choosing the right eigenvalues for buckling analysis

[futura22]

hello everyone...

I am running an eigenvalue buckling analysis from input file from abaqus documentation.

The outcome of the linear analysis is series of eigenvalues
and they will give me the magnitude of buckling load by simply multiplying with the applied load.

my problem is how i am going to know the right eigenvalue ? is it just taking the lowest value of eigenvalue or just choosing any value that will give me the closest answer to the numerical/theoretical one? if the answer is taking the lowest eigenvalue, why i am keep getting very low value of eigenvalues?

i believe there is nothing wrong with the input file from the abaqus documentation.

Thanks in advance

 

[SWComposites]

1) generally the lowest eigenvalue is the critical buckling load

2) plot the mode shape (deformed geometry) for each eigenvalue and see if they make sense - good way to find boundary condition, connection, meshing, etc problems

3) are you saying you cannot reproduce the results in the Abaqus manual?

 

[futura22]

Thanx for the reply,

yes..the lowest eigenvalues that I got are no where near the results. So assuming the input files in the ABAQUS is correct, I am just choosing the value which give the closest to the correct answer, that means not neccessarily the lowest one.Is it okay to do that?

I have running the input file several time without getting error messages but still got the same results.

What is the most probable mistakes? Is it because of the boundary conditions or something else?

 

[SWComposites]

Have you made the mode shape plots?

What specific example problem are you running? Please describe the porblem in detail.

Are you comparing to a result given in the Abaqus manual or to a theoretical result from another source?

 

[futura22]

yeah i have plot one and it looks nothing wrong in terms of the way it buckles...

i am running eigenvalue buckling analysis for cylindrical shell

i am comparing both from the abaqus manual and the theoretical result...but what confusing me was the 'very small' value of eigenvalue obtained after i run the analysis thru the input file from abaqus manual

here's the link:

http://abaqus.ethz.ch:2080/v6.6/books/bmk/default.htm

Just go to 1.analysis test -> 1.2 buckling analysis -> 1.2.3 Buckling of a cylindrical shell under uniform axial pressure

 

[dkeatley]

Are you aware that the eigenvalues are actually normalized?

 

[johnhors]

dkeatley,

EigenVECTORS are normalised !

 

[futura22]

what do u guys mean by normalised?

 

[SWComposites]

All displacements are divided by the maximum displacement in the model.