Posted By terrynew on 29 May 2010 09:41 AM

Three of my exterior walls are earth-sheltered but the south wall, mostly windows and doors, is stick frame. My general contractor has chosen an interesting approach to insulating this, and I'd love feedback to help me assess it, please. This is for a cold climate, central Ontario (like Minnesota).

The frame is 2x6 with Roxul insulation (R22). Outside the OSB is 2" of EPS (Plastispan, R8), and inside the frame is 1.5" of poly-iso (R11) and then drywall/sheetrock. I know it is said to be questionable -- in the current discussion at http://www.greenbuildingtalk.com/Fo...fault.aspx, a link to http://www.greenbuildingadvisor.com...-retrofits says "Adding board foam to the interior of a wall assembly is rarely recommended. This strategy is described as delivering the ‘lowest bang for the buck’ and is compromised by intervening partitions and floors. The board foam interior application tends to leave large air leaks unresolved and also places the framing in a more vulnerable situation, with respect to moisture."

I believe my contractor's reasoning is that the poly-iso between the studs and the gypsum provide a thermal break against the energy loss through the stud frame. Is there value to this, or should we just go with the REMOTE approach of beefing up the exterior EPS? For that matter, is a double wall (two offset 2x4 frames) a better approach? I know many of you are ICF and SIP fans but since this wall is mostly doors and windows, it seems stick-framing is simpler.

Thanks for your thoughts,
...Terry
PS The exterior finishing is natural stucco.

There's already a thermal break on the studs- the 2" of EPS, but going for the extra R11 would be a measurable improvement.  But it doesn't much matter if that extra R is on the interior vs. exterior- best practices would indicate putting it on the exterior.

When you put the foam on the interior, it makes the structural wood run colder, which means it stays below the dew point of the interior air for more hours of a season.  Putting it on the exterior reduces the  number of hours of condensation risk.  But with double foil-facers acting as the vapor retarder & proper air-sealing on the interior, as long as it can dry toward the exterior you should be fine.  (But it has to be truly air-tight on the interior, which means foam-sealing around all electrical & plumbing penetrations, all seams & edges, etc., and it has to STAY air-tight forever.)

Plastispan is good for ~ 5 perms @ 1" thickness, which means it's probably reduced to ~2 perms as 2" thickness, which should be sufficient for the assembly to dry toward the exterior.  If you increased that to 5" (R20) and skipped the interior iso it would be too vapor retardent to dry toward the exterior, but you can probably also skip the interior side vapor retarder and let it be an interior-drying assembly, since all of the mold-susceptible wood is now on the warm side of the foam, and the foam would be ~50% of the total R.  (You might have a fight with the code inspectors on this one, but it's probably about right.)  Alternatively, if you did double- layers of 2" foil-faced iso on the exterior (for about the same wall thickness of your exterior Plastispan/interior iso stackup) you'd have R26 on the exterior, raising the temp of the cold edge of the stud even further.  In any cold climate, with enough foam on the exterior you can eventually get to the point where interior vapor retarders/barriers are neither necessary or desirable, and that's probably where you should be going with this. 

If you use all external iso, taping the seams with FSK tape and staggering 2 layers (with attention to how they overlap at the corners), and foam-sealing the edges makes for an extremely air & vapor tight assembly.  As long as all of the moisture-susceptible materials are on the inside, and warm enough to (almost) never condense.  Assuming you're winter temps are similar to say, International Falls, Minnesota your average January temps are around -18C.  If you keep the interior at +20C with a wintertime relative humidity of 30% or less, condensation will only accumulate in the assembly at the point where the average temp is +2C or less, so if you have just half the R in exterior foam the cold edge of the stud will be right on the edge unless you lower the RH a bit. (25% is about the lowest you can go and still be comfortable).  But with slightly more than 50% in exterior foam the wood will stay dry.  If you used cellulose instead of Roxul on the interior (wet-sprayed would be best) you'd have slighly less R on the interior (R20-R21), but the insulation would harmlessly wick any moisture away from the structural wood during condensing hours. Blown/sprayed insulation also has the benefit of  a perfect fit- filling all voids, lower intra-insulation convection, and will usually outperform R22 batts in real-world performance. With Roxul getting the condensation line well into the foam layers is more critical when you don't have an interior vapor barrier, since convection would bring far more moisture to the condensing layers than if you used cellulose, which has high resistance to convection, and will redistribute minor condensation when it occurs.

In a nutshell, it's always safer to put all of the foam on the exterior, and with R8 + R11 (R19) you're on the verge of being able to skip interior vapor barriers, and can be less fussy about perfectly air-sealing the gypsum board.  If instead you made that R13 + R13  or R13 + R11, all-iso, you're probably there with some margin, and using cellulose to as a hygric buffer to help out with the extremes makes the assembly more resilient to condensation during cold-snap weeks.   (DO check your local climate statistics- if  the average January temp is more like -25C you may have to pad out the iso another inch.)

If you have to prove your case to the inspectors/contractors, you may have to model the stackup with WUFI.  You're spending enough money on foam board that I'm pretty sure you can do better with the same/similar materials using a slightly different stackup.  It's all about keeping the wood warm & dry-  putting R11 between the studs & the interior just isn't the best way to go when you could be putting it between the studs and the great-white north.

Another interesting thread, I wish this forum was around 25 years ago when I had built the home I'm in now! I used a 2 x6 wall, set back the rim joists on both levels and filled with foam, ground and second floor, wood foundation 2 x10 and had the exterior of the house sheathed in 2 inchs of foam, all of it taped and covered in Tyvek. cutting down costs I had to go with Vinyl siding, being the old school installers they used rolled tar paper before installing the siding, I didn't think it was going to hurt anything so I let them have at it! Then the vapour barrier 6 mil, completely sealed the inside walls including the extensive use of acoustic sealant. The basement had 2 x3 nailed on the flat to the exterior wall, in retrospec this was not the way to do it! the first and second floor had another 2 x4 wall built with the studs offset fromthe 2 x 6 wall, thats where all of the plumbing, electrical and miscellaneous runs were put.
I'm about to put the house on the market next year and will be paying someone to do a pressure test and energy audit just to see how it does and for either my amusment or a selling feature!

John

I should have added the house was built in Ottawa Canada, and we do get some pretty cold winter weeks of  -20 to -30 Celsius. I also did have a short battle of my insulation wall stack up, but at the time the code was 2/3  on the cold side and 1/3  on the heated/cooled side and the inspector agreed to my design!

John

Posted By JohnyH on 02 Jun 2010 08:12 AM
Another interesting thread, I wish this forum was around 25 years ago when I had built the home I'm in now! I used a 2 x6 wall, set back the rim joists on both levels and filled with foam, ground and second floor, wood foundation 2 x10 and had the exterior of the house sheathed in 2 inchs of foam, all of it taped and covered in Tyvek. cutting down costs I had to go with Vinyl siding, being the old school installers they used rolled tar paper before installing the siding, I didn't think it was going to hurt anything so I let them have at it! Then the vapour barrier 6 mil, completely sealed the inside walls including the extensive use of acoustic sealant. The basement had 2 x3 nailed on the flat to the exterior wall, in retrospec this was not the way to do it! the first and second floor had another 2 x4 wall built with the studs offset fromthe 2 x 6 wall, thats where all of the plumbing, electrical and miscellaneous runs were put.
I'm about to put the house on the market next year and will be paying someone to do a pressure test and energy audit just to see how it does and for either my amusment or a selling feature!

John

Yeah, sealing the studs in next to the cold-damp ground side of things can be a hazard.  As with above-grade walls, a lot can be said for putting foam on the exterior, raising the temp of of the foundation wall.  Unlike above grade walls, there's no hope of a dries-toward-exterior stackup, and the Canadian code requirement for 6mil poly on the interior of basement wall insulation can be a problem.  Vapor-sealing the interior drives ground moisture higher in the foundation wall to potentially rot the sill plate and/or create exterior efflorescence & spalling on the above-grade portion of the foundation. 

In this situation building in very good capillary breaks at the footing & sill, as well as drainage & waterproofing the below grade exterior of the foundation wall are critical.  With an interior vapor retarder the foundation wall can only dry via the above-grade exterior portion, so minimizing the migration of groundwater into the foundation wall is key.   If the foundation is instead insulated with the appropriate rigid foam on the exterior with only vapor permeable materials on the interior, the foundation stays drier, and the integrity of the capillary break at the foundation sill becomes less critical.

If the above-grade portion of the foundation shows signs if efflorescence or is spalling, the foundation (but not the sill plate) can be protected from deterioration by a sacrificial parging of a lime-cement/sand plaster.  The plaster may need touch-up or replacing every 25-100 years, but that's much cheaper than the foundation repairs that might otherwise eventually be called for should the cement between the aggregate in the concrete give up.

Thanks for that very thorough reply, Dana1. As John points out, I am indeed hindered by the Ontario building code requirement for 6-mil poly on the inside (which my contractor would place between the studs and the interior iso). My builder also added that another reason for the iso on the interior is that it covers up the thermal break SM where the concrete slab floor meets the stem wall. (See attached drawing.) If I could find another solution for that cover-up he might be more swayed.


(Background for the drawing: This stick frame is just for the south wall. It's a floating slab floor. The other three walls are Durasol ICFs, cut into a hillside and bermed to near the roof. It's a walk-out basement with a roof on top.)

I discussed your response with my builder and he still believes there is value in blocking heat escaping the interior as soon as possible with the iso. Perhaps if I can find an authoritive online document from a builders association or something? It's a bit late to switch builders and that would push the project past this yearend.

If I interpret part of John's reply correctly, would a drainage mat between the OSB and EPS solve the problem? Here is my suggested content: drywall, 1.5" iso, 6-mil poly, 2x6, OSB, two sheets of building paper, 3/8" drainage mat with weep screed below, 2" EPS, metal lath, 3-coat cement stucco. Would this drainage prevent moisture buildup?

Thanks,
...Terry

The "blocking heat from escaping as soon as possible" argument just doesn't hold water. The heat loss is the same whether the foam is on the interior or the exterior of the studs. Putting it on the interior just makes the studs run colder, but the heat flux & temp at the interior wall will be the same.

Putting the rainscreen between the OSB & exterior foam would substantially reduce the R value of the assembly due to convection between insulating layers, and partially defeating the thermal break over the studs that the external foam is providing. Plastispan at 2" is sufficiently permeable that it allows the OSB to dry toward the exterior. Snugged up against the OSB it'll keep the OSB warmer, thus increasing it's vapor pressure, increasing the rate of water release from the wood. The rainscreen rightly belongs between the metal lath and building paper (which goes on the exterior of the Plastispan), which will keep the semi-permeable Plastispan drier in summer when exterior vapor drives from the stucco are high.

Seriously- download the WUFI model and model the moisture content of the studs & sheathing with extra exterior foam only, no interior vapor retarder vs. your stackup. At some not-unreasonable thickness the exterior foam only approach beats the classic low-R/interior-poly approach. Codes requiring interior poly makes sense for classic stick-built structures insulated with fiber batting, but with the amount of foam you're incorporating (half the total R) the rationale begins to wane. If you used the old-school 2/3R-1/3R rule for where to place the vapor retarder (which still works in most of Ontario, but is overkill in many parts), going with lower-R inside the studwall, putting the poly on the exterior, then increase the amount of exterior foam, you'll be there. But with more precise models like WUFI to find the balance point you can probably keep the Roxul and just bump up the total R a bit with more exterior foam for a more ideal stackup.

As for the stemwall/slab thermal break- with sufficient exterior foam you don't need it, but slipping in a strip of 1"-1.5" XPS to float the slab at just that interface works- it doesn't need to extend up the interior of the studwall. The finish flooring, gypsum & kickboard can be set up to cover that foam fairly easily.

You may find this design a useful reference point:

http://www.buildingscience.com/documents/designs-that-work/information-sheet-minneapolis-profile/?topic=/doctypes/designs-that-work

If your local climate is colder than Minneapolis (which it may very well be), increasing the thickness of the exterior foam would be the only adjustment necessary, but you already have more foam in your stackup. Note, the only interior vapor retarder is semi-permeable latex (!), and it relies on air-tight gypsum board techniques to limit air-transported moisture.

At the foundation your design differs in that it has both exterior & interior foam, which is good, but it makes the quality of the capillary break at the foundation sill even more critical especially since the exterior foam slows the exteriror-drying of the above-grade stub of stem wall. Don't rely on foamed poly sill gasketing alone- consider going with both gasket + full-metal (thin copper flashing). You could go with gasket + pressure-treated sill plate, but that would allow more humidity into the studwall cavity.

The Minneapolis design uses polymer stucco, but you're going with the real goods, which has a very high reservoir capacity for retaining liquid water, wicking in dew or rain. When sun hits it, the vapor drives go ballistic, which is why it's essential to maintain adequate back ventilation/rainscreen-cavity to relieve that humidity immediately. There will be plenty of times where the stucco will release enough water vapor that condensation will occur on the building paper, which is why both the gap and 2 layers of paper are required. Read this:

http://www.buildingscience.com/documents/insights/bsi-029-stucco-woes-the-perfect-storm/?searchterm=stucco

Dana, at the foundation sill plate you recommend both gasket + full-metal (thin copper) flashing. Doesn't that cause a thermal bridge?
...Terry

Posted By terrynew on 18 Jul 2010 08:39 AM
Dana, at the foundation sill plate you recommend both gasket + full-metal (thin copper) flashing. Doesn't that cause a thermal bridge?
...Terry

Yes, but it's a thermal bridge broken by the foam insulation.  If you use the copper/poly laminate instead the thermal bridging is miniscule, but with flashing it's measurable.