CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Achieving convergence in model with porouse medium (http://www.cfd-online.com/Forums/fluent/31996-achieving-convergence-model-porouse-medium.html)

mneo August 19, 2003 06:12

Achieving convergence in model with porouse medium
 
I have troubles with achieving convergence during computations what can I do? regards

ap August 19, 2003 06:56

Re: Achieving convergence in model with porouse me
 
You can do many things to improve convergence. These are some general guidelines:

1) Use a good quality mesh.

2) If you're using a high order discretization scheme, try starting with the first order scheme, then switch to the higher one.

3) Reduce under-relaxation factors.

4) Change solver parameters (sweeps,...).

What model are you using? What discretization scheme? And what residuals don't reduce?

Hi

ap

mneo August 19, 2003 07:14

Re: Achieving convergence in model with porouse me
 
I use: steady and k-e model first order scheme under-relaxation factors: preasure-0,2 momentum-0,5 k,e-0,6 mesh type-hexa

jieremy August 20, 2003 04:44

Re: Achieving convergence in model with porouse me
 
And you can also conpute innitially using a simple Model.When u have got a initial value then u can change your model.I always do it and it works well.

mneo August 20, 2003 05:19

Re: Achieving convergence in model with porouse me
 
My problem is flow through perforated plate with 45% open.I start computations without turning on "porous zone" and "laminar zone" in Fluid panel in boundary conditions and I achieved convergence. With turned on "laminar zone" I achieved convergence too but I dont't know it's right. Any suggestions? regards

ap August 20, 2003 07:36

Re: Achieving convergence in model with porouse me
 
If the problem is the reduction of k-eps residuals, try to reduce their under-relaxation factors to 0.4.


All times are GMT -4. The time now is 13:29.