Flitch Beam Torsion
Flitch Beam Torsion
(OP)
Wanted to see if anyone has analyzed and designed a flitch beam (wood steel wood sandwich) for torsional loads? I am working on a project where an architect wants to hang a 4-seater porch swing from the rafter framing above. The porch rafter is unbraced at the bottom of the beam but has 2x decking nailed to the top. I'm currently proposing a 1" thick steel plate to be clad with 2x material. The porch swing is hung from an eyebolt that will be welded to the bottom of the steel plate. Therefore, I'm assuming that the seat can swing and load the rafter at an angle. To resolve this I'm designing the steel plate currently to take all of the load in strong-axis bending, weak axis-bending, and torsion. The architect is looking to thin out the plate section and hence I'm wondering if there is any benefit to using the wood.
My first reaction is it wouldn't help a whole lot. I know I would need to design the bolts for the shear flow through the section (which should be possible for weak-axis bending, but I haven't given much thought yet to how that would work with torsion). I'm also not convinced the wood would do a whole lot do to the difference in "E" and I've never designed wood in torsion before so not sure if it has very much strength and if cross-grain bending comes into play.
Just seeing if anyone else has gone down this path before I give it too much thought. Given it's a porch swing, I'm also starting to wondering if I'm overthinking the whole thing as there have obviously been lots of these hung from just typical porch framing.
My first reaction is it wouldn't help a whole lot. I know I would need to design the bolts for the shear flow through the section (which should be possible for weak-axis bending, but I haven't given much thought yet to how that would work with torsion). I'm also not convinced the wood would do a whole lot do to the difference in "E" and I've never designed wood in torsion before so not sure if it has very much strength and if cross-grain bending comes into play.
Just seeing if anyone else has gone down this path before I give it too much thought. Given it's a porch swing, I'm also starting to wondering if I'm overthinking the whole thing as there have obviously been lots of these hung from just typical porch framing.






RE: Flitch Beam Torsion
- My gut feel is that your lamination fasteners wouldn't be stiff enough to produce a composite section torsional response. Bolt slip and such. You'd get something closer to three sections twisting independently.
- To some degree, I suspect that the beam will rotate such that its strong axis remains more or less aligned with the load for modest angles of swing. Of course, that may gradually tear your beam bearing connections apart depending on the torsional give available between the hangers and the bearing supports. What are your torsional supports at the end of the rafters anyhow? Blocking?
- I'd feel better about some snug blocking either side of the support beam, either side of each hanger to deal with weak axis bending and torsion.
I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.
RE: Flitch Beam Torsion
If so, seems like you need to add some rafter to rafter blocking to prevent the rotation. I am sure the Arch. will love this.
RE: Flitch Beam Torsion
I would much prefer blocking, but the architect is really pushing hard against it.
To prevent rotation at the end there is blocking above a dropped WT clad porch beam. The flitch beam plate could potentially be fillet welded along the bottom. The top would be restrained by nailing of the 2x decking above to the flitch wood nailers. In a perfect world architecturally, we would not see any bolts between the steel plate and the wood nailers. The nailers would essentially be architectural furring that get epoxied on to the steel plate with threaded studs embedded into holes filled with glue.
I believe I can get a 1" thick x 6 1/2" deep steel plate to take all of the loading in strong axis, weak axis and torsion. The nailers would have to be assumed to brace the plate from buckling. I'm going to abandon the idea of trying to make the beam behave compositely and instead look at the steel section alone.
RE: Flitch Beam Torsion
Seems like you could get a clad T section to perform as well or better, with less weight and the same finished appearance.
RE: Flitch Beam Torsion
Perhaps, the way I was planning on dealing with torsion and weak-axis bending is flawed, but my thought was to form a moment couple between the steel plate and the sheathing above (which is plywood over decking) to resolve the torsion. We are not talking enormous loads as it is just a porch swing so I was thinking the sheathing could take the load. In this load path, I didn't think the WT section added a whole lot of benefit. The plate is more than adequate in strong-axis bending, the weak axis / torsion governs the design.
RE: Flitch Beam Torsion
BA
RE: Flitch Beam Torsion
RE: Flitch Beam Torsion
BA
RE: Flitch Beam Torsion
You could even put a big steel plate on top of the decking with some rod couplings welded to it that penetrate down to the underside of the deck. You could then thread you eyebolts into that.
RE: Flitch Beam Torsion
I am currently designing the 4'x9' sofa for 40 psf live load which equates to 1500#. This seemed reasonable (and quite conservative) for 4 "hefty" people. I am assuming that with the swinging motion that 2 cables could potentially be slack. I didn't increase the load by a "dynamic" effect above the 1500# as I felt assuming 2 cables were slack and taking the force at an angle was a reasonable design load. I've assumed that 30 degree angle from the vertical is the highest I would design for. That to me seems like a pretty steep angle for a swing. 45 seemed to be too conservative.
The 9' direction is inline with the 12' direction of the rafter span. Therefore, the swing is only ~2' from each support point. The strong-axis bending moment is therefore not very large (2 point loads close to support). As I mentioned previously, I broke up the lateral force into a moment couple between the sheathing / decking and the beam and designed the beam in weak-axis for the resulting moment couple.
I would prefer to use a 4x8 HSS but this won't fly architecturally. It's a $15 mil house and architectural intent is extremely important. The HSS would break up the look of the other rafters. I like XR250's idea of attaching to the decking but we have 5/4 African Mahogany decking which I'm nervous to use as it's not full 2x and I don't have any design values for it. Additionally, the architect will likely not like the look of hanging the swing from the decking. I think I am stuck architecturally with the eyebolt on the bottom of the beam unless I tell them "there is no way it can work structurally"
I feel like the 7/8" x 7 1/4" plate (or possibly 1" x 6 1/2") plate calcs out.
Does anyone disagree with my approach? Do you feel a WT section (which would be possible) would significantly increase the torsional capacity of the section? With the bracing from the sheathing I felt it wouldn't add a lot.
The architect is accepting of our current design, but likely would be resistant to other larger structural sections unless absolutely necessary. He seems to think the porch swing should be able to be attached to a typical rafter as he has seen it 100s of times.
RE: Flitch Beam Torsion
RE: Flitch Beam Torsion
I'm open to going back to them and saying it doesn't work, but I'm not sure I'm quite convinced.
RE: Flitch Beam Torsion
- In my heart of hearts, I agree with the architect. I don't feel that heaven and earth should have to be moved to hang a swing.
- Realistically, I think that the consequences of failure would be pretty minor.
I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.
RE: Flitch Beam Torsion
I was just trying this as a gut check to my other calc which I will use for submittal.
RE: Flitch Beam Torsion
BA
RE: Flitch Beam Torsion