Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

How to create a video of acoustic/elastic wave propogation in poroelastic material and tunnels.

Please login with a confirmed email address before reporting spam

Hello, I'm trying to simulate resonating sound waves in a rabbit burrow for my lab. I've successfully done this with the poroelastic-solid-acoustic wave multiphysics setup, but this uses frequency domain. It would be extremely helpful for us to be able to watch the propogation of the wave, so I tried using a transient acoustic block in place of the frequency domain . This resulted in an extremely long simulation time and the file reaching sizes of several dozen gigabytes.

(File PourousBunnyTimeTrans.mph)

I figured this was due to either incorrect implementation of the transient block or the poroelastic module interaction being too complex, so I tried toning it down. Instead, I just did an acoustic-solid interaction multiphysics simulation accepting that I'd lose the added accuracy of modeling soil as a poroelastic material. This went well but didn't give "Sound Pressure" as a possible output to plot, which is what we'd been using to visualize resonance up till now.

file: BunnyTimeTrans.mph

I tried to get the sound pressure back by running a simulation with both the transient and the frequency domain components by using a modal transient analysis. This yielded the following error "Time-dependent equation residual vector is not supported for modal analysis."

file: BunnyTimeTransFreq.mph

Does anyone know how to fix this error, or a better way to simulate the scenario of a block of dirt with a cylinder of air in it being excited by a plane wave source from the top so that I view the propogation of the sound wave through the dirt and/or cylinder of air?

Thank you, Daniel Magley



1 Reply Last Post 24.08.2018, 18:10 GMT-4
Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 24.08.2018, 18:10 GMT-4

Daniel,

you might try running a frequency sweep using the Fourier transform of some excitation pulse to excite the frequency domain study. Afterwards use an inverse FFT step to transform into time domain. We successfully did this for an electromagnetic, solid, acoustic coupled model.

Putting it all together will be a little tedious though. And of course the calculation time will multiply by the number of frequency steps you need to resolve the excitation pulse and runtime of the transient.

Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Daniel, you might try running a frequency sweep using the Fourier transform of some excitation pulse to excite the frequency domain study. Afterwards use an inverse FFT step to transform into time domain. We successfully did this for an electromagnetic, solid, acoustic coupled model. Putting it all together will be a little tedious though. And of course the calculation time will multiply by the number of frequency steps you need to resolve the excitation pulse and runtime of the transient. Cheers Edgar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.