[Ravenn]

Hello,

I'm trying to compute the flow over a wedge with an angle of 18° and a free-stream Mach number of 9, with fluent 14. (inviscid flow)

The domain I am currently using is printed on the attached file (case 1).

Boundary conditions:
- inlet: pressure far field
- high boundary: pressure far field
- low boundary before leading edge: symmetry
- low boundary (wedge and after the wedge): wall
I tried different combinations of boundary conditions for the outlet: pressure far field or pressure outlet.

I am wondering which one of the combination is the best, because I tried different domain length after the wedge, and there is always a gradient in pressure and Mach number on the outlet, so I think using pressure far field is not ideal. Is pressure outlet better? since in supersonic, the conditions on this boundary are extrapolated from the interior.

An other aspect which is linked to boundary conditions:
If I change the problem by changing the low boundary before the wedge to a wall and using viscous simulation, is it dangerous to use pressure far field directly near a no-slip wall?
Wouldn't it be better to place a small symmetry plane before the flat plat?
(see attached file, case 2)

Thank you in advance

[Ravenn]

Any advice for this case?

[D-willism]

Hi, Ravenn. I am new to CFD and now doing a similar CFD simulation as yours and I am also kinda confused on boundary conditions for the outlet: pressure far field or pressure outlet. As you mentioned "since in supersonic, the conditions on this boundary are extrapolated from the interior." what do you mean by "extrapolated from the interior" ?

My understanding is that if I use pressure outlet at the end of the wedge, the software can extrapolate the flow data at the outlet based on the data at the inlet, and all I need to do is to do some simple settings in Flunet interface. Is my understanding right?

Thank you and Best Regards

[Ravenn]

Hi D-willism,

I found the sentence "since in supersonic, the conditions on this boundary are extrapolated from the interior" in Fluent manual.

The complete and exact description from the manual is ( https://www.sharcnet.ca/Software/Flu...ug/node244.htm ):
"Pressure outlet boundary conditions require the specification of a static (gauge) pressure at the outlet boundary. The value of the specified static pressure is used only while the flow is subsonic. Should the flow become locally supersonic, the specified pressure will no longer be used; pressure will be extrapolated from the flow in the interior. All other flow quantities are extrapolated from the interior."

So my understanding is that the extrapolation is done on the flow field near the outlet in the interior of the domain. What do you think?

I haven't started computations on this case yet, I will during the next few days.

Regards

[D-willism]

from the manual, yes I agree with you. so Could you plz tell what kind of setting or parameter input you have for the pressure outlet, say mass flow, gauge pressue or etc.?

Regards

[Ravenn]

If the flow is supersonic, I don't know if these parameters are important. The only part where it could be important is in the subsonic layer of the boundary layer, but I have to try different values.

Regards

[A_Lyaskin]

Hi, Ravenn,

Pressure outlet is definitely the best choice in this case. As for symmetry or wall, there should be no difference. Symmetry is more strict than slip wall, as it also imposes zero normal gradient, but in your case there are no normal gradients before the shock. And it will be the same even for viscous case (if you'll keep slip wall, or zero shear stress wall in terms of Fluent).