General Mechanical

General Mechanical

Topics related to Mechanical Enterprise, Motion, Additive Print and more.

Bracket Fatigue Analysis in nCode

    • louiee1907
      Subscriber

      Hi,

      I want to do bracket analysis. I have g/Hz and g^2/Hz datas (X,Y an Z) until 2000 Hz

      Modal Analysis - Harmonic Analysis - nCode PSD Vibration

      1 - I calculated the natural frequency in modal analysis until 2000 Hz, however there is no enough frequency due to the less participication factor. Should I increase the Hz? Actually, I have datas until 2000 Hz, so I added.

      2 - After the modal analysis, I analyzed in harmonic. In harmonic analysis, I used acceleration input from my g/Hz. I converted g/Hz to mmxs^-2/Hz. I did it for X-Y-Z coordinate system.

       

      3 - In nCode, I put g^2/Hz datas from Multicolumn Input to Vibration Generator. Should I put as duty cycle or matrix load in PSD datas in FRF load case. I finished the fatigue analysis. 

      I have not done vibration analysis, is it true method? What should I do? Is there any example files like it? Please help me, I need really help.

    • danielshaw
      Ansys Employee

      Q1:  Yes, if are doing a mode-supe harmonic analysis, the modal analysis must include enough modes to capture sufficient modal mass (participation factors).  The amount of captured modal mass (75%, 90%, etc.) will affect the accuracy.

       

      Q3:  I do not know what you mean by "matrix load in PSD datas".  I would use a duty cycle, where each event represented acceleration in one direction.  Defining and using duty cycles is described in the Introduction to Ansys nCode DesignLife training course.  There are also several worked examples in the Ansys nCode DesignLife Help documentation.

      • louiee1907
        Subscriber

        I meant this, sir.

         

    • danielshaw
      Ansys Employee

      You would use that approach (the multi-PSD method) if multiple PSD loadings acted simultaneously (which is atypical).  A duty cycle is used to apply multiple PSD loadings sequentially (which is more common).

      • louiee1907
        Subscriber

        I have just X-Y-Z direction loads not XY-XZ-YZ(Cross spectrum). Can I still apply this method?  

    • danielshaw
      Ansys Employee

      Maybe, but that is not its intended use.  You might be able to specify zero (or very small) cross-terms to reproduce sequential loading, but why would you try? Multi-PSD loading is intended to be used when multiple PSD loadings act simultaneously.  If the X, Y, and Z PSD loadings do not affect each other (no cross terms), I recommend using a 3 event duty cycle;  one event for each directional loading.  The multi-PSD loading approach is very rarely used.

      • louiee1907
        Subscriber

        In addition, I added vibration repeats for X-Y-Z direction (1790-1790-1790). The life is 1.456E7 repeats. When I convert to hours, should I divide 1790 seconds, or 5370(1790x3) seconds?

    • louiee1907
      Subscriber

      I wanted to use this method because the vibrations affect the model simultaneously.

      Thank you so much, sir.

Viewing 4 reply threads
  • The topic ‘Bracket Fatigue Analysis in nCode’ is closed to new replies.