Partial retrofit?
![](https://secure.gravatar.com/avatar/742e573339ee53d84c15b034085c2616?s=52&d=https%3A%2F%2Fimages.greenbuildingadvisor.com%2Fapp%2Favatars%2F200x200%2Fj.png&r=g)
First, thanks to everyone for all of the advice and knowledge I’ve received here. This site and the forums are excellent.
I am the homeowner – knowledgeable, some prior construction experience, but not expert.
Climate is Southeastern PA – mixed humid. House is two-story Cape Cod. The existing rafters of the home are 2×6, vented from soffit to ridge with baffles and fiberglass batts. The shortcomings of this are understood regarding the lack of R-value, but the roof performs OK. Also, there has been attention paid to things such as air sealing, no canned lights in rafter bays, all HVAC in conditioned space, etc.
Situation: renovation underway. GC and I identified that we should improve roof sheathing on one entire section of an existing roof over a shed dormer (4/12 slope). A roof shingle replacement is already in the project scope.
Since replacing the shingles and sheathing is in play, we are considering overdecking with polyiso vs. replacing the sheathing and then using spray foam in rafter cavities or just replacing the sheathing and continuing with batts/baffles approach. All scenarios under consideration are suboptimal, and that bothers me, but I can only proceed with an idea if it can be implemented correctly at a reasonable cost with the team already on the job.
My specific question is whether it would be advisable for our general approach to be closed cell foam in the rafter bays in just that one section of the house where we have access during a renovation, effectively making just that roof/ceiling section “unvented”, while the rest of the ceiling/roof assembly for the other parts of the home remain vented with batts/baffles. I have concerns about the effectiveness of a “partial” job, moisture migrating from one area to another, etc.
I understand that many on this forum prefer the approach of installing polyiso over sheathing in this type of situation. In theory that approach makes more sense to me, too, but in the present situation, when it wasn’t designed up front, and the trades already engaged don’t typically do this, the ease of spraying vs. overdecking have me leaning toward a spray foam approach or just sticking with batts/baffles and instead looking for other ways to make improvements in the overall building envelope.
Perspectives on the choices before me would be appreciated.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb1.png)
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb2.png)
Replies
J.,
There is no reason that you can't design the shed roof over your dormer to be an unvented roof assembly, even if the rest of the roof remains vented.
As you guessed, I would advise you to install polyiso (or some other type of rigid foam) above the roof sheathing if you possibly can, since this approach addresses thermal bridging through the rafters and has the potential to providing enough R-value to meet minimum code requirements.
For more information, see How to Build an Insulated Cathedral Ceiling.
I'll second Martin's reply; no reason at all why you couldn't insulated this part differently from the other parts. As far as unskilled labor, they need to butt the sheets, tape the joints, and either screw sheathing through the foam to the rafters or over strapping - but none of these things are difficult or out of range of normal carpentry skills. I'd avoid fiberglass batts at all costs.
At a minimum, spray foam the roof from the inside, and fill the remaining cavity with dense packed cellulose. I'd avoid fiberglass batts at all costs.