CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Fluent porous media (http://www.cfd-online.com/Forums/fluent/96370-fluent-porous-media.html)

lostinicem January 20, 2012 10:20

Fluent porous media
 
Hello community,
as described in the title I've got some problems with porous media diffuser. I try to describe as good as possible and hope for your help:)
My problem is to create a fluid flow in a porous diffuser; which fluid is now not really important, so I took nitrogen like in tutorial.
So firstly I created a mesh with ICEM: http://img268.imageshack.us/img268/1076/fluent12.jpg
since I'm new to this whole stuff; I think I invested the right amount of time so it should be ok ..
so after that I imported it to fluent:
1) general: I scale it and click on axis-symmetric
2) models: I choose the viscous laminar
3) materials: nitrogen, so I took all datas from tutorial
4) cell zone conditions: I choose nitrogen and click on porous zone; i take also the same inertial and viscous resistance; so I dont use power law model nor do I change the porosity ( its 1)
5) boundary conditions): inlet: I enter here the fluid velocity, since I can calculate it; outlet: I dont touch anything here
6) solution methods: all I change here is the momentum: its on second order upwind
7) monitors: at residuals I change absolute criteria to 1e-6 ( its bigger than in tutorial, I tried to work with the bigger one, but still no convergence); surface monitors: I choose outlet as surface, I also keep the settings from tutorial
8) so at the last step I initialize and compute from inlet, since the tutorial also says so.

So I tried to change and play with the settings but still no convergence, so my velocity profile looks quiet disappionting; and also since I'm new to this stuff maybe I made a stupid mistake?
Hope someone could help me and sorry for my english :p

lostinicem January 23, 2012 04:17

maybe some hints? :)

SHL January 25, 2012 14:40

Hello,


I'm not sure, but maybe you can try to use the ANSYS meshing. Sometimes, incompatibility in mesh causes such problems. Also, you can start calculations with first order upwaind, and then after some degree of convergence, switch to first order.

Hope this works.

lostinicem January 26, 2012 11:35

thanks SHL,
but already found the mistake; the problem was in the boundary condition; I didnt defined it as interior.:)


All times are GMT -4. The time now is 18:42.