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

My plan to insulate a low slope roof in a moderately snowy climate (Zone 6)

RGBTV | Posted in General Questions on

I have posted 2 questions recently. However, having just signed up for a membership, I now am better read and a little more clarified. I’m in Burlington, VT, Zone 6. The more I learn, the more I realize I have a very poor candidate of a roof to insulate. However, I already stripped it down to studs and think I will move forward with the best practices I can implement. I am asking for one last review of the plan and any revisions or feedback folks may propose.

I have a low slope roof on an addition to my house (added on 2 owners ago). The roof slopes from 120″ to 110″ over 10 feet. So, I think this is a 1:12 pitched roof? I have 2×10 rafters, and it seems like one of the previous owners added soffit vents, thin foam baffles with about an inch of air space, ceiling and wall insulation,  but no other venting or heating. In fact, there’s some minor evidence of mold against the ridge beam where the rafters run from. Probably because of the inflow from the soffits and then the bafles that terminate a few inches from this ridge board, releasing air with nowhere to go? On this forum and elsewhere I’ve been pointed to adding high point exhaust venting options like Cor-a-vent. However, it appears most of these have minimums like a 3:12 roof, which I do not have. 

In Martin Holladay’s article, “Insulating Low Slope Residential Roofs,” he does describe a venting method that Joe Lstiburek recommends. However, it sounds like the execution needs to be flawless and I don’t know if this is less risky than where I’ve landed.

So, I’ve landed on going with an unvented attic, with 4″ of spray foam that has an aged r value of 30 followed by R-23 rockwool. This addition also has a gravely asphalt roof in 3 separate long sections – perhaps built up roofing or modified bitumen? I don’t believe there’s any insulation above the sheathing. I don’t see any evidence of a snow/ice membrane.

I was also warned about ice dams, including potentially on this set up. It seems from a study I read by Tobiasson, Buska and Greatorex, from either 1998 or 2001, that with my ground snow load and my plan for insulation, that I will be low risk for ice dams. That seems to conflict with some info I believe I’ve come across here on GBA.

I believe the primary recommendation would be to pull the roof apart and add insulation above the sheathing. Is that correct? That seems out of reach for my budget and timeline (I’m willing to compormise on the timeline, but can’t really on the budget). 

So, essentially, I’m looking for one last round of feedback or advice before I authorize the spray foam contractor who is ready to spray this Thursday. 

Thanks, as always.

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. Roger Berry | | #1


    This reply may be a day late and a dollar short if by Thursday you meant March 25th. Your post is marked March 25th so maybe there is time.

    First off, I think 4" of foam aging to R30 is optimistic at best. That would mean starting at R8 or better. If that is a supportable figure, I would love to know what material it is.

    From what you describe, it is fortunate that the roof sheathing is in good condition. I am puzzled by the description of "the ridge beam where the rafters run from". This would seem to infer the roof is a shed roof off the side of the rest of the house, not a shed roof that is open on both eaves. If the high side of the shed terminates against a house wall, then venting such a shed is very difficult. The flow of air on such a low pitch is also not very effective. The lack of flow is probably what saved your bacon from complete sheathing failure.

    Insulating on top of the roof sheathing is indeed the preferred method for most low slopes. I have 8 inches of EPS nail base above mine. The sheathing stays above 50 and is vapor open to the inside by using r-19 batts between two by 12 or truss elements. The sheathing has a synthetic roofing paper as first air barrier, the nail base is screwed through to the joists then high temp Grace ice and water shield seals all the nailbase, standing metal seam or membrane on top.

    The nail base OSB is presumably sealed from moisture build up and the structural sheathing on the joists is warm enough to stay dry and also is allowed to dry to the interior. It is also vapor open to interior humidity, so that is why the outside insulation layer is the dominant R value.

    I am prefacing with all this to better highlight why you are heading into potential problems.

    If you commit to the foam spray under the deck, then you have sealed both sides of the sheathing, as the asphalt roofing (roll roofing?) is not vapor open in any meaningful way. If you never get a leak through the asphalt roofing, you will probably survive for some time. How long that will be is a guess. Do you know if there are older layers of material under the visible roofing? Do you know that any flashing at the high side of the slope is perfect?

    If you are fortunate, future re-roofing will only entail removing the asphalt material and not having to replace any sheathing which will be firmly attached to the foam you intend to spray. A stick down barrier and fresh roofing will then carry you on for some additional years. If water does intruded under the roof material and rot the now unable to dry sheathing, then good luck fixing the spray foam from above. It is also possible that the top edges of the rafters will also have rotted leaving you with nailing and load support issues.

    The plan to spray foam is fraught with perils that won't necessarily present themselves for some time. The 2x10's mean you will have 1 1/2 " strips of R-15 punctuating the 14 1/2" strips of R-53 (maybe). Will the heat loss be enough to melt snow in a big way. Maybe not, it takes a lot of BTUs to melt snow. Will the heat losses contribute to ice dams, maybe not. For me, the sun heating the metal roof is far and away the main melting driver.

    Your roof sheathing will be protected from condensing any interior moisture by the (closed cell) foam so you will be ahead of that risk. The addition of batts will decidely enhance your heat retention. Your immediate goal of being warmer will be attained. Is this a good plan if you are looking to the future and re-roofing soon. Maybe, maybe not. You will be creating a forever sandwich that limits your options.

    If there is still time to better describe the roof's position relative to other parts of the house, it may be possible to suggest happier alternatives.

  2. RGBTV | | #2

    Thank you for the feedback! And by Thursday, I mean 4/1 so there is still time to revise. I've responded to what I think are the key parts of your reply. Anything I've left out is oversight so I'm certainly happy to add more or further clarify. Thank you for taking the time to respond and offer any additional insight after this reply!

    >First off, I think 4" of foam aging to R30 is optimistic at best. That would mean starting at R8 or better. If that is a supportable figure, I would love to know what material it is

    The product the company uses spec sheet is at the following link, listing an aged R of 30 at 4". Granted, if I only get 3.5 in some places or less than I haven't achieved that.

    > This would seem to infer the roof is a shed roof off the side of the rest of the house

    This is indeed the unfortunate set up I'm working with.

    >Insulating on top of the roof sheathing is indeed the preferred method for most low slopes.

    If I had the budget, I am sure this is how I would've gone. I've recently become more familiar with this method, but this would mean I'd have to pull my entire roof apart first, right?

    >If you commit to the foam spray under the deck

    I understood this to be the imperfect third option I have available after the other options: insulating above the roof or ventilating (which we know I cannot do effectively while still insulating). To my mind, it seems this is my last best hope to have a functional, conditioned room?

    >Do you know if there are older layers of material under the visible roofing? Do you know that any flashing at the high side of the slope is perfect?

    I believe this is the first layer of roofing ever installed on this addition (installed by a previous owner or their roofer when the structure was built about 11 years ago). And, there is flashing at the high side, and while there has not been moisture intrusion to date, of course I dont know if that will last.

  3. Roger Berry | | #3


    Sorry to doubt the foam specs, I hope I can find it way out here in the west. Do note that there is a second formulation marked 1880W which allows for application on surfaces as cold as 20F. The Vermont I remember is pretty nippy this time of year. Alternatively, spending some money on overnight electric heaters to really warm up the room just before the spraying starts will increase your likelyhood of successful coating. Don't use propane heaters. Aside from asphyxiation risks, they dump huge amounts of moisture into the air.

    Also note the 18% moisture content limit on foam substrates. If you have any way to borrow a moisture meter, now would be a good time to check the dryness of the sheathing. A pin type should be able to give a more convincing reading though with plywood it might mislead if stuck into a glue band. Poke the joists as long as you are at it.

    Indeed you are faced with lesser options if getting done and back to living is the current status. Trying to go batts only until you have the budget for exterior insulation would demand super perfect air sealing of the eaves and ceiling plane to prevent any air borne moisture getting to the underside of the roof deck. A very difficult task to guarantee. Any moist air making it's way in would most certainly condense. What doesn't get taken up by ply and joists will certainly be held in the batts. Drying out would be near impossible. It also would mess up any hope of doing exterior insulation in the future because satisfying the proper ratio of inside/outside insulations would make the outside layer quite nuts.

    Venting is less than effective thanks to the pitch and need to patch in any continuous venting choices against an existing wall. Pan vents in every rafter bay would be messy, ugly, violate the roofing material repeatedly, and probably remain covered with snow far too much to be of value and be subject to icing that causes backup leaks. All other ideas I might suggest have pretty much the same faults.

    I suspect that what saved you from the dead-ended ventilation you uncovered is a combination of some available air exchange and fairly high heat losses making the roof deck warm enough to not become a long term sponge. The points along the ridge board where some mold showed is likely where the roof got coldest, maybe thanks to snow blowing into the house/roof corner or being in shadow longer than the main roof.

    So you appear to be stuck with what is definitely contrary to best practices. Foam the underside and do flash and batt on a roof. It is done for walls grudgingly, largely due to the foam cost, and stud losses, GWP issues aside for the moment. The primary difference in conditions being walls are not typically sealed with completely impermeable cladding. The wall is less insulated and water runs down the wall. (hopefully quickly and safely)

    Warm moist air will go up, so ceilings tend to see higher risk from air leaks. The demand for R49 in ceilings aggravates potential condensing conditions by rendering the roof sheathing even further from interior heat transfer. At least with a potential R30 separation from exterior temps, the foam surface temp should remain above what you can call dew point. The rafters will not, as the functional R value for the four inches past the foam face is about R6. The nominal full depth of a 2x10 is 9.25 inches or about R14 for edge to edge. The rafters will be very cool relative to foam and batt and thus pickup moisture load if presented. Begin with sealing all the eave penetrations as meticulously as possible. Same for the high side ridge board. Consider spraying foam across the sheathing and down the ridge board and fill all the eave pockets.

    Use a smart membrane below the batts and under the drywall, minimize or eliminate light boxes (no can lights) and or seal very carefully and use LED pucks that fit shallow boxes. Be clean and tight with taping at transitions from ceiling to walls and it is likely that you will survive. If in the future when you are doing over the roofing and find the sheathing is not healthy, you might salvage the situation by creating a higher pitch over roof with venting on the new side gables. This would allow the snow load to be carried by uncompromised new rafters. The ceiling and insulation would remain in place. The one big fly in this idea is whether or not you have enough room below any second story windows to allow a pitch alteration. The siding of course would need removing to allow placing the ledger to carry the new rafters and for new flashing details. But that is in the future.

    Best of luck and mind the details.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |