6 Replies Latest reply on Aug 19, 2011 7:52 AM by James Pare

    VIBRATION STUDIES

    JAMES PARE
    JAMES PARE Sep 25, 2008 1:13 PM
      HELLO, TRYING TO SIMULATE VIBRATION STUDIES USING SOLIDWORKS/COSMO
      THE STUDIES WE RECEIVE SHOW CHARTS WITH ACCERLATION VS FREQUENCY
      ACCERLATION BEING "g" VS THE FREQUENCY
      BASICALLY WHAT WE ARE TRYING TO DO IS SIMULATE A A FIXTURE BEING SUBJECTED TO A SINUSOIDAL VIBRATION AT ALL FREQUENCIES FRO 10HZ TO 200 HZ , THIS BAND OF FREQUENCIES IS LOGARITHMICALLY SWEPT OVER A DURATION OF 3 MINUTES AT A ACCERLATION OF 1 (G)
      AFTER THAT IS PERFORMED I NEED TO DETERMINE IF THERE IS A RESONANT FREQUENCY ABOVE OR BELOW 200

      I'VE BEEN TRYTING TO DO A DYNAMIC FREQUENCY STUDY USING MODAL, HARMONIC & / OR RANDOM VIBRATION
      BUT I CANT GET EXACTLY WHAT I'M LOOKING FOR
      ANY HELP OR SUGGECTIONS WOULD BE VERY MUCH APPRECIATED
      • 1809 Views
      • Categories: Simulation
      • Tags:
        • VIBRATION STUDIES
          Vince Adams
          Vince Adams Sep 25, 2008 2:07 PM (in response to JAMES PARE)
          Hi James, you can do this using a Harmonic analysis and simply setting the range you are looking at. If you are concerned about frequencies near 200 Hz, I'd suggest running the simulation out to 300 Hz to make sure you pick up vibration in the low 200's. Vibration analysis on real-world parts isn't an exact science. Small variations in assembly position, bolt preload or excitation orientation can produce measurable variations in resonant response. You also want to make sure you calculate frequencies up to 400 Hz to make sure higher-order effects don't get missed. Failure to include higher frequency resonance in lower frequency responses is called Modal Truncation. Better to have more frequencies than less!

          Good luck!

          Vince
            • VIBRATION STUDIES
              Vince Adams
              Vince Adams Sep 25, 2008 2:10 PM (in response to Vince Adams)
              One more thing, the solution doesn't take into account the duration of your test sweep, 3 minutes. The solver assumes that each frequency iteration is allowed to reach steady-state before moving on to the next. Realistically, if your actual sweep was too quick to allow resonant responses to reach steady-state, your Simulation may over-predict response. At a minimum, this calculation will be conservative.

              -- Vince
            • VIBRATION STUDIES
              JAMES PARE
              JAMES PARE Sep 25, 2008 2:22 PM (in response to JAMES PARE)
              sounds good
              will try your suggestions
              thanks!!!
              • VIBRATION STUDIES
                JAMES PARE
                JAMES PARE Sep 25, 2008 2:27 PM (in response to JAMES PARE)
                should i use upper bound frequency or # of frequency option?