CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   STAR-CCM+ (https://www.cfd-online.com/Forums/star-ccm/)
-   -   Star not Using Desired Surface Sizes (https://www.cfd-online.com/Forums/star-ccm/218542-star-not-using-desired-surface-sizes.html)

Chris2337 June 24, 2019 13:10

Star not Using Desired Surface Sizes
 
Hello all,

I am running a sailplane in Star for my Master's research (the novel part being the transition model I'm using) and have had some success. I've decided to go with the trimmer over the polyhedral since it meshes faster and I can put cells where I want them pretty cheaply cell count wise.
A problem I've been having is certain boundaries haven't been using the surface size I specify. The fuselage, for example, I specify a target and min of 100% base size but Star generates a mesh with 50% the base size. This happens even if I specify the global min as 100% the base size.
I think this started happening when I realized I needed to refine the fuse nose separately of the fuse so I created a new face in the Star CAD model, then updated the parts and updated the region with a new boundary of the fuse nose. Could this corrupt the model to not care about what I specify (or did it change something I haven't noticed)? I'm pretty sure it's happening at the surface remeshing level because if just do a surface mesh the refinement cells are about 50% of the base, but I disabled curvature refinement on the fuse because of some CAD defects so it can't be refining the mesh based on curvature (not that the fuse has high curvature anyway).
Any thoughts are appreciated!
Thanks,
Chris

ashokac7 June 25, 2019 03:06

Quote:

Originally Posted by Chris2337 (Post 737118)
Hello all,

I am running a sailplane in Star for my Master's research (the novel part being the transition model I'm using) and have had some success. I've decided to go with the trimmer over the polyhedral since it meshes faster and I can put cells where I want them pretty cheaply cell count wise.
A problem I've been having is certain boundaries haven't been using the surface size I specify. The fuselage, for example, I specify a target and min of 100% base size but Star generates a mesh with 50% the base size. This happens even if I specify the global min as 100% the base size.
I think this started happening when I realized I needed to refine the fuse nose separately of the fuse so I created a new face in the Star CAD model, then updated the parts and updated the region with a new boundary of the fuse nose. Could this corrupt the model to not care about what I specify (or did it change something I haven't noticed)? I'm pretty sure it's happening at the surface remeshing level because if just do a surface mesh the refinement cells are about 50% of the base, but I disabled curvature refinement on the fuse because of some CAD defects so it can't be refining the mesh based on curvature (not that the fuse has high curvature anyway).
Any thoughts are appreciated!
Thanks,
Chris


Target size is only achieved where it has sufficient region to grow. So it will start with minimum size and grow it towards target size depending upon the growth rate we have specified.
So this will depend on your surface growth rate if surface mesh is the issue you are talking about. There is also volume growth rate.


And it is not accurate to say that poly mesher will give higher mesh count. Only that trimmer mesher is faster in terms of generation but for small complex geometries, complex flows, poly should be considered. Poly meshes also converge faster than trimmer. The only issue poly faces is when we have large geometry. The growth inside volume is slow for poly. It does not have hanging node like trimmer because of which trimmer have some advantages is large size geometries.



So it is my common practice to mesh with trimmer first, check for surface captures, and then shift to poly for final mesh iteration.
Hope this helps.:)

Chris2337 July 1, 2019 11:45

Thanks for the info! Sorry I phrased the cell count statement poorly. The polyhedral is almost always less although things like the file size (poly can be x2 for same cell count) make it harder, but even last night I had some luck with it. That said my geometry is pretty large (half of a 18m sailplane) so my trimmer mesh is like 70,000,000 cells at ~8GB which is big enough already.

Playing with the surface growth rates seemed to help, thanks!


All times are GMT -4. The time now is 17:43.