Ditra vs ProvaFlex
That’s right, I said typed it! Bring it on 
In the last month or so there has been a lot of hype about Loxscreen’s new polypropylene tile underlayment membrane. That’s just a big phrase for plastic tile underlayment. It is being marketed as a replacement product for Schluter Ditra.
Since I use Schluter Ditra almost exclusively as my preferred underlayment for floor tile I felt it would be a good idea to give this stuff a try. I did not do this in order to find a replacement for Ditra, I’m extremely happy with Ditra. I do feel, however, that due to the way this product is being ‘marketed’ to consumers and since it claims to be a replacement for a product I regularly use, people may want a professional opinion about the way it performs. AND! if there is something better out there – I want it.
There is also another very good reason I decided to do this: there happens to be one person all over the internet claiming this product to be better than a silk jockstrap. I’ll save that rant for the end of this post but suffice it to say that, at the very least, I vehemently disagree with his marketing tactics. If you would prefer to start with that rant please feel free to scroll to the end.
I will attempt to be as unbiased as I can as a firm believer in Ditra and the mechanical way in which it works. So to understand the key points of this comparison we should first begin with a basic understanding of that. I will just give an overview of a couple of key parts but you can read the official line from Schluter Here.
Tile being installed over Ditra
Schluter Ditra is a polyethylene membrane with square dovetailed cavities and an anchoring fleece laminated to the underside. The Ditra is attached to your flooring substrate by embedding the anchoring fleece in thinset (modified or unmodified depending on the substrate to which you are installing it). The dovetailed cavities on top are then filled with unmodified thinset and thinset is combed over the top and your tile is installed to it. The photo to the right shows the tile being installed.
The key component is the dovetailed cavity of the Ditra. See, the thinset does not actually adhere to the polyethylene (which, for purposes of not confusing anyone, and you know, not sounding like an uppity bastard, I will hereafter refer to as the ‘orange plastic’) This is completely normal and in no way compromises the installation – it is normal and on purpose.
Ditra utilizes a mechanical fastening rather than a chemical one. Thinset ‘sticking’ to the orange plastic would be a chemical bond to the plastic. Filling the dovetailed ‘waffles’ forms a mechanical bond by locking the thinset into the cavity.
Cross-section view of Ditra’s dovetailed cavities
The photo to the right is a cross-section view of Ditra. You can see how the cavities are angled back from the top opening of the cavity. The cavity gets wider as it gets deeper – that is the ‘dovetail’. After this is filled with thinset and the thinset cures it is nearly impossible to pull the thinset out of this cavity – it will not move, it’s locked in there. That is what I mean when I refer to a ‘mechanical’ bond.
A mechanical bond is the main reason I use Ditra. This will allow the ‘micro-movements’ in the substructure without transferring them through to the tile installation. The entire tile installation becomes one large monolithic structure which is able to move independently of the substrate. This means that the joists below your floor, or the concrete, can expand, contract and shift with the small movements inherent to structures without compromising the tile installation.
Or, more specifically and simply, when winter hits and your joists expand a little bit your tile and grout will not crack. Get it? Now I’m not talking typing about a major movement like a bulldozer crashing into the side of your house because I someone was doing roadwork while intoxicated. I mean the normal movements of any structure under seasonal changes.
That is basically how and why Ditra works. ProvaFlex, according to all the marketing and hype, is supposed to do the same thing. In fact if you read through all their (online) literature (yes, I have) it is nearly identical to everything Schluter has published about Ditra. So let’s start with the similarities.
Both products are marketed as an uncoupling membrane (this is what I’ve described above – the independent movements).
Both are marketed as having waterproofing ability when coupled with the respective band or tape for the seams. However, ProvaFlex is not recommended for use outside of a covered structure – you’re not supposed to use it on your porch.
Bottom of ProvaFlex next to bottom of Ditra
Both are marketed as having vapor management properties. This is a method of equalizing or dissipating vapor through the open channels beneath the membrane to prevent moisture build-up below your substrate. Confused? Nevermind, then, for the sake of argument let’s just assume that they both do that.
Both have an anchoring fleece laminated to the underside. The anchoring fleece is what is set into the mortar on the substrate to attach the membrane. They appear identical to me, and are identically difficult to pull off of the plastic membrane itself. I didn’t put a scale on it to gauge the difficulty, but it seems nearly identical to me – so let’s go with that. They’re the same.
Both are a form of plastic. Ditra is polyethylene and ProvaFlex is, well, I’m not really sure what ProvaFlex is. The online literature states it is polyethylene as well but the written literature included with the product states “Polypropylene material – Impermeable against other building chemicals. Easier to install and lays out flatter than polyethylene.” I do know that it did indeed seem to lay out a bit flatter than Ditra but that may be due to the smaller cavities in the top or the type of material – I can’t say which for certain. Someone apparently has their wires crossed somewhere and it is my opinion that it is the available information online because, well, why in the hell would you send erroneous information with the actual product? But, I really don’t know. There will be more about this in my rant. 
So it would seem that both products are marketed with identical benefits. But we all know that without actual hands-on use I could market a baseball bat as an environmentally friendly hammer – doesn’t make it true. So let’s move on to the differences.
Top of ProvaFlex
The ProvaFlex has an overlaying mesh or webbing attached to the top of it. I like this. It just seems like it would add more of a mechanical fastening to the membrane and stiffen up the installation itself. Whether that is actually true or not – I have no idea, but I still like the fact that it is there.
ProvaFlex with overlaying mesh removed
When I peeled this webbing off, however, I was a bit disappointed in how easily it was peeled back. I expected it to be nearly as difficult to remove as the fleece on the underside – it was not. It is attached fairly well but not nearly as well as I would have liked or been comfortable with. Maybe I’m just an anal bastard a demanding consumer – I don’t know. I still like it, though.
Ditra’s surface is comprised of 3/4″ square ‘waffles’. ProvaFlex’s surface is comprised of alternating 3/4″ circles and 7/8″ ‘flared’ squares – or whatever the hell you wanna call that shape, with raised circles in the center. Shape-wise (is that even a word?) I don’t think it makes much of a difference but I can’t say for sure one has the advantage over the other so I’m callin’ that a wash. I will say type that the ‘pillars’ created by the Ditra will have a bigger, more consistent footprint in contact with the substrate.
The thickness of both products seem identical but if you slam a micrometer on them I would guess the Ditra is just a platypus hair thicker. (A platypus hair is really thin, by the way. Don’t ask me how I know that, let’s just say it involved a midget and a case of scotch…or so I’ve heard 
)
So this far I would call it fairly even as far as a practical application indoors. If you don’t agree, well, start your own damn blog. I’m callin’ it even – up to this point. There is one major difference that is almost certainly a deal-breaker for me – the ProvaFlex cavities are not dovetailed.
The online literature and marketing hype – all of which compares it as an alternative for Ditra at a lower price – states, and my computer quotes “…with a grid structure of square, cut-back cavities…” Just a bit (ridiculously) misleading if you ask me. The shape of the cavities is most certainly not square and if by ‘cut-back’ they mean dovetailed – they most certainly are not that either.
This may or may not be a marketing ploy – it is not for me to say (yet) but the description can ‘technically’ be described as accurate if by ‘cut-back’ they mean the shape of the funny looking squares and by ‘square’ they mean the shape of the cavities in the z-axis, as it gets deeper.
Physical literature included with ProvaFlex
This, however, only seems to be on the online marketing sites. The physical literature included with the product itself specifically states “Square Cut Adhesive Cavity” pointing to a square (not dovetailed) cavity on the surface of the membrane. It does not seem to me that the Loxscreen company, the manufacturer of ProvaFlex, is attempting to pull the wool over anyone’s eyes as far as comparing it as an identical product in form and function. Not as far as the mechanical bonding process, anyway. It appears to only be the online marketing.
Cross-section views of ProvaFlex and Ditra
The photo to the right shows the cross-view of both products right next to one another. If you click on it and ignore my shitty photography skills you can plainly see the dovetails in the Ditra (top) and the square (or slightly rounded) cavities in the ProvaFlex (bottom).
If you can envision filling the cavities of each product with thinset and allowing it to cure then pulling straight up on each, what’s gonna happen? Without the mesh or webbing on the ProvaFlex that little fill of thinset will pull straight up and out – the Ditra will not, it is locked in due to the dovetail.
I’m no expert or anything (yes I am) but it appears to me that the ProvaFlex’s mechanical bond relies on the mesh or webbing attached to the top of the membrane. The method of the mechanical bond is different.
The bond for the Ditra relies entirely on a mechanical process, the aforementioned dovetails. The bond for the ProvaFlex relies also on a mechanical process, the webbing or mesh attached to the top of the membrane. This webbing, in turn, is attached by means of a chemical process.
I’m unsure what this chemical process actually is. I’m certain it is a trade secret and they would send black helicopters after me if I were to divulge it. I don’t know – I don’t care. It does not change the fact that the bond with ProvaFlex is not truly and entirely mechanical. The entire bond of your tile installation relies on the process and durability of the mesh webbing attached to the face of the ProvaFlex membrane.
So when you decide which product you want to use under your tile installation you should take into account the method of attachment. ProvaFlex seems like a decent product for a small application such as a small bathroom without excessive traffic or maybe a small countertop. It is entirely up to you whether to use it or not.
It will work to an extent and with certain applications. I just did two bathroom floors with it which I now own. That means if this stuff fails I’m paying to replace them, so it better not fail. But with any new product someone has to take that risk. If it does fail I will absolutely let everyone know – believe that.
This product will not be a replacement for Ditra for me. Anywhere I can use ProvaFlex I can use Ditra instead. I don’t believe the reverse to be true. If you install hundreds or thousands of dollars worth of tile do not nickel and dime the price of your underlayment. Saving one or two hundred dollars now may lead to replacing the thousands of dollars worth of everything later on down the road.
I am not saying it will fail – I’m saying I don’t know. With Ditra I know.
RANT-RANT-RANT-RANT!
This is not a rant against ProvaFlex, it is a rant against particular individuals marketing ProvaFlex on the internet. If you want to know who it is – google it, this jackass is all over the place.
I have no problem with marketing a product. I do have a problem with the method it is gone about. If someone needs to bash another company in order to sell their own (or one they are shilling) it is not only disingenuous, it is absolute bullshit.
The marketing descriptions and literature available online from this particular individual seems to be at complete odds with the product’s own marketing information. That is a big red flag.
This person apparently has had a disagreement or falling out with the Schluter company (as well as a couple of other very large, major companies in the industry) and has taken it upon himself to attempt to discredit everything about them.
Under the guise of ‘Schluter is crap and they screwed me so here’s a better product…’ he attempts to peddle ‘Prova’ products claiming they are better engineered and cheaper than Schluter products. They are cheaper, no question. Better engineered? You be the judge, that’s why I did this.
This guy bashes everything about Schluter, attempts to make false arguments such as ‘thinset doesn’t even adhere to Ditra’ (no shit) then turns right around and copies their marketing literature and substitutes his product in place of Ditra and Kerdi. THIS is where the inconsistency in things such as the polyethylene and polystyrene differences come from. Oops, forgot to change that part – jackass. The “…with a grid structure of square, cut-back cavities…” quote earlier happens to be exactly how Schluter describes Ditra – which is accurate.
The descriptions, methods, and even chemical makeup of the product, is at complete odds with the product’s own literature. Why would that be? I have absolutely nothing against Loxscreen or Prova-flex, hell, I tried it myself. I do take issue with assholes attempting to discredit one product in order to sell another. I believe if Loxscreen wishes to own a decent market share of this industry in the United States they should prevent this asshole from selling their products.
ProvaFlex may indeed be a product inspired by Ditra but it works differently, people need to know that. Attempting to force-feed the idea that it is an identical but cheaper product is bullshit. You are taking advantage of people inexperienced and uneducated about what the product should do for the purpose of profit. Please kindly go throw your own dishonest ass off a large cliff.
Please, please research anything you plan on using for a tile installation. Get more than one opinion – always. Even if that one opinion is mine (which is right, by the way) go get it from someone else. The better educated you are about it the better off you will be.
Now that you know how ProvaFlex works you can make a more informed decision about whether the amount you save is worth it for your particular application. If you want to use it and it sounds like the right product – use it. Just don’t buy it from the asshole.
If you have any questions please feel free to leave a comment. I’ll help if I can. Please understand I’ve only used this product for one installation so all my information is based on that. I have used Ditra (literally) hundreds of times. If you are the particular asshole I’m ranting about – and you know who you are – stop taking advantage of people’s inexperience and feel free to go find the aforementioned cliff.
Rant over. 
Hi Roger,
I’ve been reading up on the differences between ditra and pro-va flex underlayment. My concern with ditra is that you must use unmodified thin-set. We are installing 18″x18″ porcelain tiles which require modified thin-set. If we install ditra with the required unmodified thin-set, our warranty on the tile is void. I was told that if you use an unmodified thin-set on ditra, it will not dry…even up to 5 years and that modified thin-set is much, much better than modified. What are your thoughts? Thanks, Debbie
Hi Debbie,
Unmodified thinset will adhere your porcelain just fine to the ditra – 18×18 or not. Don’t know who told you that it may take five years for modified thinset to cure but that is ridiculous. I have used both modified and unmodified over ditra and they both work provided it is installed correctly. The basic differences in modified thinset are the stickiness of the higher modifieds (not relevant on a floor), the amount of flex it may allow (ditra addresses this) and the fact that the polymers will allow water to be retained in the mix for a longer period of time (ditra does not need this particular aspect since water will not prematurely evaporate from the mix).
Not sure what type of warranty you were told you have with your tile??? The ONLY warranty tile manufacturing companies give on their products, short of specialty made or glass tile, is on the product itself, meaning that it is properly baked, formed, glazed, etc. Once you put a drop of any type of thinset on it and adhere it to something every ounce of their liability is null and void. I’m certain you can read exactly that either on the box or on the manufacturer’s website.
Unmodified thinset will be more than adequate for your porcelain tile. Really. If you care to read any more of my thoughts on this specific aspect there are (literally) over 100 comments below ranging from friendly advice all the way to blatant verbal smack-downs. Feel free to peruse.
I have a question that I have meant to ask for a while and keep forgetting. I have a floor that I am about to tile. It will be 3/4″ plywood subfloor with 1/2″ plywood underlayment. The floor joists are 16″ OC and are 2×12 and are the older style full 2″ wide joists as this is an older house. Unfortunately the joists are not completely level in one part of the floor so the result is that one of my subfloor sheets is bowed downward somewhat. It is not much (1/4 to 3/8″) but I think it might be obvious once the tile is in since it happens over the 48″ width of one sheet. I foolishly did not catch this and fix it while installing the subfloor but realized it before putting in the underlayment. So I have decided to use leveling cement on top of the subfloor and under the underlayment. My question is whether this is the best way to do this and whether a primer is needed on the subfloor before putting on the leveling cement? The whole portion that I need to level seems to be only in the middle of one sheet so leveling cement will not need to cover any gaps between sheets. Thanks for the help.
Hey Buddy,
You can do it that way if you want. Whenever you use SLC you do want to use the primer. Oftentimes without the primer it will tend to crack and/or be unstable. Don’t know what you’re using for underlayment but if it’s something you need to screw through like backerboard rather than being installed with thinset (ditra) you may want to install that first, then use the SLC. It prevents the need to screw through your cured SLC.
I have read with some amusement the recent discussion on Provaflex. I will state up front I have no experience with either but have done some research on both in preparation for using one. Thats when I came across this site some time ago. The emotional ones always seem to be more from the Provoflex side as is this case. It seems to me that the subject of modified versus unmondified gets out of hand sometimes. While I profess to no experience, I am an engineer and see this type of thing happen all the time. You have two products that are similar in capability but someone latches onto one and thinks this is the ONLY way to do it when both will most likely do the job fine in many or most applications. Is one better than the other, most likely. But I suspect the differences between modified and unmodified are more like one will do the job and the other will do it not quite as well. It is arguable how much difference there is. Will most applications work with both? I suspect the answer is yes. It seems that there are those that seem to think this is a subject where it is all or nothing and it generally is not. It is a matter of degrees. Is Ditra better than Provaflex, I think it probably is. The dovetail from an engineering standpoint is sound and should provide better mechanical adhesion. Just think trying to take a cone with a solid substance inside and pull the substance through the small end.
What I do think is happening is that someone has become unhappy with Ditra and possibly some of their business decisions in the past and now this has become somewhat personal for them. I also think Schluter has possibly taken a little advantage of their arguable market leadership and keep their price high due to it and tried to tell retailers what they can/cannot do. That infuriated some but is their perogative. It is not a manufacturing cost issue as making Ditra or Provaflex is most likely a pretty inexpensive process. In any case, talking with some of the sales people that no longer sell Ditra, I see that it has become personal for them and they could care less about which product they sell and how good it is as much as they do not want you to buy Ditra. And because of that, they stress that you can not use modified with it whether or not unmodified or modified is significantly better, that is purely a sales tatic that is used to get those who buy into the significance of the two. They are obviously trying to make this into a discussion about modified versus unmodified and hope that people do not look at the facts.
Very well put Buddy. That conversation, by the way, was either that particular retailer or one of his minions. He likes to run around the internet and try to defend his non-sensical stance and doesn’t think anyone knows it’s him. Mildly amusing if it weren’t so assinine and childish. I do not defend Schluter’s operation model – I think it’s absolute BS. But I do like their products. I use both types of thinset over their products and don’t have any problems. If installed correctly neither type of thinset will fail, the differences are exactly what you’ve stated – a matter of degrees. It depends on what you are installing and where.
Thanks for your input – hope it entertained a little bit.
Hi Roger,
Really enjoyed your blog; love your style and the passion. Found it while researching ditra. We are in the process of low cost DIY kitchen renovation. DH is a quite handy (mostly self-taught) and does most of the heavy lifting in terms of execution, while I am the one with the ‘vision’ and do some researching as well.
Our next project is slate tiles for the kitchen floor. Learned from here and another site about the need to clean and then seal them (twice!) before installation. Which reminds me, is there any specific sealer that you would recommend for slate?
I’m sold on ditra. DH had not heard of it before. The cost might be an issue with him. The fact that the kitchen floor wouldn’t end up being much higher than the rest of the house (hardwood) will sway him, hopefully.
I have a few questions: We’ve already purchased the slate tiles. I found out after the purchase that slate tiles from India (which ours are) are softer/less dense. Would they still be okay to be installled in the kitchen? If I understand right, you would recommend modified thinset both under the ditra and between the ditra and the tiles (if we are not counting on schluster’s warranty) for slate. What is the basis of their recommendation NOT to use under the tiles, if the modified technically makes the bond stronger, do you know? The reason I am asking: is there any reason why using modified thinset would end up being detrimental?
The subfloor/joist advice thread was a little over my head. I’ll have to have my husband read it. Our house is about 50 years old. And I’m not sure what lies beneath the well worn and soon to be exiled linoleum floor. But the thought that we might need to put down two plywood layers before the ditra and then the slate is scary in terms of how high the kitchen floor would end up in relationship to rest of the house. Also the resulting height of the appliances in relationship to the granite countertop that is already installed. Does the ditra xl eliminate one of the layers of plywood, if necessry, without any impact on stabilty and strength or any other factor?
Sorry about asking so many questions. Thanks so for your time and I really appreciiate your expertise and your willingness to share it.!
Gratefully,
Sheila
Hi Sheila,
You only need one additional layer of plywood. The entire subfloor assembly needs to be a minimum of 1 1/8″ thick. Your current floor beneath the linoleum is likely already 3/4″ – you can simply add a layer of 3/8″ or 1/2″ plywood over that. Schluter’s recommendation (and that’s all it currently is) is due to the amount of time it takes for modified thinset to cure in the waffles (dovetails) when installed beneath something like porcelain. The modifiers in modified thinset allow water to be retained in the mix for a longer period of time allowing a stronger cure. Because water does not dissipate from the dovetails quickly there is really no need for the modified in that respect. That is the basis of their recommendation.
Ditra XL states that it can be used over a single layer of 3/4″ plywood – it is acceptable by the manufacturer. I don’t like installing anything over a single layer, but I’ve done it and it hasn’t failed. So it would likely be fine. You still need a properly stout floor for that installation, that derives from your joist structure.
My favorite sealer lately is miracle sealants impregnator pro. Very good stuff.
Thanks so much for your prompt response. and for taking the time to answer my questions.
That makes sense. I guess, it’s one of those “what was I thinking?” moments! It is not as if we were walking on thin air or clouds till now! It’s a relief though, because that means just one more llayer of plywood before the Ditra and the tiles. Just to clarify, is the 3/8″ or 1/2″ plywood you recommend assuming the use of Ditra xl vs the regular Ditra? The reason I ask: since we’ll end up with two layers of plywood anyway, can we use the regular Ditra instead, (to pacify my cost conscious husband) as long as it wouldn’t have any adverse consequences. Or do we have to use another 3/4″ plywood instead if we are going with the regular Ditra? Also, does the 1 1/8″ (I’m assuming minimal) total subfloor rule apply with the use of natural stone (slate) tiles as well?
Your explanation of the rationale behind Schluter’s recommendation of modified vs unmodified thinsets makes perfect sense, scientifically (at least to my not too scientific mind) If I understand you right ,unmodified thinset works just as well rendering the modified thinset unnecessary with Ditra’s unique features. But that begs the question: why the use of modified thinset nullifies Schluter’s warranty. Which seems to imply that modified thinset above the Ditra, according to them at least, seems to be not just unnecessary but is actually not recommended/is forbidden. It iis not a that big deal, I suppose. I really don’t need to have all the answers and I have to stop over analyzing things to death! Their reasoning could be as simple as ease of installation with non-modified vs modified and/or the lesser chance of amatuers messing up with it. On a related note, I recall reading here that with the use of modified thinset over the Ditra: applying it, filliing in the waffles and allowing it to cure overnight before laying the tiles. Did I get that right and is that something you would recommend with slate tiles?
Having read some threads on the need to beef up the joists (a week ago “sister joists” would have evoked a blank stare, from me) for natural stone tile installation here and at the John Bridge forums and learning more about joists and subfloors than I’d care to admit, I do have an inkling of what you mean by a “properly stout floor structure”. I’m sure my husband is looking forward to getting in our crawl space ( we have no basement) to take care that!
Couldn’t find Miracle Sealant’s Impregnator Pro; but with a google search found one by Stonetech. They also have an Enhancer Pro (a twofer: a enhancer and penetrating sealer in one) which might be just what I need since I want to enhance the color of the slate and seal it well. And if both can be done in one step all the better right? Do you have any thoughts on that product or think of something else which can do the same thing?
One final question (well at least till I think of something else): does the sealing of slate tiles make them significantly stronger? As I mentioned earlier, we have slate tiles from India (which someone on another forum described as being nothing but glorified mud or something like that, though he didn’t specify if he was referring to indoor or outdoor application) I hear it has a tendency to flake and break easily if something is dropped on it etc. But if it is so fragile wouldn’t it affect even placing/moving furnture over it, among other things? Which makes me question if going through all this trouble will be worthwhile if the raw material is apparently such a dud. And if I should I switch them out now rather than regret it later.
I’ve learned so much from here. For instance the reason behind the need for two layers of plywood and why the second layer is offset from the seams of the first. And that adding backerboard/Ditra is no substitute for it. Now if I can only make my husband listen. If all else fails I have one last secret weapon: pictures under the ‘Flawed’ section!
Thanks again.
Hey Sheila,
Nope, the additional layer of 1/2″ or 3/8″ with regular ditra. Yes, that applies to natural stone as well. A properly stout floor means simply that you have the proper deflection ratio for your joists – they are strong enough to support your installation. Basically means that it isn’t ‘bouncy’.
The problem with modified is that there are so many different modifieds out there – some of them take forever to cure in the dovetails and most DIY’ers don’t know the difference. If the installation is sealed up and walked on before the super-modified thinsets cure it will cause the tile to debond. Schluter is basically covering their ass since they gear a lot toward the DIY market. Everybody is lawsuit happy. I don’t agree with the manner in which it is argued about by many people – they think it’s a big conspiracy, I find it mildly amusing. For most applications unmodified will work just fine. Certain modifieds have better propertied, such as less shrinking while curing, which helps tremendously with polished marble or granite installations and the like, but it takes knowledge of the different products to use them effectively – and it will still void the warranty. It’s alla buncha BS.
Stonetech’s enhancer pro is a very good sealer as well – go with that. Sealing the tiles adds absolutely no strength to the tile at all. If you are worried about the stability of your slate then change it now before you install it. Take a couple of pieces and let them sit in a bucket of water overnight. If the water is muddy in the morning you’ve got the less-than-spectacular slate. You may want to change it. And I have about two hundred more pictures for the flawed section – one of these days I’m gonna update that page. (in my spare time…)
Hello, I am new to this tile setting thing and I am wondering what the issue is with the Ditra and the Provaflex.
I see the concepts of both but there is a guy on the net with like 28 some years of experince saying we should be using the modified thinset he says it is stronger and more resilient.
I mean I am no chemistry major but the better thinset sounds like a better idea. I don’t want my floor to fail.
He is also saying the provaflex out perfomed the ditra ina something called the robinson test, there is a pic of the machine and everything, here is a link: LINK REMOVED
I researched the test results and what he is saying is true.
Why can’t I use modified thinset with Ditra and how come this provaflex is showing greater strength than the Ditra in this test.
In my research I found that this test is like the only real ceramic tile test out there.
Please explain.
Hi Frank,
Let me begin by explaining what a modified thinset essentially does, in the most basic sense. Modified thinsets contain powdered polymers, or are mixed with liquid polymers. The polymers are added to help the thinset retain water in the mix for a longer period of time. Portland cement, the base product in thinset, cures through a process known as hydration. The longer the water stays in the mix the stronger the concrete (thinset) will be when cured.
Ditra contains dovetailed waffles which do not allow moisture to dissipate prematurely – the water is held in the mix. Essentially due to this fact the ditra does not require polymers to end up with a full-strength cure on the thinset. It’s simply not necessary – the dovetails do it.
As far as compression or (in the case of the Robinson floor test) sheer strength, modified thinset is not considerably stronger than unmodified after a full cure. The modified may allow more movement, but it is not necessarily stronger.
Concerning his ‘proof’, well, it’s very easy to convince someone without knowledge of the trade, products, standards, limits, etc. that these tests ‘prove’ that one product is better than the other. Had you read the entire post above you would have noticed in the ‘rant’ section details specifically about this type of behavior. It’s absolute bullshit – still! It’s simply another false ‘proof’ by someone with a chip on his shoulder attempting to shill his goods in a less than acceptable manner, in my opinion. I will be writing an additional post specifically about this particular ‘proof’ shortly. It will take much more than the short space available here, and I want a hell of a lot more people to see it. But I will detail some basic points.
I did read both reports. You can get both pdf’s by googling “tcna-046-03” for the ditra test and “359-10 tcna” for the prova-flex results. It should be the first pdf in the results. I’ll need to upload them both here to link directly to them, but you can google them and read them if you wish.
The ditra tests were conducted in 2003. The test over a concrete substrate passed all 14 cycles – an extra heavy rating and the same as prova-flex, however the prova-flex test did result in a chipped tile after 13 cycles. So basically they are identical.
The prova-flex tests were done in 2010. The concrete results are mentioned above. The plywood tests consisted of a substrate with 2×2 joists, 16″ on-center, and a double-layer of plywood with a combined thickness of 1 1/8″. In other words a proper substrate with correct deflection ratios for a tile or stone installation. (the 2×2 joists are representative of proper joist sizing in the robinson test, real joists should be 2×8’s or larger, or I-beams). The ditra test, however, were done over 2×2 joists, 19.2″ on-center, with a single layer of plywood totaling 3/4″! NOT a proper substrate for a tile or stone installation, and hardly comparable to the prova-flex tests.
HOWEVER! The first test, with ‘manufacturer A’s’ unmodified thinset still completed 13 cycles on the Robinson floor test! The second ‘manufacturer B’s’ unmodified completed nine. Says more about the differences in a quality unmodified and a five dollar bag than it does about the strength of the ditra.
So, when proper, quality thinsets (modified or unmodified) are used repectively, the prova-flex passed one more cycle with 16″ o.c. joists and a 1 1/8″ double layer of plywood than did the ditra with 19.2″ o.c. joists and a single layer of plywood. Which one do you actually think is stronger?
Another attempt to discredit a product someone does not like using bullshit tactics with people not educated in the nuances of the trade. You probably aren’t a chimney – don’t let him blow smoke up your ass. I encourage everyone to do their own research and read the reports for themselves. Ask as many questions as you like. If someone attempts to prove one product is superior shouldn’t they use identical testing parameters? Just because they use the same machine does not mean the tests were identical.
Did I explain?
Yes but the explanation does not compute.
I took these facts to a physics major I went to Clemson with, he is now an architect and he said it is clear that the unmodified thinset is a better thinset, he also said that this term “latex modified” is a very loose explanation since the modified thinsets today have chemicals that mimic latex, they are not true latex. He said true latex is hardly ever used in thinset anymore.
My concern is that if I use the modified thinset with ditra it will void the manufacture warranty, unless I use that ditra set stuff that is slightly modified, I also ask him about the slightly modified thinset issue, he said that the chemical components offer either Modified or unmodified, the term slightly is primarily bogus.
I am at an impasse, I do like the reinforcing mesh on top of the provaflex.
I also called the TCNA and ask about the test, they stated that the ditra test is old as you said, they also said they must stay neutral and cannot support one over the other.
They did however state the modified thinset was the premier application for any porcelain tile.
When I ask about the chipped tile they said it was due to a wheel on the test machine hitting the edge of the tile and had nothing to do with the underlayment.
I am a little concerned that you are offering a bias opinion.
I value your input, I want the best application for my job, I have 1356 ft to install. I want to be confident that this floor will last.
I saw you said you use modified under and over the ditra, but that voids the warranty and another guy told me that the modified thinset will take forever to dry at the inside edge of the ditra dovetail, he said this could lead to a green thinset application whatever that means. I assume it means that it will never dry in the ditra dovetail ?
Thank you in advance for your input, I must make a decision quickly my tile will be in by Friday. Either way this guy carries both styles I have to get it on the road right away, his prices are the best I have found.
What is your experience level ? How many years have you been laying tile? This info will help me decide what to opine to.
Frank
EDIT: I meant to say the modified thinset is the better thinset. I am overthinking this I think.
Bedtime!
Frank,
If you think I’m offering a biased opinion then why do you ask? I would argue that the person who rails against an entire company in order to make a profit from their competitor’s product would be the biased opinion – no? I’m not attempting to sell you anything, I have absolutely no vested interest in which underlayment you choose. All my comments on this site are my professional opinion. What you choose to do with them is entirely up to you. How and why would I be offering you a biased opinion? If you would like to know about my ‘experience level’, ‘ years I’ve been laying tile’ or anything else you may feel free to visit my ‘about me’ page – that is the amount of information I choose to place on the internet. The answer to both of those questions are on that page.
Forgive me if I seem a bit curt – I have been down this road plenty of times with this particular product, and it all derives from that particular website. There are currently five uncoupling membranes on the market – some of which allow modified thinset to be used in correlation with their warranty. I will give you a bit more information but most of the questions or points you bring up about the product itself are right in the post above.
There are several ways to make a thinset modified, the powdered polymers in the mix are one. You can also add liquid latex, which actually is latex, rather than water to an unmodified. Some manufacturers utilize latex – some utilize acrylics, others utilize many different types. The term ‘latex-modified’ encompasses them all. The shear strength or compressive strength of a particular thinset is not always derived from the modified or unmodified aspect. Laticrete 317 has higher strength in both than versabond. It is entirely dependent upon the particular thinset, not the modified aspects of it. While some (probably most) modifieds are stronger, some aren’t. I guarantee you that unmodified thinset bonds porcelain more than well enough. If you don’t believe that then use modified. I’m not here to talk you into one or another.
I know the chip in the robinson floor test was due to the wheel, that’s why I stated the results were the same. I did notice you make absolutely no mention of the wooden substrate differences. If your substrate is properly built your installation will last. If you are worried about the manufacturer’s warranty then choose the product with which you can use the thinset you want. It’s really that easy. Modified thinsets can be lightly modified. The simple black and white answer, which this particular vendor points out ad nauseum, is that it is modified or not. Lightly modified is indicative the particular type of polymer as well as the ratio of polymers to portland. And they vary – widely! So that simple little answer is another straw-man argument in an attempt to prove a biased opinion. And before whipping out the ‘physics major’ basis – if you were to read around my site a bit you would know that I actually used to be a physicist. And we argue with each other more than tile guys. That particular point your friend made, however, is also another of this vendor’s favorite ‘points’.
