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

Open cell spray foam under roof deck

arkitkt | Posted in Energy Efficiency and Durability on

A couple years ago we converted our vented attic to an unvented/conditioned attic with open cell spray foam applied under the plywood roof deck. We even installed an HVAC vent in the attic to keep the RH in the attic similar to the rest of the house. So far everything seems fine but I’ve recently read about issues with open cell under roof decks, including one article on this website. I live in climate zone 4, asphalt roof over plywood deck. Should I be concerned about having a wet roof deck? And what can be done to help if so – possibly over roofing with 4 or so inches of rigid insulation? Which I’ve thought about doing anyway to increase the R value and eliminate thermal bridging from the rafters. 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.


  1. GBA Editor
    Martin Holladay | | #1

    I assume you've read them, but for any readers who haven't, here are the links to two GBA articles on this topic:

    High Humidity in Unvented Conditioned Attics

    Open-Cell Spray Foam and Damp Roof Sheathing

    Q. "Should I be concerned about having a wet roof deck?"

    A. If your attic is heated in winter by a forced-air register connected to a duct, and if your attic is cooled in summer by a forced-air register connected to a duct, you are doing what you are supposed to. One more thing you could do is to install a hygrometer to keep track of the relative humidity (RH) in your attic. That way, if the RH starts to spike, at least you'll know about it.

    Q. "Possibly over-roofing with 4 or so inches of rigid insulation? Which I've thought about doing anyway to increase the R-value and eliminate thermal bridging from the rafters."

    A. Installing rigid foam above your roof sheathing is an excellent idea. For more information on that topic, see How to Install Rigid Foam On Top of Roof Sheathing.

    -- Martin Holladay

  2. arkitkt | | #2

    Thanks for the information - I do have a forced air register in the attic. If I decide to add rigid over the deck, in climate zone 4 would you recommend polyiso or XPS? I've read pros and cons of both. If I add 4" I'll be at about R45 on the roof - roughly half above and half below the existing deck. Is that sufficient in climate zone 4?

  3. Expert Member
    Dana Dorsett | | #3

    Polyiso is a lot greener than XPS (blown with pentane rather than high global warming HFCs) and will slightly outperform it inch-per inch. Over the long term 4" XPS will drop from R20 down to under R17 as the HFCs leak out. During mid winter it's fully depleted performance will average a bit higher than that, but not more than R18-R18.5-ish.

    Four inches 2lb density fiber faced roofing polyiso is rated at about R22-R23 (~ R5.7/inch) but during the coldest part of the winter it'll be performing at about R5/inch or a bit less, but a bit higher than R23 during the early & late winter seasons in a zone 4 climate.

    If you use 1-1.5lb density foil-faced polyiso it's labeled rating at 4" will be R24-R26, but when derated for temperature it's not going to make much difference in wintertime performance.

    That would be huge dew point margin on ~R20 of open cell foam in 2x6 rafters (or even ~R27 open cell foam in 2x8 rafters.) In zone 4 as long as at least 30% of the total center-cavity R is on the exterior the risk to the roof deck is pretty low as long as there is a class-III interior side vapor retarder (5.5" of half-pound open cell is usually a minimal Class-III vapor retarder) With 4" of polyiso you'd have more than 50% on the exterior, and the vapor retardency toward the interior becomes irrelevant.

    When R20 or more of the total R is continuous, not thermally bridged by rafters, an R45-ish total will slightly exceed code-min performance on a U-factor basis. The R49 code-minimum assumes thermally bridging joists or rafters at about a 7% framing fraction. On a U-factor basis you only need about R38-R39 "whole -assembly", with the thermal bridging factored in, and the R-value of the other layers (shingles, roof deck, indoor & outdoor air films etc) added.

  4. arkitkt | | #4

    A year later and we're finally ready to install the rigid insulation over our existing deck. I've been working with a local contractor and he is proposing a slightly different technique. The roof currently has 4" of medium density open-cell spray under the deck giving us an R-Value of just under 22.

    Instead of installing a couple layers of polyiso with staggered joints and then decking, he is proposing using a product like the GAF ThermaCAL insulated nailing base. It would come with 3 1/2" or 4" of polyiso adhered to a 7/16" deck. The total R value of the panel, that is insulation above the existing deck, would be 21.1 or 24.2, depending on the thickness used. The total roof would then be about 43.5 or 46.5 with about 50% above the deck. The polyiso on the ThermaCAL is tongue and groove - wil this reduce thermal bridging/heat loss enough at the seams? Should a bead of spray-foam in can be used at the joints to seal them?

