Domed roof trusses constructed of plywood

[TroyD]

I'm looking for some input on this unique roof concept. A custom home builder proposes to construct a domed roof over the front porch, using "trusses" made of multiple layers of 3/4" plywood sheathing cut to achieve the desired shape. I am contemplating what it would take to show that it works and to satisfy the local building official. I have made a few comments in red. Obviously, multiple layers of plywood glued and nailed together would make a stiff member. The clear span is barely 8'-0", so if the contractor could get 12 ft long panels they would span the full distance. Could maybe even get by with 2 layers. I assume the roof sheathing will also be plywood, but I don't know how the contractor intends to attach the roof sheathing to the "trusses", or get the desired roof sheathing curvature. Some coordination is needed to address

I have the APA's plywood structural properties info that I can use to determine the composite member's E, I, Sx, etc. the check bending, shear, and deflection. I need to research fastener spacing requirements and adhesives for this application.

What other engineering concerns might I have overlooked? Project is SE Minnesota, ground snow load is 50 psf.

-Troy

 

[Celt83]

Search APA for their guide on the design of plywood box beams

https://www.apawood.org/publication-search?q=H815&tid=1

https://www.apawood.org/publication-search?q=S812&tid=1

Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

 

[JAE]

I would suggest:
1. Offset the plywood sheet splice edges from each other in each lamination. So a different cut-out pattern for each.
2. Add 2x12's on the top edge, also cut on a curve on the top, and nailed or screwed into the side of the plywood to provide a proper nailing surface for your roof sheathing. Nailing into the edge of plywood probably has close to zero uplift resistance.
3. Look at lateral stability of the plywood truss - probably not to stable out-of-plane - several lines of bridging perhaps?

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[Triangled]

you can get a glulam fabricated with that radius

 

[TroyD]

Thanks for the input. Depending on my fee quote, the contractor may defer to a metal fabricator for a custom frame assembly. Celt83: I'll spend some time looking at the APA links you provided. JAE: I agree with you about the lack of uplift resistance in nailing into the plywood edge. Some 2x material will be needed for that as well as for lateral bracing.

 

[BridgeSmith]

What I've seen most often done for something like this, is to put pieces of 2x lumber along each edge between plywood 'facing' pieces. It leaves flat surfaces on the outside and eliminates the prying action in roof sheathing uplift.

I've also seen OSB used more commonly than plywood. Don't know if that was cost-driven or if the OSB has a higher shear capacity.

 

[Agent666]

With cutting the shape out of sheets of plywood, you also need to consider that the grain in the plys isn't going to be optimally orientated to the stresses in your "curved" member when its bending. (may be why OSB is used Hotrod, due to the more random orientations of constituent wood bits in OSB)

Like Triangled mentioned, I would have thought glulam was a better technical structural solution to achieve the form, though probably more expensive? They can form very tight radii if the laminations are thin enough (like 10mm or so).

I'd also mention simply cutting it from a slab of LVL of an appropriate thickness required for stability might also be a consideration, saves you having to glue multiple thicknesses of ply together with the inherent weaknesses at joints. You can pretty much get LVL from manufacturers at any size/thickness you desire.

 

[Ron]

Agree with JAE...would likely take 2x12's to cover the curvature, but nevertheless, they would be necessary for nailers.

 

[gte447f]

I think you have a pretty good handle on it in your original post. Add the 2x nailers along the curved edge as JAE says. Maybe some blocking perpendicular to the span. It would be nice if the ribs (trusses) could be cut from a single sheet of plywood without needing to splice. If that were possible, then I wouldn't be surprised if a single ply rib would work assuming they are spaced at 16".

 

[BridgeSmith]

If you have 2x's along the inside and outside radii, faced with plywood (or OSB) on both sides, you should be able to construct a sturdy arch by offsetting the joints between plywood pieces and the ends of the 2x pieces. Glue and screw to make a laminated assembly.