Buckling Analysis in Ansys 2

[MacCox]

I am doing a buckling analysis where the column will buckle. It has a pinned-pinned mount and a side load perpendicular to the axis at the center of gravity. Where my problem comes in is that the center of the column can only deflect 20mm. How can I model/simulate this in Ansys WB 10?

 

[izax1]

I would say, that's a challenge!

Not to analyze it, but to get realistic, usable answers. I don't know Ansys in any detail, and I don't know if 20 mm is a lot or a little (Depends on the length, stiffness, etc), but simulating buckling in FEA is generally complicated. How can you be sure that you model your geometry correctly. Any tiny out of center deflection of the column will reduce the Cr bucling load significantly. The FEA analysis will assume a perfect geometry, which you probably will not have.

So be careful. Use large Factor of Safety

 

[corus]

The better way to model buckling is to use large displacement analysis rather than euler buckling and increase the load until the displacements increase rapidly. The buckling load will be the asymptote of that load-displacement curve. Of course if your displacements reach 20mm and it hasn't buckled then you should be ok.

Tata but not yet tara

 

[Dave776]

In addition to large displacement, consider using "notional loads" in AISC Steel Construction Manual 13th ed Appendix 7 page 16.1=196. They are applied simultaneously in both lateral directions, in addition to the lateral load, and are related to the gravity-type loads. N=0.002* gravity load, with other rules as well.

 

[MacCox]

Ok maybe I should have given a little more information. This is a round beam that is horizontal with various cross sections so it has multiple moments of inertia. There is a given axial force that the beam must resist. It's length is approximately 600 inches long and average cross section is about 12 inches in diameter (for ease of hand calc). The side load is a 2G side load. With a quick Euler hand calc and no side load I have an approximate FS of 1.17 The beam once it buckles comes into contact with a support that is 20mm (nearly and inch) away. I have FEA specialist at my disposal but I wanted to try to get some initial FEA results for myself. What is stumping me is how to make the beam stop deflecting at 20mm ant then start supporting the load...

 

[cloa]

Best to post in the ANSYS forum.

 

[flash3780]

In Ansys WB 10? Good luck. Can you upgrade to 13? Ansys will chug through instabilities if you turn on auto time stepping, large deflections and stabilization. Ansys WB 10 may require command objects to do so.

It sounds as though you're trying to find the point at which your column deflects 20mm to the side. Ansys ramps the load until it makes it through the analysis, so if you're working in "time steps" and the end time is 1, you should be able to plot the directional deformation versus the loading. Stabilization adds dampers to the model nodes to help the solution converge.

If you don't have stabilization on, Ansys still may give you what you're looking for with a small enough time step, but your solution most likely won't converge completely.

Ansys also has a RIKS solver available (similar to ABAQUS), in case you cannot get your model to converge, but it requires APDL scripting (even in WB 13). Most likely, stabilization should work fine.