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

How much insulation is enough to prevent mold/rot from thermal bridging?

[email protected] | Posted in General Questions on

When there’s frost on my roof, there’s no frost over the rafters.  The location of the rafters is clearly visible.  

I have a vaulted ceiling, with 5.5″ bays/rafters.  The roof deck needs to come off anyway (don’t ask; renovator put ventilation channel BELOW spray foam (only r-12 where r-31 is Code), etc). I could put continuous rigid foam on the rafters and install the roof deck thru the foam. (Details in footnote*.)

How much rigid foam would be enough to prevent mold/rot in the rafters and in the roof deck over each rafter?  Only R-7 (the wood rafter + roof deck) will separate -20 Celsius from 20 Celcius (-4 F to 68 F).

Since mold requires warmth to grow, do I not need to worry about mold, assuming moisture in the wood escapes before spring?  What if there’s a week of warmer weather mid-winter? Or does the heat in the house provide all the warmth required for mold in the rafters?

Will the rigid foam do more harm than good, impeding moisture transportation and evaporation?  

I could have a ventilation channel before the rigid foam, and a ridge vent at the peak, but that would make the rigid foam useless or worse, giving more rafter surface area exposure to outdoor air.

If adding rigid foam is more valuable than allowing a 1/2″ ventilation channel, how much foam is enough to prevent moisture damage?

How much foam is optimal for heat loss prevention?  Heating costs perhaps 30 cents per thousand BTU.  (70% of heat from heat pump, 30% electric resistance; 17cents/kWh; 3142 btu=1kW.) 200 sq ft room, 200 linear feet of rafter, 3600 square inches of rafter-roof interface.

Answers to any or all of the above much appreciated.

Between the bays will be 5″ spray foam.

Climate zone 5b, near Toronto.

* Adding rigid foam above rafters:
Laser light can guide where to screw – leave the top section (and perhaps bottom section) until last.

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. Jon_R | | #1

    > continuous rigid foam on the rafters

    Structurally, a bad idea. Check applicable code for this and for roof insulation/venting.

    1. [email protected] | | #2

      I want to believe that the closed cell spray foam between all the joists will add more structural integrity than any structure lost by having the roof deck farther from the rafters. Also the rigid foam running perpendicular to the rafters could arguably be said to add structural integrity. Is my wishful thinking (and a little bit of logic) misguided?

      R-31 require for vaulted ceilings. No specifications on venting.
      Briefly: Code is silent about thermal bridging in roofing assemblies. If my roof was a wall, Code would consider my R-7.2 framing+roof deck enough. Should a ceiling have greater need to address thermal bridging than a wall? If the Code doesn't care, does this mean I shouldn't care?

      Details if wanted:

      Code for Walls: "Except for a foundation wall, the insulated portion of a wall that incorporates wood stud framing elements that have a thermal resistance of less than RSI 0.90 shall be insulated to restrict heat flow through the studs by a material providing a thermal resistance at least equal to 25% of the thermal resistance required for the insulated portion of the assembly."
      RSI-0.90 = R-5.11. The rafters were nominal 4", built out by the renovator with another nominal 2", so total actual 5", plus 3/4" roof deck. If R-1.25 per inch is valid for this type of wood, that's R-7.2.
      So if my ceiling was a wall, I wouldn't need insulation over thermal bridging. I would expect a roof assembly needs more protection against thermal bridging than walls.

  2. andy_ | | #3

    "I want to believe that the closed cell spray foam between all the joists will add more structural integrity than any structure lost by having the roof deck farther from the rafters."
    I want to believe a lot of things that turn out to be not true. In this case, the sheathing is such an integral part of the roof structure that any gap could compromise the shear strength. Your roof would twist under wind pressure let alone a seismic event.
    One of the sketchiest parts of residential construction is when the roof trusses go up and the sheathing isn't on yet. We put a lot of diagonal bracing on at this stage so we don't wind up looking like this:
    or this:
    Even with bracing, truss failures still happen. Once the sheathing is on, they don't.
    The foam may supply some rigidity, but it's still flexible against the forces acting on the roof. Repeated movement will break the connection, and then things can get wild.
    Anyway....nailing down a new (or leave the old) roof deck and you can put foam over that. Another layer of ply (or even just strapping if you're doing metal) and you're off to the races. It's been done, proven, and works.

    1. Expert Member
      MALCOLM TAYLOR | | #5

      That's why Zip-R isn't approved for roofs.

    2. [email protected] | | #9

      Thank you! I had wondered if the beams at the ends of the rafters (peak and fascia) would give the structural integrity in that orientation. Your point is thoroughly received (agreed upon).

      The renovator sprayed 2"-3" of spray foam up against the roof decking - R-12-R-18 when Code is R-31 for cathedral ceilings (like this) and R-60 if it was a conventional attic. Worse?: He put the ventilation on the underside of the insulation. !!! I'm mostly thinking I need to start again, although I'd rather not waste the 2" of foam he installed + the existing roof deck. If I seal and insulate the soffits, I suppose I could add three inches of rigid foam on top of the existing roof deck. 3" on top of his 2"-3" would be enough to achieve R-31, but with plywood roof deck sandwiched between, I'm not sure it would be safe from moisture accumulation.
      There is no vapor barrier above the drywall ceiling.

  3. Expert Member
    BILL WICHERS | | #4

    >"Also the rigid foam running perpendicular to the rafters could arguably be said to add structural integrity. Is my wishful thinking (and a little bit of logic) misguided?"

    Rigid foam does essentially nothing to increase structural integrity. Fasteners will just pull through the foam, and the foam itself isn't particularly strong anyway. Closed cell spray foam DOES help a little, but I have my doubts about the claims you sometimes hear about exactly how much it does. I would make the structure act as the structure, and anything the spray foam contributes is something of an extra bonus. I would NOT NOT NOT rely on the spray foam for ANY actual structural purposes.


    1. [email protected] | | #8

      The beams at the ends of the rafters will give it structure in that direction, but yes, I suppose the structure added by the roof deck's connection is necessary. Okay, I'll heed that warning. Thanks.

      Edit: Just read details from Andy S.'s comment. Makes it very clear that the roof deck is an essential component.

  4. walta100 | | #6

    I seem that since you will be replacing the roof and deck the smart move is to select and build one of the options in this article.

    To my ear R31 sounds craze low R value for Canada.


    1. [email protected] | | #7

      R50 is code for standard attics, R60 for attics over electrically-heated spaces, but R31 for cathedral ceilings.
      I spoke with a civil servant who helps people make sense of the Code - I asked if cathedral ceilings require less insulation for some reason or if the Code writers compromised to since they know cathedral ceilings are harder to insulate. He didn't say the former; it seems the latter.
      Ideally I'll go above the minimum.
      On the other hand, some people have told me R-24 should be enough to prevent ice dams and condensation. I haven't fully done the math on that, but I'm skeptical. Also, I expect the optimum insulation value will be above Code.

      I'll look at the linked article. Thanks.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |