[nirvananas314]

Hello, CFD Community,

I am currently working on a simulation of flow around a supersonic cascade using SU2. My goal is to accurately model the flow with the following conditions to match experimental data, thus, I need to make sure I have the correct inlet and outlet values

Inlet condition
Mach number: 1.2
Angle of attack: 45 degrees
Static temperature: 250 K
Static pressure: 65 kPa

Outlet condition :
Static pressure: 100 kPa


1/Using MARKER_INLET: When I specify the inlet conditions using MARKER_INLET, I notice that the velocity components Ux and Uy do not match the expected values based on the imposed inlet conditions. It seems that the simulation does not accurately reflect the prescribed Mach number (Ma=0.98 at the inlet). I thought because Ux and Uy are each subsonic, that marker type could be working

2/Using MARKER_SUPERSONIC_INLET: Switching to MARKER_SUPERSONIC_INLET seems to partially resolve the issue by correctly capturing a supersonic Mach number (not equal to 1.2 tho) and the exact Uy component. However, the Ux component and inlet pressure still do not align with the expected values. The flow indeed becomes supersonic at the inlet, but the inaccuracies in Ux and pressure yields a somewhat higher AoA and lower Mach Number.

You can see below some of the .cfg file I am using:

% -------------------- COMPRESSIBLE FREE-STREAM DEFINITION --------------------%

INIT_OPTION = TD_CONDITIONS
MACH_NUMBER = 1.21
AOA  = 45.0

FREESTREAM_TEMPERATURE = 250.0
FREESTREAM_PRESSURE = 65000.0
FREESTREAM_VELOCITY = (269.0,269.0, 0.0)



% -------------------- BOUNDARY CONDITION DEFINITION --------------------------%
MARKER_HEATFLUX = (Blade, 0.0)
%INLET_TYPE = TOTAL_CONDITIONS
%MARKER_INLET = (Inlet, 380.0, 160000.0, 0.71, 0.71, 0.0)
MARKER_SUPERSONIC_INLET = (Inlet, 250, 65000,269.0,269.0, 0.0)
MARKER_OUTLET = (Outlet, 100000.0 )
MARKER_PERIODIC = (Periodic_Boundary_1, Periodic_Boundary_2, 0.0,0.0,0.0, 0.0,0.0,0.0 , 0.0,0.04,0.0)
MARKER_PLOTTING = (Blade)
MARKER_MONITORING = (Blade)

Has anyone faced similar issues when modeling supersonic flows in SU2, especially regarding the proper setting of inlet boundary conditions? Any insights or recommendations on how to correctly implement these conditions to achieve accurate simulation results would be greatly appreciated.

Thank you in advance for your assistance.

[cristopher_morales]

Hi,

The MARKER_INLET in SU2 is used for subsonic flows, the initial velocity is estimated based on the free stream conditions that you are providing, I see that the mach number that you are providing is MA=1.2, so you are already assuming that the flow is supersonic at the inlet. In the CEulerSolver.cpp (https://github.com/su2code/SU2/blob/...ulerSolver.cpp) ,in line 6878, you can see the BC_INLET for subsonic. And in line 7006 you can see that the MACH NUMBER is prevented to be bigger than 1.0, because is subsonic.
/*--- Mach squared (cut between 0-1), use to adapt velocity ---*/

Mach2 = Velocity2/SoundSpeed2;
Mach2 = min(1.0, Mach2);
Velocity2 = Mach2*SoundSpeed2;
Vel_Mag = sqrt(Velocity2);
SoundSpeed2 = SoundSpeed_Total2 -
0.5*Gamma_Minus_One*Velocity2;

So if you already know that the flow is supersonic at the inlet, then you should only use MARKER_SUPERSONIC_INLET.
As you know already the mach number at the inlet, I suggest you to use MACH_NUMBER=1.2 in the INIT_OPTION = TD_CONDITIONS
MACH_NUMBER = 1.2 AOA = 45.0. Additionally, you might check the viscosity, thermal conductivity and heat capacity that you are using. If you do not provide them in the .cfg file, SU2 will use the default values:

% Free-stream viscosity (1.853E-5 N s/m^2, 3.87E-7 lbf s/ft^2 by default)
FREESTREAM_VISCOSITY= 1.853E-5

GAMMA_VALUE= 1.4
%
% Specific gas constant (287.058 J/kg*K default and this value is hardcoded
% for the model STANDARD_AIR, compressible only)
GAS_CONSTANT= 287.058
%
% Specific heat at constant pressure, Cp (1004.703 J/kg*K (air)).
% Incompressible fluids with energy eqn. (CONSTANT_DENSITY, INC_IDEAL_GAS) and the heat equation.
SPECIFIC_HEAT_CP= 1004.703

The mach number is computed based on the GAMMA_VALUE.

Best regards,

Cristopher Morales Ubal

[bigfootedrockmidget]

However, the Ux component and inlet pressure still do not align with the expected values. The flow indeed becomes supersonic at the inlet, but the inaccuracies in Ux and pressure yields a somewhat higher AoA and lower Mach Number.

How large are differences? These differences are probably due to the fact that the boundary conditions are imposed in a weak manner. You can expect deviations that depend on how well converged you are and also on the final mesh quality.

[nirvananas314]

Hello Cristopher, and thank you for your thorough response. I appreciate your insight regarding the use of MARKER_SUPERSONIC_INLET for supersonic flow conditions, and your emphasis on ensuring the correct setup of fluid properties. I have verified that my simulation uses the following fluid and viscosity model settings, aligning with standard air properties:

FLUID_MODEL     = IDEAL_GAS
GAMMA_VALUE     = 1.4
GAS_CONSTANT    = 287.05

VISCOSITY_MODEL = SUTHERLAND
MU_REF          = 1.716E-5
MU_T_REF        = 273.15
SUTHERLAND_CONSTANT = 110.4

These should indeed correspond to the conditions I aim to simulate. Given your advice, I'll continue utilizing MARKER_SUPERSONIC_INLET with my current settings, ensuring that the Mach number and angle of attack are appropriately represented in INIT_OPTION = TD_CONDITIONS.

[nirvananas314]

Originally Posted by bigfootedrockmidget

How large are differences? These differences are probably due to the fact that the boundary conditions are imposed in a weak manner. You can expect deviations that depend on how well converged you are and also on the final mesh quality.

Hello, thank you for your quick answer.

The relative error I obtain are as follows :
Ux: 6%
Uy: 0%
AoA: 4%
Mach: 4%
P_out: 0%
P_in: 7%
T_in: 4%

The residuals in my simulation appear to be well-converged, as they go from 1 to 1e-16 and are constant at that value.

I will try to run the simulation with a more refined mesh, but I am a bit perplex about the reason why one velocity component is exact and the other not.

Best Regards,