CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   how to calculate the mean value of a surface (http://www.cfd-online.com/Forums/openfoam/82719-how-calculate-mean-value-surface.html)

xck1986 December 3, 2010 09:32

a problem about steady state solver
 
Dear OF Users,

I have three problems when I use rhoSimpleFoam.

1. Is the relaxationFactors only used for steady state solver? And how much I should choose them. e.g. In my case when I choose the relaxationFactors for rho bigger than 0.1, the result will not converge. But why?

2. Another problem about the Courant Number. For the steady state solver, do I need to also consider the courant number?

3. And I want to calculate the mean velocity and density of the inlet and oulet of a channel. Anyone who know how can I calculate it? Directly in ParaView or use other functions?


Thanks a lot!

DiegoNaval December 3, 2010 10:27

Hi
For the first question, also some unsteady solver use the relaxation factor but in a different way, because normally they use its in the inner iteration but not in the last one (see PimpleFoam). Normally the relaxation factor should be closer to 1 but for a steady solution as the rhoSimpleFoam the value could be smaller to stabilize the solution. The results at the end should be not affected to the value, only the velocity of the convergence. I normally see that if the case blow up could be due to the mesh or the turbulent parameter.
For the second question the Courant Number is valid only for a unsteady solution in that case the solution is steady.
For the third I'm not an expert but I think you can do that in Paraview

I hope to be helpful.

xck1986 December 3, 2010 11:07

Quote:

Originally Posted by DiegoNaval (Post 285886)
Hi
For the first question, also some unsteady solver use the relaxation factor but in a different way, because normally they use its in the inner iteration but not in the last one (see PimpleFoam). Normally the relaxation factor should be closer to 1 but for a steady solution as the rhoSimpleFoam the value could be smaller to stabilize the solution. The results at the end should be not affected to the value, only the velocity of the convergence. I normally see that if the case blow up could be due to the mesh or the turbulent parameter.
For the second question the Courant Number is valid only for a unsteady solution in that case the solution is steady.
For the third I'm not an expert but I think you can do that in Paraview

I hope to be helpful.


Hi Villa,
thanks a lot for your explain.
Is that means, not matter which relaxation factor I have chosen, it has no influence on the final results? The only things is that if it can help to reach convergence or not?
And another question: can I use a transient solver to solve a steady state problem?

Bernhard December 3, 2010 11:17

Wrt 3.

You can add something like this in your ControlDict. I assume it is selfexplanatory for you.
Code:

functions
{
    massFlow
    { 
        type patchAverage;
        functionObjectLibs
          ( 
            "libsimpleFunctionObjects.so"
          ); 
        verbose true;
        patches
          ( 
            inlet
            outlet
          ); 
        fields
          ( 
            U 
          ); 
        factor 1.;
    } 
}


DiegoNaval December 3, 2010 15:09

Yes you can solve all steady problem with a transient solver, you must wait the end of the transition... but normally that needs more computational time, in fact exist some steady solver that use a virtual time marching approach. Like Star..

xck1986 December 3, 2010 16:36

Quote:

Originally Posted by Bernhard (Post 285901)
Wrt 3.

You can add something like this in your ControlDict. I assume it is selfexplanatory for you.
Code:

functions
{
    massFlow
    { 
        type patchAverage;
        functionObjectLibs
          ( 
            "libsimpleFunctionObjects.so"
          ); 
        verbose true;
        patches
          ( 
            inlet
            outlet
          ); 
        fields
          ( 
            U 
          ); 
        factor 1.;
    } 
}


Hi Bernhard,
Thanks a lot for you useful sugguestion!!:)


All times are GMT -4. The time now is 18:50.