Can a R/C moment frame ever go pass 1% interstory drift without collapsing (NLTHA)

[darrennthnl]

I've iterated over dozens of models, all up to standard (SMF). .All does very well in pushover (ductile)

But then Everytime I do a NLTHA, it either gets low drift (less than 1%) or it collapses, there is no in between, never a LS/CP case. I've analyzed it against various earthquake records, played with the scale factor, the results stays the same, it is always IO or CP.

Please tell me what I'm doing wrong? is it an inherent flaw of R/C frames? I do think 40 mm displacement for a concrete is very huge, so makes sense why it can never go to life safety.

 

[driftLimiter]

Sounds like it comes down to the hinge hysteretic settings and p-delta. The moment-curvature settings for the hinges need to be well thought out and be determined based on good seismic detailing practices.

NLTHA is a finicky beast and if your not extremely confident in your settings and good at troubleshooting it then its probably not advisable to use.

Seismically detailed columns (or moment frames) should be able to achieve more than 1% drift and sustain vertical loads.

Pushover analysis would be a good intermediate step for you to evaluate what is happening in your model.

 

[lexpatrie]

Non-linear time history analysis?

Starting to feel like an FAQ for all the user-created acronyms is needed. My Captain Crunch decoder ring broke a while back as it was non-UV resistant plastic.

 

[darrennthnl]

Sounds like it comes down to the hinge hysteretic settings and p-delta. The moment-curvature settings for the hinges need to be well thought out and be determined based on good seismic detailing practices.

NLTHA is a finicky beast and if your not extremely confident in your settings and good at troubleshooting it then its probably not advisable to use.

Seismically detailed columns (or moment frames) should be able to achieve more than 1% drift and sustain vertical loads.

Pushover analysis would be a good intermediate step for you to evaluate what is happening in your model.

I follow ACSE 41-17, I mean, how hard is it really? feels like just assigning fiber hinges (with the correct section definition) on columns and a distributed hinge at the beams does the job?

 

[driftLimiter]

Its hard man its the most sophisticated analysis technique we have. And you are the one reporting issues with it so you tell me how hard it is.