[Krao]

Hi dear cfMesh/foam users,

I am using cfMesh since last two years for my cfd applications, thanks to the great tool. So far I had very good experience with cfMesh. As of now I am generating mesh for a wing of propeller which has a similar airfoil profile throughout as shown in the figure airfoil.png. Due to very thin region near the trailing edge no matter how good my refinements are I am unable to generate a mesh with more than two boundary layers. Also with two BLs I am still having a saw-tooth shaped airfoil profile at the trailing edge. Any attempt to add more boundary layers or making more refinements in the region resulting in poor to worst mesh quality during the checkMesh. It would be very great if anyone using cfMesh share their experience for a similar application/similar problem.

My meshing procedure is as follows,

1) Generate geometry in CAD (.stl)
2) Clean/Simplify the geometry using blender(de-featuring)
3) Generate the bounding box
4) Extract/generate surfaces for surfaceRefinement
5) Extract features using surfaceFeatureEdges and storing it as .vtk which is used for edge refinement
6) cartesianMesh to generate mesh

My meshDict,

surfaceFile "dynamic.stl";

maxCellSize 0.027;

localRefinement
{


blade
{
additionalRefinementLevels 4;
refinementThickness 0.03;
//cellSize 0.05;
}

}


objectRefinements
{

cylindrical1
{
type cone;
p0 (0 0 0.06);
p1 (0 0 -0.06);
radius0 0.225;
radius1 0.225;
additionalRefinementLevels 2;
refinementThickness 0.027;
}
box1
{
type box;
lengthX 0.22;
lengthY 0.08;
lengthZ 0.045;
centre (-0.12 0.016 -0.011);
additionalRefinementLevels 4;
refinementThickness 0.00375;
}

}
edgeMeshRefinement
{
edge1
{
edgeFile "edgemesh_featureEdges.vtk"; //tip
additionalRefinementLevels 7;
//cellSize 0.005;
refinementThickness 0.0002;
}
}

surfaceMeshRefinement
{
SR1 //tip
{

surfaceFile "SR1.stl";
additionalRefinementLevels 7;
refinementThickness 0.0009;

}



SR2 //LE
{

surfaceFile "SR2.stl";
additionalRefinementLevels 7;
refinementThickness 0.0009;

}

SR3 //TE
{

surfaceFile "SR3.stl";
additionalRefinementLevels 7;
refinementThickness 0.0009;

}


}

boundaryLayers
{


patchBoundaryLayers
{

"(blade)" //patch name
{
nLayers 2;
thicknessRatio 1.2;
maxFirstLayerThickness 0.001;
allowDiscontinuity 0;

}



}

optimiseLayer 1;

optimisationParameters
{

nSmoothNormals 5;

maxNumIterations 5;

// feature size factor. Reasonable range <0.2, 0.5>
// lower values force thinner layers
featureSizeFactor 0.4;

// shall the normal vectors be recalculated
reCalculateNormals 1;

// relative thickess variation between two hair nSmoothNormals
// lower value produce thinner and more uniform layers
relThicknessTol 0.1;
}

}




renameBoundary
{

newPatchNames
{

"blade"
{
newName blade;
type wall;
}

"(patch0|patch1|patch2)"
{
newName rotor;
type patch;
}

}



}

Thank you in advance
Krao

[RaulCA]

In those cases i stop the proccess after the edgeExtraction step with the workflowControls entry. If this mesh has not the shape you want at the boundaries, you should change something un your settings. You can try with the enforceGeometry option, although it will probably generate a very bad mesh.

[Krao]

Hi RaulCA

Originally Posted by RaulCA

In those cases i stop the proccess after the edgeExtraction step with the workflowControls entry. If this mesh has not the shape you want at the boundaries, you should change something un your settings. You can try with the enforceGeometry option, although it will probably generate a very bad mesh.

Thank you for your reply. As you have mentioned enforcing geometric constraint has resulted in poor geometry. As you have mentioned about edge extraction step. If it is less refinement I end up with sawtooth edge. With more refinement I'm getting more number of skewness and bad cells. I am facing this problem so long for the first time due to very thin edge.

Krao

[nope]

Hi Kroa

Is the thin edge perfectly sharp? If so, consider to make it blunt. This is easier to mesh and in reality a perfectly sharp thin edge doesn't exist.

This does not really solve your problem, sorry about that. But it might be a way to get around the problem.

Cheers, nope

[Krao]

Originally Posted by nope

Hi Kroa

Is the thin edge perfectly sharp? If so, consider to make it blunt. This is easier to mesh and in reality a perfectly sharp thin edge doesn't exist.

Cheers, nope

Hi nope,

Yes the thin edge is perfectly sharp. I will consider your inputs, thank you. Maybe taking away a couple of micrometers would solve my problem.

Originally Posted by nope

This does not really solve your problem, sorry about that. But it might be a way to get around the problem.

Cheers, nope

True! Engineering is all about approximations and finding the best solution with the given amount of resources. Cheers!

If anyone have any other ideas I'm open to those as well, thank you.

Krao

[Krao]

Maybe this link would be useful for the future readers, page 24 of this thread http://www.arek.pajak.info.pl/wp-con...flow_cadOF.pdf discuss about having blunt trailing edges instead of sharp trailing edges.

Krao