Plywood sheathing on steel beams

[awa5114]

I have come across a frequent situation at work where we have plywood roof sheathing on top of steel beams. I would like to know if we can count on this sheathing to brace our beam top flange for bending. We are currently using Lb = 0 and Im not sure if this is correct.

 

[mike20793]

I never count on wood to brace hot rolled steel, there just isn't adequate stiffness. For that case, I know other engineers that do consider the top flange braced by the sheathing, but I would not. How far is the sheathing spanning? Is it going in it's strong axis or weak axis? What else supports the sheathing, is it all steel beams? Are you allowing OSB or are you specifying plywood?

 

[jayrod12]

It also depends on the connection details and the required bracing load. Sometime I have used it, other times not.

 

[KootK]

I'm one of those other engineers who will use wood sheathing to brace steel beams in many cases. Some thoughts:

1) There are AISC provisions for checking bracing strength and stiffness as, no doubt, you are aware. Of course evaluating them isn't a ton of fun in wood.

2) It is very common to consider a steel roof beam to be braced by a steel roof deck running perpendicular. Is a steel deck really all that much stronger -- or stiffer - than a wood deck?

3) In most cases, you have wood framing spanning perpendicular to the steel beam. In those situations, I prefer to use that framing to brace the steel rather than the sheathing.

4) While the number of brace points provided doesn't alter the strength requirement at each brace point, it does alter the stiffness requirement. For the case of essentially continuous lateral restraint, the stiffness demand is rather small.

You may enjoy reviewing this recent, related thread: Link

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.

 

[Archie264]

Wood joists can certainly be configured to brace steel beams. I never thought about using wood sheathing for that purpose. I'd rather not, actually. But that's a long way from saying it can't be done, I suppose. I look forward to reading more about it.

 

[KootK]

@Archie: when you use joists to brace steel, are you considering that torsional bracing of the section or compression flange relative bracing?



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.

 

[Archie264]

Just compression flange bracing. Nothing fancier than that thus far.

 

[Archie264]

PS: when I said I'd rather not use plywood for that capacity I was just speaking off the cuff. If the capacity's there it's there. Nothing further was meant beyond that.

 

[KootK]

Just compression flange bracing.

With this being the case, I would argue that you've already been using the sheathing as bracing for your steel beams as the stiffness of the sheathing is the only meaningful thing keeping your joists and beam top flanges from shifting laterally. If anything, I might prefer a direct connection between sheathing and steel beam. Obviously, have the joist in play does a couple of nice, less tangible things for us:

1) Stiffens the sheathing against compression/shear buckling.
2) Provides an avenue for torsional bracing as a fall back position.
3) Facilitates some appealing connection alternatives.

when I said I'd rather not use plywood for that capacity I was just speaking off the cuff. If the capacity's there it's there. Nothing further was meant beyond that.

Ditto for my statements. I didn't interpret your comments as a challenge but, rather, an opportunity for continued debate.

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.

 

[Archie264]

With this being the case, I would argue that you've already been using the sheathing as bracing for your steel beams as the stiffness of the sheathing is the only meaningful thing keeping your joists and beam top flanges from shifting laterally.

Doh! You're right! I never thought of that! But...but...well, speaking generally, I think the force required to brace the compression flange is quite small, relatively speaking, and the joists likely could serve the purpose unsheathed, i.e., nekkid. But, your point is made: the sheathing is bracing flexural members; I can't deny that. Thanks for helping me see it.

 

[NWEng1]

If the joists are perpendicular to the beam, they brace the beam. If joists are parallel, you can add full depth blocking on either side of the beam for bracing. Or you can calc to see if plywood is good enough on its own.

 

[thaidavid40]

The devil is in the details, I suspect. How does one effectively connect wood sheathing to the flange of a steel beam so that the connection carries the loads, and doesn't crush the wood (or OSB)?
Dave

Thaidavid

 

[KootK]

Here's my detailing contribution. You'd want to provide special detailing and inspection. I'd also be tempted to block the floor diaphragm in the vacinity of the beam.

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.

 

[KootK]

4) While the number of brace points provided doesn't alter the strength requirement at each brace point, it does alter the stiffness requirement. For the case of essentially continuous lateral restraint, the stiffness demand is rather small.

I did a little fact checking after work and it turns out that the above statement is not just wrong but bass ackward. The more brace points you add, the more stiffness is required at each brace. And, in retrospect, that makes sense. More braces means a higher mode shape and more energy required to impose that mode shape at each brace.

This leads me to the interesting conclusion that, in this situation, one should brace steel beams at a few discrete points rather than semi-continuously. Also interesting, the same conclusion would apply to steel beams braced by perpendicular running steel deck. Gonna have to ponder that a bit.



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.

 

[jwilki]

With the restraint detail of the top flange. My detail would be similar to KootK's, difference being;
I would have a timber trimmer bolted to the top flange at regular centers,sheathing is attached directly to this trimmer and any joist blocking is carried into the steel beam web and secured to the top/bottom flange.

 

[XR250]

I did a little fact checking after work and it turns out that the above statement is not just wrong but bass ackward. The more brace points you add, the more stiffness is required at each brace. And, in retrospect, that makes sense. More braces means a higher mode shape and more energy required to impose that mode shape at each brace.

That certainly goes against my intuition. I'm not buying it in the real world!

 

[KootK]

Not sure I buy it either. Much comes down to whether you consider the bracing to be nodal or relative. I find that classification tricky in diaphragm situations. I've been assuming relative bracing. If you go nodal bracing, then continuous bracing is an option and then brace stiffness would drop with additional bracing.

Another way to look at it might be to take swaths of sheathing as representing single, discreet braces. For example, the middle ten feet on a thirty foot beam as a mid span brace. That would have per brace stiffness and strength dropping with the addition of more braces.

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.