CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   rotor stator (

Amir Jourak July 4, 2007 19:19

rotor stator
Hi all, I'm trying to somehow use tutorial 12 "Flow in axial rotor" to solve my problem. i made a rotor with fluid domain and stator with fluid domain too. The rotor has a dimpel so i'm trying to see if i can reduce the pressure in contact or not. I got really good answer in steady state condition and the solver converged really fast. But in the transient answer it will not converge at all and fluctuating ...without any convergency. Do you know anything about velocity in stationary frame?!! and do you ever try to solve the same problem. please help. Thanks a lot.

blaca July 6, 2007 09:34

Re: rotor stator
If you followed a similar tutorial the problem should; in theory, be set up correctly. In that case you may need to check the solver settings and adjust the step size, also make sure to let the solver run for a really long time when it does start to converge to ensure it complete convergence.

Also, is your mesh refined enough in areas of complex physics?

Amir Jourak July 6, 2007 09:51

Re: rotor stator

Well the mesh is quite good. the problem is i started steady state solution using frozen rotor ans it converge really fast and with good answer. seems the two geometry interact together. but when i try to start transient solution with steady state solution as an initial conditon it will not converge at all even with changing time steps and even i let it run for 5 or several pitches. it is just fluctuating at the same frequency!

I also tried with zero velocity as an initial condition but after a while it start fluctuating again in transient case. i have a rotor with 6mm diameter and a dimpel with .1mm height,length in the rotor and also a 20mm stator. both of them are fluid domain and i have fluid fluid interface between two and a layer of film thickness of 2mm between them.

Do you think that fine mesh will influence the convergency in transient solution? and is it possible to count on transient solution when it does not converge at all and just fluctuating at residual below 10.e-4 or not?

what settings can i change in solver to make it convergence please help me on that.

Thanks a lot.

All times are GMT -4. The time now is 10:35.