moment magnification effects aci

huangyhg

moment magnification effects aci
folks,
is my understanding correct that if a concrete column is broken into a finite number of pieces along its height and i do a p-δ analysis, i can set the δns factors to 1.0?
yes, but i would do a baseline case (as shown in aisc) to make sure your finite number of pieces is accurately capturing the p-little delta effects.
and you also need to satisfy all the provisions within section 10.10.1 such as material non-linearity and cracking (first check to see if any of your framing is cracked under the load combinations and if so, you must alter their stiffness in your pδ analysis to include this).   
i was reviewing an etabs model and i found that it did not do a good job of computing the δns factors when the column was crossing a story [meaning a two story column]. the cm factor was computed based on end moments (which was not really the moments at the braced ends).
i have a spreadsheet that computes δns factors for given moments and axial loads. i conservatively set βd to 1. i found that the δns values were more reasonable.
however, i was wondering since the column was broken into two pieces due to it being a two-story column, then would'nt doing a p-δ analysis capture the effects of member curvature? i realize as jae said in a previous post that in order to do the requirements of 10.1.1 right, i would have to sacrifice my first born to the analysis gods.
how do you folks account for member curvature in your analysis? am i correct in assuming that if a non-sway column is not slender, then i can ignore the effects of   
i think etabs does the   
no, it uses δns factor to account for   
the 2008 aci code specifically allows you to use either an elastic 2nd order analysis (where the elastic stiffness is based on cracked properties) or the moment magnification procedure based on the δns values.  you need not consider both.  
unfortunately, the language in previous editions of the code (i'm looking at 2002), requires you to use the δns values for any non-sway frame.  it doesn't really give you another option.  for sway frame, they allow a p-delta type of analysis in general.  but, not for the non-sway.... unless you want to comply with all the requirements of 10.10.1:
  2nd order
  material non-linearity
  cracking
yes...your first born is sacrificed.