Hi,

I am simulating a super- and hypersonic air flow in 2D through a rectangular channel with a cavity at the bottom wall near to the inlet with FLUENT 6.1. Now I am using k-epsilon model and set the density of air as IDEAL GAS. I started the simulations from a low inlet velocity (20m/s) and I obtain reliable results, however as soon as I increase the inlet velocity to about 100 m/s I obtain backflow and it does not converge. After that I changed the under relaxiation factors but no real difference. If anybody need more information I can send it! Thank you, for any response!

tayfun

hi

first of all velocity inlet of 20 and 100 m/sec is subsonic! secondly when you are setting velocity inlet greater than Mach=0.3 you need other boundary conditions than that of velocity inlet. Try pressure far-field (read the manual) and for the outlet fixed pressure or pressure inlet-pressure outlet. Good luck!

Thank you for your fast response! I know the 20 m/s - 100m/s are not supersonic speeds. I tried to reach to 510 m/s - 800m/s step by step because I obtained everytime at those higher velocities errors. I am Sorry, I forgot to mention that. The problem is: if I want to simulate it by pressure inlet-pressure outlet I need the pressures, which I do not know. Secondly I thought the pressure far field is mainly suitable for "open flows", anyway, I tried it but no improvement. Now I decided to define the inlet mass flow rate by estimation of the density (pressure).

best regards

if the problem is such that the outlet flow is completely supersonic, the pressure that you set at the outlet pressure boundary is ignored, and the pressure is determined from the supersonic flow condition. The inlet pressures are failr easy to determine from the inlet mach number, the pressure and temperature ratio equations for isentropic flows can be found from any gas dynamics textbook.

nandu

cfd