Bad Results (Aircraft External Aerodynamic)
 

Hi , I'm now simulating some conditions of the flight envelope of a canard aircraft but I'm not getting any closer to the results that I wanted to achieve.

For example: Take off Condition (Sea level)
Lift required = 6474.6 N (Half-Body)
Lift Simulation = 1835 N (Half-Body)

the flow conditions are:
rho = 1.225 kg/m³
P = 101325 Pa
AoA = 2 deg
V = 38.5833 m/s = 75 knots.

the setup that I'm using is:
3D
Steady
Gas
RANS
Segregated-Flow
Turbulent
Constant Density
K-Omega (SST - Menter)

Turbulent Viscosity Ratio = 10; (Both initial and Velocity Inlet boundary cond.)
Turb. Velocity Scale = 3.85833 m/s
Turbulent Intensity = 0.01

The simulation converge but to a very bad value (around 2e-2)
despite the mesh is a "good" mesh (once i've already simulated it at other conditions giving me closer results to the real ones)
I don't really know what to do since i've already checked all the inputs and so far haven't found any error in the setup.


Any suggestions?

I wouldn't call what you show a 'good mesh'. are you looking at drag? lift? elaborate a bit please.

I have limited computing resources so that mesh (approximately 7.5 million cells) is the best that I've reached for good time of simulation (15 hrs). I'm looking to get lift and drag.

What other conditions did you simulate it at? It is very likely you aren't resolving the physics anymore with different boundary/initial conditions than your "close" results simulations.

I have simulated it at cruise condition:
P = 69820 Pa
T = 268.3380 K
V = 87.45556 m/s
M = 0.2663
Rho= 0.9046 kg/m³

reached L = 6040 N (using spallart Turbulence model)
The point is: I've only changed the velocity components and the pressure from the cruise simulation (i've used the same macro only changing those inputs).

1. You've changed your turbulence model which is more than likely causing differences in the calculated values and your issues with your residuals. Check your near wall treatment or lack their of and if you are properly resolving the near wall behaviors.
2. Your nodalization seems to have a huge amount of trets which might not be appropriate for this flow.
3. Changing the velocity and pressure can greatly affect your results (i.e. ability to resolve flow behavior). You can't just change them significantly and expect the same mesh to work. For every significant change you have to ensure your mesh is correctly resolving flow behavior or physics. Every change you've made through your turbulence model or velocity/pressure conditions have very likely made your meshing invalid for the conditions you're simulating now.