GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X 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

Flat roof (TPO over polyiso build-up) insulation guidance

user-5254480 | Posted in General Questions on

I am trying to figure out the best insulation build up under this roof situation.

– 27’x29′ flat roof with 5/8″ plywood sheathing.
– There is polyiso over the plywood built up 1/4:12 on all four side to the the center. There is 1/4″ of ISO at the edges built up to approx. 3.5″ in the middle.
– There is no ice/water or vapor barrier under the ISO. The ISO is covered with 60mm TPO.
– The plywood seams where not tape or sealed
– The ceiling joists are 14″ TJI 360 at 16″ OC.

During the roof slope build up it was decided to use polyiso vs hand framing. The idea was that it would be a bit faster and cheaper to build up with polyiso and we’d also get the R from the polyiso.

However, the framing plans called for bird blocked venting down the joist bays with openings on the east & west side. So, I now have a polyiso sloped roof with outdoor air running underneath it.

After addressing some similar issues with another roof on the house on this thread here:
Metal shed roof – insulation schooling

I am now realizing that I have useless R over the roof and no real top side ventilation for the decking and don’t know how to address the interior insulation, interior vapor barrier build up, and how to protect the sheathing long term.

I did re-read this post, but I am still confused.

https://www.greenbuildingadvisor.com/blogs/dept/musings/insulating-low-slope-residential-roofs

Thoughts, guidance? Thanks.

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. GBA Editor
    Martin Holladay | | #1

    Eric,
    We need to know your climate zone. If the polyiso tapers down to 1/4 inch, it's basically worthless anyway. So you'll have to treat this as a vented roof assembly. My guess is that your 14 inch joists aren't deep enough for the required thickness of insulation plus the required 8 inches of air space above the insulation layer.

    And did you remember to build a doghouse (cupola) in the center of your roof for venting?

    You have already provided a link to my article on this topic. The article explains your options.

    If you realize the extent of your errors, you may decide to install more rigid foam above the roof sheathing and to seal up the vents. But that will be expensive, because you'll have to re-roof.

  2. user-5254480 | | #2

    Hi Martin,

    I am in zone 4C (Seattle).

    No there is no doghouse in the middle. The original non-ISO design simply called out 1" air gap in the joist bay with R49 batts.

    When the decision was made to go with ISO I was not up to speed on the implications of the recommendation.

    Installing more rigid foam would require ripping off the TPO. Not the end of world. I want this done right and would rather address now when everything is opened up vs. doing this down the road and dealing with rot, leaks, etc.

    Is going with either of these solutions noted in your article an option?

    - You can install a thick layer of closed-cell spray polyurethane foam on the underside of the roof sheathing.
    - You can install a more moderate thickness of closed-cell spray polyurethane foam on the underside of the roof sheathing, supplemented by a layer of air-permeable insulation below that.

    Thanks for the help.

  3. Expert Member
    ARMANDO COBO | | #3

    FWIW, I spec flat roofs on half of the high-performing houses I design, and it’s always a conditioned attic. My typical spec for CZ4 is .5/12 trusses, plywood sheathing, 3” min. ISO on top of the roof decking, 60 mil TPO and 8” OC Sprayed foam min. under the roof decking. I always have the trusses designed with 12” min. parapet wall on the high side of the truss, and we install screened and louvered vents on the face of the parapet walls.
    If you need to run your HVAC ducts in the attic, make sure the trusses have a dedicated chase built in them so you can run a straight trunk and branch to each supply vent.
    To me it’s the “perfect roof assembly”.
    Forgot to mention, we install ISO crickets, scuppers and downspouts off the parapet. Depending on the architectural style of the house, we use metal or stone coping, or foamed stucco in Pueblo homes in the Southwest.

  4. GBA Editor
    Martin Holladay | | #4

    Eric,
    I think in your case, the remedy you suggest is the best way to go. Install at least R-10 of closed-cell spray foam on the underside of your roof sheathing, and seal all vent openings.

    Once you have done that, you can add R-39 of fluffy insulation (for example, fiberglass batts or cellulose) under the cured spray foam (and in direct contact with the cured spray foam).

  5. user-5254480 | | #5

    Martin,

    I am digging this thread back up as I think I may have missed a critical point related to the ratio of top side to underside R value for the sheathing. (Which I am now up to speed from this post: https://www.greenbuildingadvisor.com/community/forum/general-questions/57876/closed-cell-foam-unvented-ceiling-and-vapor-barrier#comment-152401)

    As I noted above the top side insulation is tapered ISO, running a 1/4" at the edge to approx 3.5" at the middle of the roof.

    If I use CC spray foam on the underside to build up my entire R49 goal does the >20% R rule still apply for the top side? Since my target is R49, I need at least R10 on top and that does hit until the roof is 8FT in from the edge.

    Should I try and taper the foam thickness down as I move towards the center of the ceiling?

    Thoughts?

    Thanks.

  6. Expert Member
    Dana Dorsett | | #6

    With as little as 2" of closed cell polyurethane on the underside the roof deck is protected from interior moisture drive.

    From the previous thread(s) it sounds like you're in climate zone 4, and only need ~20% of the R to be outside the first condensing surface. As long as you have at least 20% of the total R exterior to the fiber insulation at all points, the fiber stays dry too. Tapering it isn't all that critical.

  7. user-5254480 | | #7

    Thanks. Yes in Seattle (4C).

    For the underside I can address with at least 2".

    The top side is what I am concerned about as the tapered ISO is already laid down and covered with TPO. The edge of the ISO in 1/4" so to get to R10 that is about 8ft in from the edge. If I use cc foam underneath, should I be concerned about condensation on the top side in the 8ft area?

    Thanks

  8. Expert Member
    Dana Dorsett | | #8

    No worries about trapping moisture in that 8' of taper. With only 2" of closed cell foam it's vapor permeance is about 0.5-0.6 perms, which is slightly more vapor permeable than a kraft facer on a batt (when the facer is dry). That is sufficient protection from wintertime moisture drives, but is still plenty of drying capacity for the roof deck.

    In your stackup and climate 2" of ccSPF is both the least and the most you'd want to install there. At 2" even ignoring the 1/4" polyiso above you'd be at R12-R13, which is about 25% of the R49. With R38 of fiber under that it will be just fine. The most you can safely install in a single pass is 2", and going much more than impedes drying.

  9. user-5254480 | | #9

    Thanks. Sorry still confused :-)

    Are you saying that there is no issue with the top of the sheathing building up condensation in the 1/4" to 8' span?

    I think I get it... as long as I don't have vapor driving up into the sheathing and there is no underside heat rising into the sheathing - then there is no moisture to create condensation from the outside temperature on the decking. Correct? :-):-(

    Our plan right now would be to do all cc foam doing multiple passes - carefully - to ensure curing between passes.

  10. Expert Member
    Dana Dorsett | | #10

    Maybe it's me who was confused- are you planning to put R48 closed cell foam under the thin edge, and not a foam+ fiber solution?

    With an all cc foam solution the roof deck needs to be verified dry or very dry (under 15% moisture content, measured with a meter) if you're planning on 8" of closed cell foam.

    That much closed cell foam is about the LEAST green way to hit R49 though, due to both the HFC245fa blowing agent used in most closed cell foam, and the polymer content. If you want a greener and more resilient all-foam solution, 2" of closed cell in contact with the roof deck, and the rest as open cell foam (applied in 5.5-6" lifts) is half the polymer per R, and uses only water for the blowing agent, and takes only three passes (one for the 2" closed cell, then two passes of open cell.) With open cell you get less thermal bridging from the rafters/joist/trusses, since it's more than 1.5x the path through the R1-ish per inch wood.

    The moisture problems with roof decks under an EPDM or TPO roof are due to moisture from the interior side, not the exterior condensing (really adsorbing, not condensing) in the roof deck when the roof deck is colder than the room air's dew point. It can get in either via air leaks (not a problem with foam), or vapor diffusion (it matters what the vapor permeance is.) TPO is a true vapor barrier- any moisture that finds it's way in has to leave via the interior, since there is effectively zero drying toward the exterior. At 8" the vapor permeance of closed cell foam is 0.1-0.15 perms, a "dries practically never" near-vapor-barrier.

    Typical half-pound open cell vapor permeance is on the order of 10 perms @ 5.5" (R20-ish), around 5 perms @ 11" (R40-ish). The 2" closed cell foam would keep the moisture content of the wood bounded, and would still have a reasonable drying rate if it ever got damp, and the 10-12" of open cell would barely move the needle on that drying rate.

  11. GBA Editor
    Martin Holladay | | #11

    Eric,
    Here's my advice: to keep things simple, ignore the fact that you have polyiso above the roof sheathing. Since this polyiso is as thin as 1/4 inch in certain areas, it's best to ignore it. Just install the 2 inches of closed-cell spray foam on the underside of the roof sheathing, followed by about R-36 of fluffy insulation on the underside of the cured spray foam, and stop worrying.

    It's true that this approach encapsulates your roof sheathing between two layers that aren't very vapor-permeable, but builders do that all the time. The roof sheathing will be protected from interior moisture drives during the winter by the closed-cell spray foam, which is both an air barrier and a vapor retarder. The roof sheathing will be able to dry very, very slowly toward the interior through the spray foam, but there really shouldn't be any moisture in the roof sheathing that needs to dry out. In short, don't worry.

  12. Dana1 | | #12

    With TJIs that aren't being fully filled it's sometimes cheaper to use open cell foam rather than fluff, but Martin & I are on the same page here, with 2" closed cell max, with the rest being substantially more vapor permeable.

    With 14" of depth to work with this COULD be a full cavity fill, using 2" of closed cell foam (R12+), and 12" of cellulose R44, for a total cavity fill of R56+. The cavity fill would still have a 21% fraction of exterior foam, and any polyiso on top would be gravy, increasing the dew point margin. The wicking and hygric buffering capacity of cellulose would safely redistribute any moisture that might accumulate at a thin spot without damage or temporary loss of function. This is better than any all-foam approach, since there is also a measurable thermal mass benefit to a foot of cellulose, with a much lower environmental footprint than a foot of open cell foam.

  13. user-5254480 | | #13

    I have been delayed in replying here, but I wanted to say thank you. Thank you very much for such a valuable and truly helpful community. Martin and Dana - you both rock. I am working on bids with CC+OC and CC+fluff.

    I had been waiting to see how the sheathing dried out before moving forward. Good news... with 80 degrees of heat and a few dehumidifiers we are down to 8-10% and dropping.

    Are bats an alternative to the blown in cellulose?

    Thanks again for all of your help and guidance.

  14. GBA Editor
    Martin Holladay | | #14

    Eric,
    Q. "Are batts an alternative to the blown in cellulose?"

    A. Yes. Of course, batts have to be installed correctly to make sure there are no voids. You also need to make sure that you meet minimum code requirements for R-value, and that the batts are in direct contact with the cured spray foam, and are held in place in some manner to be sure they don't slump or succumb to gravity.

  15. user-5254480 | | #15

    Thanks. We are moving with the 2" CC on the underside with batts to fill out the rest of the cavity. As you said this is the simplest approach and the spacing and r values work out.

    Three questions:

    1) When using CC+Batts are can lights not advised. You replied to this on another thread I posted, but we were talking about a different build up. I am reading this as "No - not advised - due to air and moisture leakage into the cavity", but wanted to confirm.

    2) Do we need an additional vapor barrier between the sheet rock and the batts? My reading so far is no - since with this build up we want to enable drying towards the interior.

    3) Following on #2 - should the batts be faced?

  16. GBA Editor
    Martin Holladay | | #16

    Eric,
    Q. "When using CC+Batts, are can lights not advised?"

    A. No, recessed can lights are not advised. They create hot spots near your roofing, which is a bad idea, and they take up volume in your rafter bays that should be filled with insulation.

    Q. "Do we need an additional vapor barrier between the sheetrock and the batts?"

    A. No.

    Q. "Should the batts be faced?"

    A. Unfaced batts are best, but if you have already purchased kraft-faced batts, they can be used. In this application, the kraft facing is unnecessary but relatively harmless.

  17. user-5254480 | | #17

    Thank you Martin.

    And thanks for supporting and building this amazing community of knowledge. You rock.

  18. Jayraja | | #18

    Hi

    Sorry to bring a old thread to life but this thread has multiple parallels with my questions.

    Here are some pertinent details..

    Seattle area (zone 4c)

    Our proposed roof buildup is

    Ccspf R49 under sheathing
    Sheathing with seams taped
    Water/air barrier (ice and water shield?)
    Tapered polyiso for a 1/4” in 12 slope
    PVC roofing membrane (min 60mm)

    I understand the issue with this sandwich but do have some questions.

    How does Advantech sheathing perform between 2 vapor impermeable layers?
    Should I change the insulation under the sheathing to 2” ccspf + mineral wool (to full depth)?
    Do I need a smart membrane under the mineral wool?
    With a minimum of R10 over the sheathing, can I reduce the amt of insulation under the sheathing?
    What would be a good insulation strategy under the sheathing?

    Any advice would be much appreciated.

    Thanks.
    Jay

  19. GBA Editor
    Martin Holladay | | #19

    Ray,
    Q. "How does Advantech sheathing perform between 2 vapor-impermeable layers?"

    A. As long as you the sheathing is dry on the day that the closed-cell spray foam is installed, and as long as there are no roof leaks, the sheathing will be OK. Eventually, every roof leaks -- and your roof will be no exception to this rule.

    Q. "Should I change the insulation under the sheathing to 2 inches of closed-cell spray foam + mineral wool (to full depth)"?

    A. You can if you want. That should be considerably cheaper than R-49 of spray foam, and should be better from an environmental standpoint as well.

    Q. "Do I need a smart membrane under the mineral wool?"

    A. No.

    Q. "With a minimum of R-10 over the sheathing, can I reduce the amt of insulation under the sheathing?"

    A. See this article: Insulating Low-Slope Residential Roofs. In your climate zone, at least 20% of the total R-value of the roof assembly must come from either rigid foam above the roof sheathing or closed-cell spray foam under the roof sheathing (or from the R-value of both, added together). If you install tapered rigid foam above the roof sheathing, you must assume that the R-value of the rigid foam is equal to the thinnest layer of rigid foam.

    Q. "What would be a good insulation strategy under the sheathing?"

    A. Read the article I linked to. If the tapered rigid foam is too thin to meet the 20% target, you'll need some closed-cell spray foam under the sheathing. Once that's installed, you can install fiberglass batts, mineral wool, or cellulose. The fluffy insulation must be in direct contact with the cured spray foam.

  20. Jayraja | | #20

    Martin,

    You guys rock. Invaluable resource.

    Advantech claims that their sheathing does not absorb any moisture. I agree that roofs/windows eventually leak. If moisture accumulates at the sheathing due to a leak or diffusion, where does it go as the sheathing is flat and slope is provided by exterior insulation? This is what I am trying to understand.

    Thanks.
    Jay

  21. GBA Editor
    Martin Holladay | | #21

    Jay,
    Q. "If moisture accumulates at the sheathing due to a leak or diffusion, where does it go?"

    A. It doesn't go anywhere. It stays where it is and contributes to sheathing rot. If you don't like the sounds of that, build a building with a vented unconditioned attic.

  22. Jon_R | | #22

    And if you aren't worried about sheathing rot from the eventual leak, I'd review how hard it is going to be to replace it (with spray foam below it and polyiso/PVC above it).

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |