Couple thoughts from a Canadian self-build...

First, my folks did the double 2x4 walls offset back in the 80's when they built their first house. It was an excellent house, kept the heat in really well and when highways bought the land off them, someone actually bought the house and moved it 100kms north. It was just that well built. That being said, it was expensive and labor intensive. Thankfully they built it themselves. My dad always loved the really deep window-sills.

I recently built in a cold location and opted for 2x6's with R22 Roxul in the walls, sprayfoam on the headers. ICF basement and Roxul to R44 in the attic spaces. I'm fairly certain that yes, canadian code is to have vapour barrier on the inside. I have a forced air system and think it's doing pretty darn well out in the bald prairie! This traditional style is inexpensive and does a good job.

Last thing, I have a friend on the bald prairie with in-floor only heating...and it's a real struggle on those -40 days to keep the place warm. And the larger issue started to become the humidity, because it doesn't have the same level of air circulation as a forced air system. Stagnation of air causes moisture to build up on your windows and doors. It's also not a lot of fun to live in. An HRV system is now mandatory in Canada, but these perform best when connected to the cold-air returns, and the outside air feeds direct into the furnace. You can still use one with your system feeding it through the bathroom and kitchen fans, but you're going to want to ensure it's running on a regular basis to keep those humidity levels down.

Thank you for your thoughts Awnmyown. Perhaps the plain 2*6 is the most sensible option for us as well. I really wish I could do something about the bridging issue though...

As for the heating, yes, I'm a little hesitant to rely only on a single system. It certainly seems prudent to have a back-up system, especially for those days the temperature plummets by 40-some degrees in less than a day. I was considering a few different options (ranging from electric space heaters, to fireplace, to a furnace), depending on the cost effectiveness.

Thanks for the tip on the cold-air return for the HRV system. Maybe a furnace will the the best option if those two systems are highly dependent, especially if they can share some ducting. More questions to ask my HVAC guy when time comes.

When you say Canadian code are you after the R2000?

In IECC, when x amounts of exterior insulation is added, you can and should eliminate the interior vapor barrier, depending on what zone you are in and switch out to a high perm product, such as a simple coat of latex paint. However in zone 7, you would need more than an r10 to get there. Would have to see if IECC and other building scientist findings can assist you in your argument with code officials. I have never lost an argument with them the second I start talking about vapor drives and perms of materials. Many do not understand how it all works. Baking up claims with IECC or building scientist publications also helps.

I built an ICF home, about r28. This system if great for thermal bridging, air tightness, and no risk of vapor issues. If I were doing it again, I might opt for the r40 double stud wall which was my close 2nd place option. Double studs are easier to detail than thick exterior foam walls, and once you get over 2", typically is cheaper too. Installing windows and siding is normal as you do not need furring through the foam (easier= cheaper). Using ZIP sheathing helps to get you a good air barrier on your exterior. Vapor barrier can still be used on the interior of the walls. Walls are only part of the equation however. Design for solar gains (using correctly tuned and sized glass), insulating of the foundations and slabs, and obviously the roof system. All need to be air tight.

Building Science Corp.'s Builder's Guide to Cold Climates notes that:

"No assembly constructed in Climate Zone 4, 5, 6 or 7 requires a Class I vapor retarder (i.e. a polyethylene vapor barrier.
....
Interior vapor resistance, beyond that of latex painted gypsum board, is not required if sufficient thermal resistance is provided by insulating sheathing." p. 141

In practice, instead of a poly barrier, Class II or Class III Vapor Retarders, may be specified. These range from a layer of latex paint to "smart" vapor retarders.

This preliminary research at the University of Waterloo, Ontario, showed that while poly vapour barriers were effective in winter, they promoted interior stud moisture in the summer.

I don't have access to the Codes in question (Alberta Building Code 2006 and the National Building Code 2005, which is law in Saskatchewan), so I can't quote chapter and verse. But in Ontario I have found that a designed system with backup documentation can trump Code requirements.

This post was edited by worthy on Wed, Jan 15, 14 at 19:20

It is good to know that I might be able to use research to argue my way out of the code requirement, although the easiest would more than likely be to just follow the darn thing.
Thanks for the link Worthy. It will be handy to have in hand regardless of what I end up with.

Izerarc, what type of double stud did you consider? And would you use wet-applied cellulose?

Our basement will most likely be ICF, and yes, I will have to make sure the foundation and slab gets the appropriate insulation and air tightness (though I still need to figure out what exactly that is...lol)

I'm only vaguely aware of designs for solar gains. I had already planned on orienting my house towards the south, with minimal windows to the North, but I will have to look at the overhangs.
Will my window supplier know anything about correctly tuning and sizing glass, or is that something I would have to just muddle through on my own?

I just heard back from a local foam installer (seem reputable). He proposed 3" closed cell foam in the wall (R20.4 at low 90's% efficiency), and 4" directly in the cathedral (vaulted) ceiling (R29.2 at mid 90's% efficiency).

I asked about these, "in my opinion", very minimal thicknesses in order to meet code, and the potential for ice damming. He argued that a completely sealed envelope will not cause any moist air to escape, hence no ice damming. He also believed that 4" would not cause any snow to melt either (said that cabins up north only have 3" without issues)

What I like about this idea: the best air tightness possible (?). I don't have my drawings ready yet, so he could only give me a rough cost, but although this is more expensive than standard 2*6 with Roxul batts, is is significantly better and likely "only" a couple grand more, if I include in our own saved labour. In the grand scheme, I'm more than ok with that.

in a single 2*4 wall, electrical would have to be installed first (ok, whatever), but the insulation behind the boxes would be very thin, and the wires would likely be IN the foam. How would that work in 20 years when for some reason we'd need to re-wire the place?

Another big question: is the proposed thickness enough??? It also does not solve the thermal barrier issue. Would it make sense to increase the thickness a bit in order to cover the 2*4, then do another "interior" wall to install the electrical? (likely another stupid question, but could you also add the ventilation (HRV ducts) in this space?)

Another option might be to do 2*6, filled 3.5-4"(?) with foam, then leave the rest open for electrical to be installed after. This leaves a full bridge though, but perhaps that's not that important??

For the ceilings we aren't limited by thickness at all, and can go with whatever. I'm just not sure what would make the most sense. Any additional thickness would drive the costs up to a point where it might not be financially viable anymore.

Any input at all would be highly appreciated!
Thanks

The question is whether you're just trying to meet Code, probably R20, or insulate in a cost effective way for your climate?

Filling the walls with 3" of ccspf is way more expensive than using dense fibre insulation, as you probably know. By sheathing the exterior walls with one or two layers of polyisocyanurate or XPS followed by fibrous insulation in the stud cavities you'll get a much higher whole wall R value at a lower cost.

CCSPF in the cathedral ceiling may be the best choice as the cost and complications of properly detailing a ceiling in other materials may be higher.

See this online calculator from the Oak Ridge National Laboratories.

You still haven't said which Province you're in. Different Codes, you know.

I'm in SK (zone 7A I believe), sorry about that.
I do want good insulation, preferably above minimum code. Foam contractors claim that foam at minimum code is effectively much better that any other insulation, due to air tightness (duh, of course they say that...).
I am a cynic, so I can't fully belive that, but say if I get a blower test and IR done as part of a the spray deal, I would "guarantee" that all those little faults, that are doomed to show up with batts and regular vapour barrier, are minimized. I would be willing to pay a bit extra for avoiding that headache.

As for the roof, I understand that the code for cathedral vs flat roof is different (R28 vs R50) (why is that?). Would my roof fall under the cathedral category? It will be significant space between my drywall and the roof in several places (see quick-and-dirty picture below).

I also found a link to a previous discussion regarding this issue, and I think the contractor may have defined it as such.

Here is a link that might be useful: Previous discussion on foam r-value

This is a pretty good (semi-scientific) explanation why a lower R value of foam is superior to standard fiberglass, although I cannot verify the facts.

Worthy, I will have to look at that calculator when I get home. Hopefully that will shed some light on this as well. Thanks!

Here is a link that might be useful: why-r-value-is-not-accurate

spray foaming your walls is a waste of money, IMO. I always recommend people to not foam the walls. there are ways to get just as tight (and tighter over the long run as spray foam can pull away from the studs) using other methods, reducing thermal bridging, higher overall whole wall r value and some times cheaper cost.
R value is r value. Foamers will try and rationalize is ("sell it") that less foam (lower than code r value) is good enough. This is not true. Sure, foam performs better than batts, but that is because it stops the air movement. Anytime air is moving through insulation, it reduces the value of it. Stop the air moving through batts, and then the 2 systems perform similarly. Now the big advantage foam has batts never can is the ability to cover every little area of the wall. With that being said I always never use batt. However dense packs and damp sprayed products can achieve similar results. Again, you must stop the air movement. 2"+ of XPS if an effective air barrier, but joints must be sealed and taped. As a wood framed structure moves, settles, etc, the caps can open. That is why I like to place the air barrier on a more rigid plane, such as the sheathing layer. Caulk the sheathing to studs and plates, and use a product such as ZIP sheathing that can be taped over the seams. Then add exterior insulation for reduced thermal bridging. Tape the joints there as well, and you will end up with a high performing, tight wall system pushing r30, mid r20s whole wall. 3" of foam is r19 (closed cell) not matter how they try and spin it. It can never be higher than that. Do not let them sell you on that. Code r is code r.

The double wall I was going to use was 2 separate 2x4 walls with about 3 or 4" inbetween the 2 filled with dense packed fiberglass. when comparing $/r value, this is the cheapest way to get a high r wall with very low to no thermal bridging.