Welcome to the forums. Please post in English or French.

You are not logged in.

#1 2018-03-04 16:56:18

Yvon Goth
Registered: 2007-11-26
Posts: 20

Prise en compte de la gyroscopie en 3D


Pour les modèles de rotor, j'ai des difficultés pour obtenir les mêmes résultats en 1d (poutre et points avec inertie) et en 3D avec prise en compte de la gyroscopie.

Par exemple, cas test SDLL141 (poutre simple en rotation), comparé au même cas  traité en 3D (en adaptant le test SDNV109):
Premières fréquences propres
à vitesse nulle:
1D: 123 Hz, 491 Hz 1104 Hz, 1964 Hz  (modélisation A cas test SDLL141)
3D: 126 Hz, 499 Hz, 1104 Hz, 1921 Hz
donc comparaison  OK

vitesse 10000 rd/s:
1D: 119 Hz, 125 Hz 479 Hz, 503 Hz, 1077 Hz, 1132 Hz  (modélisation A cas test SDLL141)
3D: 76.9 Hz, 350 Hz, 918 Hz, 1143 Hz, 1768 Hz, 1987 Hz

J'ai constaté les mêmes types d'écart (effet nettement plus important de la gyroscopie en 3D qu'en 1D) pour d'autre cas test (SDNV110 traité en 1D, SDLL124 traité en 3D).

Merci de votre aide



sdll141_3d.zip, Size: 7.93 KiB, Downloads: 280


#2 2022-01-15 20:37:36

From: Argentina
Registered: 2017-07-27
Posts: 5

Re: Prise en compte de la gyroscopie en 3D

Hi, i'm having this same problem. did anyone had news on it?? please! Si il vous plaít!
I'm trying to make some work and some tutorials on rotordynamics on salome meca. I first tried some test cases and they worked the same as indicated in PDF's of test cases.
Then Tried a very simple case of a rotor of one (very rigid) poutre element and two palier 0D elements (no amortissement only radieur(stiffness)) to look for cilindrical and conical modes. it worked on many methodologies (QZ, Sorensen) on CALC_MODES and CALC_MODES_ROTATION (who appels the command CALC_MODES). I compared it with some hand calculations for rigid rotor and they agree in frequencies at 0 25 an 50 rad/sec.
After that tried to make the same with 3D Volumique elements. Trying to use QZ and Sorensen too. the results where good for 0 rad/sec, but many frequencies apear for different speeds. The message call for heavy amortissement model. I can't find where the problem is, and is the same problem stated here in this post.
Here are de files uploaded wich resulted of this examples.
I can give support on manual equations to this values
omega cilindrical=sqrt(2*k/m)    frequecy cilindrical: (omega/2*pi)
omega conical=J0*ω/(2*Iy)+-1/2*sqrt((J0*ω/Iy)^2+4*2*(L/2)^2*k/Iy)
J0:Polar to working rotating speed=0.61261, Iy:polar to transverse rotation=2.8588  k:palier stiffness=1000 N/m
material: steel, 7800 kg/m^3. vol=0.0157     mass=122.52
I use wxmaxima to compute this values and the manual frequencies.
I hope it's possible so solve this! i'd be very greatful. yours sinceresly,
Ignacio Capparelli

The frequencies (Hz) are:

0 rad/sec
--------                      .mess file     .mess file
nombre:   Manual  1DPoutre    3DVolumique
1               0.643       0.643           0.6776        Cilindrical mode
2               0.643       0.643           0.6776        Cilindrical mode
3               1.052       1.0524         1.1149        Conical mode
4               1.052       1.0524         1.1149        Conical mode

25 rad/sec
--------                        .mess file    .mess file
nombre:   Manual    1DPoutre    3DVolumique
1                0.643        0.643          0.0573       Cilindrical    Cilindrical modes should be always the same frequency.
2                0.643        0.643          8.0150       Cilindrical   
3                0.7091      0.7091        0.1691       Conical backwards
4                1.5617      1.5617        7.3493       Conical forwards
in 3D there is an extra first mode at 0.000088 Hz it seems to be a free body (spurous) rotation mode


#3 2022-01-15 21:38:03

From: Argentina
Registered: 2017-07-27
Posts: 5

Re: Prise en compte de la gyroscopie en 3D

Here are the files ( comm, med, hdf, rmed etc.)  separated in a 1D_Poutre case and the 3D_Volumique case

consulta.zip, Size: 270.57 KiB, Downloads: 45