Hello, If there is some one who are unterested in the discretisation of navier-stokes equation with finite volume method and fnite difference schemes and compressible fluid, he can contact me and discuss a bout this subject, I was developped some thing on this subject.

Hi mehdi.

I'm actually testing a CFD software (FLOWORKS 99) for a company who wishes to use it.

I'm comparing measured mass flows through pipes, with the computed ones by FLOWORKS 99.

The software offers two alternatives to run the solver : the incompressible one or the compressible one.

It appeares that both alternatives show completly different results : while the incompressible option is run, it shows a mixed subsonic-transonic-supersonic velovity field.

My question to 2u :

Is the fact that the flow is computed with a code either compressible or incompressible, while the flow is mixed compressible-incompressible, false ?

Is this the source of computationnal error ?

How can I deal with it ?

Or is the software simply unable to compute mixed flows ?

Thanks. .

(1). If the code is incompressible, then the density is constant. You should not get subsonic-transonic-supersonic velocity. (2). Some code for low Mach number flow can handle variable density problem. And some even can compute supersonic flows. So, read the instructions first to make sure that the incompressible code is really incompressible. (3). The simple way to check the flow field is to print or plot the Mach number, or contours. (4). If the incompressible code can handle the variable density case, then you can try a Mach 0.3 case using both codes to check the computed results. The results should be the same. (5). By the way, it is not a good idea to run incompressible code for transonic flow problems. Keep it for strictly incompressible flow or flows with Mach number below 0.3 (if it can handle variable density case).