GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Poly vapor barrier under XPS foam board in slab radiant pour

Hondaracer2oo4 | Posted in Energy Efficiency and Durability on

I am building a 28×40 garage in New Hampshire. I am doing an I slab radiant pour. I plan to use 2 inch xps foam board under the slab and as a thermal break to to frost walls. The issue that I have is that this install is for a garage build and will not have a roof over the floor before the slab pour leading to the possibility of rain water ending up being caught between the vapor barrier and foam board before the pour. My question is why would taping the xps foam board seams not act as an effective vapor barrier? If this is not acceptable for some reason would it be reasonable to lay poly vapor barrier down, then foam and tape the foam board seams to keep any rain water from getting in between the vapor barrier and the foam? Thank you for your insights.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. GBA Editor
    Martin Holladay | | #1

    The best approach is the install the polyethylene above the rigid foam (and directly under the concrete). For more information, see Polyethylene Under Concrete Slabs.

  2. Expert Member
    Dana Dorsett | | #2

    At about 0.5-0.7 perms, 2" XPS is still a class-II vapor retarder, and not a vapor barrier. At 6 mils polyethylene is an order of magnitude more vapor-tight.

    Using 2.5" or 3" 1.5lb density (Type-II) EPS is usually cheaper, and a LOT greener than XPS due to the differences in blowing agents.

  3. Hondaracer2oo4 | | #3

    Ok so the xps foam won't suffice as a vapor barrier. So the consensus is poly over the the foam and tubing. Do I just lay the poly up the frost wall insulation and cut it off later? Should I somehow attach it down to the foam to keep it from moving? Staple down? I am stapling the pex tubing down to the foam. Using fiber in the concrete instead of mesh.

  4. Expert Member
    Dana Dorsett | | #4

    On thicker slabs it's better to put the tubing nearer the top surface rather than stapling it to the foam. Many/most would tie it to the wire mesh, propping the mesh up off the foam.

    You don't have to staple the poly to the foam, but minor staple penetrations and tears from installing tubing (if you go that route) don't really matter. If stapling tubing to foam you'll get better staple retention out of Type IX (2lbs density) EPS than Type-II (1.5lbs). A compromise approach is 2" of Type-II EPS overlaid by 1/2" XPS (seams staggered) for the staple retention.

    Overlap any seams in the polyethylene by a foot or so and tape it to prevent a flap from flipping up and folding back during the pour.

    Run the polyethylene up and over the frost wall, with sufficient excess to go over the top of the foundation, under the foundation sill & gasket as one continuous vapor barrier & capillary break.

  5. STEPHEN SHEEHY | | #5

    Jonathan- look for reclaimed foam. It's cheaper and greener. Why only 2"? I'd go with 4" in your climate since you want to heat the garage, not the ground.

  6. Reid Baldwin | | #6

    Did you really mean to say put the poly over the foam AND THE TUBING? I would think the poly goes under the tubing.

  7. Expert Member
    Dana Dorsett | | #7

    Poly goes under the tubing, over the foam, for sure!

    If anything the tubing needs to be a couple inches above the foam (for more favorable heat transfer rates to the concrete.)

  8. Hondaracer2oo4 | | #8

    Ok well I have Lots of conflicting info. First off I was planning to put poly last(over the tubing) because how am I suppose to staple 500 staples through the poly and then Feel like it is still a good vapor barrier? I had looked into reclaimed xps board and no one has any that really fits the bill. Mostly 1/2 inch and 1 inch stuff. I thought eps isn't rated for ground contact and the psi rating isn't very high?(going under a garage). 2 inches is pretty standard in my area I believe for under slab insulating, it is going in a a garage unattached to anything that I plan to just keep at 50-55 all the time. The slab is going to be 4 inches thick. I don't need it to react quickly to a temp change since I don't plan to be changing temps so I think 3.5 inches down in the slab will be fine. I can't run the vapor barrier continuos up the inner frost wall of te garage since it will have 4-6 inches exposed on the inside of the garage to keep any water off the interior framing or sheet rock down the road.

  9. Expert Member
    Dana Dorsett | | #9

    There is no conflict here.

    Vapor diffusion is an Area x Vapor permeance sort of thing. Five hundred or even a thousand staple holes is not going to compromise the vapor retardency of the polyethylene by very much, since it's at most a few square inches of vapor-permeable hole. The steel staple is extremely vapor retardent, and you're only looking at the area of the sectional ring around the staple holes as a vapor path. Even that path would have the vapor retardency of the foam to limit the rate of vapor diffusion.

    Almost all insulated concrete forms are made with Type-II EPS (the most commonly used density in construction). It's fully rated for ground contact- nearly every PassiveHouse in the US has many cubic yards of it under the slab.

    Multiple layers of 1" XPS is better than a single layer of 2". The stuff shrinks over time, and single layers then have a temperature striping issue at the seams. When using reclaimed XPS, derate it to R4.2/inch, since that's where most of it will end up as the blowing agents escape. The R5/inch labeled performance is the guaranteed minimum performance at year 5 from date of manufacture. Owen's-Corning will only guaranteed R4.5/inch at year 20. A logarithmic fit to those two data points would indicate R4.25 @ year 50, but who the hell knows? Fully depleted it has the same R-value as EPS of equal density, which would be ~R4.2/inch for 1.5lb -2.5lb density goods.

    R10 under the slab isn't enough, even if you're only keeping the garage at 55F. That would be betting future energy costs are going to be quite a bit less than they are now. If using reclaimed foam R20 is completely rational, due to the much-reduced initial cost. What is your heating fuel here?

  10. Hondaracer2oo4 | | #10

    I think you are confused about the type of staples used in stapling down radiant pex to foam board. This is what they are.

    So you think 1 inch xps double layered would be better?

    Heating is from a wood fired down draft gasification furnace.

  11. user-2890856 | | #11

    What size buffer tank will you be using for that application Jonathan ? You will require one and a reliable mixing valve , preferably one with outdoor reset capability .

    Tubing should also be in the middle area of the slab depth no matter what you do . Stapling tubing to the foam is not an issue in slabs that are done competently . Will your slab have wire within the pour ? It should no matter what type material you will pour , just fasten the tubing to that . When your mason pulls the wire into the slab so too will the tubing be there , or he could use chairs . I know , what about sawcut expansion joints or other expansion joints . Saw cut joints should not exceed 3/4" depth , they only give the slab a place to crack . Other joints may eb installed at various places , if this is the case just duck the tubing down to 4" either side of that joint to allow tubing to move with the slab .

    Only 2 kinds of slabs on this planet , those that are cracked and those that have not cracked yet . Anyone says different is flat out fibbin .

  12. Hondaracer2oo4 | | #12

    Boiler has on board storage(200 gallons) so no buffer tank needed. Yes I will be installing at mixing valve to run the floor at the lowest temp possible(95?). The mason said he can either do wire or fiberglass strands in the mix. The fiberglass in the concrete has become very popular around here and seems to work very well. My parents pool deck was poured with the fiber 15 years ago and still has no cracks. Yes saw cut reliefs will be made.

    I am still trying to get some thoughts on the vapor barrier. Why not vapor barrier over everything before the pour? Is the only reason for the vapor barrier not to go under the insulation because of the possibility of getting wet from rain water before the pour?

  13. GBA Editor
    Martin Holladay | | #13

    Q. "Why not vapor barrier over everything before the pour?"

    A. The reason that you don't want to install polyethylene over your PEX tubing is that you want the PEX tubing to be near the center of the concrete slab, not under the concrete slab. And if you include welded-wire mesh, as many commenters have recommended, you definitely want the welded wire to be near the middle of the slab, not under the slab.

  14. Reid Baldwin | | #14

    Jonathon, you might want to go through all of the comments to the article that Martin linked to in #1. At least one of the commenters said that they routinely put poly under foam with good success. That commenter had some reasons why he believed putting the poly under the foam was superior.(Personally, I really hope that is ok because that is what the builder did in my basement.) Most of the article is about where the poly goes with respect to a sand layer and there seems to be a consensus that sand over the poly is bad. Since foam absorbs much less water than sand, the argument doesn't seem to me to be very applicable to foam.

  15. Dana1 | | #15

    Putting the vapor barrier under the foam leaves the possibility of water accumulating in the inevitable pockets between the foam & polyethylene, which then takes forever to dry. This sometimes leads to problems with the finish flooring or slab-paint. If the slab isn't going to have a finish floor or paint, don't worry about it.

    With two layers of foam installing the vapor barrier between the layers results in less volume in which water can accumulate.

    With the vapor barrier in direct contact with the concrete, the concrete wicks and redistributes any liquid water that gets in.

    To limit cracking in an unreinforced slab keep it visibly wet for several days after the pour, spraying it down every time it starts to look dry. This improves the curing of the concrete, but also increases the chance that some water would accumulate on the vapor barrier if placed below the foam.

  16. Hondaracer2oo4 | | #16

    I believe a lot if contractors are placing the poly under the foam in radiat slabs with no issues. This is a garage build so no paint or flooring over the slab. Radianttec suggests poly under the foam. To the poster that said his is built with the poly at the bottom, have you been through a heating season yet? What did you get for results. I don't want to over think this thing to death, it is after all a garage and I'm not looking to put the garage in for an energy award, just want it to work appropriately. I also read from radiant sites that all say max depth of tubes should be no more than 4 inches which these will not. I understand that you will get a bit better response out of tubes closer to the surface an less chance of overshoot but if it overshoots by a degree or two that is fine, it's in a garage.

  17. Dana1 | | #17

    Plastic staples are about as impermeable as metal. Even if you assumed it created a 0.1" x 0.1" hole per prong (which it doesn't), that would be 0.02 square inches per staple. X 500 staples is 10 square inches of total area with a vapor permeance still less than 1 perm (3" of EPS or 2" XPS) instead of 0.05 perms (6 mil polyethylene). In reality you're probably looking at less than 3 square inches. Don't sweat puncture holes, it's nothing compared to a missing strip of polyethylene or a accidental fold-back on a seam.

    Placement of the poly is only about floors that will be finished. If you were to paint the garage slab having it under the foam would increase the odds of having places where the paint bubbles & flakes. If you're never going to paint the slab, don't waste any time thinking about it.

    Reid: The interstitial spaces between sand grains adds up to a large volume, but so does the space between the foam & poly when laid over 3/4" screenings, etc. If the basement even floods or seeps it takes forever for that water to dry though the foam. I personally don't view it as a high risk to place it under the foam, but it's not zero risk- there are known failures related to that (per BSC), which can be avoided if the poly is on top of the foam.

  18. user-2890856 | | #18

    What make and model gasifier are you using ?

  19. Hondaracer2oo4 | | #19

    Heatmaster g200. Cleanest burning wood fired unit of all gassers epa tested phase 2.

    I understand now about above or below I believe. Only worry is trapped water diffusion through the cement later possibly causing flooring or paint covering floor to fail. This is a garage and I won't be painting it so the possible water leaving the concrete should e non issue.

  20. Nicolas_Bertrand | | #20

    This is an old post, I know, but I felt like things went somewhat unresolved.

    I am getting ready to do my basement slab for my personal house. I have lots of clean #2 stone, then there will be 2" EPS, then 2.5" Nudura Hydrofoam that will hold my pex in place. I plan to tape the joints in the smooth 2" EPS layer.

    So, where should I do my vapor barrier layer? Do I need one if I tape the EPS and seal my pipes well to it? Under everything is between 8" and 36" of cleaned stone, and that is over bedrock. My main concern is that things are open, and there's a good chance after everything is in, it will rain before the concrete is poured. So, where do I want the vapor barrier, or do I need one with the foam? I'm afraid if I put a 10 mil barrier under my foam, then it rains, I will have a floating-foam disaster.



Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |