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

How bad is it to use polyiso on the deck and spray foam on the bottom?

[email protected] | Posted in Energy Efficiency and Durability on

I am having my “flat” roof EPDM rubber roof replaced, which right now is only insulated by whatever amount of polyiso foam is on top of the deck. The rafter cavities are empty and the ends, only, of those cavities is currently stuffed with fiberglass batts.

I live in St. Paul, MN, which is climate zone 6 and everything I read, indicates that I must have an R of 25 in order to keep my deck sufficiently warm to avoid moisture problems with, what will become a “hot roof” (polyiso on top of the sheathing and blown cellulose underneath with no internal vapor retarder and the rafter ends of the rafter cavities sealed with either sealed in rigid foam or with sprayed in polyurethane foam).

My dilemma is that in order to achieve an R-25 at the thinnest part of the polyiso insulation of my tapered roof (required for drainage), then I would have to have a parapet wall built on the roof in order to accommodate the requied thickness of the polyiso. Such a wall would severely degrade the appearance of my modern design house. Everything I read from Martin, suggests never to have BOTH polyiso on top of the sheathing and sprayed-in foam underneath it because, if it were ever to get wet, it would not be able to dry in any direction.

So ultimately, my question is this: Am I insane to take a calculated risk and have sprayed-in foam applied to my roof deck only where the polyiso is too thin to acheive the R-25? For the rest of the roof, I would follow the good advice of Martin such that I would not use the polyurethane under the sheathing.

Thank you in advance for your reply and thank you for your good work in providing the wonderful resouce that is GreenBuildingAdvisor.com

Brian W.

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.

Replies

  1. wjrobinson | | #1

    Open cell

  2. GBA Editor
    Martin Holladay | | #2

    Brian,
    A.J.'s suggestion will work: open-cell spray foam is vapor-permeable, so it would not interfere with inward drying.

    I'm not sure why you can't increase the thickness of the polyiso layer above the roof sheathing without building a parapet. I don't see how the presence of a parapet would be connected in any way with the thickness of the rigid foam.

  3. [email protected] | | #3

    Great suggestion AJ!!! Thank you! Regardless of the parapet relatinship to the foam, the open cell idea solves the issue and brings my roof quote down significantly! Thank you also Martin and GBA!

  4. [email protected] | | #4

    Sorry - I'm back. I just received another opinion on the subject. The latest is that while open cell will indeed dry inward, it will not prevent vapor from reaching the roof sheathing, which would be cold because my top side polyiso isn't thick enough to give me the R-25 required for my climate Zone (6) and the vapor would therefor condense on the underside of the sheathing. If it matters, the polyiso at the thinnest point, is only .5", which I think is only an R-3 (right?). So how do keep my roof deck warm without adding to much polyiso to the top, which at it's thinnest point would be quite thick (4" or so) and would only get thicker as required by the water management taper. The only answer I can come up with is to use closed cell insulatin on the bottom, but that takes me back to my first question - Is it too risky to do that? Thanks again.

  5. GBA Editor
    Martin Holladay | | #5

    Brian,
    You're right: If your exterior foam is only 1/2-inch thick, the plan won't work.

    I have no idea why you can't just install much thicker rigid foam on top of the roof, followed by EPDM. What's all this about a parapet?

  6. gusfhb | | #6

    A picture would help understanding the parapet issue, but they do make drip edge in large sizes to cover foam. Even if there was a tiny parapet that the foam butted up to now, it would only be a matter of furring up the roof edge to match the new foam and installing the correct size drip edge,

  7. [email protected] | | #7

    Sorry, I have no picture at this time, but the "parapet" really is just stacked 2x4's, but the roofer says he needs an 11" wall if the polyiso were 3.5" and only 7" if it were the 1/2". Both change the exterior aesthetic of my house unfavorably, but less is better to me from an aesthetics standpoint. Let's just assume the parapet has to happen - On a scale of 1-10 with 1 being crazy and 10 being perfectly safe, how risky is it to use polyurethane on the underside of the roof only where the insulation is too thin? Are there other options? Thank you Keith and again, Martin and in advance anyone else! Brian

  8. GBA Editor
    Martin Holladay | | #8

    Brian,
    Lots of roofs have polyiso and EPDM, but no parapet.

    I am attaching a photo and a detail showing examples of low-slope roofs with polyiso and no parapet.

    The photo came from here:
    http://www.lyonscontracting.com/blog/tag/leaking-flat-roof

    The detail drawing came from here:
    http://www.buildings.com/tabid/3334/ArticleID/5436/Default.aspx

  9. gusfhb | | #9

    Not speaking for anyone else, but if this is an open roof edge, there is no reason for a parapet, so there is a significant probability that your contractor does not know what he is talking about. It also occurs to me that you might be thinking you need more pitch than you do, resulting in more foam than you need. I think it goes down to 1/8 inch per foot

    WRT Martin's post, even that is more complex than most current EPDM installs, I think it is a edge terminated[or somesuch term] rather than glued down membrane. Glue down your edge is just a equal length right angle drip edge.

  10. [email protected] | | #10

    OK, I have a picture and will attmept to upload it now. Hopefully it will help:

  11. [email protected] | | #11

    Second try

  12. Expert Member
    Dana Dorsett | | #12

    Open cell foam on the underside with a Certainteed MemBrain vapor retarder works great in a central MN climate, with or without the iso above the roof (but you'd still want the iso as a thermal break over the rafters.)

    The R25+ above the deck is necessary if you have only a class-III vapor retarder on the interior, but MemBrain is a class-II vapor retarder when the proximate air is dry (as in winter), but becomes quite vapor-open when the sun is baking the moisture out of the roof deck in the spring, raising the humidity in the open cell foam. This limits rate of wintertime moisture uptake, but allows the roof deck to dry MUCH more rapidly than any constant-permeance vapor retarder when the warmer weather arrives, thus preventing rot. A cold sub-freezing roof deck doesn't promote biological activity no matter how much moisture it contains but if is allows to dry quickly when warmed the problem is avoided.

    MemBrain will run ~0.5-0.8 perms during the cold weather unless you actively humidify the place to well over 40%RH in winter- better to keep it 30-35% during January. But when the roof deck is warm enough to be biologically active it'll be releasing moisture and the MemBrain goes to 5-10 perms very quickly if there's much moisture to purge, and your drying rate will then be determined by the vapor retardency of your ceiling paint (~2-5 perms).

    In case you're wondering, I don't work for Certainteed, nor do I sell the stuff. But I'd use it if the thin edge of the iso is only R10 when it needs to be R25 per IRC guidelines.

    A more expensive solution that works would be to do a flash-foam of 1" of closed cell polyurethane (about 1 perm) on the underside of the roof deck, and fill in the rest of the R with open cell foam or dense-packed cellulose/fiberglass. In most cases the all open-cell + MemBrain would be cheaper, and has better drying capacity.

  13. Expert Member
    ARMANDO COBO | | #13

    In NM and the SW, most flat roofs (Pueblo Style) do have 12"-24" parapets to allow better drainage with crickets directing the water to Canales. As I look at the pictures and detail in this thread, I wonder what type of effective drainage system those roof edges have.

  14. [email protected] | | #14

    Yes, the roofer is planning a system of crickets to direct the water to scuppers and downspouts as you can see from the picture, there is considerable ponding on the roof now (and I do have some water damage that I have to take care of). Thank you Dana Dorsett for your answer! I'm very interested in the MemBrain and open cell as an option. QUESTION: What would be the minimum thickness/R-Value required for the thermal break over the rafters in MN (Climate Zone 6/)?

  15. Expert Member
    ARMANDO COBO | | #15

    You need to look a code book (2009 IRC R806.4 & N1102.1) and read about unvented attic assemblies. You would need R25 minimum of rigid foam on top of the roof decking if you are going to install O.C. foam under the roof decking. Some O.C. foams are air-impermeable at 6”, 8” or 10”. Knowing the permeability of the foam, and the thickness will allow installing enough insulation to meet a required R49 ceiling insulation in your climate zone. Those R20-R25 O.C. foam under deck installation stories can be very risky, and more than likely, against code.

  16. [email protected] | | #16

    Ughh! But thanks!

  17. [email protected] | | #17

    So then, what about closed cell foam (getting back to my very first question)? It seems I can't get enough foam on my sheathing without adding a tall enough parpet to make my house look like a commercial building.

  18. Expert Member
    ARMANDO COBO | | #18

    You don't want to trap wood sheathing between two closed cell foams. Installing closed cell foam only under the roof decking is allowed, but again, you need R49, installed at 2" increments, and your installed better be really good, or you could have major issues or problems.

  19. GBA Editor
    Martin Holladay | | #19

    On the other hand, Armando, there isn't going to be any drying upwards in any case, what with the EPDM -- so I don't see any differrence between closed-cell foam below (as the only insulation) and closed-cell below plus polyiso above.

  20. wjrobinson | | #20

    Do What I and Dana detailed. Done.

  21. Expert Member
    ARMANDO COBO | | #21

    You got a point Martin.

  22. Expert Member
    Dana Dorsett | | #22

    The vapor permeance of closed cell foam depends on it's thickness, and at 2" most are still more permeable than a kraft facer on a batt, and some water blown closed cell foam (eg: Icynene MD-R-200) is about 1.3 perms even at 3" (R15). So if you want to spend the money 9" of that product would get you to R45 center cavity and still have roughly the same drying capacity of a kraft-facer on a batt. But open cell foam + MemBrain would be a lot cheaper and would work better.

  23. [email protected] | | #23

    Just signed off on a plan from the insulation guy to go with Dana and AJ's suggestions. Open cell and MemBrain it is (and I had never heard of MemBrain until participating in this forum). Thank you all!

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |