Questions on Torsion 2

[amustEIT]

how can I brace my beam for torsion at intermediate and end sections of my beam?

I read here that sometimes for simply supported beam the ends are not braced for torsion. How is it possible that there would be no restraint for torsion at the ends? I can't understand that.

will stiffeners help increase the capacity of my beam for torsion? I heard no but I would think yes like concrete beam stirrups.

what is the difference between torsion from eccentricity and torsion from LTB?

my understanding of torsion is weak. please recommend a journal for torsion.

 

[kslee1000]

271828:

Thanks. So, I am learning a modern day terminology for the age old phenomenon - lateral compression buckling, am I right or wrong.

 

[271828]

No clue. "Lateral compression buckling" sounds more like a column buckling limit state to me. If you're talking about a beam in bending, so one flange in compression and the other in tension, and the compression flange kicks out to the side, then yes.

 

[haynewp]

I thought you needed both lateral and rotational restraint to avoid LTB? Ref: Design of Metal Sturtures, Galambos p.209

 

[kslee1000]

271828:

That's exactly what I meant, thanks.

 

[csd72]

271828,

no I mean torsional restraint.

If the beam is torsionally restrained at a point then the top flange is fixed directly above the lower flange. So if the bottom flange is held in place by the tension effect then so is the compression flange directly above it.

 

[kslee1000]

Same effect, different terminology.

 

[271828]

csd72, I respectfully disagree. If a point on the cross-section has a torsional restraint, then web distortion could let the compression flange move laterally anyway. That's what all that BetaSec stuff is about in App. 6.

I thought you were coming from here: LTB is mostly twisting, so why does it do any good to restrain the cmopression flange against displacement? It's because the bottom flange is in tension, so doesn't want to move laterally. Therefore, the whole section won't twist if the compression flange can't move sideways.

haynewp, that's just what the AISC App. 6 says. Interesting that the Galambos book says otherwise. By that logic, if a simply supported floor beam has the top flange continuously laterally braced, it wouldn't be braced because it could still twist.

 

[haynewp]

I wonder if it has to do with AISC addressing doubly symmetric I shaped members and channels in that section?