I can also tell you that modified thinset will cure in the dovetails. I know this because I build stuff then tear it apart to prove or disprove points – not because I’m selling something. The bottom line is use what you feel comfortable using. In my experience they will both be fine for your installation. As far as the long-term stability I can only tell you that the ditra will last long-term, I don’t know about the prova. I’m not saying it won’t – I’m saying I don’t know. Another point I made in the post above. If you think my opinion is biased, well all opinions by definition are biased. Mine is only because I have set literally thousands of square feet of tile on ditra – I know it will last. I wrote this to explain the differences in the two products – that’s it. It is for information purposes. I have given you my opinions, the decision is entirely yours. Perhaps you should consider a person’s motivation behind their opinions. What do I gain if you choose one product over the other?
Absolutely nothing.
All well said, I have however decided to use the provaflex, the longer warranty and the supported use of modified thinset is the clincher.
EDIT: Useless subtle character jabs and inaccurate comments removed for my sanity – you know, since it’s my personal blog.
Thanks for helping me reach my conclusion.
I have learned much, I hope we all have.
Thanks again
Frank L
I honestly can’t believe I’ve allowed myself to be drug into another useless conversation about this. You have now called me biased, unprofessional and emotional. Give me a break, Frank. If you actually read through all this and pay attention I have not slandered anyone. I’ve not mentioned names, I’ve removed links and have not said anything about the person himself. The issue I have with him is the manner in which he initially marketed his product – that’s it. I have not spread lies about him, like he has about me purchasing from him. I have never purchased one thing from him, I bought the product at my local supplier. So please, give the subtle little jabs a rest.
Also, in case you haven’t noticed, this is a personal blog – I don’t sell anything nor am I attempting to talk you into anything. I simply give my professional opinion which is based on my experience. I guess nineteen years in the industry isn’t long enough to understand it, though. Custom has one of the products I mentioned previously – it isn’t ‘like’ prova-flex. Or, it’s a knockoff. Or it’s the same – all the things that can be said about prova’s relation to ditra. Incomplete comparison and another non-point on the selling path, no?
I’m glad you’ve reached a decision. I hope it works out well for you and I hope I have helped in some manner.
The Prova also test used a 11.5mm thick tile with a break strength of 1000 pounds (!). I don’t see anything about the tile used on Ditra, other than it is porcelain, but maybe not so heavy duty?
Hi Roger,
Great write-up! I was about to use Ditra until someone at a distributor recommended ProVa-Flex over Ditra. That’s what got me to researching the difference between the two. They said they only support and install the ProVa-Flex underlayment as it operates with the same principals as ditra, it has a stronger rating than ditra in the Robinson Test, and it can be installed with modified thinset as opposed to unmodified which they don’t like. I see you already answered the question of which thin-set to use during one of your previous posts so that takes care of that problem. I will use modified for under and over Ditra as well as in the dovetails (unless I misread your previous post).
They also said they have heard of issues with the dovetails collapsing from the force of the installers knees while installing the tile while this is not the case when using ProVa-Flex. Have you ever experienced that???
Thanks!!
And one other thing….I think I’m ok but wanted to run it by you first for your opinion. I have 3/4″ plywood down on 2×8’s every 16″ with room 10.5′ x 21′. I also added footers every 3 feet between the 2×8 joists for extra support. That should be fine to lay the Ditra right on that, right? I was hoping to get away with not using any expansion joists as well. I’ve read that subloors should be at least 1-1/8″ thick. I’m assuming that doesn’t apply when using Ditra and is usually the standard to go by when adding cement boards on top of plywood, right?
Thanks again for your help!
You still need two layers of ply beneath regular ditra. Ditra-XL is supposedly rated for use over one layer of 3/4″, but I still don’t like it. At the very least you should have another layer of 3/8″ ply over that offset from the current floor’s seams.
I’m not using natural stones. Do I still need two layers of plywood if using ceramic tiles (16″ x 16″)? I thought Schluter only required a single layer for ceramic tiles. You would add another layer of ply then the ditra as opposed to just adding a layer of cement board? It’s worth the extra work AND money?
Thanks for your suggestions!
Yes, you still need two layers. If you were to use backerboard as your substrate you would still need two layers. Schluter only requires a single layer under ditra-XL, I believe regular ditra still requires two layers. I know the standards do, whether ditra does or not.
The reason for the second layer is to stabilize the weak points in the single-layer – the seams. That is why it is offset. The seams over a joist create a ‘fulcrum’ of sorts, offsetting the second layer shores those up and stabilizes the movement over the joist when a load is placed between them (walking). Ditra DOES NOT compensate for vertical movement – only in-plane movement.
Yes, it is well worth the time and money.
Hey Steve,
I have experienced the crushed waffles. It happens when you aren’t paying attention to where you’re kneeling and land on tools and such. It rarely happens with normal knee-pads and mostly with the hard-surfaced pads and my pro-knees. And it happens with both ditra and prove-flex.
Roger
I live in San Diego. I have an old concrete patio slab-on-grade with cracks. Some are hair-line, others show movement both horizontal and vertical. The patio has a roof so it is protected from the rain. I suspect that I get some movement in the slab with subsurface moisture changes through the year, but I believe it would be minor. I plan on installing 18×18 porcelain tile on this slab.
What underlayment and installation options (Laticrete, Schluter Ditra, USG, Mapei…) would you recommend for moisture protection and crack isloation?
-Donald
Hey Donald,
Sorry to be the bearer of bad news but I’m gonna have to be. If your slab has cracks that are not in-plane, in other words vertical cracks where one side is higher than the other, then your slab or the foundation beneath it needs to be repaired. Crack isolation membranes of any sort are only applicable to in-plane cracks – nothing solves the issue of vertical movement short of slab repair. You need to find out why it is doing that and have that repaired – then use a laticrete installation system on your patio. You may get away with Schluter but I don’t use Schluter systems outdoors here due to the freeze-thaw climate in winter. It may be viable out there – I honestly don’t know.
Hi Roger,
Great info on your website, but I still have a few questions. We’re building a house in Colorado Springs and I want to put a tile deck over the patio. My builder doesn’t do tile decks, so I’m doing it. (I’ve done some tile work in the past and am now learning of my many mistakes . . . the tile I installed still looks good, but I sure didn’t do it right).
I was planning on using TI ProBoard as the tile substrate. Long story, but it isn’t available yet this year and may not be due to financial concerns. So now I’m looking at Schluter Ditra or Ditra XL. The deck is about 36′ long by 14′ deep with 16″ O.C. 2×12 joists. Would 3/4″ Advantech OSB covered with 1/4″ cement backerboard be sufficient or should I double those thicknesses? If the 3/4″ and 1/4″ materials are OK, would it be wise to use Ditra XL?
Why does Schluter recommend using ring shank nails to attach the OSB/plywood to the joists instead of screwing it down? Does it matter in your experience?
Thanks for your website and answers. I really want to get it right this time.
Thanks, Mike
Hi Mike,
For an outside deck here in Colorado I actually prefer not to use ditra. While Schluter does say it can be used in freeze-thaw conditions (and they have done it) I simply don’t like the fact that water can get into the waffles and freeze there. The expanding water in the waffles just makes me nervous.
For decks here I use two 3/4″ layers of exterior grade plywood, 1/2″ durock covered with Laticrete Hydroban or 9235, then tile. You need to ensure it’s sloped correctly at 1/4/ft. (but you already knew that) as well as properly lapped flashing from the structure where it meets the porch – make sure your flashing is above your waterproofing rather than below it. I have always screwed the first layer of plywood to the joists and the second layer only to the first – not through the joists. To be honest I have no idea why Schluter would want ring shanks rather than screws – maybe it’s just there to prevent people from using just regular nails? Not really sure.
Roger,
Thanks for the quick reply. I was wrong about the ring shank nails. I called Schluter and they explained that I had misread what it says on note 1 on page 23 of the Ditra installation manual. Screws and ring shank nails are OK for attaching the subfloor to the joists, so they got me straightened out on that.
I’m surprised that you don’t recommend Ditra here in Colorado due to the wide and fast temperature swings. I have no experience in this matter other than hours of reading and confusion (ref. my previous paragraph). I’m really concerned about the expansion/contraction since the deck has a southern exposure. Would it be a waste of money to install Ditra for “insurance?” How long have you used the double layer construction mentioned above? Have you done some here with Ditra? How do they stand up? Obviously, I’m a bit confused again as I believed a shear membrane was required for any outdoor tile installation especially here in freeze/thaw territory. I spoke with my builder’s engineer today (adding further to my confusion and shaking my confidence) as he said he’d never seen a successfully tiled deck here in Colorado. He was mainly concerned with the structural aspects, but also with dimensional stability. (He’d never heard of Schluter or TI Pro Board, and is still skeptical . . . as engineers and builders always are of the unfamiliar/unknown, but he acknowledged that bias). Is my structure sufficient, in your experience, to support your construction method and still support a normal load of people and furnishing on the deck? Also, with 1 1/2″ plywood, 1/2″ cement board plus tile, the surface of the tile will be above the bottom of the wall. (Builder installed the ledger board 1 1/2″ below the bottom of the upstairs stucco wall with the flashing coming down behind the stucco wall and bending 90 degrees over the top of the ledger in anticipation of installing TREX). Would it be OK to leave at least 1/4″ gap between wall and tile and fill the gap with caulk as I won’t be able to move the existing flashing?
Thanks again for your advice.
Mike
Hey Mike,
I have done ditra decks here and haven’t had problems with any of them – the oldest being about eight years old. It works, it’s simply a matter of my brain tearing apart every little aspect of an installation and trying to anticipate what MAY go wrong – and steering away from it. I’m simply more comfortable with the double layer method with Laticrete membrane (which also acts as a sheer membrane as it was originally intended for use as a crack suppression membrane). The ditra will work, I’m just more comfortable with something else.
Your deflection ratio is more than enough for a deck that size. And yes, you can leave a gap there for your flashing. And stop talking to engineers – they’re drama queens.
I’ve tiled plenty of decks here in Northern Colorado and haven’t had one fail. People don’t realize that an engineer is rarely looking at decks that haven’t failed – that’s not their job – their job is figuring out why one did fail. That’s likely why he’s never seen one successfully tiled – he doesn’t look for them.
Either method will work, I just err on the side of caution when I’m working on someone else’s house.
Hi Roger,
I had an interesting talk with a local builder/developer who installed a tile deck on his home that he is renovating in Monument, Colorado. (I was put in touch with him through my tile source). He agrees with you that maybe water could get into the waffles of ditra and cause problems. In fact, he felt that using a waterproof membrane of any kind under the tile introduces the possibility of trapping water under the tile and causing a failure. He proposed the idea of drilling holes through his durock/wood substrate about every 18″ along his grout lines to allow water to seep out below. (What do you think about that)? He happens to need to replace his grout anyway because the grout he installed this past November is failing. I asked him how it’s failing and he said it’s “dissolving” and turning white (it started as “espresso”) and coming out in “little strips”. His tile installer used urethane grout. I told him that another tile pro told me that urethane grout is a “no – no” for outdoor use since it will essentially fall apart. What kind of grout do you recommend? I told him I’d ask “The Floor Elf” these questions since you do tile decks in Colorado. He’s really bummed since he’s gonna have to remove a lot of grout and replace it, but like me, he really wanted a tile deck and loves the look (except for the grout). We bounced several ideas off each other but don’t have the real world experience to know if they’re any good. Can you help us out?
Good point about talking to engineers. The one I talked to did exactly what you said . . . investigated the cause of failures.
Thanks,
Mike
Hey Mike,
Not sure I like the idea of drilling holes through the plywood substrate. That would open the layers of the plywood substrate through every hole. I don’t like exposed plywood layers – exterior grade or not. I have built decks with Laticrete a lot of times and haven’t had problems with them. I seal my substrates completely around the edges as well and use a highly modified (read expensive) medium bed mortar. I use a 1/2″ x 1/2″ ‘U’ notch trowel and comb all my thinset lines down the slope of the deck. When you collapse the ridges to get full coverage you will end up with little ‘channels’ of sorts next to every ridge – this gives water beneath your tile a path to drain if need be. I also fabricate the ends of my deck to properly drain moisture from beneath the tile as well as on top of it. It is a ‘system’ of sorts that I’ve developed as a deck installation method.
The key to a successful deck tile installation is simply water management – not water ‘proofness’. When water hits the deck it does not get to the substrate. All I do is develop a way for water in any space above the Laticrete membrane a way to drain or dissipate. If you remove the chances for water to affect your substrate in any manner (Laticrete membrane) you eliminate half of the possible issues – after that it’s simply water management.
The ideal deck would have porcelain tile installed correctly (method above) with epoxy grout. You need to use an epoxy rated for outdoor use – that’s imperative. Regular Spectralock is not – I believe their commercial 2000 may be but I would have to check on that. If it is a deck with southern exposure that also needs to be taken into account with the brand of grout – direct uv rays play hell on color pigments. This would help eliminate as much moisture as possible getting beneath the tile to begin with, then allowing any that did to drain down the deck and out. Provided your deck is structurally sound enough for the installation as well as traffic it will last until you get tired of looking at it.
The ‘white’ on the grout you’ve mentioned is likely efflorescence due to water retention. Once it does evaporate the minerals are left behind on the surface and are nearly impossible to get rid of, especially on cementious grout as it will get into the pores of the grout. That may also be due to movement in the deck itself causing the grout to crack and retain water more than normal. I don’t know if the cracking is due to movement or simply the type of grout, I’ve never used regular or urethane grout on an exterior installation. Just so you know – even though you don’t care
– I think urethane grout is absolute garbage. I’m more than willing for a company to change my mind on that by allowing me to run theirs through the gamut, but until then I wouldn’t install it in a dog house.
Thanks in advance for your help and all the info you give us.
I’m remodeling my farmhouse kitchen circa 1886 in upstate NY. Temps swing from 95 to below zero. New 2×10 joists @ 16″, 3×4 t&g, 1/2 ext ply ( done), ditra, and ceramic plus under subfloor (thermofin plates) radiant heat (yet to do). About 300 sq. ft. I’m doing all the work myself and I’m up to the ditra install.
Should I add a concrete backerboard at this point to increase mass for the radiant? Relatively inexpensive (about $0.70/sqft).
Modified below and above the ditra? Best modified to use?
Do I tile the entire floor and add cabinets or tile up to the cabs after? I’m hearing both. Cabs will form an “L” shaped kitchen against 2 exterior walls. No penninsulas or islands.
Thanks again.
Hey Larry,
No real reason to add the backer – your floor has more than adequate deflection as is. Backerboard is not used for structural support in any form – mainly only as a suitable substrate to install the tile. Since you’re using ditra there’s really no need for it, but you can if you want.
Technically you need modified below ditra and unmodified on top. (and yes – I use modified on both – I’m an ass like that). The most readily available decent thinset would be either versabond from home depot or mapei brand from lowes. They’re both good thinsets. If you choose to use the unmodified above ditra use the mapei unmodified – the more expensive one, they usually have two different unmodifieds.
Tiling up to or under the cabinets is mostly a personal choice. The only real reason most people only tile up to the cabinets is for easier future removal when the tile is changed again. I usually prefer to tile under them if they are already removed or new ones have not yet been installed – it’s a lot quicker and easier. If you tile up to them take into consideration places like in front of an under-counter dishwasher – make sure it can still be removed and replaced or services if needed. You don’t want to trap them in so that the only way to remove them would be to remove tile.
Roger, thanks for the reply and info.
I will assume you assumed the ass on this end typed 3×4 t&g instead of the correct 3/4″ t&g. I don’t really have 3×4’s on the floor. Although if I did I could park my truck there, or my mother-in-law could visit.
I planned to tile the whole floor. I like quicker and easier.
I read where you had someone put mod under the ditra, then fill the top squares level with unmod and let dry, then put mod on to set tile. Do you not do this yourself because of the extra time involved or it’s just not necessary? Or, and I know you don’t care about the warranty, does this keep the warranty valid and still let you use mod on the tiles?
Thanks, Larry
Hey Larry,
Yes, I did assume that. I’ve never seen 3 x 4 t&g myself but I’ll bet it’s some sturdy stuff.
I won’t tell your wife you said that about your MIL – hopefully ditra can handle that particular, ummmm, ‘load’ for you. 
Unless you use unmodified (entirely) on the top of the ditra to set tile your warranty is void. As long as you get good coverage on your tile – and backbutter it, you don’t really need modified, unmodified works just fine if you’re concerned about the warranty. If you aren’t concerned about the warranty use modified and give it a couple of days to cure. You could also fill the waffles with modified, let it cure overnight, then set your tile with modified. I don’t do it mainly because of time – and I’m a rebel like that.
I’ve been laying tile for about 5 years but as with anything else I always take opinions from those with more experience. That being said, I’m fairly new to Ditra I’ve only used it on 3 jobs all turned out great and I’m sold. I’m about to lay marble in my own house and I’m a little up in the air on which direction to go. It’s a 12×12 marble being laid on 3/4″ advantec. I have 2×12 joists on 16″ centers. Transitioning into the next room will be a 5/8″ hardwood and would really like to avoid an extra layer of subfloor under my Ditra because of the height difference that I’m already getting from the marble. What would you suggest? Thanks, Ryan
Hey Ryan,
IF the unsupported span of your joists is 12 feet or smaller, in other words how far your joists go before they have a vertical support, then you meet the minimum deflection requirements for marble. That requirement, however, takes into consideration a double layer of plywood as your subfloor at a minimum thickness of 1 1/4″. If it were my house I would likely add an additional 1/2″ layer of plywood beneath all of it – the tile and wood – and end up with the same height difference you would keeping just the single layer. Simply build up the entire floor.
With that said you can use Ditra-XL which is just like ditra except, ummm, bigger – XL.
It is twice as thick as normal ditra but the manufactures specs state it can be used over a single layer of plywood substrate. So that would be an option. Although you would gain just a little bit of additional height with it I don’t think I would try marble over normal ditra above a single layer of subfloor – advantec or not.
The striations in marble create weak points throughout the stone. This simply means that it will crack more easily in those areas. You want your floor as strong and stable as you can possibly get it. The double layer of wood is your best option – Ditra XL would be the next best.
Thorughly enjoyed reading all comments. My question is what type plywood should I use on an outdoor deck in Florida. Pressure treated is all uneven and would prefer not to use it. Would exterior grade plywood be OK and do I need 2 sheets of 3/4 or one 3/4 and one 3/8. Should I use one 3/4 Pt plywood below and regular exterior grade as top layer under ditra? I am concerned about high humidity. I read somewhere that 1 1/8 is the recommended minimum thickness for a plywood base for ceramic tile? I plan on using 2×8 for beam and 2×6 on 16 in centers. deck size is 16×25. using PT for posts and beams. Based on your recommendations, I plan on using Ditra.
Ed
Hi Ed,
You should use a good regular exterior-grade plywood as both layers – 3/4″ each for a total of 1 1/2″. The 1 1/8″ recommendation is for interior spaces with proper deflection rating. And that is the minimum – just like any other code you’re likely to run across. With floors over-building is rarely a negative factor.
You did not state whether the outdoor deck is over a living area (interior living space) or not. If so it requires much, much more planning, water control, and substrate preparation than simply two layers of plywood and ditra. If it is just a patio with nothing beneath – that’s fine. You can read all about outdoor patio, balcony, and deck spaces on Schluter’s deck page. It will describe what you need according to your particular application.
Your joist plan sounds solid but you may want to consider either cutting the unsupported span in half (I’m assuming you only have the 2×8’s supporting at each end) or using 2×8’s as your joists as well rather than 2×6’s. Again, depending on your application this may not be necessary. I simply like to build decks to withstand hurricanes. Up here in Colorado that’s just a joke – down there it’s legit, no?
I am about to start my first tile project and have a few questions that you may be, possibly, hopefully… can answer. I’ll even buy you a beer the next time I am in your area for your insight.
Anyway, I am going to tile a bathroom. I plan to use Ditra under my tile and probably follow the guidelines they provide on using modified on the plywood and unmodified b/w tile and ditra. I have a few questions for you while you are between drunken rants. 
1. How important is the V notched trowels that Ditra recommends?
2. Reading your rant err discussion on prova flex vrs ditra I see the detail that I had missed on how ditra works with respect to bonding. I had wondered if it were something like a floating floor where the ditra would give and that seemed very illogical to me. But now I understand the dovetail and the mechanical bonding aspect. So my question is whether getting the thinset completely into the dovetail is something to be concerned with and if so, how do you make sure?
3. What are your thoughts on the 1/4 board thing where you put the underlayment (I will have a subfloor and a layer of plywood underlayment as well) such that the edge is not above a joist as recommended by Schluter?
4. How about the screw from the underlayment to the subfloor, into a joist or not?
5. Using ditra, is BC plywood necessary like you hear for non-ditra applications? Can you really use lower quality plywood or OSB instead?
6. I am concerned with the ditra becoming unlevel during installation since it will be put onto wet thinset. Is that really a concern or will the thinset above the ditra equalize any level issues when you level the tile?
7. Being a bathroom, what are you thoughts on putting the strip of waterproofing on the seams? Is it really that helpful if all you expect are accidental spills upon occasion?
8. Being a beginner, what non-obvious tips would you give or tools that you suggest that might not be normally known without learning through experience?
Thanks for your help.
Hey Buddy,
1. The Schluter trowels actually are very nice to have – they lay down the perfect amount of thinset. That said, as long as you can get a hold of a v-notch trowel very close to those sizes it will work just fine also.
2. Yes, you want to ensure you have the dovetails filled in every direction. The best way to do this is to run the flat side of your trowel over the dovetails with thinset in all eight directions – left, right, up, down, diagonally in all four directions. This will fill your dovetails. Then just flip over your trowel and comb your thinset over it for the tile.
3. You want to ensure that the edge of the top layer of underlayment is not above a joist. Your lower layer edges will obviously be on a joist. As long as you offset and stagger all your joints by at least four inches you will be just fine.
4. Bottom layer into the joist, top layer only into the bottom layer – not into the joists.
5. BC really is the lowest grade you should go. You can (‘they’ say) go with lower grade of even osb but I never, ever use osb and rarely less than BC grade. The voids in lower grades are where you may run into problems and you may not even know they are there until you begin to have problems.
6. The amount of thinset – thickness – is miniscule enough that it will not really throw your ditra out of level. Any areas that are not exactly flat can be made up with thinset above the ditra beneath the tile.
7. You can use kerdi-band to waterproof ditra by installing it on the seams and around the perimeter if you choose. It is more of a personal choice than anything. Occasional spills or normal water on the floor from stepping out of the shower, etc. are not going to affect anything in your floor negatively.
8. Take your time! Only mix as much thinset as you think you can reasonably use withing the given curing time (it’s on the bag of thinset – and it’s accurate) Use a straight-edge or level to embed your ditra into the thinset – this will help get it flat, then let it set overnight and cure before installing your tile. Make sure your dovetails are filled – again, take your time.
And have fun.
Roger,
One other question I forgot to ask is do you have any experience with or know anything about Dural Durabase? It seems to be very similar (like Prova Flex) to Ditra.
Roger,
Sorry for my confusion but I do have another issue that I just realized may be a problem. The bathroom I am tiling currently has a large jacuzzi in it that will be removed. During the demo, I have found this jacuzzi has a large cement pad under it. This was unexpected and now I am unsure how to proceed. Obviously I could demo the pad and add back floor joists but I was hoping to just tile over it but that raises the question of how do I transition from the plywood subfloor to the cement pad. I am concerned about the transition point being a weak spot and creating a stress fracture line. One option is to build up my subfloor/underlayment in the other part of the room such that I can either put underlayment or cement board over the whole thing providing structure that crosses over the transition. But is it a problem to just have the wooden part of the floor butt up to the pad and put thinset over that and then ditra and will the transition be an issue? Suggestions?
Hey Buddy,
I would likely build a frame around the concrete pad (providing it is at the same height as the wood floor in the rest of the bathroom) and leave a transition gap of about 1/8″ for expansion purposes, seasonal changes, etc. since those two materials will expand and contract at different rates. I would then use DitraXL which is ditra but bigger – twice as thick to be precise. This allows more support and would eliminate any problems you may have with the transition between the wood and concrete.
Ideally you should bust out that concrete enough to install a wooden subfloor like the rest of the room but if you don’t want to do that (and honestly – who would
) the ditraXL should work fine for you. Just make sure you use a good modified thinset to install the ditra to the two substrates. And yes, regular ditra should work, but I wouldn’t sleep as well at night if I did it that way.
Hi Buddy,
I have not used the Dural but I have examined it. It seems nearly identical to the ProvaFlex and appears to work in the same manner.
Hi Roger
I’ve pretty much read your entire website and feel like I know almost everything there is to know about tiling…and sarcasm ;-)
I have a small problem though. I am in South Africa and I have asked many tile stores regarding a tile underlayment membrane like Ditra. No-one here has a bloody clue what I’m on about. I’m guessing because all houses here are brick and cement, there isn’t really much movement.
You have mentioned that movement joints are catered for with a soft-joint running over the joint, but only if the length is over 25 feet, but my problem is that the foundations under my home are pathetic, and in the room I need to tile (3m x 7m) I have no movement joints, but I do have 2 cracks running across the room at a slight angle. The angle makes it bloody impossible to follow with a soft-joint grout line.
Is there anything else I can do here? Like maybe put down a flat bed of mortar , cover with thick building plastic, and then mortar again for tiling?
I am stumped.
Hey Bjorn,
You can ask them if they know (or can get a hold of) any type of ‘crack suppression membrane’. These are basically sheets which span the cracks and allow separation of movement from the concrete through the tile. I have no idea if they could acquire that or if they have it. That’s one option.
Another would be a mud deck for your tile. This would add significant height to your floor, however. A mud floor needs to be a minimum of 1 1/2″ to be effective. This would separate any movement from your concrete through your tile, though. The cracks would not transfer. This mud is mixed exactly like deck mud for a shower floor at a 5 or 6 to 1 ratio of sand to portland cement.
A soft joint can be placed wherever and however you want to do it. The 25 feet is a minimum requirement. Basically if your floor is over 25 feet (interior) it requires a soft joint. If you want to, though, you can place one every 2m. That’s an option but not really one that would solve the problem. If there is significant movement in your concrete it won’t make up for that. The soft joint is only for normal seasonal and climactic expansion and contraction – not really effective for moving concrete.
I think your best option would be a mud bed. Would you have enough height to do that? As far as a crack suppression membrane you can use a topical membrane like RedGard (That is actually what it was originally created for) and brush or roll it on the cracks and about six inches out on each side. When thick enough it would actually give you a fair amount of separation from those cracks and would probably be sufficient. Would that be an option?
I’m tryin’
Amazing! I was worried you might not have a solution, but you ended up providing me with multiple!
Your mud deck solution sounds perfect, although you are correct that height is a problem as I have already added a good inch of cement screed to level out the floor and this would set the floor higher than the door :-(
I will enquire about crack suppression membranes, but if I can’t find that, the topical membrane idea sounds great! Also not sure if we have RedGard, but I do still have a few litres of flexible vinyl waterproofing I reckon which should do the job! What do you think?
Thanks again for your help. It is almost unheard of to find a professional taking the time to help complete strangers. You rock!
I wish you were based here, I would hire you without a second thought! Maybe I could buy you a return ticket?
Hey Bjorn,
I don’t know that I would try the flexible vinyl waterproofing, I’m not at all familiar about how it would hold up under the tile, adhere to the substrate or tile, nor how the alkali in the thinset would affect it. Redgard and products like it (most) are Elastomeric-based rather than vinyl based. Part of the reason for that is that the alkali in thinset will not chemically compromise the membrane. A vinyl-based membrane may break down over time (chemically). An example is the sun shining on vinyl like a car dash. Over time the vinyl breaks down and begins to dry and crack. If the alkali in thinset caused this to happen beneath your tile you would end up with more of a problem than you may currently have.
If it’s a no-go on the crack-suppression membrane you should be able to order redgard or a like product online. I think at this point that would be your best bet.
Also, I’ve always wanted to visit South Africa.
Good point. In that case, I won’t be using the vinyl. ;-)
I have done some calling around and actually found a company which produces their own version of RedGard here although it’s bloody expensive and only sold in large quantities to tradesmen.
Regarding the crack-suppression membrane, I have asked why we don’t use them here and the guys at the tile places have told me the norm here is to use a flexible tile adhesive instead of standard mortar on tiles going ver the cracks. Sounds good to me, what are your thoughts on that?
Let me check some ticket prices… if I get the tickets, the tiling is free right?
A flexible tile adhesive would be defined as a highly modified mortar with an extraordinary tolerance for movement without debonding. Or, you know, in English, it allows a lot of movement without coming loose. They exist but they are really, really, really expensive (and a dream to work with btw) They are probably talking about a normal modified mortar but I dunno. If it’s really good stuff it should work fine for you. Just be weary of the ‘we do it this way all the time’ line of BS. As I regularly state just because it’s done that way doesn’t necessarily mean that’s the right way to do it.
Tile installation for a ticket to South Africa, hotel for a week, and all the cheeseburgers I can eat??? You bet your ass!
You are definitely correct. There is a product available in caulking gun tube format called Sika FC11 and costs roughly R170 ($25) per tube which covers less than 1m2! Way too expensive, so it’s likely the highly modified mortar you mention. The other is the modified mortar which costs roughl 1.5x as much as standard mortar.
The search for the crack suppression membrane continues in the meantime.
Daaaamn, tickets, hotel and cheeseburgers! You can’t be THAT good
Also, if Americans really love their cheeseburgers as much as we see on TV, I might need to sell the house to cover the costs. 
Just kidding
I have 2 different projects that needs your advice…
I am working on covering a shower wall with tile. Currently, the shower is lined with fiberglass panels that have little (not high) bumps on them. The shower is waterproof and intact.
I was hoping that I could use an adhesive to apply Ditra to create a smooth surface, then unmodified in the Ditra cavities and modified thin set for the tiles. Do you think this will work?
On a similar note, I have a piece of very thick plexiglass that we use for a patio table top. It is ugly and I would love to tile it. Actually, I tried but the tiles and mortar fell right off after being stored over the winter. I’m not sure what made me think mortar would still to plexi.
I was hoping to avoid screwing cement board to the top of the plexi. Is there any way I can adhere Ditra to this surface so that the tiles will stick? What can I use to adhere the Ditra? I would like this table to be able to stay outside in the elements.
Thanks for your help.
Hi Sheryl,
You can not apply kerdi (not ditra – that’s stuff for the floor) to a fiberglass substrate – it won’t stay. The best thing to do would be to remove the fiberglass, install regular drywall to the wall studs, then kerdi. That will make your shower waterproof.
As far as your plexiglass table epoxy may work to stick tile directly to it but I can’t think of anything that would make ditra stick. If you do choose to try epoxy I would rough up the face of the plexiglass real well so the epoxy has something to ‘grab’. As long as you rough up the surface and use epoxy setting material it will be able to outlast the elements without a problem.
Glad I came across this comparrison. I briefly thought about the Prova line based on cost and lack of modified mortar. I did come across a huge rant that initially got me doubting Schluter but I then came to my senses.(I am sure you know which one I am talking about).
I did have a question regarding mortar. I really like the Laticrete line and used their 226 Thick Bed Mortar with 3701 Additive to create my unbonded mortar bed for the shower. Wondering what is the best mortar to use to put on the Kerdi? I thought their 272 line would be best for both the Kerdi and the tile.
Was also wondering what you recommend for Ditra in regard to mortar. I have seen the D25 Ditraset which is promoted by Schluter but again I like the way the Laticrete line works.
I do plan on using the Spectralock grout throughout the shower, floor and a glass block window in the shower. Want to ensure all products are compatible.
Thanks for the advice in advance.
Hi Dan,
You can use the 272 – it’s absolutely great thinset. If you cannot get the 272 the 317 is also a great thinset. The Ditraset is good stuff but given the choice I would use 272, then 317, then Ditraset – in that order.
If you have any more questions please don’t hesitate to ask. (send pictures
)
Remodelling a house. One year ago, I’d never laid a tile. One bathroom with a tub, all walls tiled to chair rail height, and tiled counter top. Existing tile had been there 35 years, Good shape, and very well adhered. To sheetrock on the walls and particle board on the floor. (Adhered well enough that I found it easier to remove the sheetrock than just the tile. Floor tiles were stuck WELL also. Hated the color, so, out it came.
I knew no better, and used the past 35 years as a guide, and installed new tile directly to the new drywall. Floor went on hardi-board though (and yes, I got rid of that particle board).
Now for the next bathroom, tiled similar to the prev bathroom, EXCEPT, a built in full tile shower. Bathroom tile same fashion, on drywall walls, and directly to the wood on the floor. Shower was lead pan, mortar walls. FOUND rotten wood at the ENTRANCE to the shower. As if water got to it from outside of the curb. (okay, I can see how that could happen with the way it was tiled). 35 years and the wood “UNDER” the shower looked like new with the exception of the very front, outside of the curb.
The point? Well, I had one when I started. Oh! That’s right, “shower”…
Research began a month ago while demolishing this bathroom. (Wow, a mortar wall on mesh shower is NOT easy to destroy!).
Trying to Google up the best method to build a waterproof shower so that a tile wouldn’t fall and stick in my head while showering next year. I didn’t find the best method. I found gobs of different methods, all claiming to be as good as the next. I’ve since purchased a few Kerdi products, so, guess that’s going to be my “chosen” method. I must have read a hundred forum posts, articles, watched videos, and, and, and…over the course of three weeks. Settled on Schluter/Kerdi after reading some Provaflex “blog/article/rant” or WHATEVER you call WHATEVER it was that I found. See, I know nothing about tile, but, I can assure you that whoever wrote that Provaflex “thing” is a real bonehead. IHe seems to think Ebay, Schluter, and the rest of the tile world is targeting him. (Are ya kiddin’ me?? Ebay’s out to get him for three auctions??) LOL!!!
Anyway, That same day I ordered my Kerdi Drain and Floor slope.
Now, my plan is to Kerdi the shower and then continue over the curb with Kerdi membrane, go ahead and Kerdi the 40sq ft of bathroom floor, and then 4ft up the walls before tiling it all. (I sure don’t want water to get in from “outside” the shower like it did before).
Wish I’d known about Kerdi a year ago in the “tub bathroom”.
And, Is Kerdi okay for the bathroom floor, OR, do I need to use Ditra?
Hi Dan,
Congratulations! You have now had the great displeasure of tearing out a mud shower.
I still build those when someone wants one – and wants to pay for it. The guy you found on the internet is the very jackass I’ve ranted about above – piece of work, isn’t he? 
It would be better to use Ditra for your floor. You can use kerdi but you really aren’t getting any benefit from it rather than the waterproofing aspect – which ditra has also. You can install the ditra as normal, use strips of kerdi or kerdi-band on the seams and the floor to wall transitions and you will have a totally waterproofed installation on the floor as well as the wall. Along with that, if you use ditra, you will also get the decoupling properties which will help your floor to last longer and be more durable. Most vendors sell ditra at the identical price as kerdi so I would go with the ditra on the floor and kerdi everywhere else.
Thanks for the advise. For a DYI’er (Dan Yer’n Idiot) (Not to be confused with DIY), This shower stuff can be daunting to say the least. Just when I settle on “this is the way”, I see another method or another product and…here we go again.
I do appreciate the help in the Ditra decision, AND…most of all, for providing a place where I could say “That Prova guy is Insane”. (and I’ve been wanting to say that in public for three weeks now..lol).
We are planning to lay 16″ by 16″ slate on an outside concrete patio. The concrete extends about 10 feet deep by 35 feet along the back of the house, and then curves to the detached garage, running along the complete side of the garage an average of about 7 feet deep. The concrete has 2 spacer joints along the back of the house, and the concrete along the garage is one piece with no spacers.
Your input on the following questions would be appreciated.
1) Should I powerwash the concrete before laying the slate?
2) Do I need an underlayment?
3) What is the minimum size grout line recommended?
4) Do I need to place expansion joints along the house and the garage?
I would also greatly appreciate any other advice you have on preparation and installation of slate on an outdoor, uncovered concrete patio.
Many thanks for your excellent website.
Hi Connie,
Sounds like you have a project ahead of you.
1) Yes, the more foreign substances you can remove from the concrete before setting your tile the better.
2) Ideally yes, you should use an underlayment such as Schluter’s Ditra. This will separate much of the normal seasonal movement in your concrete from your tile installation so the movement does not transfer through and crack your grout, tile, etc.
3) Minimum grout line is entirely dependent upon the tile you use. If your tile is rectified (that means it is all EXACTLY the same size because it is cut to size after it is fired) then you can probably use 1/8″. This will require an absolutely flat concrete substrate, though. You would probably be better off using 3/16″
4) Yes. You need expansion joints along the entire perimeter along the house and garage as well as every 10-12 AND over every concrete expansion joint (the ‘spacers’ in the concrete).
If your concrete is sealed it will need to be scarified – this takes the coating off the concrete to ensure a bond with the thinset. This can be done by sandblasting as well as physical removal with a grinder or something similar. You also want to make sure to use a good thinset beneath your slate to ensure any freeze-thaw cycles or radical seasonal temperature changes do not debond your tile installation.
Hope that helps. If you have any more questions just let me know.
Many thanks for your response.
To date, I have purchased 120 of the approximately 330 slate tiles we estimate needing, and thanks to your website, I am busy cleaning and sealing them twice before we lay them. I have 60 sealed & ready to go so far, and I’m working on cleaning the next 60. Just think, after that, only 210 to go!!!
I have 2 more questions if you don’t mind:
1) Is there some sort of test that will identify if concrete has been sealed?
About 220 square feet of the concrete we are covering is new, and the remaining 300+ square feet is about 30 years old. The new concrete has not been sealed, but I have no idea if the old concrete was.
2) Regarding the the expansion joints we need to place over the existing concrete “spacer” expansion joints, do they need to be as wide as the existing concrete “spacers,” or can they be narrower? I think the concrete spacers are 3/4 – 1 inch wide. Based on your advice, we will probably be using 3/16″ grout lines.
Once again, many thanks for your help. I want this floor to last a long time, so I want it to be done right the first time!
Connie
Hey Connie,
Slate is fun, huh?
1) There is a test – it’s very technical and arduous process but I’m sure you can get the hang of it. This is what you do:
Splash water on it.
That’s it.
If the concrete soaks the water in it has not been sealed, if it simply sets on top of the concrete in a puddle and is not eventually absorbed by the concrete than the concrete needs to be scarified either chemically or mechanically.
2) The expansion joints do not need to be the same size. They only need to be the size of your grout lines son that if the expansion in the concrete causes movement at those joints it will not crack grout or tile.
Thanks a lot. That test sounds very technical, but I will try to control all other variables while conducting it!
ok,
I’ll buy shorter scews and screw em into the t@g only.
I’ll go with experience on this one…..
Thanks ,
Dave
We tiling our bathroom walls and have discovered areas that bow in and are uneven(surprise,surprise)! How do we level those areas? Thinset doesn’t seem to be enough. Is there a trick to evening out these areas?
Hi Cooper,
You can use either SLC (Self-Leveling Cement) or simply set your tile with a medium bed mortar rather than thinset. Medium bed is simply the same thing as thinset with more sand added which enables it to be built up much thicker than regular thinset.
Very entertaining, educational also, at least the parts I understood.
Hey Ron,
The parts I understood entertained me too!
Thanks for the comments.
Schluter has the patent on the dovetailed design. First was Blanke Unimat with several redesigns which now has the mesh on top, followed by Tec’s “skill set” underlay which looks like a licensed copy of Unimat, Custom Building products has “Spiderweb” their version of plastic cavity/fleece bottomed/mesh coated underlayment just now coming out and now this product. Ditra…imitated, not duplicated.
Very well put, Brian! I’m certain that after that patent is gone (here in the States) we will see a whole slew (that’s a lot, right?) of knockoff products hit the market virtually overnight. Strap in!
Same thing’s gonna happen with the kerdi drain, I think. Prova’s drain does not have lateral movement and it may be because of patent protection – although I’m not positive on that.
Ummm, so if I pre-fill my waffles, is it not then only a chemical bond between the 2 different thinsets, there is no monolithic nature to the floor at that point. Slightly confused.
FULL DISCLOSURE! Mr. Gueuzeman up there is a professional tile contractor in NY (a very good one) and right now he’s fucking with me.
As thinset cures it ‘grows’ crystalline structures like little spikes which will work their way into the microscopic crevices and pits of the layer of thinset in the already filled waffles of the Ditra. So technically it is still a ‘mechanical’ bond.
A chemical bond requires separate molecules to share one or more atom therefore ‘chemically’, or more specifically, through the atomic properties of magnetic force, are locked together.
The discussion of the bond of this type is moot anyway since nothing above the membrane is attached to anything below the membrane directly. The specifics of how the tile is adhered to the membrane is the major difference. The mechanics of the dovetails or the mesh are what define the effectiveness of the membrane.
And you damn well know it.
Re your Ditra / Prova flex comparison: Ditra instructions say don’t use modified thinset between ceramic/porcelin tile and Ditra or you will void your warranty. The prova flex spec says you may use modified thinset between tile and plastic. That seems like an important difference but it didn’t get mentioned in your review. Is it a significant difference, in view of your statement above:
“Modified thinset has better adhesion due to the polymers in the mix but it also allows for a bit of flex. Different materials expand and contract at different rates (naturally). Any tile whether slate, granite, porcelain, etc. will expand and contract at different rates than anything it is attached to. The tiny bit of flex as well as the increased adhesion is why I will only use a modified for any tile I set.”
If there is an advantage to being able to use modified thinset all the way, I’d like to know about it. Rather than peeling back the mesh on the prova flex and saying that wasn’t so hard to do, why not adhere one tile with modified thinset, let it cure and then see how easy it comes off, since the mesh will be bonded in the thinset? The mechanical bond of the Ditra may be great but really it too is limited by the bond of the fleece underneath the plastic, obviously.
One last question, I had cracks in my concrete floor and used a two part epoxy called concrete mender to repair the cracks and spalding. Will a modified thinset adhere to that epoxy if I skuff it up?
Hi Mark,
My post is not specifically about the use of modified or unmodified mortars with either membrane – that’s why it was not mentioned. It is specifically about the difference in the bonding of the two different membranes – the mechanics of it. Ditra relies on dovetails and provaflex relies on the overlaying mesh.
Both products rely on the fleece on the bottom side of the membrane. Not really an issue as I stated they seemed identical or nearly so. The difference is the MECHANICAL BONDING PROCESS. Using modified or unmodified will not make one iota of difference in how well the overlaying mesh is attached to the membrane since thinset does not adhere to the membrane.
