|February 3, 2014, 13:06||
Axis treatment in axisymmetrical problem
Join Date: Nov 2013
Posts: 20Rep Power: 4
I am trying to run simulations on an axisymmetrical geometry. To set it up I used Salome to generate a flat, one layer mesh of NetGen2D prisms. Applying makeAxialMesh and collapseEdges I produced a wedge type geometry which looks good, aside from checkMesh complaining about a node being located ~10e-8 outside the respective plane. I don't know if that is an issue alltogether.
To set up the case I assigned the following patch types:
- old front/back group: empty
- new old/back patches: wedge
- symmetry axis: symmetryPlane
- other patches (walls) : normal boundary conditions (fixedValue, zeroGradient etc)
When I now run the case I run into problems. I think it boils down to the symmetry axis being regarded as some kind of wall type. Specifically, I am trying to model the one dimensional sinking of a solid body in a liquid (parallel to the symmetry axis). The result is an unphysical pressure field towards the axis (see attachment), which, I think, stems from the material not being able to move, even though a force/acceleration is applied.
Since the balance of applied forces appears to be calculated correctly, the problem must lie within the symmetry condition of the axis. If the wall conditions would represent a correct symmetry, the sinking process should be modelled correctly on "both sides of the symmetry". But apparently the solid "gets stuck" at the axis.
Now I wonder if treating the axis as a symmetryPlane type patch is the correct choice. And do I even need to specify the axis as a seperate patch at all? Shouldn't the definitions of the wedge planes provide enough information? The axis patch seems more like an artifact of the mesh generation process to me.
Did anyone ever face similar issues? Would be really great if someone had some hints on how to handle this problem! Thank you in advance!
|February 3, 2014, 16:54||
Join Date: Aug 2011
Location: Nancy, France
Posts: 1,419Rep Power: 25
If you take a look at
In my 2D axisymmetric simulations I also use 'empty' patch type for axis without any problems.
|August 25, 2014, 03:09||
Join Date: Feb 2014
Posts: 7Rep Power: 4
i want to get axi-symmetrical mesh with extrudeMesh,after i run autoPatch ,i cannot find the axis ,so i cannot define it .how can i get the axis
|Thread||Thread Starter||Forum||Replies||Last Post|
|[DesignModeler] axis display disparity of model and triad||ram881973||ANSYS Meshing & Geometry||12||January 7, 2014 12:51|
|problem with axis boudray and mixture flow||m.beh||FLUENT||0||December 5, 2010 03:13|
|Can I solve this problem by Fluent?||Kai_kc||FLUENT||1||October 27, 2010 05:29|
|Enhanced Wall Treatment Problem||Andy Cong||FLUENT||0||November 13, 2005 23:55|
|UDF and the wall treatment problem||Andy Cong||FLUENT||0||November 11, 2005 03:12|