Condensation or roof leak in low-slope, spray foamed, TPO covered roof in cold climate?
I’m a contractor in Vermont typically involved in high performance new construction and renovations along the lines of the Pretty Good House model. I recently got a call from someone with a fairly new house, less than ten years old, about what they think is either a roof leak or possibly condensation. The house is a two story, 2×6 walls, 2×12 almost flat roof with TPO membrane roof, closed cell spray foam insulated, HRV but only on the first floor, bath fans ( one per floor) what appears to be well built house. The homeowner discovered some ceiling water spots on the second floor. Upon my first site visit, on a 20 degree sunny day in winter, there was significant condensation on the windows in the second floor room where the water damage was appearing. Upon going up on the roof, which wasn’t totally cleared of snow and ice, I discovered some soggy feeling roof sheathing under the roofing membrane in a few areas. One area being directly in the area that corresponded with the noticeable water damage below and another in the area where the upstairs bath vent exits the building. The other thing I noticed is that the primary “vein” of water damage was located directly in line and below where the plumbing vent stack exits the roof. All the above makes me think a number of things:
– The bath fan exhaust is condensing on the underside of the roof sheathing as it exits through the soffit creating that area of compromised sheathing.
-There may just be an undiscovered leak somewhere.
– More balanced ventilation is needed on the second floor. A perfect spot for some Lunos as the area is probably only 500 square feet.
-The 1/4″ per foot minimal roof pitch in combination with the vent stack is allowing water to back up into a less than perfect seam of the roofing somewhere.
– There is a compromised area in the spray foam, in the vicinity of the plumbing vent stack that is allowing warm moist air up into the boot of the stack where the condensation collects and then drains down on top of the sheathing. The boot looks good from what I can see.
It could be any combination of the above factors too. I haven’t dug into the ceiling yet and removed any drywall to get a better look yet. The idea of roof sheathing encountering any amount of moisture as its sandwiched between spray foam and TPO rubber doesn’t get me excited at all.
Thoughts?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
WL, as you say, any of the items you list could be factors. Condensation on windows on a 20°F day is not much of a smoking gun; 70°F air at 40% RH will condense on a 45°F surface. Can you get accurate relative humidity and temperature readings?
Bulk water usually causes more damage, more quickly than condensation issues, so I would start by looking carefully at any seams or penetrations in the roofing membrane. After that, I would look at the closed cell foam--it often pulls away from the framing, allowing a narrow channel for warm air to condense. I would also look at the bath fan duct--is it working properly, sealed an insulated?
The roof leak is my gut call as to the culprit considering that the interior water damage is very localized. I'm sure the bath fan duct work is causing some some problems as well. As for the possible condensation pathway through the foam, I was curious as to the possible volume over time and the fact that there is no escape route once it breaches the failed vapor/air barrier of the interior foam. Another reason I'm not keen on foam unless necessary and also for ventilated roofs.
On another note, Michael, I enjoyed your most recent article. Thanks.
After you mentioned shrinkage I found this video Michael.. Narrow channel? LOL. Frankly, I cant believe it's still recommended and I'm reading shrinkage rates are in the 8-14% range which spoils my plans.
https://www.youtube.com/watch?v=ieLm9NwE22c
Mike, I have become pretty anti-foam except when there is no other good option, but don't believe what you see in that video. It looks to me like UFFI, an injection foam used in the 1970s and early 80s that did have major shrinkage issues. It also becomes brittle, and releases a lot of formaldehyde. Closed-cell foam can shrink a little, but when that happens I usually see 1/16" to 1/8" of separation, not a full inch. If the substrate is clean and dry, the mix is correct and installation temperatures in a good range, it shouldn't shrink at all, but conditions are not always perfect.
I came across a situation similar to this recently as well although the foam hadn't shrunk, it had become incredibly brittle to the point you could practically blow it out of the stud bay cavities with little effort. The folks I was working for mentioned that a renovation had happened in the early 2000's and that was when it had been installed.