Return to index: [Subject] [Thread] [Date] [Author]

RE: effective length factor for sway moment frames? K for frames

[Subject Prev][Subject Next][Thread Prev][Thread Next]

I confess that this is a topic that has troubled me for a long time. I do not
think it proper to assume K=1 just because I turn on the P-delta feature on
software, for reasons I shall describe below.  And since the normal AISC
equations are intended to account for P-delta effects, it seems overly
conservative to include the additional stresses/deformations in the forces when
I do a code check.  So the P-delta is an interesting exercise that is basically
used to check drift limits.

My problem with the P-delta analysis in computer programs is that they do not
include the erection tolerances for out-of-plumb conditions and permitted camber
or sweep that is permitted in fabrication - and these constitute a large part of
the P-delta effects found in the AISC equations.  So the computer is only able
to calculate the P-delta deformations of a theoretically perfect structure, that
is erected perfectly plumb and with totally straight members.  In theory you
could add nodes at mid-points of member spans and shift coordinates slightly to
include these effects, and then in theory the P-delta analysis might come out
just right, but who the heck is going to do that?