We are building a new house and of course I have loads of questions...my first relates to wall structure. I live on the southwest coast of BC, Canada. What Americans would call the Pacific Northwest. Winters here only drop below zero (Celcius) maybe a couple weeks out of the year, otherwise hover around or above zero. But they are pretty humid. I live in a micro-region with very warm summers, often reaching 30 degrees or higher. So my heating needs are not heavy, and while I can live without air conditioning quite happily, I wish to avoid our current problem of a house that is unbearably hot at the end of a long hot day (due to lack of proper insulation). I'm looking to use Roxel mineral wool batts for insulation. I've been researching wall structures and I'm currently considering a 2x6 wall 24" oc, with an exterior truss wall (Larssen type) with another 6" of roxel batts on the outside of the plywood sheathing. According to the manufacturer, this would eliminate the need for a vapour barrier b/c the dew point would not be reached at the sheathing. However, to do this is quite expensive and I'm not sure the increased insulation is necessary, given our mild climate. So my question is sort of two-part...first, how much extra insulation do I really need in my walls given our mild climate, and second, is there some way to achieve this and have walls that dry in either direction to avoid a vapour barrier? TIA

I can't give you the amount of insulation you'll need, but as a general rule you are far better off spending your money on the house "envelope" than for a more expensive heat/AC/dehumidification system and the continuous ongoing expense of running it. A better envelope = more comfort for less cost.

Put just rigid foam and diagonal steel straps on the exterior and you won't have any moisture sensitive exterior sheathing to worry about. Use EPS or unfaced polyiso and the wall can breath/dry in both directions. If code doesn't allow straps, then perhaps other non-wood structural sheathing or structural sheathing on the inside (next to the wallboard).

Rain gap, good air sealing (two barriers outperform one), cellulose and overhangs are also important.

Using Building Science Corp's handy cheat-sheet, at a whole wall R of R25 you're approaching the financially rationality balance point.  See the zone 4 row of Table 2, p.10.

An 2x6 24" o.c. R23 Roxul wall with 1/2 CDX or OSB sheathing and half-inch wallboard comes in at about R14-R15.  If you add 2" of foil-faced polyiso or  3" of rigid Roxul ComfortBoard on the exterior of the sheathing you'd be at R25, and you would have HUGE dew point margin at the sheathing for being able to skip the interior vapor barrier.   According to the IRC prescriptives you only need R3.75 on the exterior of 2x6 walls for dew point control in zone 4C, and you'd have more than 3x that much. 

From a dew point control perspective you could cheap-out and only install 1" or 1.5"  of exterior polyiso, but the installation labor is the same at 1" as it is at 2".   Even though R3.75 is the minimum, it's much better to buy some margin- don't go any thinner than 1" polyiso, or 2" rock wool. (Polyiso is a LOT cheaper per R than rigid rock wool in my neighborhood- YMMV.)

That's a lot simpler & cheaper to build than a Larsen Truss. The R23 x 2 Larson Truss wall would run about R28-R30 whole-wall. It would in fact do the trick, but it is probably past the financially rational point for wall performance due to the high cost of the complexity of building with trusses.  You'd be better of spending the extra money on other performance aspects- like heating with high efficiency mini-splits (as recommended on your other thread).

See:  http://www.greenbuildingadvisor.com...rior-walls

http://www.greenbuildingadvisor.com...-sheathing

http://www.greenbuildingadvisor...side-house

A passive solar house will overheat like crazy if you're not VERY careful about window type, size & orientation, so be careful when building in passive gain.  Getting rid of west facing windows entirely can take quite a bit off the peak cooling load, since it's hard to shade from the low evening sun angles, and the gain is occurring late in the day after the air temps have heated up and the exterior of the house has been sun-drenched and well above the ambient air temp most of the day.

You guys are great, thanks so much.

Bob, I agree with you 100%, which is why I'm heavily researching wall structure. But I know at some point there is diminishing returns on adding more R, so I'm trying to figure out where that is for me.

jonr, I did not know that you could use rigid foam as a sheathing material. But I never though to use anything other than plywood, so your advice about using a non-wood sheathing product is very interesting. We will definitely have a rain gap and a big overhang (planning on 24").

I read the Building Science paper on wall structures where they tested a bunch of combinations (http://www.buildingscience.com/documents/reports/rr-1014-high-r-walls-pacific-northwest-hygrothermal-analysis)...one message I got from this is that adding rigid foam significantly increases the wintertime sheathing moisture content. I'm guessing this is because rigid foam doesn't breathe and so drying can only occur to the inside of the home.

And when writing my OP, I forgot that Roxul makes boards as well as batts, so could a rigid Roxul work better, given it breathes and so presumably the sheathing moisture content measure would be much lower than with rigid foam?

Dana, thanks for the detailed response. So 2" of roxul comfortboard would suffice. I don't know what polyiso is, but is it breathable like rock wool (and I also love that the rock wool is also very flame resistant)?

As for the passive solar, we will have a fair amount of windows on the south wall because that is where the beautiful views are. Fortunately, the entire western side of our property is tall forest, right up the property line, so we don't need to worry about west exposure heating. And as it turns out, there will only be one small bathroom window on the west wall, anyways. And the south windows will all have overhangs (I did the angle calculations) or other summer shading (e.g. trellis with canvas "curtain"). So I think we will be good there.

"one message I got from this is that adding rigid foam significantly increases the wintertime sheathing moisture content."

You're completely mis-reading the BSC paper- you've flipped it! Adding exterior foam REDUCES the sheathing moisture content, provided that it's R value is high enough relative to the total R for the climate. See the discussion around Figure 9, p.16 (PDF pagination) and look at the graphs carefully.

The worst-case wall was wall 3c, which had the plywood sheathing on the exteiror of the 1" EPS and R21 fiberglass, with NO foam exterior to the sheathing.

When they flipped that, with the 1"/ R4 foam on the exterior (wall 3b) the peak moisture content didn't even quite make it to 17%, and was under 15% by April.

When they swapped the EPS for 1"/R5 XPS (wall 3), the absolute peak moisture was about 16% despite the much lower drying capacity toward the exterior.

When they bumped the XPS up to 2" / R10 (wall 4) the absolute peak was 13%. Doubling that to R20 dropped the peak to 11%.

The take-away isn't that exterior foam INCREASES the moisture content, but rather that it REDUCES the moisture content, despite the fact that it hampers the ability to dry toward the exterior. More exterior R is consistently better than less exterior R, independent of the vapor retardency of the exterior foam. (At more than 2" XPS would meet the Canadian code definition of a vapour barrier.)

Polyisocyanurate is a rigid foam, usually sold with foil facers, and even more vapor retardent than 6 -mil polyetheylene. It's the opposite of breathable, but if the R-value is high enough, it doesn't matter, as long as the assembly can dry toward the interior.

Putting Roxul on the exterior would be somewhat better for drying capacity than using foam. If you are going with less exterior R than would be necessary for dew-point control and you had to use a polyethylene vapor barrier on the interior it would be essential.

More exterior R is consistently better than less exterior R, independent of the vapor retardency of the exterior foam

I wouldn't conclude that in the general case. For example, I'd take R6 of EPS over R6.1 of foil faced polyiso. Especially in light of BSC's warning (page 33) about foil and the superior drying ability of EPS (figure 19).

BSC gives lower constructabilty ratings to rigid foam > 1.5"; I'd also consider that.

Derating 1" / R6.1 labeled polyiso for temperature in climate zone 5 or colder would deliver lower mid-winter R-performance than 1" / R4.2 EPS, so sure. But now we're picking nits.

I was talking about the measured performance in the BSC document referenced, which used only polystyrene insulation, but of significantly different vapor permeance. (1" EPS at about 2.5-3 perms, vs 1" XPS at about 1.2 perms or 4" XPS at about 0.3 perms.)

Despite having only marginally higher R value but less than half the vapor permeance, the 1" XPS resulted in lower peak & average moisture content in the sheathing than the 1" EPS tested. It's unlikely that reason for that difference is the higher vapor permeance of the EPS exposing the sheathing to more mid-winter moisture drives from the exterior, but maybe. That could be determined from further specific experiments. The more likely explanation is the marginally higher sheathing temp from to the higher R of the XPS keeping the interior moisture drives at bay.

Hi Dana, thanks for clarifying. That was a particularly embarrassing mistake on my part - that's what I get for browsing quickly through the paper, planning to read it in detail later, and then not doing so.

I've read that latex paint is a class III vapour barrier. Doesn't pretty much everybody paint their interior walls? And if so, doesn't that increase the risk if one is relying on drying to the interior?

If I used Roxul on the outside, I would definitely make sure I'm keeping everything inside the wall above dew point. In our humid climate, especially in winter, I just don't feel comfortable relying solely on a vapour barrier to keep the inside of the walls dry. I'd rather allow for drying in both directions.

Also interesting to read in that paper that fibre cement siding can hold a fair amount of moisture that can result in summer inward vapour drive. I have been weighing the pros and cons of fibre cement vs. wood siding, and there aren't many pros on the wood side other than beauty (I do so love the look of real wood!). This may be another pro to add to that column - less risk of summer inward vapour drive.

The study assumed latex paint and various walls worked well with it. With ventilated cladding, summer inward vapor drive is a non issue, even with walls that are highly permeable to the exterior. See page 22.

See fig 5 for why interior and exterior air barriers are beneficial. See page 9 for why cellulose typically outperforms bat insulation.

A class-III vapor retarder is fairly vapor permeable, and can move quite a bit of moisture when needed. In your climate the outdoor dew points are always WAY below you indoor temperature, so there is no exterior vapor drives that might be "trapped" by an interior vapor retarder, nor does it impart a significant latent cooling load (ever!).

That's why you can get away with interior side polyethylene vapor barriers in your climate. In places like Florida the outdoor dew points can hit north of 25C, and an interior-side poly sheeting would collect condensation in the walls in a house air-conditioned to 22-23C. But if you don't NEED the interior polyethylene sheeting and can use class-III vapor retarders, the wall becomes much more resilient to occasional bulk-water intrusion, since it dries readily in both directions.