Amplification of accidental torsion when using dynamic analysis as per ASCE-7-05 1

[ayelamayem77]

Dear all,

When the dynamic analysis method of ASCE-7-12.9 is used, all the requirements of horizontal shear distribution, given in 12.8.4, including torsion calculations that may be accounted for by displacing the calculated centers of mass of each level (see 12.8.4.1 and 12.8.4.2) also apply.
However, 12.9.5 states that amplification of accidental torsion, need not be amplified by Ax where accidental torsional effects are included in the dynamic analysis model.


My specific question,

when doing response spectrum analysis using etabs program,the starting value of eccentricity ratio has to be 0.05 (in response spectrum case data )
so based on the ASCE-7-12.9.5 and if i found tortion irregularity type 1a or 1b so no modification has to be made to the previous eccentricity ratio.
Please correct me if i am wrong.

 

[ayelamayem77]

please help

 

[yasir1989]

the horizontal shear distribution shall be in accordance with requirement of section 12.8.4 expect that amplification of torsion per section 12.8.4.3 is not required where accendital torsional effects are included in dynamic analysis model

 

[MrHershey]

Who starred the above response? It's word for word what the code says. Doesn't clarify anything at all.

Was just reading up on this myself to make sure I had it right. The way I read 12.9.5 is that the 5% is sufficient, no amplification of the torsion is necessary. Unlike equivalent lateral force, modal analysis should pick up inherent torsion as any natural torquing will be picked up in the various modes (make sure you have more than 90% participation, by the way), whereas ELF will not unless you iterate through almost like a horizontal p-delta analysis. You still want to include the 5% as 'accidental torsion', but amplification of this moment to account for 'inherent torsion' of the structure would be redundant given modal analysis already implicitly includes these effects.

For instance, the building I'm currently looking at would definitely qualify at least for the torsional irregularity. Probably for the extreme torsional irregularity as well. And I could have told you this without even looked at the ratios because my primary three mode shapes aren't weak axis->strong axis->torsion like a typical rectangular building, but are weak axis->torsion->strong axis because of an irregular plan and strong axis elements being primarily on one side of the building. The inherent torsion in the building was already being picked up in the modes and that the building exhibits a torsional irregularity presented itself in the periods of the mode shape (torsion mode had higher period than strong axis mode) without me even needing to explicitly include the effects by amplifying the torsion.