CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Segregated or coupled? (

Pierluigi Allegro September 23, 2000 04:38

Segregated or coupled?
Hi All, I'm solving a steady 3d axial flow turbine problem using standard k-eps model. The balde is film-cooled by 3 cooling holes; fluid is rotatind at 314rad/s (399m/s in the middle) and the higher Ma is 0.88. The problem is that in the compressible simulation the segregated solver diverges for any relax factor, so I have too use the coupled solvet that takes a lot of time ( I have 500.000 nodes). Is it possible to use the segregated solver or the problem prevent this? Thankyou all.

John C. Chien September 23, 2000 13:16

Re: Segregated or coupled?
(1). All I know is, for subsonic flow it is better to use pressure-based formulation. (2). Since your problem is in the transonic range, the right approach is the density-based transient formulation. I am somewhat confused with their new names of algorithm. (3). So, ths suggestion is: use the density-based transient formulation. Based on my previous experience, it may take more than a couple of weeks to converge using compressible, transient formulation. But, that is life. (4). Even if you can obtain the solution ,500,000 nodes may not be adequate at all for mesh independent solution.

Jin-Wook LEE September 25, 2000 23:51

Re: Segregated or coupled?
Dear Pierluigi Allegro

In Fluent, coupled solver means 'algorithm for compressible flow'. It is better for you to use 'coupled solver' because compressible effect can't be neglected for your problem, Mach Number =0.88. In general, as you may know, the flow is considered as compressible when Ma>0.3.

Sincerely, Jinwook

All times are GMT -4. The time now is 14:51.