final sub floor over concrete plan (I hope)

Well I am finally putting down the new sub floor in my basement here in Connecticut.

After the advice I got here this is the plan:

6mm of ploy over the concrete floor. Then 1/2 inch XPS rigid foam board with 3/4 plywood (probably tongue and grove) on top. Then either power actuated nails or tapcons through the plywood and foam into the concrete floor.
There is a lot of advice on vapor barriers that they should be on the warm side of the insulation. Does this mean I should put the XPS on the concrete and the vapor barrier on top of the XPS. This way the vapor barrier is between the XPS and the plywood?

Tom

If you expect instant responses, hire a consultant! It's SUnday night and many of the people have to get up and go to work in the morning...

My feeling is you want ot try to ensure you don't get moisture into your subflooring, so block it at the slab.

Have to agree with Jim. According to Dana's advice the 6 mm vapor retarder should go against the concrete, then the XPS and then your plywood sub-floor. Cheers.

ctkeebler said:

Well I am finally putting down the new sub floor in my basement here in Connecticut.

After the advice I got here this is the plan:

6mm of ploy over the concrete floor. Then 1/2 inch XPS rigid foam board with 3/4 plywood (probably tongue and grove) on top. Then either power actuated nails or tapcons through the plywood and foam into the concrete floor.
There is a lot of advice on vapor barriers that they should be on the warm side of the insulation. Does this mean I should put the XPS on the concrete and the vapor barrier on top of the XPS. This way the vapor barrier is between the XPS and the plywood?

Tom

Click to expand...

"Vapor barrier on warm side" is applicable only to above-grade structures, as a means of keeping interior-air moisture from finding it's way to an element cold enough to condense upon. With basement slabs you have ground moisture issues to deal with as well, both capillary draw and vapor-drive from below.

The solution:

The 6 mil poly on the concrete keeps the subsoil moisture issues under control. The XPS provides protection from room-air moisture, as long as it has sufficient R value that cold side of the subfloor stays well above the dewpoint (condensing temperature) of the room air, staying dry enough to avoid mold conditions. If you intend put a thick shag carpet & carpet underlayment with an R3 insulation value, you need to boost the R-value of the XPS under the subfloor accordingly. The warmer that wood stays, the drier it stays, and the heat flow is from the interior-air down, not from the earth up. The moisture drives are from both sides, but manageable.

If you put the poly between the XPS and subfloor it's probably not a disaster, but moisture CAN build up in XPS, reducing it's R value, and you'd placing it in the ground-moisture side of the humidity environment. In most cases this won't be a problem- XPS and EPS are used under well-drained slabs without vapor retarders all the time. But if your slab isn't well drained &/or very close to the water table the rigid insulation itself has the potential to lose R value from water saturation. With the vapor retarder between the ground moisture and the XPS any moisture that finds it's way to the XPS will dry toward the interior. That way its' in the same hygric environment as the room- it'll stabilize at the average room humidity rather than the average ground humidity.

On basement walls it's trickier: If you apply a Class-I vapor retarder like 6 mil poly to the basement walls the humidity in the concrete rises, causing potential rot conditions at the foundation sill, and spalling & efflorescence on the exterior the above grade portion. Only modest vapor retarders should be applied the interior of foundation walls (like 1-2" of rigid XPS insulation) in all but the coldest of climates. You can then further raise the R-value on the interior with a studwall & batts/cellulose, but it must be vapor-permeable to allow the studwall & foundation ground moisture to dry toward the interior. The balance of the rigid-insulation R and studwall R has to be such that the average mid-winter temp of cold edge of the studs is above the room-temp air's dew point, or you'll have mold issues in the studwall. (If you expect the basement to flood occasionally, use open-cell foam or unfaced fiberglass in the studwalls, not cellulose or close-cell foam.)

For Dana. How the heck did you get all this knowledge about moisture containement and management and the consequences? Very impressive. You've put me in the right direction with my moisture containement projects in a 200 year old house in NY. And you've enabled me to do things correctly vis-a-vis all the many moisture problems we have with this old house. Cheers and best regards, Rughead.

Rughead said:

For Dana. How the heck did you get all this knowledge about moisture containement and management and the consequences? Very impressive. You've put me in the right direction with my moisture containement projects in a 200 year old house in NY. And you've enabled me to do things correctly vis-a-vis all the many moisture problems we have with this old house. Cheers and best regards, Rughead.

Click to expand...

There's been a lot studied & written up about moisture control issues since the early 1980s, after mold & rot problems started showing up in droves in newly constructed housing using newer insulating & vapor barrier materials. (I've been reading up on it for decades.) Some of the best easy-reading on these topics can be found on the D.O.E. sponsored Building Science Corporation website. But most of it boils down to the basic physics of water, and how water vapor differs from air, how capillary action works on liquid water in different materials, etc. Surprisingly few people in the construction trades understand even the psychrometric chart, and how to apply it with climate data to the stackups of building assemblies to avoid moisture buildup in susceptible materials.

It's not rocket-science, but it IS basic water-science. Wood that stays below the dew point of the adjacent air takes on moisture, wood in contact with hygroscopic matierials with access to ground water or rain water are also at risk. Dry, warm (warmer than the dew point) wood has very low risk of rot & mold. Keeping the wood dry isn't always possible, but designing the assembly that it CAN dry in a reasonable amount of time after a bulk-wetting incident. Capillary breaks, vapor retarders, and (very importantly) vapor-permeable air-barriers all make a huge difference in the long term moisture performance of a building. But how they're best used in above grade walls & roofs will vary by climate zone, if less so in basement slabs & below-grade walls. The key is to use them in such a way that it keeps chronic airborne or vapor diffusion water from building up in the assembly, yet still allows it to dry from bulk water incursions. (Limiting bulk water incursions is also important, but that's a bit more obvious, eh? )

For above grade assemblies it's often pretty easy to estimate condensation risk in different stackups based on seasonal climate temp & humidity averages, but not always. Using more precise modeling like WUFI can sometimes better determine whether a seasonal or chronic mold risk will occur in a given stackup & climate.

The vapor barrier is to prevent humidity from accumulating in the insulation, so the vapor barrier goes between the warm area and the insulation.

hj said:

The vapor barrier is to prevent humidity from accumulating in the insulation, so the vapor barrier goes between the warm area and the insulation.

Click to expand...

That's the problem with rules of thumb- it's possible to hit 'em with a hammer!

More important than keeping moisture out of the insulation (where it does no harm, if the insulation is XPS foam) is keeping humidity out of the wood, where it can create mold & rot conditions. The wet-insulation issue is secondary, and unlikely to occur from vapor permeation alone with non-hygroscopic insulations like fiberglass. Air-leaks and capillary draw are orders of magnitude larger moisture transport mechanisms than vapor diffusion. AIR BARRIERS are far more important than vapor retarders/barriers for keeping moisture out of the assembly. (A square inch of air leak is worth more than a whole wall of vapor diffusion through unpainted wallboard.)

But on a basement floor you have vapor pressure from both directions- diffusion from the warm room air into the not-so-super-permeable XPS foam insulation, but more importantly, the near-100% humidity of ground moisture coming up from the slab. Here the foam is only slightly permeable, but the higher drive will be from below (even though it's the cold side!)- the foam will stay drier with the poly between the XPS and the slab. If the poly is between the OSB and the XPS you'll end up punching more holes in the poly, since it's within reach of the nails that secure the wood flooring to the nailer-deck OSB. Since it's above the XPS it'll be warm enough that it won't condense, but the humidity level of both the OSB & flooring will be higher.

In above-grade walls, etc the rule of thumb is true- the higher vapor retardency layers in the stackup go on the side that is warmer for more hours on an annulaized basis, but in all but the very coldest (or most hot-humid) US locations, using highly retardent layers such as polyethylene or foil is a double-edged sword, since it effectively prevents the assembly from drying toward the usually-warm side, effectively cutting it's ability to dry from bulk-water incursions in half. Yes, it prevents condensation from vapor-permeation, but it doesn't let anything out either. Most of the time semi-permeable or semi-impermeable vapor retarders (not barriers) like kraft backing, an inch or less of XPS or vapor-retardent latex will result in drier assemblies, and greater resiliency from the occasional rain-drip that finds it's way into the wall during a gale wind. (And using hygroscopic insulation to redistribute the drips that make it in results in an even more resiliant assembly, as long as it's drying capacity hasn't been crippled with vapor barriers.)

Tapcon to concrete

Hi everyone.
First timer here. I have been reading allot of info here regarding subfloor installation. Everything is answered clearly (Thanks Dana & all) What I get from this thread is --- 6mm poly, 1" XPS, 5/8 Plywood/OSB. I'm hoping OSB is OK as it is cheaper (thoughts?). My question is - Do I need to screw this sandwich to the cement or can I have it float using tongue & groove OSB?
What I have planed is a 10' x 23' Synthetic ice rink on 1 half - carpet or laminant on the other. Also do you think I could forgo the OSB on the rink side as the panels are strong enough to be layed outdoor on grass ( Is the XPS strong enough?)
Thanks and I'm looking forward to any advice.
Chris.

ChrisH said:

Hi everyone.
First timer here. I have been reading allot of info here regarding subfloor installation. Everything is answered clearly (Thanks Dana & all) What I get from this thread is --- 6mm poly, 1" XPS, 5/8 Plywood/OSB. I'm hoping OSB is OK as it is cheaper (thoughts?). My question is - Do I need to screw this sandwich to the cement or can I have it float using tongue & groove OSB?
What I have planed is a 10' x 23' Synthetic ice rink on 1 half - carpet or laminant on the other. Also do you think I could forgo the OSB on the rink side as the panels are strong enough to be layed outdoor on grass ( Is the XPS strong enough?)
Thanks and I'm looking forward to any advice.
Chris.

Click to expand...

If it were something stouter like hardwood or bamboo that would tie it all together well you could float it, but with laminate or carpet you'll have to secure it to the slab.

OSB instead of plywood is fine in this app.

If the synth-ice works fine on sod it'll be even better on slab-supported XPS or EPS- no underlayment necessary. The compressive strength of sod is at least 3 orders of magnitude lower than standard density XPS. You may need a slip-surface like 15# felt or rosin paper between the panels & insulation though. (Hard to say for sure without reading the specs.)

Dana.
Thanks so much for the quick & informative response.
Two more questions if you would be so kind.
Can you clarify why the OSB + xps needs to be fastened into the concrete?
You figure the XPS under the rink panels will hold up under the strain on people skating on it a few hrs a day?
Chris

A shift of even 1mm will show with laminate flooring. Under carpet there could be more significant movement of the OSB, which may cause damage at the OSB/XPS interface or a slight wave in the carpet if it's bad enough.

You can drive a 10 ton truck over XPS under a 2" slab without exceeding the ~25psi rating:


http://www.insulfoam.com/images/stories/docs/6065_Below_SlabInstalGde.pdf

The analysis is for EPS, but the math works for any similar material. The typical 25mm XPS sheathing is most similar to Type IX EPS from a compressive strength & elastisicy point of view.

If the panels are rigid enough for skating with a highly compressible sod underlayment without creating huge rockering & edge height issues, the dynamic weight distribution of a skater will be at least a few orders of magnitude lower than the compressive elasticicity limits of XPS.

Dana said:

If it were something stouter like hardwood or bamboo that would tie it all together well you could float it, but with laminate or carpet you'll have to secure it to the slab.

OSB instead of plywood is fine in this app.

If the synth-ice works fine on sod it'll be even better on slab-supported XPS or EPS- no underlayment necessary. The compressive strength of sod is at least 3 orders of magnitude lower than standard density XPS. You may need a slip-surface like 15# felt or rosin paper between the panels & insulation though. (Hard to say for sure without reading the specs.)

Click to expand...

Hi, I am putting a similar floor on an above grade slab. I would like to keep the floor floating if possible. Can I do poly, foam, osb (stapled together with quarter inch space between sheets), with an engineered wood floor floating on top of all this. I would prefer not to have to hammer drill all this to the ground. Also, will I help if I glue the foam to the osb (kind of making a home made Barricade type floor system). Thanks, and appreciate any feedback you can give me.

sara9 said:

Hi, I am putting a similar floor on an above grade slab. I would like to keep the floor floating if possible. Can I do poly, foam, osb (stapled together with quarter inch space between sheets), with an engineered wood floor floating on top of all this. I would prefer not to have to hammer drill all this to the ground. Also, will I help if I glue the foam to the osb (kind of making a home made Barricade type floor system). Thanks, and appreciate any feedback you can give me.

Click to expand...

If the slab is nice level I don't see any reasons why you couldn't float the whole thing, and just glue tongue & groove OSB to the XPS with foam-board construction adhesive, as long as the OSB seams are at least a foot or so away from the XPS seams- no need for any gaps or spacing, but with T & G you can give it a little to allow for seasonal hygric contraction/expansion without risk of vertical displacement at the edges of the OSB.