Download PDF version (536.4k) Log In or Register to view the full article as a PDF document.

Building Codes

You — and your building code inspectors — may be unaware that the 2006 version of the IRC for one- and two-family dwellings permits attic construction with no ventilation of the attic cavity. This new provision, R806.4, is largely due to the efforts of Joseph Lstiburek, Armin Rudd, and their colleagues. In brief, unvented conditioned attic assemblies are permitted when an air-impermeable insulation such as rigid foam is applied in direct contact to the underside/interior of the structural roof deck, with sufficient thickness — given the climate — to prevent condensation on the underside (see "Insulating Unvented Attics With Spray Foam," 3/07).

This new provision is a direct challenge to the rule of thumb that has been in place for 50 years, which says that you have to vent a steep-roof attic so the ratio of net free vent area to the projected roof area is 1-to-300 (or 1-to-150 when using "cross ventilation" rather than soffit and ridge vents). This ratio arose from observations of frost on protruding nail points in Wisconsin homes by researchers at the Forest Products Laboratory in 1937, and frost on aluminum plates in research "doghouses" at the University of Minnesota in 1938, under "outdoor" conditions of -13°F.

The Federal Housing Authority turned these findings into the famous 1-300 ratio in 1942, to be applied as a minimum building requirement for the small homes in its financing program. The requirements were picked up by model codes and others following World War II, and the rest, as they say, is history. Shingle manufacturers did not begin piggybacking their warranties on venting regulations until reports of shingle problems began piling up following the change in asphalt sources in the early 1980s.

To Vent or Not

Every designer and builder should be able to produce good attic and roof assemblies, both with and without ventilation — or anything in between — with just part of a conventional ventilation system. For example, from our studies, roof assemblies that have holes but not necessarily straight airflow paths (one gable end vent, or soffit-only) should also be candidates for good performance. And although unvented roof assemblies can perform well, there are still good reasons to vent: The truss-framed, steep-roof attic with an insulated ceiling has been the workhorse of single-family construction, and ventilation works well with this construction, at least in the northern United States.

In some cases, there are also good reasons not to vent: in wildfire areas, in complex cathedral ceiling assemblies, in existing and historic buildings that have never had ventilation, in shed roofs beneath clerestory windows, with foam insulation (foam and ventilation do not go together — think fire), and in complex roof assemblies that combine steep and low-slope construction. I've also heard persuasive arguments against venting in hurricane-prone regions, but I'm not an expert in that area. In short, since critical performance doesn't hinge on ventilation, then either vent, no-vent, or an in-between "kinda"-vent can be taken as the starting point. Whether the choice works or not depends mostly on other factors.

So you should vent where venting is appropriate and not vent where it is not appropriate. As it turns out, the worst-performing, most mold-ridden attics I have seen were vented — with a flooded crawlspace and a direct path for air movement from the crawlspace to the attic. You can mess up a vented attic by allowing such airflow. You can mess up an unvented attic as well, usually by not providing vapor protection appropriate to the climate and indoor moisture levels. Tight ceilings would be a great first step toward moisture control, summer and winter.


The father of a colleague of mine says that when the word "ventilation" comes out, people stop using their heads. Vented assemblies often perform well, but not always. Sometimes roofs appear to be vented but actually aren't. Still, we can take comfort in the observation, based on years of experience, that our attic assemblies are pretty darn good, and — in my opinion — they're getting better. We need to constantly be on the lookout for new conditions and new problems, as they crop up.

Those of you working in the trenches should continue to build in a way that complies with code and that you know works for your climate. For more information about ice damming, summer cooling load, shingle service life, and moisture issues, visit (TenWolde and Rose, "Issues Related to Venting of Attics and Cathedral Ceilings"). For all four of these concerns, ventilation makes a contribution that is generally more positive than negative, but it hardly ever makes the difference between success and failure.

For the most part, the focus of codes, researchers, designers, and builders on roof ventilation is misplaced. Instead, the focus should be on building an airtight ceiling, which is far more important than roof ventilation in all climates and all seasons. The major causes of moisture problems in attics and roofs are holes in the ceiling and paths for unwanted airflow from basements and crawlspaces. People should focus first on preventing air and moisture from leaking into the attic. Once this is accomplished, roof ventilation becomes pretty much a nonissue.

William B. Rose is a research architect with the Building Research Council at the University of Illinois at Urbana-Champaign, and the author of Water in Buildings: An Architect's Guide to Moisture and Mold. This article was adapted from The JLC Guide to Moisture Control.