Torsion 1

[SPR Baker]

Dear everyone!
Could you tell about torsional constant modifier value in Etabs to as code requirement?
when reduce it, effect on other local and global stability design?

 

[DaveAtkins]

I don't use ETABS, but it appears they are saying you must release the torsional degree of freedom in the SUPPORTING member in order to redistribute the torsion back to the SUPPORTED member.

DaveAtkins

 

[JoshPlumSE]

This is something we ran into as well while I was over at RISA. I'm speaking off the cuff right now, because I don't have any references in front of me and I no longer work for RISA.

As I recall ACI has some varying provisions for design of torsion in beams.
a) If the torsional moment is below some threshold amount then it can just be ignored.
b) If above the threshold then it must be considered in one of the following ways

i) The member must be designed to resist the demand (i.e. closed ties)​

ii) A portion of the torsional moment may be "redistributed"​

It's the 2nd option that this section of the ETABs help file is talking about. They are trying to point out that they do not do anything in the program to automatically re-distributed these torsional moments. My guess is that they also don't consider the torsion in the design of the beams. But, that's just a guess as I'm not an ETABs expert. They are emphasizing that you (as the engineer) are responsible for adding this moment into the adjacent beams or designing / detailing the beam to resist the torsion.

IMO this is merely good practice in documenting some program limitations that the user needs to be aware of.

 

[Archie264]

I have a quick question on terminology. They mention "statistically indeterminate structure"...is that an autocorrect typo? I know what statically indeterminate means but what is a statistically indeterminate structure? Or is it a more generalized term for the same thing?

 

[JoshPlumSE]

LOL I didn't even notice that typo. Yes, they definitely mean statically indeterminate structure.

 

[Archie264]

Ah good. I didn't know if there was some more nuanced meaning.

 

[SPR Baker]

JoshPlum
when design, Etabs have option for user to consider torsion or not.
but if we let Etabs ignore torsion in design phase, it will be our fail because moment in slab will be smaller than we release in analysis phase.
so we should release in analysis to ignore torsion transfer to slab resist.
The question for engineer how much we need release for torsion cause by slab deflection(statically indeterminate structure) ? It depend on your judgement to make sure the torsion Tu=min(Tcr,Tu)for statically indeterminate structure.