Article on cool roof failures in Phoenix?
This is going to be a super esoteric question for most people reading this but hopefully for not at least one person.
I’m trying to find an article describing the failure of the roofs of some upscale homes in Phoenix due to their use of either white roofs or radiant barriers but without handling ventilation properly. I read it within the past year but don’t know if it was written recently. I’m asking here as a last-ditch effort if only because y’all have a much higher likelihood of reading an article like this than most.
Okay, more details that I can remember. The builder built up the entire community with green building principles but after a very short amount of time (6 months to a year), new homeowners were complaining of visible condensation on the ceilings. When they investigated further, they found the roof decking to be often soaking wet with notable rot and mold already forming. The explanation given was that white roof reflected so much heat that the top of the deck was actually cooler than ambient air and so condensation gathered on the underside. This caused massive problems, maybe, because the ventilation was done completely wrong? My memory starts to fail at this part of the story and it’s the part that I want to remember the most.
Basically, I want to re-read what precisely went wrong and how they ultimately fixed it.
Anybody remember reading such an article?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Kurt,
Q. "Anybody remember reading such an article?"
A. I have written several articles in which that case was discussed. One such article was titled "Night Sky Radiation."
To read about the failures related to white roofing, scroll down to the section with the heading, "Radiation to space happens during the day, too."
Thanks, Martin! That's awesome.
Okay, I want to make sure that I understand what happened and what the ramifications are. Don't assume I understand this at all, in the following, I'm very much still learning.
With the heavily reflective white roof radiating so much heat on a clear day, the OSB sheathing ended up being a good 8 degrees colder than the ambient temperature both inside and out. Say the ambient outside temp was 40F and the dew point was 34F. With the OSB sitting at 32F, we're almost guaranteed condensation.
But since the top of the roof is presumably well sealed (flat roof and all), it's unlikely that any vapor can reach the OSB from the outside to be able to condense. That leaves us with the inside air.
In this case, there was fiberglass batts against the underside of the OSB. I'm not sure what the temperature of the air would be at that point but since it's treated indoor air (ultimately), it's definitely going to be more than even the 40F outdoor ambient. And since the fiberglass isn't even a little bit vapor impermeable, there is plenty of warm moist air up against the underside of the OSB... thus condensation and no real way to dry it off.
The solution those builders went with was to install rigid foam on the outside. On that same 40F day, the outer-most part of the foam would be lowered to 32F (since still painted white, likely) but it's still impermeable and so the moisture has nowhere to go. Meanwhile, the foam has insulated the OSB and so it's nowhere near 32F or even 40F. Problem solved.
I want to explore this idea some more. What if, instead of putting rigid foam on the outside, the builders sprayed closed cell foam on the inside? Now the OSB is impermeable on both sides and even though the temp is below the dew point, the only place moisture can condense on is the exterior sealed surface. That seemingly would also solve the problem?
What if there was a ventilation channel between the fiberglass batts and the underside of the OSB? Now we'd have moving(?) air under the OSB. Would that be enough to warm the OSB above the dew point? Or failing that, would the air be enough to dry out the OSB?
I've read a few documents from the Dept of Energy and BSC about unvented attics that reference using permeable insulation (like blown cellulose) up against the roof deck but ONLY in hot-dry places like Phoenix or Las Vegas. Those don't make any reference to roof color, though. Presumably the assumption with that recommendation is that the roof would be a much darker color? Or... are they saying it's okay because those are sloped roofs and maybe that changes the equation? It certainly seems like that recommendation is describing a use-case just like the known roof failures in Phoenix.
This source reports sheathing being an average of 1C cooler than ambient, which doesn't have much impact.
http://rdh.com/wp-content/uploads/2015/03/The-Problem-with-Ventilated-Attics-RCI-March-2015-Nov-5-2014-For-Printing.pdf
Most wetting comes from air movement (even interior to sheathing and back to interior) - so an interior side air barrier is critical to limiting the amount of moisture.
> fiberglass isn't even a little bit vapor impermeable
But painted drywall (or Membrain) can be.
Kurt,
Q. "What if, instead of putting rigid foam on the outside, the builders sprayed closed cell foam on the inside? Now the OSB is impermeable on both sides and even though the temp is below the dew point, the only place moisture can condense on is the exterior sealed surface. That seemingly would also solve the problem?"
A. Yes. That approach is used frequently. For more information on that approach, see these two articles:
"How to Build an Insulated Cathedral Ceiling"
"Insulating Low-Slope Residential Roofs"
Q. "What if there was a ventilation channel between the fiberglass batts and the underside of the OSB? Now we'd have moving(?) air under the OSB."
A. Yes. Again, that approach is very common, as you will know if you read these two articles:
"How to Build an Insulated Cathedral Ceiling"
"All About Attic Venting"
Note, however, that low-slope roofs are different from steep roofs -- if you want to vent a low-slope roof, you need an entirely different approach from the approach used for a steep roof. Again, this article explains what you need to know:
"Insulating Low-Slope Residential Roofs"