Attic insulation depth above wall is less than wall insulation
I live in Madison, WI (cold climate). I have an old ranch home with 2×6 rafters with a deep soffit overhang (2ft). The rafters are notched at the top plates and I have about 2 inches of free space between the top plate and the plywood roof sheathing (assuming I keep an inch free for ventilation). The walls are 2×4 with plywood and Homasote sheathing.
I have read here and elsewhere (i.e., buildingscience.com) that I should not insulate the wall more than the depth of the attic insulation. I’ve got to pull off a of siding this summer to replace with fiber cement and was thinking of adding 2″ of XPS foam, boosting the wall value to R13 batt + R10 XPS + Homasote value (maybe 1). I’d end up with a wall at R23 and an attic right above it at R13ish at best (assuming closed cell insulation). Rest of attic is combination of loose fiberglass and fiberglass batt, R30 at best.
My question is, why would this be a bad thing? I’d air seal the heck out of things (airtight drywall) and if there is 2x more heat transfer through the attic than the wall, what is so bad? Wouldn’t the heat transfer through the attic be unchanged regardless of the heat transfer through the wall? Or do I need a course in thermodynamics? Ice dams are a plague in our climate, so I don’t want to screw things up, but would like to boost the wall’s insulation value. But that contradicts most guidance I’ve seen on the subject.
Am I stuck with R-13 walls forever without reframing my attic rafters?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
John,
The biggest error made by the builders of your house was to frame the roof in such a way that there isn't much room for insulation at the perimeter of your attic. This problem could have been solved by attaching the rafters to a plate nailed to the tops of the attic floor joists, or by using raised-heel trusses.
It could be argued that the second-biggest error was to build poorly insulated walls. But the basic problem that Joe Lstiburek is railing against (for example, in his list of Lstiburek’s Rules for Venting Roofs ) is insufficient attic insulation.
His rule of thumb -- insulate the perimeter of your attic with AT LEAST the same amount of insulation that you have in your wall -- is a guideline to get builders to please, please remember to include insulation above the top plate of the wall.
His intention was not to limit your energy retrofit project to improve your wall insulation. Feel free to improve the R-value of your walls.
That said, you have a real dilemma with your attic insulation at the perimeter of your attic. Your roof is a candidate for ice dams. The best you can do with the current situation is to use closed-cell spray polyurethane foam above your wall top plates. Two inches of spray foam gives you only R-13 insulation above your top plates. That's not much.
The long-term solution is to convert your attic into an unvented, conditioned attic by installing at least two layers of thick rigid foam insulation on top of your roof sheathing, followed by a layer of plywood or OSB and new roofing. That job will probably have to wait until you need a new roof.
If you have ice dams next winter, and you spend a lot of time chopping the ice off your eaves with an ax, you may do so much damage to your roofing that it will soon be time for a new roof anyway.
Notched rafters with insufficient room over the top plates for attic insulation are real candidates for converting to sealed-conditioned attics, but it's not cheap- it would likely take 6" or more of exterior rigid foam to get you out of ice-damming danger due to the localized heat leak at the tops of the walls in a WI climate. But starting with 2" of closed cell spray foam on the underside of the roof deck and sealing the attic venting may be a good start, even if it's only R13.