critique my wall assembly
Trevor_Lambert
| Posted in General Questions on
Hi,
I’m planning a workshop/garage in zone 6. It will be fully conditioned. I will be doing most of the building. The wall will be resting on some variety of shallow frost protected foundation, the exact details I haven’t worked out yet, possibly something like this
https://www.jlconline.com/how-to/foundations/a-frost-protected-shallow-foundation_o#:~:text=The%20insulation%20around%20a%20frost,foundation%20from%20freezing%20and%20heaving.
but with glass gravel under the slab being the compacted fill.
I’ve attached the proposed wall cross section, one from the side and one from above.
I welcome any suggestions or criticisms.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Im sure it will keep the place warm, but it looks overly complicated and difficult to construct. There are 2 assemblies that are considered standard construction and high performance- frame wall with exterior insulation, or double stud wall. I would consider one of these choices.
I would say this is a version of a double stud wall. The only difference I can see between this and a more traditional one is it overhangs the slab. What specifically would you recommend changing? Moving it all onto the slab kills a pretty significant amount of useable floor space, and I'm not sure how much work it really saves.
Exterior insulation isn't very environmentally friendly, nor is it cheap.
A 3-inch-deep rain screen is poinless overkill cut it back to ¾ of an inch.
Note it is unlikely that your R45 wall will ROI in under 25 years unless your fuel is very expensive and impossible if heated intermittently.
Note the foot thick wall increase the square footage of the building as measured by the tax man by 5 to 10%
Walta
The rain screen is 1.5 inches, not 3 inches. The reason for 1.5" was to account for the bulging that the dense packing of the cellulose will cause.
I haven't plugged the numbers in yet, but the R-value of the assembly is more in the ball park of 38 than 45.
My first critique would be that it just looks really complicated, and time consuming to build. I'm sure you could, but you've got to plan for the appropriate amount of time, which involves weather, and potentially multiple seasons.
My first priority for a workshop would be to keep critters out. I'm not sure how well that is possible with fiberboard. If you are looking for permeable options, 1/2" denseglass might be better, it is also very easy to tape to seal it up and quick to nail up with a roofing nailer or an autofeed drywall screw gun.
I can't see anything above a 2x6 or 2x8 wall being worth it for a shop. If you want more R value, I think something simple like a 2x6 load bearing Mooney wall with 2x4s on edge for extra insulation and service cavity.
The idea is that critters won't get to the fibreboard layer. It's blocked on the face by the siding, and at top/bottom with bug screen. The permeable layer is housewrap, and the only purpose of the fibreboard is to help contain the dense packed cellulose in combination with the housewrap.
As for "worth it", the intention is that this workshop will be conditioned like a house 24/7. I feel like I already scaled back, since our house has R-56 walls.
I looked up Mooney wall, and it seems like something created mainly as an easy retrofit, not necessarily optimized for initial construction. Lots of thermal bridging, and with an effective thickness of only 9", not really what I'd consider high performance.
I'd consider the PERSIST approach if you're looking for something that's easy to build and has the potential for higher R value.
For a shop with loads that are easily manageable, why not something like a 2x4 cavity wall, with as much low density insulation on the outside as needed? It's continuous, buildable, and common to trades. It's not all that different from your proposed assembly, just without all the gussets, caivities, and chases.
I'm not sure what you mean by "loads that are easily manageable".
I've taken a look at the description of the PERSIST approach. My immediate reaction to it is twofold: it's an environmental catastrophe, and it's expensive. That's a massive amount of foam for both the walls and roof, plus the peel and stick membrane.
By loads that are manageable, I only meant that a 2x4 wall can easily support the weight of the roof of the structure.
I wouldn't call it an environmental catastrophe, by any means. Building anything at all is pretty rough on the environment, and EPS or polyiso isn't the worst choice you could make in terms of insulation.
2 x 2" layers of polyiso gets you an automatic true R24 wall. Add R13 fiberglass insulation to the stud cavities and you'll get another R10 ish. That's a true R34 wall, for somewhere around $2.25 / sq ft in insulation, and can be easily sided using 1x4 furring strips with 6.5-7" headloks.
The total thickness, drywall included is around 9.5", with most of it being on the exterior, saving your interior shop space. It uses less wood, and is far easier to construct.
I admit that "catastrophe" was overly dramatic, and would only apply to something like XPS or HD poly foam. However, cellulose is still far better than polyiso foam.
Polyiso is not common around these parts, and I don't see prices in line with what you're saying. The one place I could find it put the price at $7 a square foot for the polyiso alone. Cellulose comes in at around a dollar.
I think I'd choose a traditional double stud wall on a 2x10 over that assembly, yielding similar R-value at lower cost and much lower embodied carbon. Using the inner wall as the load bearing, the same amount would be on the exterior, relative to the slab footprint.
I'm a bit west of you and 2" roofing polyiso is about $1/sqft. So roughly an R30 wall needs only about $3/sqft of polyiso. Not cheap but not that expensive either.
I think how you can make this work in terms of cost is if you can get lumber and board sheathing for cheap from a local sawmill. Much easier to hit studs through thick foam when they are real 2x.
The one benefit of the Persist is you don't need an interior finish which does save significant dollars. You can use the stud bays as shelving as a bonus.
Wiring (either AC or conduit) won't be fun though.
For a house there is a comfort argument for high R value walls. For a workshop, it is mostly about operating cost.
Say you are building a 40x50 shop with 10' walls.
Assuming you are around 7000HDD, the loss through a 2x6 wall in a season is about:
7000HDD*24h*1800/R18=168Therms or 5216kWh.
Assuming my local rate of $0.16/kWh for electricity, heated with COP 3 heat pump, the cost is ~$280 per year.
Even if you bump up your walls to R36 to halve that loss, the savings is only $140/year. Even if you DIY, you'll never recoup the cost of a more complicated wall.
My suggestion for the Mooney wall was based on the assumption of wanting a horizontal service cavity with a buried mid wall vapor barrier.
I've done the math on cross strapped wood framing and it is very efficient, the only thermal bridging is where the studs cross which is very small fraction of the overall surface area, you get about 95% of the assembly R value a similar wall with continuous insulation.
I don't think building any wall with a service cavity is worth the cost though, including the Mooney wall, I go back to my original suggestion of a simple 2x6 (R18 assembly) or 2x8 (R24 assembly) walls 24" oc for a shop.
I'm planning it becoming an ADU at some point in the future.
2x6 and 2x8 walls are not R18 and R28. I plugged the numbers, using 24" stud spacing and R3.7/inch insulation, into an r-value calculator and got R15.7 and R20.4 respectively.
A double stud 2x4+2x3 wall on 2x8 plates comes in at somewhat less than R24 (the calculator uses layers, so I wasn't able to account for the top and bottom plates being continuous).
I'd be curious to see your math for the cross strapped wood framing. Seems to me that it doesn't matter a whole lot that the contact area on the studs is minimal, you still have a framing factor for each section that is equivalent to a standard stud wall. Imagine you have two stud walls, each with an effective R15. If you place them back to back, I think you'd agree that you'd have R30. I don't think rotating one wall 90 degrees changes all that much.
You can run it through https://www.ekotrope.com/r-value-calculator . This was using HD batts, so R4.2/inch. R3.7/inch would be a bit less. The results are pretty close the worked numbers from:
https://cwc.ca/en/design-tools/effective-r-calculator/
Once you include siding, drywall and air films you do get R18 for 2x6 and R24 for 2x8.
Find attached the calculator. This is set for 16OC for both horizontal and vertical studs. Currently set for R 3.6/inch insulation, you can edit the formulate to use your own.
>"Seems to me that it doesn't matter a whole lot that the contact area on the studs is minimal, you still have a framing factor for each section that is equivalent to a standard stud wall. Imagine you have two stud walls, each with an effective R15. If you place them back to back, I think you'd agree that you'd have R30."
The series heat flow for any give cross section makes it so that the minimal contact area does matter. Rather than picture this like layers of an onion with each layer being a homogeneous R value, it needs to be pictured like a bunch of parallel paths with independent series analysis for each path (or each unique path, weighted by area). Like stacked cord wood.
Similar to an electronic circuit with 4 resistors, 2 in series and 2 in parallel:
Resistors 1 and 2 are in series. Resistors 3 and 4 are in series. Resistors (1 + 2) are parallel with (3 + 4). To calculate total circuit resistance:
1/Rtotal = 1/(R1+R2) + 1/(R3+R4)
I question whether things like siding and drywall should count, especially if the drywall is not air tight and the siding is outboard of an open air gap. In my case, I'm not even going to have drywall. In any case, in order to keep an apples to apples comparison, it makes sense to leave out such items that either a)will be common between different wall types or b)may not apply.
Your drawing shows both interior and exterior finish. No matter if it is air tight, drywall is the same R value per inch. Even if you use OSB/CDX for the interior finish it is about the same R value as drywall.
Each is only about R0.5, so not a lot but not zero. You can subtract R1 from each wall, so 2x6 is R17, 2x8 R23, 2x6+2x4 Mooney wall with R3.7 cellulose R31.
Interesting assembly. I've done similar, though without the added service cavity at the interior or the fiberboard at the exterior, with good results. I had spec'd a double 2x12 mudsill but my engineer required a solid 3x12. And I had already spec'd taller gussets so they would better transfer loads on the outer wall to the structural inner wall. The licensed engineer and the strict code official were both happy. We're using the same system again on another project now, though no engineer is involved.
I don't see a need for the exterior fiberboard. It will slightly limit outward drying without significantly raising the temperature of the condensing surface, and it's expensive. I love wood fiberboard but I would spend that money elsewhere.
The only purpose of the fibreboard is to reduce the bulge of the dense packed cellulose between studs. It can certainly be done without it, and I've seen an example house. I haven't priced out the fibreboard*, but if it is indeed expensive then I'd consider deleting it. Maybe there's some other material I could use that is more economical. I just need to make sure the house wrap doesn't come within 3/8" of the siding, which should be doable with a 1.5" gap.
I have a service cavity in my house, and I really appreciated during the build not having to worry about stuff inside the walls, during or after the build.
*I may be using the wrong term for this material. The stuff I'm talking about is far less dense than even cardboard, you could literally bite through it even you had no teeth.
We did a lot of exterior cellulose insulation where I used to work. We used Pro Clima Mento on the exterior, stretched drum-tight, with two layers of strapping to prevent the bulging from being a problem.
I've been thinking about a similar plan. I don't have a ton to add for your wall options. What are you doing for doors? I think one of the biggest losses in a shop is through the garage door. In my case, my car doesn't come in and out every day or anything like that, so I was thinking of building some carriage doors to get a good seal.
On the power side - personally I'd like to have conduit run, keep the wiring semi accessible. Easy to add an outlet in an odd place when needed. I have a number of 3 phase tools that need to be hardwired where they sit, so it makes sense to keep it easily accessible to add a circuit or branch for that.
The overhead door is something I've given thought to, but don't have a final solution for. At the very least, I'm going to use a system that closes the seal tightly. I may add insulation to the door panels, if that seams to make sense. I may have a secondary door inside the main one that would be closed when the vehicles are inside at night. That would account for the period of greatest heat loss, and it wouldn't be much of a burden to open and close on the way to and from the vehicles.
Where I am, I think you can actually have power wire just stapled to open studs in a garage. For me, I it would be much more accessible to have wires inside a service cavity than in a conduit. Making changes inside a conduit seems like an unfun proposition. And so does running the wires in the conduit in the first place, for that matter.
Wall assembly rev 2.
I've taken on some suggestions and pared back and simplified the wall. I haven't been convinced to switch from green cellulose to beige exterior foam. I agree the foam, assuming it's not XPS or polyurethane, isn't as bad as I had first thought, but it's still not great. Kind of hard to beat a negative GWP. I also priced out the materials for my proposed wall and an equivalent r-value stud/exterior foam wall and it's $2000 more for the foam wall, so no incentive there.
Any glaring errors? I know the 2x10 width may seem like overkill, but I'd rather overkill than underkill. The total cost difference between this one and using 2x8 is only about $400.
That looks like a pretty standard double stud wall.
7/16" osb doesn't quite qualify as a vapor barrier according to the OBC and your inspector might still want either poly or vapor barrier paint.
2x3 are hard to find in longer length and even harder to find straight. For the small cost delta, I would go to 2x4s. If you must have 2x3, try to find the finger jointed ones as they tend to be straighter.
Even though the interior OSB will provide bracing for the wall eventually, it won't go up until the wall is wired and insulated. I would include some diagonal bracing somewhere inside the wall.
P.S. There is also no attractive way to air seal OSB, most of the better tapes won't hold paint either. If you want a nice interior finish, you might need another layer of something over the taped OSB.