Hardwood slab on grade
house is slabish on grade, built into a hill so the back wall has about 6' of dirt against it.
looking to install hardwood on a slab (on grade) here is what i am thinking of doing:
6-15mil poly > 1/2" poly iso > 5/8" Plywood (or osb?) > Insulayment or Quiet floor (has vapor barrier) > nail down t&g 3/4 hardwood or 3/8".
Would screw the ply to the slab thru the foam.
I also was thinking of just using 1-3 strapping with 3/4 iso in between as the subfloor but i think the strapping to concrete would defeat the whole purpose of warming the floor. This would be cheaper.
Any comments or criticism is appreciated
First, I'd take a 2'x2' piece of plastic and tape it securely to the floor...maybe do this in several places. Leave it a couple of days to see if any or how much moisture is trapped underneath the plastic. Then, I'd evaluate my choices. Personally, if things stayed dry, I'd seriously consider an engineered hardwood product rather than solid wood (and probably still would once I had moisture controlled). There's a huge amount of it available in various price ranges and various thicknesses. I'd look for one that can be installed as a floating floor. Quite awhile ago, I was at a home show in LA. I asked a guy, after telling him I lived on the east coast and wasn't going to buy anything from them, what he would install, if he were going to install wood flooring. He said something from www.kahrs.com . He said the company he worked for did not sell that brand, but he had worked elsewhere, sold it, and found it to be topnotch stuff. They have a huge selection, but you may have to search for a dealer. They do not sell online, so you have to find a dealer. Both glue down, floating and nail down stuff, in lots of woods, finishes, thicknesses, and choices...all real wood surfaces. Adding a lot of thickness to the floor, if there are stairs getting to that level, may mean rebuilding your stairways since there are some strict requirements about consistency of the rise of each step.
The stackup using will work fine except that you need to avoid any vapor-barrier underlayment. Putting a vapor barrier above the subfloo would create a moisture trap for the susceptible wood putting it at high risk for mold/rot at some point in the future. Using something that is more vapor open than 2 perms (or inherently vapor open) is highly recommended be. (Rosin paper works.) Insulayment Quiet Walk is not a vapor barrier. According to the manufacturer it wicks and disperses moisture, so that would be perfectly fine.
You may want to use real 23/32" or 3/4" t & g subfloor rather than 5/8" over 1.5lb iso, since it's rated for 15psi compared to the 25psi of similar density rigid polystyrene. The T & g limits the amount of "leaf-curl" or cupping that it might otherwise experience, and the thicker wood will distribute the floor loading better. Iso is fine in this application as long as it has the poly underneath as a slip-surface and additional vapor barrier against ground moisture. (The foil facers are also powerful vapor barriers, but any nicks on the underside could potentially wick moisture into the iso if it didn't have the poly.) There are higher 2 & 3lbs density higher-strength iso products used in the roofing industry, but they're usually only 2" or thicker panels. There is even higher strength 4 -6 lb density goods out there, but they're lower R and higher $. The standard foil-faced 15psi stuff is fine the way you're using it though. (Cheap 5/8" ply/osb sheathing would probably be fine, but real subflooring is slightly thicker and better engineered for the application.)
The total average R-value of the assembly is higher going to be somewhat higher with the 1/2" iso than it would be with 1/3 sleepers 16" o.c., due to the high thermal bridging factor of the sleepers. With 1x3 sleepers 16" o.c. the fractional percentage comprised by the wood adds up to about 16%, reducing the "whole-assembly" R of the sleepers & foam layer to about R2.75. With a continuous plane of half-inch iso that layer is R3. Add about R1.5 for the subfloor + finish flooring and you're pushing R5, which is enough for a comfort difference, if less than ideal for a VT climate.
Furthermore the sleepers would be susceptible to moisture accumulation/mold on the underside (even with poly) since the temp of the now-insulated slab would be well below the summertime outdoor air dew points. With R3 under all of the wood it stays warm enough to not accumulate moisture in summer, for a much lower risk of mold.
Do you see any value in adding "quiet walk" type underlayment? I was thinking that it would add a tiny bit of comfort without any real height, they say a .5R plus might deaden some sound.
Is it worth doing the Iso? is the trade of rvalue/psi worth it? or just do XPS, my main motivation is the higher R value but if it comprimises the structual integrety then it isnt worth it.
One other detail, there is some existing wood flooring that is going to redone. It has been down since ~1986, here is the stackup: Carpet tiles glued to the slab > Poly > 1/3 pine sleepers 12"oc > 3/4" hardwood. Interesing but when i first say it i thought for sure it would be moldy and rotten but it only smell musty, i haven't taken it all up yet so i am not positive but seems pretty dry. That floor is mostly cold about 2'-3' from the outside walls I suspect when i take off the drywall, air seal and reinsulate with iso on the inside studs it will improve. The middle of that floor isn't that bad thats why i figured the r5 would be better then nothing.
Would I make the foam fit tight against the walls and caulk, foam it? Would be butting into 1"iso on the walls.
I could bump up the foam thickness but I have one transition between two rooms that would be a problem. the slate is 1/2" off the slab.
here is the breakdown:
1/2"F +3/4ply+3/4"hardwood my transition to the above would be 1.5"
3/4"F... Transition = 1.75"
1"F...Transition = 2: (This is the max foam I could consider)
This is why i want the ply to be @ 5/8"
I could redo the slate in this area is about 30 sqft to bring it up but i just did it last year, it's not the money in this case it's the effort to redoo it.
Would it be better to just bump up the foam to 3/4 or an 1? I can't redo the floor once it's done...
is OSB ok? Should i be concerned with moisture issues? spills, condensation...
If I went with 5/8" I could biscuit the edges so they stay on the same plane,does that make sense? or 3/4 still better?
Do I need to screw the assemby to the concrete? or can it float?
If screwed do the screws going thru the poly cause an issue?
>If I did the sleepers, would using sill plate foam do anything for the moisture that might accumilate under them?
>I could use 1" foam and cut into 12" strips cut a lap joint on one long edge to accomidate a 1x3, this would give a 1/4" of foam under the sleeper, is that worth considering?
>Sleepers would be the most inexpensive option, but if i don't have to screw the Plywood/osb it probobly isnt worth it.
What do you think about using 3/8" t&g naildown hardwood? Would really help out on the height issue but i can't get 3/4" out of my head!!!
Thanks alot for your help!
The R-value of the quiet-type underlayment is nearly nil, and won't be buying ANY comfort. It may dampen some nail-squeaks if the flooring warps slightly and rises up, that's it.
The difference between 15psi foam and 25psi foam when fully supported by a slab, and with a subfloor + hardwood flooring above is mostly academic. At your very low R-values the extra R0.5 is actually substantial. XPS has the additional downside of having about 200x the lifecycle global warming potential (GWP) of polyiso, since it's blown with HFC134a at ~1400 x CO2 GWP, whereas iso is blown with pentane at ~7 x CO2 GWP. The long-term (50+ years) R-value of half-inch XPS is also about R2 rather than the rated R2.5, as the blowing agent slowly escapes over time too. The pentane in polyiso is gone within months of manufacture, and it pretty much retains it's labeled R-value. (When it's fresh off the presses half inch iso performs at about R5 or 6, but 18 months later it's down to it's labeled value, whereas with XPS it takes decades, so they are allowed to label it to a presumed "lifecycle average".)
Adding 1/4" or ~R1 under sleepers doesn't add much R, but it adds some, but it's not necessarily worth the trouble of the extra work, since it barely affects the structural rigidity.
If you can get truly flat high quality 5/8" t & g plywood it will probably work out just fine as a subfloor in this app. I've seen people get away with 7/16" OSB with other types of flooring, but I'd worry about the long-term nail/staple retention for hardwood flooring. I've read of thinner subfloors developing edge curl & waves when it wasn't t & g in this kind of stackup, but that may be less of an issue when you have the rigidity of the hardwood flooring working for you.
There are some OSB designed for subflooring that can withstand being rained on without warping (!), but most OSB will wick and retain water much more rapidly than plywood. If you have a history of flooding OSB is probably less desirable, but trenching in drains, sumps, & pumps is probably on the critical list.
Putting gasketing under sleepers adds maybe R0.1, which will not raise it's temperature substantially. The poly sheeting is sufficient to block groundwater wicking, but does nothing for air-adsborbed summertime moisture. With R3 insulation under the subfloor the slab will stagnate at only slightly above the subsoil temperature, which is in the 40s Fahrenheit for most of VT. In summertime the outdoor (and presumably indoor) dew points are in the high-50s/low 60s, and over time moisture from the air will end up in wood that is always below that dew point temp. (he dew point of 75F 60% relative humidity somewhat air conditioned summertime conditioned space air would be ~60F, WELL above the deep subsoil temps in most of VT, and even if the bottom side of the sleepers was 55F it would be taking on substantial moisture over the warmer months of the year.
Were it new construction it would be worth putting R10 of EPS (also blown with pentane, not HFCs) under the slab in a VT climate, but even half that is still pretty good. If you used 1" iso you'd be at about R6, which is a huge uptick over the bare-concrete condition. With 1" iso, 3/4" subfloor and 3/8" flooring you'd be at 2-1/8", and with a 5/8" subfloor you'd be at a dead-even 2" rise. Using 3/8" hardwood and a thicker subfloor isn't a bad way to go, since the thicker subfloor would do more for distributing the load along both axis' than 3/4" subfloor, which is really stiff along the length of the board, but totally flexy laterally from board-to-board (which is why it's preferable to run them perpendicular to joists, in timber framed floors.)
so will using .5, .75 or 1" polyiso under 5/8" or 3/4" ply (any combo) eliminate condensaton problems? if so which ones.
also do i screw( can i powernail?) the sibfloor to the slab, or can it float
thanks for your help, mark
Even R3 between the subfloor and concrete is sufficient to eliminate "condensation" issues for the subfloor at VT subsoil & dew point temps.
It's really adsorption into the wood fibers, not condensation onto the wood surface but that's just a technicality. The water isn't in a liquid state until the wood reaches a very high moisture content. Adsorbed water is more akin to molecules dissolved into a solid- it can move by vapor diffusion, but doesn't exactly flow or wick with liquid-like surface tension until the wood is fairly saturated (and WAY beyond mere mold-inducing levels.)
I've always TapConned the subfloors to the slab 24" o.c. with a minimum of 3/4" penetration into the concrete. YMMV. Nails might work, but would probably require more fasteners (= more thermal bridging, more holes in the poly for ground moisture to enter.) If you simply floated it without tacking the subfloor down to the slab, normal seasonal humidity changes in the flooring & subfloor are likely to cause some amount of lifting or wave/curl in the floor that changes with the seasons.
if i use advantc 5/8 t&g with 3/8" hardwood and 1" poly iso i would be at around r7.5 and only have a 1.5" transitions do you think that would be robust? i think the thermomax brand poly iso is 25psi. also slab appears to be very dry, i think the overpour ontop of poly must have done that so if i am scwreing thru the poly int the concrete i might possibly poke thru the poly within the concrete, will that cause moisture problms? what about in the iso?
Poking through a vapor retarder and filling it with a steel screw isn't much of a moisture path- don't sweat it.
R7.5 isn't terrible- you'd easily feel the difference.
Any 25psi iso is likely to be 2-3lbs density, but that's a good thing for this application.
Dana, the Advantech isnt a vapor barrier of any type due the the water resistant is it? won't cause any issues in my stackup?
Originally Posted by Dana
Also, i read in an earlier post (sorry, for not responding to that one but it was old) you stated "By contrast polyisocynurate is downright WIMPY when it comes to pressure loading, and should not be used as slab insulation under a floor. It also can absorb water over time in a no-loading slab (no floor) application."
It was in the context of someone asking about putting a washer/dryer on top of xps and ply. I am not sure if maybe you were referring to it being under a slab maybe?
It's still ok to use poly iso ? hope so i bought it and it's going down in the morning along with the 5/8 advantech! Don't know if this means anything but i have had plastic taped to the floor in a few diff spots for a few weeks with no moisture.
Additionally, I just read an earlier post where you said it was possible to float a similar build up, (1" xps, 3/4 t&g and construction adhesive with a few fasteners if needed) My stack up will be 1" polyiso, 5/8" advantech what do you suppose the fewest fasteners i could use in conjunction with adhesive? you had suggested 24"oc but that would be 18 per 4x8, is there a way to reduce that using a glue? could the screws be eliminated because it is t&g?
Under a slab iso at any density is a disaster due to the moisture accumulation loss of performance.
Above the slab 15psi goods under some types of washing machines may need a heavier/stiffer subfloor to distribute the dynamic loading of the vibration during the spin cycles to avoid localized compressions. (With 25psi goods you don't even have to think about it.)
Tacking it down 24" o.c. along the t & g seams (staggered across the seam) -only you might be able to get away with it, but it's better to have at least a few in the middle of the 4/8 sheet.
The vapor permeance of Advantech isn't super-relevant in this app since it's the very subflooring that you're trying to protect, but highly moisture tolerant. At 1-2 perms it's about half as permeable as standard OSB, but that's sufficient to allow spills or moisture that gets between Advantch & foam to dry at very reasonable rates. The "no cupping or warping' aspect of it will probably let you get away with just gluing it to the foam in the center, and TapConning it along the seams. If you give the seams about 1/16" of space to allow hygric & thermal expansion the warping stresses will be very small.