Ultrasound FSI Simulation, Lack of Convergence

[UarkGradStdnt]

Hello all,
I am interested in running simulations investigating ultrasound wave behavior in fluid-filled porous solids (could be metals or biomaterials like bone). I was able to successfully run the 'fluid flow past a flap' example on this forum. I was also able to propagate an ultrasound wave in a copper plate in the FEBio Structural Mechanics module with results comparable to those of a reference publication [Chen, Hanxin, et al. "Simulation of ultrasonic testing technique by Finite Element Method." Prognostics and System Health Management (PHM), 2012 IEEE Conference on. IEEE, 2012.]. However, when I attempt to run a similar simulation using the Fluid-FSI module of a copper plate with a water filled circular pore, the simulation does not converge/gets hung up and I still cannot determine why. Am I setting up the simulation conditions incorrectly? Is it more theory related, where perhaps I didn't realize FEBio is not well-conditioned for the type of simulation I want to run? For some reason I am unable to attach the files, so I can email the simulation files if anyone is interested. Any help or suggestions would be greatly appreciated! Thank you!

[jshim777]

Hi

If the attachment feature does not work, you can send me the files at (jjs2215 at columbia dot edu) and I can take a look

Thanks

[jshim777]

Hi

I was able to get it to work by setting vtol and ftol to 0, ie turning them off. I think the issue was that for a long time in the simulation the fluid is not doing anything so the solver is basically trying to solve for numerical round off errors, so as the result, the residuals for v and f were extremely small and was hard to meet the tolerance. Later on when the wave hits the fluid, you may want to turn them back on if the fluid solution is not sufficiently converged when the tolerances remain disabled. You can do this by having two analysis steps: one where vtol and ftol are turned off up until the wave hits the fluid, and one similar step afterwards where the only different is they are turned back on. It seems like your wave hits the fluid at around time t=1.6e-6. Hopefully this helps!

[UarkGradStdnt]

Hello,
Thank you very much for your help! I thought it might have something to do with the conditions I had on the fluid domain since the copper plate simulation worked fine. I will keep a closer eye on my choice of fluid tolerances.