Hyperelastic 2-way FSI fails due to large deformations

Manatee

Hi guys,

I am looking at a hyperelastic artery with pulsatile flow. Ideally, I would like to study the effect of the wall deformation on the fluid flow so I am trying to set up a 2-way FSI. However, the model fails due to large deformations/distorted solid model (I have large deflections turned on). Until now I have checked the following:

• The structural model solves and converges just fine in Mechanical (by itself and when applying a simplified pressure curve as present in the fluid as boundary condition) with realistic results
• The fluid model solves and converges just fine in CFX with realistic results
• A 1-way FSI with CFX: force --> Mechanical also solves and converges

So now I am stumped as to what else I can try. Apparently, when including the other data transfer (Mechanical: displacement --> CFX), the deformations in the solid model just becomes too large? I would have understood that if the error occurred for the fluid mesh but I cannot think of what to do the other way around. Does anyone have an idea?

Cheers!

Kai

Here are 2 things that you can try: 1. use smaller time step size; 2. drop mid node for solid elements