CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > OpenFOAM Meshing & Mesh Conversion

"did not succesfully snap mesh. Giving up"

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

Like Tree6Likes
  • 2 Post By chegdan
  • 1 Post By chegdan
  • 1 Post By chegdan
  • 1 Post By chegdan
  • 1 Post By chegdan

Reply
 
LinkBack Thread Tools Display Modes
Old   July 8, 2013, 13:17
Default "did not succesfully snap mesh. Giving up"
  #1
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
Hi Dear All;
i use snappyHexMesh for meshing the domain around two element airfoil.
during the running of snappyHexMesh, this message is appear that ""did not succesfully snap mesh. Giving up"" . is it related to the parameters in snapControls section of snappyHexMeshDict?
if yes, what changes should i do in my snappyHexMeshDict that i have attached in the following?
thank you
Attached Files
File Type: gz snappyHexMeshDict.tar.gz (3.6 KB, 4 views)
File Type: txt snappyHexMeshlogfile.txt (32.3 KB, 2 views)
s.m is offline   Reply With Quote

Old   July 8, 2013, 15:00
Default
  #2
Senior Member
 
chegdan's Avatar
 
Daniel P. Combest
Join Date: Mar 2009
Location: St. Louis, USA
Posts: 543
Rep Power: 18
chegdan will become famous soon enough
I would make your base mesh smaller (i.e. change your blockMeshDict to have more subdivisions), decrease your levels on your edges from 6 7 to 5 6 or lower depending on your base mesh sizing and change

Code:
    tolerance 1.0;
    nSolveIter 300;
in snapControls. A lot of errors with SHM are resolved by changing your base mesh sizing.
s.m and MtnRunBeachBum like this.
__________________
Dan

Find me on twitter @dancombest and LinkedIn

Last edited by chegdan; July 8, 2013 at 15:55. Reason: i noticed you were actually using 2.2.x
chegdan is offline   Reply With Quote

Old   July 9, 2013, 02:28
Default
  #3
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
Quote:
Originally Posted by chegdan View Post
I would make your base mesh smaller (i.e. change your blockMeshDict to have more subdivisions), decrease your levels on your edges from 6 7 to 5 6 or lower depending on your base mesh sizing and change

Code:
    tolerance 1.0;
    nSolveIter 300;
in snapControls. A lot of errors with SHM are resolved by changing your base mesh sizing.
Hi Dear chegdan,
i am using the OF 2.2.0. this is my blockMeshDict, would you please look at it and tell me is it necessary to increase the subdivisions?
Attached Files
File Type: txt blockMeshDict.txt (1.6 KB, 10 views)
s.m is offline   Reply With Quote

Old   July 9, 2013, 10:39
Default
  #4
Senior Member
 
chegdan's Avatar
 
Daniel P. Combest
Join Date: Mar 2009
Location: St. Louis, USA
Posts: 543
Rep Power: 18
chegdan will become famous soon enough
The closer you make your base mesh cell aspect ratio to unity (equal length in x y and z direction), the better results you will see in SHM. Try:

Code:
blocks
(
    hex (0 1 2 3 4 5 6 7) (880 480 1) simpleGrading (1 1 1)
);
and maybe change your z-thickness to 0.025 to match the aspect ratio of the x and y-directions. With your current settings of levels 6 7 you have cells that are on the size of 0.025/(2^6) = 3.9E-4 [m] to 0.025/(2^7) = 1.95E-4 [m].
s.m likes this.
__________________
Dan

Find me on twitter @dancombest and LinkedIn
chegdan is offline   Reply With Quote

Old   July 9, 2013, 11:47
Default
  #5
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
Quote:
Originally Posted by chegdan View Post
The closer you make your base mesh cell aspect ratio to unity (equal length in x y and z direction), the better results you will see in SHM. Try:

Code:
blocks
(
    hex (0 1 2 3 4 5 6 7) (880 480 1) simpleGrading (1 1 1)
);
and maybe change your z-thickness to 0.025 to match the aspect ratio of the x and y-directions. With your current settings of levels 6 7 you have cells that are on the size of 0.025/(2^6) = 3.9E-4 [m] to 0.025/(2^7) = 1.95E-4 [m].
thank you,is my cells size good or i should increase the level of refinementSurfaces?
s.m is offline   Reply With Quote

Old   July 9, 2013, 11:50
Default
  #6
Senior Member
 
chegdan's Avatar
 
Daniel P. Combest
Join Date: Mar 2009
Location: St. Louis, USA
Posts: 543
Rep Power: 18
chegdan will become famous soon enough
Quote:
...is my cells size good or i should increase the level of refinementSurfaces?
This ultimately depends on your y+ and turbulence model you are using so I can't comment on it. To snap to the surface, it should be fine.
__________________
Dan

Find me on twitter @dancombest and LinkedIn
chegdan is offline   Reply With Quote

Old   July 9, 2013, 12:07
Default
  #7
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
Quote:
Originally Posted by chegdan View Post
This ultimately depends on your y+ and turbulence model you are using so I can't comment on it. To snap to the surface, it should be fine.
i have both high-Re and low-Re number turbulent model, "k-wSST & lunderSharmaKE" would you please guide me?
is this cell size belong to all the cells around my airfoil? i ask this question, becuase for computing the y+ i only need to know the first cell size of my mesh.
thank you again for your advice
s.m is offline   Reply With Quote

Old   July 9, 2013, 12:14
Default
  #8
Senior Member
 
chegdan's Avatar
 
Daniel P. Combest
Join Date: Mar 2009
Location: St. Louis, USA
Posts: 543
Rep Power: 18
chegdan will become famous soon enough
From my experience

lunderSharmaKE needs y+ <= 1

but it really can't be calculated until your simulation has been run to know how much further you need to refine your mesh. You can use the yPlusRAS utility to calculate the y+ value.
s.m likes this.
__________________
Dan

Find me on twitter @dancombest and LinkedIn
chegdan is offline   Reply With Quote

Old   July 9, 2013, 12:37
Default
  #9
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
Quote:
Originally Posted by chegdan View Post
From my experience

lunderSharmaKE needs y+ <= 1

but it really can't be calculated until your simulation has been run to know how much further you need to refine your mesh. You can use the yPlusRAS utility to calculate the y+ value.
yes you are right. for reducing my y+ to be appropriate for low-Re number turbulent models, i should increase the subdivision in blockMeshDict, is it right?
is it better to use level (6 6) for the refinementSurface, to have squre cells size?
s.m is offline   Reply With Quote

Old   July 9, 2013, 13:02
Default
  #10
Senior Member
 
chegdan's Avatar
 
Daniel P. Combest
Join Date: Mar 2009
Location: St. Louis, USA
Posts: 543
Rep Power: 18
chegdan will become famous soon enough
the (6 6) would mean that if a featureEdge is encountered then the same refinement (a level 6) will be used. If you had done a (6 7) then a further refinement of 7 would be done around the "turn" of the edge.
s.m likes this.
__________________
Dan

Find me on twitter @dancombest and LinkedIn
chegdan is offline   Reply With Quote

Old   July 9, 2013, 13:20
Default
  #11
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
Quote:
Originally Posted by chegdan View Post
the (6 6) would mean that if a featureEdge is encountered then the same refinement (a level 6) will be used. If you had done a (6 7) then a further refinement of 7 would be done around the "turn" of the edge.
Thank you i am running the case, with new blockMeshDict and also new snappyHexMeshDict. i'll put the resulted logFile after fininshing.
s.m is offline   Reply With Quote

Old   July 10, 2013, 08:08
Default
  #12
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
Hi,
i didn't recive that message anymore, thank you very much for your kind help.
i put some picture of my mesh, would you please tell me is this mesh looks good?
and i also attach the snappyHexMesh and checkMesh logfile.

as you see in checkMesh, "Mesh non-orthogonality Max: 59.5431 average: 7.46241" is it fine?
thank you very much again
Attached Images
File Type: jpg main-corner.jpg (48.6 KB, 11 views)
File Type: jpg flap-corner.jpg (47.6 KB, 10 views)
File Type: jpg slat-corner.jpg (42.0 KB, 10 views)
Attached Files
File Type: txt checkMesh.txt (6.2 KB, 1 views)
File Type: txt snappHexMesh.txt (27.5 KB, 3 views)
s.m is offline   Reply With Quote

Old   July 10, 2013, 08:15
Default
  #13
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
and these are, another picture of this snappyHexMesh, as you see ot just doesn't capture the main trailling edge corner angle true,if you have a comment to fix it, please tell me
thanks.
Attached Images
File Type: jpg main-leading-edge.jpg (37.4 KB, 17 views)
File Type: jpg flap-leading-edge.jpg (55.4 KB, 11 views)
File Type: jpg main.jpg (34.6 KB, 11 views)
File Type: jpg slat.jpg (44.6 KB, 16 views)
File Type: jpg 3.jpg (90.6 KB, 7 views)
s.m is offline   Reply With Quote

Old   July 10, 2013, 12:38
Default
  #14
Senior Member
 
chegdan's Avatar
 
Daniel P. Combest
Join Date: Mar 2009
Location: St. Louis, USA
Posts: 543
Rep Power: 18
chegdan will become famous soon enough
Quote:
as you see in checkMesh, "Mesh non-orthogonality Max: 59.5431 average: 7.46241" is it fine?
This is fine, anything above a 70 would be a concern.

also...if you are doing a 2D case, you will be better off doing a 2D mesh. have a look at

Code:
tutorials/incompressible/pimpleDyMFoam/wingMotion/wingMotion_snappyHexMesh
for some inspiration on how to generate a 2D mesh.

For resolving the edge, you could try drawing a refinement box around the edge and then make the cells there really small. Lastly, to make your mesh look better in screenshots (i.e. get rid of diagonal cuts in hex cells) you can check "use VTKPolyhedron" in Paraview object inspector when you first import results into Paraview.
__________________
Dan

Find me on twitter @dancombest and LinkedIn
chegdan is offline   Reply With Quote

Old   July 10, 2013, 13:17
Default
  #15
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
Quote:
Originally Posted by chegdan View Post
This is fine, anything above a 70 would be a concern.

also...if you are doing a 2D case, you will be better off doing a 2D mesh. have a look at

Code:
tutorials/incompressible/pimpleDyMFoam/wingMotion/wingMotion_snappyHexMesh
for some inspiration on how to generate a 2D mesh.

For resolving the edge, you could try drawing a refinement box around the edge and then make the cells there really small. Lastly, to make your mesh look better in screenshots (i.e. get rid of diagonal cuts in hex cells) you can check "use VTKPolyhedron" in Paraview object inspector when you first import results into Paraview.
yes, i looked at it some times ago, i didn't check "use VTKPolyhedron" anyTime, but i examine it now, it was really nice. thank you very much
i have another problem yet
as you see, the mesh around the airfoil is extremely tiny but, i get the pressure countour around it really bad.
i write my problem in the solving forum of openFoam.
Home > Forums > OpenFOAM Running, Solving & CFD > wrong contour of pressure at the airfoil leading edge
would you please take a look at it,thank you.
Attached Images
File Type: jpg main-leading-edge.jpg (18.4 KB, 10 views)
File Type: jpg flap-leading-edge.jpg (15.0 KB, 8 views)
File Type: jpg p.jpg (15.8 KB, 12 views)
s.m is offline   Reply With Quote

Old   July 10, 2013, 13:33
Default
  #16
Senior Member
 
chegdan's Avatar
 
Daniel P. Combest
Join Date: Mar 2009
Location: St. Louis, USA
Posts: 543
Rep Power: 18
chegdan will become famous soon enough
I will leave that thread to the experts. However, you may want to add some boundary layers cells to your mesh and/or increase the refinement near the surface. If you give as much info as possible on that thread and also search the forum....i bet you will get it solved. Good luck!
s.m likes this.
__________________
Dan

Find me on twitter @dancombest and LinkedIn

Last edited by chegdan; July 14, 2013 at 13:58.
chegdan is offline   Reply With Quote

Old   July 10, 2013, 13:47
Default
  #17
s.m
Senior Member
 
saeideh mohamadi
Join Date: Aug 2012
Posts: 229
Rep Power: 5
s.m is on a distinguished road
Quote:
Originally Posted by chegdan View Post
I will leave that thread to the experts. However, you may want to add some boundary layers cells to your mesh and/or increase the refinement near the surface. I'm not an expert at airfoils so if you give as much info as possible on that thread and also search the forum....i bet you will get it solved. Good luck!
ok, thank you
s.m 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
snapEdge - failure Tobi OpenFOAM Native Meshers: snappyHexMesh and Others 33 March 18, 2014 04:58
Star CCM Overset Mesh Error (Rotating Turbine) thezack CD-adapco 3 December 11, 2013 04:09
[ICEM] surface mesh merging problem everest ANSYS Meshing & Geometry 39 June 5, 2013 19:02
Layers:problem with curvature giulio.topazio OpenFOAM Native Meshers: snappyHexMesh and Others 10 August 22, 2012 09:03
[ICEM] Problem making structural mesh on a surface froztbear ANSYS Meshing & Geometry 1 November 10, 2011 09:52


All times are GMT -4. The time now is 07:10.