|
[Sponsors] |
two different motion solver for dynamicMeshDict |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
December 1, 2015, 08:40 |
two different motion solver for dynamicMeshDict
|
#1 |
New Member
Join Date: Oct 2014
Posts: 24
Rep Power: 12 |
Dear foamers,
I hope you can help me about an intersting mesh motion in openfoam. Simply I would want to use two different motion solver to move mesh, for example: motion1: dynamicFvMeshLibs 1("libtopoChangerFvMesh.so"); dynamicFvMesh movingTopoFvMesh; and motion2: dynamicFvMesh dynamicMotionSolverFvMesh; motionSolverLibs ( "libfvMotionSolvers.so" ); solver displacementLaplacian; i would like to put these two into dynamicMeshDict together. Can you help me to edit the file? Thankyou in advance. |
|
December 1, 2015, 16:50 |
|
#2 |
Retired Super Moderator
Bruno Santos
Join Date: Mar 2009
Location: Lisbon, Portugal
Posts: 10,981
Blog Entries: 45
Rep Power: 128 |
Quick answer: As far as I know, this is not yet possible up to OpenFOAM 3.0, namely to have two different types of dynamic mesh operations at the same time.
__________________
|
|
December 2, 2015, 03:51 |
|
#3 |
New Member
Join Date: Oct 2014
Posts: 24
Rep Power: 12 |
Thanks for your reply Bruno,
I have a couple of question for you: first: how can you know that of3.0 has this possibility. second: in this case how can i find the instructions to edit this dynamicMeshDict (at the moment I only saw the dynamicMeshDict in the tutorials cases but I didn't find anything of similar). All the best |
|
December 2, 2015, 10:32 |
|
#4 |
Super Moderator
Tobias Holzmann
Join Date: Oct 2010
Location: Bad Wörishofen
Posts: 2,711
Blog Entries: 6
Rep Power: 52 |
Dear Floyed,
as Bruno said, up to version 3.0 there is no motion library that couples the mesh motion (normally moving the boundary and solving the laplace equation for the cell centers for smoothing the internal mesh) and the topoChanger (which actually change the topologiy of your mesh by removing or adding cells). If you want to do sth. like that you have to couple the mesh motion solver and the topoChanger libs. I also combined these two libs but still I did not get a working lib. Compiling is fine but after starting I get an segmentation fault error. Up to now I have not too much time to investigate into this but it is possbile because a guy from Linz did it already and presented it in PFAU9 last year in Linz.
__________________
Keep foaming, Tobias Holzmann |
|
December 2, 2015, 11:34 |
|
#5 |
New Member
Join Date: Oct 2014
Posts: 24
Rep Power: 12 |
Dear Tobi,
thankyou for your reply. I understood that for the coupling i have to modify the source code, that is put in the same source file both codes (motion and topological changes), is it right? it would be great if you explain me the first steps to do for the coupling. Best regards. |
|
December 2, 2015, 12:33 |
|
#6 | |
Super Moderator
Tobias Holzmann
Join Date: Oct 2010
Location: Bad Wörishofen
Posts: 2,711
Blog Entries: 6
Rep Power: 52 |
Quote:
Hi, not really. You have to write your lib in a way that you first call the motion lib which will actually move the mesh due to boundary conditions. Then you have to write a code that will check the topology of the mesh (like in movingConeTopoFvMesh) which also include your conditions when you will remove cells and so on and then you have to call the topoChanger lib with the necessary information. This will at least change the topology. Kind regards, Tobi
__________________
Keep foaming, Tobias Holzmann |
||
December 23, 2015, 02:54 |
|
#7 |
Member
Cem Albukrek
Join Date: Mar 2009
Posts: 52
Rep Power: 17 |
Hello Dr Floyd,
I have been struggling with this problem myself. Depending on your goal, I have a somewhat quick and certainly dirty workaround: If your goal is to apply the two mesh modifications sequentially, but not in a coupled manner and if the first motion has to happen only once or less often (in my case, I wanted to move an airplane wing flap down and then move forward with a 6 DOF body motion solver for the whole system), here are the steps: 1. Run case with mesh motion 1 alone. 2. Export the dynamic mesh output with the point/cell displacement files into a new "fixed"/basic mesh 3. Edit your dynamic mesh dictionary and point displacement files to reflect your motion 2 goals 4. Run the case with mesh motion 2 alone. As you see, the 2 motions will be run sequentially, not together... Now here comes the dirty part which is in step 2: I could not find 1 single OpenFOAM command to export a dynamic mesh to fixed mesh (no point/cell displacement files). So I converted the dynamic mesh to Fluent format (foamMeshToFluent) and then back to OpenFoam format (fluent3DMeshToFoam). It all worked in the end, but I would love to hear of a hidden command for the dynamic to fixed mesh conversion within native OpenFoam formats... Regards, Cem |
|
June 21, 2017, 12:22 |
two motion solvers...
|
#8 |
New Member
Join Date: Dec 2012
Posts: 3
Rep Power: 13 |
Hi Foamers,
I am reading this thread because I have a similar issue - trying to simulate a two-stroke engine, and need to include cell layer removal/addition in the cylinder as well as constant rotation of the volume around the crankshaft. I hope somebody can point me to the answers for my many questions below please: 1. Where and who could provide detailed insights into this modeling technique? 2. Is it possible to combine a movingConeTopo and a constantVelocity motion in one model, for the cylinder and crankshaft mentioned above? 3. Would it be worth starting to experiment with movingCone (without cell removal/addition) for the cylinder combined with constantVelocity for crankshaft rotation? I am using OpenFoam 4.x and the rhoPimpleDyMFoam solver. I have already spent a few months trying to figure out the model setup process. In the meantime, I have setup the whole model in Fluent in half a day... Kind of frustrating. Many thanks, nimrod |
|
July 9, 2018, 12:05 |
Two types of dynamic mesh at same time.
|
#9 |
New Member
Weidong Dai
Join Date: Oct 2016
Location: Edinburgh, Scotland, UK
Posts: 9
Rep Power: 10 |
Hi, wyldckat, I met a similar problem, I just wonder how OpenFOAM 3.0 achieve this. Is there a tutorial for that? My problem is a turbine blade rotating at a constant angular velocity, but at the same time, it will passively pitch. I was thinking using sliding mesh for the rotation and 6 DOF for the passive pitch, can those two solvers apply at the same time? Thank you.
|
|
July 10, 2018, 07:47 |
|
#10 | |
New Member
Weidong Dai
Join Date: Oct 2016
Location: Edinburgh, Scotland, UK
Posts: 9
Rep Power: 10 |
Quote:
|
||
February 3, 2019, 16:52 |
|
#11 |
Retired Super Moderator
Bruno Santos
Join Date: Mar 2009
Location: Lisbon, Portugal
Posts: 10,981
Blog Entries: 45
Rep Power: 128 |
Just a quick note: As of the 23rd of January 2019, it has been implemented into OpenFOAM-dev the ability to have multiple motion solvers: https://github.com/OpenFOAM/OpenFOAM...949a8ebc4d7d01
|
|
April 12, 2019, 12:51 |
|
#12 | |
New Member
Jianming
Join Date: Dec 2016
Posts: 3
Rep Power: 9 |
Quote:
I have a similar issue as yours. I need to combine the sliding mesh solver with the 6DoF rigid body motion solver. Have you solved your problem? Thanks, Jianming |
||
September 12, 2023, 13:27 |
|
#13 |
New Member
Huang, Ching-Chan
Join Date: Jan 2023
Posts: 16
Rep Power: 3 |
Is it possible to use two kind of motion solvers to simulate the [displacement] and [rotation] at the same time. First move the boundary patch up and down, then rotating the cell inside boundary area. Moreover, how to set the refinmemtSurface of the boundary patch area.
|
|
Tags |
displacementlaplacian, dynamicmeshdict, movingtopofvmesh |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Working directory via command line | Luiz | CFX | 4 | March 6, 2011 21:02 |
Getting too many iterations by velocity solving (aborting). Changing U - Solver? | suitup | OpenFOAM Running, Solving & CFD | 0 | January 20, 2010 08:45 |
Motion diffusivity solver has problems with patches moving toward each other | bfa | OpenFOAM Running, Solving & CFD | 2 | July 8, 2009 22:35 |
why the solver reject it? Anyone with experience? | bearcat | CFX | 6 | April 28, 2008 15:08 |
Convergence moving mesh | lr103476 | OpenFOAM Running, Solving & CFD | 30 | November 19, 2007 15:09 |