P-Delta Question

[ToadJones]

Basic question...

If using P-Delta analysis for a secondary analysis is the general rule of thumb that the P-Delta effects only need to be accounted for in the Lateral Force Resisting System (LFRS)?

In other words, secondary members like girts, purlins, or infill beams that are "along for the ride" don't need to be considered in a secondary analysis do they?

My issue is not exactly this straight forward, but....

 

[ToadJones]

second-order, not "secondary"...sorry.

 

[msquared48]

Toad:

I don't think that is true if the beams or purlins that take gravity and lateral loads are part of a frame that sees additional lateral deflection due to P-Delta effects.

Othwewise, I would tend to agree.


Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[ToadJones]

Well, anyone out there designing purlins or girts using a second order analysis?
Seems ridiculous.

 

[aggman]

Toad,
I have been designing grits and purlins for a long time and never considered them in the second order analysis. This is assuming that the grits and purlins are not carrying compressive forces that would have any effect on the stability of the system (i.e. only used as beams spanning from main member to main member)

 

[aggman]

Just for the record I have been designing "girts" not "grits". Sometimes I really hate the spell check on apple products!

 

[BAretired]

"grits" is a perfectly good word.


BA

 

[cliff234]

It sounds as if you are performing a lateral load analysis on a structure and are performing a second-order analysis to determine whether the structure has sufficient strength and stiffness to withstand the applied loads (lateral loads + gravity loads). If that is the case, then you need only consider the second-order effects on those members that are part of the lateral load resisting system. Floor beams and other members that are "along for the ride" and do not contribute to the global structural stability do not have to be analyzed in the second-order stability analysis (and I am not sure how they would be analyzed anyway, since they are not part of the lateral load resisting system). Your comment that your issue is "...not exactly this straightforward..." tells me that there is something else going on. What is the missing piece of the puzzle? If a member is bracing another heavily loaded member and that brace is subjected to significant second-order effects then you could have a problem. You can't have wimpy braces. I think there was a bridge constructed in Quebec in 1907 that collapsed during construction because braces bracing the main members had insufficent strength and stiffness.

 

[WillisV]

I take p-delta effects into account whenever there is a "P" in combination with a "delta"

 

[ToadJones]

Basically what I have is a series of T-shaped frames that form a canopy with roof purlins spanning over top. The frames are cantilevered and designed for lateral stability in both directions.

For various reasons the structure is not perfectly symmetrical. As a result, there is some differential lateral deflections in the model that results in bending of the purlins, primarily weak axis bending. A P-Delta analysis exacerbates the problem. I plan to stiffen the frames so as to mitigate this "secondary" effect but this all got me wondering if it was legitimate to ignore these "secondary" effects in the purlins. Hell, a few years back, I would have designed the frames and purlins independently and not take the interaction into account anyway! I realize that as a structural system this is a real phenomena, but I am not sure if it is necessary to go this far.

 

[haynewp]

Chapter C of the AISC spec states when P-Δ and P-δ analyses are required.