Replies

1. GBA Editor
   Martin Holladay | Oct 15, 2018 06:29am | #1

   User 7196488,
   First of all, can you tell us your name? (I'm Martin.)

   Q. "I was thinking fiberglass or similar between rafters, with foam board on exterior of roof. My question is what should be the layers from drywall to shingles to properly insulate?"

   Briefly:
   1. Drywall installed in an airtight manner.
   2. Fiberglass-filled rafter bays.
   3. Roof sheathing installed in an airtight manner.
   4. An adequately thick layer of rigid foam insulation.
   5. A second layer of roof sheathing (OSB or plywood).
   6. Roofing underlayment.
   7. Shingles.

   For more information, see this article: "How to Install Rigid Foam On Top of Roof Sheathing."

1. 
   Csmaine | Oct 15, 2018 03:01pm | #2

   Thanks Martin,

   Sorry, name is Chris.

   I was thinking 5 1/2 inches of polyiso on top of roof. With r-30 rock wool batts between rafters.

   Does there need to be any sort of barrier between drywall and insulation batts?

   Should there be anything on first layer of sheathing before polyiso goes on?

   My wall sheathing has same cold sheathing issues. I intend to add exterior foam there later. What is best way to transition from hot roof to cold wall? Or should I really do walls at same time?

   I’m not a fan of spray foam, but how do you feel a roof done this way compares to it?

   Thanks, Chris

2. Expert Member
   Dana Dorsett | Oct 15, 2018 03:37pm | #3

   What part of Maine?

   At 5.5" derated for temperature you're looking at R28-ish performance on the polyiso, which isn't quite enough for dew point control in the US climate zone 6 parts of Maine (which is most of the state)- it really wants to be at least 50% or more, expecially near the colder half of zone 6. A simpler solution would be to go with R25 fiberglass batts for the 2x8s, which performs at R24 when compressed into a 7.25" deep rafter bay. That way you'd have the derated R28 on the exterior, R24 in the rafter bays for R52 total, and ~54% of the total.

   For the northernmost counties which are in zone 7 it's quite a bit shy of the ~60% exterior-R prescriptive. It'll still work in zone 7 if using vapor barrier latex and air tight drywall or a "smart" vapor retarder on the interior. Bumping the exterior foam to 6.5" (R33-ish) with R25 batts (=R24) in the rafters bays would get you to ~58%, which is close enough for the warm edge of zone 7.

1. 
   Csmaine | Oct 15, 2018 04:06pm | #4

   Thanks, Dana

   I’m in southern Maine, I would like to do this right so if it needs more polyiso so be it.

   Is this the best way to insulate a 2x8 rafter cathedral, or is there a better route?

   Thanks, Chris

3. 
   Klucas06 | Oct 15, 2018 10:33pm | #5

   Without trying to hijack the OP thread can someone answer one of the questions asked in post # 2 and maybe explain why if it does to help educate. I'm nearing the same point in my construction.

   "Should there be anything on first layer of sheathing before polyiso goes on?"

   Thanks

4. GBA Editor
   Martin Holladay | Oct 16, 2018 05:22am | #6

   Chris,
   Q. "Does there need to be any sort of barrier between drywall and insulation batts?"

   A. No. This type of assembly is designed to dry inward, so you don't want interior polyethylene. However, the drywall should be installed in an airtight manner.

   Q. "Should there be anything on first layer of sheathing before polyiso goes on?"

   A. You need an air barrier at this location. The usual approach is to tape the seams of the roof sheathing. (Zip sheathing works nicely, but ordinary sheathing can also be taped.) If you have sheathing that doesn't lend itself to being taped, you can install a peel-and-stick product like Grace Ice & Water Shield.

   Q. "My wall sheathing has same cold sheathing issues. I intend to add exterior foam there later. What is best way to transition from hot roof to cold wall? Or should I really do walls at same time?"

   A. The only "transition" issue between walls and roofs are the usual ones -- air barrier continuity and insulation layer continuity. These continuity issues have nothing to do with whether the walls or roof are "cold" or "hot." You don't want gaps in your air barrier, and you don't want discontinuities in your insulation layer.

   Q. "I’m not a fan of spray foam, but how do you feel a roof done this way compares to it?"

   A. Using an adequately thick layer of rigid foam is better than installing spray foam between the rafters, because the rigid foam addresses thermal bridging through the rafters.

5. GBA Editor
   Martin Holladay | Oct 16, 2018 05:23am | #7

   Kevin,
   Q. "Should there be anything on first layer of sheathing before polyiso goes on?"

   A. See my answer to Chris: "You need an air barrier at this location. The usual approach is to tape the seams of the roof sheathing. (Zip sheathing works nicely, but ordinary sheathing can also be taped.) If you have sheathing that doesn't lend itself to being taped, you can install a peel-and-stick product like Grace Ice & Water Shield."

1. 
   Csmaine | Oct 16, 2018 04:58pm | #8

   Thanks Martin,

   That was a very helpful and detailed answer.

   I think this is my last question. Should I use XPS instead of polyiso? Last night I read a article you did on that topic. I never knew insulation r-value changed with temp. So it seems XPS is better for my climate.

   Thanks, Chris

1. Expert Member
   Dana Dorsett | Oct 16, 2018 06:30pm | #9

   In a 50/50 roof stackup XPS is not better for your climate than 2lb roofing polyiso inch for inch, and significantly worse for the planet. XPS loses performance over time, and is only warranteed to perform at R4.5/inch (and may drop as low as R4.2/inch over the lifecyle of your house.)

   Brand-X 2lb roofing polyiso will perform at about R4.5/inch when the mean temperature through the foam is +25F, and it'll be significantly higher when warmer.

   When it's -2F, Lewiston's 99th percentile temperature bin, and +70F indoors, with a ~50/50 split and 6" of polyiso ( derating the foam to R5 for average temperature) and R30 in the cavity, the warm side of the foam is about 34F, the cold side -2F with a mean temp through the foam of about +16F. At that point the cheap polyiso is only performing at about R26- sounds bad, right?

   But the mean outdoor temperature in Lewiston even in January is about +20F, so with a 50/50 split at a 70F interior the warm side of the foam is 45F, cold side 20F, with a mean temp through the foam of+33F, a temperature at which it's performing at about R4.7/inch, which is about what age-derated XPS would be performing at. That's in JANUARY, the coldest month of the heating season. The rest of the heating season the polyiso would be outperforming it.

   From a dew point/interior moisture drive protection point of view it's the average performance over the whole winter that matters, not what happens during the 1% or even the 15% coldest hours of the year. (January is only 8% of the hours in a year.)

   For a visual on this, take a look at Figure 1, p1 of this bit o' marketing fluff from DOW:

   http://msdssearch.dow.com/PublishedLiteratureDOWCOM/dh_098a/0901b8038098a7e2.pdf?filepath=styrofoam/pdfs/noreg/179-00379.pdf&fromPage=GetDoc

   You can see that FRESH XPS (the orange line) will outperform the world's crummiest polyiso sample, but over time XPS performance sinks to about performance of rock wool (the gray line).

   In a wall stackup it's usually less than a 50/50 split between foam/fiber, so for thinner wall sheathing derating 2lb polyiso to R4.5/inch might be necessary for getting the minimum thickness protection on the walls. For a 2x6/R20 wall it takes R11.25 minimum (about 36% of the total) on the exterior, for 2x6/R23 (rock wool batts or high density fiberglass) it takes R13 Derated to R5/inch that would be 2.25"-2.6", but it's really performing worse than that since the mean temp through the foam at a 36/64 split is lower that is in the 50/50 on the roof. For R20 cavity fill 2.5" would deliver ~R11.25, but for R23 cavity fill it's prudent to bump that to 3".

1. 
   Csmaine | Oct 16, 2018 07:01pm | #10

   Thanks Dana,

   That was very detailed and helpful. I never knew there was so much to insulating. Obviously whoever insulated my house first time didn’t know either.
   Thanks again for all the great help.