CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   how to simulate orifice in supersonic flow? (http://www.cfd-online.com/Forums/fluent/117849-how-simulate-orifice-supersonic-flow.html)

 YING May 16, 2013 05:30

how to simulate orifice in supersonic flow?

4 Attachment(s)
Good afternon,

I have some problems in simulating orifice in supersonic flow.

As shown in the following picture is my model. The left side is entrance, and the right side is exit. The diameter of entrance and exit are 5mm. The diameter and thick of the orifice are 0.12mm and 0.2mm, respectively. http://www.cfd-online.com/Forums/att...1&d=1368692051

ideal gas, k-e viscous model

The inlet is set as pressure-inlet with total pressure of 1atm, and the outlet is as pressure-outlet with an gauge pressure of 600Pa.

if I use pressure based as solver, the results are shown in the following picture.
streamline
http://www.cfd-online.com/Forums/att...1&d=1368694294
velocity contours
http://www.cfd-online.com/Forums/att...1&d=1368694300
The mass flow rate of inlet and outlet are 2.1621258e-6kg/s and -2.1409824e-6kg/s, respectively.

if I use density based as solver, the results are shown in the following picture.
streamline
http://www.cfd-online.com/Forums/att...1&d=1368695154
velocity contours
http://www.cfd-online.com/Forums/att...1&d=1368695163
The mass flow rate of inlet and outlet are 1.9779529e-6kg/s and -2.0015871e-6kg/s, respectively.

The question is which solver should be selected to simulate the orifice in supersonic flow.
If I select pressure based as solver, the streamline in the picture looks good, while velocity contours looks not good. If I select density based as solver, the streamline in the picture looks not good, while velocity contours looks good.

Thanks in advance to all community for support.

Best regards.

 oj.bulmer May 16, 2013 09:58

Hope you have done adequate mesh independence study.

Try density based coupled solver here. Try with first order discretization and get a nice convergence and then switch to second order.

Have you tried realizable model? It tends to model the jets better than k-eps.

OJ

 YING May 16, 2013 10:41

dear OJ