    Do you see any other issues with using this system instead of standard polyiso and decking?

    My other question is, when looking at details for over-roofing on the US Dep of Energy website, the detail shown lists an R value of 60 minimum, but the energy code requirements for Climate Zone 4 list R30 minimum with R45 as the high target for a compact roof. Can you clarify which is correct? Will a total roof of R43 to R46 work with about half above the deck with no problems of ice damming, etc.?
    Thanks again.

  5. charlie_sullivan | | #5

    "will this reduce thermal bridging/heat loss enough at the seams?

    I think it would work fine.

    "Should a bead of spray-foam in can be used at the joints to seal them?"

    That would be a good idea. What does the manufacturer of the panels recommend.

    In a retrofit, you can usually treat code as a recommendation. I think that the combination you are planning will do a great job of avoiding ice dams, and that the combination of interior and exterior insulation will do well at avoiding moisture problems.

  6. arkitkt | | #6

    Thanks for the comments Charlie. Any thoughts on my earlier post of whether I might have any issues with a wet roof deck if I don't add the rigid on top? On frosty mornings, depending on how heavy the frost is, I can see the rafters where the frost in melting a little faster and a few other small spots but overall it's pretty good. And of course it melts a little faster around the skylights. There are also few spots over the eaves on the gable ends where the spray foam has not stopped heat loss from the top of the outside walls.

    I've received a quote for over-roofing is not cheap but I also don't want a wet moldy roof deck.

  7. arkitkt | | #7

    One more question - if I go with the GAF ThermaCAL product, they have both vented and unvented options. That is a space above the decking/below the shingles. Would I be better off going with the vented and sacrifice R value - I'd be at at R17.5 above the deck if I go with the 4 1/2" product. Or better R value of 24.2 if I went with the unvented 4 1/2" product?

  8. user-2310254 | | #8


    I'll respond to your post to give it a bump. It's my understanding that venting a cold roof is most beneficial in regions where homes develop ice dams over the winter. I don't imagine that is much of an issue in your area, but I could be wrong.

  9. charlie_sullivan | | #9

    I think you will get much lower risk or roof deck rot with either the vented or unvented product. I think it's a tough call deciding between the two. The vented shouldn't be necessary, but in practice, there seem to be more failures of assemblies like that without venting. So venting is a good idea, in my opinion. and I think even with the lower R-value you have enough to protect the deck below it. Of course the best would be the thicker version, but with venting added above it. But that's going to add even more cost.

  10. arkitkt | | #10

    Charlie, You mentioned more failure on unvented roofs - are you referring to shingle failure or roof system failure, like deck rot? I thought about the thicker version with venting but it adds cost and makes the existing roof that much thicker and trickier to detail. From what I've read, venting under teh shingles is required in high snow load areas to prevent ice damming but we are not in a high snow load region.

    I could possibly go with a 5" vented product with an R value of 20.50 - this would give me a total R value of just under 43. This puts almost half the R value above the deck and gives a total R value just under the high target R45 suggested by the 2012IRC/IECC. Minimum for climate Zone 4 is 30.

    Also, the minimum thickness to eliminate roof deck condensation in climate Zone 4 is R15 so I'd like stay comfortably above that, leave a margin for error. That's why I'm concerned with using the vented 4 1/2" product because the R value is only 17.5.

  11. arkitkt | | #11

    And thanks for the comments - this is a pretty major project so I want to get it right the first time.

  12. arkitkt | | #12

    So I've been more research (this is a very expensive undertaking so I'm still justifying if it's really necessary or if I'm just being compulsive) and understand that if the roof deck is insulated with open-cell in climate zone 4, as long as the attic is conditioned to control humidity there really shouldn't be any problems with a wet/rotting roof deck. Well, most of my attic is conditioned - it's a 1 1/2 story house so there are a few areas under the sloped part of the ceiling where the space is not really conditioned. All little are probably leaks in from the conditioned areas (around the framing, through the ceiling, etc.) of the attic but there's really no way to measure the humidity in those areas.

    Any thoughts on if I should be concerned about those areas possibly having a 'wetter' roof deck? I still plan to over-roof someday, I was just thinking about waiting a few years (the roof is only 5-6 years old) until the roof is in worse shape and I could justify a replacement. Thanks again.

  13. GBA Editor
    Martin Holladay | | #13

    If most of your attic is conditioned, the chance of sheathing rot is quite low. That said, R-22 isn't much roof insulation, so if you can possibly afford to install the nailbase above the existing roof sheathing, you'll have a much better performing roof. I think unvented nailbase would work fine for you.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |