'Lockheed SR-71' simulation project by roozbehmousavi


I created a new simulation project called 'Lockheed SR-71':


More of my public projects can be found here.


Geometry from: https://grabcad.com/library/sr-71-blackbird-2


@roozbehmousavi - I think the inlet face of the computational domain is too close to the aircraft, what do you think? Will you continue working on that project? Would be great to see a compressible flow sim of it.


@dheiny - Hi Mr. Heiny. In fact, I planned to perform a supersonic (M = 3) simulation and as you know, in the supersonic flows the region outside the Mach cone does not receive any information from the inside (You can not hear a supersonic airplane before it passes you and you get located within its Mach cone). As because of the sharp nose of this airplane, the airplane is supposed to be located downstream of an oblique shock wave (conical) initiated at the nose, the inlet boundary condition can be started even at the nose. In this way you can save some computational costs as well!

But here the problem is that I can not get a stable solution for the supersonic flows including shock waves. In fact I started with a simple nozzle (https://www.simscale.com/projects/roozbehmousavi/de_laval_nozzle) supposed to make a supersonic flow at the outlet or forming a shock wave at the diverging part. Although the solution was not stable at all using a steady solver, but I could perform a stable simulation using an unsteady solver for the case with no shock wave. But for the higher outlet pressure corresponding to the formation of a shock, the solution does not have a long term stability even with dt = 1e-7s !

So, if I can resolve the nozzle problem, I come back to this airplane for sure. It would be great if I could receive some suggestions from you. Thanks



That’s a nice Aircraft, but for a High Super-Sonic flow , you will have to use the " Density based" solver in the ‘Compressible’ analysis type. I see that the current runs were with the ‘Pressure based’ solver which is Not designed for such high speed flows.

The ‘density based’ is only available with a ‘Laminar’ turbulence model (for now). So you may go with a Laminar assumption and try with that.

There is some description about the solver in my earlier forum post: Trans-Super Sonic Compressible Flow projects

To select it see the snap below:

Also, you can use non-dimensional values for the variables (velocity, pressure etc).