Open cell foam help
Was this open-cell spray foam install correct? (Zone 4A – Metal Roof, Closed Attic)
I built my house 9 years ago (Zone 4A) with a 30-year metal roof and a closed, unvented attic insulated with 5.5″ of open-cell spray foam directly under the roof deck. No vapor barrier was installed (not required by code at the time).
This year I started noticing major leaks. I assumed bad screws or washers, but after pulling panels I found every screw badly rusted, 30 year metal panels corroded through, and severe rot in the purlins and trusses. Interestingly, the wraparound porch using the same metal and screws (but no foam) is perfectly intact.
Inside the attic, the foam is saturated. I found:
-
Gaps in the foam with exposed metal
-
~½” delamination throughout the entire roofline with striations still attached (looks like it happened at install)
-
Gable foam only 2–3″ thick
-
Fiberglass used as soffit baffles
-
Some blistering on the surface of the foam
-
Encapsulated wood members nearly rotted through
Foam was applied in a single 5.5″ pass.
My theory is a dew point gap formed over time from vapor drive and delamination, trapping moisture and accelerating damage. The installer said the job looked fine, but clearly something failed.
Does this sound like an improper install or assembly issue? Appreciate any insight from others who’ve worked with foam + metal roof systems.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Open cell foam installs need a conditioned attic. Was the attic conditioned?
It wasn't. This was before it was required.
The problem is open cell is fairly vapor open, even a big lift of it like 5" is 6 perm which is about the same as painted drywall (ie fairly vapor open).
The questionable install, reduced thickness and missed spots did not help things but not the ultimate issue.
It also did not help that there is no roof deck or underlayment under the metal. This means if any areas de-laminate (which can happen from thermal movement) the air gap there becomes a moisture pump as any air that gets in there will condense at night time from radiative cooling and gets soaked into the foam/trusses.
With open cell foam the important part is to have a way to remove moisture from the assembly. This means conditioning. Conditioning the space most likely would have saved the roof as it does work for a regular unvented house roof.
Overall, with your type of assembly, the safest is closed cell spray foam directly over the metal. CC SPF forms a much stronger bond so it won't delaminate and it is a vapor barrier so moisture can't get to the roof and create the issues you are now seeing. It also does not rely on conditioning to keep it safe. It doesn't have to be all closed cell just enough closed cell for condensation control. The rest can be fluffy or open cell.
Are you looking for a way to fix the existing install or does it have to all come down anyways?
It all has to go. It rotted the trusses.
A few more pictures. There are several of these gaps in the foam that go straight to the metal roof. The last picture I used a snake cam to confirm that the delamination was throughout the entire roofline.
Those big holes all the way to the roof are due to a poor quality installation. This doesn't necassarily mean the mix was bad, but it does mean the crew didn't do a very good job with the application. Spray foam should ideally be applied in a very even layer over the entire surface you're insulating. My crew will go back and fill any low spots either with more of the "main" mix, or canned foam for smaller gaps. You could easily use canned foam to fill these small holes.
I think your main problem here is that the open cell foam got saturated with moisture. You really need to be using closed cell spray foam for a safe unvented roof assembly. It's difficult to know why the spray foam would have seperated, but my guess would be poor surface preperation prior to spraying. If the surface is very dusty, the dust can sometimes form a sort of release agent for the spray foam, meaning that the spray foam will adhere to the dust and not really adhere well to the actual structure. Think of this as spraying against wax paper, then you can seperate the material from the wax paper relatively easily. Other issues that can cause problems with adherence is an excessively damp surface, or a very cold surface.
I don't know of any way to deal with the seperation/delamination issue. That void will likely accumulate moisture over time, which means mold and rot risk for the wood. The only real solution here is likely to physically remove the old spray foam and replace it with new closed cell spray foam, being sure the surface is in good shape to accept the new spray foam. The only good news is the old foam should come down more easily if it's already delaminated over most of the structure.
Bill
Did you ask for open-cell foam?
Nope. I just asked the insulation company to insulate my home to code. At the time I knew nothing about foam insulation.
Remove all open cell ASAP, assess damages to any wood framing and correct, use closed cell foam if you want foam or rockwool if you feel jaded about foam. My demo of open cell foam starts this coming week and roof osb, shingles, flashing etc to immediately follow. I am going to use rockwool but I don't have a metal roof. I am converting to vented attic also now after this drama