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
Q&A Spotlight

Fixing Problems With a New Metal Roof

Adding a layer of exterior insulation looks likely, but the question is how much?

The metal roof is new, but problems are cropping up inside in the form of mold. The homeowner is looking for the right fix. Photo courtesy Cavallero.

Cavallaro, a GBA reader from Arkansas, replaced a wood shingle roof on his 1975 A-frame house last year with standing-seam metal, but far from enjoying the upgrade Cavallaro is ready to tear off the new roof and start again.

The problem? Mold.

“The new vapor-impermeable roof caused moisture to build up inside,” he reports in this recent Q&A post. “In addition, the metal changed the temperature of the interior roofline causing condensation build up and mold to grow all along the interior roof during winter.”

The roof consists of 2-in.-thick tongue-and-groove boards—no attic, no air gap, and apparently no insulation. Just wood. Mold is one issue, odor is another. In the heat of summer, Cavallaro says, the metal roofing heats up the underlayment and makes a “horrendous” smell inside the house. All of it, he says, has been an emotional challenge.

Mold accumulating on the interior of the uninsulated roof deck.

Cavallaro has concluded that the metal roofing has to come off. He’d add 3 in. of polyiso insulation, OSB sheathing, a self-adhering waterproofing membrane, underlayment, and then, finally, reinstall the metal roofing.

“Is this the correct solution in addition to reducing interior moisture?” he asks. “Is an air gap necessary directly below the metal roof for efficiency gains? Will we smell the second layer of underlayment in the summer? How much more efficient is a light-colored roof than a darker (charcoal gray) roof?”

Those questions are where we start this Q&A Spotlight.

A roof guaranteed to fail

As built, writes Akos, the roof was bound to fail. The layer of Ice & Water Shield installed over the T&G roof is vapor impermeable, so winter condensation and mold should come as no surprise.

“The solution with exterior rigid insulation you are looking at…

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. nickdefabrizio | | #1

    I am no building expert so I will not comment on the insulation issue. However, it seems from the pictures that you have quite a bit of vegetation right next to your foundation/crawlspace that may be retaining a lot of moisture. Also, the A frame roof line is so steep it seems to direct a lot of rainwater right next to your foundation and I wonder if the water flows right over yor gutters because of the velocity flowing down the slick metal roof. I suspect you would need very wide gutters and leaders on the ground that effectively divert most of the rainwater far from the house. Given how steep the roof is, I wonder if on windy days rain just bounces off the metal roof and settles right next to the foundation?? And the valley between the two A frames deposits right on your deck so I imagine there is a lot of water deposited on or under the deck when the rain comes from this side.

    Again, maybe this is just the way the pictures look, but if not I would consider these issues as well as dealing with the insulation. When my house had moisture problems in the basement, I removed vegetation close to the foundation, attached downspoutso water moved away from the foundation and in one area installed some plastic tarps sloped away from the house buried just below ground cover so bulk rainwater moved away from the house...These were cheap solutions that reduced water intrusion into my foundation significantly. That can't hurt the overall mositure problem in the rest of the house.

  2. frankcrawford | | #2

    You can look at the Quik-Therm Matrix roof system, which combines EPS foam, vapour open films and integrated wood strapping in two layers to remove thermal bridging from the screws.

  3. charlie_sullivan | | #3

    First, a simple suggestion: If you want a low-cost way to get to whatever R-value target you choose, or a way to get a higher R-value from your budget, EPS or GPS is going to be cheaper than polyiso.

    If you are looking for code justification for less than R-38, I note that this is steep enough that you might be able to call it a set of walls.

    In fact, that's how it was working before, as a wall, vapor open without a completely waterproof layer. It's possible consider going back to something like that, which then could allow the use of different, cheaper insulation, although by the time you are done with setting up for that it's probably cheaper to use EPS or GPS.

    Kohta, that graph of dew point vs. height is very interesting! Is there a report that this is in or is this its first publication? Is there work on developing a model that can predict that? It seems like one less from that is that the practice of taking exhaust for HRV/ERV from high up is good practice, and perhaps we should put return ducts for whole-house dehumidifiers there too.

    1. Expert Member
      KOHTA UENO | | #5

      "Kohta, that graph of dew point vs. height is very interesting! Is there a report that this is in or is this its first publication?"

      The graph there is from a client report, but Joe Lstiburek wrote about this phenomenon here--scroll down towards the end.

      BSI-006: No Good Deed Shall Go Unpunished

  4. maine_tyler | | #4

    Would it be wrong-headed to think of a dehumidifier as capturing the latent heat from the (unwanted) moisture and depositing that heat indoors where it is needed-- Instead of venting it?
    Of course, there may be other reasons to just vent instead, such as other air quality issues.
    Perhaps the answer to this depends on the efficiency of the heating system the dehumidifier heat would be replacing? And I suppose it would depend on the efficiency of the heat recovery (if any) of venting.

    1. charlie_sullivan | | #6


      The answers to your questions are yes, it's smart to think of it that way, yes, there are other reasons to vent, and yes, which approach uses less energy depends on the heat system and dehumidifier efficiencies, as well as the indoor and outdoor temperatures and humidities, and the HRV characteriestics. I've run numbers and found that the dehumidifier can be attractive when you want some heat, and when the outside is high humidity and only a mild temperature.

  5. user-1020940 | | #7

    We are remodeling a 1970's A Frame, have stripped the old roofing off and are going back with all new materials; see drawing for detail. The tongue and groove is the existing structural ceiling on the interior.

    1. rlangley | | #8

      Hoping you, or someone, wont mind answering a few questions with regard to your drawing:
      1) In neither the air barrier or weather barrier have you identified a material...wouldn't the Polyiso (taped) be an air barrier?
      2) I am assuming that the tongue and groove decking is mounted on rafters (every four feet?) and that you will be using long (ten inch?) screws to secure the strapping and polyiso to the rafters? If I have that wrong, how are you securing the Polyiso to the building?
      3) Your 1x3" strapping appears to be horizontal...even if you use vented soffits in the rakes would this actually result in venting? I am just wondering why bother with the OSB?
      Roofers in my area (Ontario Canada) add strapping over a weather barrier which is over OSB or Plywood and then fasten the steel roofing to the strapping. Apparently the strapping does not rot in this it really makes me wonder what the value of the OSB is over strapping....and if protecting the strapping is not a concern then wouldn't taping the top layer of Polyiso act as a weather barrier?
      I am trying to understand the safest (performance) way to insulate my poorly insulated cathedral ceiling. Thanks

      1. Expert Member
        MALCOLM TAYLOR | | #9


        Just some thoughts. Take them with a grain of salt.

        1) I've never done this, but I wonder if it isn't easier to get the air-sealing out of the way at the t&g and not have to worry about taping the rigid foam, and worry about interior air moving along the underside and out the edges - or having to tape each layer off a ladder?

        2) Would it be worth the small hit in R-value to run three layers of foam (3", 1 1/2" and 3") with the second layer spaced by 2"x4"s on the flat. That shortens the length of screws necessary, and diminishes the chance of missing the rafters, and ruining the t&g.

        3) I agree. The strapping or sheathing is probably best as an either/or choice. Whichever you choose, I would still use a high-temperature sheet WRB as an underlay. Some attention also needs to be paid to how that strapping or sheathing is secured at the peak, as there is no backing to secure it to.

        1. rlangley | | #10

          Malcom Taylor,
          Thanks for your response. I have a question open here really appreciate your input as it is different from this thread in that the the interior is drywall.

    2. user-1020940 | | #11

      -The T&G is attached to the exterior of the A-Frame members and is structural.
      -There is no internal insulation, the T&G is the ceiling inside the A-Frame
      -We stripped off all of the old roofing and exposed the exterior side of the T&G.
      -We patched some rotted areas with T&G from a local architectural salvage company
      -We put down a layer of Intello-X on top of the T&G
      -We had to add a layer of plywood on top of the Intell0-X, we found the A-Frame was not square after removing the legacy roofing. It was a bit of an exercise using straps and come-along, etc. Adding the plywood "locked in" the roof to square so we didn't have out of square issues going forward.
      -We ran 2x4's vertically up the roof and screwed them through plywood into the A-Frame members.
      -We installed rigid poly-iso insulation between the 2x4's (vertically)
      -We ran a second layer of poly-iso horizontally and taped all of the seams, staggered seams.
      -We ran 1x4 battens vertically on top of the 2nd layer of insulation, screwed into the 2x4's, etc.
      -We ran 5/8th ZIP roofing horizontally up the roof and screwed it into the vertical battens/2x4's, etc. and taped the seams and fastener locations.
      -There is cora-vent at the base of the ZIP to give us a vertical air chase to the ridge, between battens.
      -New metal roof will be installed on top of the ZIP with vented ridge.
      -Structural engineer reviewed the assembly, specified fastener type, screwing pattern, etc.
      As you can see, we had to make some adjustments in the field. :-)

    3. user-1020940 | | #12

      I forgot to add:
      -We installed a Zehnder ERV and Mitsubishi Mini-Spits.
      -We upgraded the windows to triple-pane tilt-turn
      -We installed exterior venetian blinds.
      -A-Frame faces due South, so solar heat gain mitigation was necessary.

    4. Cavallaro | | #13

      Hi, thanks for the drawing detail. Can I ask how much you are spending and how many sq ft of your house or roof is for comparison? Also why are you adding the air barrier?

      1. user-1020940 | | #14

        1,700 square feet under roof with loft, main floor and basement.
        Close to $70K with taking off the old and putting on the new metal roof (labor/materials).
        The Intello X is a vapor retarder, not air barrier, in this application.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |