|
[Sponsors] |
March 21, 2010, 09:57 |
Convergence Problem?
|
#1 |
Senior Member
Join Date: Nov 2009
Posts: 111
Rep Power: 16 |
Hello
I'm trying to compare 2 geometries and I'm having "problems" with the 2nd one. Here is the "round" geometry : http://yfrog.com/0vroundnp It converges in approxmimately 500 iterations. Settings are : 2D, SIMPLE algorithm, pressure standard, 2nd order for all the other discretizations, k-epsilon Enhanced wall function. Y+<1 everywhere With the "square" geometry and the same settings, it converges in approximately 7500 iterations and at the beginning I have weird things appearing for the pressure field as you can see in this screen : http://yfrog.com/8dconvergence1p Velocity doesn't seem to have this problem : http://yfrog.com/5zvelocityconvergencep And I don't understand why I have such behaviour... Could you give me some ideas? Thanks for the help |
|
March 22, 2010, 01:04 |
|
#2 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
you have a problem with the mesh at the right side from the square geometry.
Check the connectivity from your edges over there. You can display the default wall (without grid) and check if you don't have an edge in the fluid domain.
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
March 22, 2010, 02:37 |
|
#3 |
Senior Member
Join Date: Nov 2009
Posts: 111
Rep Power: 16 |
That's what I think too but I can't find the error in the mesh.
And what's most weird is that if I let fluent iterates more (say 7500 iterations) then everything seems to be ok ... |
|
March 22, 2010, 02:43 |
|
#4 |
Senior Member
Join Date: Nov 2009
Posts: 111
Rep Power: 16 |
I defined the edges inside the canal as "interior" for the boundary condition type in gambit
Is it correct? |
|
March 22, 2010, 03:04 |
|
#5 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
you don't need to define them.
If your splits are correct, gambit sets them automatically as interior. If you need to define them, then there is a problem. Click and click and click on the edge, and check how many edges are picked. If there is more than one, then you have superposed edges.
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
March 22, 2010, 04:04 |
|
#6 |
Senior Member
Join Date: Nov 2009
Posts: 111
Rep Power: 16 |
Hi,
Again thanks for the help I checked egdes in gambit : I deleted edges to see if there were more than one and there was only one edge so no problem here apparently |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
convergence problem when use pisoFoam, LES for wind tunnel case | Forrest_Lei | OpenFOAM | 3 | July 19, 2011 06:00 |
convergence problem | commonyue | Main CFD Forum | 1 | December 1, 2009 03:54 |
Convergence of CFX field in FSI analysis | nasdak | CFX | 2 | June 29, 2009 01:17 |
3D Fluid Flow Convergence problem | Emily | FLUENT | 2 | March 21, 2007 22:18 |
Non Convergence of 3D Heat transfer cfd problem | Balraj | Main CFD Forum | 3 | December 9, 2004 00:24 |