Replies

1. Expert Member
   Dana Dorsett | Nov 12, 2018 03:45pm | #1

   The short answer is "no" on the vapor barrier. I assume there is soffit to ridge venting in each rafter bay?

   Is it a full 12.0", or is it 11.25", like dimensional lumber?

   Batts aren't designed to fit trusses well. With an I-joist type truss using batts designed for steel framing works OK, but batts sized for dimensional lumber rafters/joist spacings will be too shy on width. If it's open webbing the voids need to be filled with something otherwise it becomes a thermal bypass of the insulation.

   R38 may meet current NC code, but IRC 2012 and later code for US climate zone 4A would be R49, or U0.026 (= R38.5 "whole assembly, after thermal bridging, with credit for air films, wallboard & roof sheathing, etc.).

   Some low density R38s are only good for R37 when compressed to 11.25" in a 2 x 12 bay. Even if your trusses are a full 12.0" there would only be 11.0" after the mandatory 1" air space created by the baffles, so a low-density R38 just isn't going to cut it, even for meeting the letter of NC code.

   A rock wool R30 is 7.25" of loft, a rock wool R15 is 3.5", which when stacked adds up to R45 @ 10.75" of thickness. That's a lot closer to R49.

   With a continuous layer of 1" foil faced polyiso on the underside of the trusses, long-screwing/nailing the wallboard through the foam would then get you to R51, meeting IRC code on an R-value basis. If it's an 11.25" deep truss the R45 rock wool compresses against the baffle to about R43 ag, and with R6 polyiso it still meets code. If it's a true 12.0" it would work with 3/4" polyiso. Foil faced foam is easy to reliably air seal using foil tapes.

   Yes, foil facers are a true vapor barrier, but it's the ease of air sealing we're after not vapor retardency, and the assembly is relying on soffit to ridge ventilation to keep the roof deck's moisture under control.

2. GBA Editor
   Martin Holladay | Nov 13, 2018 05:41am | #2

   B. Weaver,
   Dana's answer assumed that you have some kind of trusses. But you described these "trusses" as 2x12 -- so I'm guessing they may be rafters, not trusses.

   In any case, you don't need an interior vapor barrier. As with all cathedral ceilings, you have to pay attention to airtightness. (Air leakage causes far more problems than vapor diffusion.)

   For more information, see "How to Build an Insulated Cathedral Ceiling."

3. Expert Member
   Dana Dorsett | Nov 13, 2018 07:54am | #3

   If it's just a 2x12, not a truss, a high density "cathedral ceiling" R38C fiberglass batt gets you to R38 in the cavity at 10.25", with the necessary inch of air space. But that's still R11 shy of the R49 prescribed IRC 2012 or later codes. A layer of 1.5" foil faced polyiso would be enough to get you there on a perfromance basis, despite adding up to "only" R47 in that stackup. That takes up anogther 3/4" or so out of the headroom from the rock wool solution in my original post here, and is a little bit harder to build, long nailing the ceiling gypsum to the rafters through 1.5" of foam rather than 3/4"- 1" of foam.

4. 
   Jon_R | Nov 13, 2018 09:11am | #4

   Lstiburek: "No interior attic assembly side vapor control is required or recommended in climate zones other than Climate Zones 6 or higher..."