Floating wood floor no membrane

[Thatspellstrouble]

Hi guys

Hope you had a lovely Christmas!

In the new year I’m getting around to putting a woodfloor down on the concrete downstairs.

I’ve not noticed any moisture coming from it but to be on the safe side I was thinking of doing a floating floor so if any moisture was forthcoming it’s spread across the floor instead of concentrating under a membrane and going into the walls? Though I’ve seen posts on here suggesting it’s better the latter way.

Also would it matter if it was solid wood or engineered wood for this? Assuming no moisture then I suppose it makes no difference but if there’s a slight chance… would solid wood fare better?

 

[jamest]

It is generally recommended that you don't float solid wood flooring. The reason for this, I assume, is that it's much less stable than engineered flooring or laminate, which doesn't expand and contract nearly as much with humidity and temperature changes. So yes, it makes a difference and engineered flooring would fare much better.

Regardless of the type of flooring, if you don't have a membrane of some sort and your floor is either damp or can transmit moisture, then the wood will get moist underneath and then will rot and/or get attacked by insects (eventually). What you need is some means by which moisture in the floor can evaporate away if you want to float your floor. The only place this can really happen is at the edges of the flooring, so you need a means by which moisture can migrate to the edges and then disperse. My solution to this was to use a 'cavity drain' type membrane that is semi-rigid (I used a slimline membrane from Newton that was designed to allow wood flooring to be installed over concrete floors that have not yet cured) and leave gaps at the edge for moisture to evaporate. You have to leave expansion gaps at the edge anyway, but you need to ensure that the skirting (or whatever) is not sealed with caulk or the like. I have used cork strip in some places and unsealed skirting in others. If you don't like tiny gaps along the top of your skirting, you'd have to fit vents into the skirting itself.

Whether this solution works or not, I cannot say for sure as I haven't ever checked under the floors. However, I've had not the slightest signs of any problems in all the years they've been down - I did have problems before when the floors were carpeted. The standard approach in these situations when using a solid wood floor, is to use a liquid dpm and then glue the floor down on top of that (like parquet) - if you take that path and you're worried about the walls, you could have an evaporate strip of (say) breathable tiles over limecrete around the edge of the room (with no dpm) as some have done.

 

[Thatspellstrouble]

Thanks for the comprehensive reply - the reason I thought no membrane would be necessary was, I asked some limecrete providers what would be put inbetween a wood floor and limecrete (since if that’s vapour permeable then logic follows I treat concrete similarly unless I’m not understanding something?) and they said no membrane needed…

 

[jamest]

Trouble with floated wood floors is that it is recommended that they be glued along the tongue and groove so the whole thing acts like a single sheet, but clearly it isn't vapour permeable (no gaps between boards for vapour to escape). If vapour/moisture rises through the limecrete, where does it go? If it's dry, fine; if it isn't, there will be a problem somewhere down the line. A standard membrane is not appropriate as it defeats the whole purpose of a limecrete floor, which is probably why providers don't recommend it. They probably wouldn't recommend floating a floor either, as it effectively does the same thing. The newton-type membrane is one possible solution. There are others, but not for floated floors.

 

[Toby Newell]

Trouble with floated wood floors is that it is recommended that they be glued along the tongue and groove so the whole thing acts like a single sheet, but clearly it isn't vapour permeable (no gaps between boards for vapour to escape). If vapour/moisture rises through the limecrete, where does it go? If it's dry, fine; if it isn't, there will be a problem somewhere down the line. A standard membrane is not appropriate as it defeats the whole purpose of a limecrete floor, which is probably why providers don't recommend it. They probably wouldn't recommend floating a floor either, as it effectively does the same thing. The newton-type membrane is one possible solution. There are others, but not for floated floors.

This is quite seriously not true. Wood is famously and tremendously vapour permeable. Hence the popularity of the leaf. Also why it moves. If it were not permeable it would not absorb moisture and thus would not move. Trees are designed to be some of the most vapour permeable things on the planet. Over the last two millennia General Sherman has transported several million tonnes of water from the ground to it's leaves. Sherman is a tree. Wood is made from trees.

The glue in the gaps makes almost zero difference (and there will be gaps in the joint open up), of more importance is what finish and how much of it you apply to the surface. All finishes are slightly permeable, unless you coat in a thick layer of molten PVC. Wax, shellac or single part oil finish are the most breathable.

This is all moot as you should always use a dpm under a wooden floor and this should go right to the very edge. You should also use an underlay. Yes, you can float solid floors but you will get more movement, the wider the boards, the greater the movement. 150mm is a good maximum width to stick to for solid, for engineered floors you can go up to 320mm.

However. This is all mooter than moot. Limecrete is specifically chosen to be very permeable and thus fitting a dpm underneath a wooden floor, which is essential, entirely defeats the point of the limecrete. I have recently refused two such jobs.

You don't fit wooden floors over limecrete, you fit permeable unglazed ceramics or natural stone, favourite options are sandstone and terra cotta.

If however you do not have any damp issues and chose limecrete for another reason, then just fit a dpm and underlay, I normally use a liquid (paintable) dpm and a metalised underlay.

 

[jamest]

I am duly chastened, I would indeed have to be some kind of idiot to suppose that wood is impermeable to water (vapour or otherwise). If it were, then it would be waterproof and no membrane would be needed! What I meant to say was that it is relatively impermeable, but relative to what? Possibly relative to a limecrete floor, but more importantly, relative to the atmosphere - if moisture can't get to the atmosphere without having to get through the floor boards, then you're going to be in trouble - the moisture content of the boards will gradually increase as the moisture passes into them. It might eventually evaporate from the top surface, but by that time the boards will be pretty damp, allowing for fungal growth and attack by beetles. My original point was that you could try to provide a more direct route to the atmosphere either by having gaps between the boards (not really an option for floating floors) or else using something like the newton-type membrane, which was designed for that purpose. I promise that never again will I suggest, either by word or deed, that wood is impermeable to moisture!

 

[Toby Newell]

I am duly chastened

Hi James. Chastened? You don't appear terribly chastened, which I am happy for and is a good thing, however, your elucidation is still based on false premises and thus still quite wrong. Reductive reasoning by lay people deriving theory from false premises is very common, everyone does it, including me, it just so happens that I know a little bit about this particular subject and these posts are intended to help people.

What I meant to say was that it is relatively impermeable,

Wood is not relatively impermeable, it is very permeable, even compared to limecrete.

Possibly relative to a limecrete floor, but more importantly, relative to the atmosphere

Relative to the atmosphere? Again, this makes no sense. Physically wood is more permeable to the transfer of moisture than air. At it's highest dew point air at sea level can hold around 80 grams per metre cubed of water, but thats at over 50 celsius! At 20 C its closer to 17g. Timber can hold up to 200kg, which is 10,000 times more, three orders of magnitude. Wood is hygroscopic, air is not.

Few people use a newton type membrane for above ground level installations. They are designed for exterior or excavated below ground level areas where there is a pervasive high moisture environment that cannot be mitigated. Such membranes are also often used to line the walls. No contractor in his right mind would fit a wooden floor to green concrete. The build is usually scheduled so that the slab has 3 to 6 months to dry before any timber floor is fitted. On new builds there will be insulation and a dpm under the slab itself and so for a glued installation which can act as its own dpm, there is no need for another. For a floating or sprung (battoned) install a liquid dpm is applied, up to the edges and preferable behind any skirtings.

Insulation, which reduces substrate (wall, floor, ceiling) surface temperatures, is arguably the most important thing as it matters not how toasty your property is, a beer straight from the fridge still instantly condenses water from the air...

if moisture can't get to the atmosphere without having to get through the floor boards, then you're going to be in trouble

This is nonsensical. Wood is hygroscopic, it absorbs water. (Most) Wood is incredibly permeable. Of course water is going to move through the wood. Any gaps have a negligle effect.

the moisture content of the boards will gradually increase as the moisture passes into them

No. Again, this makes no sense. The moisture content of the wood has less to do with the amount of water entering it at any given point, rather it is the equilibrium it achieves with it's surroundings. This has more to do with surface temperature and relative vapour pressure. Only if the air in the room above it is totally saturated, ie 100% humidity, no further water can then escape from the timber surface until the wood becomes totally saturated. This is only the case in very cold areas where the dew point is low and the air reaches constant saturation or an area with a lot of evaporating moisture, eg unventilated kitchen or shower room, which thusly can reach 100% saturation at normal domestic temperatures. A well ventilated and heated interior has a RH of 40 to 60% (can be as low as 20% in winter with strong central heating) and so this is rarely an issue.

Let us assume the worst case, the moisture content of the wood reaches a maximum of 30% by weight, this will look and feel wet to the touch, much like wood left outside. The rate limiting factor for moisture transport are not any gaps, it is simply the vapour pressure, ie the amount of water molecules already bouncing around the air above and the energy they have (temperature). If the vapour pressure in the air above is lower, ie less than 100% RH then there is space for water molecules to escape. If not then no water will evaporate, not from the wood, not from the limecrete below not from anywhere. Gaps are then utterly irrelevant. What matters is the temperature of the air that surround the solid materials. Higher temperatures = more kinetic energy = more water molecules can remain in the gas phase. The only factors that detetmine equilibrium wood moisture content are air pressure, air temperature and how saturated the air already is with water.

Yes, gaps DO affect the rate of change of equlibrium, but not by much.

It is therefore incredibly rare to find worm in modern buildings. Worm also need a point of ingress, this would obviously be from above, not below, in a floating floor. Mold is a different issue, but again, doesn't have much to do with wood, that is simply a food source, the conditions for mold are a pervasive high relative humidity and a cold condensing surface. Mold is happy to start growing on paint, paper, vinyl, anything really, wood is simply a substrate where it can proliferate as it contains high levels of food.

Wood, being incredibly vapour permeable, already provides a means of transport for moisture, gaps are irrelevant, more water is wicked through the solid timber than evaporates through the gaps. Hence the popularity of the paper towel (say vs 'air drying'). What is more important is the relative humidity of the adjoining substrate, the air above and the surface tempratures of the substrates leading to condensation.

Cold surfaces condese water at lower relative humidity, a beer from the fridge will condense water in a normal home.

To avoid pest attack and rot it is best practice then to:

1. Insulate (as per building regs) under your slab, this reduces condensation.
2. Apply a dpm under your slab, this reduces water transmission from below and from the walls.
3. Apply a dpm on top of the slab. This reduces any moisture wicked from the walls to the slab or absorbed by or condensed on the slab above the dpm.
4. Provide sufficient heating and ventilation to maintain a stable relative humidity of 40 to 60% in the room.
5. Provide adequate heating and insulation of the walls to reduce condensation.

Gaps in the floor, around the floor, are almost 100% irrelevant. The important caveat being, suspended, penetratively fixed floors, ie old pine floors on joist for example, these can normally benefit from vents to increase convective flow to augment air bricks, although it it incredibly rare to see them. Such vents are more often seen in very large commercial buildings with suspended floors of several hundred sqm.

Pretty much all of the thousands of floors I do have gaps. Ironically, the floors with the largest gaps often have the most worm damage. I recently conserved an old elm floor with three quarter inch gaps, it was worm eaten to death, so much so the client put a chair leg through it and the ceiling!

I hope you found this useful!

 

[jamest]

Ha! That's brilliant Toby- I did find some of it interesting, particularly your observation that gaps between boards make very little difference. I had imagined that if boards were laid directly on a solid concrete or limecrete floor, then having gaps between them would allow more moisture to escape from the subfloor than if there were no gaps and that this might reduce the amount of moisture absorbed by the wood boards. Not that the OP would have gaps, so it's not important to the particular issue in hand. Anyway, I can only imagine that in your long career, you have heard so much B.S. and misinformation about wood and its behaviour that you're fed up with it.

Now that the technical details are out of the way, and despite the inaccuracies of my exposition, my only real points were that (a) you need to protect your wood floor from moisture and (b) that with a limecrete floor, using a liquid membrane or a polythene sheet would sort of defeat the purpose of using limecrete in the first place (i.e., to provide a vapour permeable subfloor), so some other method would be better. That brings me to the only point you made with which I would take issue. You're right that Newton (other manufacturers are available) make membranes primarily for walls in cellars and basements (also for drainage of flat roofs and a couple of other more esoteric applications) and I hope you're right that most contractors would wait until a slab was sufficiently dried out before laying a moisture sensitive floor, or possibly any floor. However, Newton themselves state that their slimline membrane was primarily designed for a specific purpose, stating that it "can be installed above slabs that are not yet cured, and as soon as the concrete is walkable" and that the primary application is "To protect flooring from not yet cured slabs". Given that like the tanking membranes there are voids between the membrane and the subfloor, moisture can evaporate from the floor (at least until the void air is saturated). Some of this vapour can gradually escape from around the edges, but given that there is no actual ventilation or air flow in the membrane voids, that would probably be negligible, certainly in the short term. Whether using this membrane is in practice superior to simply painting the subfloor surface with a liquid membrane or laying a polythene sheet (once it's dry, of course), I really don't know. Newton seem to think so. On the face of it, it would seem to be marginally better and if you ever wanted to take your wood floor up and replace it with tiles (say), the breathability of the subfloor would not have been compromised, as it would be if you were to use a liquid membrane. This applies only to floated wooden floors, of course, which is what the OP was thinking of doing.

I am definitely chastened - I accept the lack of correctness in several of my statements - but I stand by my original claim that if you are going to insist upon laying a floating wood floor over a limecrete subfloor, then you might prefer to use a Newton slimline membrane (or similar) instead of a polythene sheet or a liquid dpm. It may well be that the difference between the Newton membrane and the alternatives is negligible, but that is an empirical question that I can't answer.

 

[Toby Newell]

This is all slightly off topic. Wood is not really a suitable covering for limecrete. That is my point. If done correctly sometimes a limecrete install can reduce damp problems so much that an engineered floor would be ok without a dpm, but that is a risk few sensible professionals would take. With UFH it might be ok if you had a sophisticated RH controlled HVAC system. Adding a dpm seems to defeat the breathability part of the limecrete and so is not advised.

Applying a partial dpm under battens (leaving periodic gaps inbetween the battens to increase air movement) and fixing an engineered floor onto those is an option. Then you could fit cork to the perimeter and vents at each side to promote convection. Most of the protection from movement would come from the raising of the floor from touching the slab and any condensation and this would minimally affect rising damp being pushed to the walls.

I am just correcting your reasonable, yet incorrect assumptions that timber laid directly onto a concrete slab, lime based or otherwise, would be fine if it had a few gaps in it. The wood would still absorb more moisture on it's underside from the breathable lime and thus cup. So you would be left with a loose, cupped and warped floor. I think that would be more of an issue than mould and woodworm.

I also wanted to make the observation that limecrete, as fantastic as it is, is only part of the equation and that damp issues are caused mostly by condensation (lack of heat, insulation and ventilation) and high relative humidity (lack of ventilation and heat) And that small gaps to increase expiration rates of a substrate are relatively insignificant, compared to these other factors. If the air cannot absorb any more moisture then no moisture can evaporate and if a surface is cold, moisture will condense on it.

Your premises and explanations are still totally wrong. The purpose of most limecrete installs are to reduce rising damp, ie water being pushed to the walls, this is exactly what a membrane would do and so increases rising damp. Cavity membranes are normally thicker and thus even less permeable, they work below the ground as hydrostatic pressure forces water through the walls which at the bottom are often 100% saturated anyway. Cavity membranes allow water to flow to the edge where there is normally a pumped perimeter drain.

I don't think you have grasped it yet. By fitting ANY membrane you are reducing the expiration rate from the slab significantly, whilst the moisture can eventually migrate to the edges there is not much airflow inbetween the 'dimples' and a massive surface area on the underside of the membrane for the moisture evaporating from the slab to recondense, so subsoil moisture under the floor will still tend to migrate towards the walls, albeit maybe slightly less, defeating the point of the limecrete.

I also don't think you know what the cavity membrane is used for in your suggested uncured concrete application, it is to protect the concrete, not the wood. The wood cannot tell the difference between a flat or dimpled membrane, it's just dry and happy. A normal membrane will pool water underneath and this will affect the proper curing of the slab, leading to increased porosity, increased uptake of salts possibly leading to carbonation and a loss of strength. Concrete is very sensitive to heat and moisture during curing.

None of this is advised for use with limecrete! The wood might be fine, but your walls not so much.

Fitting a wooden floor therefore needs to be done without a vapour barrier which would likely lead to movement, warping, cupping etc. I would not expect mold or woodworm necessarily, limecrete can often negate that, but still the floor would fail.

Your install was done on a concrete floor, your barrier is just a higher performing dpm, but not suitable in this case. As I said, I hope this helps. I provide this infomation to help people, but feel free to look it up, it might take you a while. I did read materials engineering for four years which does help a little.

 

[Toby Newell]

Thanks for the comprehensive reply - the reason I thought no membrane would be necessary was, I asked some limecrete providers what would be put inbetween a wood floor and limecrete (since if that’s vapour permeable then logic follows I treat concrete similarly unless I’m not understanding something?) and they said no membrane needed…

Apologies. How foolish and lazy I am. You have concrete not limecrete. Doh. They are very different, as I have discussed at great length. Applying a dpm to a normal portland concrete slab is fine once it is sufficiently dry.

The general rule is to allow one month per inch of slab. This can be shorter with better humidity control and more in some cases. Point contact moisture meters are useless unless you drill for probes or use a hygrometer with hood. A cheap method is to cut several foot square sheets of dpm plastic and tape them in several places over the surface (seal all around) after 24hrs if the surface underneath is wet then the slab isnt dry enough. Another cheap method is the CaCl salt method. You basically seal pre weighed pucks of hygroscopic CaCl salt in various places over the floor and reweigh them afterwards to work out the water expired, they have a look up table that equates to the approximated moisture content. Me? I am very lazy, I simply refuse to fit a floor until after 6 months.

Yes, once your slab is sufficiently dried out you apply either a physical or liquid dpm. I use liquid dpm.

I also use a metalised underlay like timbermate excel etc especially on concrete, this has sound and thermal insulation benefits as well as walking comfort and resistence to joint cracking. Simply glue the joints with a good D3 PVA. Some floors are clicklock, I never use these. Laying onto battens as discussed in my other replies is the most breathable option for the slab. Glueing a floor direct to the slab is fine and many modern adhesives also function as their own dpm, but I always apply a dpm coat to be on the safe side.

Google is getting a lot worse these days for information research. You get a lot of erroneous facts copied and promulgated over dozens of sites, I guess they just copy each other. Yes you can fit solid wood as a floating floor but it will move more, it depends on the quality of the wood and the cut of the grain, rift and quarter sawn oak move less than crown cut, and prime and select grades normally are more stable than character / rustic or barn grades. You would be ill advised to float wider than 150mm solid, some say 120mm max but remember they are covering themselves. This is for 20mm floors, I would not float a 15mm or less solid floor at all. Having said that most of my work is conserving old floors and fitting reclaimed ones, I never fit modern solid wood as a high quality engineered one is longer lasting and better for the environment as the show wood goes further.

For engineered oak you can float up to 320mm no problem. I have seen, but not installed myself, super wide flooring 450mm+ this is best glued and screwed and plugged.

If asked, I always fit 20mm wide board engineered oak with a 6mm wear layer. I use Upton Wood Flooring. I was the one who made Jonny publish his bespoke page. He uses sawmills in Ukraine and the UK although gets a lot of wood from Lithuania when Ukraine has issues. All of the wood I have got from him has been the best I have ever seen and used. He can bespoke mill straight edge without the horrible microbevels, up to 320mm, although you have to wait until the mill gets enough preorders to run a batch. His character grade is beautiful with amazing knots and medulary rays. Of course, this could all change, at any time. I used to use Staki but they stopped doing straight edge and now their prices have gone up and quality down, they dropped their top layer on their 20mm product from 5.5mm to 4mm so, meh.

Good luck.