CFD Online Discussion Forums
Page 3 of 3 12 3
Show 40 post(s) from this thread on one page

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Diverging simulation (https://www.cfd-online.com/Forums/openfoam-solving/161844-diverging-simulation.html)

czhongrong November 20, 2015 05:01
Alright, so I will set my deltaT as 0.25E-4, maxCo as 0.5 and make the changes in the PIMPLE block, and see how it goes.

Just wondering, would maxCo = 0.5 make my simulation to run for ages? Because initially I set it as 0.5 and its really crawling and I do not have that luxury of time.:confused:

mkraposhin November 20, 2015 05:57
I saw this strange place in your fvSolution just now:

Code:
relaxationFactors
{
    U              0.1;
    nuTilda        0.1;
    p                0.1;
    k                0.1;
    omega                0.1;
}
If you are doing transient simulation, you must set relaxation factors for all fields to 1. But, if you need only a steady-state solution, you can set relaxation factors as folows:
Code:
relaxationFactors
{
    U              0.3;
    nuTilda      0.5;
    p                0.2;
    k                0.5;
    omega        0.5;
}
Then, you must set nOuterCorrectors to larger value, let's say about 10-50. And this will let you to increate deltaT to values with Co >> 1
Code:
PIMPLE
{
    nOuterCorrectors    20;
    nCorrectors        2;
    nNonOrthogonalCorrectors 0;
    pRefCell            0;
    pRefValue          0;
}

czhongrong November 20, 2015 07:08
Yes, I'm running transient simulation. If I set all the relaxationFactors to 1, do I still need to set the nOuterCorrectors to 20 as well?

mkraposhin November 20, 2015 07:15
Quote:
Originally Posted by czhongrong (Post 574185)
Yes, I'm running transient simulation. If I set all the relaxationFactors to 1, do I still need to set the nOuterCorrectors to 20 as well?
No, if you are running transient simulation, you don't need to use outer corrections, so you msut set nOuterCorrectors to 1. And all relaxation factors must be set to 1.

czhongrong November 20, 2015 23:22
It seemed to be able to simulate past the error point. But its taking way too long~(about 6hours to reach 0.01) What are some ways where I can speed up the process without having it to diverge again?

mkraposhin November 21, 2015 07:17
1) Increase maxCo to 0.8-0.9, but at the cost of chance of divergence
2) Use more CPU


All times are GMT -4. The time now is 00:46.
Page 3 of 3 12 3
Show 40 post(s) from this thread on one page