Learning Forum

[Jox]

Hello,

I am trying to model a balloon that contains water and expells it due to an applied force on the outer surface, similar to the attached picture.

I have successfully done this with a one way fluid structure interaction, where only the deforming stucture imposes a force on the fluid. 

Now I am trying to complete a two way fluid structure interaction but have run into an issue. The first couple of time steps solve perfectly fine until an output of the following in produced:

I have attempted some fixes mentioned in this form such as

• Reducing number of solving iterations in fluent to avoid pressure spikes
• Using a compressible fluid
• Using solution stabilisation (in fluent and global)

But the same failure occurs with the transient structural component outputting a huge deformation that causes the fluent to fail. I have attached a screen shot of the transient mechanical solution right before failing and the failure solution where you can see with the legend that massive deformations occur.

Mesh: the mesh is as fine as I can make it using the student software so I doubt that is the issue.

Setup:

       Transient Structural:

• Has been modeled as a surface with 4 mm thickness to save on computational costs.
• Has a pressure (MPa) applied (starting from zero and increases using a cube function)

     Fluent:

• Uses a pressure outlet
• Dynamic mesh: Smoothing and Remeshing and the FSI wall with solution stabilization (.25 coefficient-based)
• Model: k-epsilon, as expecting turbulence
• Pressure-Velocity Coupling: PISO 
• Spatial Discretization: Using 2nd order everywhere and least squares cell based
• Max iterations: 5

     Coupled:

• Step size: .0025, Max iterations: 10, RMS convergence: .01, Under Relaxation Factor: 1.

Any help with this would be much appreciated

[Rahul]

It is a common error in FSI simulations, where an element formulation error due to excessive element distortion in the Mechanical APDL solver. When this happens, it is often because unreasonable forces were sent to Mechanical APDL. 

Check whether the force sent to Mechanical was too large(you can check this from SyC log file). This can happen if Fluent didn’t converge, or if there is an FSI instability due to similar densities between solid and fluid.

Using under-relaxation on the data transfers can help here. This will make the solution more stable at the expense of more COUPLING ITERATIONS to reach convergence.