the installation price. Because the floor prep is often “free,” many often try to build-up the thinset under the tile using products they are familiar with instead of using an appropriate underlayment. All cement products shrink; basic modified thinsets shrink a lot when used to level floors. If the profile thickness of the thinset exceeds roughly 3/8 of an inch, it may shrink to the point of fracturing the tile. Standard thinset has greatly reduced strength in thicker layers, think making a road out of brick mortar. This is an excellent place for a medium bed product which is designed for thicker applications while maintaining its strength. If you are fortunate enough to be presented with, or get paid for, a floor flat enough to receive the ceramic tile with no further preparation a contact mortar may be just the product you need to get coverage without back buttering greatly reducing installation time. One last consideration on large tile is drying time. If you are using a large body porcelain tile drying time is greatly extended by size. Latex or polymer modified products need drying time for the latex to coalesce and harden. When used over a conventional membrane system, think Elmer’s glue between 2 pieces for plastic wrap. Until it dries and becomes clear in the center, the bond of the tile is at risk with any exposure to traffic. Thinset mortar made for large or porcelain tile recognizes this and is formulated for faster drying times.
Large tiles on walls are a case for lightweight and non-sag mortars. When using large tile on walls the installation concern goes to bonding and slippage. While non-sag mortars have been on the market for at least 30 years they have been used little. The mixing proportions have always been exacting to get that non-slip performance. Mastic is not a viable option for large tile on walls. It can not be built out thicker to adjust for out of square conditions nor can existing basic non-sag materials. Drying time with mastic and large tile is usually never, many never dry! Mastic was designed for tile up to about an 8x8, a few more expensive products can even go 12-by-12, but that is about it. In wet applications, mastic should never be used at all. Setting large wall tile with thinset required a much greater skill level than floors. The new lightweights and non-sag products have now made walls as simple as floors plus most can be used in a thicker profile to allow for truing of the wall, within reason.
Unfortunately there are very few of these wonderful properties that are covered under industry testing. This does not mean that these products do not work as advertised, with rare exception they do and we are very lucky to have them available. However, being an industry kind of guy I think it would be wonderful if we could establish a core base of requirements that truly reflect today’s installation needs. For instance, when testing to current industry requirements for shear bond a 4”x4” cut tile is affixed to a 4”x 6” concrete block. I see two things here. One, 4” tile does not represent how a 12” tile will dry and two, you can throw thinset at concrete block and it will stick. There is very little data using large porcelain over typical concrete, plywood, membranes, or any other substrate. There have been some International Standards (ISO)established for thinset products that are a vast improvement over our current American National Standards system but in my opinion don’t go far enough plus, we do not currently use any ISO standards for tile and related products in the US. Until those standards are established it will be difficult to for anyone to specify performance and understand exactly what we are buying. Till then, buy products form reputable companies you know and trust. Take a systems approach when ever possible, and please read the instructions, your not using your grandfather’s thinset anymore.