Eastern Canada ‘pretty good house’ Insulation strategies
I’ve read so many articles my head is spinning. I’m building in New Brunswick Canada which appears to be equivalent to your climate zone 7. I’m endeavouring to build a pretty good house meaning efficiency to the point of diminishing returns aka maximum value. One of the major challenges I have here is reconciling good building practices I’m reading about here with the capabilities of local contractors. With that said, I have a few questions.
House is 32 x 48, story and a half on an ICF foundation. Good solar orientation with front of house facing due south. No fancy detail on roof other than switch from 12/12 pitch in front of house to 6/12 in back.
1. Main Floor Joist – On foundation or hung inside? I have a spec’d a 9′ ICF foundation to accommodate hanging the main floor inside of it to eliminate the thermal break. Neither my ICF contractor or my framer wants to do this and both say it will cost extra. I have friend in spray foam business and am leaning towards just having him spray the rim joist instead of fighting to have it hung. Considering I am aiming for pretty good and not perfect, I think I have answered my own question. Good talk.
2. Exterior wall – Again, anything other than 2×6 framing is out of the question. I would like to have at least R30 in my walls. Since I can get a good deal on closed cell spray foam I am leaning towards doing all walls in 5″ of closed cell but am unsure of best way to deal w/ thermal bridging. I am 90% sure I will strap the interior wall w/ 2×3 to create a service cavity since studs are full of insulation anyways. On the outside I would have a layer of OSB then housewrap then cladding. Is that enough to deal w/ bridging or should I look into 1″ of rigid on the exterior as well? If I were to use this system I would not need implement any other air/vapor barriers would I?
3. Roof. If I’m this far into spray foam I may as well stay the course right? The steep roof for sure I think I should spray but once I’m into my flat attic I’d like to blow in cellulose to R60. Do I carry the spray all the way to the ridge even with cellulose below it? Do I need to vent this roof at all if its all done in closed cell? Do I need a second air barrier (poly) underneath the cellulose? I would think so because I don’t want my attic to be a conditioned space do I? This seems to be the most confusing detail for me, any advice is greatly appreciated.
Barry
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Barry,
You seem satisfied with your answer to Question 1, so I'll assume that's settled.
2. You're right to worry about thermal bridging through your studs. If you're framing with 2x6s, you really should install a continuous layer of exterior insulation (usually rigid foam). Your suggestion to install 1 inch of exterior rigid foam won't work, as this article explains: Calculating the Minimum Thickness of Rigid Foam Sheathing. If you really are in Climate Zone 7, you'll need exterior rigid foam with a minimum R-value of R-15. If your Canadian location turns out to be more like Zone 6 -- and it may -- then you need a minimum of R-11.25 rigid foam.
Once you install rigid foam on the exterior of your walls -- and I think you should -- the question arises: Why use spray foam between the studs? Cellulose, mineral wool, or dense fiberglass batts probably makes more sense there. To learn why, see How to Design a Wall.
3. If any section of your attic has cellulose on the attic floor, there is no need to install spray foam under the roof sheathing in that section of attic.
If you want to learn more about all of the different ways to insulate sloped roof assemblies -- either vented or unvented -- see How to Build an Insulated Cathedral Ceiling.
To expand on Martin's point #2. I recently had a conversation with a code official in Ottawa, who claimed that that you need 2/3 exterior insulation on the sheathing/vapour control layer. He did however say, that a dew point analysis can be submitted for review. I personally feel this is excessively "safe", and am currently trying to revise my double stud wall plan to suit. The IRC, and many requirements in the states is 1/3 exterior insulation.
However, Table 9.25.5.2 (Ontario building code, but should be similar section in NBC), states "Minimum ratio, total thermal resistance outboard of a materials inner surface to total thermal resistance inboard of a materials inner surface"
Up to 4999 HDD requires 0.20
5000-5999 HDD requires 0.30
6000-6999 HDD requires 0.35
7000-7999 HDD requires 0.40
When i read this section, it would seem that the ratio identified is exterior insulation. Example, Ottawa is 4600 HDD per an old OBC i have laying around, so 0.20 outboard thermal resistance ratio. In an R40 wall, that would be R8 exterior. I have sent in a follow up to this, and haven't heard back.
Closed cell foam between studs is a waste of good foam. The thermal bridging swamps the benefit of the high R/inch foam. At a 25% framing fraction (typical for 24" o.c. stud construction), 5" of R6.5/inch has a "whole wall R" of about R17.3, (that's including the R value of the gyprock, the sheathing the siding, and the interior & exterior air films!) despite a center cavity R of R32.5.
If you instead installed a full 5.5" of R4/inch 0.7lb density open cell foam (or 1.8lb density fiberglass) in the same wall the whole-wall R comes in at about R15.75.
The ~R1.55 difference that can be made up FAR more cheaply (and at a far lower environmental impact) with 3/8" fan-fold XPS or EPS siding underlayment on the exterior, and 2 mil nylon (rather than polyethylene) under the gyprock as the interior side vapor retarder, or using 2x4s instead of 2x3s for your insulated service chase.
A 2x3 insulated service chase adds about R7 to those numbers, so you're looking at R24-R25 whole wall as-described.
Mind you, none of these are anywhere near approaching the limits of financial sanity on a lifecycle basis. Something closer would be the Zone 6 or Zone 7 whole-assembly R-values found in Table 2, p.10 of this document:
http://buildingscience.com/file/5806/download?token=GouEIX9Y
Using zone 6 & 7 recommendations in that table, an R35-R40 whole wall using 2x6 framing with high density fiber plus a 2x3 insulated service chase would take ~3-4" of EPS exterior to the sheathing. The cost of going much beyond that would be more expensive on a lifecycle basis than the energy it offsets using rooftop PV and heat pumps as the energy source, but over the lifecycle of the house going that high still usually "pays off". There are significant error bars in the calculation, depending on what you anticipate future energy price inflation/deflation to be, but dropping back to R25-ish whole wall type construction, would need signficant future energy price deflation (or significant climate warming) to be at the limits of financially rational. At R40 whole-wall type construction, even in the foggy-dew maritimes you would be able to fit enough solar on the roof (if correctly oriented) to cover 100% of the annual energy use (unless you're the type that leaves the windows open all winter, or host 3 video gaming parties per week for mega-hours of 5kw power draw using world class rendering engines for the graphics. :-) )
Thank you all for your replies. Responding to Martin first.
2. This is the realization I'm coming to. I was only onto the spray because I could get a good deal from a friend but even w/ that deal, blown in would be cheaper. I am in fact in climate zone 6 and believe I can meet my goals with R20 blown in, R10 in rigid exterior foam and R5ish in my service cavity. The only downside of this design is that I will have to carefully manage the integrity of continuous poly as the inner air barrier correct?
3. I have read the insulating attic and insulating cathedral ceiling articles but will still scratching my head when picturing spray foam on steep roof and blown in on flat ceiling. As you mention, if there is blown in anywhere, I should blow it all in, even the steep and leave a continuous vent from soffit to ridge. For my ceiling, under the rafter I would have poly as air barrier again and tie that into the wall poly. I wish I wasn't wrestling so much poly but it seems the most logical solution. Thanks again.
R10 on the exterior is insufficient for dew point control at the sheathing layer when you have R20 (studwall) + R9 (insulated 2x3 service chase), and would require a Class-II vapor retarder. If you placed that vapor retarder between the studwall and service chase it would have fewer penetrations to seal up.
Using polyethlene instead of 2-mil nylon is a bit risky, since the exterior foam is also vapor retardent. 2.5" of unfaced Type-II EPS runs R10.5 and will have a vapor permeance right at the edge of Canadian code definitions for vapour barrier. R10 XPS or polyiso would be even tighter. Sheets of 2-mil nylon meet the code definition when the proximate air is dry, but if the air in the cavities ever hit mold/rot levels it becomes vapor open, and allows the assembly to dry. Certainteed MemBrain can be purchased through Lowes Canada for ~CDN$180 for an 8' x 100' sheet if you get stuck, but it's usually considerably cheaper if you can find it through distributors. Detailing it as an air barrier is the same as with polyethylene- the labor cost is the same. In the grand scheme of things at 23 cents per square foot it's cheap insurance. There are other membrane type "smart" vapor retarders (eg Intello Plus) most of which are European imports at a much higher price point, and harder to find.
Response to RYAN MAGLADRY
Reading the building code you cite:
https://www.ontario.ca/laws/regulation/120332
Table 9.25.5.2 seems to say something different than you interpret.
Your example seems to be that the ratio equals (exterior insulation)/(total wall R value). I think the code is that the ratio equals (exterior insulation)/(cavity R value). Thus for an R40 wall, it would be R6.6 exterior and R33.3 in the cavity.
Barry,
I'm confused by what you wrote in Comment #4: "I have read the insulating attic and insulating cathedral ceiling articles but will still scratching my head when picturing spray foam on steep roof and blown in on flat ceiling. As you mention, if there is blown in anywhere, I should blow it all in, even the steep and leave a continuous vent from soffit to ridge. For my ceiling, under the rafter I would have poly as air barrier again and tie that into the wall poly."
This is hard to decipher. You have some insulated sloped roof assemblies -- also known as cathedral ceilings -- right? With those insulated sloped roof assemblies, you have several options when it comes to insulation details: vented or unvented, with spray foam or without spray foam.
When it comes to the unconditioned attics -- the spaces with insulation on the attic floor -- the situation isn't complicated. You just install a pile of cellulose on the floor, as deep as you want. (Usually 12 to 14 inches is a good idea.)
Can you please explain your confusion? Because the last time you tried to explain your confusion, I couldn't follow your train of thought.
Thank you Dana Ryan and Martin. As you can tell I'm a bit out of my depth. Please allow me to reread everything including the 61 pager Dana posted and come back to you tomorrow after I've collected my thoughts. Sincerely.
C.B.
That is precisely how i interpret the table, for Ottawa with 4600 HDD, i would need 20% exterior insulation.
However, I can't seem to gain any traction with the local building officials. They are insisting on VB on "warm side" 9.25.4.3., and are saying at least 2/3 exterior insulation to the VB, which would mean you'd have 2 vapour control layers in order to follow that table. In my opinion, this is asking for trouble... should moisture ever get into the wall, it is not inherently difficult for it to escape.
Thanks for your patience everyone. I'll try walls first.
We agree rigid foam on exterior is helpful to eliminate thermal bridging. I can get a 2" Codebord from owen's corning at reasonable price giving me R10 exterior which is not quite the R11.5 prescribed for my climate zone but I am calling it close enough. Inside the stud cavity I could do batt, spray or dense pack cellulose w/ preference to cellulose because it is less expensive and environmentally harmful than foam but fits tighter and eliminates potential for thermal cycling
So, assuming dense pack cavity insulation, I have a R32 wall, not an effective R32 but probably pretty good. That bring me to question of air barrier. The Codebord product can be used as part of their air sealing system however they suggest that one stlll installs an internal poly layer and not seal it on the sides which I don't understand. Why install it at all if you aren't going to seal it? The code says a sealed continuous layer is needed and if I understand the building science correctly, it says another air barrier isn't needed and can actually trap moisture in the wall since the exterior foam is an air barrier as well. To get around this all, I can use a smart air barrier like Membrain that will let the wall dry to the inside if moisture were to get in there. Is that correct?
Barry,
I stand by the standard GBA advice:
-- in your climate zone, you need a minimum of R-11.25 of exterior rigid foam, not R-10;
-- it's a good idea to seal the seams of exterior rigid foam insulation with an appropriate tape;
-- when a house has exterior rigid foam, you don't want interior polyethylene;
-- if a local code inspector insists on some type of interior vapor retarder, you should install a "smart" vapor retarder like MemBrain, which is unnecessary but harmless, and which will usually satisfy the building inspector;
-- it's always a good idea to have an interior air barrier (usually, gypsum drywall installed with attention to air sealing).
If you're going to cheat the exterior R relative to the IRC prescriptive, use either 2-mil nylon (MemBrain) as the interior vapor retarder or half-perm latex primer, not polyethylene, not standard latex paint. You can get away with using polyethylene with only R10 on the exterior since it meets or beats the NBC R-ratio prescriptives, but it's not nearly as resilient as something more vapor-open.
Martin, Dana, Ryan I have to say you guys are suckers for punishment answering the same questions over and over again. I have to tell you, you certainly are appreciated and unfortunately this is what it's going to take to bring these efficiency details into the mainstream. I think you know that - thank you.
I sat down tonight to do 30 minutes of reading 4 hours ago. Back into the rabbit hole I went, there are a lot of contradictory theories out there and some questions where the jury is still out. So again, thanks for wading through comments to help clear the air where possible.
Today I spoke to a guy to get some energy modelling done. We got talking about construction details and I explained I was going for drywall, membrain, 5.5" dense pack, osb (taped), 2" XPS (taped), Tyvek, 1x3 strapping, cladding of some sort. My wall would be air tight and dry to the inside. I am still considering adding more exterior foam and likely switching from XPS to EPS.
Anyways, he told me that another good way to do it is to it is to have it dry to the outside. His assembly was drywall, 2x3 strapping, 2.5" batt, air tight poly, 2x6 structural wall, dense pack cellulose, osb w/ 1/8" gap between boards, 2" of EPS (not taped), Tyvek, 1x3 strapping, cladding. His air/vapor barrier was on the inside and all the insulation except the 2.5" batt was outboard. He was adamant about gapping the OSB and not air detailing the foam so the wall could dry to outside and I know he has designed and built passive spec houses before. I can see how that assembly would dry nicely but wouldn't we want to keep it from getting wet int he first place? I suppose the WRB would do that to a degree but wouldn't that be susceptible to wind washing? The only true line of air defense would be the poly layer wouldn't it and at that point you already have moisture in your wall right??
Maybe I just misunderstood him that has been known to happen.
Barry,
Q. "He was adamant about gapping the OSB and not air detailing the foam so the wall could dry to outside."
A. The person you were speaking with was confusing diffusion drying with ventilation drying. If you want to dry an assembly by ventilation drying, you can. But that approach carries so many risks -- risks of poor energy performance and moisture accumulation -- that no one recommends that approach. When we speak of "drying to the exterior," we're talking about diffusion drying, not ventilation drying.
You never want to encourage air to flow through your wall assembly. Air leaks are bad. They increase your energy bills and raise the risk of condensation and rot.
To limit air leakage, it's important to seal the seams of the different layers of your assembly. OSB seams should generally be taped. So should rigid foam seams. That reduces air leakage. The lower the air leakage, the better.
We encourage diffusion drying by choosing vapor-permeable materials. Encouraging air flow through your wall is nuts.
There are a few applications for ventilation drying. Ventilation drying is used to remove moisture from the underside of roof sheathing, and ventilation drying is useful in a rainscreen gap to help dry the backside of the siding and to help dry damp wall sheathing. But you don't want ventilation drying through the layers of your insulated wall assembly.
If you want to employ ventilation drying on the exterior side of your wall assembly -- by including a ventilated rainscreen gap -- you must have an exterior air barrier to prevent the ventilation air from sucking heat from your wall assembly. That's why OSB seams and rigid foam seams need to be taped.
The only thing gapping between the foam & OSB does is provide another capillary break, but at the expense of creating a thermal bypass. The best location for the capillary break is on the exterior (where rain/dew can actually accumulate)- a gap between the siding and foam (aka "rainscreen" assembly.)
The OSB will stay warmer & drier if you keep it all air tight, with the foam tight to the OSB, and a smart vapor retarder rather than polyethylene on the interior.
If you are mounting the windows "innie" with the glass roughly coplanar with the OSB rather than with the siding, using a crinkle type housewrap between the foam and OSB, lapping the window flashing properly with the housewrap is a good idea. A crinkly housewrap provides a drain path and a moderate capillary break between layers with a negligible amount of thermal bypass.