|
[Sponsors] |
December 2, 2008, 01:27 |
mesh quality to CFX
|
#1 |
Guest
Posts: n/a
|
Hi all, Can any tell about TET and Prism mesh minimum quality criteria (such as Aspect ratio &hellip to export to CFX solver from Hypermesh.
thanks in advance |
|
December 2, 2008, 11:18 |
Re: mesh quality to CFX
|
#2 |
Guest
Posts: n/a
|
Look up mesh quality in the CFX documentation (ANSYS CFX-Solver Modeling Guide | Advice on Flow Modeling | Mesh Issues | ).
|
|
December 3, 2008, 06:44 |
Re: mesh quality to CFX
|
#3 |
Guest
Posts: n/a
|
It's interesting theme. Would you mind if I add my question? What ICEM quality criterion should be used first for elements' smoothing? This mesh is for CFX modeling. I heard ICEM criterion "Quality" is main quality characteristic for Fluent. Is there similar ICEM criterion(s) for vertex centered CFX solver? And what is reasonable interval of it? Please share your experience. Thanks for replies.
|
|
December 4, 2008, 00:21 |
Re: mesh quality to CFX
|
#4 |
Guest
Posts: n/a
|
Really, the same criterion apply. Only FLUENT is more sensitive to mesh quality than CFX (because the vertex centered control volumes are effectively polyhedral elements). The main things to avoid are:
1. Sudden mesh expansion (i.e. high volume ratio), 2. Highly skewed elements (look at dihedral angle, which is the angle between two faces, as opposed to face angle, which is the angle at the corner of a face between two edges), 3. Clusters of poor quality elements. As a general bit of advice, don't spend too much time smoothing and fixing your mesh. If the mesh needs that much smoothing or repair you are better off going back and fixing the mesh controls that led to the problem. Include sufficient resolution in areas of high geometric complexity or, if the region is not important, remove the complex features. If you are using Hexa, focus on the quality of the blocks. Hexa offers tremendous flexibility for creating hex meshes, but if a poor quality block will result in many poor quality elements. Individually these may not affect you convergence or solution, but collectively they will wreak havoc. If you can't ensure the block quality, make it a swept or unstructured block and move on. There is no point in wasting 8 hours perfecting a mesh that is only saving you 2 hours of run time . Lastly, check out Workbench meshing too. You'd be surprised how good it is and it is especially good when it comes to changes in geometry. Workbench meshing allows you to update your geometry while maintaining your mesh controls. Most of the time you can just update the geometry and regenerate the mesh without having to do any re-work! -CycLone |
|
December 4, 2008, 01:55 |
Re: mesh quality to CFX
|
#5 |
Guest
Posts: n/a
|
Thanks a lot CycLone for your comprehensive answer! Georg
|
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[ICEM] CGNS mesh failing in CFX | siw | ANSYS Meshing & Geometry | 5 | November 23, 2017 20:07 |
simulating 2D airfoil mesh in CFX | iffy | ANSYS Meshing & Geometry | 7 | January 24, 2017 04:05 |
Gambit problems | Althea | FLUENT | 22 | January 4, 2017 04:19 |
increasing mesh quality is leading to poor convergence | tippo | CFX | 2 | May 5, 2009 11:55 |
hex mesh using cfx mesh. | mehul | CFX | 3 | December 12, 2005 23:06 |