CFD Online Discussion Forums - Problem : Cell expansion ratio (blockMeshDict)

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)

- OpenFOAM Pre-Processing (https://www.cfd-online.com/Forums/openfoam-pre-processing/)

- - Problem : Cell expansion ratio (blockMeshDict) (https://www.cfd-online.com/Forums/openfoam-pre-processing/104784-problem-cell-expansion-ratio-blockmeshdict.html)

Problem : Cell expansion ratio (blockMeshDict)

Hi, guys.

I try to make my mesh for simulating 2D flow over the airfoil using blockMesh, so I compiled my blockMeshDict file. But when I check in paraView, I got a problem that the mesh is twisted.

First, I inputted small expansion ratio about 20 in normal direction to the airfoil surface, and there is no problem.

And then, I increased the expansion ratio (about 100). blockMesh was good working and no error existed. However, in paraView, The mesh had big trouble.

I attach a part of my blockMeshDict (related to definition of block) and picture about bad mesh.

How can I resolve it?

PHP Code:


		
			
blocks
(
    hex (11 10 0 1 27 26 16 17) (             60            256              1) simpleGrading (1 1000 1)
    hex (12 11 1 2 28 27 17 18) (             32            256              1) simpleGrading (1 1000 1)
    hex (13 12 2 3 29 28 18 19) (             36            256              1) simpleGrading (1 1000 1)
    hex (5 13 3 4 21 29 19 20) (            256            256              1) simpleGrading (0.001 1000 1)
    hex (10 15 9 0 26 31 25 16) (             60            256              1) simpleGrading (1 1000 1)
    hex (15 14 8 9 31 30 24 25) (             32            256              1) simpleGrading (1 1000 1)
    hex (14 13 7 8 30 29 23 24) (             36            256              1) simpleGrading (1 1000 1)
    hex (13 5 6 7 29 21 22 23) (            256            256              1) simpleGrading (1000 1000 1)
);

What is the output of checkMesh?

The output of checkMesh is

PHP Code:


		
			
/*---------------------------------------------------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.1.0                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
Build  : 2.1.0-0bc225064152
Exec   : checkMesh
Date   : Jul 16 2012
Time   : 22:26:18
Host   : "wtcfd"
PID    : 12055
Case   : /home/wtcfd/MSthesis/MATLABcodetest
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time
Create polyMesh for time = 0
Time = 0
Mesh stats
    points:           394752
    internal points:  0
    faces:            787200
    internal faces:   392448
    cells:            196608
    boundary patches: 4
    point zones:      0
    face zones:       0
    cell zones:       0
Overall number of cells of each type:
    hexahedra:     196608
    prisms:        0
    wedges:        0
    pyramids:      0
    tet wedges:    0
    tetrahedra:    0
    polyhedra:     0
Checking topology...
    Boundary definition OK.
    Cell to face addressing OK.
    Point usage OK.
    Upper triangular ordering OK.
    Face vertices OK.
    Number of regions: 1 (OK).
Checking patch topology for multiply connected surfaces ...
    Patch               Faces    Points   Surface topology                  
    inlet               768      1538     ok (non-closed singly connected)  
    outlet              512      1026     ok (non-closed singly connected)  
    airfoil             256      512      ok (non-closed singly connected)  
    frontAndBack        393216   394752   ok (non-closed singly connected)  
Checking geometry...
    Overall domain bounding box (-30 -30 0) (30 30 0.2)
    Mesh (non-empty, non-wedge) directions (1 1 0)
    Mesh (non-empty) directions (1 1 0)
    All edges aligned with or perpendicular to non-empty directions.
    Boundary openness (-2.14328e-19 -9.24849e-20 2.84901e-16) OK.
 ***High aspect ratio cells found, Max aspect ratio: 1000, number of cells 4
  <<Writing 4 cells with high aspect ratio to set highAspectRatioCells
    Minumum face area = 9.46298e-08. Maximum face area = 0.644086.  Face area magnitudes OK.
    Min volume = 2e-300. Max volume = 0.128817.  Total volume = 643.668.  Cell volumes OK.
    Mesh non-orthogonality Max: 179.984 average: 20.1941
   *Number of severely non-orthogonal faces: 914.
 ***Number of non-orthogonality errors: 2500.
  <<Writing 3414 non-orthogonal faces to set nonOrthoFaces
 ***Error in face pyramids: 8016 faces are incorrectly oriented.
  <<Writing 5730 faces with incorrect orientation to set wrongOrientedFaces
 ***Max skewness = 23.7527, 71 highly skew faces detected which may impair the quality of the results
  <<Writing 71 skew faces to set skewFaces
    Coupled point location match (average 0) OK.
Failed 4 mesh checks.

Only thing I get think of is the writePrecision in controlDict, maybe you want to set that to a larger value.

Thank you for your advice :)
Unfortunately, I can not correct my problem with your suggestion.
I increased writePrecision till 20, but results of the mesh that I can see in paraView are all the same.

Do you need more information to resolve this trouble?

Than you should share everything one needs to reproduce the problem, so that we can try ourselves.

I can not understand this problem.
SAME blockMeshDict file, but Different expension ratio.
For the small expansion ratio, there is no matter.
For the bigger expansion ratio, there is bad results.

First, I'll try more things by myself.
If I do not find a solution, I'll upload blockMeshDict.

Once again, I heartily thank you for your advice :)

Bernhard, I decided to make a blockMeshDict file again with different structure for blocks.

.Bernhard, I decided to make a blockMeshDict file again with different structure for blocks.

Can you please attach your whole blockMeshDict, so that we can have a look of that.
It can be problem with your blockMeshDict.

Siva

I resolve my problem.
Output of my blockMeshDict shows some errors about the order of definition of block. But the mesh has no errors in paraView.
So I thought that is not critical and continued my calculation.
When I use simpleFoam, the calculation is perfect.
Thank you for your concern.:)