CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > OpenFOAM Native Meshers: blockMesh

EdgeGrading changes geomety

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Display Modes
Old   April 8, 2008, 09:13
Default Hi, I have a 2D case with m
  #1
New Member
 
Anita K.
Join Date: Mar 2009
Posts: 25
Rep Power: 9
anita is on a distinguished road
Hi,

I have a 2D case with mesh generation with blockMesh. The solver didn't solve momentum (U) equation. According to this thread http://www.cfd-online.com/OpenFOAM_D...es/1/6884.html
my domain is bent.

I analyzed my mesh and found the problematic block. In this block I use edgeGrading. If I change all ratios to 1 the block is straight. When using combinations of ratios the block gets fatter in the middle.

Parts from blockMeshDict:

convertToMeters 0.001; //mm

vertices
(
//a
(-31.7 9 -0.5) // Vertex a1b = 0
(0.0 9 -0.5) // Vertex a2b = 1
(0.0 15.775 -0.5) // Vertex a3b = 2
(-31.7 15.775 -0.5) // Vertex a4b = 3
(-31.7 9 0.5) // Vertex a1t = 4
(0.0 9 0.5) // Vertex a2t = 5
(0.0 15.775 0.5) // Vertex a3t = 6
(-31.7 15.775 0.5) // Vertex a4t = 7
);

blocks
(

hex (0 1 2 3 4 5 6 7) (50 22 1) edgeGrading (0.25 1 1 0.25 1 4 4 1 1 1 1 1 )//block a
);



CheckMesh: (on z-axis the define points are at
-0.5 mm or 0.5 mm)

Checking geometry...
Domain bounding box: (-0.0317 0.009 -0.000515173) (0 0.015775 0.000515173)
Boundary openness (1.90575e-18 -4.52277e-17 2.57335e-16) OK.
Max cell openness = 2.22755e-16 OK.
Max aspect ratio = 7.16218 OK.
Minumum face area = 4.29135e-08. Maximum face area = 1.16818e-06. Face area magnitudes OK.
Min volume = 4.29163e-11. Max volume = 3.55129e-10. Total volume = 2.17623e-07. Cell volumes OK.
Mesh non-orthogonality Max: 43.2975 average: 31.2745
Non-orthogonality check OK.
Face pyramids OK.
Max skewness = 0.418386 OK.
Min/max edge length = 0.000141259 0.00116818 OK.
All angles in faces OK.
Face flatness (1 = flat, 0 = butterfly) : average = 1 min = 1
All face flatness OK.

Mesh OK.

Is this a bug with edgeGrading?

Anita
anita is offline   Reply With Quote

Old   April 8, 2008, 10:50
Default Sorry, I don't see the problem
  #2
Senior Member
 
Hrvoje Jasak
Join Date: Mar 2009
Location: London, England
Posts: 1,784
Rep Power: 22
hjasak will become famous soon enough
Sorry, I don't see the problem: "the block gets fatter in the middle". Can I see this from the picture above?

I originally wrote all this grading stuff, it is used in the pitzDaily simpleFoam tutorial and it works in 2-D without trouble. Can you investigate further?

Hrv
__________________
Hrvoje Jasak
Providing commercial FOAM/OpenFOAM and CFD Consulting: http://wikki.co.uk
hjasak is offline   Reply With Quote

Old   April 9, 2008, 02:55
Default Hi Hrv, you don't see it from
  #3
New Member
 
Anita K.
Join Date: Mar 2009
Posts: 25
Rep Power: 9
anita is on a distinguished road
Hi Hrv,
you don't see it from the above picture.
I added two pictures looking sideways on the mesh. The z-axis is stretched to see the curved surface better.

You can also see it from the Domain bounding box (-0.0317 0.009 -0.000515173) (0 0.015775 0.000515173) which is larger in z direction as the hex in blockMeshDict.

I also looked in the points file: Only few point at the start and the end of the file have the desired z-position.

Anita

Looking from right side:


Looking from the bottom:

anita is offline   Reply With Quote

Old   April 9, 2008, 05:09
Default Hi, I looked at the utilitis
  #4
New Member
 
Anita K.
Join Date: Mar 2009
Posts: 25
Rep Power: 9
anita is on a distinguished road
Hi,
I looked at the utilitis of OpenFoam and found flattenMesh. This removed the curved surface and now my mesh is real 2D.

Also the solver now solves the U-equation.

By now I can go on with simulating my case, but I still want to now if it was my fault and how I can avoid this problem in future.

Anita
anita is offline   Reply With Quote

Old   December 11, 2008, 07:43
Default Hallo Anita, I have the sam
  #5
Senior Member
 
Wolfgang Heydlauff
Join Date: Mar 2009
Location: Germany
Posts: 136
Rep Power: 10
wolle1982 is on a distinguished road
Hallo Anita,

I have the same problem that by using the edgeGrading the bounding box gets greater. flattenMesh does NOT remove it, poorly.

And further advice?
wolle1982 is offline   Reply With Quote

Old   February 26, 2009, 10:36
Default Hi, I can also confirm this
  #6
New Member
 
Andreas Feymark
Join Date: Mar 2009
Location: Gothenburg, Sweden
Posts: 18
Rep Power: 9
feymark is on a distinguished road
Hi,

I can also confirm this behavior for my 3D-cylinder.

/Andreas
feymark is offline   Reply With Quote

Old   February 27, 2009, 11:38
Default For my case the error comes an
  #7
New Member
 
Andreas Feymark
Join Date: Mar 2009
Location: Gothenburg, Sweden
Posts: 18
Rep Power: 9
feymark is on a distinguished road
For my case the error comes and goes depending on what value I use on the edge.
A value less than 12.5 for instance gives bad behavior and a value greater than 13.2 works well.

Hrv, do you have any clue why this error might occur?

/Andreas
feymark is offline   Reply With Quote

Old   March 2, 2009, 12:16
Default A blockMeshDict like the one b
  #8
New Member
 
Andreas Feymark
Join Date: Mar 2009
Location: Gothenburg, Sweden
Posts: 18
Rep Power: 9
feymark is on a distinguished road
A blockMeshDict like the one below creates this error. The vertices are ordered according to http://www.opencfd.co.uk/openfoam/doc/blockMesh.html .
/*--------------------------------*- C++ -*----------------------------------*\
| ========= | |
| \ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \ / O peration | Version: 1.5 |
| \ / A nd | Web: http://www.OpenFOAM.org |
| \/ M anipulation | |
/*---------------------------------------------------------------------------*/
FoamFile
{
version 2.0;
format ascii;
class dictionary;
object blockMeshDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

convertToMeters 1;
vertices
(
( 0 0 0 ) // vertex number 0
( 1 0 0 ) // vertex number 1
( 1 1 0 ) // vertex number 2
( 0 1 0 ) // vertex number 3
( 0 0 1 ) // vertex number 4
( 1 0 1 ) // vertex number 5
( 1 1 1 ) // vertex number 6
( 0 1 1 ) // vertex number 7
);

blocks
(
hex (0 1 2 3 4 5 6 7) (2 2 2) edgeGrading (4 3 3 4 2 1 1 2 1 1 1 1)
);

edges
(
);

patches
(
);

mergePatchPairs
(
);

// ************************************************** *********************** //

/Andreas
feymark is offline   Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
aerofoil geomety in ICEMCFD using script files said CFX 0 December 21, 2007 15:38


All times are GMT -4. The time now is 15:14.