A static analysis and a buckling analysis were performed, revealing anomalous behavior:
_By performing the static analysis alone, the results also include the Contact Output data.
_ By performing the static analysis and the buckling analysis, the Contact Output results are no longer produced, and the buckling analysis deformation does not appear correct.
To verify, a frequency analysis was performed, and the deformation does not appear correct.
I don’t understand what the error is.
Thank you in advance for your help.
Linear buckling and natural frequency extraction are linear perturbation procedures. They can’t take nonlinear contact into account. Instead, the contact state is frozen (fixed) from the previous static step. It may need to have Nlgeom enabled to be used for preload, but you also need to enable the Perturbation setting for the Buckle and Frequency steps.
Of course, you could also use tie constraints - they work in linear perturbation steps too:
1) Replacing the contacts from Hard to Tie does not resolve the issues.
2) The compound model was generated to avoid dividing the bolts at the contact points between the outer tube and the inner plate. This allows us to understand the contact stresses between the bolt and the inner tube and plate. The compound model allows us to model the bolt as a single element and generate the relevant sections automatically, reducing modeling time. Is there another way to do this?
All steps run to completion. But you won’t get contact stresses this way. Perhaps you should run static step with contacts and linear perturbation steps with tie constraints as separate analyses. Or analyze nonlinear buckling (with a static step having Nlgeom enabled) if you want to take changing contact into account in buckling calculation.
So you enabled the Split compound mesh option on purpose to be able to use contacts ? Ok, I see. At first, I thought that you only wanted to merge the geometries of the parts and generate a continuous mesh.
Yes, the Compound model was used to also determine the stress on the bolts; the Hard model and the Tied Contact model do not work, only the Tie Constrains model works, which does not allow us to determine the stress on the bolts.
Linear buckling can’t provide real displacements and stresses (including contact stresses) anyway - such outputs are normalized during the eigenvalue problem solution. So if you are interested in buckling and contact stresses or other outputs of this kind, you should run a nonlinear buckling analysis. So keep the static step, make sure it has Nlgeom enabled and apply load expected to cause buckling. This is GNA: Nonlinear buckling explained simply - Enterfea
