BC in cylindrical coordinates

[elkhorn]

Hello:

I have issues with boundary conditions for a quarter-cylindrical shape model.

Can we apply a BC constrain in cylicrical coordinate? e.g., rotational constrain.

Thank you for your insight.

 

[PamCrash]

1) *TRANSFORM nodes into cylindrical cs
2) Apply *BOUNDARY CONDITION

 

[corus]

If you have a quarter solid or shell model of a cylinder and are applying symmetry boundary conditions then you can just apply the boundary conditions in the global XY co-ordinates, presuming the sides of the model are on the global planes.

ex-corus (semi-detached)

 

[elkhorn]

Thanks to both of you (Corus and PamCrash) for your respondings.

I wanted to give some additional detail. Responding to Corus: my 1/4 model is such that the right section is 45deg in positive-x and the left section is 45deg in negative-x.

Therefore, the sides of the model are not paralel to the global X-Y coordinates.

Again thank you for your insight.

 

[johnhors]

You could rotate the model by 45 degrees to get the faces to line up with global axis planes.

http://www.Roshaz.com

 

[bues0022]

If you don't mind me bringing this issue up again and hijacking your thread....


I am doing a very similar analysis. I have a quarter symmetry model of a washer. I am putting a cylindrical displacement BC to the ID in the first displacement step. However, I cannot use the X- and Y-symmetry BC's because the displacement and symmetry planes are in different coordinate systems. How can I either 1) apply a cylindrical displacement in x,y,z CS, or 2) get a radial symmetry BC for the quarter-symmetry model? I'm a GUI guy and don't know how to write script into the input file (infact an input file can't be generated because of this problem).

Thanks!

 

[MechIrl]

@bues0022
Have a look at using an MPC Slider constraint. It constrains nodes between to given reference nodes to remain in a straight line, allowing symmetry constraints to be placed independently of co-ordinate system or part deformation.

Its not supported by CAE but c'mon...its one line one code in the INP file!

 

[bues0022]

I'll check into that, thanks for the tip. I think I could probably wrestle with one line of code!