[giorgianig]

I am using mapField to map a coverged solution from a coarse mesh to a fine one. When I re-start the simulation on the fine mesh, the simulation crashes at the beginning of the time loop.
I have discovered that there are a bunch of points (like 5 points) where omega (the turbulence variable) is very large, and I think this is the reason for the crash. Probably mapField is generating this large values do to the fact that the boundaries are not exactly the same (the two meshes are generated with snappyHexMesh).
Is there a way to locate these few points and correct omega manually?

Thanks in advance

[unilord]

Originally Posted by giorgianig

I am using mapField to map a coverged solution from a coarse mesh to a fine one. When I re-start the simulation on the fine mesh, the simulation crashes at the beginning of the time loop.
I have discovered that there are a bunch of points (like 5 points) where omega (the turbulence variable) is very large, and I think this is the reason for the crash. Probably mapField is generating this large values do to the fact that the boundaries are not exactly the same (the two meshes are generated with snappyHexMesh).
Is there a way to locate these few points and correct omega manually?

Thanks in advance

Easiest way would be to use upwind divScheme for turbulence constants, and cellLimited Gauss Linear 1 for grad Schemes until it starts converging.

[giorgianig]

Originally Posted by unilord

Easiest way would be to use upwind divScheme for turbulence constants, and cellLimited Gauss Linear 1 for grad Schemes until it starts converging.

The problem is that it diverges before starting the time loop. The limiting would kick-in after the first iteration.

[Severus]

Hello,
You have two options,
Option 1:
You can use limitFields functionObject: https://www.openfoam.com/documentati...mitFields.html
Note that the functionObject gets called in the beginning of the time loop, so it should work before things are computed in the first time step.
Check here: What is the trigger of the function objects??

Option 2:
If you can add piece of code inside your solver and recompile your solver, then I suggest you add something like this, before the time loop.
Lets say that the maximum allowable value of omega is max_value.
Then use

omega = min(omega,max_value);

Thanks