Problems with AMR and compressible multiphase flow

[George Hunter-Brown]

Hi everyone,

I want to simulate bubble collapse due to a planar shockwave, so have been having a look at Daniel Fuster's sandbox to simulate compressible multiphase flows. I currently can't get either the oscillation or shockwaveadapt examples to run. I have been able to modify the spurioustest example to allow for surface oscillations, but if I try and include adaptive mesh refinement it seems to crash the code. 

I'm wondering is it possible  to use AMR with the compressible multiphase solver, and if anyone's had similar issues, and if so, how have you managed to resolve them? 

many thanks, 

George

[MANDEEP SAINI]

Dear Gorge

It is indeed possible to use AMR with two-phase-compressible.h from Dr. Daniel Fuster's sandbox. The trick is that you always stay at max level at the interface, you can do this by using a copy of scalar 'f' in the list which is an argument adapt wavelet function for example:

scalar fscalar[];

event adapt (i++) {

foreach(){
     fscalar[] = f[];
 }
boundary({fscalar});

adapt_wavelet((scalar *){fscalar},(double[]){0.0001,},maxlevel = MAXLEVEL); 

}

Best

Mandeep