GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X 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

New Approach to Insulating a Low-Slope Roof

Moderate | Posted in Green Building Techniques on

Regarding Josh Salinger’s article “a new take on insulating a roof” in issue 299 of Fine Homebuilding.  I am in the process of selecting the envelope details for a new home soon to be constructed and I’m wondering if this approach would be effective for ventilating a 2:12 single pitch roof in zone 5 of northern California?  This looks like a thoughtful solution to eliminating above grade plastic foams, but based on other articles on GBA highlighting the ineffectiveness of ventilation channels on low slope roofs I am wondering if this roof assembly would be appropriate for my application.

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


    I know this isn't the answer you want to hear, but the simplest way to eliminate worries about potential failures of low-sloped roofs is to design them out of your project.

    That's true of the three or four most common problematic building assemblies the come up here on GBA all the time. None of them are inevitable, they are all designed into houses. Which seems like something we should probably stop doing.

  2. Jon_R | | #2

    Based on European building codes and basic physics, it's my opinion that 2:12 roof venting will work better than 3:12 if you up-size the vent size to 3" and similarly increase the vent openings. Always air seal well and provide a good interior to exterior perm ratio.

    "3:12 works fine and 2:12 can't be vented" isn't logical.

    1. Expert Member
      MALCOLM TAYLOR | | #3


      You may be right. There may be no bright line, and the advice against venting roofs at that slope may have simply come from having to set the limit somewhere. However it's worth noting that a 3/12 roof is proportionally a lot steeper than a 2/12 one in a way increasing the pitch between a 7/12 to an 8/12 one isn't.

      My point is that I don't see much benefit to using any assembly or material close to it's limits. It introduces risk the doesn't need to be there. I have a set of things I just don't design into a house here in the PNW. They are all possible to do, and some are quite common here, but to me they just aren't good ideas, and not having them doesn't adversely affect the houses I design.
      - Low slope roofs
      - Decks over conditioned spaces.
      - Unvented roofs
      - Roofs with no overhangs,
      - Masonry as cladding or chimneys.
      - Walls that aren't vapour open to the outside.
      - Open cladding
      - Doors with no cover above.
      - Patios or decks that rely on floor drains.

      1. Jon_R | | #6

        > a 3/12 roof is proportionally a lot steeper than a 2/12

        Sure, the latter has 2/3 the height driving stack effect draw. But a 3" vent is say 2x the vent size - which will more than make up for it, causing flow as good as or better* than a 4:12 roof (which isn't at the limit).

        * - better because wind is the more important driver and this doesn't change much with small changes in height. This is why even flat vented roofs can work (with about 6" vents).

  3. GBA Editor
    Kiley Jacques | | #4

    To further Malcolm's point, bear in mind that Josh's method is for a vented roof, and venting becomes less effective on low-slope roofs. So, while it is possible, it's probably not best.

    1. verygood | | #8

      I tend to land somewhere in-between Kiley and Jon R on this. Although I don't have any first hand hard data on this there has been some good discussions on roof venting including a recent B.S. and Beer show with Ben Bogie and Martin Holladay discussing this. There is also some good wingnut science with Peter Yost about roof venting here:

      I completely get there are perfectly good reasons to have a lower sloped roof. I also agree that it is more risky than a higher pitched roof. That being said, there are good ways to manage the risk and the most important is making sure the roof assembly is air-tight first and foremost. Secondly, I would go for the larger venting space to cut down on static pressure and allow for more air movement. I would also choose the most vapor permeable membrane for the top of the rafters I could find- at 58 perms regular Tyvek isn't a bad option, but something like Delta Vent may be more durable. I would couple this with balanced heat recovery ventilation to control pressure differentials in the building and also manage RH. Cellulose insulation can act as a buffer, so I would prioritize that over fiberglass or other hydrophobic insulations.

      It's all about managing the risk and what type of risk you are comfortable with. These are some good belt and suspenders options to help mitigate that risk. A higher pitched roof would certainly be my first choice.


    2. Expert Member
      MALCOLM TAYLOR | | #9


      I think threads like this represent much of what's best about GBA. A fairly comprehensive overview of the topic, and some good advice for how to proceed if that's the plan.

  4. jberks | | #5

    I don't disagree with Malcolm.

    However I am the type that enjoys the use the engineering and technology in design. A lot of times I am not keeping it simple stupid, BUT, there are times where that is highly merited and this is the balancing act. We as humans fly structures the size of houses up in the air across the world on an abounding and regular basis, I like to think that I can handle designing technically for a low slope if it brings more value.

    For example, when you have a vast open plot to do whatever design you want, a high slope standing seam for the roof is an easy design win. But when you design in my parts, maximum usable space is critical to value. This means doing a flat roof to make a usable 3rd floor and having a rooftop patio.

    So to this, you could say I like flat roofs, as long as they're designed and applied well.

    The perception is that a low slope roofs leaks, but in my limited experience, I'd like to add that I find shingled high slope roofs also leak just as much.

    I once read in a roofing association report (don't ask me to name it or find it, I've already tried) that the three major things that cause roof leaks are:
    1. UV damage
    2. Water erosion
    3. Poor workmanship

    So how do I eliminate for those?:
    1. Put a green roof on it to absorb UV, protect the main membrane. (This is stipend by many major municipalities, Toronto in my case)
    2. Use a polyurethane liquid applied roof membrane (water doesn't hurt it)
    3. Also the polyurethane roof membrane solves this, it eliminates seams and relayilance on roof workers to makeup those seams, common with torchdown modbit or EPDM or TPO roofing (Note there are still areas that need special attention, like in any roof, skylights, plumbing stacks etc. PU isn't a catchall to good practice)

    I also throw in some redundancies, like using self adheared ice and water shield on the decking. the layers of polyiso are glued down, no fasteners (unfortunately it's not green & I couldn't find reclaimed polyiso on my last build) Also, the Green roof, above the PU membrane requires a root barrier, which is a heavy duty poly roll material, to which I shingle and tape the seams of and it ends up being the bulkwater layer.

    I'd also like to add that I don't ventilate the roof, I have it dry down towards the conditioned space.

    I'm sure that complicates things, but I guess that's how I like it lol


    1. Expert Member
      MALCOLM TAYLOR | | #7


      I agree with a lot for what you say. I grew up in Montreal which has predominantly flat roofs, and they generally work fine. But I think that is largely due to them doing what you do in your projects and not pretending they were something else and using the strategies to keep them safe that work on sloped ones.

      Any list like the one I posted is intensely climate specific. What works in one place doesn't in another. However I think the general proposition that assemblies become riskier at the margins of their usual use is self-evidently true. That extends to assemblies that rely on perfection of execution and that they stay in that state in perpetuity for them to work.

      Kudos to people like you who innovate and extend the boundaries of building science. I'm all for that and benefit enormously from it, but I don't see the point of integrating risky assemblies into houses when it yields no appreciable gains.

      1. jberks | | #10

        "I don't see the point of integrating risky assemblies into houses when it yields no appreciable gains."

        Agreed. It's all about the value gain.

        Well isn't it nice that people on the internet can respect each other's position these days... What a forum...

        Thanks Malcolm

        1. Expert Member
          MALCOLM TAYLOR | | #11

          Cheers Jamie,

          There wouldn't be much point c0ming here if we only listened to ourselves talk.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |