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
Q&A Spotlight

Rescuing a Problem Cathedral Ceiling

Installing spray foam and a new drywall ceiling won’t be cheap, prompting this homeowner to look for alternatives

Nasty ice dams. The red circle marks the area where Kacey Zach's roof develops ice dams, in part because of a poorly insulated cathedral ceiling.
Image Credit: Kacey Zach

You could call it the $6,500 problem, because that’s what it’s going to cost Kacey Zach to re-insulate a cathedral ceiling with closed-cell polyurethane foam and hang new drywall.

Writing at Green Building Advisor’s Q&A forum, Zach explains the situation: a cathedral ceiling framed with 2x12s and insulated with fiberglass batts to R-38 “with no regard to air sealing.”

The result? Nasty ice dams where two sections of roof intersect. “I’ve gotten quotes to spray foam the ceiling to R-38 with closed-cell foam for $4400,” Zach writes. “I also got quotes to re-drywall the ceiling after it is all done for $2k. I could be into this project for nearly $6500…yikes! Do I have any other less costly options?”

That’s the subject of this Q&A Spotlight.

Despite the cost, foam is the best option

Even though the closed-cell foam isn’t cheap, GBA Senior Editor Martin Holladay thinks it’s Zach’s best bet for curing the problem.

“It can be installed from below — a job that will require new ceiling drywall — or it can be done from above — a job that will require new roofing,” Holladay says. “If your current bids seem high to you, you can always contact other contractors for more bids. But I wouldn’t be surprised if the job ended up costing what you were told.”

Zach thinks the foam will be between 6 inches and 7 inches thick, and plans on filling the rest of the cavity with R-19 fiberglass batts. In order to spread the costs out, Zach wonders if it’s OK to leave the insulation exposed for the winter and tackle the drywall later.

Unfortunately, that’s not an option, Holladay says. Exposed foam is a fire hazard that must be covered with gypsum drywall to meet code requirements. There will be some thermal bridging through the 2×12 rafters, he adds, but not enough to cause any real problems.

Less foam, more fiber insulation

Dana Dorsett has a slightly different suggestion: “Rather than an ungodly R-38 of closed cell foam, it’s fully code-legal, moisture-safe, and cheaper to go with R-20 in closed-cell foam against the roof sheathing, and fill the rest of the cavity with fiber insulation to bring it up to code-minimum R-value.” Dorsett adds a link to the section of the International Residential Code dealing with roof assemblies.

The reason to dial back the amount of closed-cell foam is the HFC-245fa blowing agent it uses, Dorsett says, a chemical with 1,000 times the global-warming footprint of carbon dioxide. “Using any more than the minimum required for meeting the other goals is the way to go,” he adds. “This is getting there with half the foam in a safe and responsible manner that can’t be argued by inspectors.”

Another option, Dorsett adds, is an Icynene water-blown product called MD-R-200, which has an R-value of 5.2 per inch and is a Class 3 vapor diffusion retarder (1.8 perms at a thickness of 3 in.), according to the manufacturer.

“You’d still be able to get to R-38 even in 2×10 framing without burning up the planet,” Dorsett says, “and it would be sufficiently vapor retardant to protect the roof deck.” Aloha Energy, in Saratoga Springs, New York, also has a water-blown closed-cell foam, he adds, and isn’t too far from where Zach lives.

Vented or unvented roof assembly?

As David Sheard points out, there are two competing theories on how to insulate this kind of ceiling.

“The two different theories are to leave an air space above the foam, below the roof sheathing, to allow air to vent from soffit areas to a ridge vent, which would keep the roof deck cooler and avoid condensation to build,” Sheard says, summing up a classroom discussion in his Canadian Home Inspectors’ course. “The second theory is to fill the entire cavity tight with foam making contact with the bottom side of the roof sheathing.”

One key question about the unvented assembly, he adds, is whether it affects the life expectancy of asphalt shingles on the roof, and whether manufacturers will honor warranties when shingles are installed over an unvented roof.

“Either way will work,” Holladay replies. “If you want to have a ventilation gap between the top of the cured spray foam and the underside of the roof sheathing, you’ll need to install very rigid insulation baffles before the foam is installed.”

As to the warranty issue, he says, it varies by manufacturer. Many won’t provide a warranty if there is no ventilation channel beneath the sheathing. “There is little logic to justify their position, however,” Holladay says. “Fortunately for contractors, the roofing warranties are almost worthless, so this issue doesn’t matter too much.”

Installing rigid foam is yet another option

Zach has found a deal on 3-inch-thick foil-faced polyisocyanurate board, and suggests putting two layers of this insulation against the roof deck between the rafters. “This will be a cut-and-cobble job from below, but each of the 11-foot runs are straight and wide open, so putting up rigid foam seems like it won’t be too bad of a job,” Zach says. “Rip the sheet on the table saw, spray foam the rigid to the rafter bay.”

That approach will work, says William Geary. The foam can be cut so it fits tightly between the rafters, or loose with foam filling the gaps around the edges of the foam.

“If your interior trim details allow, I suggest you strap across the bottom of the rafters with 1×3 or 1×4 (this will help reduce thermal bridging and reduce nail pops from seasonally moving 2x12s) and fill the remainder of the cavity with dense-pack Spider fiberglass or cellulose under Insulmesh netting,” Geary adds.

Holladay advises that, If Zach insists on the “cut-and-cobble” approach — a method that some homeowners like, but that no professional contractor uses — it would be better to combine “cut-and-cobble” with a continuous layer of foam on the interior, on the underside of the rafters, to address the problem of thermal bridging. As long as there’s enough rigid foam against the roof sheathing, Zach shouldn’t have to worry about any moisture problems in the ceiling with this “foam sandwich.”

Our expert’s opinion

GBA technical director Peter Yost credits the good points made by Martin and other commenters, adding these points:

Use a certified spray foam company and trained installer. Many if not most of the problem spray foam installations I have seen are due to a substandard level of expertise by the installer or the company in general. In the U.S., we don’t require certified spray foam companies and trained installers the way the Canadians do, but we should. Look for certification and training such as this one from the Air Barrier Association of America.

Go with water-blown foam. The benign water-blown spray foams have way less of an environmental footprint than the R245fa-blown foams. But you should be aware that switching to water-blown spray foam is not likely to save you any money. It should also be mentioned that flame retardants are still an issue even with water-blown spray foams.

It’s more important to get a continuous air seal than it is to vent. Venting all those valley rafter cavities is really tough to do, of course. But if you don’t have that topside drying potential of the venting, than the vapor profile of your roof assembly becomes critical. Take a look at this blog I wrote on the subject.

An unvented roof assembly must dry in one direction. I like the idea of continuous rigid insulation on the interior, as both an interior air barrier (if all seams and margins are taped/sealed) and a thermal break on the roof framing. But since the top side of the assembly — your asphalt roofing shingles — form a layer with very low vapor permeance, using foil-faced rigid foam eliminates any drying potential to the interior of your roof assembly. To provide some ability for the assembly to dry to the interior, you will need to pick a rigid insulation with at least some vapor permeability, such as EPS or rigid mineral wool. (I must say, I have never used rigid mineral wool on the interior of a roof assembly and can imagine there is a learning curve there).


  1. Bennett Sandler | | #1

    How about dense pack cellulose?
    We have a client with a similar problem. We batted around the idea of pulling a band of sheetrock mid way across the 12 foot cathedral slope to allow us to hook out the old fiberglass and then dense packing the cavity at 4 lbs. I understand that the dense pack has to be right on for this to work, but it seems simpler, cheaper, and healthier than a full demo and foam?

  2. Bruce Wilson | | #2

    Rescuing a Problem Cathedral Ceiling
    I have built cathedral ceilings using dense pack cellulose and never had a problem with ice dams, but must admit that I used ice and water barrier roofing underlayment on the whole roof to avoid any problems. The owner have never mentioned any ice dams.

  3. Stephen Wik | | #3

    rigid foam moves? and how to combine cheap/expensive insulation
    I've been following the posts on cathedral ceilings (and low slope ceilings - lucky me, mine is both) for a while now while I procrastinate dealing with mine. I too would like to combine some lower cost insulation with a layer of closed cell spray foam.

    I don't think using rigid foam is a good idea in any of the above mentioned assemblies because I believe it will change sizes due to temperature changes and ageing; eventually air will leak in the gaps - even if it was urethane-foamed into place or taped - reports of 1" or more changes in size could be even worse in the roof. Am I wrong about this Martin? You must have more confidence in rigid foam as an air barrier than I do, or you wouldn't have recommended the foam sandwich approach.

    In my roof, I only have 7.5" space between the rafters, and no room inside to add more insulation so I'll have to add from the top. What I want to do is 1) remove the existing roof deck and insulation 2) spray 2" to 3" the closed cell foam against the ceiling drywall from above 3)fill the rafter bays with cellulose 4) reinstall the old roof deck 5) add 2" of polyiso above the roof deck 6) install new roofing, probably metal. (is there a roofing material/method/assembly that can rest directly on top of my polyiso? or do I need another layer of sheathing?)

    I considered just filling the 7.5" space with spray polyurethane foam and thats it- problem being that it is very difficult to entirely fill the cavity with foam, and I keep reading that you want it in contact with the roof sheathing to prevent moisture. (and it only gives R40ish and doesn't address thermal bridging at all) That requires going from the inside right? Martin, in your article here: you state that SPF can be used from the top side, but I can't figure out how to do it without leaving a small air gap below the roof sheathing.

    So I said what I WANT to do , but I'm not sure I SHOULD do it. It seems like a risk to have loose insulation between layers of rigid or spray foam insulation. What if water does get in there from either leaks in the roof or moist air infiltrating cracks in an imperfect air barrier? It can't dry in either direction, right? Also what if we want to add some new lighting to our ceiling at a later date? This assembly is basically resistant to any future renovation unless we are going to run conduit or something similarly ugly on the surface.

    I have considered adding all the insulation above the existing deck in the form of rigid foam and using the existing 7.5" cavity for wires and can lights (just kidding, no can lights... but there would be electrical boxes of course). But this sounds very expensive and still offers no guarantees with regard to air sealing.

  4. User avater GBA Editor
    Martin Holladay | | #4

    Response to Stephen Wik
    You raise some good questions. As someone who answers questions about insulating cathedral ceilings almost every day, I am regularly besieged by homeowners who suggest sub-optimal solutions, asking, "Can I get away with this?" I am regularly asked, "What about cut-and-cobble?" or "Can I get away with a foam sandwich?" -- and I have to determine which option is less bad.

    Here's the bottom line: when insulating a cathedral ceiling, the right location for insulation is above the roof sheathing. Ideally, you want at least two layers of rigid foam, with staggered seams. That's the right way to do it.

    Unfortunately, for a variety of reasons, very few homeowners want to do it the right way.

    Yes, rigid foam can move. It can shrink. Yes, rafters can move. Yes, gaps can open up after a cut-and-cobble job is complete. There is risk with many of these approaches. Determining the level of risk is a judgment call.

    I have seen jobs done as you propose -- opening up the roof cavity from above and spraying closed-cell foam against the top side of the ceiling drywall. I wouldn't worry too much about the air space above the cured foam if I were you; it's no different from a ventilation gap. You can even deliberately ventilate the gap if you want.

  5. User avater GBA Editor
    Martin Holladay | | #5

    Response to Bennett Sandler (Comment #1)
    You suggested using dense-packed cellulose in an unvented roof assembly. That approach violates the building code and is considered risky.

    For more information, see How to Build an Insulated Cathedral Ceiling.

  6. John Winkler | | #6

    ice dam
    Howdy you all.

    Down here we don't have ice dams, but we are pretty tight with a dollar. Is there a way to run a small heat duct to the area where the ice dam forms? This seems as though it is a $200 solution to a $6500 problem. Maybe heat tape.

    My other thought is that is spray foam uncovered is a fire hazard, why is it not when covered?


  7. User avater GBA Editor
    Martin Holladay | | #7

    Response to John Winkler
    I don't know where "down here" is, but if you lived somewhere that had ice damming, you'd know that every hardware store sells electric heat tape as a proposed solution to ice dams. The idea is that you staple the heat tape to your roof, you plug it in, and — voilà! — the ice will be melted as your electric meter spins.

    As I'm sure you have guessed, this is a poor solution. Your house is leaking lots of heat, so why not waste more heat to solve the problem? It's the American way!

    For better ways of solving ice dam problems, read this article: Prevent Ice Dams With Air Sealing and Insulation.

  8. Brooke Guthman | | #8

    The real culprit
    It's interesting no one has brought up what might be the real culprit of this homeowners dilemma. How is all of that air getting into the cathedral ceiling? Unless it is a 'vented' cathedral ceiling, sheetrock that is taped and finished should be enough of a vapor retarder after a couple coats of paint to not cause this much damage. I'd most likely bet a week's salary there are recessed can lights involved. Recessed can lights should never be installed in ANY ceiling. Put in an enclosed soffit, okay, but never in the ceiling. Use surface lighting. Surface lighting offers more options as to look and ability to direct the light as well as normally not contributing to air movement in the enclosed ceiling.

  9. Kacey Zach | | #9

    There were recessed can
    There were recessed can lights which are now gone. I will say tho... the worst of the ice damns was on the other side of the roof (actually complete opposite side). When I tore down the drywall there was alot of rodent damage to the insulation. This house was a vacation home before we moved in and it was rather infested. Last year I sealed up the walls, now I'm doing the ceiling. I have two more sections of roof to take care of but that is next years project.

  10. Dirk Gently | | #10

    Which system was used?
    Kacey, What system did you ultimately use for insulation?
    Damn rodents can cause havoc on fiberglasss already poor performance. A point I have yet to come across till now.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |