[dylewiczk]

Hi OpenFOAMers!

I recently implemented adaptive mesh refinement to my simulation using rhoCentralFoam solver.

However, I encounter problem with the maximum number of cells that the AMR generates. My 'base' mesh has ~ 1e6 cells. And I want to refine the field up to ~ 2e6 cells. However, during the simulation it went up to nearly 8e6 cells!

One of the messages prompted by OpenFOAM:
Selected 777925 cells for refinement out of 2269126.
Refined from 2269126 to 7714601 cells.

Here is my dynamicMeshDict:

/*--------------------------------*- C++ -*----------------------------------*\
========= |
\\ / F ield | OpenFOAM: The Open Source CFD Toolbox
\\ / O peration | Website: https://openfoam.org
\\ / A nd | Version: 7
\\/ M anipulation |
\*---------------------------------------------------------------------------*/
FoamFile
{
version 2.0;
format ascii;
class dictionary;
location "constant";
object dynamicMeshDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dynamicFvMesh dynamicRefineFvMesh;

// How often to refine
refineInterval 1;

// Field to be refinement on
field MachGradMag;

// Refine field in between lower..upper
lowerRefineLevel 1e4;
upperRefineLevel 1e6;

// If value < unrefineLevel unrefine
unrefineLevel 1e8;

// Have slower than 2:1 refinement
nBufferLayers 3;

// Refine cells only up to maxRefinement levels
maxRefinement 5;

// Stop refinement if maxCells reached
maxCells 2278148; // twice the base mesh

// Flux field and corresponding velocity field. Fluxes on changed
// faces get recalculated by interpolating the velocity. Use 'none'
// on surfaceScalarFields that do not need to be reinterpolated.
correctFluxes
(
(neg none)
(pos none)
(phi none)
);

// Write the refinement level as a volScalarField
dumpLevel true;

End of dynamicMeshDict

Could you guys help me out understand why OF is doing what it is doing? Is it something to do with buffer layers and maximum refinement?

Thanks!

[Volkanatar]

Hello have you figured it out why it is like that? I am having exactly the same problem with a steady state solver.

Thanks