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

What are your recomendations regarding rock wool, dead air space, reflective polyisocyanurate and vapor and air barriers?

goodnatured | Posted in Energy Efficiency and Durability on

May be trying to educate myself and reading too much. First off I am in zone 6, Portland Maine. The outer roof consists of OSB, ice and water shield and asphalt shingles. No attic space, a complete living area. The wall outer layer is cedar shingle.

A portion of the roof (12’x26′) is low slope roof (4/12) 2×10, have installed rock wool to an R38. The rest of the roof is 9/12 pitch 2×6, rock wool to R23. The 2×4 walls have rock wool to R15.

I really want to use foil faced (both sides) polyisocyanurate also on the interior. The final system must include the rock wool that is already installed.

I am trying to get a radiant barrier mostly to prevent heat transfer into the home during the summer, hence rigid polyiso-, and dead air space. OK, my plan, from the rock wool, 6 mil polyethylene sheeting, 1″ dead air, 1 1/2 polyiso, 3/4 dead air, then 5/8 gypsum.

My concerns are the polyethylene sheeting causing trouble with exterior drying. But isn’t the foil face of polyiso an air/moisture barrier? Is using both just too redundant? Am I really accomplishing anything with two dead air spaces to insure a radiant barrier?

A second idea is polyiso agaist the rock wool, then dead air space followed by the gypsum.

I think my biggest concern is exterior drying. Also on the low slope roof, it can’t be vented so I am trying to keep it as cool as possible for winter snow, ice damming. Should I just simplify and install the gypsum right up against the rock wool?

Lots of complexities and questions. Suggestions ?


  1. GBA Editor
    Martin Holladay | | #1

    Greetings again. We've been down this road before, haven't we? (I never forget an Armenian name.) The last time you wanted to build an unvented roof using air-permeable insulation, I advised against it. (Remember the Q&A Spotlight that Scott wrote about your roof insulation plans in 2011?)

    So, at the risk of repeating myself: you can't use an air-permeable insulation like rock wool to insulate an unvented roof assembly. For more information on your options, read these two articles:

    How to Build an Insulated Cathedral Ceiling

    Insulating Low-Slope Residential Roofs

    If you install a radiant barrier adjacent to an air space, the R-value of the air space will be in the range of R-1 or R-2. That isn't very much, so I think you are worrying too much about radiant barriers. For more information on radiant barriers, see Radiant Barriers: A Solution in Search of a Problem.

  2. Expert Member
    Dana Dorsett | | #2

    What he said. And more...

    With out a vented roof deck, with Ice & Water shield on the exterior of the roof (or even #30 felt + shingles) there is no such thing as exterior drying. The I & W is 0.05 perms max, a felt and shingle stackup is about twice that, or ~0.1 perms. To have any chance at all of drying it needs to be more than 0.5 perms, but that's ignoring the fact that during at least 30% of all annual hours the surface of the shingles are wet with liquid water from snow/ice/rain/dew, which would prevent drying toward the exterior at any vapor permeance of those materials.

    Installing either polyethylene sheeting OR foil faced goods in an unvented assembly with Ice& Water Shield would create a severe moisture trap, whether there's an (unvented) air gap between the foil & fiber or not. Either of your solutions would all but guarantee moisture build up in an rot in the roof deck, probably sooner than later.

    To not have moisture problems there needs to be a vented air space of at least 1.5" (2" would be better) between the rock wool and roof deck, with venting at both the eaves and at the ridge. A 4:12 pitch is sufficient for this venting to work (2:12 would be pushing your luck.) With the vented space above you can go ahead and use foil faced iso under the rafters to reduce thermal bridging, making a very solid interior side air-and-vapor barrier.

    The value of the radiant barrier effect is near-zero at the delta-Ts your talking. I works great in outer space with nothing but ultra low density hydrogen between the RB and the 5800K sun on one side, and the 3K deep space on the other, but when you have adjacent layers in a stackup like that with delta-Ts of less than 2 degrees it's pretty useless. At temperature differences low enough that the convective transfer rates are low it works great, but it's a "who cares?" kind of deal, and at temperature deltas that matter convective heat transfer starts to dominate. The best you're EVER going to do in a perfectly horizontal application is about R8-R10 equivalent, with an air gap on both sides of the foil, along with foil facers on the far sides of the gaps. And in a heating dominated climate like Portland ME there's more value in filling those gaps with real insulation to reap the wintertime performance, since in a cold-side up scenario the convective heat transfer is ~100x that of a warm-side up stackup. Average annualized performance of the foil facers would probably be R2 or maybe a bit less, as Martin suggests.

  3. albertrooks | | #3

    Martin & Dana,

    This would be a good application for about 100mm (4") of Agepan THD woodfibre insulation board. The 100mm would be 10 perms and up. The thermal lag of woodfibre boards would meet his summer over heating goals. Heck, even a 52mm (2") layer would do it. That would place the permeability at 18 US perms.

    A decent membrane (or sheet goods) to control air and vapor on the warm side + framed cavities w/ MW + Agepan THD 52 in one or two laters (air sealed) + high perm roof membrane + battens for gap + roof underlayment + roofing would give him a thermally broken, unvented, drying to the exterior, underlayment vented assembly that would have enough thermal mass to increase thermal lag and significantly slow down summer over heating.

    Here is the board:

    And oh yeah... No foam or foil. :)

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |