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

Poorly insulated cathedral ceilings

Montgomery413 | Posted in Energy Efficiency and Durability on

We have recently purchased a Cape style house. The front (north side) of the house has very narrow 2×6 framed cathedral ceilings. The housing was at one point reinsulated with fiberglass bat insulation and a plastic vapor barrier. We had an energy company come out a do an audit and we decided to have them dense pack the cathedral ceiling with cellulose. And reinsulate the attic with blown cellulose also.

As they say hindsight is 20/20 and we are now looking at other options. First off they have been back multiple times to fix their blunders. They missed the bay closest to the chimney and when they came back to fill it they found the insulation was soaked. Myself and a contractor friend determined that the moisture was from the heating and cooling cycle of the roof being close to chimney and poorly insulated. The company is planning on insulation around the chimney (very tight space) with rockwool insulation and then dense packing the rafter bay with cellulose.

My wife and I are looking for options moving forward. We don’t want to rip the ceilings down in the upstairs of the house. I suspect that the insulation company did not remove most of the old fiberglass insulation because it was not reachable without taking the drywall down.

Our concerns are did the product they sell us really help or hurt our insulation situation? Also I had the following idea to increase our R value on the cathedral ceiling. I want to remove the shingles, replace any water damaged roof sheathing, put 3 inches of foam board over the existing roof sheathing, and build a new roof deck on top of the foam board insulation and shingle.

I come here asking for advise. I am by know means an insulation or building expert, i have a good understand of our situation but would like any possible suggestions to our current situation. Thank you very much.

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. GBA Editor
    Martin Holladay | | #1

    Your insulation contractor has chosen a risky method of insulating your ceiling. It sounds like your rafter bays are unvented. If that's true, then the insulation method used by your contractor would violate the building code if the method were used on a new house.

    Your approach -- installing rigid foam above the roof sheathing -- is the right way to proceed.

    Unless I missed it, you didn't tell us your climate zone. The minimum thickness of the required rigid foam layer varies depending on your climate zone; colder climates require thicker foam.

    Here is a link to an article that explains everything you need to know: How to Build an Insulated Cathedral Ceiling.

  2. Montgomery413 | | #2

    How can they sell a product that is not up to code? The more I look into this process the more questions I have. We are considered climate zone 5b, we are in Western Massachusetts but at around 1100' of elevation, so it's cold in the winter.

    How would I setup the foam over roof sheathing? Do i need a vapor barrier? We already think there is a potential moisture problem. Or would the extra insulation mitigate the temperature extremes? Thank you.

  3. GBA Editor
    Martin Holladay | | #3

    Cold roof sheathing can absorb moisture during the winter (or in some cases, condensation can occur on the cold sheathing). The source of the moisture is the warm, humid interior air.

    The reason that installing rigid foam above the roof sheathing solves the problem is that the rigid foam keeps the sheathing warm during the winter. Warm sheathing is dry sheathing. (Of course, if you have a roof leak, that's an entirely different problem.)

    Here is a link to an article with installation details: How to Install Rigid Foam On Top of Roof Sheathing.

  4. Expert Member
    Dana Dorsett | | #4

    Western MA is zone 5A, not 5B, not that it matters much...

    Dense packed cellulose as PRODUCT is a perfectly code legal product, it's the assembly with vapor barriers on both sides (shingles & felt on one side, poly sheeting on the other) and no roof deck ventilation isn't code compliant. Right now it's fine as long as moisture NEVER gets into the rafter bays, but once it's in, it pretty much never leaves, so there is high risk of problems eventually showing up somewhere.

    It's fine to dense pack over old batts, and the compressed batts still provide substantial R-value comparable to (or even higher per inch) as cellulose.

    There are a couple of approaches to take moving forward. On sections where you can rip down the cathedralized ceiling you can remove the poly sheeting and replace it with Certainteed MemBrain or Intello Plus smart vapor retarders. If the cellulose is dense packed, it will hang there just fine while you're doing that work.

    As for installing rigid foam above the roof deck- don't, unless you've replaced the interior side vapor barrier with a smart vapor retarder. If you're only working above the roof deck, strip it down to the deck, and install a vapor-permeable underlayment, and only insulate with rock wool above, with at least a 1" vented air space between the rock wool and the nailer deck. Rock wool is extremely vapor permeable, and with a permeable underlayment and a vented nailer deck, the structural roof deck can dry toward the exterior, despite the poly sheeting on the interior.

    In a zone 5 climate the R-value of the above-deck insulation needs to be a minimum of 40% of the total center-cavity R to protect the roof deck from interior moisture drives by keeping the average winter temp at the roof deck above the indoor air's dew point. With standard milled 2x6 rafters dense packed cellulose (or cellulose on top of batts) comes in at about R20 in the rafter bays, so it takes a minimum of R13.3 above the roof deck. R15 rock wool is 3.5" deep, and would be good enough for dew point control, but the total R would still be sub-code (R49). R23 rock wool is 5.5" deep, and would be a better choice, but it could be awkward to implement, depending on your roof lines. At that point you'd be at R43, but if done as a continuous or nearly continuous layer rather than banging in 2x6 rafters it would likely meet code-min on a U-factor basis.

    You'd also have to figure out how to establish the ventilation gap etc, and the total stackup would be adding at least 7.5-8" to your roof thickness. You may opt for a continous R15 solution that's only 5-5.5" of additional thickness.

    With the vented rock wool and an interior side vapor barrier there is a minor concern about summertime condensation potential on the vapor barrier. As long as you don't have roof leaks keeping the cellulose humid, you'd probably have to air condition down to 65F or so for that to really be an issue in a MA climate. Midsummer average dew points in western MA run in the mid-60sF, and even if the vapor barrier is 70F on those few hours & days when the dew points climb into the 70s the cellulose safely redistributes the small amount of moisture accumulation, and releases it over the following days. The structural roof deck itself is a smart vapor retarder, and the amount of moisture coming in will be pretty tiny unless the vapor barrier's temperature is chronically at or below the outdoor air's dew point.

  5. Montgomery413 | | #5

    Thanks Dana. I can only assume the majority of the interior cathedral ceiling has plastic lining. We redid one room and the whole ceiling was covered in plastic, interior side, below the fiberglass bat. There was no obvious signs of moisture. When we got close to the chimney that is where we found soaked fiberglass insulation.

    Now we have been waiting to see if the roof is leaking or if it was just the extreme temperature difference caused by the warm chimney and the cool roof. Have not found or seen any leaks. I really have no plan to explore anymore of ceiling in search of plastic vapor barrier.

    You are saying that the best bet would be to rock wool insulate about the existing roof sheathing because of moisture control? We were planning on just doing this to the front roof for now because the other roof is only a couple years old and it seems silly to replace it. The front roof is only 10 years old but we are concerned with mold and moisture build up.

    We would have to frame to get all the rock wool secured correctly right? Seems like a really big project.

  6. Expert Member
    Dana Dorsett | | #6

    Yes, it's a really big project to strip it down to the roof deck and add a vented rock wool insulation above that. With crappy R19 batts in 2x6 framing with air-leaks at the soffits it's sorta-vented, since the R19 has very low air-retardency- it'll let the air move. Cellulose is an order of magnitude more air-retardent, so any tiny seep from a roofing nail leak just gets redistributed within the cellulose, but doesn't leave via air leakage.

    There are a number of ways to install the rock wool above in a somewhat continuous fashion, but all would involve significant amounts of framing, and you'd have to really design it to meet wind and snow loading conditions, etc.

    The north side is the most susceptible pitch since it gets the least amount of direct solar heating, which means lower average roof deck temps, and much slower drying. If you're not going to add exterior insulation, strip the roofing down to the bare wood, lay down a semi-permeable under layment (#30 felt is fine, and install 2x3 or 2x4 furring through-screwed to the rafters with 3.5" or 4" pancake head timber screws on which to mount a new nailer deck, with a standard roofing layup on top of that. If the roof has hips and valleys, install the furring in 20-24" segments with 8-12" gaps so that it cross-ventilates. Use the compressible roll mesh commonly used for ridge venting at the soffits to mitigate against turning the 1.5" vent gap as a critter-condo. Install ridge venting on all ridges and hips too.

    That adds only ~2-2.25' to the roof thickness, but allows both the structural roof deck and new nailer deck and won't look TOO wierd if you don't get around to the other pitches until later. That's a pretty low performance roof relative to current code minimums (MA code R-values are based on IECC 2012, which calls out R49), but it's at least not a moisture trap.

    I can all but guarantee that the moisture problem at the chimney is in no way related to the temperature difference. A warm chimney would tend to drive moisture off, not attract moisture. Masonry chimneys are usually moisture permeable, and chimney flashing & cricket mis-installation is common, as is leakage trough flashing seams from snow melt when the snow depths are above the top of the flashing. Unless the masonry is sealed periodically with a decent sealer and the flashing is installed high enough to be above the typical roof snow depths you'll almost certainly be getting at least some seepage in via the masonry. The problem is aggravated by the fact that it can't dry toward the interior through 6 mil poly sheeting, nor to the exterior through wet shingles and a layer of snow.

  7. Expert Member
    Dana Dorsett | | #7

    It occurs to me that if the local inspectors would go for it, using 4" wide 1.5' thick EPS or XPS strips centered directly over the rafters and through-screwing the new nailer deck directly to the structural rafters with 4.5" pancake head timber screws would improve the thermal performance over using wooden 2x furring, roughly doubling the R value of the framing fraction. It would still be well below code-R but would limit the amount of visible frost-striping compared to doing it with 2x4 furring, and help limit the ice-damming potential as well.

  8. Montgomery413 | | #8

    That's a ton of good information, thank you. Currently the bay downhill from our chimney is open and uninsulated, not sure if I mentioned that before. So we are losing a lot of heat through the roof at this point. I inspected the roof deck from the inside and i could see that after 2 days of running the wood stove some of the deck had dried out but other sections have not and I notice nail seepage, no obvious leaks but not drying out completely either.

    I'm new to being a homeowner and dealing with this style of house. I was told by my experienced builder friend that he doesn't want to get on the roof and check for leaks until spring and honestly I do not blame him, it's a steep pitch and has snow, ice and the roof and shingles are cold. His suggestion was to call a professional roofer and have them see what is going on. I'd just hate to spend the money fixing something that I will end up replacing anyway.

    I guess I'm kind of at a lose of what to do. I'm assuming we have wet insulation throughout the cathedral side of the house and now we have fresh cellulose that is most likely adsorbing moisture too with no where to go. Of course that has to happen a month before our first baby is born and in the dead of winter! At least this winter is not like the last.

  9. Dana1 | | #9

    Don't panic.

    When it comes time to do whatever roof-over scheme you decide on, drill a few inspection cores in the roof deck in suspect areas with a 1.5" hole saw. Even if the cellulose feels a bit damp in your hand, if it doesn't LOOK wet you won't have to remove it. Save the plugs from the hole saw and can-foam seal them back in place before putting down the semi-permeable underlayment.

    If it turns out that the cellulose is saturated wet everywhere you'll have to get rid of it and replace it, but it should be on the insulation installers' dime, not yours.

  10. Montgomery413 | | #10

    I'm paniced yet. It's just strange to me that the insulator went about this process without any investigation. I think my plan is get a roofer up there and figure out what is going on with that section of roof.

  11. Dana1 | | #11

    Really, DON'T panic. There are many roofs stuffed with cellulose and vapor barriers on both sides that have gone more than a decade wtih out problems. Even if problems develop, it doesn't fall apart in a year.

    If you're able to see to those damp nail points the moisture is condensation, not seepage. Nails are thermally conductive, and it just means the nail points are below the dew point of your interior space air, and exposed to the interior space air. Nails generally won't leak until/unless the roof is in really rough shape.

    Roofers & insulation contractors don't always have the best grasp of the building science. The roofer should be able to figure out bulk water managment issues, but no necessarily drying paths.

  12. Montgomery413 | | #12

    Thank you. Our roof front roof is around 10 years old. It looks to be in really good shape and even after we look at it carefully the flashing around the chimney and the roof itself look to be in good shape. I only saw moisture on the nails, so it makes sense that it is the nails conducting colder air.

    Do you have any suggestions that would work to better insulate the roof other than rock wool? I would love to add 3inches of foam insulation above the roof deck. I remember you saying before to avoid using foam board. Just trying to educate myself on all options for the future project. Thanks again.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |