Ceramic Tile Advice Forums - John Bridge Ceramic Tile

Ceramic Tile Advice Forums - John Bridge Ceramic Tile (https://www.johnbridge.com/vbulletin/index.php)
-   Tile Forum/Advice Board (https://www.johnbridge.com/vbulletin/forumdisplay.php?f=1)
-   -   Front hall tiles over plywood, too much movement (https://www.johnbridge.com/vbulletin/showthread.php?t=131272)

TPay 07-26-2021 11:47 AM

Front hall tiles over plywood, too much movement
Hi all. I'm trying to determine if I can redo my front hall tiles myself. Twice in the past, shortly after we moved in, the adhesive failed, leading to loose tiles. We gave up and have put up with poor tiles for decades. I'd like to tackle the project now.

My first question is whether or not newer, more flexible adhesives and grouts will be enough to solve the problem without major work on the subfloor plywood (which seems to have a noticeable wave to it throughout my hall and kitchen at least - I'm thinking the builders didn't fasten it properly). I'd like to not have to pull it up and redo it entirely, if possible. At the very least though, I'm willing to install many additional screws in the hope that it significantly reduces movement.

My Second question is whether I have to remove all of the old adhesive from the floor and the back side of the tiles, or if I can simply use a particular adhesive, like an epoxy or other non-cement type, to reglue each tile back into the exact spot where it was. The old adhesive is pretty difficult to remove from the floor, and appears to be impossible to remove from underneath each tile. My concern is that if left in place, the old thinset will simply find a different point of failure, but it would save a huge amount of labor and cost if this proposal were possible.

I have a lot going on with this floor, so I've saved some questions for later, but if there's a way to get away with simpler solutions, I had to ask about those first. I'd appreciate any guidance someone can offer.

Thanks a lot

Lazarus 07-26-2021 12:14 PM

Tyler....as you have discovered, tiling over plywood is a real crapshoot and looks like you came up "snake eyes."

While it "can" be done, "should" be done is another matter. You say nothing about the thickness or number of plywood on the floor. As to it being wavy, that is a red flag. It shouldn't be. :scratch:

Can you pull up tiles, try to clean them and the plywood and stick 'em back down? Not too freakin' likely.

Truth be told, the only reasonable and permanent option is to tear this crap out, probably to the studs....a minimum of a layer of 3/4" plywood...or two layers of 1/2" covered with a cement board, bedded in morter and screwed down. It's the only way I would guarantee the floor.

Yeah...it's money, but do you ever want to have to tear it out and do it again?

cx 07-26-2021 01:05 PM

Welcome, Tyler. :)

We need a good deal more information to help you much.

1. What is this "adhesive" you currently have?

2. What is the plywood subflooring? Thickness and edge style?

3. What is the joist spacing?

4. Have you evaluated the joist structure? The Deflectometer in the dark blue bar near the top of the page can be helpful in providing an initial go/no-go calculation.

Let's start there. Photos are always helpful.

My opinion; worth price charged.

jadnashua 07-26-2021 03:54 PM

Tiling requires the structure to be stiff enough, and that's what the 'Deflecto' tool can help you evaluate. Then, a significant part of that is how strong the subflooring is so that the structure in BETWEEN the joists can be stiff enough. The modifiers in thinset are more to help with seasonal dimensional changes and differences between dissimilar materials...not so much to help with deflection. That's one reason why a modified thinset is required to succeed in bonding tile directly to plywood. You can overcome that issue if you use some intermediary substance, like a cement board, or suitable membrane to decouple the two.

FWIW, to successfully tile directly to plywood requires TWO layers of plywood, both installed properly over joists that meet industry standards.

If the backs of the tile retain a solid coat of thinset, that implies that at least the workmanship of setting them was down well, but the materials used or the structure was not sufficient to support tile long-term. Other things that can cause an install to fail are things like improper or missing expansion joints, and the tile can break bonds due to tenting issues.

Without knowing more, it's impossible to say why yours failed.

TPay 07-27-2021 07:59 AM

Thanks for all of your responses. I just measured the plywood thickness, and it is just under 1 inch. I'm guessing this means it's two 1/2" sheets. The old adhesive is some kind of thinset, but I have no idea if what was used is considered flexible or not.

The joists are 16" apart. I'll pull some insulation down in the basement to investigate them further. One side of the hallway has a heating pipe directly underneath the floor, which has to be a problem on that side at least. When the insulation batts were installed in the basement (which I've heard is not absolutely necessary or advised anymore, at least for unfinished basements), the pipe was trapped between the fiberglass and the floor above it. If I had known more at the time, I wouldve moved the insulation between the pipe and the floor.

I have heard of these decoupling mats. I think this may be a good solution, once the floor is stabilized as much as possible. I like this possibility because it won't raise the level of the floor as much, for one thing.

cx 07-27-2021 08:30 AM

Tyler, a geographic location in your User Profile will be helpful in responding to some of these issues.

We don't want to guess at the subflooring. Two sheets of nominal half-inch plywood is not a good thing at all unless they are properly installed and glued together with a full spread of wood glue. Unlikely you have that. Perhaps while you're taking down insulation in the basement you might find a grade stamp on the subflooring.

I recommend you put the information you gather on your joist structure into the Deflectometer in the dark blue bar near the top of the page to get an initial assessment of that part of the subfloor package.

Originally Posted by Tyler
The old adhesive is some kind of thinset,..

Some clarification: Thinset is a method of installing ceramic tile, not a product. Thinset mortar, usually Portland cement or similar based, is one of the products used to bond ceramic tile using the thinset method. There are also organic adhesives, commonly referred to as mastic, used for such bonding.

The uncoupling mat is your choice. Keep in mind that there is no tile industry standard for such materials. Manufacturer's claims are all you have to base your decision upon.

My opinion; worth price charged.

TPay 07-27-2021 09:29 AM

2 Attachment(s)
The deflection calc gave me these results:

"For joists that are SYP or Douglas Fir, in good condition, 9.25 inches tall, 1.5 inches wide, 16 inches on center, and 13.5 feet long between supports, the deflection calculated is 0.389 inches.

This translates to a deflection of L / 416.

Since the maximum deflection for tile is L / 360, and for natural stone is L / 720, your floor is rated for Ceramic tile, Congratulations!"

The longest unsupported span is approximately 13.5 feet, but some of the joists are shorter as they come up against the framing for the stairs. This is about half the width of the hallway.

There is some dried adhesive under the floor that may be mastic.

I found some stamping, but I'm not sure if its a grade stamp.

I'm located in eastern massachusetts.

cx 07-27-2021 09:46 AM


Originally Posted by CX
a geographic location in your User Profile will be helpful

If you don't put the location into your User Profile, the information will be lost before we leave this page.

Don't know what the stamp represents. It is not a grade stamp.

That deflection calculation would be correct if you had SYP or Doug Fir. Not really enough lumber in view to make a good assessment, but I'd bet heavily it's neither of those two species.

Always up to you what you're willing to tile over.

My opinion; worth price charged.

TPay 07-27-2021 09:49 AM

What are you suggestions?

cx 07-27-2021 09:55 AM

First, determine what you have for subflooring. The subflooring is usually more critical than the joist structure when it comes to a ceramic tile installation.

If your house was built within a code compliance jurisdiction at the time of construction, it's fairly safe to assume the joist structure met code when built. That technically satisfies the tile industry requirements. Our Deflectometer is more conservative than residential building code, partly because our visitors are nearly always dealing with old, used structures, rather than new construction. Your structure might be fine, but I can't tell that from over here.

My opinion; worth price charged.

TPay 07-27-2021 10:17 AM

Thanks. Yes, the builders probably constructed to code, but I wouldn't rule anything out. As I've thought about it, placing tile over wood seems strange. The longest joints look solid from below, as there are only two instances of single joists over the longest span. On one side of these are three joists side by side, and on the other side, two joists side by side. At the bottom of the stairs where the joists are short, the tiles are loose. This is clearly a high traffic area. I don't see any obvious signs of the plywood having lifted from the joists in a way that would account for the unevenness of the floor, but clearly something isn't working. I hear some creaking of the floor in one area, but I won't be able to fully detect wood movement until I have more of the tiles up, since the grinding of the tiles themselves makes too much noise.

cx 07-27-2021 11:38 AM


Originally Posted by Tyler
As I've thought about it, placing tile over concrete seems strange.

Some of us will need a little help with that, Tyler. Thought we were dealing with tile over a wood framed floor.

TPay 07-27-2021 12:46 PM

Pardon my error. It is a plywood subfloor. I've edited my reply to correct it.

jadnashua 07-27-2021 02:22 PM

There is an acceptable way to lay tile directly over TWO layers of plywood, but that also requires a highly modified thinset material. But, the industry also calls for an expansion joint for all tiled surfaces. On a floor, depending on the size of the area tiled, and any environmental issues (that could be uneven heating from the sun, or in your case, maybe from the heating duct), they also can call for an expansion joint there, plus it calls for an expansion joint at any change of plane or materials...that means that on a floor, you need an expansion joint all around the perimeter...it cannot just butt up against a wall or other material without a 'soft' joint in between. That can be a literal air gap (usually if you can cover it with say the baseboard trim), or maybe a caulked joint, or an engineered one in the field, or against another material.

Your structure might be fine, but if the materials used were not sufficient, or they did not abide by the industry standards about expansion joints, that can cause a tiled floor to fail.

What holds a tile down is the interlocking crystalline spikes that the cement grows as it cures. Those spikes, create a tangled, interwoven mesh that pokes into microscopic imperfections. Movement beyond certain limits can literally shatter those crystals, and it can be gradual, so over time, the bond becomes too weak to hold things together. In the right circumstances, a modifier can cushion those crystals, but if the stress is too great, things will still bend enough to break them, loosening the bond. The modifier also can act somewhat like a glue, augmenting that bond, but the primary bond is still from the cement.

TPay 07-27-2021 04:44 PM

That's a big thing to consider, thanks. When I get to the edges I'll check to see if they left the proper gap.

I have some concerns about placing tile back onto this surface. The parts that I've scraped of mortar are still covered with the smooth remnants filling the grain of the wood. Your comments on crystalization of cement make me think that it's no longer a porous enough surface for good bonding with a tradition thinset (though clearly it didn't work too well to begin with).

Two upstairs bathrooms have tile floors with no problems whatsoever, incidentally.

I think I'll be paying a professional to at least consult on it, and possibly do the whole thing, if the correct solution looks like it's totally beyond me.

All times are GMT -6. The time now is 04:49 PM.

Powered by vBulletin® Version 3.7.4
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.
Copyright 2018 John Bridge & Associates, LLC