[Matt1986]

Hi everyone,

I have a problem with my first Openfoam model.
I have modified the motorbike example, which can be found in the Openfoam-tutorials.
I have changed the bike geometry to a ball geometry and located the geometry in the center of the cavity. Blockmesh and snappyHexMesh are working fine. The simpleFoam solver is also running, but there is no convergence. How can I fix this problem.
I have tried to change the size and the parameters of the mesh whitout effect.
I am not sure, what could be changed in the the fvSolution file, without losing precision.

Looking for help
Matthias

[owayz]

Hi Mathias,
Can you share your fvSolution and fvSchemes file.
Boundary conditions could be the initial guess for the problem. If you can explain how and which B.Cs you are using it would be easier to help.
In my experience schemes and solver has less effect. But boundary conditions are the 'USUAL SUSPECT' in a convergence problem. Also the under-relaxation factors effect the convergence alot.
Regards,
Awais

[Matt1986]

Hi Awais,

Thank you for your answer. In the attachment you find my fvSolution and my fvSchemes files. I have haven't changed sth. in these files. They are the originals from the motorbike tutorial (incompressible simpleFoam). When I start simpleFoam the terminal tells me that no convergence criteria was found, although I have only modified the example of the tutorial.

Here are my boundary conditions:

The cavity consists of the following 6 walls and the ball:
inlet
outlet
front
back
upperwall
lowerwall
ball

inlet :
k / omega / U :type fixedValue;
value $internalField;

nut : type calculated;
value uniform 0;



p: type zeroGradient;

outlet :

k / omega / : type inletOutlet;
inletValue $internalField;
value $internalField;



U : type inletOutlet;
inletValue uniform (0 0 0);
value $internalField;



nut : type calculated;
value uniform 0;



p: type fixedValue;
value $internalField;

front back upperwall lowerwall:

k / omega / U : type slip;

nut : type calculated;
value uniform 0;



p: type slip;


ball:


k : type kqRWallFunction;
value $internalField;


omega: type omegaWallFunction;
value $internalField;
U : type fixedValue;
value uniform (0 0 0);


nut: type nutkWallFunction;
value uniform 0;


p: type zeroGradient;


Regards,

Matthias

[Matt1986]

The attached Solution files

[owayz]

Hallo Mathias,
Apparently I can't find anything wrong. The only thing I could think of is Mesh. How does your mesh look like?
If you could share your mesh and transport properties file as well.
Regards,
Awais

[Matt1986]

Hi Awais,

Thank you for your help. In the Attachement you find the files and a screenshot of the mesh. The fluid flows from the right to the left side.

Regards Matthias

[Matt1986]

Hi Awais,

I have checked the Mesh again and finally set the refinement level in the snappyhexmeshdict to level 3. After a clocktime of 4130 seconds the Solution has convergenzed after 871 iteration steps. Thank you very much for your hints and your support.

Regards Matthias