Vented cathedral ceiling assembly
Location: Tacoma, Washington (climate zone 4c)
I am planning to convert an unvented cathedral ceiling to a vented assembly by adding coravent raftavent at the wall rafter blocking and continuous ridge vent. See attached sketch.
Primary insulation is intended to come from 2 layers of staggered polyiso with great stuff spray foam at perimeter gaps. Below this, I will fill with un-faced fiberglass batt. This will allow me to run misc electrical wiring below the rigid layers and not worry about serious thermal gaps.
Question 1a: Costs aside, Is polyiso going to be more efficient than XPS or EPS?
Question 1b: assuming I use polyiso, is foil-faced or paper-faced preferred?
Question 2a: instead of 2x blocking to create the 1 1/2″ vent depth, is it better to make this blocking out of ripped strips of polyiso?
Question 2b: to avoid screws/nails, can I use glue to bond the foam boards? What kind of glues are best?
Thanks a lot, Ben
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Ben,
In your climate zone, the 2012 IRC requires a minimum of R-49 roof insulation. Your proposed roof assembly is at about R-37.
You only need one layer of rigid foam to create the ventilation baffle. If you moved the second layer of rigid foam -- making it a continuous layer on the interior side of your rafters -- the performance of the roof assembly would be much improved.
For more information, see Site-Built Ventilation Baffles for Roofs.
In theory, it's better if your ventilation baffle is relatively vapor-permeable. That would dictate one layer of rigid foam, not two layers, in that location, and it would dictate a preference for paper-faced polyiso over foil-faced polyiso. As my article explains, however, in practice the vapor permeance of the ventilation baffle it may not make much difference, especially if your drywall layer is relatively airtight.
In zone 4C you only need at 20% of the total R to be on the exterior side of fiber insulation for this to work without interior vapor retarders. If you used a single layer of 2" polyiso, derated for temp you'd be looking at R11 for the polyiso, and it would leave enough room for a 5.5 batt, which would be at most about R23, for a total of R34, so you would have PLENTY of dew point control at the foam/fiber boundary with just 2" of polyiso.
But it's still well below code-minimum for total R value, and well above the code-max U-factor.
If you cut some of the 2" polyiso into 1.5" wide strips and and glued them to the stud edges, then installed 1x furring onto the bottom of the edge strips, through-screwed or ring-shank nailed to the rafters you'd have room for 2.75" more fiber, or 8.25" total. If you took an R30 batt and compressed it into that space it would perform at about R27-R28. for about R39 total. You would still have plenty of dew point control. Being on the warm side of the assembly the edge strips would perform at their full-rated R, or about R12-R13, so the ~7% framing fraction would now be about R20, more than doubling, and with the R-value of the air films, roof deck, and ceiling gypsum added in it would come pretty close to meeting code on a U-factor basis.
Thank you for the response D Dorsett! Good to know about the 20% rule related to vapor retarders. I do like your idea for adding to the bottom of the rafters. Seems a little scary sandwiching the middle batt insulation between upper and lower layers of rigid though? Adding the bottom continuous rigid seems great thermally - but this makes any penetrations for lighting, electrical conduits much less forgiving.
Regarding my original question 2a: seems like you think creating the 2x blocking out of polyiso makes sense? Regarding question 2b: do you have any recommendations for fastening/adhering the polyiso?
Thanks! Ben
Thanks for the quick response Martin! This is a remodel scenario so my target is R38.
I'm a but confused because what you're calling my vent baffle is even more importantly acting as my primary thermal insulation. That is why I'm thinking 2 layers of rigid foam is better than one so I can stagger the seams to minimize the potential for air and thermal leak. Does this make sense? I realize moving the lower rigid foam layer to below the rafters eliminated the thermal bridge of the rafters - but this seems to create a scary foam sandwich with batt in the middle?
MOST IMPORTANTLY, how important is it to vent a cathedral ceiling? To buy more insulation, what if I eliminate the venting and push two layers of 2" rigid foam tight to underside of sheathing which will then buy me 5 1/2" of batt insulation, bringing me close to R45. Would this perform better than having a vented roof with R38? (it would save me the cost of having to add a ridge vent!)
Ben,
Venting your cathedral ceiling is important, unless you want to use closed-cell spray foam.
Your suggestion to use rigid foam insulation up against the sheathing, with no ventilation gap, is called the "cut-and-cobble" method. When used for unvented cathedral ceilings, this is a discredited insulation method that had led to failures (condensation and sheathing rot). For more information on the method as well as the failures, see Cut-and-Cobble Insulation.
The "sandwich" of rigid foam that worries you (a) isn't as dangerous as you think (as long as there is a ventilation channel to carry away incidental moisture), and (b) is in any case avoidable if it bothers you. (Just use a cardboard baffle instead of a rigid foam baffle, and move all of your rigid foam to the interior. It's always better to install rigid foam as a continuous layer of insulation, rather than to cut it up into narrow rectangles and insert it between the rafters.)
If you're putting foam on the interior side, you MUST vent the roof deck. The 20% rule for your climate zone is for foam on the exterior side, and a class-III (or at the tightest, a class-II vapor retarder) on the interior.
I was not talking about making a foam sandwich- the interior foam is on the rafter edges only, not covering the fiber. If the interior-side foam is only on the rafter edges, not covering the fiber layer, it doesn't increase the risk, as long as the cavity foam/fiber ratio is adequate. This presumes you are using foam as the baffle for a VENTED roof deck assembly.