Bug with Two-phase.h FILTERED

[Bradley Boyd]

Hello All

I found a bug in the two-phase.h file when using FILTERED. If FILTERED is used, then the initial capillary time constraint (in tension.h) is far more strict. The reason is that the rho is computed  as rhov[] = cm[]*rho(sf[]) in the properties event. However, sf[] is not computed until the later tracer_advection event. This means that the variable ‘rhom’  is then incorrectly determined in tension.h stability calculation.

Attached is a simple example.

Cheers

Bradley Boyd