Negative pressure error. Supersonic wing simulation.

Rodrigo Martín
Rodrigo Martín Altair Community Member
edited February 15 in Community Q&A

Hello. I am trying to model supersonic flow over a wing in Simlab, to later study aerodynamic heating and optimize the wing based on this.

I'll provide some background over the steps I took so far for the simulation.

Import wing tunnel geometry (solid cube with the wing removed)

Perform surface mesh, to later CFD mesh with boundary layer. 

Material: Ideal gas

Boundary conditions:

Inlet: 500 m/s (x) 20 m/s (z). 101325 Pa. 273 K.

Outlet: 101325 Pa

Symmetry

Slip for the wind tunnel walls

No slip for the wing wall.

Initial conditions; 

Eddy viscosity: 0.0001

Pressure: 101325 Pa

Temperature: 273 K

Velocity: 500 m/s (x) 20 m/s (z)

I attach the log file and the database file just in case it is of any help.

The simulation works fine for Mach 1.1, but as soon as I increase the inlet velocity to Mach 1.5, the solver crashes. In the log file, the error seems to come from a negative pressure arising. 

Is something wrong with my boundary/initial conditions? Maybe should I use far-field conditions instead of inlet/outlet? I have performed some attempts at it but seemed to get similar results, but I got to say I do not really understand that BC. 

Thanks beforehand. Any help is appreciated.

 

Tagged:

Answers

  • acupro
    acupro
    Altair Employee
    edited February 15

    First - you'll need to use the mildly compressible or supersonic solver for this which from the Log file I see you're already doing - as in this tutorial:
    https://help.altair.com/hwcfdsolvers/acusolve/topics/tutorials/acu/acu_2400_intro_sl_r.htm

    That being said - external supersonic flows are not good applications for AcuSolve.  The target and tested applications for the compressible solvers are internal flows (compressors, turbos, etc) - not external.  Your experience may vary, since it is really beyond the scope of the solver.  You'll need very fine mesh - Y+ on the order of 1 with a growth rate around 1.1 to 1.2, fine resolution in the volume including the wake region, etc., and that particularly as shocks start to form.  Maybe smaller time increment, etc.

Categories