I have a question about insulation plan recommendations. This will be for a new construction 2-story house in climate zone 6 (southern Minnesota). My question is specific to the above-grade (main and second) levels of the house. (For the foundation, I am leaning toward ICF at this point.) Reading here, on GBA and BSC, and elsewhere I recognize the benefits of external (rigid foam) insulation, but I also understand it must be thick enough for the climate zone to avoid the risk of reduced wall-assembly drying capacity. For our zone, I would go conservative and want 2.5" of XPS minimum. The thing about this that gives me pause is the tricky nature of getting all the finishing details right, not to mention the increased material and labor cost of even attempting to get them right. Sheathing will be plywood instead of OSB (owing to my moisture paranoia). WRB will be high quality -- ideally #30 felt, or Tyvek or R-Wrap with careful attention to correct installation. Exterior cladding will likely be steel lap siding, or possibly the thickest vinyl we can get. I considered fiber cement but opted against it. I am also hoping to use a rain screen technique between the cladding and the WRB. If we forgo exterior foam on the sheathing, I am leaning towards spray foam on the interior walls (all 2x6). So to get our code-required R value, either a full 5.5" of open-cell or 3" of closed-cell. (If we can find a contractor who is experienced and good with blown-in wet cellulose, I would also consider that, but cellulose does not seem to be popular in our region, which is unfortunate.) So assuming a choice between open- or closed-cell in the wall cavities, I like the vapor permeance (and increased drying potential) of open-cell foam. Would a thicker application of 5.5" of open cell still have significantly better drying capacity than 3" of closed cell? Or would it be a wash at that point? Cost would not be a major criterion in choosing between the two, nor would closed-cell's benefit of adding structural strength to the walls. Thank you for any advice.

PS. Apologies for the long paragraph without line breaks. I may have a plug-in that is messing things up in my browser.

I see I can resort to hard-coded markup tags and will do so in the future.

First, don't use XPS. The R5/inch is partially due to the HFC134a blowing agent, which is a powerful greenhouse gas (~1400x CO2), and as it loses that blowing agent over a handful of decades performance drops to about R4.2/inch, the same as EPS.

EPS and polyiso are blown with pentane (about 7x CO2), and are cheaper. Polyiso as the outermost layer has performance issues in your climate, but if you stack a layer EPS to the exterior of it, performance is preserved. A 1" layer of polyiso on the exterior would perform at about R5.5 in mid-winter if you installed 1.5" of EPS on the exterior. The EPS gains performance at colder temps, whereas polyiso falls off a cliff when the average temp through the polyiso drops below freezing. The labeled R of 1.5" EPS is R6.3, but when it's cold out it'll be doing better than R7. The labeled R of 1" polyiso is R6, or R6.5, but it'll average about R5.5 if kept warm by the EPS blanket. (If you swapped the layers the 1" polyiso would average about R3 in January in your climate- it's that bad if you let it run too cold!)

So performance-wise 1" + 1.5" stackup you'd be at or above R12 for the foam layers under all conditions, just like the labeled R of 2.5" XPS. But unlike XPS, it won't lose that performance over time, and it does a lot less damage.

Closed cell foam is almost as bad as XPS, since it's blown with HFC245fa (about 1000x CO2). Worse yet with only 3" of stud as the thermal bridge instead of 5.5" an R20 closed cell solution dramatically UNDERperforms an R20 full-fill solution of open cell foam or cellulose. Ignoring the R-value of the exterior foam layers, it's the difference between ~R10.4 whole-wall-R, and ~R14.4 whole-wall-R due to the higher thermal bridging of only 3" wood vs. 5.5". The thermal performance difference is the equivalent of a full inch of EPS(!).

And yes 5.5" of open cell offers dramatically faster drying than 3" of closed cell- about an order of magnitude faster! (About 5-7 perms instead of 0.3-0.5 perms for 3" ccSPF) As long as you have the insulating sheathing you won't need to do anything other than latex paint on the wallboard as an interior side vapor retarder, and it'll have very good drying capacity.

Dana, Thank you kindly! My SPF preference for wall cavities, assuming cellulose is not practical owing to lack of an experienced installer, was 0.5 pound open-cell -- for the environmental reasons you mention, among others. Hearing your perspective solidifies this choice.

If our contractor (not yet chosen) has experience with exterior insulation and we're confident they can get the finishing and flashing done correctly with foam outside the sheathing, I would like to go that route. And we'll use EPS instead of XPS! Above-grade exterior foam is not common in our area, but it seems a good, conscientious builder could readily get the detailing right.

There are lots of installation details for exterior rigid foam in both the text and hyper-texted links in this bit o' bloggery. 

A key decision early on is whether to go with "innie" vs. "outie" windows, since that determines which layer the housewrap needs to be on in order to lap correctly with the window flashing.

You could use a mineral fiber exterior insulation sheathing like Roxul Comfortboard. It is vapor open so you don't have the moisture concerns that you have with foam.

In my neighborhood rigid rock wool is ~2x the cost of rigid foam per R. YMMV.

But it's also fireproof, and has many good attributes relative to rigid foam. If "cheating" the IRC prescriptives on exterior R for dew point control by very much it may be "worth it", but installing a smart vapor retarder on the the interior to mitigate insufficient dew point control would usually be cheaper.

I'll research it and discus with our contractor once we hire one and get to that point. My guess is rock wool is a peculiarity in residential construction, if heard of at all, in our area. If we do go with insulation outside the sheathing, I expect we'll use EPS (type I sufficient for exterior sheathing, or worth upgrading to type VIII or II?).



And yes, absolutely will use MemBrain or an equivalent smart vapor retarder for any area where local building code requires it.

Type-I EPS (1lb per cubic foot nominal density) is a bit fragile- it's easy to ding the corners off, etc. It's often sold with aluminum or plastic facers, which would limit breakage, but would also reduce/elimnate any drying to the exterior. Typical R/inch is about 3.85- 3.9.

Type-II (1.5lb density) noticeably more rugged in handling, usually sold without facers, and has a higher R, about R4.15-R4.2 per inch. At 3"/R12.5-ish it's still about 1-perm, the boundary between Class-II and Class-III vapor retardency, and will not create a moisture trap. It CAN dry toward the exterior, seasonably but slowly.

Type VIII is 1.25 lbs nominal density and about R3.9-R4/inch, but not dramatically more vapor permeable than Type-II.

Type-II is probably your best choice for an all-EPS stackup on the foam. When the average temp through the EPS layer is 40F or lower it will perform at better than R4.5/inch, so as exterior insulation even 2.5" would get you to the bare-minimum IRC prescriptive for dew point control in climate zone 6, and 3" would have real margin.

Yep rockwool sheathing is peculiar and unheard of in my area too. I had to special order it through a Roxul dealer. Everyone thinks I am a weirdo. But i am very happy with the results and would do it again on the next project.