Insulating an attic ceiling
We're researching the best way to insulate our circa 1900 home, which currently has no insulation except for a thin layer of blown-in cellulose between the second floor ceiling and the attic. There is also a crawl space above much of the attic. We'd like to renovate the attic and turn it into living quarters. We've been considering an expanding foam product called Icynene. Is the need for a vapor barrier eliminated? And do we need some type of passageway in the rafters above the attic to ensure proper ventilation?
We're wading into what has been a contentious area of building science, but I think you can skip the vapor barrier as well as the air vents if you go with Icynene. As you suggest, the conventional approach when insulating a roof has been to install vents at the soffit and at the ridge. Baffles installed against the roof sheathing create a passageway for air that carries moisture away.
This detail is standard when fiberglass batts are used between the rafters. You can buy the molded foam baffles, plus ridge and soffit vents, at just about any building supply dealer.
When the ceiling between the attic and the rest of the house is insulated -- either with batts or blown-in fiberglass or cellulose -- baffles are used only at the top of the outside walls where the insulation might otherwise block the flow of air. In this case, the attic is not a "conditioned space."
Foam stops air leaks
Spray-in foam insulation has changed the rules. Icynene is the trade name for one kind of low-density foam. It has an R-value of about 3.6 per inch, which is slightly better than fiberglass or cellulose. These foams expand instantly on application to 100 times their liquid volume (making this the only kind of insulation job I know of that's actually fun to watch).
High density urethane foam has an R-value of about 6.5 per inch and doesn't expand as much. Both types are better at stopping air leaks than either blown-in cellulose or fiberglass, and far superior to conventional batt insulation in that respect. They're great at filling every nook and cranny to create a monolithic air and thermal barrier, and for that reason I think they're a much better choice than batt insulation.
Closed-cell and open-cell foams have different permeability to moisture -- a fancy way of saying they allow the passage of water vapor at different rates. We want to keep moisture away from roof decks and wall sheathing. But research has shown that the most significant carrier of moisture is bulk air leakage, not vapor permeability.
Skip the vents
Insulation contractors often spray foam directly on the underside of the roof deck, without installing vents or baffles. The idea is that if air can't move through the insulation, moisture isn't going to be much of a problem either.
Icynene spells out exactly such a scenario in a case study on its web site (www.icynene.com). Fiberglass batts with an R-value of 50 were replaced with 5 1/2 in. of Icynene with an R-value of only 20. But the company says energy costs declined by 24%.
It's still a good idea to eliminate all unnecessary moisture in the house -- make sure the dryer is vented to the outside, that each bathroom has a fan, that you have a vent hood over the range and that your basement isn't wet.
Spray-in foam will cost you more up front. And it must be applied by a professional. I don't pretend to be a building scientist, but I think either Icynene or one of its competitors will do a superior job -- without vents and without a vapor barrier. If you want to read about this in more detail, one place to check is the web site for a U.S. Department of Energy program called Energy Efficiency and Renewable Energy (www.eere.energy.gov). Good reading.
An accomplished woodworker and carpenter, Scott Gibson is the former editor of Fine Woodworking magazine, and a former editor at Today's Homeowner and Fine Homebuilding magazines. He also is former managing editor of the Kennebec Journal, a daily newspaper in Maine.