Vapor barrier with open cell foam?
Hello all,
I have found some old posts on this issue, but I’m curious what the more recent opinions are regarding this topic. I am building a colonial house in Scarborough Maine zone 6. I have attic trusses and I would like to insulate the underside of the roof deck (already shingled otherwise I would have put sandwiched foam above). My spray foam contractor tells me it’s fine to use open cell foam. I’m hesitant to do this, but I’m curious if simply putting a vapor barrier below the open cell foam would prevent any moisture vapor penetration from potentially rotting out my roof sheathing? Thoughts? I realize closed cell foam would be ideal but much more expensive so I’m curious if anyone has done this? My roof rafter portion of the truss is 8” thick. I planned on putting strips of 1” foam under the rafter portion and screwing half of a ripped 2×4 (1.75”) along the length to prevent thermal bridging and still have a surface suitable for strapping and drywall. This route would allow me to have 10.75” of open cell foam equating to r-43.
so would anyone advise against adding a class 1/2 vapor barrier below the open cell foam?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
>"My spray foam contractor tells me it’s fine to use open cell foam. I’m hesitant to do this, but I’m curious if simply putting a vapor barrier below the open cell foam would prevent any moisture vapor penetration from potentially rotting out my roof sheathing?"
Installing a true vapor barrier on the underside of an open cell foam insulated unvented roof deck creates a moisture trap. While you might be able to get away with a "smart" vapor retarder such as Intello Plus or Certainteed MemBrain or half-perm "vapor barrier latex" paint (to avoid the moisture trap) it's still riskier than a vented roof deck.
>"I planned on putting strips of 1” foam under the rafter portion and screwing half of a ripped 2×4 (1.75”) along the length to prevent thermal bridging ..."
There is no such thing as preventing thermal bridging. An inch of foam is a thermal break that MITIGATES some of the thermal bridging, but isn't much of a thermal break on 8" of wood. You may want to bump the foam thickness and use 1x furring as the nailer, AKA "Bonfiglioli strips". For many tips on how to go about that, see:
https://www.finehomebuilding.com/membership/pdf/9750/021250059.pdf
So with 1.5" of foam (R9, if polyiso, comparable to the R value of 8" of wood), and 3/4" thick 1x furring Bonfiglioli strips you're adding 2.25" of depth for a total of 10.25".
In zone 6 it takes about 51% of the total R on the exterior of permeable insulation to provide adequate dew point control at the permeable/impermeable boundary. Using 4" of HFO blown closed cell foam ( ~R27-R28, about $5 per square foot), and 6" (R22) of half pound open cell foam or blown cellulose (much greener- it's sequestered carbon) gets you to code-min on an R-value basis with reasonable dew point margin.
At 4" the vapor permeance of most 2lb density closed cell foam is about 0.2-0.3 perms- a powerful class-II vapor retarder, but more than 5x the drying rate of 6 mil polyethylene. As long as the roof deck is dry when the foam is applied, the roof deck is protected.
You could back off to 3" of HFO blown foam (R21) and 5" of open cell foam or cellulose (R18-R19) and with no Bonfiglioli strips still have reasonable dew point control, and a "reasonable" total R from an energy use point of view. It might even meet code performance on a U-factor basis since the thermal bridging webbing of an I-joist is less than half that of dimensional lumber. (The U-factor would have to be calculated based on the framing plan and I-joist specs, but it would be close.) The downside risk to that in snow country is the ice-dam starting potential of the 8" deep low-R thermal bridge.
So would you say that 1/2 pound open cell foam at 11.25” In my zone would suffice for a reasonable dew point margin? Can you explain that a little more for my poor brain? I just don’t want any regrets in having the contractor do all open cell foam.
I’m getting the 11.25” with the following method.
7.25” actual depth of rafter
1.5” of polyiso
1.75” strip covering the polyiso using 5” screws
.75” strapping hung perpendicular below the rafters and strips for hanging drywall.
Nevermind I totally misread that.
I have built about 200 units in Minnesota south part of zone 7 . All low density foam. The only vapor barrier any had was the latex paint. Going back 25 years. I did not use rated paint as we where of the opinion any latex was fine. I will add though that every one used 25/32 Bilt Rite sheathing . R2.06 and very vapor permeable. Makes me sad you can't get BiltRite any more. Those buildings punched above there weight.
Insurers now not insuring spray foam homes in UK
https://www.theguardian.com/money/article/2024/aug/12/they-encouraged-us-to-insulate-our-home-now-its-unmortgageable
link to inspection protocol: https://www.property-care.org/write/MediaUploads/Technical%20Documents/SprayedFoamInsulationInspectionProtocol_March2023.pdf
Closed cell to my understanding is a vapor barrier in and of itself. Just had the underdeck of my roof quoted at $7 for a 7 inch deep square foot which came out over $21,000 and decided to use R30 + R19 over the first floor ceiling. The R30 fits perfect in 10 inch roof trusses and the R19 will layer over. The job was so inexpensive now, the installer wants me to do all the walls with open cell saying it costs less than fiberglass. My feeling is I rather have R23 Rockwool in the 6 inch walls so if I have to open a wall for a repair, I am not chipping away and dealing with foam. I could just pull out and reuse the rockwool, mineral wool, Roxul - seems it has different names.
Open cell requires a vapor barrier according to the insulation company rep. - unlike closed cell.
Code does not even require a vapor barrier over fiberglass if by barrier you mean poly.
We have been misled over the decades. Most of the moisture in our walls is carried in air that is leaking in or out. Low density foam stops air movement. Low density foam is easy to remove if you need to.
>"Closed cell to my understanding is a vapor barrier in and of itself. Just had the underdeck of my roof quoted at $7 for a 7 inch deep square foot which came out over $21,000..."
Even at 7" closed cell foam is still only a Class-II vapor retarder at about 0.15 perms, not quite a vapor barrier (less than 0.1 perm). While a Class II vapor retarder in air-tight contact with the roof deck is sufficient protection for the structural wood, it doesn't take more than 2" to get there (although 2" isn't sufficient for dew point control on a fiber layer at R49 total.) See Table 3:
https://www.buildingscience.com/sites/default/files/migrate/pdf/BA-1001_Moisture_Safe_Unvented_Roofs.pdf
Also, the thermal performance of 7" of closed cell foam isn't the same as R49 of lower R/inch materials. The thermal bridging of 7" of wood is pretty severe, undercutting the performance of closed cell foam. Installing any more closed cell foam than is necessary for dew point on the vapor permeable layers is a waste. Do the math:
https://www.finehomebuilding.com/membership/pdf/184243/021269086NRGnerd.pdf
>"...and decided to use R30 + R19 over the first floor ceiling."
Without a vented air space between the fiber and the roof deck the assembly will almost certainly fail due to high moisture content in the roof deck, primarily the north facing &/or fully shaded pitches.
If there is an air space between the R30s and R19s, only the R30 counts toward code compliance.