CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Problems modeling subsonic engine air inlet with flow separation (https://www.cfd-online.com/Forums/cfx/63083-problems-modeling-subsonic-engine-air-inlet-flow-separation.html)

 TWaung March 27, 2009 17:42

Problems modeling subsonic engine air inlet with flow separation

Hi,

I need to model an air inlet for an air breathing engine (RBCC engine). Flow velocities through the duct are subsonic (Mach 0.2 - 0.5). The average static pressure at the outflow plane of the air inlet is fixed. I am having convergence problems in the regions of flow separation and I was wondering if anyone has any suggestions on how to improve convergence.

I have tried increasing mesh density in the regions of separated flow but so far this has not helped. (I am also reaching the memory limits of the computers I have access to.)

I have been hesitant to extend the domain past the true outflow plane since I need to achieve a given average pressure at this plane. Is it typical to extend the domain past the true outflow plane for simulations? If so, how does one achieve the desired pressure at the true outflow plane location? (iteration?).

The simulation details are as follows.
Domain Settings
Using SST turbulence model, Ideal Gas, Total Energy model

Boundary conditions
- subsonic velocity inlet
- opening pressure outlet. Using the entrain setting since I expect there to be regions of separated flow crossing the outflow plane. I have been hesitant to extend the domain past the true outflow plane since I need to achieve a given average pressure at this plane.

Solver Settings
Advection scheme: High Resolution (when using UW scheme, the simulation converges)

Timescale control: Local Time scale

Any suggestions would be greatly appreciated!

Thanks

 ghorrocks March 29, 2009 17:52

Have a look at:

http://www.cfd-online.com/Wiki/Ansys...gence_criteria

Try the recommendations on that page but I bet in your case you have transient flow structures so you will need to run a transient simulation.

Glenn Horrocks

 TWaung March 29, 2009 18:42

Thanks Glenn,

I'll give the transient simulations a try.

 All times are GMT -4. The time now is 19:54.