GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Mystery mold around a hot roof…

davidbailey | Posted in Energy Efficiency and Durability on

Hi Everyone!
I have a real head scratcher (for me) that I’m trying to problem solve. Bear with me, there’s a fair amount going on here:
I recently looked at a roof assembly in Cabot, VT (climate zone 6A) that has some very intense molding on the underside of the sheathing, though nothing is rotted out just yet. It’s a steep pitched (I’m guessing a 16in12 or steeper) gable roof that el’s into an attached unheated room over the garage, what would be a “bonus room” in a McMansion, for example. The unvented roof assembly is insulated with roughly 3-4″ of closed cell spray foam, which I think isn’t enough, but maybe that’s besides the point? The thermal enclosure of the roof ends at the roof’s last common rafter before the valley rafter, and the CCSF was installed down the gable end wall, again about 3″ worth. It was installed with Typar house wrap as a backing on the cold side of the wall. Post facto, half the garage was converted into a mudroom. This has foil faced polyiso foam of an unknown thickness over the framing on the interior side with about 10″ of cellulose over it which is a little denser than loose filled, but not “dense packed”, covered by 3/4″ t&g plywood.
Generally speaking, all the work seems to have been done okay, and I wasn’t able to find any obvious point source of moisture that shouted at me as being the dragon’s head of all the mold that was present. There does not appear to be a roof leak, based on the total lack of evidence of dripping inside. Admittedly, the asphalt shingles were covered in snow so I couldn’t see their condition.
The mold appears to stop at the point where the CCSF is in direct contact with the roof deck. The sheathing is moldy into the valley peak and works it’s way down the garage roof deck (it’s a California-style valley) spilling into the garage room and back up the underside of the sheathing in the garage. At first blush one would almost think there had been a fire and it was all soot. There was a half-assed hatch cut with a sharp knife through the foam into the hot roof’s attic that rodents have chewed a couple little tunnels into, but otherwise it fit into it’s hole okay. This seemed to be a point of moisture exfiltration from the house, but I’m inclined to think it’s not leaky enough to justify the quantity of mold. The attic has collar ties that have fiberglass insulation between most of them, kraft paper facing up, which makes me believe they were installed as an afterthought. I don’t understand why they are there, as the roof rafters are insulated up to the peak.
Finally, there were some small puddles of standing water in a fold of the Typar, which was of an unknown origin. Less than a couple cups worth of water, but still, why was it there?
Anyone have any guesses as to what the heck is going on here?
The homeowner wants to blow loose fill cellulose over the attic floor, which I think is not a great idea as it will restructure the thermal enclosure to some degree without addressing venting, or moisture control. Am I missing something here?
Thanks to all who have read this far! I really appreciate any and all feedback as I am coming up empty handed with a solution I feel confidant with.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Expert Member
    Michael Maines | | #1

    David, your description is hard to follow, and it sounds like a complicated assembly. Could you post photos of the exterior, additional interior photos, and/or ideally a sketch of the existing situation?

  2. davidbailey | | #2

    Hi Michael,
    My apologies if my post is difficult to follow - with something so complicated I'm finding it challenging to be clear. I have a few more photos of the structure, which I hope will help illustrate the situation, and I can grab a pencil and try and sketch it out if needs be. That might take me a little bit to draw it clearly.
    In IMG_5118 you can see the mold creeping up the sheathing from the awkward unheated space created by the roofs intersecting and into the unheated room over the garage. There's also a buddy there which helps give a sense of scale.
    In IMG_5120 is the peak of the roofs intersecting. Everything that is black is the mold, which hasn't affected the solid wood as much as the plywood (yet).
    In IMG_5125 is a sample snapshot of the hot attic when I removed the cut out section of the foam.
    I will endeavor to put a pencil to paper and sketch it out. If it turns out decipherable I will post it!
    Thanks again!

  3. walta100 | | #3

    I think by having 2 separated layers of insulation they have made the attic a cool moist attic perfect for growing mold.

    They need to decide if they want a conditioned attic and get all the insulation up to the roof deck and conditioned the attic with supply and return duct work. Or a unconditioned attic by installing vents at the peak and soffits with no less than R38 insulation on a well sealed attic floor.

    From the photos it seems did well and kept the ductwork and HVAC equipment out of the attic, is that true?

    Where do the bath fans vent to?

    How many can lights are in the ceiling?


  4. Expert Member
    Michael Maines | | #4

    David, your additional information helps a lot, especially the last photo. I still don't understand exactly what's going on but generally agree with Walter's advice and questions. A sketch would be great, but a photo of the exterior would probably tell most of the story.

  5. davidbailey | | #5

    Thanks so much Walter and Michael!
    I'm asking for photos of the exterior and should hopefully have something soon. I'm assuming you mean the exterior of the structure and of the area I'm asking about. I didn't get any photos of it as it was completely covered in snow, but it's been a little warmer lately so I might be in luck. I drew a sketch of the roofs as they intersect, a stripped down wall section, if you will. I'm attaching it.
    Walter, the bath fan that I looked at, which is in the mudroom area that used to be the garage, is a Panasonic WhisperLite that vents to the outside, though not very well. It's run thru semi-rigid flex ducting which makes a hard 180 degree turn. I pulled up a sheet of the t&g subfloor to inspect as it seemed like a suspicious culprit. However, it is well sealed to the polyiso and the vent pipe is taped and seems solidly done at least as far as I could see it, which was well out of the range of the trouble area. The cellulose was also dry in the vicinity of it, so I ruled it out as my point source. I gave a recommendation that it be vented properly and verifiably, and believe it will happen.
    I have had the homeowner purchase some cheap hygrometers to set around the house to see where the %RH hovers, but so far haven't heard back with numbers.
    I didn't inspect any other bath fans and don't know that there are any. There aren't any can lights in the attic ceiling, but there are some in the rafter bays (I know... I know...). There is no HVAC or anything like that in the attic.
    Oh, the house does have an HRV but the homeowner never or almost never runs it because it's noise is bothersome. It might be acting in place of a bath fan for the other bathroom. I recommended installing one or several duct silencers, but don't know if that advice will be taken.
    Let me see if I'm understanding this. I'm hearing that by adding the fiberglass to the attic floor they've altered the temperature of the attic but not necessarily the moisture content, hence bringing the RH% up as the temp drops, which is maybe a bit laymen in my terminology but basically accurate? So then, is the moisture heavy air all escaping out of the attic and into the trouble spot, but not necessarily affecting the hot attic? Is the slightly higher temp of the (not-exactly) hot attic saving the rafters in there from molding up?
    It seems that removing the fiberglass is the way to go here - establishing venting above the foam and can light filled rafter bays feels like a very hard sell. Would you recommend topping off the spray foam with a couple more inches and making sure it wraps over the rafters? Code here for roofs is R49, which is 6.5 inches of the next-gen foam folks are using, or so I've been told. Would the absence of fluffy stuff in this attic do the trick on it's own, or does there need to be active conditioning of the space? The house is currently primary heated with a wood stove and has oil-fired hydronic baseboard as a backup. Could cutting a grill and allowing for passive venting into the attic do the trick?
    Thanks again for your feedback, this has been enormously helpful!

  6. walta100 | | #6

    The reason they have the mold is any part of the house that falls below the dew point of the air it exposed to will get wet.

    The attic whole needs to be warm and well above the dew point or cold and vented.

    The thing I dislike about conditioned attics is that it increases the surface area by 30-40% making it 30% less efficient and materials cost 100% more.

    Personally I like a cold vented attic.
    My advice
    1 Abate the mold with dry ice blasting.
    2 Abandon the spray foam on the roof leave it in place with the understanding it serves no usefully purpose.
    3 Vent the ridge and soffits.
    4 Blower door test the house, looking carefully for leak in the ceiling.
    5 Insulation on the attic floor to R60 with cellulose.


  7. ohioandy | | #7

    David, it's still very difficult to figure out exactly what the layout is you're describing. Sounds like this attic area is halfway to being conditioned, but you can't do any of this sort of thing half way. Either the thermal/air barrier is the attic floor OR the house roof. Can't be a halfway combination of both or the whole thing fails (as you are seeing.) A passive vent would not be enough; the attic would need to be actively supplied according to its actual heat losses. I think Walter's giving the best advice--abandon the ccSPF on the roof and go with the cellulose in the floor, especially since the roof area is huge, half the area below is unheated garage, and the existing insulation on the roof sheathing is very inadequate. I'd do what I can to create an air barrier on the attic floor above the mudroom, vent the attic with lots of soffit and ridge venting, and then blow in a mountain of cellulose.

    As for the bath fan, dry cellulose around it does not mean it's not the culprit. This is a warm spot, so if it's leaking a ton of damp air there won't be any condensation here--it will be on a nearby cold surface like the roof sheathing.

    Wow, good luck convincing the homeowner to spend some money on this!

  8. Expert Member
    Michael Maines | | #8

    I agree with the advice Walter and Andy are giving you, and would add that generally speaking, each of the control layers in a building should be continuous. The control layers are, in no specific order:
    WRB (stormwater and wind)

    The closer these layers are to each other, the fewer problems are generated. You have various control layers of all sorts scattered about, with interstitial conditions that are hard to predict. Choose a single, continuous location for the control layers and allow the other, odd spaces to ventilate to the exterior.

    More images would help, but it's such a three-dimensional problem that you may want to consider getting a qualified energy consultant there to look at it with you in person.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |