Condensation in roof steel vs. wood vs. foam
In an unconditioned attic space above a residence, most of what I’ve read on GBA seems to agree that steel sheet roofing might develop some condensation on its underside, even if the ceiling below is properly sealed and insulated, just from the outdoor air warming faster than the surface of the metal, and this is part of the argument for installing steel sheet roofing over underlayment and plywood or OSB sheathing. Even in an installation with sheathing, though, it seems likely that there would be times when the air inside the unconditioned attic space would warm faster than the surfaces of the sheathing and framing exposed to it. Does something about the greater R-value of wood (compared to steel) keep condensation from forming, or, more plausible to me, is condensation on the sheathing and wood framing not a concern because the wood in those surfaces would absorb it enough to keep it from dripping, after which it would eventually evaporate back out of the wood?
I ask partly out of curiosity, but also because I’m contemplating an unusual installation: metal sheet roofing over purlins on trusses, but instead of using sheathing underneath, sealing 2′ wide strips of EPS or polyiso between the purlins to catch any drips. I’ve gotten helpful comments on other aspects of my idea in another thread, but here what I’m wondering is this: Would the underside of EPS or polyiso board develop condensation that would drip like metal or not drip like plywood, OSB, or wood?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Jeff,
When condensation forms on the underside of steel roofing, the source of the moisture is usually exterior air, not interior air. During the spring, steel roofing can be cooled by snow or by nighttime radiational cooling. When the weather warms up during the day, outdoor air enters the channels under the metal roofing and moisture condenses when it contacts the cold roofing.
When warm, humid air contacts cold plywood or cold OSB sheathing, the result is usually not condensation. Rather, the plywood or OSB takes on moisture which is adsorbed. (In other words, the moisture content of the plywood or OSB increases.) However, if the plywood or OSB is below freezing, frost can form on the surface of the plywood or OSB.
If warm, humid air contacts cold rigid foam, condensation is certainly possible.
I agree with Martin that warm humid air contacting cold foam could in theory lead to condensation, but in the scenario he describes (snow on the roof), the insulation would prevent significant condensation. The foam also has much lower thermal mass than plywood or steel, so it wouldn't stay cold as the weather warms. So I doubt you'd ever have a problem with condensation on rigid foam.
Are you using the insulation for r value or just to catch water? Seems like a fussy idea if only to catch water. I haven't personally used it, but Fabral makes a metal roofing with a coating called Condenstop. Maybe someone else can comment on its effectiveness. Lots of pole barn builders just install tyvek under metal roofing when using the truss and purlin method.
I have steel on purlins over rafters on one building (and over OSB and plywood decks on two other buildings) on the edge of climate Z 5 and 6. Very heavy dew and ground fog south side of Lake Ontario. The quantity of bulk water due to condensation is not enough to pour off the steel to the space below. Instead, it runs the length of the steel and drips off at the eaves.
Thank you, all, for helpful replies.
Rick: The foam would just catch water, but I have an increasing sense that in my dry 2B climate zone it's probably an overabundance of caution.