Rigid foam & cathedral ceilings
I live in a house built in 1994 by a production builder just outside of Boston, MA (ie: poorly insulated and poorly detailed). Since moving in a couple years ago, I have been thinking that my goal is to eventually up the insulation in the main attic space (framed with engineered trusses) by putting in a lot of blown-in cellulose. There are 3 areas with cathedral ceilings where I expected to put rigid foam on top at the time we re-roof. I started looking in to solar, and the original roof is due for replacement. All of the local roofing contractors I have talked to looked at me like I was crazy when I mentioned rigid foam, except for 1 who quoted me double the price of everyone else just for the shingles (he wanted another $8,500 for the rigid foam on top of that). I mentioned it to an architect we have consulted with, and he also gave me the “you’re crazy” look, recited an advanced framing anecdote/horror-story, and said to just leave it alone, or pull the sheetrock and spray foam.
I’m looking to find a happy medium between improving energy efficiency and being cost-effective… Am I just talking to the wrong roofers, or is adding rigid foam really cost-prohibitive? I’m also happy to take referrals to any contractors who service the Greater Boston area who are comfortable with doing rigid foam.
Thanks,
-Chris
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Hi Christopher,
This question is asked frequently. When contractors are asked to do something they are not familiar with, it is common for them to either refuse or to bid the job excessively high. And that makes sense; they are protecting their business and trying to ensure a profit on the job. Whether it is a smart way to go or not, insulating above a roof is not a common detail. And the truth is, you probably don't want to hire a contractor who has never done it before.
You can look for a general contractor or remodeling contractor with experience with high performance homes and/or energy retrofits to manage the project and make sure it gets done right. However, it may be just as expensive.
The average residential roofer is the wrong type of contractor to be handling roof insulation detail. Most are only competent to install what goes above the top side nailer deck, and I'd be loathe to trust a newbie to do it right, let alone price it correctly. It's not cheap to insulate cathedral ceiling properly to current code spec, no matter which way you go about it.
David Joyce of Synergy out in Lancaster MA has done several deep energy retrofit projects in the Rt 128 region (and elsewhere, including a project in Worcester I consulted on a handful of years ago) and would be able to get the details right on roof foam-over. He's listed on the MassSave contractor list, as are several other deep energy retrofit contractors who should be able to get this right.
https://www.masssave.com/-/media/Files/PDFs/Contractor-Lists/Deep_Energy_Savings_Contractor_List.pdf?la=en&hash=DDF55643A7FC8AA928126E8C1491B492B146F4AC
How deep are the rafters/trusses in the cathedralized section area?
If the rafter bays are vented (they should be), drilling & filling the rafter bays completely with cellulose from the exterior would be necessary to avoid the vent becoming a thermal bypass. In US climate zone 5 a minimum of at least 40% of the total R would then need go on the exterior.
So if they're 2x10s (9.25" deep), a combination of compressed batt and cellulose fill would yield about R34, and there would need to be at least R23 of rigid foam above to keep the roof deck happy & dry. That would take 4"-4.5" of foil-faced polyiso, or 5" of 2lb fiber faced roofing polyiso.
With 2x8 rafters (7.25" deep) it would be about R27 of cellulose + batt, and it only takes R18 above the roof deck for dew point control, but shoot for R20. That would be 3.5" of foil faced or 4" of roofing polyiso, a somewhat easier build.
Whatever the truss depth, multiply the depth-inches x R3.7 as the cavity fill when topping up with cellulose, and multiply the cavity-R number by 0.67 for the absolute minimum exterior foam R (x 0.7 is better.)