[billy]

I tried to produce a lid driven cavity flow using interFoam. I set initial values of gamma in some cells to 1.0 and the rest to 0.0. I wanted to simulate a mixing of two fluids, however I get always the same message even using smaller time steps:

...
Reading transportProperties

Selecting incompressible transport model Newtonian
Selecting incompressible transport model Newtonian
Calculating field g.h

time step continuity errors : sum local = 0, global = 0, cumulative = 0

Starting time loop


Max Courant Number = -0
Time = 0.001

Liquid phase volume fraction = 0 Min(gamma) = 0 Max(gamma) = 0
Liquid phase volume fraction = 0 Min(gamma) = 0 Max(gamma) = 0
Liquid phase volume fraction = 0 Min(gamma) = 0 Max(gamma) = 0
Liquid phase volume fraction = 0 Min(gamma) = 0 Max(gamma) = 0
time step continuity errors : sum local = nan, global = nan, cumulative = nan
ExecutionTime = 0.12 s

...

Does anyone know what might be wrong?

[billy]

Actually the error looks like this:

...

Max Courant Number = -0
deltaT = 0.001
Time = 0.001

Liquid phase volume fraction = 0.0625 Min(gamma) = 0 Max(gamma) = 1
Liquid phase volume fraction = 0.0625 Min(gamma) = 0 Max(gamma) = 1
Liquid phase volume fraction = 0.0625 Min(gamma) = 0 Max(gamma) = 1
Liquid phase volume fraction = 0.0625 Min(gamma) = 0 Max(gamma) = 1
time step continuity errors : sum local = nan, global = nan, cumulative = nan
ExecutionTime = 0.1 s



Max Courant Number = -0
deltaT = 0.001
Time = 0.002

Liquid phase volume fraction = 0.0625 Min(gamma) = 0 Max(gamma) = 1
Liquid phase volume fraction = 0.0625 Min(gamma) = 0 Max(gamma) = 1
Liquid phase volume fraction = 0.0625 Min(gamma) = 0 Max(gamma) = 1
Liquid phase volume fraction = 0.0625 Min(gamma) = 0 Max(gamma) = 1
time step continuity errors : sum local = nan, global = nan, cumulative = nan
ExecutionTime = 0.22 s

...

[mattijs]

You seem to get a division by 0.

Also you have no flow? Nothing is solved for it seems? Do you have valid boundary conditions? Valid initial conditions?

If you are interested in where it goes wrong: set the environment variable FOAM_SIGFPE to 1 which will make the code coredump and then look at the core in a debugger.

[billy]

I don't know what happened, but I know it was due to the mesh file. I created one using blockMesh and it worked OK. The original mesh file seemed OK when viewing it in paraview though. Does OpenFOAM automatically orient the face normals?

[mattijs]

Yes, all boundary faces point away from the owner cell. There is a section in the user guide about the mesh structure.

[billy]

Hi,

After I got this case working (Lid-driven cavity flow with two phases) I noticed that altering the surface tension produces little effect on the output. I tried positive, negative and zero values of the surface tension and there seems to be no mix between the two fluids. Also, both fluids have same material properties such as density and viscosity.

I assume that:

A postive surface tension -> immiscible
A negative surface tension -> miscible

Is this right?

PS: I wish to thank the OpenFOAM team for making available such a great software.

[henry]

interFoam uses a VOF-type method to capture the interface which of course does not allow mixing of the phases, that is the point. Negative surface tension has no physical meaning and may very well destabilise the solution algorithm.