I'd suggest you have the loadbearing wall sitting on the concrete part of the foundation, which you could easily do by moving everything in and having only one sheet of exterior foam extend past the ICF EPS. Looks like the thermal bridging issue is taken care of pretty well.

I would think hard about eliminating the sheet foam altogether. If you are already building a double wall, put it to good use by thickening it up. You can solve the thermal bridging at the rim by setting your rim back enough to apply a thin strip of rigid at that area only, and thickening the wall will give you the R values you are looking for.

You will save a tremendous amount of cost and difficulty in detailing your window and door openings along with the cost of the sheet foam itself. If you are worried about cold sheathing and drying direction, shift to a breathable exterior sheathing (structural fiberboard, diagonal 1x boards, etc) and shift your air barrier to the inside with thin CDX taped plywood under your drywall.

I think that you are over complicating this.
 If you are willing to commit to a double wall, or maybe a Larsen truss double wall in your case, exterior foam is just an added expense. Just make the double wall 12-14” deep with offset studs and Dense packing with cellulose. You have the blower in anyway for the densepacking the additional cost to go 4 more inches of depth is not much. If you did a Larsen truss the inner wall can braced to resist wracking, be load bearing, the outer wall can extend over the foundation and be denspacked over the sill. The outer wall only needs to hold its own weight+ insulation+ siding so can be lightly built.

The thing I see with your assembly is that you have peal and stick on the outside, I assume that is a vapor barrier, and your drainage plane then 3 inches of foam, then tyveck?
 I don’t think the tyveck will have any value in this assembly. It is not the drainage plane, the peal and stick and foam can be the air barrier. I don’t think it will hurt, it just won’t help that much.

Posted By eric anderson on 20 Oct 2010 01:14 PM
I think that you are over complicating this.
 If you are willing to commit to a double wall, or maybe a Larsen truss double wall in your case, exterior foam is just an added expense. Just make the double wall 12-14” deep with offset studs and Dense packing with cellulose. You have the blower in anyway for the densepacking the additional cost to go 4 more inches of depth is not much. If you did a Larsen truss the inner wall can braced to resist wracking, be load bearing, the outer wall can extend over the foundation and be denspacked over the sill. The outer wall only needs to hold its own weight+ insulation+ siding so can be lightly built.

The thing I see with your assembly is that you have peal and stick on the outside, I assume that is a vapor barrier, and your drainage plane then 3 inches of foam, then tyveck?
 I don’t think the tyveck will have any value in this assembly. It is not the drainage plane, the peal and stick and foam can be the air barrier. I don’t think it will hurt, it just won’t help that much.

It may/may not be highly retardent (depends on the material, which isn't specified), but putting the a low perm WRB, or even the 3" of exterior XPS on the outside of 8" of dense-packed cellulose reduces the outward drying capacity to nil or near-nil, without providing enough exterior R to reliably keep the exterior studwall above the dew point of interior-space air in  colder areas.  At R15 it's about 1/3 of the total R plenty for zone 5, could be real iffy for zone 6.   If you did it in higher-perm fiber-faced iso instead of XPS, it could dry into your 3/4" rainscreen under the siding as long as it's a high-perm WRB (like Tyvek) and you'd have a better shot at keeping moisture from building up in the cellulose. Iso would give you 20% more exterior R, raising the average winter temp of the exterior studwall a few degrees, with a somewhat higher vapor permeability better allowing it to dry, but it's still a bit lower perm than ideal.  The other way to go would be to make it unfaced 4" EPS, which has about the same perm  ratings of 2" XPS , but with the same R-rating of 3" XPS (still a bit lower perm than ideal.)

I'm with eric and others- since you're already in double-studwall-land, just making the walls thicker would be easier/cheaper/more reliable.  If you don't put any low-perm layers on the exterior, and rainscreen the cladding to maximize the exterior drying capacity, you'd be able to use standard latex on the interior as your vapor retarder, and it could dry faster in both directions.  Using nothing lower perm than #15 felt as the drain plane on the exterior is about right for a fat cellulose wall.

Putting the sill plate & studs over the concrete is right- if need be it can be thermally broken on the exterior with SPF, or extending the exterior ICF foam plane upward to lap the studwall wall sheathing by 4-6" using iso for maximum R.  Make sure the drain plane/flashing laps correctly if you take that route.

The amount of exterior R required to be safe for all of zone 5 might be R7.5 for 2x4 (~R13) construction maybe, but not for 8" of dense-packed cellulose (~R30). To be sure you can run the numbers for the ratio you'd need for your exact climate data, not generalized recommendations. But in southern OH you are probably just fine from a condensation point of view with a one-third foam, two-thirds fiber, R-ratio, and can probably get away with reducing the foam to just 1/4 of the total R. In much of ND having only 1/3 as foam would be asking for trouble.

But from an overall moisture resilience point of view it's better to be more vapor-permeable than 3" of XPS. In mixed climates highly retardent vapor barriers create as many problems as they cure. Moisture get in by any number of methods, but the primary way it leaves is by vapor diffusion. Thick cellulose walls can buffer a LOT of moisture, but if you inhibit it's ability to release that moisture it can build up. As long as it can leave as fast or faster than it's coming in, everything stays happy. Using wide metal flashing for both a capillary break between the foundation & sill as well as the termite shield wouldn't be a bad idea, to impede the ground-moisture path into your studwall. But leave both the exterior & interior surfaces of the studwall relatively vapor-permeable.

Question for all:
I haven't used ICFs but it seems counterintuitive to have the structural 2x4 wall only sitting on 1-1/2" - 2" of concrete. For those who have used ICFs - is this common?

I was looking at R-15 foam/R-30 cellulose for a finish R-value of around R-40 after framing losses. I'll look at using 2.5-3" of Polyisocyanurate instead of XPS. I was going XPS for the higher R-values than EPS, going a little better won't hurt. Price wise there does not seem to be much difference between XPS and Iso. I am still considering ditching the OSB and use bracing instead. Or perhaps only using it at the corners for the required bracing.

Posted By Bob I on 20 Oct 2010 07:08 PM
Question for all:
I haven't used ICFs but it seems counterintuitive to have the structural 2x4 wall only sitting on 1-1/2" - 2" of concrete. For those who have used ICFs - is this common?

While I have yet to personally build with ICFs I have seen several details from the manufacturers with the sill plate flush to the face of the ICF. The only one I have at the moment is Quad-lock's. Detail QL-122 in particular (SIP wall flush to the outside of the ICF, cantilevered over the foam). I believe you can register at their website and get access to all their details.

Bob,
The detail for having the wall partially off the bearing surface is the same as as a minimal cantilever. The ends of the I joists need proper squash blocks, but all the manufacturers supply this detail.
 Often they will use a sill plate that is a 2X10 and hang one side over the foam a bit on the outside. Only 8 inches of it bears on the concrete. The rimjoist is often upsized also.
This is the detail from my own house It is a 1.5” overhang

skip the foam if you do a double wall. cheaper and better (it will easly dry to the outside). run a WUFI analisys and you'll see what I'm talking about

Compare the 2 wall assemblies in question

Your assembly is 3” XPS + 0.5” OSB +8” doublewall+ ½” drywall= 12” Crude r calculation Assume 30% framing factor

1/[0.70* 1/(3*5+0.5+ 8* 3.5 + 0.5) + 0.3* 1/(3*5+0.5+ 8*1 + 0.5)] = 35.2 R value for wall assembly

Now by comparison if we made a 12” total thickness double stud offset wall,

1/[0.70* 1/(+0.5+ 11* 3.5 + 0.5) + 0.3* 1/(0.5+ 3.5*1 +7.5*3.5 + 0.5)] = 36.3 r value for same thickness and simpler assembly.


Use sip sheathing on the outside, tape the seams for your air barier, use 15 lb felt over that for your drainage plane, on the inside use drywall with laytex paint.

Assembly can dry to the inside and outside, R35+ whole wall assembly.
Since you can easily cantilever out 6” over the ICF foundation, you should lose no interior space. Both assemblies will work, I think mine is cheaper to build

I reached the exact same conclusion and a lot of other people as well.

for example:

http://energetechs.com/wp-content/uploads/2009/12/Wall-assemblies-cost-analysis.pdf
http://greenedmonton.ca/mcnzh-insulate-and-seal