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

Detailing a Slab in a Hot-Humid Climate

JRoyal | Posted in General Questions on

Hello folks,

I’m in the process of finishing my basement and wanted to hear some opinions about floor assemblies. First some quick info: Climate zone 3a, poured concrete slab, 9′-6″ ceiling, no insulation under slab, no bulk water issues so far, new house constructed in 2016.

Based on what I have read in these forums and on Fine Homebuilding, I thought my floor assembly was going to be: slab>3/4″ rigid foam (seams taped)>3/4″ OSB (secured to slab w/ tapcons)>finished floor of engineered wood or LVT.

However, I have seen some sources suggest using a air gap membrane under the OSB subfloor, similar to a Delta FL or DMX. Would this take the place of the rigid foam? I can’t imagine an assembly with both of them, so is there an advantage to either approach? In my climate zone I don’t think any insulation is required under the slab, any my primary intent of using the rigid foam above was to reduce the risk of moisture problems (condensation), not necessarily comfort…though that would be nice.

Also, I did not plan to put a layer of poly down between the slab and either assembly, because there is one under the slab. Bad idea? Poly is not that expensive of course, so if it would help, then its something I would probably include. I just wouldn’t want it to complicate any of the approaches listed above (I think Martin mentioned an imperfect knowledge of building science creating big problems in the FHB podcast…makes me cautious).

Anyways, would love to hear folk’s thoughts about the two approaches and what would work best in my relatively warm, humid climate.


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

    If you have a dry slab, and you need to choose between rigid foam and dimple mat, go with the rigid foam. Rigid foam has R-value, which is good -- the R-value means that the rigid foam will lessen the chance of condensation (reducing the chance of mold or funny smells), and the R-value will slightly reduce your energy bills.

    The value of a dimple mat between a slab and OSB subflooring is often exaggerated, and explanations are often based on magical thinking. The dimple mat physically separates the OSB from the slab, and stops capillary transfer of moisture. But it doesn't provide a drying path, so it doesn't reduce the humidity level below the OSB.

    If you know for a fact that you have polyethylene under the slab, you don't need any polyethylene above the slab.

  2. JRoyal | | #2


    As always, thanks for taking the time to answer questions in these forums.


  3. gozags | | #3

    Martin, I am wondering about the comment on dimple mats and no path for drying. I have an old slab (basement) and some areas pass the plastic tape down moisture test, others don’t. Clearly no insulation or poly below. Items that breathe (cardboard box, etc) or something perched up on legs, etc don’t accumlate moisture but items like plastic, etc can after a period of time.

    It makes sense to me that a dimple mat would alow that moisture to dry on some level, is that not the case? No skin in the game other then our flooring at some point, just wondering.

  4. Jon_R | | #4

    A dimple mat can allow bulk water to redistribute before it gets the OSB wet. For example, to a drain. I'd use it in addition to insulation.

  5. Expert Member

    Surely if you have sufficient bulk water intrusion that it can flow on the surface of the slab to a drain, you need a lot more drastic solution than a dimple mat?

    1. Jon_R | | #18


  6. GBA Editor
    Martin Holladay | | #6

    Sean Cotter,
    The fact that dimple mat creates a capillary break, interrupting upward wicking, provides a limited benefit. But there is still no drying path (as there is, for example, in a rainscreen gap with openings at the bottom and the top). If the slab is moist, the air between the dimples will also be moist. That moist air won't magically dry.

  7. Jon_R | | #7

    The low perms of something like Delta-FL over 1000 sqft allows ~7 gallons/year of drying (depending very much on details). More permeable mesh mat/EPS foam/flooring would allow much more. And this isn't accounting for air flow (there is always some) or drains.

    A few gallons/year happens to be how much bulk water leakage I get in my basement. Concentrated in one area, it's a problem. Spread out and slowly removed, it's not.

    That being said, the right solution is for me to have zero gal/year of basement leakage.

