Fluid Structure Interaction with deformable structural part

[AlessiaB]

Hi,
I have a problem with a FSI of a cylindrical pipe in wich a fluid flows. There is no problem if the structural part is steel and the fluid is water. But if I set a lower Young Modulus for the pipe (e.g. 1Gpa instead of 200GPa), the simulation doesn't go to convergence and the error message is:
ERROR #001100279 has occurred in subroutine ErrAction.
Message:
CFX encountered the error:
0. A folded mesh has occurred. This process will shut down as soon as possible.

The same message is generated even if I refine the mesh.
Can anyone tell me what is the problem with deformable materials in FSI?

Thank you.

 

[seymours2571]

CFX does not remesh the fluid domain for each FSI iteration. Instead it performs a mesh skewing operation where the individual cells are modified to conform the existing CFX mesh to the newly deformed boundary of the solid structure. There are some settings within CFX that allow you to control the way in which the mesh skewing occurs by adjusting the mesh stiffness, either as a constant or more appropriately as a function of distance from the structural body.

If adjustment of the CFX mesh stiffness does not work you can try meshing a large CFX domain to distribute the displacement of the structural body boundary over a greater number of CFX cells.

You could also try smaller timesteps, but my experience in the past with FSI simulations using Ansys and CFX the issue was always with the CFX mesh skewing.

I believe if push comes to shove you could remesh the fluid domain at each FSI interval, but I would suspect that would exponentially increase the time to solve. However, if this is the route you take could post back. I have never tried remeshing between FSI iterations and would be interested in getting info if someone has done this before.

Good luck,

Steve