[jpina]

Dear all,

I have a general question. In a multiphase VOF transient simulation and with the interested in tracking the interphase position through time, should I get the residuals converge at each time step?

I have found that if I monitor my residuals and set an "absolute" convergence criteria, it is common that, once reached the target, the solver goes into the next time step. However, sometimes it happens when the residual is still decreasing steeply.

I have thought that maybe I should use the relative convergence criteria for the residuals.

Well, my question to sum up is:

- Should all the residuals' plot be flat before going into the next time step?

[jpina]

I mean, would this residuals be correct for a VOF simulation where I am interested in tracking the interphase through time?



Or should I get flat residuals before going to the next time step?