GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Audio Play Icon Headphones Icon Plus Icon Minus Icon Check Icon Print Icon Picture icon Single Arrow Icon Double Arrow Icon Hamburger Icon TV Icon Close Icon Sorted Hamburger/Search Icon
Musings of an Energy Nerd

Sliding Insulation Between Rafters From Above

If you need to insulate the sloped ceiling of a Cape Cod home, you may be able to slip in baffles and insulation from the third-floor attic

Working from above, it's possible to slide ventilation baffles and insulation into uninsulated rafter bays. But this type of work can be tricky, especially if the attic is cramped and if the sloped ceiling you are trying to insulate is long. [Photo credit: Joe Ducheck]

It’s harder to weatherize an old Cape Cod house than it is to weatherize other types of houses. The second floor of a Cape Cod house has sloped ceilings, typically interrupted by 4-foot-high kneewalls. In most cases, a Cape also includes a horizontal ceiling in the center of the second floor, with a cramped third-floor attic above the horizontal ceiling.

If you live in an old, uninsulated Cape, you know how hard it is to come up with a good way to insulate your home’s sloped ceilings. Among several possible methods, these approaches are typically recommended:

  • It’s possible to install continuous insulation (usually, rigid foam insulation) above the existing roof sheathing. Of course, this means that you’ll also have to install a second layer of roof sheathing above the rigid foam, as well as new roofing.
  • It’s possible to demolish the drywall or plaster ceiling on the sloped portion of the ceiling, so that the insulation installers can gain access to the rafter bays.

These approaches work, but they are expensive and potentially disruptive. What about a third approach — namely, trying to slide ventilation baffles and insulation into the rafter bays from the third-floor attic? Is that even possible?

The answer is “maybe.”


Before we provide tips on how it may be possible to slide insulation down the rafter bays from the attic above, let’s list the factors that may make this work difficult or impossible:

GBA Prime

This article is only available to GBA Prime Members

Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.

Start Free Trial


  1. Doug McEvers | | #1

    This sounds like the kind of project I would get involved in, cramped quarters and nearly impossible. The one thing I will say about insulating ceilings, flat or cathedral, the work must be impeccable including air sealing. We have had a lot of light, fluffy snow in the Twin Cities in the last week and the ice dams are really showing up. Some of these on fairly recent construction that should have been built better according to MN Code.

  2. David B | | #2

    The Title of this Article should be "Mission Impossible". This is one of those "I have more time on my hands than money" projects that you get 50% through, throw up your arms and do what you should have done in the first place, which is remove the drywall. In reality, if you need to do this work in the first place, it's probably leaking inside air anyway. If you can do this easily, then your roof probably wasn't nailed on properly. Save yourself some aggravation. My advice: Do not try this at home.

    I can't count the number of stupid ideas I've come up with to save money or save from destroying one thing to fix another. I think I am batting a 1000 on my stupid workarounds actually wasting more time and ending up in the same darn place. Do it once, do it right. Great article to read as I got to reflect back on some of my more interesting workarounds.

    1. GBA Editor
      Martin Holladay | | #3

      Thanks for sharing your perspective.

      I have provided the following advice to countless homeowners asking about this type of work: "Remove the drywall on your sloped ceiling." As you might imagine, the pushback is significant. No one wants to do it.

      The steps in this article describe an alternative path -- but as you point out, the alternative path has its own pitfalls.

      1. marg0049 | | #19

        Would this method work for a house with no vented soffits or ridge vents? We have gable vents and a couple scattered eyebrow vents only.

        1. GBA Editor
          Martin Holladay | | #20

          For a vented approach to work for an insulated sloped roof assembly, you need air inlets low in the roof (usually soffit vents), and a continuous air path directly under the sheathing, and air outlets high in the roof (usually a ridge vents, although gable vents in an attic can work).

          If you don't have a vented soffit vents, you might consider installing vented drip edge flashing or shingle-over vents. If you're not familiar with these products, you can use Google for more information.

          1. marg0049 | | #21

            I'm wondering about cutting channels in our soffits with a circular saw, then adding vents to that. Our current R value is something like 0.5 in climate zone 6, and unfortunately our rafters are only 3.5 inches deep. I'd use the method in this article to add 2.5 or so inch thick bats of rock wool, which i would get by cutting 5.5 inch thick rockwool batts roughly in half, leaving an inch for air movement. This would all be an attempt to avoid the thirty thousand dollar fix of ridged foam on the roof deck. If installed reasonably well, do you think this would have have a noticeable enough impact on regulating our upstairs room temperature to be worth while? Hypothetically this might be R-10 of insulation at this point.

          2. GBA Editor
            Martin Holladay | | #22

            Whether or not the fix you describe will make much difference depends to a large extent on the quality of the installation job. Getting batts properly installed -- filling the rafter bays without bunching or voids -- is tough in the best of circumstances. If you are very conscientious, the effort would be worth it. But remember -- even if the installation is conscientious, you'll end up with a low R-value compared to code recommendations.

            For more information on possible fixes to this problem, see "Insulating Tight Spaces at the Eaves."

  3. Expert Member
    Kohta Ueno | | #4

    Just so folks are aware of current research, there is a Building America project looking at injecting closed cell spray foam into closed rafter cavities as a retrofit. The application to this geometry would be if the roof were being switched to an entirely unvented roof; their approach would inject ccSPF through holes drilled in the gypsum board/plaster at the "compact" roof portion.

    Of course, if the compact roof section is short enough, a spray foam contractor might do slow-rise injection/poured foam from above.

    I'm always wary about shooting foam "blind"--seems like many things can go wrong from a quality control point of view. But I'll look forward to whatever results these guys get.

    1. GBA Editor
      Martin Holladay | | #5

      Thanks for the information on the Building America research. Like you, I'm interested in the results.

      In addition to all of the problems inherent in trying to "work blind," there is another potential problem with injecting foam -- the problem of blowing out the drywall or plaster. Of course, that's not supposed to happen with a skilled installer. Yet, amazingly enough, it still happens...

  4. Andy Kosick | | #6

    Thanks for this article, it makes me feel like I'm not crazy having done this several times professionally. In response to David B above, I think there is room for a solution like this. These houses aren't going away, nobody wants to tear apart the inside of their house, and usually the call comes just after a new roof was put on (and there's still ice dams). Also, if the rafters are at least a 2x6 it works. By works I mean there's very little ice damming and interior surface temperatures are reasonable. If you push the R-values up everywhere else except that sloped section the overall situation is pretty good. I'm considering one right now with 2x4 rafters that needs a roof, and I'd like to add foam above the roof deck, but I'm actually struggling to find contractor to work with in the area. As much as GBA talks about this method, I don't think anybody's doing it in my neck of the woods yet.

    We use 3/4" polyiso with clips made from F-channel (see picture). The foam seems to take dense-pack cellulose with only a little deflection, even at 2 ft centers. I think the trick to this work is to prepare everything possible outside the attic and then think through the process very carefully to minimize movement. At least two people, one installs, the other feeds tools and materials to the installer.

    The ccSPF sounds good but I have my doubts. Foam just doesn't fill the way fiber does, the amount pressure needed to fill every nook and cranny tends to be more than the wall/roof can take. I thermal imaged some walls where a slow rise ccSPF was used and they looked like swiss cheese. That said, if they can make it work, I'm open to solutions.

    1. GBA Editor
      Martin Holladay | | #8

      Thanks for sharing your experience with this method. Reports from the field like yours are valuable.

  5. David B | | #7


    I'll stand partially correct, but with a caveat. The pics shown by you is a post 1952 house I presume. I would guess this would be easier with trusses for sure. Still a PITA and still has many of the same issues (air leaking). But I get it. It's circumstance dependent and if your circumstance is that you can't/don't' want to remove drywall , this is a valid option in some cases. Cheers

  6. Jon Nicholson | | #9

    I have a 1940s cape (Portland OR) and have been curious if this technique would be feasible in the sloped portion of my plaster ceilings. It’s just under a 4ft stretch with access only from the underside - the ‘upper’ attic has no access and is maybe 20” tall from if I did my geometry right.

    2x4 rafters so I would like to maximize the R per inch of whatever I can stuff in there. Does anyone have experience attempting this procedure with 2” of foam board beneath a ventilation chute? I imagined the foam board would have to be cut quite loose and gaps filled with canned foam (albeit blindly...I’ve seen applicators with 4ft extension)

    Attic gets quite uncomfortable in the summer/winter and as much as I’d like to address that I’m not sure it’s worth moving out of our upstairs bedroom to embark on some plaster demo

    A few more details: vented attic, unconditioned; about 20yrs ago somebody came in with blown-in fiberglass filled the attic floor and sprayed it up into vaulted ceiling portion - what’s left of it is ineffective at best and smells terrible.

    Appreciate any input that’s out there

    1. Sal Oriti | | #11

      If you fill it with foam board against the plasterboard, there will be an air gap on top of that, no need for a baffle! One 2-inch thick XPS board would work for you with your 2x4 rafters

      1. Expert Member
        Dana Dorsett | | #13

        >"One 2-inch thick XPS board would work for you with your 2x4 rafters"

        A 2" foil faced polyisocyanurate foam board cut'n' cobbled between rafters would do even better, and a LOT greener (even though putting high-R/inch foam between framing is generally wasted.)

        XPS is the heaviest impact insulation material in common use today, with nearly 10x the CO2e footprint per R as EPS or polyiso, the difference between low-impact hydrocarbon blowing agents vs. extreme global warming potential HFC blowing agents:

        But any high R/inch insulation is robbed of it's performance potential when installed between framing:

        An R13 slab of foil faced polyiso in a 2x4 frame performs worse than well fitted R13 fiberglass (but the fiberglass won't leave room for a vent channel in a roof assembly.)

        1. Sal Oriti | | #15

          Wow, great info thanks. I hadn't gotten nearly that far to consider the effect of MAKING the insulation. So polyiso is better insulator, AND greener? Sounds like a win win. So what do you think is the right insulation to put in here, while maintaining the vent channel above it?

  7. Sal Oriti | | #10

    What about vapor barrier placement? Shouldn't there be a vapor barrier up against the drywall in each rafter bay?

    1. GBA Editor
      Martin Holladay | | #12

      There are a lot of misunderstandings about vapor barriers. Building codes in the U.S. do not require a vapor barrier on the interior side of walls or ceilings -- only under concrete slabs.

      Most building codes require that homes in cold climates have a vapor retarder -- a less stringent layer than a vapor barrier -- on the interior side of some walls. (There are exceptions.) Vapor retarder paint meets this code requirement.

      A ceiling adjacent to a vented cathedral ceiling is not required to have a vapor retarder.

      For more information on this issue, see these three articles:

      "Vapor Retarders and Vapor Barriers"

      "Do I Need a Vapor Retarder?"

      "All About Vapor Diffusion"

      1. Sal Oriti | | #14

        So there's no need for a vapor retarder on a upstairs ceiling, to stop movement of moisture into the attic?

        1. Expert Member
          Malcolm Taylor | | #16


          What Martin was saying is you can't use the term Vapour Barrier and Vapour Retarder interchangeably. A Vapour Barrier isn't required, but a Vapour Retarder may be. If it is a variety of materials will work, including some paints.

        2. GBA Editor
          Martin Holladay | | #17

          Read the articles. By far the most important mechanism that carries moisture from the interior of a house into an attic is air leakage, not vapor diffusion -- and to stop air leakage, you need an air barrier, not a vapor barrier.

          In any case, the ventilation channel in a cathedral ceiling (or the ventilation openings in an attic) take care of the very small amounts of moisture that may enter these areas by vapor diffusion.

  8. Deleted | | #18


Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |