Buckling 3D analysis of tubular structure

JRPJRP
Hello,

I have a model of tubular structure which I use for static 3D analysis. FEM model is built with line2 beam elements. I would like to perform 3D buckling analysis and I found out it can't be done with line2 beam elements.
My question is what are my possibilites in model redefinition. Which elements can I use so I can obtain as simple model as I can? If I want to use shell elements as tubes I need to import surface model of tubular structure into Mecway and then mesh it with shell elements. If I understood manula correctly this should work. Is there any more simple procedure? 

Comments

  • VictorVictor
    An easier way would be to use the CCX solver with Mecway version 6 and line3 elements (Mesh tools -> Change element shape) which can be circular tubes. Here's a simple example.

    Be aware that the beam elements won't capture local buckling like shells will.
  • JRPJRP
    Thanks Victor, I will try a few things with new version and probably upgrade afterwards. Talk to you soon.
  • JRPJRP
    One question. What does Mode 1 represent in Solution results for 3D buckling analysis? Value is zero, but I get some deflection also, but deformations are a lot smaller (20 times smaller than in Mode 2).
  • VictorVictor
    That's a good question. I don't think it's really a buckling mode. It appears to be the solution of the static solve that CCX does first. I'll need to look into whether this should be suppressed from the solution. Perhaps you could say the very first buckling mode is the shape it deform to when the load is too small for actual buckling. By that definition, it would be correct. Thanks for bringing it up.

    Keep in mind that the scale of displacements can be meaningless and it doesn't tell you much/anything if one mode has 20 times the displacement of another.

  • AndreaAndrea

    In ""in a linear buckling analysis the initial stiffness matrix is augmented by the initial stress matrix corresponding to the load specified in the *BUCKLE step multiplied with a factor"" CCX Manual

  • AndreaAndrea
    So if you calculate the stress from buckling analysis (add S to *Node File) for the first buckling mode (Bukling factor =0) you will find the same stress derived from static analysis.
  • JRPJRP
    Thanks for answers. It makes sense this way.
Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register