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

Attempting a cost-effective roof insulation plan – will it work?

dhammond | Posted in Green Building Techniques on

Hey guys, this site has been awesome.

I am building a modern home in Chattanooga (zone 4) with a low slope metal roof. Pitch is 1/12, unvented assembly with 2×12 rafters 24″ OC. I had more extravagant plans to build up polyiso on top of my roof deck but I’ve had to scale back a bit. Here’s where I think I’m landing with the assembly, from inside out:
– 5/8 drywall ceiling
– R-30 fiberglass batts
– 2.5″ closed cell foam against roof deck (vapor barrier)
– 7/16 OSB
– ice and water shield
– R-5 3/4″ polyiso, taped
– metal roof applied directly over foil face of polyiso

Any suggestions or tweaks to this? I initially wanted to hit R-60. This will hopefully get me close to 50. I punted on more polyiso above the deck due to costs associated with the polyiso itself and the labor + materials tied up in a purlin system for the metal roof. Will this assembly hit the important aspects of controlling air/vapor/water while also achieving a higher R value? Most of my peers want me to throw R-38 fiberglass in the rafters and skip the rest.

I have confirmed that my polyiso and roof are compatible with this type of installation.

Thanks for your help.

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

    In Zone 4A, you want a minimum of 31% of the total R-value of your roof assembly to be in the form of some type of foam insulation (assuming you are combining foam insulation with fluffy insulation in an unvented assembly).

    Your planned assembly has a total R-value of about R-51, with 41% of the total R-value in the form of polyiso plus closed-cell spray foam. So this assembly is safe from a moisture perspective.

    With the closed-cell spray foam under the roof sheathing, you have a good air barrier, so you could save money by skipping the Ice & Water Shield. Instead of the Ice & Water Shield above the roof sheathing, I advise you to install a layer of synthetic roofing underlayment between the polyiso and the metal roofing.

    Some metal roofing manufacturers might advise the installation of 1x4 or 2x4 purlins above the polyiso -- whether or not to install the purlins depends on the type of roofing and the preferences of your roofer.

  2. dhammond | | #2

    Thanks Martin! I would love to save money and time by omitting the ice & water shield. My thought on using it was for sealing around the roof screws and to give me a secondary guard against water intrusion. But, I'd be happy to go without it if it seems like the synthetic underlayment, taped polyiso, and properly installed metal roof will provide enough insurance against water coming in. If my pitch wasn't so low, I wouldn't even consider adding a self adhered membrane. I've talked to a number of roofers in the area and there's no consensus on whether to add ice & water or just go with synthetic underlayment. Seems like many of them like it for the extra insurance but most agree that it's overkill - especially down south where the product isn't used very much (compared to the great white north).

  3. NateSc | | #3

    David I just want to mention that 1/12 is very shallow for a metal roof. Manufacturers will typically recommend 3/12 - check the installation manual.

  4. ChrisInCR | | #4

    It was my understanding that doing some sort of impermeable material on either side of the roof decking was a bad idea (the polyiso on one side and closed cell on the other)? In the event the decking does get wet, where will it dry to?

  5. dhammond | | #5

    Chris, your caution is granted. As far as I understand, you have to be absolutely certain that the roof deck is dry. This includes monitoring the moisture level of the deck before spraying the foam. I can't find it now, but there was a good discussion on GBA on this topic. The case study below was posted in that thread. It details how to use foam on the underside of the deck.

    Here in Chattanooga, spray foam installers tell you not to use spray foam against the deck but a lot of people do it. It's really up to the roof installation to make sure the deck stays dry. It's a gamble for sure.

  6. Expert Member
    Peter Yost | | #6

    Hi David -

    In this previous Q&A I addressed low-slope standing-seam roof detailing:

    Read down a bit.


  7. user-6988032 | | #7

    I have a similar question:

    Working on an unvented roof detail in climate zone 5. I have an 8/12 roof with 2x12 rafters. Would like to use 4" of continuous EPS of on the exterior and fill the cavities with mineral wool or fiberglass with Cerainteed Membrane. I'd like to install horizontal strapping above the foam to facilitate the installation of metal roofing. Do I need more exterior foam? Not sure that it's financially possible in this case. Might be possible to work some venting in with the horizontal strapping.

    1. GBA Editor
      Martin Holladay | | #8

      This article explains the math: "Combining Exterior Rigid Foam With Fluffy Insulation."

      You're proposing R-16 rigid foam and R-41 of fiberglass, for a total of R-57. You end up with 39% of the total R-value in the form of rigid foam.

      [Later edit: Oops! Math error. I meant to write, "You're proposing R-16 of rigid foam and R-41 of fiberglass, for a total of R-57. You end up with 28% of the total R-value in the form of rigid foam -- not enough, since you are aiming for a minimum target of 41%."]

  8. user-6988032 | | #9


    I don't quite follow your numbers in the article linked vs. the numbers listed above. I could use r38 fiberglass with r16 eps, 16/38 is 42%, correct? The total assembly would be R54. If I multiply R54 by .41, according to the table, I'd need more like R22 eps and R32 fiberglass. Where is my math wrong?


    note: my BA was in writing!

  9. GBA Editor
    Martin Holladay | | #10

    I did, indeed, make a math error. See my corrected comment, above.

    If you choose to install R-16 EPS and R-38 fiberglass, you'll have slightly different numbers that I assumed. (I assumed 11.25 inches of fluffy insulation at R-3.7 per inch.) Here's the calculation for your latest proposal: Total R-value of R-54, with only 30% of the total assembly R-value coming from the EPS -- and that still won't work, since you are aiming for a minimum target of 41%. You need thicker EPS or less fiberglass.

  10. user-6988032 | | #11

    OK! Yes, that's a bit off. Any way to make the system work with a combination of foam and fiberglass within the cavities? I only have enough 4" EPS for one layer over the sheathing- 5/8 Zip. We have 4" EPS on the walls, too, and cut the rafter tails flush with the wall framing so that the EPS is continuous.

    I can possibly vent the assembly by strapping vertically, underlayment would be installed over the strapping.

    1. GBA Editor
      Martin Holladay | | #12

      Q. "Any way to make the system work with a combination of foam and fiberglass within the cavities?"

      A. Yes. That approach is called the flash-and-batt approach. Here is a link to an article with more information: "Flash-and-Batt Insulation."

      It's even possible to use a hybrid approach, with some rigid foam above the roof sheathing, supplemented by additional closed-cell spray foam on the underside of the roof sheathing. If you do that, you can add up the R-value of the rigid foam and the spray foam, using their combined R-values to determine whether you've hit your percentage target.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |