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

Venting a cathedral ceiling that can’t be vented — and concern with ice damming

Sal_123 | Posted in Energy Efficiency and Durability on

I am building in Zone 5, have referenced Martin’s article on venting a cathedral ceiling and looked at best practice guidelines but am challenged by a particular area. I have a first floor room, 20 ft by 40 ft, half of it sticks outside the house’s main footprint, with a cathedral ceiling and a dormer that starts at the lower perimeter of the roof.
The upper part of the cathedral dies midway into the second floor wall.
It has created an area 15 feet wide by 5 feet that has microlams at the bottom and the top of the rafter bays. Thus interrupting any venting option from soffit to ridge. The rafters are 2x10s. Where the upper peak meets the wall from the second floor, I built out an 18″ ledge, this accommodates a 3″ round vent from below, allowing these bays to vent into the space under the ledge safely (as in no leak issues). But that would only be 1 vent at the top. Does it suffice? Does it need air to cross flow?
The only other way I can see venting these bays to allow air to cross vent, is by going up on the lower part of the bays, up though the roof. To have a lower vent to allow air cross flow, I would need to penetrate the roof plane. The roof is a Spanish barrel tile, the thought of creating leaks is very concerning. Plus, with the layers of insulation, see below, I may never see a leak until it is too late.  To do this, the lower vent opening would have to be at the bottom of the foot rafter bays to allow cross flow from a lower vent hole. Perhaps running a PVC pipe up through the hole, so the actual opening ends above the roof plane, raising it about an inch, ending just under the tile, which I installed atop a grid system of 1x3s and are 1.5″ above the plywood. On the exterior, the roof plywood is covered first with a GAF peel and stick membrane and a synthetic underlayment atop that. I am very wary of making any penetration in the roof plane. Given the GAF peel and stick  membrane on the entire roof surface, the plywood will not breathe upward and the idea of spray foam insulation or solid rigid XPS concerns me with possibly compromising the plywood’s longevity.
Alternatively, trying to think out of the box, I considered running 1.5 inch perforated PVC though the lower part of the bays, horizontally.  Perpendicular to the rafters, and then Tee-ing off PVC pipe segments to run up and terminate up under the ledge alongside the vent holes of the upper bay areas in an attempt to allow cross venting. But would it?
Another variable, that same room will house a large wood stove, Jotul CB600, 81,000 btu. The peak of that room will have air temps in the 80′ and 90’s in the dead of winter (despite an oversized paddle fan right above the stove to move air), thus it is concerning for water damming. That roof needs to stay cold. In the 9 1/2″ of the rafter span, I plan:
from top to bottom
1.25″ vent space
2.5″ solid rigid XPS, tightly fitted and caulked on all sides with polyurethane
5″ of R21 fiberglass insulation (it is 5.5″ but compressing it by 0.5″)
0.75″ EPS tightly fitted and taped
for an R of about 36 with meticulous attention on blocking any air movement.
Does the insulation suffice?
I am considering perpendicular strapping to add insulation to the roof plane but would really prefer to avoid the extra labor. Is it needed?
I am DIYing this as much as I can and time is an issue.
Any suggestions/ideas/constructive criticism.

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. Peter Yost | | #1

    Hi Sal -

    love to help with this but without any images, pretty hard to work through with just text. Can you support with visual(s)?


  2. Sal_123 | | #2

    Thanks Peter.
    You'll see in the pic, 3 bays were insulated. At he bottom there is a microlam, at the top it ends against the wall of the second floor. How would you suggest I create some form of cross ventilation?

  3. Expert Member
    BILL WICHERS | | #3

    It looks like you might have some moisture issues already (darker spots on the underside of the sheathing?

    I’d be concerned about those recessed lights. Even the “airtight” ones are still leaky, and they’re risky in this kind of ceiling assembly. If you really have to use them, at least use the shallow depth kind so you can fit more insulation on top of them.

    I think your best option here is probably closed cell spray foam. I wouldn’t trust cut and cobble rigid foam.


  4. GBA Editor
    Martin Holladay | | #4

    If you want to create a vented roof assembly, every single rafter bay needs to be vented at the bottom of the rafter bay and the top of the rafter bay -- not just a few of the rafter bays here and there.

    With microlam headers blocking air flow pathways, you need to detail this section as an unvented roof assembly -- which means that you need some closed-cell spray polyurethane foam. More information is available in the following two articles:

    "How to Build an Insulated Cathedral Ceiling"

    "Flash-and-Batt Insulation"

    1. user-723121 | | #12


      Does an unvented roof assembly absolutely have to include closed cell foam under the roof deck? What about foil faced polyisocyanurate tight against the roof deck, cut to fit between the rafters and sealed at all joints?

      I believe there is some seasonal drying through the roofing felt and shingles as Jon has stated. Winter condensation on the underside of roof decking is most often a result of air leakage through the ceiling or unvented bath fans. Attic vents getting covered by snow also contribute as now the flow or air out of the attic is stopped.

      1. GBA Editor
        Martin Holladay | | #13

        The method you suggest is called the "cut-and-cobble" method. Researchers have discovered that the cut-and-cobble method of insulation is risky in unvented cathedral roof assemblies -- it is associated with moisture accumulation and rot (apparently due to the fact that rafters change shape with seasonal changes in humidity and temperature, and with changes in roof loading due to snow accumulation and melting).

        For more information on roof assembly failures associated with the cut-and-cobble approach, see this article: "Cut-and-Cobble Insulation."

  5. Sal_123 | | #5

    Thanks for the input Gentlemen. Question, if the roof sheathing (5/8" plywood) on the exterior has GAF peel and stick weather membrane and on the interior side is closed cell spray foam, doesn't that compromise the plywood? I was under the impression a wall (or a roof for that matter) has to vent/dry/breathe, to either exterior or interior?
    I am thinking a 1.25" vent channel using rigid 2.5" XPS then spray the rest of the bay with closed cell. If that vented space has a 3" opening, if nothing else but only at the top, it should still allow that plywood sheathing to vent. Yes?
    Zephyr7, the ply looks wet or stained but it is not. Your comment on the high hats is correct. I will switch them out with LED wafers. I gotta keep the roof cold, it has a 6/12 pitch. I like Martin's take on cans, "They should be removed from your ceiling and deposited in front of a moving steam roller."

    1. GBA Editor
      Martin Holladay | | #6

      Q. "If the roof sheathing (5/8 inch plywood) on the exterior has GAF peel-and-stick weather membrane and on the interior side is closed-cell spray foam, doesn't that compromise the plywood?"

      A. No. Everything will be all right, as long as the plywood is dry on the day it is encapsulated in spray foam. This situation -- roof sheathing that can't dry to the exterior and can't dry to the interior -- happens all the time on roof assemblies that are insulated on the interior with closed-cell spray foam. These assemblies work.

      There are certainly some builders who prefer to see some mechanism allowing their roof sheathing to dry in at least one direction. But that type of builder needs to plan ahead -- and you didn't. So you don't have many options.

      1. Jon_R | | #11

        More drying occurs than people think:

        "...ccSPF on plywood sheathing with cellulose insulation on the interior has the capability according to WUFI to safely dry a leak up to 0.6% of the rainfall ..."

  6. DavidfromPNW | | #7


    I have the same exact problem. First, after getting educated here, I removed the four super slope can lights that my electrician installed. And, for that section, I am doing a flash and batt per code. with closed cell foam. It's a small enough section (about 70 sf) for me to just order it from tiger foam and DIY. I've done it once before and it's pretty easy.

    There is just no other "right" way to do this. I've talked it over with my inspector and we've brainstormed ideas and have come to the same conclusion. Best route if you are averse to foam is to remove all penetrations (can lights very bad), flash it with closed cell foam to code minimum and fill the cavity with batts. Airtight is a must. Good luck.

  7. Sal_123 | | #8

    You are right Martin, I did not plan it. I hired a professional whose job it is to plan, to specify details, a person who has devoted their career and livelihood to plan for me. Unfortunately the professional "planner" regurgitated the same dogma that has been followed for the past 50 years. The concept of outsulation, caulking the framing and paying meticulous attention to details that make a significant difference are not part of his dogma. Therefore non-planners, who are not professionals in the field, who did not attend graduate level education in such planning, such as myself, seek education and knowledge from sources like,, and awesome folks like Dana Dorsett.
    Had I planned it Martin, I would have planned it right.

    1. GBA Editor
      Martin Holladay | | #9

      I'm sorry to hear that you hired a professional who didn't plan how to insulate your cathedral ceiling before the roof was framed. That is certainly unprofessional, but probably not as unusual at it should be.

      Here at Green Building Advisor, our job, day in and day out, is education. We strive to help homeowners, builders, and architects. Our goal is to help all of these people figure out what they're doing before construction begins -- and before it's too late to achieve the building they want.

      I wish you the best of luck.

  8. natesc | | #10

    Is the outside done? If not, the easiest way to vent at this point would probably be to run furring strips and vent the endwall flashing. If you're shingling you will need another layer of plywood on top of the furring, if using metal may be able to run horizontal furring over the vertical.

  9. Sal_123 | | #14

    Hi Nathan, Thanks for participating in the discussion. The plywood already has GAF peel and stick membrane over the entire roof plane, with another layer of Grace Triflex XT synthetic underlayment on top of that. I then installed furring strips in a grid pattern upon which Spanish style barrel-tile sits. So yes, the outside is done. This is the first house I have ever built, live and learn. Furring strips and another sheathing layer may have been the best solution on this area. I sought to educate myself rather then just go along with the architect and subcontractors in all regards and I am shocked by the difference in what the best practice guidelines I got from reputable sites like GBA, finehomebulding, etc... and what the "norm" in my area has been.
    The architect specified the cathedral ceiling to have an inch of a vent channel, a thin styrofoam barrier and R-30 fiberglass kraft faced insulation. I know from sites like this, the critical importance of stopping any air inflitration and this does not cut it. This particular situation is made worse by the incredible amount of heat that will be generated by the wood burning stove sitting below. I'd rather diy-it and try to achieve some modicum of efficiency and good design than pay subcontractors who are clueless or just want to bang it out the same old way since "that's how we always do it". I will spray it, like David B suggested, get tanks of diy spray foam and seal it. It seems others have been stymied but a similar circumstance.
    I am grateful a resource like GBA, that brings other's experience and input to the table, exists.
    Caveat emptor

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |