CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Convergence does not stable (https://www.cfd-online.com/Forums/fluent/223356-convergence-does-not-stable.html)

kcd96 January 7, 2020 11:16

Convergence does not stable
 
Hello dear friends,

I'm doing a vehicle simulation in Fluent 19.2. I put all of inputs and doing hybrid initialization then starting the calculations and my drag coefficient, drag force etc. graphs being up and down always change with the iteration values. When the calculation done the value which gives me last unit of the iteration. For example if I do 500 iteration, results give me 500th. values and I can not believe that results because it depends with the among of iteration how many iteration I entered. So if the graphs converge correctly I can believe the solution values. Please help If you have any idea about that problem.

Editional Info: The X coordinate values are expected values, the problem is only with none converging.

NonStopEagle January 12, 2020 06:38

Quote:

Originally Posted by kcd96 (Post 754011)
Hello dear friends,

I'm doing a vehicle simulation in Fluent 19.2. I put all of inputs and doing hybrid initialization then starting the calculations and my drag coefficient, drag force etc. graphs being up and down always change with the iteration values. When the calculation done the value which gives me last unit of the iteration. For example if I do 500 iteration, results give me 500th. values and I can not believe that results because it depends with the among of iteration how many iteration I entered. So if the graphs converge correctly I can believe the solution values. Please help If you have any idea about that problem.

Editional Info: The X coordinate values are expected values, the problem is only with none converging.

what type of simulation are you running?
Steady or unsteady?
If you are running a steady simulation, consider switching over to a transient simulation.
Sometimes when the flow around a body is inherently unsteady, there is a possibility that the steady solver might not be able to converge, hence giving you the oscillating behavior that you mentioned.
If you're already running a unsteady simulation, consider decreasing the time step( putting it as adaptive may help).
of course this is assuming that your mesh is good, the y+ values are good and your choice of physics is correct (i.e. turbulence model etc)


NonstopEagle

kcd96 January 12, 2020 07:11

Dear NonStopEagle,

Firstly I thank you to reply my post. I'm working with steady withsome changing I got stable values but they are very low then I expected. So that reason I called my ANSYS desk. They told me those y+ values and meshing problems. I hope to solve with these info's. Thank you very much.

Best regards.

NonStopEagle January 12, 2020 08:13

Quote:

Originally Posted by kcd96 (Post 754404)
Dear NonStopEagle,

Firstly I thank you to reply my post. I'm working with steady withsome changing I got stable values but they are very low then I expected. So that reason I called my ANSYS desk. They told me those y+ values and meshing problems. I hope to solve with these info's. Thank you very much.

Best regards.

No problem mate, feel free to post if your problem isn't solved.

cheers

NonStopEagle


All times are GMT -4. The time now is 14:42.