Rotor Dynamics of a High Speed Shaft with Gyroscopic Effects

- April 21, 2020 at 10:50 am
  
  aleksi_tamura
  Subscriber
  
  Hello All,
  
  I am trying to perform dynamic analysis of high speed shaft assembly with disk attached at the middle of the assembly. Assembly consists of two shafts modeled with line bodies and one 3d disk.
  
  In order to add effect of centrifugal load i add pre-stress analysis. In order to do that i define rotational velocity in static structural (learnt it from one of the discussion in here). Then again in the modal analysis part i defined rotational velocity again in order to see the effect of rotational speed and forward and backward whirl frequencies(coriolis effect is on). After completing the analysis i performed the same analysis with Abaqus and take pretty close results(large deflection is closed).
  
  My problem is with the large deflection option. I know that and as Ansys suggested i should open large deflection option. When i opened it i lost some of the modes and results seem irrelevant. I was expecting small changes but the results are not like that.
  
  My question is a little bit general. Is there something wrong with my whole approach. Thanks for the help.
- April 30, 2020 at 4:03 am
  
  Cheta Rathod
  Ansys Employee
  
  Hello:
  
  When large deflection is turned on in static analysis, 'spin softening' is included. And this can affect the modal frequencies downstream. So one cannot always expect small changes in the Modal frequencies when comparing between small deflection and larger deflection settings for the base static analysis.
  
  You may refer to this help manual example for such analysis:
  
  https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v201/en/ans_rot/Hlp_G_ROTEXCAMPPRES.html?q=centri
  
  Note, in the example 7.3 above: turn off the Coriolis in static to get the centrifugal force. Then in PERTURB restart we need to activate it to include the gyrosopic effect.
  
  Thanks,
  
  Cheta
- May 16, 2020 at 4:36 pm
  
  aleksi_tamura
  Subscriber
  
  Thanks for the reply. I will check the source you menti?ned. Thanks also for that.
  
  I performed the same analysis both with large deflection on and off. As i mentioned i called the results irrelevant because first i checked the static steps results and for both cases the results are pretty same. For example for my system max deformation is 0.091667 mm for large deformation off and 0.091795 mm for the large deformation on case. This is why i was expecting very close results for the frequency step.
  
  Thank you very much.