[Judith1437]

Hi
everyone. recently I was troubled by the problem "MUT/MU clipped for xxx cells" in numeca calculation. Some topic can been found before in the forum, and may be we can change the expert number MUT/MU to a large number, but it can not solve the question, and maybe the result is far away from the right one. so please give me some suggestion to solve this problem. my case is an axial compressor, and the boundary condition has no problem for it can calculate air correcttly. while the problem appeared when i changed the working fluid to condensaion fluid.

[DarylMusashi]

Hello Judith,
The turbulent viscosity MUT/MU is clipped if it exceeds a certain value which is defined by the user. As you mentioned with integer expert parameter MUCLIP (default value 5000 I think) you can change this value.

The appearance should not just be ignored, as you are right. You should find out by studying your results and convergence history:

1. is MUT/MU still clipped at your last iterations? If yes, go to point 2. If not, you are fine and don't need to take care of it. What matters is the converged solution, not the calculations between.
2. Where (in which cells) does this clipping appear. You may analyse your result file.
3. How many cells are affected / how large is the region of clipped MUT/MU?

As you post in the NUMECA forum and calculate a turbomachinery you use Spalart-Allmaras turbulence model? A value of MUT/MU = 200.000 is acceptable, then. Some of my colleagues even use 500.000 and get reasonable results. If you want to get further SA application hints you may read page 81 at "Turbulence models and their application to complex flows" (http://people.nas.nasa.gov/~pulliam/...uide_v4.01.pdf)

Kind regards / Viele Grüße
Holger

[Hamidzoka]

Hi Judith;
Either improper mesh distributions or boundary conditions may cause this warning. As you mentioned, you have changed the fluid to a condensing one. This may need a review on mesh sizes and also boundary conditions.

regards

[Judith1437]

Dear Holger Dietrich,
Thank for your detailed advice,and your commendatory reference. It is really useful and important. I will try it again after checking on my result file and figuring out the requirement of different turbulent model. Thank you.
Regards

[Judith1437]

Dear Hamid Zoka,
Thank you. You are right, the condensation fulid seems more sensitive to the mesh quality than air, but I do not kown why. I meshed the geom based on the ywall requirement with the flow condition of condensation fulid.
Regards