Usage of contact planes to model a special boundary condition

[kathrinsb...@googlemail.com]

Hi,

for my linear elastic problem (3d, with a fixed button or top region and some forces on the other side) I want to define another special condition:

The condition is given by a direction as a normal vector of a 2d plane. Now I want to prevent node movement in that direction.

I saw in the examples that I cannot use the essential boundary conditions, because I am blocking an arbitrary direction and not a dimension (x, y, z).
The idea is now to use a contact plane instead. Is that a good idea for defining this condition?

Thank you for your help. 

[kathrinsb...@googlemail.com]

Is it probably a better idea to use the LCBC condition with the kind 'normal_direction'. Could you provide an example how this could be defined? As far as I understand the explanation about lcbc kind normal direction it means the velocity is bound to the normal direction. In my case I would like to define the opposite, which is not allowing the movement in the direction of the normal vector. Thank you.

Regards,

Kathrin

  

[Robert Cimrman]

Hi Kathrin,

Maybe you can adapt [1] for your purposes - it allows a general LCBC in each node.

r.

[1] http://sfepy.org/doc-devel/examples/linear_elasticity/nodal_lcbcs.html

[kathrinsb...@googlemail.com]

Hi Robert,

this is working. Thanks a lot.

Regards,

Kathrin