  8. GBA Editor
    Martin Holladay | | #8

    According to a document on this page:

    "DELTA®-FL functions as an effective vapor barrier in conjunction with wood laminate floors and many other common floor coverings, and is a simple new solution to create a beautiful, high-performance basement floor."

    Another page:

    Delta-FL "Creates a sealed vapor barrier subfloor that prevents moisture rise and makes basements feel warm and dry."

    1. Jon_R | | #10

      It also says < .3 perms - so while it may be an "effective vapor barrier"/retarder, it's apparently not an absolute vapor barrier. Unless it's far less than .3 perms (and other mechanisms contribute nothing) , my point stands - basement floor dimple mat can have little but sometimes significant and sufficient drying.

  9. Expert Member

    "And this isn't accounting for air flow (there is always some)"

    Jon, this is a similar argument to the one you make about mesh under metal roofing panels. What is the mechanism that would create air-flow on a slab with a dimple mat, rigid insulation and flooring above it in an enclosed space?

    1. Jon_R | | #13

      "... it was found that barometric cycling caused the wetting of cavities about 10 times faster than by vapour diffusion alone." and " It can cause both wetting and drying of cavities".

      1. Expert Member
        MALCOLM TAYLOR | | #15

        Can you expand on that? What would cause a pressure differential between cavity and the room above?

        1. Jon_R | | #16

          Barometric pressure changes cause changes in room-above pressure because houses aren't sealed perfectly. The cavity can only change pressure to match the new room pressure if there is flow. Such flow causes drying.

          Over large areas and long time periods, it adds up to a lot of gallons of drying. Said gallons might otherwise cause problems (eg soaking OSB).

  10. gozags | | #11

    I always assumed that moisture, not unlike love, always finds a way. I would assume that some of the moisture laden air would make it's way to the edges of the dimpled products and be released into the room, then handled by dehumidifier/AC. But that's probably not much different then just putting down 6mil (or whatever) and then your vinyl flooring (or OSB then flooring). I don't have the head height to go with the proper rigid foam, then OSB then flooring - so it's poly then rigid vinyl flooring for us.

    I had contemplated DMX or something like it (I might have even asked about it here at some point), but sounds like an unnecessary cost.

  11. MarkAno | | #12

    Dear Martin & GBA Community,

    Further Jamie's "Use of dimpled membrane on basement floor" question;
    Jamie could "imagine an assembly with both (dimpled mat AND rigid foam) of them".

    I would like to ask if there IS an advantage/need using BOTH of them for my imminent basement build?

    I too am in Climate Zone 3a (Atlanta) with the identical poured concrete slab, 9′-6″ ceilings, no insulation under slab, no bulk water issues so far, new house constructed in 2010.


  12. MarkAno | | #14

    Dear Martin & GBA Community,

    ...continuing from previous post:

    In re-reading Martin's article titled:
    "The High Cost of Deep-Energy Retrofits
    A pilot project generates cost data on deep-energy retrofits of four buildings in Utica, New York"

    ...this basement floor insulation assembly is quoted as follows: "the contractors first installed a layer of Platon dimple mat on top of the existing concrete slabs, followed by R-10 rigid insulation and a layer of Durock cement board." This is the first time I have ever come across using Durock as a basement subfloor above EPS/XPS insulation. Being inorganic, I thought it brilliant and ask (out of ignorance) why this is not done more?

    Typical current assemblies are:
    - Concrete Slab
    - EPS (Martin's preferred substrate) /XPS
    - Ply/OSB (***organic substrates***)
    - Owners choice of finish (LVP/LVT/etc...

    I am approaching this basement build through a worst case scenario lens. In the event of a bulk water event (burst washer/dishwasher/water pipe flood, future foundation slab or wall leak , surely an assembly of rigid foam-cement board-inorganic final layer (e.g. LVL) would perform best and survive any severe/catastrophic bulk water events?


  13. Jon_R | | #17

    While it may be more than needed, it's an energy saving and robust approach. See below regarding dimple mat on basement floors. Cement board being more moisture robust than wood is obvious - perhaps others know why it isn't more common.

    Edit: I turned off italics with less than, /I, greater than.

  14. Expert Member
    Peter Engle | | #19

    (Why are the replies now in italics?....and how do you turn it off/on on this platform?)

    I think the choice of subfloor depends on finish selection. Cement board would be a great choice if the finish floor is going to be ceramic tile, vinyl, laminate, or maybe a glue-down engineered wood floor. The Utica project above would be pretty bulletproof with a ceramic tile floor, even in a clean water flooding situation. But everyone seems to want wood floors in their basements. For a nail-down floor, you need a good nailing base, and that's going to be plywood or OSB. That's fine, just consider them sacrificial layers in flood events.

  15. Dmitry_R | | #20

    Hello GBA community,
    I would like to revisit this question and expand on it:
    I am also in climate zone 3a and looking to finish the basement walls and slab.
    To address the water intrusion issue I had I have excavated 2 sides of my walkout basement with grade sloping down to 3rd side and installed dimple mat with exterior French drain system, Routed all of the downspouts away from foundation and graded the slope away from the building. I couldn't excavate the 4th side due to having slab on grade adjacent to the wall. I have performed moisture testing via taped plastic squares on that 3rd wall and slab and don't see any condensation. I also would like to note the basement slab is cracked (see photo). I am assuming that there is no poly under the slab and certainly no insulation.
    I have several questions that I would greatly appreciate your help with:

    Q1) Since I am finishing the basement - do you absolutely advise to install the radon mitigation pipe via cutting the slab and installing few ft of perforated pipe with outlet leading to the outside?

    Q2) Seal all the cracks. What crack filler do you recommend?

    I was thinking to use something elastomeric to allow for slab expansion. something like this:
    or this:

    Q3) If there is a better product out there? please let me know what it is....

    Then I plan to Install either:

    a) sealed joint dimple mat/ 2 layer of advantech flooring or cement board/ finish floor
    b) eps foam /2 layers of advantech/ finish floor
    c) Dimple mat/eps/2 layers of advantech flooring/finish floor

    Q4) What option is recommended ? Obviously less thickness is preferred

    Q5)In case of water intrusion from the ground up or wall will eps will soak up the water and will hold it? Will this require eps and the whole floor assembly to be replaced to avoid potential mold heaven?
    In this case option a) and c) make more sense since the water will stay between slab and dimple mat. But in these cases water will have to dry out or go somewhere. There is no interior French drain or pump so:

    Q6) What is going to happen with this water?

    Since the idea of installing rigid foam is to prevent warm moist air reach cold slab/wall- although I have read that EPS is somewhat vapor permeable. I am assuming the rate of permeability is not make the real world difference...otherwise the whole insulation idea would be bad If that is the case then

    Q7) what happens with moisture in in the slab if it can't dry thru eps? (I am assuming there is always will be some moisture that gets transferred from the ground to the slab and to the interior but will be stopped by eps. ) Isn't that a formula for mold growth between slab and eps?

    Reading the pdf The Stay Dry , No Mold Finished Basement:,%20No%20Mold%20Finished%20Basement_FHB169.pdf

    I noticed that the plywood installed against the concrete wall and the wall insulation sits on top of plywood (see attachment here).

    Q8) Isn't it better to have wall insulation sit on top of slab insulation for uninterrupted envelope with no thermal bridge and to avoid any potential moisture at the edge of the plywood where it is most venerable?

    It make sense to install basement partitions on top of the subfloor to avoid interruption in the slab insulation if installed directly on slab. What if I want to erect the partition that will supplement as support wall to stiffen 1st floor joist above to reduce the 1st floor "bounce". I can do the calculations with some assumptions of the loading to check if foam will not be crushed but wanted to ask if
    Q9) Is it a common practice to install wall on top of plywood resting on eps ?

    1. GBA Editor
      Martin Holladay | | #21

      You have more questions than can be easily answered on a web forum. What you need is a consultant who can visit your site.

      In general, finishing a basement is risky. If you don't want to accept the risk, don't finish your basement.

      If you are worried that rising water might damage the materials you are about to install, then you need to install an interior French drain connected to a sump with redundant sump pumps, or an interior French drain connected to a drain pipe leading to daylight.

      Note that you can't prevent water from entering through slab cracks by sealing the cracks.

  16. Dmitry_R | | #22

    I am afraid that the consultant might be less knowledgeable than you and people on this forum.
    I absolutely agree that that interior French drain would solve the bulk water intrusion worry .
    However I think I have effectively eliminated this issue by work done outside as it has rained extensively and my basement staid dry and moisture free based on plastic squares test.
    I was thinking about best option to finish the basement just in case some moisture gets thru the slab which it most likely will as vapor or not bulk water.

    Taking that in the account would that be easier to help me answer my questions ?

    I am looking to seal those cracks to minimize radon escape path. Is’nt this what’s recommended?

  17. Expert Member
    Peter Engle | | #23

    Have you tested for radon, and was it high, or even marginal? If the radon is low now, you don't have to worry much about it changing significantly over time. If it is marginal or high, you should consider an active radon system as a part of your basement finishing. If there is an active radon system, the cracks are usually sealed to help the radon system achieve proper suction and to avoid pulling conditioned house air down through the cracks into the radon system. Your finish floor layers will do that more effectively than sealing the cracks. In fact, if you have an active radon system, the suction would tend to pull moisture from the under-floor cavity down into the soil below the slab through the cracks - the cracks would act like drains.

  18. Dmitry_R | | #24

    Peter, thank you . This absolutely makes sense . I was planning to do the radon testing prior to deciding how to proceed with or without mitigation.
    Do you have any thoughts on any other of my contemplations ?

  19. Jon_R | | #25

    Here are some general comments:

    a) dimple mat distributes moisture and increases drying rate
    b) vapor permeable insulation (say unfaced EPS vs foil faced EPS) results in dryer concrete
    c) wet concrete can grow mold
    d) yes, thermal bridging is best avoided
    e) don't put foundation drains above the slab level (as shown)

    1. dnnrobert | | #26

      What would be the minimum R-value for a basement slab with poly underneath in zone 5? Finished floor would be synthetic carpet. Slab does not get especially cold in the winter, so comfort isn't a major concern. Just want to make sure the slab stays warm and dry enough to not worry about mold. It seems like DMX Airflow would be a simple solution, but it only has an R value of 1.9.

    2. dnnrobert | | #27

      Never mind -- looks like the answer (R-3 minimum) is here:

  20. rob_g_1 | | #28


    I also want to finish my basement flooring with a dimple mat for water management and insulation. I read in this article ( that you should use r-10 insulation (I live in climate zone 7) so 2" of XPS, a sub-floor, and then finishing.

    However, on the Building America Solution Center they recommend not using mechanical fasteners through the insulation and dimpled matt (intrusions of gas and thermal bridging?).

    My question is: if you should avoid mechanical fasteners, what should you use? An adhesive? Will the weight just hold it in place? Something else?

    I also wonder if I could reduce the r-value to 5 if I wanted to save 1" of headroom. It seems it would be okay since I have r-zeroish right now. :-)

    Any feedback would be appreciated.

  21. Expert Member
    PETER G ENGLE PE | | #29


    The way you get away from fasteners is to use a floating floor as specified in that document. There are several ways to do this. One of my favorites is to use 2 layers of 1/2" plywood with staggered seams, glued and screwed together. This results in one big sheet of plywood that stays in place just because of its weight. It provides a 1" nailing base for any flooring to be installed on top of it.

    The R-10 insulation is based on the code minimums for your climate zone and also on the amount of insulation necessary to keep the subfloor warm enough that it is unlikely to experience condensation with typical floor coverings. R-5 insulation would be a riskier assembly, would have significantly more heat loss to the ground, and may not meet local codes.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |