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
Musings of an Energy Nerd

Rethinking the Rules on Minimum Foam Thickness

Exterior rigid foam doesn’t always have to be as thick as the rules recommend

How risky is “too thin” foam? In theory, builders should make sure that any rigid foam installed on the exterior side of wall sheathing is thick enough to keep the sheathing above the dew point during the winter. But in practice, most walls with “too thin” foam won't fail.
Image Credit: Image #1: Fine Homebuilding

When builders ask for advice about installing rigid foam on the exterior side of a wall, I usually refer them to one of my articles, “Calculating the Minimum Thickness of Rigid Foam Sheathing.” The article explains that the R-value of the rigid foam layer needs to be high enough to keep the OSB or plywood sheathing above the dew point during the winter. For example, a house with 2×6 walls in Climate Zone 6 would need rigid foam with a minimum R-value of R-11.25.

Here’s the reasoning behind the advice: thinner foam would be risky, because the thinner foam doesn’t have a high enough R-value to keep the sheathing dry during the winter — but is still thick enough to limit the rate of outward drying during the spring and summer.

So far, so good: Just follow the rules and you’ll stay out of trouble.

Recently, however, building scientist Joseph Lstiburek — the guy who explained these rules to me in the first place, a decade ago — has been stirring the pot. In a 2016 article called “Doubling Down,” Lstiburek wrote, “I grew up with this wall in Ontario [Climate Zone 6] — 2×6 wall with R-5 foam sheathing coupled with unfaced fiberglass batt cavity insulation covered with an interior 6 mil poly air/vapor barrier. This wall is currently being built all over Minnesota and Wisconsin. It is still being built in Ontario. And it works. We know that it works because we have been building it for so long without problems.”

In this article, I’ll explore the following question: If a 2×6 wall in Climate Zone 6 with R-5 exterior rigid foam works well, why have I been advising Zone 6 builders to install foam with a minimum R-value of R-11.25?

Three relevant issues

There are at least three factors to explain the disparity between my…

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. charlie_sullivan | | #1

    flashing leaks
    In addition to the considerations mentioned, people considering an interior vapor barrier plus thin exterior foam should consider how confident they are in their window flashing. Water that gets in through a flashing leak will lead to persistent dampness if drying is impeded in both directions. Poor planning and poor execution of flashing are both rampant problems.

  2. Expert Member
    ARMANDO COBO | | #2

    ASHRAE Fundamentals or WUFI
    Learning to calculate wall and roof assemblies condensation through ASHRAE Fundamentals or WUFI, teaches you that it’s not just about the ratio of exterior and cavity insulation with in a climate zone, but also a correlation between R-values with indoor and outdoor temperature and relative humidity. I believe that the Code (2015 IRC R702.7.1) and any rules of thumb can be misleading if not completely understood.
    If the Temperature and/or RH on the inside of your house is maintained much higher than your neighbor’s house (assuming both houses are built the same), the amount of exterior insulation will need to be higher, regardless of your climate zone. I don’t know many homeowners that are taught by their Architects/Builders/Realtors/HVAC Contractors the importance of maintaining proper temperature and RH in their houses, in relation to how the house was built.

  3. Expert Member

    Building Science
    Joe's work, and that of John Straub have been game-changing in the way in which we look at and build houses. I do wish they were a bit less strident about their preferred methods and knowledge at any given time though. The whole "you are an idiot if you don't build our way" stuff isn't always helpful.

  4. Jon_R | | #4

    > keep the sheathing above
    > keep the sheathing above the dew point all winter long

    Picking R19 insulation in a stud wall and R11.25 of exterior insulation, we get 37%.

    Pick a simple temperature of 0F (zone 6 gets much colder than this).

    .37 * 70F = 26F

    70F air at 30% RH (lowest recommended level) has a dew point of 37F, much higher than 26F.

    In this generous example, R11.25 of rigid foam fails to keep the sheathing above the dew point all winter long.

  5. Expert Member
    Dana Dorsett | | #5

    It's the mean winter temp, not the peak low @ JonR, Martin
    "...keep the sheathing above the dew point all winter long..." is clearly wrong. It's the mean temp at the sheathing over weeks/months that matter. Wood sheathing can take on quite a bit of moisture as adsorb before it becomes problematic, and cycling moisture into & out of the sheathing when it's cold as the temperature varies doesn't really matter much.

