Instabilities running a partitioned FSI- Simulation with pimpleDyMFoam
 

Dear all,

I am currently working with a partitioned approach for FSI- Simulations. I adapted the pimpleDyMFoam solver to compute the solution of my fluid part. There are several benchmarks for validation. But every time I try to run the coupled simulation, my Foam solver blows up (calculates unphysical force values). I use the movingWallVelocity at the wet surface. In my example, I obtain displacement values in the first time step of ~5e-6. Then I divide this values by the time step length and use it as boundary condition for the pointMotionU. As you might suggest, the mesh movement is computed by the velocityLaplacian solver. But as you can see in the picture, I get a velocity peak at the boundary and the simulation finally calculates unphysical force values, although my Co Number is small enough. My Problem is: The displacement values seem to be physical, since the deformation is very small (I increase the velocity at the inlet with the time). So, maybe anyone has a suggestion, how I can run this case more stable. Any help is appreciated.

Best regards
David