CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Mesh Generation & Pre-Processing Software > ANSA

ANSA mesh for sliding meshes (pimpleDyFoam

Register Blogs Community New Posts Updated Threads Search

Like Tree2Likes
  • 1 Post By tomf
  • 1 Post By vangelis

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   July 11, 2014, 10:19
Default ANSA mesh for sliding meshes (pimpleDyFoam
  #1
New Member
 
Join Date: Nov 2010
Posts: 9
Rep Power: 16
yosuu is on a distinguished road
Dear users,

I was wondering if you could give me some tips/"best practices" to create a mesh for OpenFOAM with a cylindrical interface where a fan lies in it. I have used the this method:

- I meshed everything and define the interface as baffles, then exported to OpenFoam case. Then in the polyMesh/boundary I change manually the baffles patches to cyclicAMI type. I use then the pimpleDyMFoam solver for the simulation, it works perfectly when I start from 0, but then if I write the solution and I restart the simulation, it is not possible due to some error in the interpolation between the faces. (this is due to the following: when I export the mesh to openfoam, the faces of the two patches that baffles creates share the same points instead of one duplicating these points, so writing the solution with the moved mesh does not modify the points of the patches but the points of the cells inside the moved mesh are updated).

- I use then the tool of OpenFOAM to duplicate the points (mergeOrSplitBaffles -split -overwrite), it duplicates the points but then I cannot even run the solver.

Is there a good method to create sliding meshes for OpenFOAM with ANSA? I think I am doing something wrong.

Kind regards,

yosuu

pd; ansa support in my country cannot give me any tips since they don't have any experience in such meshes.
yosuu is offline   Reply With Quote

Old   July 14, 2014, 06:14
Default
  #2
Senior Member
 
Tom Fahner
Join Date: Mar 2009
Location: Breda, Netherlands
Posts: 647
Rep Power: 32
tomf will become famous soon enoughtomf will become famous soon enough
Send a message via MSN to tomf Send a message via Skype™ to tomf
Hi yosuu,

Our workflow is a bit different: Define the interface as a wall, so you get two different volumes in ANSA. Next do the meshing of both volumes, with the volumes having a different PID. Only show one of the volumes and all the boundaries (including the interface) of only that volume and use save visible as function. Press invert, show the interface again, rename the interface as interface_slave and use save visible as again. Now open the first "save visible as" saved file and use merge to include the second saved ansa file. You should now have the complete mesh with the interface split in 2 equal patches (interface and interface_slave), which can be wall in ANSA. Now export to OpenFOAM and again manually change the interface patches to cyclicAMI and off you go.

I hope this is clear enough?

Kind regards,
Tom
Yann Scott likes this.
tomf is offline   Reply With Quote

Old   July 15, 2014, 10:29
Default
  #3
Senior Member
 
Vangelis Skaperdas
Join Date: Mar 2009
Location: Thessaloniki, Greece
Posts: 287
Rep Power: 21
vangelis is on a distinguished road
Hi there
I just want to add that you can set these two properties
as cyclicAMI type in the PID list in ANSA and output the mesh ready to run
without need for manual changes afterwards.

Then in auxiliaries>Interface
create a new interface of type AMI
and set it to noordeting type

Vangelis
tomf likes this.
vangelis is offline   Reply With Quote

Old   June 18, 2016, 06:35
Default
  #4
Member
 
alex
Join Date: Feb 2016
Location: chennai
Posts: 48
Rep Power: 10
crusen mind is on a distinguished road
Quote:
Originally Posted by tomf View Post
Hi yosuu,

Our workflow is a bit different: Define the interface as a wall, so you get two different volumes in ANSA. Next do the meshing of both volumes, with the volumes having a different PID. Only show one of the volumes and all the boundaries (including the interface) of only that volume and use save visible as function. Press invert, show the interface again, rename the interface as interface_slave and use save visible as again. Now open the first "save visible as" saved file and use merge to include the second saved ansa file. You should now have the complete mesh with the interface split in 2 equal patches (interface and interface_slave), which can be wall in ANSA. Now export to OpenFOAM and again manually change the interface patches to cyclicAMI and off you go.

I hope this is clear enough?



Kind regards,
Tom

hello Tom
Is there any tutorial material for this propeller
crusen mind is offline   Reply With Quote

Old   June 20, 2016, 04:52
Default
  #5
Senior Member
 
Tom Fahner
Join Date: Mar 2009
Location: Breda, Netherlands
Posts: 647
Rep Power: 32
tomf will become famous soon enoughtomf will become famous soon enough
Send a message via MSN to tomf Send a message via Skype™ to tomf
Quote:
Originally Posted by crusen mind View Post
hello Tom
Is there any tutorial material for this propeller
No, this is just a general workflow description.
Regards,
Tom
tomf is offline   Reply With Quote

Old   June 25, 2016, 07:43
Default
  #6
Member
 
alex
Join Date: Feb 2016
Location: chennai
Posts: 48
Rep Power: 10
crusen mind is on a distinguished road
Quote:
Originally Posted by vangelis View Post
Hi there
I just want to add that you can set these two properties
as cyclicAMI type in the PID list in ANSA and output the mesh ready to run
without need for manual changes afterwards.

Then in auxiliaries>Interface
create a new interface of type AMI
and set it to noordeting type

Vangelis
In Auxileries interfce > new we have to select patch and neighbouring patch. how to select patch and neighbouring patch
crusen mind is offline   Reply With Quote

Old   June 27, 2016, 05:44
Default
  #7
Senior Member
 
Vangelis Skaperdas
Join Date: Mar 2009
Location: Thessaloniki, Greece
Posts: 287
Rep Power: 21
vangelis is on a distinguished road
Type ? in each field (patch and neighboring patch)
and select the PID of each side


Quote:
Originally Posted by crusen mind View Post
In Auxileries interfce > new we have to select patch and neighbouring patch. how to select patch and neighbouring patch
vangelis is offline   Reply With Quote

Old   August 5, 2016, 10:58
Default
  #8
Member
 
alex
Join Date: Feb 2016
Location: chennai
Posts: 48
Rep Power: 10
crusen mind is on a distinguished road
Hi vengalis

thanks for the help so far
my case involve meshing a propeller, or wing which has sharp edges, when ever i press fix quality it automatically deforms shape of mesh by deleting some mesh regions which dont represent the shape that we want to simulate.

I have one more question-how to reduce non orthogonality and skewness ?.
crusen mind is offline   Reply With Quote

Old   August 10, 2016, 10:32
Default
  #9
Senior Member
 
Vangelis Skaperdas
Join Date: Mar 2009
Location: Thessaloniki, Greece
Posts: 287
Rep Power: 21
vangelis is on a distinguished road
If the sharp angles are very acute then it is not possible to satisfy openfoam criteria for skewness and orthogonality by definition of these criteria.
Can you post an image of the surface mesh where you have the problems?
vangelis is offline   Reply With Quote

Old   August 12, 2016, 10:58
Default
  #10
Member
 
alex
Join Date: Feb 2016
Location: chennai
Posts: 48
Rep Power: 10
crusen mind is on a distinguished road
Thanks Vengalis
when I am using scripts for post wrap fix and quality criteria i am getting good results. I will post picture of my propeller, I want to generate a good quality o-grid using hexa_block meshing. your suggestion would be invaluable.
Attached Images
File Type: png model and domain.png (8.2 KB, 24 views)
File Type: png propsurface.png (9.5 KB, 26 views)
crusen mind is offline   Reply With Quote

Old   August 12, 2016, 11:16
Default
  #11
Senior Member
 
Vangelis Skaperdas
Join Date: Mar 2009
Location: Thessaloniki, Greece
Posts: 287
Rep Power: 21
vangelis is on a distinguished road
Hi there
It seems that you have a domain made out of Faces
while your propeller is probably in STL mesh and you used wrap
to create a CFD surface mesh.
It would help a lot if you also had the propeller geometry as Faces.
Could you do that?
Creating a Hexablock mesh is not an simple task and it would
also require the presence of the actual geometry of the propeller as Faces.
vangelis is offline   Reply With Quote

Old   February 14, 2023, 11:28
Default
  #12
New Member
 
Vishwas
Join Date: Jul 2019
Location: Graz, Austria
Posts: 2
Rep Power: 0
Vishwas is on a distinguished road
Quote:
Originally Posted by tomf View Post
Hi yosuu,

Our workflow is a bit different: Define the interface as a wall, so you get two different volumes in ANSA. Next do the meshing of both volumes, with the volumes having a different PID. Only show one of the volumes and all the boundaries (including the interface) of only that volume and use save visible as function. Press invert, show the interface again, rename the interface as interface_slave and use save visible as again. Now open the first "save visible as" saved file and use merge to include the second saved ansa file. You should now have the complete mesh with the interface split in 2 equal patches (interface and interface_slave), which can be wall in ANSA. Now export to OpenFOAM and again manually change the interface patches to cyclicAMI and off you go.

I hope this is clear enough?

Kind regards,
Tom

Hi Tom,
I'm as well trying a Propeller simulation with ANSA mesh in OpenFOAM. I followed the workflow of the two ANSA files as explained. The Dynamic motion is applied to the selected zone. The AMI are generated with AMI weights as perfectly 1. But I get following error just at the first iteration of PIMPLE foam - Not implemented
From virtual void Foam::cyclicAMIGAMGInterface::write(Foam::Ostream& ) const in file AMIInterpolation/GAMG/interfaces/cyclicAMIGAMGInterface/cyclicAMIGAMGInterface.H at line 160


Any ideas what it might be? And how to resolve it?

Thanks a lot!
Best regards,
Vishwas
Vishwas is offline   Reply With Quote

Old   February 14, 2023, 12:58
Default
  #13
Senior Member
 
Tom Fahner
Join Date: Mar 2009
Location: Breda, Netherlands
Posts: 647
Rep Power: 32
tomf will become famous soon enoughtomf will become famous soon enough
Send a message via MSN to tomf Send a message via Skype™ to tomf
Hi Vishwas,

This message is too general to tell you what is wrong. However it does seem to be on the OpenFOAM side, not ANSA. Maybe check your setup against the setup of the propeller tutorial of OpenFOAM, for version 2206 I have it at:

$FOAM_TUTORIALS/incompressible/pimpleFoam/RAS/propeller

Good luck,
Tom
tomf is offline   Reply With Quote

Old   February 16, 2023, 11:16
Default
  #14
New Member
 
Vishwas
Join Date: Jul 2019
Location: Graz, Austria
Posts: 2
Rep Power: 0
Vishwas is on a distinguished road
Hi Tom,


Thanks for your reply. There were some minor differences in the fvSchemes (my old fvSchemes were adjusted for the MRF simulations). Also the includeEtc "caseDicts/setContraintTypes" was missing in the boundary condition files (which could've been influential point for the AMI patches). But thanks everything is working now.



Cheers,
Vishwas
Vishwas is offline   Reply With Quote

Reply


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Moving mesh Niklas Wikstrom (Wikstrom) OpenFOAM Running, Solving & CFD 122 June 15, 2014 07:20
ANSA mesh quality report bondmatt ANSA 2 February 18, 2013 07:35
[ICEM] Problem making structured mesh on a surface froztbear ANSYS Meshing & Geometry 4 November 10, 2011 09:52
Check the skewnesses of your face meshes and make sure the face mesh sizes are not to sophie-l Main CFD Forum 1 April 13, 2009 20:16
Check the skewnesses of your face meshes and make sure the face mesh sizes are not to sophie-l ANSYS Meshing & Geometry 0 April 13, 2009 18:27


All times are GMT -4. The time now is 16:49.