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

Smart or not? Combining closed-cell and open-cell foam

user-509810 | Posted in Energy Efficiency and Durability on

I had a recent conversation with a BPI certified energy auditor. We were talking about encapsulation solutions for my 1950’s attic spaces. He suggested the use of 1 inch of closed cell foam on the roof deck followed by 8 to 10 inches of open cell fully encapsulating the rafters. Is this a good or bad idea?

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've heard worse. Where are you located?

    First of all, when your wrote of installing "1 inch of closed-cell foam on the roof deck," I assume that you mean on the underside of the roof deck -- not on top of the roof deck. Am I right?

    The closed-cell spray foam layer will help prevent interior moisture that diffuses through the open-cell foam from reaching the cold roof deck. That makes sense. As with any spray foam job, success depends on choosing a skilled contractor who installs the foam carefully and doesn't skimp on foam thickness.

  2. Expert Member
    Dana Dorsett | | #2

    Climate really matters here. If the closed cell-R to open cell-R ratio isn't big enough you can end up with some soggy under-performing layers in the open cell foam by the end of winter.

    No matter what climate you'd need to spray on "vapor barrier latex" on the interior side of the open cell foam which ends up at about 5 perms (as opposed to 0.5 perms, when applied to wallboard.) To limit moisture adsorption rates in the open cell foam.

    Be sure that the contractor installs the open cell in two lifts, not a single shot of 8-10", or you'll have adhesion/shrinkage issues, or (worst case) a fire shortly after installation due to the heat generated during curing.

  3. user-509810 | | #3

    Sorry climate zone 4 and yes under the roof deck for CC and OC application...

  4. davidmeiland | | #4

    I'd be skeptical of getting enough closed cell when they're shooting for 1"--you could easily end up with thin spots. I would accept the cost of spraying 3" of closed cell and play it safe.

  5. Expert Member
    Dana Dorsett | | #5

    In any variation of US zone 4, bumping up the closed cell to 2" with 8-10" of open cell would be fine, as long as you use the vapor barrier latex directly on the interior surface of the foam.

  6. user-509810 | | #6

    I thought I read somewhere on this site the vapor barrier paint applied directly to foam does not work.

  7. GBA Editor
    Martin Holladay | | #7

    Unlike Dana, I'm skeptical of the performance of vapor retarder paint on cured spray foam. Joe Lstiburek told me that he tested it and it doesn't work.

  8. wjrobinson | | #8

    Have never seen moist spray foam. I take care of two homes with open cell, dry foam everywhere, roof, attic.... Cellar 1st floor, rim joist... Dry.

    Please post the data, number of failed installs in zone 6a. I know of none.

  9. Expert Member
    Dana Dorsett | | #9

    Martin, Woody: Lstiburek et al have written in several documents that vapor barrier latex on open cell foam, delivers only about 5-perm performance.

    That's an order of magnitude more permeable than the specified 0.5 perm performance of the same stuff when applied to wallboard. But "it doesn't work" is only in the context of expecting class-II vapor retardency. The foam itself has a MUCH higher vapor permeance- and the vapor-barrier latex really DOES work, just not at class-II vapor retardency. The reported performance when applied to open cell foam still solidly in the middle of the Class-III range, making it comparable to standard latex on wallboard. At that performance, stackups that work with interior side Class-III vapor barriers under code-prescriptive R-ratios would still work with vapor-barrier latex on open cell foam.

    See the discussion under figure 9 on p18 (p 27 in PDF pagination) of BA-1312 ,.authored by Aaron Grin, Jonathan Smegal and Joseph Lstiburek, released late last year. Note where it states:

    " A 5 perm coating was modeled, as this represents the actual effective achieved perm value based on BSC experience with spray-applied Class II vapor retarder coatings."

    (That document is also VERY relevant to this thread topic.)

  10. user-509810 | | #10

    Thx Dana

  11. wjrobinson | | #11

    The site of the following quote has much more that is interesting to read.

    "Our opinion is also based on logic. Open cell foam has been installed in this manner in upstate New York for decades now. We are unaware of any roof failures resulting from this application method in the region. Additionally, we are only aware of a few failures in the entire country and those were all in extremely cold winter climate areas where heating degree days surpass 8,000. The applicator in those cases either put an insufficient amount of insulation in the roofline or they failed to use a vapor barrier (which must be used with open cell in climates that are that cold). Even if you don’t subscribe to our opinion, open cell foam sprayed directly to a roofline can be coated with a vapor retardant paint for $0.10-$0.15 per SF, which will effectively give it the moisture permeability properties of closed cell foam"

    The quoted site states the way I feel about spray foams and moisture and whether there even are any failures due to moisture. Like I stated before, the projects I did are doing well and are much more energy efficient than any prior projects of mine.

    Personally I am limiting my use of spray foams. Air sealing and insulating is key to energy savings. It can be done with out spray foam.

  12. Expert Member
    Dana Dorsett | | #12

    Unless they can show you third party data, take statements from a foam installer/vendor with a big grain of salt. This is especially the case for statements like:

    "Even if you don’t subscribe to our opinion, open cell foam sprayed directly to a roofline can be coated with a vapor retardant paint for $0.10-$0.15 per SF, which will effectively give it the moisture permeability properties of closed cell foam. "

    Really? For that you'll have show me, not tell me!

    Unlike someone hawking a product, the folks at BSC have used the vapor barrier paint on open cell approach, then MEASURED the moisture performance of the o.c. foam/v.b. paint combinations, which turn out to be an order of magnitude more vapor permeable than closed cell foam, as shown by the measured moisture content cycling of the roof deck.

    The numbers aren't an opinion- it's data.

    Engineering instincts make me inclined to design stackups based on data, and to ignore armchair theories and "If it's really a problem where are the dead bodies?" type arguments. It's important to view the relative risks as a probability function with lots of moving parts- average indoor humidity levels will vary (a lot) from house to house, and the rot potential of the roof deck also varies with the material, OSB being more susceptible than CDX, and CDX being more susceptible than ship lap planking. Even the orientation and color of the roofing material affects the average temperature/moisture content of the roof deck, north facing light-colored decks being more of an issue than south facing dark shingled roofs.

    Most roof decks only get the most modest amount of inspection. Short of catastrophic failure events that's usually at the time of re-roofing. Even then unless somebody puts their foot through it the roofers are inclined to just add a shingle layer and move on. Just because not much prima-facie evidence of the problem has yet to surfae in the ~25-30 years since half-pound polyurethane began to be installed under roof decks in US climate zone 6 doesn't mean that what's is under the shingles is destined to make it to age 50. Only the absolute worst vapor diffusion problems show up as catastrophic failure in a mere decade or less, but that doesn't mean it's not a real problem.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |