Dynamic mesh refinement in a 2D or axissymmetric case

christian_kunkelmann · August 22, 2008, 12:34

Hello everyone,

is there any way to suppress dynamic mesh refinement (with interDyMFoam) in a certain direction?

I am running an axissymmetric (y-axis) case with one or two levels of refinement. Unfortunately the solver refines the mesh in the rotational direction, too. This leads to a wedge geometry with a thickness of 2 or 4 cells (between the wedge boundaries). The simulation continues but produces unphysical results like velocity vectors pointing out of the wedge boundary.

Thanks a lot for your help!

Christian

mattijs · August 22, 2008, 14:06

The 1.5 dynamicRefineFvMesh (used in the interDymFoam tutorial) always does 2x2x2 splitting. Writing a directional one would not be trivial.

christian_kunkelmann · August 23, 2008, 09:15

Mattijs,

thank you very much for the quick reply!

Christian

musahossein · September 8, 2008, 23:21

Hi Mattijs:

Where can I find the interDyMFoam tutorial?

http://www.cfd-online.com/OpenFOAM_D...ons/tree_m.gif

Thanks
Musa

marico · January 19, 2009, 06:15

Hi,

1. I also want to use mesh refinement in one direction. The simulation is a (transient) enlargement of a crack, which normally would create elements with bad aspect ratio. Refinement (2x2x2) doesnt seem to help for this.
Is there any other "out-of-box" method like adding layers (removing, if crack is closed)?
(ANSYS hasn't any... ;)

2. For that simulation it would be great to specify the position of the moving boundary by an analytical low of motion (dependent on time and probably XYZ). Whats the proceeding for this?
The fvMotionSolver for this would be "displacementLaplacianFvMotionSolver", I guess?!

Thanks a lot for your answers.

Marco

deepsterblue · January 19, 2009, 10:58

Marco,
Is your case 2D or 3D?

marico · January 19, 2009, 11:35

3D

deepsterblue · January 19, 2009, 12:02

I've done a fair amount of work on triangular/tetrahedral re-meshing. The 3D work is still a little preliminary, but take a look at the video and let me know if this is what you're looking for:

3D Tetrahedral Flipping

marico · January 19, 2009, 12:08

Hi,

wow, looks fine. That seems to be the thing, I'm looking for! Only sounds are missing.. ;)
How does it work? ;)

Thanks,
Marco

deepsterblue · January 19, 2009, 13:15

Derived mesh class which performs topological changes based on input criteria. It also uses the mesh-motion solver to minimize the number of topological changes for each time-step.
Doesn't do field mapping yet, though. I'm working on that.

August 22, 2008, 12:34	Hello everyone, is there an	#1
christian_kunkelmann New Member Christian Kunkelmann Join Date: Mar 2009 Posts: 8 Rep Power: 17	Hello everyone, is there any way to suppress dynamic mesh refinement (with interDyMFoam) in a certain direction? I am running an axissymmetric (y-axis) case with one or two levels of refinement. Unfortunately the solver refines the mesh in the rotational direction, too. This leads to a wedge geometry with a thickness of 2 or 4 cells (between the wedge boundaries). The simulation continues but produces unphysical results like velocity vectors pointing out of the wedge boundary. Thanks a lot for your help! Christian

August 22, 2008, 14:06	The 1.5 dynamicRefineFvMesh (u	#2
mattijs Senior Member Mattijs Janssens Join Date: Mar 2009 Posts: 1,419 Rep Power: 26	The 1.5 dynamicRefineFvMesh (used in the interDymFoam tutorial) always does 2x2x2 splitting. Writing a directional one would not be trivial.

August 23, 2008, 09:15	Mattijs, thank you very muc	#3
christian_kunkelmann New Member Christian Kunkelmann Join Date: Mar 2009 Posts: 8 Rep Power: 17	Mattijs, thank you very much for the quick reply! Christian

September 8, 2008, 23:21	Hi Mattijs: Where can I fin	#4
musahossein Senior Member musaddeque hossein Join Date: Mar 2009 Posts: 309 Rep Power: 18	Hi Mattijs: Where can I find the interDyMFoam tutorial? http://www.cfd-online.com/OpenFOAM_D...ons/tree_m.gif Thanks Musa

January 19, 2009, 06:15	Hi, 1. I also want to use m	#5
marico Member Marco Müller Join Date: Mar 2009 Location: Germany Posts: 94 Rep Power: 17	Hi, 1. I also want to use mesh refinement in one direction. The simulation is a (transient) enlargement of a crack, which normally would create elements with bad aspect ratio. Refinement (2x2x2) doesnt seem to help for this. Is there any other "out-of-box" method like adding layers (removing, if crack is closed)? (ANSYS hasn't any... ;) 2. For that simulation it would be great to specify the position of the moving boundary by an analytical low of motion (dependent on time and probably XYZ). Whats the proceeding for this? The fvMotionSolver for this would be "displacementLaplacianFvMotionSolver", I guess?! Thanks a lot for your answers. Marco

January 19, 2009, 10:58	Marco, Is your case 2D or 3	#6
deepsterblue Senior Member Sandeep Menon Join Date: Mar 2009 Location: Amherst, MA Posts: 403 Rep Power: 25	Marco, Is your case 2D or 3D? __________________ Sandeep Menon University of Massachusetts Amherst https://github.com/smenon

January 19, 2009, 11:35	3D	#7
marico Member Marco Müller Join Date: Mar 2009 Location: Germany Posts: 94 Rep Power: 17	3D

January 19, 2009, 12:02	I've done a fair amount of wor	#8
deepsterblue Senior Member Sandeep Menon Join Date: Mar 2009 Location: Amherst, MA Posts: 403 Rep Power: 25	I've done a fair amount of work on triangular/tetrahedral re-meshing. The 3D work is still a little preliminary, but take a look at the video and let me know if this is what you're looking for: 3D Tetrahedral Flipping __________________ Sandeep Menon University of Massachusetts Amherst https://github.com/smenon

January 19, 2009, 12:08	Hi, wow, looks fine. That s	#9
marico Member Marco Müller Join Date: Mar 2009 Location: Germany Posts: 94 Rep Power: 17	Hi, wow, looks fine. That seems to be the thing, I'm looking for! Only sounds are missing.. ;) How does it work? ;) Thanks, Marco

January 19, 2009, 13:15	Derived mesh class which perfo	#10
deepsterblue Senior Member Sandeep Menon Join Date: Mar 2009 Location: Amherst, MA Posts: 403 Rep Power: 25	Derived mesh class which performs topological changes based on input criteria. It also uses the mesh-motion solver to minimize the number of topological changes for each time-step. Doesn't do field mapping yet, though. I'm working on that. __________________ Sandeep Menon University of Massachusetts Amherst https://github.com/smenon

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
InterDyMFoam dynamic refinement	ala	OpenFOAM Running, Solving & CFD	12	September 28, 2016 19:51
Choise of ddtSchemes regarding Dynamic Mesh Refinement TaylorTayleighInstability	sega	OpenFOAM Running, Solving & CFD	0	March 9, 2009 10:34
A dynamic case for stressedFoam	fw407	OpenFOAM Running, Solving & CFD	2	July 27, 2008 08:00
Variable inletoutlet for dynamic mesh mixer2D case	soeren	OpenFOAM Running, Solving & CFD	0	May 11, 2008 18:22
Dynamic Mesh in an In-cylinder case	venkitaraman	FLUENT	0	March 26, 2007 18:15