Personal tools
You are here: Home Forum Use Viscous Layers on a complex geometry

Viscous Layers on a complex geometry

Up to Use

Viscous Layers on a complex geometry

Posted by Andrea Bartolini at November 20. 2012

Hi, the object that i'm trying to mesh is a fluid domain consisting of a sphere with an airplane geometry (DLR-F6) cut in it, while ' im getting relatively satisfying 3d meshes via Tetrahedron and Netgen 2D-1D algorithms once i add the viscous layers hyphothesis the computing crashes immediately (viscous layers failed at the very first...) or most of the times it takes like hours to get to another crash or an error message.

The plane is scaled to be 1 long whit a spherical domain of 13 diameter, i tried varying the layers total thickness (0.01 to 0.001)  and launching the hypothesis from a coarser basic mesh but nothing so far.

I'd really appreciate some fresh ideas. Thank you so much.

Re: Viscous Layers on a complex geometry

Posted by SMESH expert at November 20. 2012

Hello Andrea

As a developer of this algorithm I can't say anything without even seeing a model. Can you provide a snap-short of it? or, better, a study?

I suppose in the model there could be small cavities or something like this. In principle they should not be obstacles, but there must be not killed bugs, as for about 2 years of the algorithm age there were only 5 bugs reported - suspiciously few :)

Best regards

Edward

Re: Viscous Layers on a complex geometry

Posted by Andrea Bartolini at November 22. 2012

Plane's geometry is at this link (http://aaac.larc.nasa.gov/tsab/cfdlarc/aiaa-dpw/Workshop2/DLR-F6-geom.html), that had to be transfered in a Cad environment in order to get the solid to cut from the sphere, that cause i wasn't able to get it from the geometry module (friend of mine told me that was an original geometry' s issue)

Study is at the following link http://www.megafileupload.com/en/file/375542/study-hdf.html

Again, Thank You

 

Re: Viscous Layers on a complex geometry

Posted by Andrea Bartolini at November 22. 2012

Plane's geometry is at this link (http://aaac.larc.nasa.gov/tsab/cfdlarc/aiaa-dpw/Workshop2/DLR-F6-geom.html), that had to be transfered in a Cad environment in order to get the solid to cut from the sphere, that cause i wasn't able to get it from the geometry module (friend of mine told me that was an original geometry' s issue)

Study is at the following link http://www.megafileupload.com/en/file/375542/study-hdf.html  the solid that i need to mesh is PLANE_FINAL

Again, Thank You

 

Re: Viscous Layers on a complex geometry

Posted by SMESH expert at November 26. 2012

Hello Andea

I have fixed some bugs and now your case works. Would you like I send you a patched source code and you re-build SALOME from sources? or I can compute your mesh and sent you your study?

SALOME 6.6 is issued soon (~ in a month), I hope my fix will be included. So another option is to wait SALOME 6.6 release.


Best regards

Edward

Re: Viscous Layers on a complex geometry

Posted by Andrea Bartolini at November 26. 2012
The study could be enough but since i'm doing this as work for the university i might be needing to change some settings myself depending on what my teacher says. Could you send me both?
 
 

Re: Viscous Layers on a complex geometry

Posted by SMESH expert at November 26. 2012

With what parameters of Viscous Layers hypothesis to compute your mesh?

Re: Viscous Layers on a complex geometry

Posted by Andrea Bartolini at November 26. 2012

total thickness: 0.01

n layers: 3

stretch factor: 1.2

every face except the spherical outer one, i renamed it as SFERA 

Re: Viscous Layers on a complex geometry

Posted by Andrea Bartolini at November 26. 2012

sorry for bothering again, could you make a mesh of PLANE_FINAL_2 with the same parameters as well? again i really appreciate your help!

Powered by Ploneboard
Document Actions