I know what Schluter says about unmodified. I also know that it is a great source of argument in the trade about which to use. I use whatever I deem necessary for a specific installation. I have used modified throughout – a lot. The difference is I never rely on a manufacturer’s warranty – ever. My installations are warrantied by me. If something goes wrong I fix it – period.
The decision on certain installations to use modified throughout are based on the specific installation and my seventeen years of experience installing tile. I have no way to describe which I would use in a specific condition without actually looking at the job. If you want Schluter’s warranty – use unmodified. That’s the bottom line. I don’t care about Schluter’s warranty.
Provaflex (as well as every other Ditra knockoff) will absolutely tell you to use modified. It’s a selling point for them, why wouldn’t they? If they rely on the properties of polymers in the thinset so their membrane will do what it says then it isn’t the same as ditra, is it?
You can use modified throughout with ditra. It takes longer to cure before you can walk on it and you won’t have Schluter’s warranty. That’s all.
I do not think thinset will adhere to epoxy of any sort well enough to be comfortable with. But I honestly don’t know, I’ve never had any reason to try. I don’t think I would do it. You may want to go over that with a liquid anti-fracture membrane before installing your tile (or membrane) above it. Redgard is one and there are several others. That may be your best bet.
Thanks for the reply….
About sealing twice,if it’ll help keep grout out, we’ll do it! Seal twice before laying or after? If sealing after laying, is the sealer a factor if it gets in grout area? We are planning on going with sanded grout, given a .375 tile gap. Is the sanded grout going to scratch the sealed slate tiles?
Our initiall plan was to seal once after laying, grout, then seal again, but I have not looked into sealing with a special grout sealer or the same sealer as the slate? We have also talked and read about color enhancers?
Given the slates different thickness, around .3-.5, I do have a concern about what trowel size to apply the unmodified thinset. I’m leaning to .5? Should I butter the thinner slate or all? Man, I’ve got alot of questions. It keeps me up at night thinking about it!
I’m going into new areas on this one and if I didn’t have the net, I’d be screwed.
Thanks,
Dave
Hey Dave,
If you have the time it is always better to do it before you install it. That just makes it easier. If you seal it after you lay it you want to take care not to get an excessive amount of sealer onto the sides of the tile (in the grout joints). A little isn’t going to hurt anything but you don’t want a lot in there. Sanded grout should not scratch the slate tile.
You can use the same sealer on the grout that you do on the slate. Seal it twice before you grout then once more after you grout to seal the grout. If you want the enhancing sealer you can use it as well, just make sure you get the impregnating enhancing sealer rather than the topical enhancing sealer.
Yes, use a 1/2″ x 1/2″ square notch trowel for slate. And yes, backbutter all of them, it’s the only way to get 100% coverage on ungauged slate.
One more thing, and yes, it goes against the ‘rules’ (well, kind of). Schluter specifies unmodified with ditra, this is because of the dovetail shape of the ‘waffles’. The polymers in modified thinset require air to cure properly and that doesn’t happen (quickly anyway) with the dovetails. But I absolutely refuse to use unmodified thinset with any type of natural stone – I’m just a paranoid ass like that. So here’s what I do (and, coincidentally, what I think you should do) Use modified thinset to attach your Ditra to the plywood then fill the waffles with unmodified flush with the top and let it cure overnight. This will leave you with a filled Ditra underlayment fully cured and flat. Then use modified thinset to set your slate to it. This gives you the benefit of ditra as well as having the piece of mind with modified thinset for your natural stone. Win – win.
Now you can get some sleep.
I really appreciate your advice. I’ll go with it. Can I ask your reason of not wanting to go with an unmodified thinset only between ditra and slate? Not as much adhesion? Your logic about taking a while to cure makes sense and I’ve read that alot. I dig the combo. The modified will adhere to the cured unmodified….no problem? Hmmmm.
We’ll seal twice before laying and once after grouting. I’ll set the Ditra with modified on my plywood and then fill the waffles with unmodified on the same day, wait overnight and start setting some tiles! I wish my tile date was sooner!
Thanks alot,
Dave and Pam
Hey Dave (and Pam
)
Modified thinset has better adhesion due to the polymers in the mix but it also allows for a bit of flex. Different materials expand and contract at different rates (naturally). Any tile whether slate, granite, porcelain, etc. will expand and contract at different rates than anything it is attached to. The tiny bit of flex as well as the increased adhesion is why I will only use a modified for any tile I set. And yes, the modified will adhere to the unmodified without any problems at all.
It’s August and after rebuilding my floor joists, ceiling joist, total demo, rewiring, drywall new wall and pass thru, cabinet install….we are going to do the slate install in two weeks. My question is this…..given I’ve shimmed the base cabinets .75 plywood for cleanance for my dishwasher, I’m leaning to the thicker xl instead of the thinner .125. MY thoughts are that the layer of modified between the ditra and floor will be less than .125? and give I use unmodified to flush the waffles….how thick will the modified be between the ditra and my slate if I butter them and use .5 trowel?
Thanks,
Dave Finley
Hi Dave, sounds like all the hard work is done.
I would probably use the XL as well. The thinset below it will be less than 1/8″ and using a 1/2″ trowel and backbuttering the slate (depending on the slate) will leave you with 3/8″ to 1/2″ mortar above the waffles.
Thanks for your help!
My wife just asked me about whether we need to seal the plywood before applying the modified and ditra? What’s your take or thoughts on this? Will this lengthen my drying time of the modified sandwiched between the sealed plywood and ditra?
Thanks,
Dave
Hey Dave,
Absolutely DO NOT seal the plywood before installing the ditra on it. This will severely limit the amount of grab the thinset will be allowed to have on the plywood. It will not allow the crystals in the cement to grow much into the plywood and weaken that bond to the point that may not be acceptable. If you seal the plywood you seal OUT the thinset’s holding power. And yes, it would lengthen the curing time, but that doesn’t matter since you’re not gonna do it, right?
Hey Roger,
no, I’m not gonna seal it. I’m looking foward to laying some slate next weekend. After all the freaking work….it’s been two months since total demo, man, I’m sick of eating out! I’m gonna use a .5 trowel for above the ditra….what do you recommend or use for plywood to ditra for trowel size?
Thanks,
Dave
Hey Dave,
I use a 3/8″ v-notch trowel for kerdi to plywood. You can also use a 1/4″ U-notch if you need to. As long as you get full coverage on the fleece you’re fine (pull it up to check).
Interesting info. I’m researching a kitchen redo. Slate tiles, 12″, on old 2×8 floor joist, 8′ span with 3″ T@G subfloor. As of now I’m gonna sister the floor joist with plywood, add .75 plywood underlayment, and .125 Ditra….no slate or ditra installed under cabinets.
Dave
Hi Dave,
Sounds like someone’s been doing some research.
Your plan sounds good! Looks like your deflection is about 820 so you are good to go for slate and you are adding the additional 3/4″ ply – I feel useless. 
How about a tip: Seal your slate tiles with a good impregnating sealer twice BEFORE you grout your tile – it’ll make your life sooooo much easier. But you probably already knew that…
I’ve got a good one for ya….After researching for a couple of months, we ripped the old pos kitchen out and we are havin fun! Now my question…I have added or sistered my 2×8 floor joists with additional joist instead of plywood. After rebuilding the freakin floor joist and fixing my ceiling joists, I’m ready to add .75 plywood to our existing 3.25 t & g old floor. For my .75 plywood, do I attach it to the floor joists like I want to or just screw it to the .75 t & g like I’ve read online and in the ditra handbook that they sent me. My thought is to make a strong stiff structure and then the ditra will help my slate “float” thru any lateral or vertical stresses? Either way I’d like to screw or ring shank the old t&g flooring to the floor joists before adding my plywood.
Thanks,
Dave
Hi Dave,
You should screw the plywood only to the plank. The reasoning behind this is that it will not transfer every little movement of your joists all the way through to the bottom of the ditra. It would allow an extra ‘decoupling’ between the two subfloor layers. That said, you CAN do whatever you wish to do with it. I am one that will periodically break the rules BUT I only do that when I have all the information, reasoning, and facts behind why the rule is a rule. And anything I do I own it – if it goes to hell I replace it out of my pocket.
You would be better off only screwing to the plank. As long as the plank is securely fastened to the joists this would be your best option. Ditra will take the lateral stresses very well and even some of the vertical stresses. Screwing your plywood to the planks only will add an additional layer of absorption for any movement. As long as you’ve met the correct deflection by sistering your joists there really is no need to fasten the ply to the joist, your floor should be plenty solid without having to do that. Remember – most (probably 85% or so) of your deflection or ‘stiffness’ of your floor is from the joists – not the extra plywood. The purpose is to allow for as much movement as possible without transferring that through to your tile installation.
But yeah, you can do that if you want to. No real reason to, though. And yes, I know that’s counterintuitive – that’s why you wanna do it.
I think that what messes with my head on this…. is that if you in my old basement and put your hands up between my sistered joists and touch the old t&g and have my 130 lbs wife walk above, I can feel the t@g deflect downward. It is like I’m scewing or nailing to something that isn’t stout or stong? It goes against what I’ve learned or thought I’ve learned. LOL! I’ve got to make a call on this soon…….
Thanks,
Dave
HAH! You’re actually getting your head wrapped around why this is the correct procedure – you just don’t know it yet.
If you screw the plywood to the joists you are only strengthening the spot directly above the joist – those spots are already as solid as they will get. If, instead, you screw to the spaces BETWEEN the joists it will firm the areas of the t@g flooring UP TO the plywood which are in turn supported by the areas above the joists. When they joists expand a little with seasonal changes, etc. (and they will) they will push up into the t@g. If the plywood is not screwed to the joists it will not move the plywood UP the same amount as it pushes the t@g up – this is where the separation properties come in. The plywood, on the other hand, will actually stiffen a bit between the joists because the joists are pushing and the plywood is pulling the t@g BETWEEN the joists up the same amount – it actually stiffens your floor more than if you were to screw the plywood to the joists.
Does that make sense? Probably not.
The t@g supports the plywood above the joists – the plywood will actually pull the t@g up between the joists to stiffen it. If you screw the ply into the joists you are gaining no measurable support between the joists. If you don’t – you do. (That helped, huh?)
Screw it only between the joists and have your wife walk across it again. (Don’t tell her I said that)
Hey Rodger,
I just wanted to say how much I appreciate your advice. We did our kitchen per you help, with slate and ditra xl. It looks awesome and no we do not have any cracks anywhere! We are going to use the same slate and ditra on our hallway and bathroom in a couple of weeks and it got me to check back at your site. We went exactly as you said, modified under the ditra, filling in waffles with unmodified., then using modified to apply the slate. Any questions feel free to contact me via my email fincity1964@hotmail.com.
Thanks,
Dave Finley
Glad it worked out well Dave. If you have any questions when you start your bathroom you know where to find me.