    Five perm paint (standard latex on gypsum board) limits the magnitude of the hourly, daily, weekly moisture load swings in the sheathing, but if the mean temp of the sheathing over say 10-12 weeks is sufficiently below the dew point of the conditioned space air enough moisture can accumulate to become a problem once warmer mold-growing temperatures arrive.

    While 0F events (or even 0F daily highs during a cold snap) are common in US climate zone 6, mean January temps in zone 6 are always above 10F, and the mean temperature of the 10 coldest weeks is usually north of 20F, even at the cool edge of zone 6, warm edge of zone 7, such as Duluth, MN:

    Get a bit deeper into zone 7 and the winter temperature averages are lower, requiring more exterior R for dew point control:

  6. Expert Member
    Dana Dorsett | | #6

    The argument against active humidification @ Armando Cobo
    "If the Temperature and/or RH on the inside of your house is maintained much higher than your neighbor’s house (assuming both houses are built the same), the amount of exterior insulation will need to be higher, regardless of your climate zone."

    That is SO true!

    I can't count the number of times I've had to tell people to get rid of their whole house Aprilaire on their hot air heating systems (or at least turn it down) to avoid the springtime mold or wintertime window condensation problems (or how much push-back I get!) Some seem convinced that 50% relative humidity is the "right" number all year round, and anything less isn't healthy or comfortable.

  7. Jon_R | | #7

    If air can flow from the
    If air can flow from the interior, to the exterior sheathing and then back to the interior, then large amounts of moisture could be deposited during periods where the sheathing is below the dew point - independent of interior side perms. An interior air barrier plays a role in wall moisture.

    As Lstiburek says, building pressure is also a factor. Tripling the average pressure is similar to moving RH from 30% to 60%. Keep Winter monthly average pressure negative and I expect that some highly non-recommended walls would work well.

  8. Expert Member

    I agree, and that brings up an interesting question: How resilient do you make a house to the behaviour of the occupants?

    There must be a sweet spot between designing a house that can function whether the occupants control the humidity and pressure or not, and one that works only if the occupants carefully maintain certain conditions.

    A corollary question is to what extent should a house rely on mechanical systems for the robustness of its basic structure?

  9. Jon_R | | #9

    I think that control of
    I think that control of important parameters should be automated, with alarms for when they malfunction. But this isn't new ground. Consider:

    crawl space exhaust fans and dehumidifers
    bathroom/shower fans
    radon and HRV fans (important even if not a structure issue)
    accidentally sealing off a furnace return (causing a perhaps 10x increase in room pressure and much more wall moisture)
    FPS foundations that rely on building heat

  10. Jon_R | | #10

    Interesting and seldom
    Interesting and seldom discussed is Lstiburek's comment "If the outside is “leakier” than the inside we reduce the risk". I interpret this as:

    a) the common practice of taped exterior sheathing or foam with no interior side air barrier increases risk.

    b) As Martin writes, a smart vapor retarder and interior side air sealing would offset the need for some unknown amount of exterior foam (ie, maintain the same risk level).

    I'll add that somewhat vapor permeable external insulation (eg, EPS) is lower risk than impermeable (eg, foil faced). Perhaps another offset here.

    It would be great if there were a simple, proven rule such as "with a smart vapor retarder and interior side air sealing, one can reduce the recommend levels of foam by 50%". But until then, perhaps a simple approach for anyone wanting to use less than the recommended thickness of foam is to put it on the interior side. Tape it. Keep the exterior side perms high.

    Even if one uses the recommended exterior thickness (IMO, this isn't infallible), still consider the other factors that further reduce risk.

  11. Expert Member
    MALCOLM TAYLOR | | #11

    Addressing the downside of relying on automated controls with alarms sounds like a really good idea.

  12. mangler66 | | #12

    Is a foam with perm rating of
    Is a foam with perm rating of 3.48 enough to make a difference and help moisture migrate to the outside? The assembly is a R-19 2x6 wall (R-19) with 2" R-10 foam in Zone 5, so it should be "safe" regardless, but it would be nice to have extra safety margin. According tho the manufacturer the 2" EPS foam is foil faced with micro-perforations, making it a decent water barrier while still being vapour permeable.

    I investigated rock wool insulation, as I know this is safe (extra permeable), but it is prohibitively expensive in my area vs EPS.

  13. mercifullyfree | | #13

    Laws of physics don't change
    Thank you for what in IMHO is a balanced examination of the issues surrounding this construction detail and it's effects. FWIW, I think Joe L and John S perceived stridency is probably born of trying to stem the "tide" of incorrect "good-ole-boy" construction traditions and the resultant push-back. I laugh and tell clients that I know all the wrong ways to do things, being brought in after things "break." So I applaud the guys out there trying to build them right and participating in discussions like this. Thank you!

  14. Expert Member
    Dana Dorsett | | #14

    The laws of physics are also self-enforcing... @ Bruce
    ... and there's no enforcement discretion that mere humans can apply.

    Now if only some of the folks making policy had a clue about that...

    'nuff sed. ;-)

  15. Jon_R | | #15

    breathable is better
    In a cold climate, 1.74 perms (3.48/inch @ 2") to the exterior is better than say .03 perms to the exterior. In your "safe" case (although R10 is suspect), it might be reasonable to use WUFI to quantify the difference.

    This isn't well researched, but I interpret page 9 here to suggest that if you have less than 1 perm on the interior side (eg a smart retarder) and > 1 perm to the outside (eg <= ~2" of EPS and the right sheathing), then any amount of external foam is safe. Similar to any amount of Roxul being OK on the exterior.

  16. GBA Editor
    Martin Holladay | | #16

    Response to Dana Dorsett (Comment #6)
    You're right that I shouldn't have used the phrase "keep the sheathing above the dew point all winter long," since the phrase is technically inaccurate. Thanks for keeping me on my toes. I have edited the article to reflect your point.

  17. _Stephen_ | | #17

    Roxul Comfortboard?
    Am I wrong here, or would Roxul Comfortboard not work way better for this type of application? Or, at least, it's a lot more vapor permeable. If you're covering it with an air barrier like tyvek it would give the wall a lot of drying potential to the exterior without having to be worried about ensuring the wallboard has sufficient insulation to keep it above the dew point. You can get your thermal break cake, without having to eat the cost and complexity of 2 layers of 1.5" of EPS.

  18. GBA Editor
    Martin Holladay | | #18

    Response to Stephen G
    Roxul Comfortboard (or other brands of semi-rigid mineral wool) is neither better nor worse than rigid foam. It's different.

    Here's what I wrote in one of my articles, How to Design a Wall:

    "Mineral wool insulation can be substituted for rigid foam insulation on the exterior side of wall sheathing. One advantage of mineral wool over rigid foam: because mineral wool is vapor-permeable, it doesn’t inhibit wall sheathing from drying to the exterior. That means that builders can install mineral wool of any thickness on the exterior side of their walls. You don’t have to worry whether exterior mineral wool meets any minimum R-value requirement. (Of course, thicker insulation always does a better job of resisting heat flow than thinner insulation.)"

    It's true that you can use mineral wool of a lower R-value as continuous exterior insulation than is required for rigid foam. But if you go that route, remember that the sheathing will still be absorbing moisture in January and February. While the mineral wool will allow the damp sheathing to dry in April and May, it would probably be better to install a thick enough layer of continuous exterior insulation to prevent the moisture accumulation in the first place.

    In other words, if you want to keep your sheathing dry, it's best to follow the minimum R-value rules, whether you are using rigid foam or mineral wool to provide the continuous exterior insulation. And rigid foam has a big advantage over mineral wool: it's less squishy, so it's much easier to install furring strips that are co-planar.

  19. charlie_sullivan | | #19

    Winter performance with thin mineral wool

    While I agree that following the minimum R-value rules with mineral wool will give more margin on keeping the sheathing dry than a thinner layer of mineral wool would, and in that sense it's best, I think your second to last paragraph above is a little overly pessimistic. Compared to s simple old-fashioned wall with sheathing on the outside, any thickness of exterior mineral wool will only reduce the mid-winter moisture accumulation in the sheathing. That's in contrast to a thin layer of foam, which can make things worse compared to no exterior insulation.

  20. GBA Editor
    Martin Holladay | | #20

    Response to Charlie Sullivan
    I agree with your analysis. Thin mineral wool is better than no exterior insulation at all, and not quite as good as mineral wool (or rigid foam) that meets the minimum R-value levels designed to prevent significant moisture accumulation.

  21. gingging65 | | #21

    I am a homeowner to be, hoping, praying and scheming ways to get a new house built in these crazy times. But I digress. I am excited about the 2021 IRC updates (which my state, Connecticut has adopted) because they significantly increase the wall R-value to 30 or 20+5ci (much better than the previous R-20). But since I have memorized Martin’s tables from the aforementioned 2010 “Calculating…” article, the new code makes me nervous that builders will default to R-5 rigid, and a lot may simply switch from the now ubiquitous Zip sheathing to Zip-R, which I have read about on GBA as not being appropriate for Zone 5. I’d love to see this issue specifically addressed by GBA, as it applies to the new code, and material options, so I can go to my builder with the scientific backup I might need in order to convince him.

  22. MartinHolladay | | #22

    Ging Ging,
    Q. "I’d love to see this issue specifically addressed by GBA."

    A. It has been. See these two articles:

    "The 2012 Code Encourages Risky Wall Strategies"

    "Vermont Addresses the 20+5 Wall Problem — Sort Of"

    1. gingging65 | | #23

      Thanks Martin. I had seen the first article, and then found the second one right after posting today. GBA has such a wealth of information, it can take a while to get through it all — a great “problem” to have. Thanks!

      1. MartinHolladay | | #24

        Ging Ging,
        You might also want to read the following article:

        "Updating Vermont's Energy Code" (Note that there is a section near the end of the article with this heading: "The '20+5' issue.")

        Note also that the 2021 version of the IECC attempts to address the 20+5 issue by changes to the requirements surrounding vapor retarders. This is a contentious section of the code, and the code is not well written. But if you want to know more about the issue, see "Building Codes Update Vapor Retarder Requirements." In that article, I wrote:

        "The new code attempts to address this issue—the need for inward drying when walls have exterior rigid foam—by adding two confusing footnotes.

        "Footnote (b) applies to Class I vapor retarders in Climate Zones Marine 4 and Climate Zones 5 through 8. According to footnote (b), “Use of a Class I interior vapor retarder [for example, polyethylene] in frame walls with a Class I vapor retarder [for example, foil-faced polyisocyanurate] on the exterior side shall require an approved design [in other words, a design using accepted engineering practice for hygrothermal analysis].” Presumably, if builders want to avoid the expense of hiring an engineer to perform a hygrothermal analysis, they won’t use polyethylene on walls with exterior rigid foam.

        "Footnote (c) applies to Class II vapor retarders installed in Climate Zones 3 through 8. According to footnote (c), “Where a Class II vapor retarder is used in combination with foam plastic insulating sheathing [rigid foam] installed as continuous insulation on the exterior side of frame walls, the continuous insulation shall comply with Table R702.7(4) and the Class II vapor retarder shall have a vapor permeance of greater than 1 perm when measured by ASTM E96 water method (Procedure B).” This convoluted language—the language referring to Procedure B—is a very roundabout way of prescribing a vapor retarder with variable permeance—in other words, a so-called smart retarder.

        "In simpler language, footnote (b) and (c) say this: “If you install a continuous layer of exterior rigid foam in Zones 3 through 8, use a smart vapor retarder on the interior, or else you’ll have to hire an engineer to perform a hygrothermal analysis that gives the thumbs up to your plan.”

        "In essence, the code defines a smart vapor retarder as a material which passes two tests: (1) when tested by the dry-cup method (ASTM E96, Procedure A), it is categorized as a Class I or Class II vapor retarder, and (2) when tested by the wet cup method (ASTM E96, Procedure B) it has a vapor permeance greater than 1 perm. Clear as mud."

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |