CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Siemens (
-   -   Appropriate turbulent model (

Ashish Joshi March 27, 2003 12:55

Appropriate turbulent model
Hi, I am simulating a converging-diverging nozzle, with a compressible flow, and Re=50,000 - 75,000. It suffers an adverse pressure gradient. It is told to me that a K-E high reynold no. model wont work for this type of flow. What is the appropriate turbulent model? K-W can be a choice, but is not offered by Star-CD? Do reply, thanks in advance Ashish

cjtune March 27, 2003 22:49

Re: Appropriate turbulent model
The k-w model is available (in v3.15 onwards, albeit a beta feature). You just have to turn on the appropriate switches and constants. You can find the instructions in the Supplementary Notes but for your convenience:

1. Select the low Reynolds number k-e model in the Turbulence Models panel. 2. Write the geometry file using command GEOM or an equivalent GUI operation. 3. Select the high Reynolds number k-e model in the Turbulence Models panel. 4. Set Constant 88 to 3 using command RCON 88 3. 5. Write the problem file using command PROB or an equivalent GUI operation. Ignore any warnings about file This file can be deleted. 6. For a STAR-HPC analysis, the geometry decomposition process will create .bshl files in each sub-domain subdirectory. These files may be deleted.

The above steps actually activate the _high_-Reynolds form of the 'standard' k-w model although you have chosen the low-Reynolds k-e model options in GUIde.

Thomas March 28, 2003 02:02

Re: Appropriate turbulent model
try to use the qubic - k-eps modell. it should work fine as well. from my experience you should use the two -layer version.

regards, thomas

All times are GMT -4. The time now is 16:20.