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

Re: CQC and SRSS

[Subject Prev][Subject Next][Thread Prev][Thread Next]
>I think you'll find that SRSS is totally inappropriate for any three 
>dimensional analysis of any structure with even a moderate degree of  plan 
>irregularity. This results from losing signs when squaring displacement or 
>force quantities and then taking the square root; signs get lost and result 
>in very significant errors.
I'd say the reason that SRSS is inappropriate is because it doesn't 
account for the additive effects of closely spaced modes. All the 
combination methods I've ever used lose track of the signs, including 
CQC. Actually this is proper since the seismic response is 
oscillatory--it has no sign. The problems comes from the order in which 
nodal forces get calculated. ANSYS (I think STAAD does this, too) figures 
the combined element loads then calculates the nodal forces by adding the 
loads from the adjoining elements. This approach gives nodal forces that 
are usually way too large. What should be done is to figure the nodal 
forces from the adjoining elements for each mode and then combine the 
modal results. When I need a particular set of nodal forces I do this 
manually--it isn't difficult. Or else use combined element loads and add 
them as I see fit.

Christopher Wright P.E.    |"They couldn't hit an elephant from
chrisw(--nospam--at)skypoint.com        | this distance"   (last words of Gen.
___________________________| John Sedgwick, Spotsylvania 1864)
http://www.skypoint.com/~chrisw