R-value at roof/wall intersection
I am located in the Albany, NY area zone 5A.
I purchased a dilapidated house and will be doing a deep energy retrofit. The siding will be torn off, most of it is rotten or soft, and I will install huber zip which should provide the air and vapor barrier. Larsen trusses will be installed and the cavity depth will be 12″, ~R-40.
The roof is 12:12 and peak is roughly 10′ high. New asphalt shingles were supposedly installed approximately three years ago- they appear to be in decent shape. Roof framing is dimensional 2×4’s 24-30″ O.C. There is a shoddily built dormer which will be removed. From exploring around the dormer it looks like there was 1x decking, wooden shakes, 1/2″ plywood, tar paper, then shingles. The exterior walls are balloon framed with a double 2×4(dimensional) top plate. The second floor ceiling joists are approximately 7″ deep.
I have attached a an image with a basic drawing. Brownish color represents framing, greenish is cellulose (didn’t extend it into the ceiling joist like I did for the roof rafter for clarity), orange line air barrier.
My concern is the R-value achievable around the top plate and roof/wall intersection. Once the 2″ ventilation channel is considered, the top outside corner of the ceiling joist has about a 2.75″ cavity that can be used for insulation which ultimately presents moisture (cold spot on the ceiling) and energy issues. However the minimum depth from the top of the double plate when including the joist cavity is 7″ which is acceptable if spray foam is used.
Now since this is a complete rehab project, there is no issue with running a continuous interior soffit to allow for the proper insulation depth. Anyone have any input on whether that would satisfactorily address the issues? It seems like it would reduce, but not necessarily eliminate, the issues.
I have contemplated tearing the whole roof structure off and going with a flat roof (labor to be done by me and not to fond of working on steep roofs) because the dormer has to come off which represents about 1/4 of the roof area. I was originally concerned with 2×4’s 24-30″ O.C, but it seems to have worked for the past 100 years. It does have a pretty new 30 year roof on it. I do like the 12:12 appearance. It would increase the expenses. Ultimately, I am struggling to justify whether the energy problem at the roof/wall intersections merits redoing the whole roof structure.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Upstate,
That's definitely a cold corner. It will always be subject to humidity and mold. You often see moldy drywall ceilings under corners like that -- especially in bathrooms.
The best solution is to install several layers of rigid insulation on top of the roof sheathing. Of course, that means you need new roofing, which is expensive.
Upstate,
You might be interested in an article I wrote on this topic for the October 2008 issue of Energy Design Update. The article was called "Preventing Mold Growth In Cold Ceiling Corners."
Reading it will take patience, because I'm attaching the article as 11 image files. I apologize for the awkward format.
Here are the last 5 image files.
I guess there's no chance the added wall depth could go to the inside?
Is raising the roof and cobbing in an energy heel in a situation like this even remotely feasible?
One more thought - fill the space above the wall plate solid with high-R foam and introduce ventilation further up the roof using these: http://www.cor-a-vent.com/in-vent.cfm. I have no idea though if this might introduce extra problems in snow country - Martin?
I think the issue is more with the thermal bridging than the joist cavity. The joist cavity provides about 9" of depth which spray foam could provide an adequate R-value. It would be a relatively small volume so I think the expense would be tolerable.
Once the flooring and ceiling is finished I will have about 7' 3" ceilings on the second floor which would make an interior insulation approach very difficult.
I am hesitant to insulate the interior of the walls due to cold sheathing concerns, loss of space, and maintaining the air barrier. Also, there is brick in all of the stud cavities which I am leaving exposed on the first floor and unsure about the second.
The retrofitted energy heal is an interesting idea. I have attached another image with an idea. Basically continue the larsen truss up to the depth of the ceiling insulation. The top of the larsen truss could be essentially flat, though slightly sloped for drainage or it could integrate at another angle. Venting can be achieved through channels above the insulation. I do however wonder about aesthetics and of course the building inspector since the larsen truss would now carry a roof load.
Upstate, you could probably make that work, in fact it's similar to an elegant traditional detail called a swept eave. I'd go with a minimum 4/12 pitch for the sweep so you could stay with the same roof finish. Eaves load could be taken back to the original wall plate with a simple strut. Don't know what Martin will think of the brokeback profile, it's similar to one of the no-no's on his roof design article, from concerns about snow retention. I'm told though that the swept eave is a classic traditional detail in Denmark, where snow is hardly unknown.
By the way I don't think you need to go as high as the full depth of the ceiling cellulose. IIRC Dr. Joe suggests you'll be OK with the same R over the plate as you have in the wall. It could look something like this:
I'm not a fan of swept eaves in snow country, because they definitely increase the chance of ice dams, and they interfere with ice clearing from the roof.
But the idea could work. If you do it, don't forget to install a generous band of Ice & Water Shield, or consider roofing the eaves with metal flashing for the first two or three feet before switching to conventional roofing.