Torsion on Wood Members

[jheidt2543]

I was looking in the AITC Timber Construction Manual for information on torsion and was suprised that torsion isn't even in the index! Does anyone have a good reference paper on how to handle torsion in combination with bending?

Say you have a wood beam carring a floor load with joist of vastly unequal spans framing into it, one side 8' and one side 16') isn't that beam in combined bending and torsion?

Or how about a gabled end truss with a wind loading and roof loading. Doesn't that truss have to handle loading in both the vertical and horizontal planes? So, we brace the truss against the horizontal load. But, don't we have to make an assement of how much horizontal load the truss members can take so we kown the spacing of the braces?

 

[DaveVikingPE]

The beam you describe will not be in torsion provided the framing connection isn't a moment connection.

There's a good section in the Wood Engineering and Construction Handbook, 2nd ed., page 4.17, that addresses your question and includes an example calculation.

 

[jheidt2543]

DaveViking,

I don't have that ref. book, but will look for a copy, thanks for the note.

The case I was thinking of is where the joist frame into the beam with joist hangers and one span is 8' and the other 16'. So the loads are significally different from each side, I would think some torsion would be induced into the beam, which would be the net difference between the joist loads x (half the thickness of the beam + half the width of the joist hanger).

 

[DaveVikingPE]

The book is in-print and available from the usual suspects. It's written by Faherty and Williamson, published by McGraw-Hill. I got it for $15 on eBay - thanks to my guardian angel. It's a good supplement to the TCM and Breyer's "Design of Wood Structures."

1. Consider joist hangers as pinned connections, so there's no moment at the beam's supports - i.e., where they frame into the girder.

2. The end reactions of each beam are additive where they frame into the girder - that must be of enough section to handle the shear, and bending due to this total load, 'natch.

3. Now, considering the eccentric loading of an 8' span and a 16' span acting on the girder. Is the loading really that eccentric? If one of the joists is a cantilever, then maybe - but the joist would be designed as a propped cantilever and your connection hardware would have to handle any torques rather than them being applied to the girder itself.

4. That being said, the girder will be braced against lateral torsional buckling at its ends and perhaps where the joists frame-in to the girder.

5. Consider fastener pull-out prior to a twisting failure of the girder. The loads will probably have to be so high that the joists will fail in shear or bending before the girder fails in torsion.

My own experience with torsional loads on timber members is thus: I built a deck using 4x4 columns. Home Depot sold me the wood and, caveat emptor, said wood had a lot more water in it than you'd think. One of the 4x4 columns, after a couple of months, looks like it's been twisted nearly 360 degrees! - but there's no evidence of a failure in the member itself. I tried to stop this twisting (silly me) with heavy steel plates, angle connections, whatever... The screws popped out and the angles bent. Argh!