why the bending moment decrease after the P delta analysis on ETABS 2

[ibrahemnasr]

after i make P-delta analysis on Etabs i get the bending moment decreased on column however it must be amplified

I did the following:

- I decrease the stiffness due to cracks according to ACI,i use 0.7 for I2 and I3 in columns , and 0.35 for I3 in beams, and 0.25 for f11,f22,f12 in slabs.

-then I turn on P-delta option with load combination = 1.2D+1.6L
and max iteration 5

the analysis converge in the 2nd iteration

after the analysis completed I found that results:

- the period of the structure increased
- the story drift also increased
-but the column moments and axial force decreased

the structure is unbraced so the moment must be amplified due to the P-delta effect

so my question is why the moment is decrease?
best regards

 

[logicway]

which load cases did you introduce? have you checked the moments against each load case? or only against 1.2D+1.6L Factored ?
It also depends whether you are iterating with P-Delta for seismic lateral forces or just P-Delta effects due to sidesway, wind lateral or lateral deflection effects. It depends upon the type of structure and location of columns beams also.

 

[TLycan]

If you have tensile axial forces in the columns,From the 1st order analysis, it will reduce the bending moments.
Since you use ETabs you can try run the model in SAP2000 to see if the will converge close answers.
I always feel SAP2000 is more easy to capture errors and structural instabilities

 

[ibrahemnasr]

Mr logicway and TLycan

I've checked the model again for combinations that include gravity loads only and other that include lateral loads , also i checked different column positions ( interior-edge-corner) and also for ground story and other for upper story and i noticed that:

the change in the straining actions is due to two reasons or in other words occure in two steps
1-change due to the reduction that happened to the stiffness ( and this effect is different according to the column position so for interior and edge columns the axial load increase and the moment and shear decrease and for corner columns all internal forces is decreased ).

2-change due to P-Delta effect which will cause the straining actions to increase but the increase in the straining actions due to P-Delta effect does not compensate the decrease happened due to the stiffness reduction which end us up to internal forces that are smaller than that from elastic analysis in which we ignore cracks.

so i have to questions :


to consider P-Delta effect i use the option in Etabs in which you have to define combination load and iterative numbers and tolerance NOT the other method in which you have to define nonlinear P-Delta load case
so i think the first method i use did not capture the effect of P-Delta completely so Should i have to use the second method "nonlinear"?
what is your opinion about that?

from the above should i have to design the building for the reduced internal forces from P-Delta analysis or design for the unreduced internal forces from elastic analysis?

 

[TLycan]

Mr Ibrahemnasr

The method you are using is alright, I think. I've revised an Etab File of a Formidable Consultant Office and they are using the same method as you.
my question is
what do you mean by the reduced stiffness

the 0.7, 0.5,...etc factor to account for the cracked section or something else

 

[ibrahemnasr]

ya i mean the reduction in stiffness due to cracks so i will take the opportunity to ask you about the factors of reduction i used

i use 0.7 for I2 and I3 in columns
and 0.35 for I3 in beams
and 0.7 or 0.35 for F22 in walls
and 0.25 for F11 , F22 and F12 in slabs

by the way my first question did not answered untile now
for any results should i have to design columns ,the outputs from elastic analysis (which are the larger in bending moments) or the outputs from analysis taking into account the cracks ( which are the larger in axial loads)?
or ia have to design for the max from the two cases?

 

[TLycan]

According to which code you are working on, that is:

Codes require/allow that you use the cracked section and therefore the lesser bending moment. still you have another problem that is the reduced stiffness has led certainly to an increase in the sway. check if you have exceeded the limits or not according to the applicable code