|
[Sponsors] |
[ICEM] Checking the mesh -> problem of single edges |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
September 4, 2012, 10:49 |
Checking the mesh -> problem of single edges
|
#1 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
Hi!
(I apologize for the mistakes) I'm a beginner regarding ICEM-CFD. I have to mesh a simple 2D axisymetric geometry. My mesh quality is very good (>0.8). But when I check the mesh, it diagnoses that there are single edge elements. I can delete them, put them in a new subset or ignore. I tried all options but it doesn't work (I can import the mesh in Fluent if I create a new subset, but it doesn't solve my problem) . When I import my mesh under Fluent, it well appears in the window but when I check the mesh there are errors, and of course I can't run calculation. If you have any idea to solve this problem, I'd be really grateful! I have: - built my geometry - created different parts my different bocos - initialized blocking - splited block to match with my geometry - associated edges and curves - associated vertices and points - updated size (update all) - pre-meshed - adjusted edge params - converted to unstruct mesh - created bar (element topology) to match with geomtry and parts - defined my bocos Thanks a lot, Clémence |
|
September 5, 2012, 05:27 |
geometry
|
#3 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
Thank you for your reply.
I join geometry. Here my mesh is coarse, but it isn't the problem and it changes for single edgesnothing when I refine. (INLET_TERTIAIRE = there are 2 inlets of tertiary oxygen at the top of the tube INLET_SECONDAIRE = on the zoom, there is an inlet bottom right INLET_CHARBON = coal inlet in the middle INLET_FUMEES = flue gas inlet) My problem is the presence of single edges when I check the mesh Maybe I missed a step when I built my mesh? Thanks a lot again |
|
September 5, 2012, 05:28 |
|
#4 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
* it changes nothing for single edge when I refine the mesh
|
|
September 5, 2012, 06:23 |
|
#5 |
Senior Member
AB
Join Date: Sep 2009
Location: France
Posts: 323
Rep Power: 22 |
Maybe do a build topology in order to have only red curves.
Then recreate your part with those curves. Maybe you have two curves for the same part. You should not have a GEOM part by the way. Single edge is a possible problem. I don't know how FLUENT works, but maybe the problem doesn't come from the single edges. |
|
September 5, 2012, 08:26 |
|
#6 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
Thank you, I'm going to try this, I'll tell you later if it is the solution!
|
|
September 5, 2012, 10:00 |
|
#7 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
Do you mean that I have to a build diagnostic topology (in "Repair Geometry")? (I tried this, but I don't know which option I had to choose)
|
|
September 5, 2012, 11:05 |
|
#8 |
Senior Member
AB
Join Date: Sep 2009
Location: France
Posts: 323
Rep Power: 22 |
Yes : Geometry -> Repair Geometry -> Build Diagnostic Topology
Tolerance should be something as 10 time your smallest element mesh size. Select the option "join edge curves". |
|
September 5, 2012, 12:46 |
|
#9 |
Member
Yon Han Chong
Join Date: Jun 2012
Posts: 77
Rep Power: 14 |
The geometry look very simple. Can you upload the tin file?
|
|
September 6, 2012, 04:24 |
|
#10 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
Yes it is! But even with an easier geometry I have the problem of the single edges (I tried with a square yesterday...). I don't understand where I'm wrong...
I join the tin file. I tried the build diagnostic geometry as you told me Alexandre, but it deleted my inlet of tertiary oxygen (even if I put a very small tolerance). I'll try again, maybe I have done a mistake. Thanks again! |
|
September 6, 2012, 09:11 |
|
#11 |
Member
Yon Han Chong
Join Date: Jun 2012
Posts: 77
Rep Power: 14 |
I am sorry. I forgot to ask the blk file as well.
|
|
September 6, 2012, 09:30 |
|
#12 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
Here it is!
I fail and fail, and I don't understand why ... Don't hesitate if you have any idea to get me out of there!! Thank you (again ) |
|
September 6, 2012, 09:34 |
|
#13 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
I forgot the corresponding geometry
|
|
September 6, 2012, 09:50 |
|
#14 |
Member
Yon Han Chong
Join Date: Jun 2012
Posts: 77
Rep Power: 14 |
Actually, I found what your problem is.
It is a 2-D mesh you are trying to generate and you don't have surfaces. You use bodies for 3-D mesh. I don't know how you created the geometry but you can go back and output that or if it was generated in ICEM then follow the step to generate a surface.
|
|
September 6, 2012, 09:57 |
|
#15 |
Member
Yon Han Chong
Join Date: Jun 2012
Posts: 77
Rep Power: 14 |
In fact, you might try
Repair Geometry -> Build Diagnostic Topology after creating the surface. I might automatically project the curves to the surface and segment the surface for you. |
|
September 6, 2012, 10:57 |
|
#16 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
Thank you. I tried this with my current geometry but it failed (when I check the mesh in fluent, "check mesh failed").
But I am starting a new project (I restart from the beginning to avoid mistakes) where I am going to follow your procedure, I hope it will work! I'll tell you if it was the solution. Thank you very much anyway! |
|
September 6, 2012, 11:00 |
|
#17 |
Member
Yon Han Chong
Join Date: Jun 2012
Posts: 77
Rep Power: 14 |
Did you deleted the old geometry?
I have built a mesh as I have described using unstructed mesher and read into Fluent without any error. Good luck with your new project though. |
|
September 6, 2012, 11:59 |
|
#18 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
There is no error when I check the mesh if I choose 2D planar in general model in Fluent. If I choose Axisymetric or Axisymetric Swirl, check mesh failed!
Thank you for your help, maybe I can run my simulations with 2D planar, we'll see! |
|
September 21, 2012, 06:48 |
|
#19 |
Senior Member
AB
Join Date: Sep 2009
Location: France
Posts: 323
Rep Power: 22 |
Sorry, I was in holiday and wasn't able to answer to you before.
1) You don't have any surface, so that won't work. Even if it's a 2D scenario, you need surface (as you need volume when it's a 3D scenario). 2) If I understand your project, there are 2 inlets which are ponctual (mean modelled by a point), 2 inlets which are linear (mean modelled by a curve). I guess this is more a solver issue than a mesher issue. How do you model ponctual source with your solver ? Maybe it's by coordinate ? Are you sure you need to specify a point within your mesher ? 3) The geometry is very simple, so your blocking is OK. 4) Try to run your simulation without any symmetry first, and see if it's work. Next, you will have to define which part is symmetry etc ... But, you should first try to acheive a simple simulation. |
|
September 21, 2012, 09:38 |
|
#20 |
New Member
Join Date: Sep 2012
Posts: 13
Rep Power: 14 |
Thank you for your reply.
I finally found my mistake a week ago: my axis was not the X-axis, so it didn't work on Fluent with the Axissymmetric model (because the axis of symetry have to be the X-axis on Fluent, I don't know why). I have always single edges, but I think it's not a problem for Fluent, because the check mesh on Fluent is OK. All my inlets are curves, but some are very very small. Now I have other problems on Fluent, so I have to ask to an other forum ! And thank you to Yonchong, you allowed me to continue my project! |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[CGNS] CGNS converters available | mbeaudoin | OpenFOAM Meshing & Mesh Conversion | 137 | December 14, 2018 05:20 |
[ICEM] Problem making structured mesh on a surface | froztbear | ANSYS Meshing & Geometry | 4 | November 10, 2011 09:52 |
[ICEM] Problem making structural mesh on a surface | froztbear | ANSYS Meshing & Geometry | 1 | November 10, 2011 09:52 |
Problems in compiling paraview in Suse 10.3 platform | chiven | OpenFOAM Installation | 3 | December 1, 2009 08:21 |
unstructured vs. structured grids | Frank Muldoon | Main CFD Forum | 1 | January 5, 1999 11:09 |