|
[Sponsors] |
October 30, 2019, 03:22 |
Clarification about Mesh orthogonality
|
#1 |
Member
Join Date: Sep 2019
Posts: 33
Rep Power: 6 |
Hello everybody,
I have a question concerning the Mesh Orthogonality. The situation is as follows: - I have a pipe where I have a sudden change in diameter. At one position I have an increase in diameter (from small to larger diameter) and at another position the opposite (from large to smaller diameter). - At both locations I get cells with bad orthogonal quality. In the picture the location is displayed where the change from the large to smaller diameter happens. The definition of Mesh orthogonality is clear and makes sense. My question is: Is it even possible to have cells with good orthogonal quality in regions like that? If yes how do you obtain them? Thanks a lot for your answer and have a nice day! |
|
October 30, 2019, 09:01 |
|
#2 |
Senior Member
Erik
Join Date: Feb 2011
Location: Earth (Land portion)
Posts: 1,166
Rep Power: 23 |
This should probably be in the meshing forum. But anyways...
That looks like a step in diameter? I would be using hex elements and mesh that larger volume like a hollow cylinder. A full structured mesh would have perfect orthogonality. I would slice it up into multiple cylinders of the same diameter, end to end, split where the change in diameter occurs. Then a hollow cylinder around the cylinder that should be larger. Use a structured mesh on all 4 bodies. |
|
October 30, 2019, 17:33 |
|
#3 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,665
Rep Power: 143 |
One approach, as Erik states, is to use a hex mesh in that region.
If you want to use a tet mesh over everything then you will probably need to make the element size smaller at the step to improve mesh quality. You only have 1 or 2 elements defining the step, if you made that 4 or 5 the quality will improve quite a bit, especially if you are using inflation layers.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum. |
|
October 31, 2019, 05:21 |
|
#4 |
Member
Join Date: Sep 2019
Posts: 33
Rep Power: 6 |
First of all, thanks a lot Erik and Mr. Horrocks for your answers! (I know that should be in the meshing forum, but already posted it there and didn't get an answer)
@Erik: I'm not quit sure if I get your idea with the hollow cylinder. For clarification I made the following sketch (see attachement), where: - black lines --> geometry - red lines --> splitting lines - green lines --> also splitting lines, with which the hollow cylinder is achieved Is that what you meant? @Mr. Horrocks: I'll definitely give it a try and compare it to the structured option. |
|
October 31, 2019, 08:59 |
|
#5 |
Senior Member
Erik
Join Date: Feb 2011
Location: Earth (Land portion)
Posts: 1,166
Rep Power: 23 |
Yes, that is what I meant.
I always slice up my geometry into regions where I can control the mesh easily, and get a good quality mesh. If you have a complex geometry, it is very difficult to control the mesh, so you slice it up into simple, easy to mesh shapes. |
|
October 31, 2019, 10:08 |
|
#6 |
Member
Join Date: Sep 2019
Posts: 33
Rep Power: 6 |
Thank you one more time and your picture made it even clearer.
Have a nice day! Patrick |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[snappyHexMesh] SnappyHexMesh/splitMeshRegion : region1 in zone "-1" | GuiMagyar | OpenFOAM Meshing & Mesh Conversion | 3 | August 4, 2023 13:38 |
[snappyHexMesh] snappyHexMesh does not create any mesh except one for the reference cell | Arman_N | OpenFOAM Meshing & Mesh Conversion | 1 | May 20, 2019 18:16 |
[snappyHexMesh] Snappyhex mesh: poor inlet mesh | Swagga5aur | OpenFOAM Meshing & Mesh Conversion | 1 | December 3, 2016 17:59 |
[ICEM] Improving mesh orthogonality in ICEM or Fluent | Harshal | ANSYS Meshing & Geometry | 0 | December 2, 2013 10:04 |
[ICEM] Problem making structural mesh on a surface | froztbear | ANSYS Meshing & Geometry | 1 | November 10, 2011 09:52 |