Poking holes in a double-wall wood-on-gravel frost-protected shallow foundation assembly in 4A
Please help me poke holes in this assembly to see how metaphorically (and literally) air-tight it is. I have obviously gone down quite a few rabbit holes around here, but my AHJ is supportive and happy to let me do anything that fits into code even if he is a bit skeptical.
(Climate zone: 4A – Snow load: 25lbs – Wind design: 115mph – Seismic category: B)
We are designing and building our first home, a 24’x36′ gable relatively easily construct-able rectangle avoiding concrete, foam, and engineering where possible – under the 2018 IRC.
(Edit: Updated detail following feedback in comments also attached – Moves walls onto floor, adjusts floor insulation layering and protection and other changes)
The design (attached) is a 2″ gap 24″oc double 2×4 wall with plywood sheathing beneath a rainscreen and metal cladding. The floor is 16″oc 2×6 with a rim joist against the inner (structural) 2×4 wall. The wall sits on a treated 2×10 footing plate, the floor on treated 2×6 sleepers, and both sit on layers of XPS above the compacted gravel drainage pad. The site is on undisturbed earth at the smooth top of a hill with clear effectively unlimited drainage in all directions.
Taped sheathing is the intended primary air barrier, and drywall the secondary. The WRB is intended to be vapor permeable to maximize drying while controlling water.
I’m also looking for feedback on the clarity of my details. SweetHome3d wasn’t made for cross-section, but has still been much easier than sketching or mspaint.
The inspiration for and bulk of this design comes from the details towards the bottom of this article: https://buildingscience.com/documents/building-science-insights-newsletters/bsi-020-wood-foundations-picasso-does-foundations
Also, if you have any ideas to help me finish the “I’ve never seen that done” bingo card, I’m happy to see them.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
The biggest potential issue I see is that your walls and roof are bearing on only 5" of foam and soil. Dr. Joe's designs all have at least 7 1/4" of bearing area. I would want to know the actual load on the foam and soil at the perimeter to ensure there will not be settling. I'm almost certain that regular (15-psi) XPS won't stand up to the load long-term. 25-psi foam might be enough. In my area, with snow loads, it would need to be at least 40 psi.
I am also concerned about the potential for the top of the gravel to push the vertical foam outward under load, over time. By code, untreated framing lumber and sheathing needs to be at least 6" above grade, and I aim for 8-12." Joe is using foam and plastic boards to get around this requirement. There are plenty of slab-on-grade buildings out there with the framing right at grade, but they almost always have moisture-related problems sooner or later.
Be sure you can fasten your trusses to your top plates. Truss screws from above are usually, but not always, enough. Sometimes hardware is required.
15 psi foam is probably cutting it close under the floor as well. I use at least 25 psi foam in similar situations.
Overall you have a creative, thoughtful design and I don't see issues with your other details.
Thanks for the feedback!
5" wall & roof bearing: I may not need and am not set on that much foam - 4" was an arbitrary bump over the ~1"1/4" likely minimum I could gather from the revised guide (1) due to adding under-floor insulation. Admittedly, I did not absorb most of it.
I could also bring in the bearing plate 3.5" and set the outer wall on another 2x4 and tie them together.
Gravel pushing foam out: I'll try to think of ideas to mitigate that, nothing is coming to mind and I'd welcome ideas. I saw a picture of a sizeable rock or rock-like paver skirt next to the structure instead of simply compacted granular fill.
Untreated wood 6" above grade: Aha, I didn't notice that! You say framing and sheathing, and I see the plastic board section, but the framing lumber itself doesn't seem special. Am I missing something? Would using ground contact sheathing for the bottom 2' fulfill the same role as the plastic boards?
Moisture problems: That makes sense. Do you think a well sloped site and attention to detail especially in the water control layers and drainage is enough to reasonably mitigate this?
Truss fastening: I liked tying the two walls together at the top with the plywood, but can bring it in to the inner wall and use hurricane ties, or perhaps slice some holes for them through it? Honestly, I'm not sure how I am planning to actually accomplish the plywood layer.
(1) https://www.homeinnovation.com/~/media/Files/Reports/Revised-Builders-Guide-to-Frost-Protected-Shallow-Foundations.pdf
Unfortunately, I agree with Malcolm's advice--you have a great start on a creative design that is close to being buildable. I recommend finding a structural engineer to work with on the remaining details.
Fair enough. I appreciate you giving it some consideration and sharing your concerns!
I may try to work on some of the feedback here and check in with the AHJ, and failing that there's always a traditional stem wall.
Sometimeslcy,
It is an interesting design. I would be a lot more comfortable if the finished grade was a few inches below the lower layer of foam, meaning it didn't rely so heavily on the vapour-barrier to stay dry.
I know you want to avoid engineering, but I doubt this fits within any of the prescriptive paths in the IRC. You are probably going to need to need some calculations on the bearing capacity of those exterior footing-plates, and the substrate below.
The house is also unrestrained by the grade around it against horizontal loading from things like high winds and seismic if those are concerns where you are.
I agree, Malcolm. There are a lot of unusual ideas here and failure of any one of them would be a big problem. Structural engineers are worth their fee, though it can be hard to find one willing to think creatively or take on the risk of a project with this many non-standard details. I'm fortunate to have two I can call on. In fact a chunk of my time today was corresponding with one of them, and several highly experienced builders and architects, for advice on a more conventional slabless slab; there are a lot of variables to consider.
That's reasonable. Would you perhaps be willing to share contact information for you two engineers if they are licensed in Virginia?
Finished grade below foam: I'd be happy to use comfortboard too, but I don't think my AHJ has seen it used for pad insulation - so I should find the appropriate documentation for it, I know I've seen it recommended and used in quite a few underslab projects around here.
That said, would you be more comfortable with it knowing the house is on a pretty large hill with clear drainage in all directions?
Code & Engineering: I was optimistically hoping it would fit with Section R504 for the floor and R403.2 for the footing plates. Narrowing the foam and opting for a higher pressure rating like Michael noted.
Horizontal loading restraint: Wind specification is 115mph, seismic B. I should add that to the question! I was planning to look into anchoring the footing plates with something like this: https://www.foundationtechnologies.com/products/manta-ray-earth-anchors/
Sometimeslcy,
I'm intrigued by this, much like I was with concrete-free slabs. In some ways it seems like their logical extension, but it takes some time to get my head around it.
I wonder if it wouldn't be simpler to make it closer to conventional platform construction by building the floor, and setting the walls on top? That would:
- Tie the whole structure together.
- Leveling and straightening the walls would be less difficult.
- Make building and standing the walls easier.
- All the walls, exterior and interior would be the same height.
- Mean you had limited the unusual features down to basically the sizing and composition of the footings.
Malcolm_Taylor,
That's a really interesting idea, especially any implication it has on the acceptance of footing details, which have been the most contentious area in this plan so far. See (A)(including its previous incarnation, which was.. I believe the building official referred to it as a "Balloon framed wood foundation".)
I feel the same way, but coming at it from having no background in architecture or structural engineering. I love learning new things and optimizing, though. It makes me wish I had a building science nerd friend to theorycraft with.
From my perspective, it's "just a simple combination" of three well documented conventional details: Frost-Protected shallow foundations, Double walls, and Wood foundations.
The trick being wood foundations are very unusual as you said, and almost always combined with a basement, due to the trends at the time from what I can tell.
The floor is effectively a PWF basement floor, in code as "Pressure Preservative-Treated Wood Floors (On Ground)" I believe.
----
Making it more conventional by building walls atop floor:
Coincidentally, I just watched this (1) lecture by Steve Baczek today, and in it he shows his WIP for his "Next generation Perfect Wall" (2).
Would it? It's certainly something that could be sacrificed to make the plan more easy to achieve. I hadn't considered it, due to habitually trying to avoid the floor-exterior thermal bridge.
- I was expecting to use equipment to anchor the studs to the rim joist and the rim joist to the floor joists. The two walls are tied together with the footing plate. Sitting directly on the floor would simplify the connections, though, that's true..
- Leveling and straightening them seems like an advantage, building them and standing them could be easier.. Either way, the footing plate has to be extremely level at some point. (A) I do wonder what it would look like, though. I'll have to work on a version like that and see! It should at least be more visually understandable.
- Could finish the floor first without it being beneath the walls, too.
- Both walls are currently the same height structurally, or am I missing what you mean?
- I'd be happy to have it be more usual. Considering it, the foam could start beside the footing plate and go downwards, instead of being beneath it. Theoretically, that's where it belongs anyway. Either way, it has to withstand the compressive force of compacting the gravel on both sides, but that would get it out from under the plate.
(1) https://www.youtube.com/watch?v=yoOQOjQ8XTM
(2) attached
Sometimeslcy,
I like Malcom's detail, the only change I would do is to use a layer of stone (something like 8x8 limestone blocks) as the base around the perimeter. You can infill this with gravel and build on top.
This now gets everything that 6" off the ground, plus the stone will hold up to a lot more abuse.
Akos,
Just playing around with the idea of eliminating as much extra stuff as possible. Two layers of sub-floor floated over the foam. Being risk adverse, this is well into the territory that would make me quite nervous
Malcom_Taylor,
The more time I spend with your suggestion, the happier I am with it the idea. Thoughts about this version based on your sketch?
I also incorporated truss to wall ties from Michael Maines and a stone block perimeter from Akos in post 13 (Thanks for the suggestion!)
Sometimeslcy,
That looks good. I think Akos was suggesting the pavers as a short perimeter retaining wall to contain the gravel above grade. I guess you could also use PT 2"x for that.
Re #17
Aha, that makes sense. Adjusted that on my end. Thank you!
I considered the idea your other sketch myself, but for my part even I balked at only having two (lapped and fastened) plywood layers as the only lower lateral bracing for the walls.
I would try it with a stem/concrete foundation wall, though.
Sometimeslcy,
Agreed.
I always worry about termites and other vermin when you build anything on the ground without a concrete "rat slab". I have a barn built with an oak floor over gravel and with foam sheets behind the siding...It is fine for a barn but as a house......I would worry....
If you want to build a concrete free (or almost free) foundation, have you thought about doing it the way it has been done for decades: on wood pilings driven into the ground or augered in like a pole barn and with a raised wood floor a few feet off the ground?
On pilings is something I've considered, just not very much. Adjacent to a raised house on groundscrews, which I have avoided due to the goal of being able to do all of it myself and avoiding the necessity of engineering (while pushing up to the edge of it, admittedly)
It just isn't the way I went with it, but I am interested, similarly to being interested in timber frame structures. Those just require engineering outright, unless I'm mistaken. Seem to have several nice advantages, though.
Termites aren't very present here, and rats.. that's fair, but in this case there will be metal all the way to the bottom of the vertical insulation buried in compacted gravel. I hope that's enough.
Sometimeslcy,
You could always start your floor stack-up with a layer of hardware cloth below the foam. For insects you could use Stego's VB under the foam too.
https://www.stegoindustries.com/pango-wrap-termite-vapor-barrier
Malcolm_Taylor,
That's a good idea. The one thing I'm keeping in mind is all the PWF floor documentation I've found strongly suggests not sealing the vapor barrier to allow eventual drainage of any unexpected liquid.
I was reading somewhere else on here earlier tonight and there was a snippet of a very old book about wood floors for factories which suggested venting the subfloor area to allow drying. Perhaps more like an insulated crawlspace in this case, and still very useful in case of internal flooding, and could allow the fully sealed Stego barrier.
That looks like interesting stuff. I noticed that it claims to be a radon barrier-another important consideration.
As for hardware cloth for rodents, that is also a good idea. It is interesting how small you need to make the mesh openings given how tiny mice can make their bodies to slip through openings that seem impossibly small.