[foxcfd]

Hi guys,


I am trying to simulate the phase transition from water to ice using
buoyantPimplefoam and the fvOption solidificationMeltingSource.
Using this i created a simple Stefan-Problem consisting of adiabatic
walls and a single fixed temperature wall. As the temperature of this
wall is lower than the liquid, the liquid start to freeze.

I have the problem that the analytical and numerical data do not match.
I have attached several files, the first two images are my thermophysical
properties and my fvOptions file and the third image shows the simulation
progress, top is the temperature and bottom is the phase fraction of liquid/ice.
0 stands for ice whereas 1 stands for liquid. Lastly the comparison between
the simulated and numerical results.



Does anyone have an idea what could have caused the difference
between the two solutions? The exact same values from the simulation
were used to calculate the numerical solution.


Thanks in advance.

[foxcfd]

Hi everyone,

just a quick follow up. I have significantly reduced
my mesh and time step, which has resulted in my
numerical solution approaching that of the analytical
one. I assumed that mesh and my time step would
be fine enough, but that was apparently not the case.