Dear CFD programmer/user:

I am currently developing my unstructured hexa code for compressible flow.

The problem that I faced when I try to simulate the flow in centrifugal compressor is the occurence of backflow at pressure outlet (where I fix the outlet pressure and extrapolate the remaining).

Is there any special treatment on backflow at outlet? What are the variables we need to fix/extrapolate at outlet if backflow happens? Any smoothing method that we need to employ to avoid backflow at outlet?

Please help....

thank you.

-khai ching-

Hi; It may phisicaly happen.You can test it by enlarging outlet to obtain fullydeveloped flow hadi.

Hi:

In some cases, backflow may happen during the iteration phase, but not physically. For example, the centrifugal compressor that I extract from FLUENT tutorial.

During the solution process in FLUENT to solve the flow in the same centrifugal compressor, reverse flow is detected at both inlet and outlet. But, after some iterations, reverse flow is no longer exist.

I am sure FLUENT has employed some algorithm to treat backflow at both inlet & outlet.

Anybody has the idea to solve this problem?

tq.

-khai ching-

Hi; During iteration that reverse flow occure you can chang intensity and turbulence intencity(%) parameter. this may help you.

How about if I employ only Navier Stokes Equations but not turbulence models?

Hi; use Higher order disctretization method and let residuals becom smaller than 10^-4. hadi.

hi khai ching when you fix the outlet pressure reverse flow may be occure. pls send me your B. C. . anyways. I had this problem for moving mesh simulation, when I increase inlet velocity my problem is solved