Learning Forum

[lm32]

Hello,

I am trying to run a 2D axisymmetric simulation. I have modelled a lagrangian 2D shell interacting with an S-ALE mesh, all on the plane Z=0 as required for an axisymmetric run. However, after starting the run, I receive the following warning for many nodes:

*** Warning 21432 (STR+1432)
     Node ID XXXXX has an invalid z-coordinate value of -4.55E-06.
     In an axisymmetric analysis, all nodes should lie in the xy-plane
     with a z-coordinate of zero.

The simulation is running, but it stop earlier than required, shortly after many shell elements suddenly begin to quickly erode. what could be the problem?

Thanks in advance.

[Ushnish Basu]

It seems your mesh is slightly offset from z=0.0. Please use your preprocessor to move all nodes to the xy-plane exactly.

[Ian Do]

Hello Im32,

 

May I ask what your name is & which organization are you with?

[Q1] ...  invalid z-coordinate value of -4.55E-06.

[A1] This may mean that in the process of mesh generation, there may be small numerical error introduced into the Z coor of some nodes and the formulation, being axismmetric, does not like it. This can lead to instability later. So I would redo or clean up the mesh to make sure all Z coor are at 0.0

 

[Q2] shortly after many shell elements suddenly begin to quickly erode. what could be the problem?

[A2] LAG elm erode when they reach failure limit. Somehow they are loaded ==> mat accummulates damage ==> eventually fail. This is due to your definitions for mat data & BC's & IC's, etc. If there are FSI (fluid structure interactions) then that can be checked, too.

Regards,

Ian Do