Convergence issue for a model with driver nodes

[ruzfactor]

I am working on a model, where I have few independent nodes (driver) kinematically coupled to few nodes on a part (driven). The first step is gravity loading. I have constrained the the driver nodes (3 DOF). In the second step, I will be using velocity time history (Implicit Dynamics) for the driver nodes. However, in the static step the solution is not converging. It convergences if I constrain other nodes in the part (not ideal as I don't they are not close to driver nodes). It may be under constrained or the driver nodes are not working in gravity loading. I have added tiny bit of point mass on the driver nodes, to check if that's causing any issue. Any suggestion on how to tackle this issue would be very helpful. Thanks.

 

[FEA way]

I have constrained the the driver nodes (3 DOF).

I assume that you constrained only the translational DOFs (1-3). Why did you leave the rotational ones free ?

 

[ruzfactor]

The model considered is a part of a global structure. That's why assumed the rotations may need to be included while translations are constrained.

 

[FEA way]

Ok, but you have to eliminate rigid body motions to achieve convergence in a static analysis. There are various ways of dealing with that. Here, you could e.g. try adding soft springs to stabilize those unconstrained directions.