3 Replies Latest reply on Oct 12, 2007 9:19 PM by genexxer genexxer

    Contact Set vs Global Bonding

    Lee Drozdenko

      I have a 3 part assembly and the global contact is set to bonded. I'm running a frequency analysis to find the natural frequencies.

      Can I assume since the parts are touching that they will be bonded or do I have to define a contact set?

      I ran both simulations with these results:
      Just global:
      Mode No. Frequency(Hertz)
      1 122.65
      2 138.95
      3 279.87
      4 387.1
      5 846.1

      with contact set:

      Mode No. Frequency(Hertz)
      1 -1.7596
      2 -0.10258
      3 -0.028497
      4 -0.0098229
      5 0.055411
      6 0.70618
      7 997.02
      8 1101.8
      9 2860.8

      Thanks for any tips.


        • Contact Set vs Global Bonding
          Vince Adams
          Lee, take a look at the deformed images. These should tell you what is working and what isn't It is unlikey that the displaced shapes at each of these modes are identical with such different frequencies. If this doesn't answer your questions, let me know.
            • Contact Set vs Global Bonding
              Lee Drozdenko

              I looked at the modes and also contacted our reseller tech support. They informed me that I should ignore the first 6 modes if they are ~0 due to the 6 DOF's.

              Even if I did that the simulated results are far off the measured lab results.
              The first simulation seemed to be on the right track so that's why I was confused on why the results of the 2nd round were so far off.


            • Contact Set vs Global Bonding
              genexxer genexxer
              Experience shows that bonded by default ties to my experimental results. I never went to the other thing you mention, except when using shell elements. Another question is if both studies used the same mesh. I find great errors using draft elements in natural frequency modal analysis.