*BOUNDARY on coupling reference node yields trivial solution

Analysis issues
1

Hey everyone,

I’m stuck with the follwing problem: I have a simple beam consisting of C3D10-Elements. At both ends of the beam, I define surfaces. There are two nodes outside the beam. I couple these nodes to the surfaces with *Distributing. I want to fix one of the nodes and apply a (large) displacement to the other node. Here’s the structure of my .inp:

*NODE
... 
Node definitons for the mesh
...
1118, -3.000000, 10.000000, 5.000000 ** Coupling Node
1119, 103.000000, 10.000000, 5.000000 ** Coupling Node
**
*ELEMENT, TYPE=C3D10, ELSET=SOLID
...
all Nodes except 1118 and 1119
...
*SURFACE, NAME=SURF_X0, TYPE=ELEMENT
...
*SURFACE, NAME=SURF_X100, TYPE=ELEMENT
...
*COUPLING, CONSTRAINT NAME=CP_LEFT,  REF NODE=1118, SURFACE=SURF_X0
*DISTRIBUTING
*COUPLING, CONSTRAINT NAME=CP_RIGHT, REF NODE=1119, SURFACE=SURF_X100
*DISTRIBUTING
...
*AMPLITUDE, NAME=DISP_PATH
0.000, 0.000
0.200, 0.200
0.400, 0.400
0.600, 0.600
0.800, 0.800
1.000, 1.000
*STEP, NLGEOM
*STATIC
0.05, 1.0, 1e-5, 0.1
*BOUNDARY
1118, 1, 3, 0.
*BOUNDARY, AMPLITUDE=DISP_PATH
1119, 1, 1, -10.
*NODE FILE
U
*EL FILE
S, E, PEEQ
*END STEP

When I run the simulation, I receive a trival solution (no displacement applied to the model).

I tried to replace the displacement by a *CLOAD-Card:

*CLOAD
*1119, 1, -50000.

When I do that, the simulation works.

Can anyone assist here? I’m not sure if I have a worng syntax or if Calculix can’t apply displacements via a coupling?

Thx in advance!

2

*Distributing: the purpose is to redistribute forces and moments defined in a reference node.

1 Like
3

Got it. Thank you. The alternative would be to apply the displacement via *DISTRIBUTING COUPLING. But according to the docs, it is not recommended to use more than 50 Nodes for the *DISTRIBUTING COUPLING. In my original problem (not the beam-minimal–example above), I need to couple around 2500 Nodes and apply the displacment. What would be the best practice here?

4

Check this thread:Different coupling constraints and their limitations - #3 by Calc_em

1 Like
5

Thanks everyone for your help so far. Based on the iput above, I tried a few setups. First I tried to constrain the two external points with *DISTRIBUTING COUPLING. Since *DISTRIBUTING COUPLING does not fix the rotational DOFs, the model was not fully constrained. I fixed all nodes of one of the surfaces and applied the dispalcement via the coupled node. That setup worked with reasonable results, but that is not what I was aiming for. I tried to add another external node, coupled it with half of the surface and coupled the remaining half with the exisisting extenral node. This setup does not converge.
My goal is to mimic the behavior of RBE3-Elements (aka Distributing Couplings) in Abaqus. It seems like, rotations can be locked in Abaqus so with two points, my model would be completely fixed. Is there a workaround in Calculix for this issue?