Flange Warping in collector beams

[NParikh]

I am doing torsional analysis on my collector beams that are framing into a column. Since its a collector beam, the flanges are not welded to the columns and hence warping is not restrained. Is there a way to restrain warping of these flanges?

 

[JoshPlumSE]

You do not generally need to restrain warping at the connections. It is very, very difficult to truly restrain warping in anything other than a side plate moment connection. The Euro guys have a design guide on torsion (SCI P385) that shows some examples on how to restrain warping. I suppose you could use welded side plate towards the end to turn it into more of a tube type section. That should effectively restrain warping. But, is it really necessary?

 

[NParikh]

@JoshPlum Thank you for the response.

I have a wind load acting at the bottom of the flange causing torsion. I did my torsional analysis considering both the beam ends to be torsionally fixed. (I assume that's how my beam is going to behave because we did provide a detail at those collector beam connections where the top flange is restrained by the deck above and the bottom flange by the angle plate shown. I believe this would make it torsionally restrained. Wouldn't it?)

However the REVIEWER at DSA responded back with comments stating that "Your analysis indicates that warping of flanges is restrained. Since these are collector beams, the flanges are not welded to the columns thus warping of the flanges is not restrained."

 

[Lomarandil]

Probably not. The amount of flange movement in the axis of the beam for a "torsionally free" end condition is pretty small, so you need to be really stiff to effectively restrain that. Depending on your deck detail, that can be possible. But for your bottom flange, the fact that you have slotted holes (so presumably not slip critical bolts) means I wouldn't call that torsionally restrained.

I don't know the whole story, but you'll most often be best to find another solution that doesn't require your beams to be torsionally restrained.

----
The name is a long story -- just call me Lo.

 

[KootK]

I agree with your reviewer and the other gentlemen here: ineffective warping torsion restraint.

It's always tough to know just what another person is thinking based solely on what they've written. That said, there's often no way to help meaningfully without making some assumptions in that regard. So, forgive me if I get this wrong but, when I read your comments, I get the impression that you don't fully grasp the difference between a torsionally pinned restraint and a torsionally fixed restraint. Try this:

1) A torsionally pinned restraint has what it needs to resist torsion. It isn't free to just spin around in space about a longitudinal axis. It sounds as though this is the kind of torsional restraint that you've provided with your clip angle. It's not entirely clear to me how you're laterally restraining the top flange at he connection but that's not important to this part of the discussion. Perhaps slab connection etc.

2) A torsionally fixed restraint has the features of #1 but also goes an important step further by adding cross section warping restraint. You know how we assume that "plane sections remain plane" for normal, strong axis beam bending? In order to have warping restraint, you basically need plane sections to remain plane about the weak axis at the restrained support as well. In the sketch below, that is the case at the left end of the beam but not at the right.

AISC has an excellent design guide on torsion which I'd recommend checking out if you haven't already. The cover graphic shows a beam cross section warping up a storm.