CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > CFX

Liquid Volume Error when VOF + Deforming Mesh Used

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Display Modes
Old   June 5, 2012, 10:14
Default Liquid Volume Error when VOF + Deforming Mesh Used
  #1
Member
 
Nick Cleveland
Join Date: Mar 2012
Posts: 34
Rep Power: 5
NCle is on a distinguished road
I am writing to post a type of error I found with VOF and deforming moving mesh used in conjunction. What happens is that the mesh elements change volume with the moving mesh, but the VOF vol frac in each element stays the same, as if the mesh wasn't moving and the volume not decreasing overall. The result is that the liquid inside each cell gets "compressed" with the cell/element artificially, implying that the total mass of liquid in the domain is decreasing when it should stay completely constant.

Has anyone come across this error also in your work? I am now trying to find a way around it. possibly CFX doesn't have one and only FLUENT does? I'll post a video and picture also to clarify my explanation. Can't post a video though

I'm posting just to say that if anyone has strange fluid loss errors when using moving mesh that this could be the cause.
Attached Images
File Type: jpg VOF+Mov+Mes_error.jpg (18.7 KB, 12 views)
File Type: jpg vof_MM-err.jpg (31.1 KB, 13 views)
NCle is offline   Reply With Quote

Old   June 5, 2012, 19:43
Default
  #2
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,937
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
Which version of CFX are you using?

Before anything else you should check whether you simply need tighter convergence or smaller time steps. Check whether you still get the effect with far tighter convergence and/or tighter convergence.

Then test some other options, such as coupled versus segregated VF. The first implementation of the coupled VF solver had some conservation issues but I think it was resolved in later releases of CFX.
ghorrocks is offline   Reply With Quote

Old   June 6, 2012, 07:53
Default
  #3
Member
 
Nick Cleveland
Join Date: Mar 2012
Posts: 34
Rep Power: 5
NCle is on a distinguished road
Quote:
Originally Posted by ghorrocks View Post
Which version of CFX are you using?

Before anything else you should check whether you simply need tighter convergence or smaller time steps. Check whether you still get the effect with far tighter convergence and/or tighter convergence.

Then test some other options, such as coupled versus segregated VF. The first implementation of the coupled VF solver had some conservation issues but I think it was resolved in later releases of CFX.

I'm using the CFX 12.1 version that comes with ANSYS 12.1.

Do you know by any chance if the convergence can be tightened by reducing the "residual target" value under "convergence criteria" in the solver controls? (I'll just try that to test for result). I already tested the effect of smaller timestep. The effect was negligible on this error. Also, I coincidentally already experimented with the coupled/segregated VOF multiphase settings to no avail. I'll try that convergence criteria tightening now.
NCle is offline   Reply With Quote

Old   June 6, 2012, 08:01
Default
  #4
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,937
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
You should consider upgrading to the current version. I think they made some big improvements to the coupled VF solver in V13. I remember V12 had problems with VF conservation - which is exactly what you are seeing.

Quote:
Do you know by any chance if the convergence can be tightened by reducing the "residual target" value under "convergence criteria" in the solver controls?
Of course. Let me guess, you are new to CFD.....

And forget about reporting it as a bug if you are on V12. No point fixing a bug on an old version which is probably already fixed anyway. (Get my drift? Using old versions of the software means you may well be wasting your time on a problem which is already fixed.)
ghorrocks is offline   Reply With Quote

Old   June 6, 2012, 11:52
Default
  #5
Member
 
Nick Cleveland
Join Date: Mar 2012
Posts: 34
Rep Power: 5
NCle is on a distinguished road
Gaining experience by the minute. Thanks, I'll try 13
NCle is offline   Reply With Quote

Old   June 6, 2012, 18:56
Default
  #6
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,937
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
Why not try V14?
ghorrocks is offline   Reply With Quote

Old   June 7, 2012, 10:01
Default
  #7
Member
 
Nick Cleveland
Join Date: Mar 2012
Posts: 34
Rep Power: 5
NCle is on a distinguished road
Ok, got another amateur question: If I run a model in CFX solver 13 that I setup in CFX-pre 12.1, shouldn't that solution have the newer VF coupling corrrectness? I'm pretty sure it should, but I just did that (ran a 12.1 setup in 13 solver) and it had the same exact mass conservation issue, no improvement.

I ran it again with a much tighter conservation target and the same exact solution was produced. Even if I could get access to v14, not sure if it would be any better. Going to try Fluent v13 now.
NCle is offline   Reply With Quote

Old   June 7, 2012, 18:53
Default
  #8
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,937
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
It is starting to look like you have found a bug in VF conservation in moving meshes. But forget about doing anything about it unless you run in the current version, V14. If it also happens in V14 then I would report it as a bug and hopefully you can get it fixed.
ghorrocks is offline   Reply With Quote

Old   June 8, 2012, 13:07
Default
  #9
Member
 
Nick Cleveland
Join Date: Mar 2012
Posts: 34
Rep Power: 5
NCle is on a distinguished road
Ok, got it. The ANSYS technical support said that the surface tension model causes problems at this small a scale (sub-micron). When I turned it off, there was less, but still some, mass loss.
NCle is offline   Reply With Quote

Reply

Thread Tools
Display Modes

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 On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
how to set periodic boundary conditions Ganesh FLUENT 13 January 22, 2014 05:11
3D Hybrid Mesh Errors DarrenC ANSYS Meshing & Geometry 11 August 5, 2013 06:42
[ICEM] surface mesh merging problem everest ANSYS Meshing & Geometry 39 June 5, 2013 19:02
BlockMesh FOAM warning gaottino OpenFOAM Native Meshers: blockMesh 7 July 19, 2010 14:11
Convergence moving mesh lr103476 OpenFOAM Running, Solving & CFD 30 November 19, 2007 15:09


All times are GMT -4. The time now is 16:38.