|
[Sponsors] |
[mesh manipulation] Converting a 2Dmesh to axisymmetric |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
January 14, 2009, 08:07 |
Go to
http://openfoam-extend.
|
#21 |
Assistant Moderator
Bernhard Gschaider
Join Date: Mar 2009
Posts: 4,225
Rep Power: 51 |
Go to
http://openfoam-extend.svn.sourceforge.net/viewvc/openfoam-extend/trunk/Breeder_ 1.5/utilities/mesh/manipulation/MakeAxialMesh/ and click on "Download GNU Traball" Bernhard
__________________
Note: I don't use "Friend"-feature on this forum out of principle. Ah. And by the way: I'm not on Facebook either. So don't be offended if I don't accept your invitation/friend request |
|
January 15, 2009, 07:14 |
Dear Bernhard,
Thanks for t
|
#22 |
Member
Mahendra
Join Date: Mar 2009
Location: Pune, Maharashtra, India
Posts: 65
Rep Power: 17 |
Dear Bernhard,
Thanks for the makeAxialMesh Utility. I am now able to make a axial mesh in OpenFOAM-1.5. |
|
January 17, 2009, 09:54 |
Dear Bernhard !
One quick q
|
#23 |
Member
Mahendra
Join Date: Mar 2009
Location: Pune, Maharashtra, India
Posts: 65
Rep Power: 17 |
Dear Bernhard !
One quick question, how to specify axis to a grid? |
|
January 19, 2009, 01:50 |
Dear Bernhard,
I am not abl
|
#24 |
Member
Mahendra
Join Date: Mar 2009
Location: Pune, Maharashtra, India
Posts: 65
Rep Power: 17 |
Dear Bernhard,
I am not able to use collapseEdges utility, It is giving me segmentation fault.... Can u shed some light on this? Regards, Mahendra |
|
January 19, 2009, 02:26 |
The error i am getting is like
|
#25 |
Member
Mahendra
Join Date: Mar 2009
Location: Pune, Maharashtra, India
Posts: 65
Rep Power: 17 |
The error i am getting is like this:
Cell:14998 uses faces:6(29649 30248 60346 60347 29349 29648) of which too many are marked for remov al: 29649 30248 60346 60347 29349 29648 Cell:14999 uses faces:6(30249 30299 60348 60349 29350 29649) of which too many are marked for remov al: 30249 30299 60348 60349 29350 29649 Morphing ... --> FOAM Warning : From function boundBox::boundBox(const pointField& points) in file meshes/boundBox/boundBox.C at line 52 Cannot find bounding box for zero sized pointField, returning zero --> FOAM Warning : From function boundBox::boundBox(const pointField& points) in file meshes/boundBox/boundBox.C at line 52 Cannot find bounding box for zero sized pointField, returning zero Collapsing 0 small edges Collapsing 0 in line edges #0 Foam::error::printStack(Foam:stream&) in "/home/lcfd/OpenFOAM/OpenFOAM-1.5/lib/linux64GccDPOp t/libOpenFOAM.so" #1 Foam::sigSegv::sigSegvHandler(int) in "/home/lcfd/OpenFOAM/OpenFOAM-1.5/lib/linux64GccDPOpt/lib OpenFOAM.so" #2 ?? in "/lib64/libc.so.6" #3 collapseHighAspectFaces(Foam::polyMesh const&, Foam::PackedList<1> const&, double, double, Foam ::edgeCollapser&) in "/home/lcfd/OpenFOAM/OpenFOAM-1.5/applications/bin/linux64GccDPOpt/collapseEdg es" #4 main in "/home/lcfd/OpenFOAM/OpenFOAM-1.5/applications/bin/linux64GccDPOpt/collapseEdges " #5 __libc_start_main in "/lib64/libc.so.6" #6 __gxx_personality_v0 in "/home/lcfd/OpenFOAM/OpenFOAM-1.5/applications/bin/linux64GccDPOpt/coll apseEdges" Segmentation fault lcfd@clust5:~/OpenFOAM/lcfd-1.5/Mahendra/Pipe_trial> Regards, Mahendra |
|
January 19, 2009, 12:31 |
Hi Mahendra!
The axis is de
|
#26 |
Assistant Moderator
Bernhard Gschaider
Join Date: Mar 2009
Posts: 4,225
Rep Power: 51 |
Hi Mahendra!
The axis is defined using the axis-option (it assumes that you have a planar patch of that name). With the offset-option you can move the axis away from that line. I'm not sure what happens if the patch is not planar @the collapseEdge-crash: no idea. At first try to run checkMesh on your original (axial) mesh to see whether it is valid. It seems to me that the utilitiy is trying to remove too many edges. Decrease the tolerance parameter (what value are you using now?) Bernhard
__________________
Note: I don't use "Friend"-feature on this forum out of principle. Ah. And by the way: I'm not on Facebook either. So don't be offended if I don't accept your invitation/friend request |
|
January 20, 2009, 02:53 |
Dear Bernhard,
I am using a
|
#27 |
Member
Mahendra
Join Date: Mar 2009
Location: Pune, Maharashtra, India
Posts: 65
Rep Power: 17 |
Dear Bernhard,
I am using a edge length as 3m and angle as 5 degree, what i found was when I used edge length as 0.001m, the collapseEdges utility worked fine saying that 0 edges and 0 faces modified. I assumed that since my pipe length is 3m and wedge angle is 5 degrees I should use them while specifying the arguments to collapseEdges utilty and it resulted in segmentation fault. Was I correct in using it or not??? Regards, Mahendra |
|
January 20, 2009, 13:12 |
Hi Mahendra!
The first para
|
#28 |
Assistant Moderator
Bernhard Gschaider
Join Date: Mar 2009
Posts: 4,225
Rep Power: 51 |
Hi Mahendra!
The first parameter means "all edges shorter that this will be removed". Which would mean almost all edges. Vary the feature angle Bernhard
__________________
Note: I don't use "Friend"-feature on this forum out of principle. Ah. And by the way: I'm not on Facebook either. So don't be offended if I don't accept your invitation/friend request |
|
January 20, 2009, 16:27 |
Hello!
I have just learned
|
#29 |
Senior Member
Sebastian Gatzka
Join Date: Mar 2009
Location: Frankfurt, Germany
Posts: 729
Rep Power: 20 |
Hello!
I have just learned that my meshes created with makeAxialMesh are not 1 cell but 5 cells thick (in the wedge direction). Did I do something wrong? Do you have any idea what may be the problem? Greetings. Sebastian.
__________________
Schrödingers wife: "What did you do to the cat? It's half dead!" |
|
January 21, 2009, 16:33 |
Hi Sebastian!
Have you chec
|
#30 |
Assistant Moderator
Bernhard Gschaider
Join Date: Mar 2009
Posts: 4,225
Rep Power: 51 |
Hi Sebastian!
Have you checked the original (2D planar) mesh? This shouldn't happen unless the original is that thick, too Bernhard
__________________
Note: I don't use "Friend"-feature on this forum out of principle. Ah. And by the way: I'm not on Facebook either. So don't be offended if I don't accept your invitation/friend request |
|
January 21, 2009, 17:10 |
Yes I did.
The mesh is just
|
#31 |
Senior Member
Sebastian Gatzka
Join Date: Mar 2009
Location: Frankfurt, Germany
Posts: 729
Rep Power: 20 |
Yes I did.
The mesh is just fine in both 2D and wedge. Due to the big amount of cells and their small size this was just my mistake in seeing them correctly. Sorry.
__________________
Schrödingers wife: "What did you do to the cat? It's half dead!" |
|
January 21, 2009, 17:34 |
Mahendra: Please have a look a
|
#32 |
Member
Ola Widlund
Join Date: Mar 2009
Location: Sweden
Posts: 87
Rep Power: 17 |
Mahendra: Please have a look at my post of September 9, 2008: http://www.cfd-online.com/OpenFOAM_D.../126/9114.html
I think your problem is in large differences in sizes of cells/faces in your mesh. Try the modified "collapseEdgesBetter" in my post. Use the option -areaFactor to specify an "areafactor" smaller than the default 1e-9. Maybe 1e-11? /Ola |
|
January 21, 2009, 18:07 |
No, that is not the problem.
|
#33 |
Senior Member
Sebastian Gatzka
Join Date: Mar 2009
Location: Frankfurt, Germany
Posts: 729
Rep Power: 20 |
No, that is not the problem.
I was just too blind to see them correctly. But thanks for your advice.!
__________________
Schrödingers wife: "What did you do to the cat? It's half dead!" |
|
January 22, 2009, 00:48 |
Hi Bernhard,
Still I am getti
|
#34 |
Member
Mahendra
Join Date: Mar 2009
Location: Pune, Maharashtra, India
Posts: 65
Rep Power: 17 |
Hi Bernhard,
Still I am getting the segmentation fault error with using collapseEdgesBetter. Cell:49999 uses faces:6(100949 100999 201048 201049 97950 98949) of which too many are marked for removal: 100949 100999 201048 201049 97950 98949 Morphing ... --> FOAM Warning : From function boundBox::boundBox(const pointField& points) in file meshes/boundBox/boundBox.C at line 52 Cannot find bounding box for zero sized pointField, returning zero --> FOAM Warning : From function boundBox::boundBox(const pointField& points) in file meshes/boundBox/boundBox.C at line 52 Cannot find bounding box for zero sized pointField, returning zero Collapsing 0 small edges Collapsing 0 in line edges #0 Foam::error::printStack(Foam:stream&) in "/home/lcfd/OpenFOAM/OpenFOAM-1.5/lib/linux64GccDPOpt/libOpenFOAM.so" #1 Foam::sigSegv::sigSegvHandler(int) in "/home/lcfd/OpenFOAM/OpenFOAM-1.5/lib/linux64GccDPOpt/libOpenFOAM.so" #2 ?? in "/lib64/libc.so.6" #3 collapseHighAspectFaces(Foam::polyMesh const&, Foam::PackedList<1> const&, double, double, Foam::edgeCollapser&) in "/home/lcfd/OpenFOAM/lcfd-1.5/applications/bin/linux64GccDPOpt/collapseEdgesBett er" #4 main in "/home/lcfd/OpenFOAM/lcfd-1.5/applications/bin/linux64GccDPOpt/collapseEdgesBett er" #5 __libc_start_main in "/lib64/libc.so.6" #6 __gxx_personality_v0 in "/home/lcfd/OpenFOAM/lcfd-1.5/applications/bin/linux64GccDPOpt/collapseEdgesBett er" Segmentation fault lcfd@clust5:~/OpenFOAM/lcfd-1.5/Mahendra/Pipe_2D> Regards, Mahendra |
|
January 22, 2009, 11:23 |
Hi Mahendra!
Try skipping t
|
#35 |
Assistant Moderator
Bernhard Gschaider
Join Date: Mar 2009
Posts: 4,225
Rep Power: 51 |
Hi Mahendra!
Try skipping the collapseEdges-Stuff. Simply set all scalars to zeroGradient and the vectors to slip on the axis-patch (I think that are the settings that work) and try to use the mesh that way Bernhard
__________________
Note: I don't use "Friend"-feature on this forum out of principle. Ah. And by the way: I'm not on Facebook either. So don't be offended if I don't accept your invitation/friend request |
|
February 6, 2009, 06:35 |
Hi,
I have created a 3D mes
|
#36 |
New Member
Age Nammensma
Join Date: Mar 2009
Posts: 10
Rep Power: 17 |
Hi,
I have created a 3D mesh with Gmsh, 1 element thick, and used gmshToFoam to convert it. checkMesh gave an "OK" Now I want to use makeAxialMesh to convert my mesh into an axi-symmetrical one. This was the result: Create time Create mesh for time = 0 Plane of the grid: (0 0 1) (0.105253 0.0174027 0.0005) The rotation-axis: ((-0.06 0 0.0005) (0.94 0 0.0005)) Creating wedge with an opening angel of 5 degrees Radius to axis: min = 1.0842e-19 max 0.05 #0 Foam::error::printStack(Foam:stream&) in "/home/age/OpenFoam/OpenFOAM-1.5/lib/linuxGccDPOpt/libOpenFOAM.so" #1 Foam::sigSegv::sigSegvHandler(int) in "/home/age/OpenFOAM/OpenFOAM-1.5/lib/linuxGccDPOpt/libOpenFOAM.so" #2 Uninterpreted: [0xb7ff3400] #3 getDistance(Foam::polyPatch const&, Foam::line<foam::vector<double>, Foam::Vector<double> >&) in "/home/age/OpenFOAM/age-1.5/applications/bin/linuxGccDPOpt/makeAxialMesh" #4 main in "/home/age/OpenFOAM/age-1.5/applications/bin/linuxGccDPOpt/makeAxialMesh" #5 __libc_start_main in "/lib/tls/i686/cmov/libc.so.6" #6 _start in "/home/age/OpenFOAM/age-1.5/applications/bin/linuxGccDPOpt/makeAxialMesh" Segmentation fault What did I do wrong? FYI, I'm not only an OpenFoam newby, but also a Linux newby, so please explain the whole process step by step. |
|
February 6, 2009, 09:30 |
Hello,
i suppose you have f
|
#37 |
Member
Etienne Lorriaux
Join Date: Mar 2009
Location: Compiegne, France
Posts: 45
Rep Power: 17 |
Hello,
i suppose you have followed the wiki instructions and checked the different constraints : http://openfoamwiki.net/index.php/Contrib_MakeAxialMesh As you have already your model in gmsh, you can directly revolve your mesh in gmsh : - rotate -2.5° your plane from your main 2D plane - extrude (revolve) 5° - save mesh - gmshToFoam I've already done that with gmsh, it works fine. You can also have another solution by revolving your mesh in OpenFoam with the extrudeMesh tool, but it's not the easier way to do, i think. Good luck. |
|
February 6, 2009, 09:35 |
Hi Age!
No idea (havn't see
|
#38 |
Assistant Moderator
Bernhard Gschaider
Join Date: Mar 2009
Posts: 4,225
Rep Power: 51 |
Hi Age!
No idea (havn't seen that stacktrace). How big is the mesh? Could you pass the mesh to me and I'll have a look Bernhard
__________________
Note: I don't use "Friend"-feature on this forum out of principle. Ah. And by the way: I'm not on Facebook either. So don't be offended if I don't accept your invitation/friend request |
|
February 6, 2009, 10:49 |
@ Etienne:
So you don't use m
|
#39 |
New Member
Age Nammensma
Join Date: Mar 2009
Posts: 10
Rep Power: 17 |
@ Etienne:
So you don't use makeAxialMesh at all? Just extrude using rotation in Gmsh? @ Bernhard: The mesh-file is very large. I will try some other things first, like simple geo's etc. Thanks guys |
|
February 6, 2009, 11:52 |
It depends on the case, if my
|
#40 |
Member
Etienne Lorriaux
Join Date: Mar 2009
Location: Compiegne, France
Posts: 45
Rep Power: 17 |
It depends on the case, if my geometry is done in gmsh, i revolve the mesh in gmsh. If i already have a mesh in gambit or a blockMeshDict, i use makeAxialMesh.
All those procedures should work, i've tried successfully many different ways to build axy cases in OpenFOAM. If your msh file is too large, perhaps you can zip the .geo, it should be easier to transfer. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Simplest way of converting a 2d Navier Stokes code to a 2d axisymmetric one | mseka | Main CFD Forum | 1 | September 18, 2017 15:53 |
Axisymmetric Boundary condition | Mohit Singh | SU2 | 3 | July 15, 2015 10:19 |
[mesh manipulation] Converting axisymmetric mesh into fully 3D mesh | tomloh | OpenFOAM Meshing & Mesh Conversion | 0 | April 29, 2012 21:31 |
Difference of final temperature between a plane and an axisymmetric geometry | douchka | FLUENT | 0 | July 7, 2011 09:38 |
URGENT ! Need help on Axisymmetric Flow ! | Suman Kumar | Main CFD Forum | 1 | November 20, 2001 15:51 |