9 Replies Latest reply on Apr 25, 2014 9:18 AM by Bill McEachern

    Buckling Analysis

    Phil Perlich

      Two Questions:

      1) Does buckling analysis only use euler's buckling formula?

      2) Do you always need to use soft springs in the buckling module?

       

      Thanks,

      Phil

        • Re: Buckling Analysis
          Prasad Bhonsule

          Hi Phil, my answers to your questions below

           

          1) Does buckling analysis only use euler's buckling formula? as I understand it, if you are running a buckling study in Simulation Professional, it assumes Linear (Euler) buckling. If you have Simulation Premium, and run a Non-Linear study, then you can take into account Non-linear buckling.

           

          2) Do you always need to use soft springs in the buckling module? No, you don't need to always use soft springs. In fact (I may be wrong), I believe SolidWorks reccommends using fixtures wherever possible to restrain models.

           

          I hope that helps. Please let me know if you require more information.

           

          Kind regards,

           

          Prasad Bhonsule.

            • Re: Buckling Analysis
              Phil Perlich

              Thank you for the reply.

               

              1) That seems very limited. I currently have Sim Pro. I'm dissappointed.

               

              2) OK, I put an immovable restraint at the bottom of my column. I then used reference geometry to restrain the top of the column in the transverse directions. Put my load on the top of the column downward in the axial direction. The solver falied. When I added soft springs, it solved.

                • Re: Buckling Analysis
                  Shaun Densberger

                  1) That seems very limited. I currently have Sim Pro. I'm dissappointed.

                   

                  In FEA, there are two ways to perform a buckling analysis: linear and non-linear.

                  • Linear Buckling: This solves for the perturbation of a linear static pre-load problem via an eigenvalue problem. The results from a linear buckling analysis are un-conservative by an unknown amount. Better results can be obtained by re-running the entire process with a new pre-load value this is closer to the previously calculated buckling load. This type of analysis is best when used for studying what modes of buckling exist in your design and how the modes of buckling and the buckling load factors are effected by design changes.
                  • Non-Linear Buckling: This isn't "special case" like a linear bucking analysis is. Instead, large displacement and strain formulation are used in conjunction with a non-linear material model (if needed). This method is more accurate compared to a linear buckling analysis, but will (most likely) still be un-conservative due to assumptions on the user's end (homogenous material, free of defects, etc.)

                   

                  2) OK, I put an immovable restraint at the bottom of my column. I then used reference geometry to restrain the top of the column in the transverse directions. Put my load on the top of the column downward in the axial direction. The solver falied. When I added soft springs, it solved.

                   

                  Can you post a screen shot of your model (or the model itself)?

                  • Re: Buckling Analysis
                    Jared Conway

                    Immovable on a solid element?

                • Re: Buckling Analysis
                  Bill McEachern

                  there are 3 types of bucking:

                  1) Linear - eigenvalue extraction - is based onthe undeformed shape (no load applied, though it may have one inthe analysis to allow a clac of hte load factor) of the structure.

                  2) Non-linear (so called) - you run a static cases of hte load onthe structure and use the resulting deformed stiffness matrix to extract the eigen value (vector). So the load factor is based onthe deformed shape.

                  3) Post buckling - youloaqd the structure till it collapses and then you can see at which load it lost its stiffness - no eigen value procedure. This procedure can be quite accurate in my experince - well as accurate as you material properties and the fidelity of your boundary conditions allow. I have done calcs where this is in the order of 1% of test data.