Double stud wall advice for Zone 6
I know Martin wants us to move on from wall design, but I really need help.
I am in St. John’s, Newfoundland, zone 6, lots of driving rain, little use for A/C in a home, supposedly like Bangor, Maine.
Double stud seems like the least expensive option for a high R value wall. I’ve already priced dense pack cellulose installed, and quite frankly I can buy and self-install mineral wool batts at about 70% of the cost, so given the advantages of mineral wool, that is my preferred option.
Interior vapour control has to be 6 mil poly. So it can’t dry to the inside, and I have no buffering capability in the mineral wool, so it must dry to the outside.
I see 2 options for sheathing – 1. plywood, 2. fibreglass faced gypsum like Denseglass Gold. OSB seems to be too risky from everything I’ve read, so although cheaper, it’s not an option.
So, from the inside to outside the wall will look like the following:
1. gyproc
2. 6 mil poly
3. service cavity of 2×3 or 2×4, with R8-R14 insulation
4. full insulation space of 2 1/2″ or 3 1/2″ with another R8-R14
5. 2×4 exterior load bearing wall, with R12-14 (I may go with fibreglass batts)
6. plywood sheathing
7. tyvek
8. vinyl siding
Most every article I see uses cellulose and OSB, and they are considered risky.
Questions:
1. is this a robust wall?
2. if not, what do I need to change?
3. If I need to change anything, can it be Tyvek to #15 felt? Can it be plywood to Denseglass Gold?
Thanks,
Steve
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Steve,
Relax. It's a robust wall.
Either Tyvek or #15 felt will work. Either plywood or DensGlass Gold will work.
Steve: Is 6 mil poly a specific requirement or could you use a smart vapor retarder? Whatever you do, make sure you carefully detail your air barrier so it's continuous from walls to ceiling. Our primary air barrier was a membrane attached to the exterior face of our interior stud wall. It was pretty easy to do and allowed almost all wiring and plumbing to be inside the air barrier. We also carefully taped the (Advantech) sheathing as a redundant air barrier.We're in Maine.
thanks Martin, that was amazingly quick.
So is one better than the other?
In the articles and studies I've read (and there are many), is it the OSB or the cellulose that makes those other walls risky, or is it the combination of cellulose and OSB?
To Stephen - comment #2
The 6 mil poly is a specific requirement. That said, I need to see if say Certainteed Membrain has a CMHC (Canadian Standards) test report because if it does, I can probably substitute.
I also want to use the plywood as an air barrier.
We are limited apparently to only 20% of the insulation inside of our VB, so I can't put it on the outside of the interior wall. If I do, I would need to leave the interior wall insulation free. I know people talk about the 2/3 rule, but apparently that is a generalization and depends on the climate zone.
Steve,
Q. "Is it the OSB or the cellulose that makes those other walls risky, or is it the combination of cellulose and OSB?"
A. Technically speaking, neither the OSB nor the cellulose makes the walls risky. What makes the walls risky is the lack of an adequately thick layer of continuous exterior insulation to keep the sheathing warm during the winter. Cold sheathing accumulates moisture.
Building scientists now know that the best way to avoid moisture problems in walls is to protect the sheathing with an adequately thick layer of continuous exterior insulation. If you skip the exterior insulation, you have to choose your materials carefully to avoid problems.
According to the short-sheet spec for Canada the MemBrain product is covered in CCMC Evaluation Report #13278-R :
https://www.certainteed.com/resources/30-26-074.pdf
https://www.certainteed.com/resources/MemBrain_CCMC_13278-R.pdf
Cellulose is somewhat protective of wood sheathing & framing due to it's ability to wick, redistribute & store quite a bit of moisture without losing it's insulation function, but that capacity is not infinite, so some sort of interior side vapor retarder is necessary in cooler climates when there isn't sufficient exterior-R.
OSB isn't particularly more risky than CDX plywood. Though the latter can offer somewhat more drying capacity toward the exterior when the local humidity is high, there is very little practical difference from a moisture handling point of view.
https://www.greenbuildingadvisor.com/sites/default/files/Vapor%20permeance%20of%20plywood%20and%20OSB%20-%20APA.jpg
Densglass is much higher permeance, and also not much affected by mold, though even that would require an interior either ventilated/rainscreened siding (vinyl siding meets those requirements) or an interior in zone 6 under IRC 2015 prescriptives.
https://up.codes/viewer/wyoming/irc-2015/chapter/7/wall-covering#R702.7.1
The lowest risk would clearly be Densglass sheathing, cellulose, and MemBrain, but I doubt it's possible to dense-pack the cellulose against Densglass without it buckling/bowing under the pressure. With rock wool that is not an issue. With 6 mil polyethylene (and not MemBrain) rock wool insulation and highly permeable DenseGlass there could be condensation issues during the cooling season if the house were air conditioned, but for the 20 minutes out of the year that you might want to air condition down to something below the outdoor dew point it's not a big concern.
https://weatherspark.com/d/29672/8/8/Average-Weather-on-August-8-in-St.-John&%2339;s-Canada#Sections-Humidity
Steve M.
Why is 6 mil poly a specific requirement? I'm not familiar with Newfoundland's code, but unless they have modified the NBC in way quite different than BC's, the required vapour barrier can be a number of materials, including coatings on the interior drywall, rigid insulation, membranes etc. The only requirement I can see in 9.25.4.2 is that they have a permeance not greater than 60 ng/)(Pa.s.m2)
To Dana #6 and Malcolm #7 - that report speaks to the 2015 code, and also mentions something about limits on wind pressure. I tried to find the wind pressure for St. John's by searching for Appendix C, but I can't find anything. As the windiest city in the country, I wonder if that would cause it to not be approved in St. John's?
Also, St. John's, as of today, but expected to change very soon, is under NBC2010.
The other point to note is this is my first build in St. John's, and first in 25 years. It would be almost impossible to find someone in the city that has used anything other than 6mil poly, so inspectors are likely not familiar with it in the field. I'm not looking to trail blaze here because I'll have enough on my plate to keep on top of things.
To Martin #1 - I've read your comment elsewhere that you err on the side of caution, so when you say it is robust, that products need to be carefully chosen when not using exterior insulation, I'll stick with the original list of products.
Thanks all for your input.
Steve,
You have highlighted the main advantage of using poly as an air and vapour barrier in Canada - which for now I don't think it should be discounted: Poly is part of a successful air-sealing strategy that the industry is used to using, can be inspected and tested easily and has been integrated into the regular sequence of construction.
I know Martin disagrees with this for good building science reasons, and in the long run the advantages of other methods and (hopefully) code changes will probably see its use diminish or disappear. But for now, in climates where cooling is rare, I still see it as a viable option.
We've built a bunch of houses in southern Maine with double stud, cellulose, and either Zip or regular advantech as exterior sheathing. We've opened several up again. No problems.
As long as your insulation installer is good and gets to proper density, you get your exterior details right (flashing, window installs, etc) and keep an air gap (which your vinyl siding will do), you'll be fine.
I see the science behind exterior foam but haven't seen the problems that modeling suggests in the field. As I've said before, I call the "cold sheathing" problem the Yeti of building science - much discussed, rarely seen.
Dan #10 - thanks for the practical field experience. 2 differences, and I don't know how important they are, is plywood vs some form of OSB, and mineral wool batts vs dense pack cellulose. I've found an installer for cellulose, but the price is prohibitive (maybe it's a lot of work or I'm just going to pay for the 'learning curve'), and I can't justify it.
Are you using poly?
Not anymore, Steve, but we did on the first house I built this way. We did a fairly extensive reno a couple of years ago, turning the 3rd floor into an artist's studio, and found no signs of condensation or deterioration (after about 7-8 years).
I'm submitting plans today for the double stud, as follows:
1. vinyl siding (I'd like to do a rainscreen and local spruce clapboard painted 6 sides, but price is likely an obstacle)
2. tyvek (I wonder if there is any benefit to putting a layer or 2 of #15 felt on top of the tyvek? Or maybe this is a definite no-no?)
3. 3/8" plywood horizontally , mid wall blocking, acoustic sealant on all edges. Inspectors may require 1/2", so if they do, 1/2" it will be ($8 more per sheet!).
4. 2x4@16"oc wall with R12 fibreglass or R14 mineral wool (I really want to do the mineral wool, I just need to bite the bullet on the extra cost by saying the advantages are worth it.
5. 2 1/2" space with R10 mineral wool. I priced it yesterday (all comfortbatt), it's the version for steel stud and is significantly cheaper per R4 than the residential version of R14 calculated at R4. By this I mean R10 locally is about 65 cents per square foot, while the R14 is about $1.10 per square foot. So the R10 is about 26 cents per R4, while the R14 is about 32 cents per R4. I'd be curious as to the relative cost of the versions in your area, because it seems to me that Roxul is gouging the residential (wood) compared to commercial (steel).
6. 2x3@17", probably finger jointed because hopefully it's less likely to twist and cause finish wall issues. I know, 17"?, but to use R10 mineral wool, it has to be steel, so 17" it is when using wood. I'll hang the gyproc horizontally on those walls, and it's only an issue when the wall is longer than 12'. But even then, it's not much of an issue. Let's say your wall is 18' - take a 12' sheet and cut off enough so it falls on the centre of the stud nearest 12', so maybe you need to cut off about 5". The remained is covered by most of an 8' sheet. Just remember about baseboard, etc for the 17"oc spacing, but it's only for the outside walls anyway.
The other option is to cut 1" off the steel batts, which apparently given their stiffness is an option and gives you small pieces to place in other spots if needed. I'd need to make a cutting table for this. I'm thinking a couple pieces of 2x6 about 2' long, stand on edge on a mitre saw, cut down about 3 1/2 to 4" (so you can use it for the R14 as well), get a piece of plywood about 2' long x 20" wide, nail it to the uncut edge of the 2x6, making sure you line up the 2 curf marks on the 2x6, and you have a decent cutting board to make sure you cut that mineral wool batt straight both ways. Just think of an old style mitre box.
7. poly, all sealed up nice and tight. I've put down Membrain as an option on the plans, we'll see what the building inspectors say when they review the plans. Maybe they allow it, maybe not. Even if they do, as Dan said, he's opened a house done 7-8 years ago with poly and didn't see any evidence of issues, so I might just stick with the poly. But at least I'd have that option and can weigh it out for a bit before I need to actually make a decision.
8. gyproc
Originally I was thinking R14 in all cavities, for an overall width of 10 1/2", but that gives up a lot of floor area over the standard 5 1/2" walls. My design gives up 8 1/2", and the full cavity of R10 helps immensely with thermal bridging. I think it's a very good compromise, and given the R10 is so much cheaper relative to the R14, price and performance will likely be a very good compromise as well.
Just to clarify, if Roxul were pricing the residential the same as steel, the R14 wood should be about 91 cents per square foot. For my house, I need about 6000 square feet for the cavity and 2x3 wall (or 2x4 if I were using R14). With a difference of 20 cents per sq ft, that's $1200 more.
One final option, I could put 2 layers of R10 in between the 2 walls, that would mean an 11" wall (I'm not doing this because I don't want to give up the space). Comparing the 10 1/2" 3 layer of R14 insulation cost only for R42 nominal would be $1.10 per sq ft per layer - 2000 sq ft per floor of walls - 6000 sq ft of insulation at $6600. For 11" 1 layer R14, 3 layers R10, 2000 sq ft of R14 at $1.10 is $2200, 3 layers R10 at 2000 sq ft per layer for a total of 6000 sq ft at 65 cents per sq ft is $3900, total $6100 for R44. R2 better, $500 cheaper, less thermal bridging (probably insignificant), and all for just a 1/2" loss of space. That's not counting taxes and the lower cost of 2x3 over 2x4.
Maybe Roxul prices it differently in your region and so this analysis may not be relevant. We're on an island, so transportation is the reason roxul is so expensive here compared to elsewhere.
Steve,
From my local big box store, Roxul ComforBatt for 2 x 4 walls (r15) is $.72/sf by the bag or $.50/sf if you buy by the pallet. I use ComfortBoard 80 that I get from local building supply stores cheaper than the big box and that is just under $.60/sf/inch thickness (R4).
Re: 17" spacing. Sounds ok, as long as you're doing the drywalling. If you're hiring a sub for that, they'll probably gripe about it.
Re: felt over Tyvek - why would you do this? Not sure it's a definite no-no, but it's probably a waste of time and money. If you want to spend some more money on that outer layer, sub the Tyvek out for Typar Metro Wrap. I don't know the exact price difference, but I was told it's not huge. The majority of my house was done with Metro Wrap, with only the triangles below the gable roof ends being done in the regular stuff because it was done at a later time. The regular stuff looks it's been through a war, while the Metro Wrap looks brand new, despite being exposed to the elements twice as long, and the regular stuff having some added protection from the eaves.
Steve,
Q. "I wonder if there is any benefit to putting a layer or 2 of #15 felt on top of the tyvek?"
A. I agree with Trevor: Adding one or two layers of asphalt felt to the exterior side of your Tyvek housewrap would be a big mistake.
The exterior sheathing installed on double stud walls gets damp in February and March, and the sheathing needs to dry outward in April and May. Adding one or two layers of asphalt felt (which, when dry, has a vapor permeance ranging from 0.5 to 5 perms) is not a good idea, since the asphalt felt would slow the rate of outward drying. You want one layer of housewrap, and that's it -- and you want housewrap with a high vapor permeance rather than housewrap with a low vapor permeance.
For more information on this issue, see these articles:
Monitoring Moisture Levels in Double-Stud Walls
How Risky Is Cold OSB Wall Sheathing?
Two Views of Double-Stud Walls
How to Design a Wall
Steve,
I'd suggest going with 7/16" or 1/2' plywood regardless of what the inspectors want. 3/8" plywood is a nightmare to work with. It is wavy right off the pile, and almost impossible not to over-drive nails into. I haven't seen it on a site, except as floor underlayment, since the '80s.
Are you sure you can't compress the batts enough to go with 16" oc interior framing? Jumping to a spacing like, 19.3", that still divides by 8 ft is one thing. having one that has no modular basis is - well - strange.
Rooms longer than 12 ft always have the drywall sheets run off the studs, even with 16" spacing, because the interior studs don't start at each partition, they are continuous along the whole exterior. Most drywallers use a backer board, rather than cut sheets. This helps them get a flush finish at the butt- joint
The previous house I lived, built in 1889, had random spacing. I found this out when I had to locate the studs to blow in insulation. Now that is really weird. I suspect the reason he doesn't want to compress the R14 batts into the cavity is because of cost. It's 40% more mineral wool, apparently also costs more per R to begin with, and my understanding of batt insulation is if you compress it you don't get much more R value than the value of uncompressed stuff at that thickness (i.e. you still end up with barely over R10, in this case).
To Trevor #16 and Martin #17 - ok, just a single layer of housewrap. I'll have to check in Metrowrap.
As to spacing, I won't be doing the gyproc, so I'll have to see if it can be compressed without causing it to not fit properly.
To Malcolm #18 - 1/2" adds another $500 to the cost. But hey, it's only money. I'd rather have the siding looking flat than wavy, so it's money well spent.
To Trevor #19 - not sure I follow you. My outer 2x4 wall will have R14, so that fits fine. The inner 2x3 wall is where I plan to use R10, but it's sized for steel studs, so an inch wider than the spacing for wood studs. What I need to deal with is the extra inch in width. Suggestions seem to be don't put it 17"oc, keep to something standard and either compress the R10 an extra inch (which might not give a good install), or cut it off. I think I'll go with 16"oc and then figure out from there the best option to deal with the extra inch. Either that or go to Roxul and try to get them to reprice the R14 for me so it is the same price per R4 as the R10. Or I could go to another mineral wool supplier.
With the R14 batts, the design depth is 3.5". If you go with 16" spacing using 2x3, you'll have 2.5" depth. The R14 batt will easily compress into that depth, but you don't get R14 out of it, it will be closer to R10.
It's generally a bad idea to try to compress batts sideways, because they can't compress much in that direction, so they have the tendency to fold instead, leaving voids. However, 1" isn't a lot, it might turn out to work just fine.
I've learned that contractors don't like doing anything different, even if it's no extra work. So if you can get the 16" spacing without too much hassle to yourself, it's probably worth the effort.
ok, now I see what you were saying and I understand about the sideways compression. When the time comes, I'll have to experiment a bit and see what works best. hopefully I don't end up with 3000 sq ft of R10 I can't use, and have to tear down 2x3 walls to install 2x4!
The price per R for the R14 is about 25% more than the price per R for the R10 - that's probably confusing! Let me say that for the R14, I would pay about 8 cents for each R1, whereas for the R10, I would pay about 6 cents for each R1. Hopefully that's more clear.
You've all convinced me on the 16" centres. I don't mind manual work, but gyproc is too heavy for 1 person, even if all I did was the outside walls.