Low Reynolds, compressible flow

andre burdet

Dear all,

I am currently developping a CFD code to simulate internal compressible flow, using FVM coupled with Maccormack type scheme (centered).

I am trying now to validate the implementation of the viscous part for laminar case (no turbulence model implemented yet).

Therefore I am simulating the straight channel test case. The aims is to reach the steady state solution of the flow with a constant inlet velocity (development of the parabolic profile inside the channel). Meanwhile I am encountering some problems :

example : if I am doing the simulation with a Reynolds of 35000 and Mach of 0.3, the simulation is converging. However this is not a laminar case. If I am doing the same simulation but this time with a Reynolds of 1500, the velocity at the inlet, on the wall (L.E.) (which should tends to zero) starts to oscillate more and more until it eventually blows up.

Therefore it seems that the diffusive part has a important impact on the simulation.

Has anyone any idea, comments, etc.. on what is happening ?

Andre.