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

Mansard-Style Vented Roof

maxwell_mcgee | Posted in Energy Efficiency and Durability on

Hi all,

We’re in the planning stages of a new build in Southern Ontario (climate zone 5? 6?) and are starting to translate the schematic design into construction details.

Roughly speaking, the house is an L-shaped house framing a courtyard in the back. There’s a long-skinny single-storey wing running North-South on one side of the house which holds the great room (family room + kitchen) and has a simple gable roof on it (will have high cathedral ceilings inside). Let’s call this the GABLE wing.

It’s connected to a rectangular 2-storey East-West running wing that holds the rest of the house (living/dining/garage/bedrooms). This East-West wing has a mansard style roof on it. It has a 16/12 slope on all sides, and rises up about 8′ and then is flat across. Let’s call this the MANSARD wing.

Met with the architect and builder recently and we talked about how to build the roof. They both seemed to think that a vented roof for both the gable portion and the mansard portion would make sense, and they’d build it with a service cavity on the underside to house ductwork, electrical, lighting, etc. within the conditioned space.

This generally makes sense to me for the gable wing. I can imagine how a sufficient flow of fresh air will make its way from the soffit to the ridge to ventilate that roof.

But what about the Mansard wing? How does the whole idea of venting a roof work if there’s no ridge? There will be a series of mushroom vents (or maybe a different shape…but same idea) spread over the flat part of the roof.

Is venting this Mansard roof a good idea? Or should I be pushing the architect and builder for an unvented roof?

One last point. I’m planning to install solar panels on the flat portion of the Mansard roof. So most of the roof will be in the shade — which I understand may further impact the drying potential inside the attic.


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 would look at it this way:

    The flat roofed portion doesn't know about the mansard area. All it knows is that it experiences some ventilation from around it's edges. It's just like any other flat roofed attic - and they can be vented, but it's tricky.

  2. Expert Member
    Michael Maines | | #2

    A true Mansard roof has a sloped upper roof, like a gambrel but on all sides. In that case, you would still be able to use a ridge vent, though you probably wouldn't get enough area in the ridge to reduce the overall venting to the IRC-required 1/300 of the floor area; you'd be stuck with having to provide venting equal to 1/150 of the floor area. (I know Ontario is not on the IRC but it's still a useful guide and I bet Ontario's code is similar.)

    With a low-slope upper roof--maybe call it Mansard-inspired?--I agree with Malcolm; the same rules apply as on any low-slope roof.

  3. maxwell_mcgee | | #3

    Michael, you're right. I guess it's more Mansard-inspired than true Mansard. So there might be a small "ridge" in the middle but minor.

    I guess my question would be, does the 8' of space in the cavity make any difference to the amount of venting that will occur? Is that enough height to create the type of stack effect pressures such that if we do a decent job of placing soffit vents at the bottom, we'll see enough ventilation through there that this assembly should be safe?

    As an amateur, I feel like I'm confusing myself also. If I go down the unvented path, that makes intuitive sense, but then the advice I've read suggests to keep the roof in dark colors and don't shade it -- so if I then go and install a bunch of rack-mounted solar panels on that flat part of the roof, do I lose all my drying potential and compromise the unvented roof assembly and run the risk of moisture and rot at the ceiling? Which pushes me back towards the vented option?


    -- Confused in Canada

    1. Expert Member
      Michael Maines | | #4

      Unfortunately there's no way to confidently answer your question. The simpler the structure, the easier it is to get the details right, and a low-slope roof with steeply pitched sides is getting a bit complicated, and it's unusual so we don't have a pile of examples to learn from. Professionally, as a designer I have to play it safe and show assemblies I am confident will be safe. There is a good chance that your low-slope roof is small enough that a few mushroom vents will be enough to keep the assemblies dry, but there is also a chance they won't. (I assume you've considered the challenge of making mushroom vents on a low-slope roof watertight?)

      There is some literature out there that says if you meet ten requirements and nothing ever changes that you can do an unvented, low-slope assembly with fluffy insulation, but that is way too many "ifs" for me. I avoid using foam (and other materials with high levels of embodied carbon, and potential health impacts) whenever possible, but I won't do a low-slope roof without foam. As such, I avoid designing low-slope roofs whenever possible, though I like how they look.

  4. Expert Member
    Akos | | #5

    Do you have a height limit, is that the reason for the flat roof? This type of roof is unfortunately very common in the city with strict height limits and it is always built as an unvented roof with spray foam. Unless there is a very good reason to do this, I would simplify the build and use a simple sloped roof. Spray foam plus complicated roof shape, you are probably adding $20k to your build.

    If you must have the shape, you can also build it without spray foam using exterior rigid insulation. Flat roof crews are used to using rigid insulation and this add very little extra labor cost. It is also simpler to build the roof slope using tapered rigid insulation. You want 40% of your assembly R value above the roof deck, so with an R31 roof, that is R12.5 rigid above and R19 batts bellow. Depending on your rafter size, the 2x6 batts won't fill the entire rafter space. This is not a problem as long as there is no gap between the insulation and the roof deck, you can hold the batts up using insulation wires.

  5. maxwell_mcgee | | #6

    Yes exactly, city height limit restrictions and therefore the flattening off of the roof.

    The reality is I quite like the aesthetic that this roof provides, plus the fact that it gives me a large flat portion on which to mount a solar array rather than having to mount it on the sloped parts and ruin the look from the street. (The front of the house faces South so it'll be a pretty big eye-sore to have a massive solar array visible from the street) .

    But it sounds like the consensus push is towards an unvented assembly in this case.

    I also found this webinar from Building Knowledge Canada which was quite useful and also pushed me towards unvented.

    We're going to target something closer to an R-50 roof, so I guess we should be aiming at R-20 rigid above and R-30 below?

    My preference is definitely to avoid spray foam, but am willing to accept that it may be necessary in certain parts of the roof assembly -- especially at the edges of the roof and around skylights and other penetrations. But am hoping we can use something like bat insulation for the majority of the underside of the roof.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |