CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Meshing & Mesh Conversion (https://www.cfd-online.com/Forums/openfoam-meshing/)
-   -   [snappyHexMesh] Smoothing the surface (https://www.cfd-online.com/Forums/openfoam-meshing/128135-smoothing-surface.html)

Tobi January 4, 2014 11:57
Smoothing the surface
 
2 Attachment(s)
Dear all,

I am working with sHM a long time but I never solved the following problem!
If I am using multi-region cases I got bad surface approximation on curved surfaces (see the attached pictures).

If I mesh just one region - everything is fine.
Does someone know any solutions for this? I think this depend on the multi-domain function of sHM. Maybe it happend after splitting the mesh?!

Regards
Tobi

alexeym January 4, 2014 18:24
But can you exactly post what is the problem? Wasn't able to figure out it from the pictures you've attached.

Tobi January 5, 2014 12:32
Quote:
Originally Posted by alexeym (Post 468615)
But can you exactly post what is the problem? Wasn't able to figure out it from the pictures you've attached.

Hi,

the problem - maybe its not a problem - is that the surface is not smooth enough. If I just mesh the cup region then I get a extreme accurate surface approximation. If I am home again I will do this again and post a new picture so that the differences are clear.

To clear it up:

- one region: accurate and smooth surface
- multi region: no smooth surface (see pictures) - snatchy surface.


Regards Tobi

Tobi January 5, 2014 17:10
3 Attachment(s)
Okay.

In the attachment you find the surfaces after meshing but just the cup (no multidomain with cellZones and faceZones).

The question now. Is it possible to get such a surface with the multidomain meshing too or not?

Regards
Tobi

mirx January 9, 2014 03:53
I do not know where these issues come from, but I was able to fix it by trying different mesh settings.

You have to play with the values: nSolveIter, nRelaxIter, nFeatureSnapIter. explicitFeatureSnap true and multiRegionFeatureSnap false did a better job for me than implicitFeatureSnap. Furthermore the meshQualityControls may have an influence on the result. As a starting point you may use my settings:

Code:
snapControls
{
  nSmoothPatch 3;
  tolerance 1.0;
  nSolveIter 60;
  nRelaxIter 5;
  nFeatureSnapIter 5;
  implicitFeatureSnap false;
  explicitFeatureSnap true;
  multiRegionFeatureSnap false;
}
Code:
meshQualityControls
{
  maxNonOrtho 100;
  maxBoundarySkewness 40;
  maxInternalSkewness 8;
  maxConcave 80;
  minVol 0;
  minTetQuality 1e-30;
  minArea -1;
  minTwist 0.02;
  minDeterminant 0.001;
  minFaceWeight 0.02;
  minVolRatio 0.01;
  minTriangleTwist -1;
  nSmoothScale 4;
  errorReduction 0.75;
}

Tobi January 9, 2014 08:50
Hi,

I will check it out.
I used explizit Feature snapping and have more iterations on smoothing as you mentioned but I will check you Settings.

It only appears with multi-Region!

mwaqas June 4, 2018 10:18
Hello Tobi,


Have you figured out something to smooth surface while using SHM for multiregion meshing?
I have seen your tutorials on your website for mutiregion meshing. In those tutorials, surface is not smooth.
Did you face any problem in simulation due to non-smooth surface?
Thank you


Regards


All times are GMT -4. The time now is 10:27.