Mud Bed Flakes Away [Archive] - Ceramic Tile Advice Forums - John Bridge Ceramic Tile


View Full Version : Mud Bed Flakes Away

07-13-2019, 12:41 AM
Hi all! I'm doing my first tile project, so I'm as amateur as it gets. Apologies in advance. Also, apologies for the novel, but I understand you want all the relevant info possible. I just put in my shower's mud bed for the second time. I used the dry pack deck mud method. Both times have resulted in a really flaky result that makes me worry about it's structural stability. While it's firm enough to walk on, it chips out super easy. After sitting for 24 hours, I used my air hose at 90 PSI to blow off the surface for a clean base to start laying my tile. But that resulted in the top layer of the mud bed come up as well. I have no doubt that if I continue to blow/sweep off any dust/loose top sand, the entire thing will eventually come out. Simply pivoting on my feet as I cleaned also dug into the floor. Is this normal or should I redo the entire floor again? Or is this all about compression strength and totally fine once the thinset and tile are placed? This is the same thing that happened the last time with a different product (Custom Building Product's pre-blended CustomFloat Beddign Mortar). If I need to start over or make repairs, please advise on what I should do differently. I'm thinking about just using regular reinforced cement if I have to do it over again.

I uploaded an unlisted 15 second video clip to YouTube of the floor blowing away, but it doesn't look like I'm allowed to post links yet. I'm not sure if it will help, but maybe I can send it in a direct message if it would help anyone.

Here's all the info I can think of that may help:
It's a curbless shower on a slab. While replacing my drains (the reason for this project) I cut an inch into the slab at the start of the shower and dropped 1/4" per foot to the back of the shower (5') where I have a linear drain. Essentially, the foundation is my pre-slope.


Over that, I placed several layers of redgard that wrap up and around the entire shower.


So, my mud bed is 1 inch thick for the full length of the shower (all the products I used advertise they're effective as thin as 1/2" and I don't want to drop my slab lower to risk overall structural issues. This time, I followed the instructions laid out in an article by John Bridge on the website, precisely and watched several YouTube tutorials. Again, I can't post the link, but it was titled "Deck Mud (Floor Mud)". Search that on the website and it should give you the article. Basically, I dry mixed 2 parts Quikrete Sand/Topping Mix and one part Play Sand and then slowly added small amounts of water. It was just like everyone said - slightly drier than what would be used to build a sand castle. This was in my house, so the temperature was around 72F. Someone recommended I lay down some fresh thinset first to help with adhesion. I didn't do that the first time, but did this time. Same result either way.

The only thing I can think of that may have been an issue was the time it took me. I'm new, so I'm slow. But I knew that going into it, so I did a couple things to compensate. Firstly, I enlisted the help of my wife to make my mud deck mix in 3-4 stages after showing her the first time. (It took me a while a to get around the drain just right - I might be too much of a perfectionist, to be fair). So, the mix I was using at the end was fresher than the mix I started with. I started at the drain, did the perimeter, and worked my way out. The other thing I did to help with the prolonged working time (I think), was to immediately spread each mix out over the entire floor and stamp the crap out of it. So at the end, I was just smoothing it out. Also, the mix I laid down first is flaking the most.

I didn't get any pictures of the process since I was in a rush, but here are some close ups of the end product now as well as when I easily chipped out the first attempt (which turned out to be a good thing since I didn't put in the waterproof membrane that time).

From when the first attempt chipped out with little effort using only a flat shovel:

I've been working on this project after work for a few months now and have learned a lot. But I'm very concerned with getting everything perfect so I never have to do it again. Any advice would be greatly appreciated. Thank you so much!


**Edit: sign off**

Sponsored Links

07-13-2019, 05:56 AM
When you mud it back, add just a little more water in the mix and when you are finished, use a flat trowel to slick it down on the surface. You can go over it lightly with a vacuum the next day but don't blow it out with compressed air. That was probably your biggest problem. You have to remember, dry pack will never be as hard as your driveway.:tup1:

07-13-2019, 06:42 AM
Thanks for the reply, Davy. I'm not looking forward to starting from scratch again, but definitely will if that's what it takes. If it's no surprise that the compressed air caused damage, that makes me think it's structurally sound if it had remained flat. Is that the case? If so, couldn't I treat this like any other slab that has excessive pitting and just fill in the 1/8" holes with thinset or some pre-mixed repair cement I used on the actual slab beforehand? Then, once it sets, just tile over it like normal?

While the extra time, labor, and materials it would take to start over aren't ideal, my bigger concern is the near certainty of ripping through my water proof Redgard seal and ending up with a likely imperfect patch job that leaks. The entire reason I'm doing this project is to fix a leak from this shower drain. I really don't want to rebuild it with even the slightest potential for another leak. Obviously, I don't want to have an unstable foundation from repairing something that needs replaced, either. I'm just not sure the thinset would even stick to this loose, sandy surface even if I hadn't touched it.

07-13-2019, 08:36 AM
Welcome, Glenn.

Couple points of concern I see here.

You say you've chipped out your concrete SOG an inch at the front of your shower and chipped out enough to have a minimum of 1/4" per foot to the back, or a total of at least 2 1/4" cutout at the drain in the back. That would leave not much structural concrete at that drain area if the slab were poured like those common in my area.

Then in that thin section of concrete you've placed some sort of clamping drain and installed RedGard, using a reinforcing fabric to clamp in the drain per the manufacturers' instructions? Then you installed some sort of linear drain that connects somehow to the clamping drain? And you protected the weep holes in the drain and placed a mud bed of a consistent 1" thickness?

Sounds a bit improbable to me, but I suppose it's possible. Then, after your first mud bed attempt you removed the mud bed without damaging the RedGard on the concrete floor? Or did you re-do the RedGard and, if so, did you re-do the portion in the clamping drain as well?

It appears you intended to make a curbless entry to this shower and I'm curious to know why you wouldn't have put your RedGard on top of your final mud bed, bonded to your linear drain and extending outside the shower a couple feet as we would generally recommend for such a curbless design.

For your mud bed problem, I've tried to duplicate the problem you're describing after reading probably hundreds of similar posts here over the years and my best guess is the lack of proper compaction. You indicate you "immediately spread each mix out over the entire floor and stamp[ed] the crap out of it. So at the end, I was just smoothing it out." It's not necessary to beat deck mud into submission, but it's important to have it well compacted when you cut it to shape. I'm envisioning your procedure and think you were still likely to have been forming loose sand as your finished surface rather than carving compacted mud. I usually pack my mud for a shower receptor by simply whacking it with a wood float while cutting it to shape. The surface should be relatively tight and can be finished a bit tighter by troweling at the finish as Davy has recommended.

But I'm most concerned with your receptor waterproofing as it sounds as though you might be replacing one leaking shower with another. Perhaps you can tell us more about your waterproofing method, especially at the drain?

My opinion; worth price charged.

07-13-2019, 09:44 AM
Thanks for the reply, CX. You've summarized my process pretty accurately. What part do you consider unlikely? Please keep in mind this is the first time I've ever done anything like this. I'm a thorough researcher and but of a perfectionist, but may have missed something. That's why I'm here.

It seems I have a fairly thick slab. There is a little more than 2 inches of slab thickness left at the thinnest part near the drain. But even though that thinness is only a small area, I feel like that's pushing it. That's why I didn't want to cut any deeper. I spent about a week carefully grinding down slowly as to not cause any fractures and to get as smooth of a slope as possible.

Also, If you notice in the first picture, there's a 1 foot wide trench of new concrete. That's where I placed the new drain and had to bust all the way through the foundation. It's fairly deep. I have roughly 6 inches of concrete over pipe that was first back filled with smooth river gravel and vapor barrier. I filled around the existing slab the best I could and epoxied in plenty of rebar as deep as possible to tie it all together. Also, that deeper central trench with thicker concrete backfill allowed me to essentially sit the linear drain right on the pre-sloped slab with all the connection pieces beneath it. There's no way it would fit within the 1 inch I had for the mud bed. If you look closely at the pictures before I added the mud bed you can see the slab sinks down quite a bit at the drain o accommodate for those connection pieces.

As for why I placed the Redgard under the mud bed, I actually didn't on my first attempt. I planned to do exactly as you suggested. But I realized, after the fact, that wouldn't work with this drain. In order for the tile to be flush, the mud bed had to be below the lip of the drain by the thickness of the tile. Obviously, that would defeat the purpose of the pre-slope since it wouldn't allow any actual drainage. It would be hindered by the drain, itself. So I had to redo the entire thing that time anyway.

Also, there was no damage to the Redgard since I hadn't placed it yet. I just shoveled up the mud bed and used my grinder to ensure I had a smooth surface before putting up my cement board and Regarding the entire thing (including a foot past the boundaries of the shower on the floor and walls that aren't shown in the pictures since that wasn't relevant to my question).

I figured if I had gone the old school route of using sheet membrane, I'd end up clamping it in below the linear drain as they intended. This linear drain actually came with the pieces that clamped it to the actual drain pipe below and instructed to do it that way. I figure that once dry, Redgard is basically the same thing as a sheet membrane. So I painted it over the lower portion and screwed the top portion down on it as the instructions indicated. I painted quite a few layers of Redgard on the floor so I think it's actually thicker than a typical sheet membrane and after a flood test, it didn't drain very well. So, I decided to place a few thin plastic shims between the 2 pieces and clamp them back together. I did 2 more flood tests and was happy with the results.

Hopefully that answered all your questions concerning my waterproofing. But about the mud bed, it sounds like I just need to use more water. Based on all the videos I saw of guys doing this, I don't think it's possible that I didn't pack it well enough. If I redo it, I'll certainly keep that in mind, though.

Right now, I'm leaning toward trying something different, though. These might be stupid amateur ideas, but what do you think about skimming the surface with something made for thin repairs? Or do you think that wouldn't be structurally sound? It seems to be fine for compression strength. It's just the scraping of the surface that causes problems. Once the tile is on, that won't be a problem.

07-13-2019, 02:14 PM
Glenn, have you got a link to the drain you used? I'm trying to figure out exactly how you have the Redgard tied into the linear drain.

When I mention more water, I mean very little. You want it so slicking it down with the flat trowel forces a little water to the surface. That little water will make the surface a little harder.

07-13-2019, 06:01 PM
Unfortunately, I didn't think to get pictures of the Redgard painted into the drain, but I assure you it's good. That isn't my concern. But I'm always up for verification, so here's the linear drain setup I purchased:
This is how I assembled it with the exception of using Redgard in place of the membrane in the picture:

I poured the concrete around the Floor Mounted Base Flange and ensured a smooth transition from slab to flange. I then painted the Redgard over the flange up to the lip of the drain so there's one smooth, continuous, unbroken waterproof membrane starting 1 foot outside the shower all the way to the lip of the drain with a 1/4" drop per linear foot. I performed multiple satisfactory flood tests. I'm confident my waterproofing is good to go. I'll be less confident if I have to pull out the mud bed, but will do it if necessary. So, again, can I just smooth out the top layer of the mud bed with some cement repair or is that too dangerous and need me to start from scratch with a wetter mix?

07-13-2019, 06:46 PM
Glenn, while Redgard is approved as a pan liner material (according to Custom's spec sheet), you need to show me industry-issued documentation that just painting over the drain flange will sufficiently tie it in with the drain assembly in order to convince me that your assembly will stay waterproof for a long time. Lots of Youtubers out there showing it like that but that doesn't convince me at all.

I believe this is not an industry-approved method and I am concerned about that connection right there between Redgard and drain. Also, did you use the reinforcing fabric alongside the walls and corners? Potential movement would show up in these areas and put stress on the Redgard which might result in small tears. If you share my concerns (and it seems Davy has some concerns as well), now is still time to address this.

07-13-2019, 07:38 PM
makethatkerdistick, thanks for your input. Yes, I did use reinforced mesh in all corners and joints. I haven't seen anyone do exactly this on YouTube or anywhere else before. It's something I came up with out of necessity/preference. Regardless of industry approved methods, I don't thin it's a good idea to put the waterproof membrane behind the backerboard. And since I already had Wonderboard (which isn't already waterproof like Kerdi system, etc.), I elected to go with a liquid membrane over the top. I painted over the drain because of the logic I'll outline below. Industry standards are great and all, but they're slow to change and aren't accepted until they're accepted. Someone has to be first :)

With that being said and with your prompting, I re-read the Redgard product data sheet to see if I missed anything. It looks like I did. My initial decision was based on all the combined possible uses of the product. After all, it's not just for waterproofing. It's also designed for crack prevention and used for up to 1/8" cracks (much much larger than the microscopic joint between my cement and flange situation). But I'm glad I looked at the data sheet again. They have actually linked to 3 application examples for shower installations. one of them shows an identical setup to what I have already concluded and completed:
As you can see, the Redgard goes under the mortar bed and over the slopped fill (my slab) and flange, joining them together. The only difference is I don't yet have any of the layers above the mud bed since I'm still trying to get advice on it from you all.

I understand waterproofing is crucial and really do appreciate your concern, But I would also very much appreciate someone addressing my actual question concerning repairing the mud bed. I mean no disrespect and truly value and appreciate all the advice and concern with my waterproofing methods. Is there a reason no one wants to comment on repairing my mud bed?

07-13-2019, 08:05 PM
I know Redgard and other paint on membranes are much stronger when used with the fabric. Looks like to me the set up you have is just like the drawing shows as long as you use the Megalite thinset, is that correct? I've never installed a mud bed over a paint on membrane. I trust the PVC liner more.

It's not designed for it but you can use thinset to fill the spots in your mud that are 1/8 deep. I just hope the mud is strong enough to not crumble under the thinset. Something we can't see.

07-13-2019, 09:00 PM
Industry standards are great and all, but they're slow to change and aren't accepted until they're accepted. Someone has to be firstThing is, Glenn, that particular wheel has been reinvented a number of times over the years and even the oldest iteration is still accepted as one of the best. You're certainly free to reinvent it again if you like, of course, we can only tell you what the industry consensus is and where the smart money is betting. Personally, I'm a large fan of the sheet-type direct bonded waterproofing membranes for shower waterproofing, but would never consider a liquid applied membrane in that application. Different strokes.

I'd like some context for that drawing you have there and would appreciate a link to where you found that. That doesn't appear to be a clamping drain in your drawing and I've never seen Custom recommend anything but a clamping drain for use with RedGard. The most current online TDS I can find still says the following:
RedGard® Waterproofing and Crack Prevention Membrane / TDS*104

Drains should have a clamping ring with open weep holes for thin-set application. Apply the membrane to the bottom of the flange. The drain should be fully supported, without movement, and should be even with the plane of the substrate. Apply the RedGard membrane around drain. Embed a 12" x 12" (30 x 30 cm) fiberglass mesh into the membrane, making sure it does not obstruct the drainage weep holes.Then apply an additional coat of the membrane and smooth. After curing, clamp the upper flange onto the membrane and tighten. Use a silicone caulk around the flange where the membrane and the upper flange make contact. That's been their recommendation for many years and I'll be curious to see where that drawing you have might have originated.

But, to your question. You're using a receptor construction method that requires the top mud bed to be solid, but porous. Known for decades as a water in/water out system, it requires that liquid water be able to pass readily through it to reach the weep holes in the drain. Patching the surface with non-porous materials will slow that process, especially the part where the remaining moisture is required to evaporate through the tile surface. Again, you're certainly free to do as you like, but keep in mind the purpose of using deck mud in that application and weigh whether it might not be a good idea to replace it rather than repair it. Deck mud is dirt cheap; DIY labor is free.

My opinion; worth price charged.

07-13-2019, 10:33 PM
That's my concern, exactly. I'm going to have to really think on it. My biggest issue is that I've never seen this before, other than pictures and videos online. Those aren't quite the same as in person, so I'm not sure exactly what it should look/feel like. I very well may redo the entire thing just to be on the safe side. I'd rather be safe than sorry for sure.

The local supply store actually gave me CBS's VersaBond for my thinset. That's what I've been using to fill the joints and flatten the walls so far. I've had other stores tell me that was the best as well. So, I assumed the MegaLite in the diagram was just for the example and not required. But with that being said, I don't really trust what people tell me without putting in my own research and getting a few opinions first. So, I probably need to look into that as well.

That's actually how I got into this predicament in the first place. I had originally hired all this out, but my tile guy decided to suddenly stop doing tile and focus on counter tops only and my plumber didn't show for 3 days. Then they told me they didn't have a saw to cut the slab with, so they were just going to start jack-hammering at my foundation without scoring it. I didn't know much, but I knew that was a big no go. So, I let them go. Apparently, we have a shortage of plumbers in the area because I couldn't find another one. So, I rushed to the local home supply store to get the materials to do it all myself over my vacation. They recommended a lot of supplies that I wouldn't go with now that I know more. But even if even if they aren't state of the art, these products are made for this and used ever day by professionals. My philosophy is that nearly anyone can do nearly anything with enough will power and training. It's just taking a while since I wasn't prepared to train myself in multiple trades. But anyway, that's why I have a hard time trusting others with this stuff now and appreciate makethatkerdistick needing proof of something he might not be familiar with. Plus, no one around my area seems to have any experience at all with curbless showers.

I'm not purposely reinventing anything. Just trying to work with my limitations. With that being said, I did chose not to put the sheet up behind the backerboard. I just don't see how it's a good idea to allow water to penetrate through the backboard before being reflected to the drain. I want the waterproofing as close to the surface as possible. I actually had a sheet membrane from the home supply store, but it wasn't big enough (even though it was the largest one I could find at any of the stores in this area) and I didn't feel comfortable using it even though some "pros" were pushing me to do so. Either way, there are lots of different methods of doing anything. It's OK to be a fan of only one particular method. I don't mind doing things differently so long as it works.

Here's the direct link:

And here's the RedGard data sheet that links to it under "Application of Product", "SHOWER RECEPTORS INSTALLATION":

Ether way, the process outlined in the CBS quote you referenced is exactly what I've been saying I did - just like the reference image I posted previously showing the drain manufacture's instructions with the sheet membrane in place of my RedGard application.

But, to your question. You're using a receptor construction method that requires the top mud bed to be solid, but porous. Known for decades as a water in/water out system, it requires that liquid water be able to pass readily through it to reach the weep holes in the drain. Patching the surface with non-porous materials will slow that process, especially the part where the remaining moisture is required to evaporate through the tile surface. Again, you're certainly free to do as you like, but keep in mind the purpose of using deck mud in that application and weigh whether it might not be a good idea to replace it rather than repair it. Deck mud is dirt cheap; DIY labor is free.

Right, exactly. But all cement is porous. That's why we have to use a waterproof membrane in the first place. Otherwise, the water would weep out into the rest of the foundation and possibly up into adjacent rooms (which was my original issue and reason for this project). Also, remember that even if my mud bed was perfect, it would still need thinset over it to adhere the tile. So, if I use the same thinset to repair the 1/8" imperfections first, there will be absolutely no difference at all as far as porousness and water flow is concerned. If that's the only issue, it wouldn't make sense to start over. I'm more concerned about the structural integrity of the mud bed - specifically the compressive strength. The forces acting against it during sweeping, blowing, etc. are more akin to properties dealing with shear strength. The only thing that really matters with any concrete product is compressive strength. Tensile, shear, and torsional strengths don't really matter for this application. So if it's just a surface imperfection and it's not going to crumble from the inside out, we're all good. That's the direction in which I'm leaning since it can't really crumble due to the force of the surrounding slab. It all makes sense based on what I can remember from my biomaterials class way back when. But this is my first time dealing with this in person and thought I'd get some input from those more experienced than I.