|
[Sponsors] |
May 17, 2016, 09:35 |
Strange non-hexahedral cells?
|
#1 |
New Member
Join Date: May 2016
Posts: 3
Rep Power: 10 |
Hi,
I'm quite new to openfoam and I am seeing a strange behavior on a very simple blockMesh I want to use for snappyHexMesh.. It is just a cuboid with cube-cells, but when I view the Mesh with paraView, I see very strange cells, that I can't explain. I only see them in the middle of the block in the XZ-Plane. Attached you find an image of ParaView showing these strange cells, if you want to reconstruct using the Dicts below, create a slice in ParaView, click "Y normal" and don't move the plane... Would be nice, if someone here has an idea, what I am doing wrong.. Thanks a lot for your help! system/blockMeshDict Code:
FoamFile { version 2.0; format ascii; class dictionary; location system; object blockMeshDict; } convertToMeters 1; vertices ( ( -0.02 -0.02 -0.032) ( 0.02 -0.02 -0.032) ( 0.02 0.02 -0.032) ( -0.02 0.02 -0.032) ( -0.02 -0.02 0.013) ( 0.02 -0.02 0.013) ( 0.02 0.02 0.013) ( -0.02 0.02 0.013) ); blocks ( hex ( 0 1 2 3 4 5 6 7) ( 80 80 90) simpleGrading ( 1 1 1) ); edges ( ); patches ( ); mergePatchPairs ( ); Code:
FoamFile { version 2.0; format ascii; class dictionary; location system; object controlDict; } startFrom startTime; startTime 0; stopAt endTime; endTime 1000.0; deltaT 1.0; writeControl timeStep; writeInterval 1000.0; purgeWrite 0; writeFormat ascii; writePrecision 100; writeCompression uncompressed; timeFormat general; timePrecision 6; graphFormat raw; runTimeModifiable yes; Code:
FoamFile { version 2.0; format ascii; class dictionary; location system; object fvSchemes; } divSchemes { } gradSchemes { } laplacianSchemes { } Code:
FoamFile { version 2.0; format ascii; class dictionary; location system; object fvSolution; } solvers { } |
|
May 17, 2016, 10:16 |
|
#2 |
Senior Member
Join Date: Aug 2013
Posts: 407
Rep Power: 16 |
Hi,
Do a checkMesh and see the type of cells that it reports. My guess is that it will still be hexes. The reason you see the weird cells is because Paraview decomposes polyhedral cells automatically and when you take a slice, it triangulates the slice. Uncheck both options and you will see that they are regular hexes only. Cheers, Antimony |
|
May 17, 2016, 11:17 |
|
#3 |
New Member
Join Date: May 2016
Posts: 3
Rep Power: 10 |
Hi,
thanks for your help, Antimony! checkMesh reports only hexes, you are right. If I uncheck decomposing and triangulating, it does look better, but still weird, see attached image. @Mods: It seems, this is more likely a problem with paraView than with blockMesh, could you please move the thread to the right subforum? |
|
May 19, 2016, 09:56 |
|
#4 |
Senior Member
Mikko
Join Date: Jul 2014
Location: The Hague, The Netherlands
Posts: 243
Rep Power: 13 |
Hi,
Try using ExtractCellsByRegion filter when you want to visualize the mesh. If you want slice like behavior check Extract only intersected and Extract intersected marks. If you compare the results from ExtractCellsByRegion filter to the results of Slice filter, maybe you can find a reason for the "strange" behavior. Regards, Mikko |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[snappyHexMesh] SHM Layer Addition Phase | dickcruz | OpenFOAM Meshing & Mesh Conversion | 4 | November 1, 2018 08:05 |
Strange cells in Trimmed mesh | yatsh94 | Siemens | 0 | February 28, 2017 18:40 |
[snappyHexMesh] snappyHexMesh Segmentation Fault | avd28 | OpenFOAM Meshing & Mesh Conversion | 11 | May 11, 2015 21:32 |
[snappyHexMesh] snappyHexMesh aborting | Tobi | OpenFOAM Meshing & Mesh Conversion | 0 | November 10, 2010 04:23 |
7 Node Hexahedral Cells | Mark | FLUENT | 0 | February 5, 2007 12:10 |