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

How to Build a ‘Perfect Wall’

Exterior foam will limit thermal bridging, but choosing between cellulose and spray foam can get contentious

Reduced thermal bridging

Rigid foam can be part of a 'perfect' wall system as long as it's thick enough for the climate. What to use for cavity insulation is another question.
Image Credit: FHB

Andrew Homoly is building a house in Kansas City, Missouri, and plans to use a “perfect wall” system consisting of 2×6 studs, Icynene open-cell foam insulation and an additional 1 1/2-in. layer of rigid foam on the building’s exterior.

His question in this Q&A post at GreenBuildingAdvisor concerns the best way to install window and door flashing, and where to install housewrap.

“Should the window flanges go over the foam at the window openings or should you line the window openings with a 2×4 so the window can be mounted into a solid piece of wood?” Homoly asks. “This would provide a more stable connection, but negates the benefit of the thermal bridging provided by the foam.”

Homoly is planning on using Huber’s Zip wall system under the rigid foam, and applying housewrap over the foam, an approach he recognizes as “a bit overkill” but reassuringly redundant.

The debate over how best to attach and flash windows in walls with exterior foam has been going on for years. What does the forum have to say? That’s the topic for this week’s Q&A Spotlight.

Sorry, this wall system isn’t ‘perfect’

For starters, Brett Moyer points out, there are three problems with the wall system Homoly is proposing, leaving it something short of perfect. He cites these shortcomings:

â— There is “zero moisture buffering capacity.”

â— The Zip System is expensive and unlike more traditional methods it relies on tape rather than overlapping layers of material to keep moisture out of the wall. Further, Huber hasn’t revealed what the perm rating of the material is.

â— There’s no need to use petrochemical foam where it isn’t needed.

Moyer seconds a suggestion from Armando Cobo, who recommends cellulose insulation in the wall cavities rather than foam. As does David Meiland, who says he sees “very little benefit to the foam, especially low-R-value open cell, with a lot of risk and cost if/when something gets wet.”

Cobo’s recommended wall would include a weather-resistant barrier (WRB) over the sheathing, then the windows, the rigid foam, a ventilated rainscreen and then the wall cladding. Or, Moyer adds, Homoly could save money and eliminate the flashing issues by using a double wall or a Larsen truss; either system can defeat thermal bridging in the same way that rigid foam does.

Why is cellulose better than foam?

Homoly, however, isn’t ready to abandon ship on the foam.

“It seems to me spraying a wet substance into a wall is just a bad idea in general,” he writes. “Surely the wall insulation will settle over time leaving gaps at the top (physics working against it). It won’t be able to fill cavities like an expanding foam. I would think newspaper will mold if it gets wet.”

Open-cell foam, on the other hand, seals cavities more effectively, repels water, and will permit drying if the wall gets wet, Homoly says. Although it doesn’t have the high R-values that closed-cell foam has, it’s more forgiving. “I would certainly pick cellulose over batts,” he adds, “but not over foam.”

Cellulose is preferable for a number of reasons, Cobo replies. It has a much lower global-warming impact than foam, should not settle if installed correctly, and it’s cheaper. In addition, cellulose can absorb some moisture at times of high humidity and release it when the air dries out.

Although foam makes sense in some situations — at the hard-to-insulate connection between floor joists and rim joist, for example — it’s not the best choice every time.

Further, Cobo says, the 1 1/2 in. of rigid foam Homoly is planning is a minimum; 2 in. would be better for the Kansas City climate. “The rigid foam outside of the sheathing and on top of the roof decking for a conditioned attic is for thermal bridging and to keep condensation to build on the inside of the wall and roof sheathing,” Cobo writes. “You also should install your wall and ceiling sheetrock with an Airtight Drywall Approach or ADA.”

Attaching windows and siding

One of the reasons Homoly isn’t eager to increase foam thickness is the difficulty of attaching siding, shingles, and trim. “I could see siding and shingles flying off much easier in high winds (I am in tornado alley),” he writes. “Unless there is some kind of excellent new technique for attachments through this amount of foam, I don’t feel comfortable beyond 1.5 in.”

Cobo wonders whether an engineer’s report would be required for attaching Hardie fiber-cement siding through foam any thicker than 1 inch. But GBA senior editor Martin Holladay assures him that the calculations for attaching furring to foam up to 4 in. thick have already been published by FastenMaster. “I think your worries are baseless,” he says.

As for installing windows in a wall sheathed with rigid foam insulation, Holladay points to several articles he’s written on the topic. They explain how windows can be installed in the same plane as the back of the siding (“outies”) or in the same plane as the sheathing (“innies”). Both methods can work, and Holladay offers step-by-step instructions in a related blog.

Our expert’s opinion

We asked GBA advisor Michael Chandler for his take on this question. Here’s his response:

The important thing here is for Andrew to feel comfortable in his house. If he intuitively feels that more than 1 1/2 in. of foam would leave him vulnerable to wind damage, then he should stay with a system that makes him feel comfortable. But in my opinion, he should think through a few other issues.

First, he doesn’t need to provide more than 1 1/2 in. of attachment for his window’s nailing flanges, so “outie” windows can hang on 2x2s screwed to the sheathing. Doing this doesn’t “negate” the thermal break at all the studs and rim joists, and you already have the thermal break of the window frame that you are accepting as part of the cost of the solar gain and daylighting.

But if he is planning on wrapping the house with OSB, then I would recommend that he question the Huber ZIP and just use regular OSB and housewrap, nail the window flanges to the OSB, and do the regular housewrap detail. Then he can build up exterior jamb extensions and run the foam and tape those seams if he likes — or not. That’s much easier than using flex wrap to get out onto those 2x2s.

As for the foam versus cellulose isse, I think it’s a good idea to question spray foam in this assembly. But damp-spray cellulose isn’t the only alternative. You can also do a blown-in-batt approach with a scrim and dense-packed (dry) cellulose or dense-packed (dry) Spider micro-filament fiberglass, either of which has the same R-value as open-cell foam.

I prefer the Spider because it doesn’t hold water in the event of a bulk water leak (I’m in hurricane alley, so I relate). At Johns Manville’s Spider website they are all about promoting their damp-spray application but the scrim and dense-packed Spider is cheaper and better in my opinion.

Finally, I’m a big fan of double-wall construction. it’s surprisingly inexpensive, storm-worthy, and I think it would be good for your locale. But the extra-thick walls are an acquired taste aesthetically, and if that’s not your thing, the 2x6s with 1 1/2 in. of foam and “innie” windows is a great, if perhaps not perfect, solution.


  1. Glen M | | #1

    Alternate Sheathing Option
    Excellent comments.

    I agree the Zipwall system doesn't seem to make sense in this application as well considering the Spider. Although we have been using the Spider with good success, Andrew may consider "equal" products such as AsureR Plus, Jet Stream Ultra, or Propink L77. Cellulose is fine too (I have it in my house) and the concern over it holding moisture can be overblown, notwithstanding hurricane zones, although that doesn't hold much creditability for me either since how often does that occur and how much can be saved/prevented in the aftermath is a big unknown.

    Andrew didn't comment on what siding product is being planned, but does imply it is a nailed-up siding such as fiber-cement or similar.

    One other option to consider is structurally insulated sheathing. I am only familiar with Dow's product (SIS) which comes in 1/2 and 1-inch thickness, so R-value is less, but eliminates the wood sheathing and simplifies window/siding installations.

    I would have a hard time promoting a double-wall system in this application. Yes it has all these great benefits, but it costs more and uses more resources. In this climate I just don't think it can be justified from a cost-benefit perspective. If you are looking to increase the R-value, use the rigid with closed-cell foam.

    If you use the 1 1/2" rigid, I would go with an "outie" design with the 2x2s (as noted). In addition to ease of installation it gives you a nice "deep" window well.

  2. Michael Roland | | #2

    Rock Wool Rigid Board and Batts
    An alternative to exterior rigid polyiso as a thermal break are rock wool boards. Two mfgs. I'm familiar with are Roxul and Thermafiber. They both make batts as well, which have a slightly higher R value than fiberglass and allow less air circulation. They're sometimes difficult to get hold of since they're used mostly in commercial construction. But they can be found. Lowe's is one source for Roxul.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |