Attic insulation: how much of a concern are uncapped wall cavities?
bencarsan
| Posted in Energy Efficiency and Durability on
On a retrofit of a leaky balloon-framed building in upstate NY, we found that most walls, both interior and exterior, did not have top plates and were open to the attic. All had fiberglass insulation in them already. Assuming that these walls are tight and aren’t exfiltrating, I am wondering whether heat loss through convective air movement in this situation is likely to be significant. In some areas access is very difficult, so I’d like to have a sense of how much benefit there is in sealing these cavities.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Bennett,
Fiberglass batts do little to limit air movement, so ideally all of these walls would be capped with an air barrier material.
You mention that the interior partitions have fiberglass insulation in them. That's unusual. Or do you mean that the interior partitions have attic batts laid on top of the open stud cavities? If the latter is true, that would be a huge air leak.
Even under the hypothesis that there is perfect air sealing between the house and the wall cavities, which is very unlikely in an old house, air can leak in from the exterior or basement into the exterior walls, and convect up through the fiberglass, into the attic. It starts as cold exterior air, sweeps through the wall picking up heat, and then carries that heat into the attic. Even poorly sealed top plates are a problem--wide open tops with no plate are a serious problem, especially with fiberglass insulation. The interior walls are a problem as well, as there are opportunities for basement or interior air to get into them. And the tops of the interior walls are easier to close off, so that's certainly worth fixing.
For energy purposes, a membrane taped over the top is a fine solution, but wood blocking would provide a fire blocking function as well.
Thanks Martin. Right, interior partition walls are insulated with fiberglass for some reason, not just FG overlaid on top. And I think communication is not an issue here--these are second story walls with a plate between floors, first floor walls are dense packed, there is good sheetrock, and no dust signatures in the FG to indicate air leakage.
It's an odd scenario. Open cavities ring alarm bells, but our energy modeling software software can't help with the question of how much benefit we might buy in sealing these cavities.
Bennett,
Obviously there are a lot of variables -- so software isn't much use in determining the rate of air leakage. Depending on how old the fiberglass batts are, they can be helpful to indicate air leakage (if they are dusty) -- but if the batts are only a few years old, they may not show much dust. But you are correct that clean batts usually indicate less air leakage.
If you are doing the work for a client, the energy savings may be too low to justify the air sealing labor. If it's your own house, though, and you are like a typical GBA reader, a good attic air barrier may be important to you -- so important that you are willing to crawl around your attic for many hours, sealing things up, so that you can sleep better at night.
Not all batts or batt installations are created equal.
Low density R11 aren't nearly as air retardent as R13s.
R15 are QUITE air retardent, comparable to cellulose.
Batts of R13-ish density or higher that are fitted perfectly in the cavites with few compressions or voids can still have reasonable thermal performance without top plate on the balloon framing, but perfection isn't likely even with standard milled studs 16" o.c., and downright impossible with batts for full-dimension 2x4s, or non-standard stud spacing.
I'd bet real money that
A: the batts aren't R15s and
B: the installation is far from perfect.
Stud walls without top plates are an air leaky disaster, but the magnitude of the air leakage can vary by over an order of magnitude- it's a "known unknown". With blower door testing it's possible to at least get the order of magnitude right, and with blower door + IR imaging guided air sealing you can chase down the bigger leaks pretty quickly. If experience is any guide, it's "worth it" to air-seal the tops of all studwalls (exterior & partition) even before running a blower door test.