Buckling analysis

[rgduarte]

Hi.

Is seismostruct capable of making a buckling analysis ? i.e. using a Euler's column model, is there any option that can provide Euler's critical load and deformed shape of the element ?


I've search for "buckling"; "buckling analysis" and "Euler column" on the search topic but i didn't get the info i need.

Thank you.

[huffte]

I haven't tried it, rgduarte. But, since geometric nonlinearities are included by default in the settings, I would imagine than a static analysis of, say a frame under vertical loads and some small lateral load could be analyzed to estimate a buckling load. You would, I believe, need to gradually vary the vertical load manually and plot the load versus displacement to try and identify an asymptotic load value.

[rgduarte]

Thank you for your answer.
Anyway, i can't even get the eigenvalues on a simple Euler Column:

http://postimg.org/image/kmiy1r7gv/

I'm using standard wizard material and infrmfb element type, I've just changed the section to a 15*2 mm2 rectangular section.
Restrains:
n111 x+y+z
n112 x+y

I've tried with other elements type, bur with no sucess.
Is there any modeling option that is making impossible to run such a simple model ?

[huffte]

Buckling modes and natural frequencies of a structure may both be reduced to an eigenvalues problem. They are not the same eigenvalues problem, however. I do not believe that you will be able to solve the buckling eigenvalue problem with SeismoSoft.

[rgduarte]

Yes, I know. Anyway, i can't get eigenvalues of a simple Euler column on Seismostruct ? Because, that's what is happening with my simple model, i can't get the eigenvalues nor any result from a force solicitation.

Thank you

[seismosoft]

Have you tried subdividing the the member to 5-6 smaller elements? This might work. In that case, it is better to use the displacement based, rather than the force-based, elements.

Seismosoft Support

[ruipinho]

Hi rgduarte,

In order to manage to get the eigenvalue analysis to work, and in alternative to the member subdivision suggested by Seismosoft's support team, you may also want to perhaps try changing the solver and/or the eigensolver, or also introduce a (very) small distributed mass along the element.

For what concerns the estimation of the buckling load, you may obtain this by subdividing the member in more elements, introducing a small imperfection at mid-height, and then apply an incremental compression force.

Best,

Rui