OpenFOAM – SECOND ATTEMPT. AIR FLOW THROUGH A SUGAR BEET CLAMP

After getting familiar with OpenFOAM by building a model of airflow around a clamp, it was time to start looking in the clamp. To be honest, I’ve been waiting 2.5 years for this moment, so I was pretty keen to get a model that worked, regardless of whether it has the right parameters or not.

Taking the above mentioned airflow around a clamp case as my starting point, the first step was to get the clamp area back into the mesh. For a mesh built with snappyHexMesh, the trick was to change the clamp patch from just a surface refinement to be a faceZone and include a cellZone. Note that when the cellZoneInside part was not included, no cells were defined in the polyMesh/cellZone file.

CLAMP EXCLUDED FROM MESH  
"clamp.*"
  {
      // Surface-wise min and max refinement level
      level (4 4);
  }
_____________________________________________________________
CLAMP INCLUDED IN MESH
"clamp.*"
  {
      // Surface-wise min and max refinement level
      level (4 4);
      faceZone clamp_face;
      cellZone clamp;
      cellZoneInside inside;
      insidePoint (1.001 0.01 0.01);
  }

Win. Then onto the physics.

I’m running OpenFOAM v2012, so to run this porous model as an extension of the simpleFoam solver model I ran previously, it is possible to basically just add a fvOption dictionary to the CASE/constant directory, which can be used to defined the porosity properties of the cellZone.

porosity1
{
 type explicitPorositySource;
 active yes;

 explicitPorositySourceCoeffs
 {

  selectionMode   cellZone;
  cellZone        clamp;

  type           DarcyForchheimer;

  D 50;

  DarcyForchheimerCoeffs
  {
   d ($D $D $D);
   f (0 0 0);

   coordinateSystem
   {
      type  cartesian;
      origin  (0 0 0);
      e1 (1 0 0);
      e2 (0 1 0);
   }
  }
 }
}

You’ll note that the Darcy (D) part of the model is here set very low and that the Forchheimer (f) part is all 0s – again, I just wanted to see if it ran, not necessarily produce a defensible model of the system. And with inlet velocity of 1 and 5m/s, and nu of 1e-2 and 5e-2, it did run. 122 and 225 iterations respectively.

• 

•