Rotor model with tangential force and symmetry

1

Dear all,

I have a 2D rotor model with centrifugal load, symmetry conditions (no displacement in angular direction) at the two red lines, a penalty contact condition at the blue lines and fixed radial displacement at the green line. This model uses CPS6 elements and works fine.

Rotor_Model
Rotor_Model793Ă—804 82.2 KB

Now, I want to add a tangential force at the rotor surface (outer radius). Therefore, I think I need to alter my symmetry conditions. I tried to use the *cyclic symmetry card on the two red lines, but it gives me the following error:

"*ERROR reading *CYCLIC SYMMETRY MODEL
sector angle cannot be determined:
there exists no dependent node
with the same axial position as
an independent node (a tolerance
of 1.e-10 length units is applied).
The smallest axial distance of 1.1679476431456434E-003
exists between dependent node 2
and independent node 15176 "

I already checked the coordinates of the two nodes:
node 2: 7.409513850740e-02,3.069121127568e-02,0.000000000000e+00
node 15176: 2.250125640914e-02,5.332673006825e-02,0.000000000000e+00

If “axial distance” is refering to my z-axis, it doesn’t make sense. Maybe it has something to do with conversion from 2D to 3D elements in CalculiX?

I tried to use my 3D version of the model (C3D4 elements), but I get the same error.

  • Is the cyclic symmetry card the right approach for my problem?
  • If yes: How can I get rid of the error? Maybe someone has an idea?

Here are my files: Google Drive

Thanks in advance!

Best regards,
Alex

2

Under the *CYCLIC SYMMETRY MODEL keyword you should specify the coordinates of 2 points defining the axis of symmetry. So it can be something like:
0, 0, 0, 0, 0, 1
if it’s Z axis.

Cyclic symmetry can be used only when the geometry, boundary conditions, loading and response exhibit this kind of symmetry.

1 Like
3

@Calc_em Thanks for your answer! I misunderstood the needed coordinates of the 2 points, now it does work.

I want to model a tangential force at the outer rotor surface in counter-clockwise direction, so I need a periodic boundary condition. Is *CYCLIC SYMMETRY MODEL suitable for that?

Best regards,
Alex

4

Yes, it can be used to model torsion for instance, as long as the assumptions listed above are fulfilled.

1 Like
5

with most fe software, cyclic condition requires the edge meshes to be EXACTLY same. Don’t know if it is the case with CalculiX but worth checking out. I had to mesh one edge region and mesh copy it to meet this condition (in ANSYS).
hth
Osman

6

No need for that in Abaqus and CalculiX:

The *CYCLIC SYMMETRY MODEL card triggers the creation of cyclic symmetry multiple point constraints between the slave and master side. If the nodes do not match on a one-to-one basis a slave node is connected to a master face. To this end the master side is triangulated.