Hi!
The compressible solver available in the platform is pressure-based, so it’s best for velocities under Mach 1. For transonic/supersonic simulations, a density-based compressible solver would be best, however it’s currently not available for community users.
In short, mach 3 would be very unstable and very prone to divergence using solvers available for community users - feel free to provide feedback/suggestions here!
Cheers