Kelvin-Helmholtz instability

pierre

Hi,

I'm tying to simulate a simple incompressible free shear layer where kelvin-helmholtz may develop (using DNS and Fluent 6.0.20). My parameters are the following ones:

*momentum thickness: 2,4e-05 m *vorticity thickness: 1e-0.4 m *(U1-U2)/(U1+U2) = 0.33 *Re = 180 (based on momentum thickness and velocity gap)

I use a udf to declare my inlet velocity profile. My problem is the fact that i do not manage to obtain a quite good Strouhal number (~ 0.033 following the stability theory of parallel flows).

The first thing I found is that Fluent do not keep the inlet velocity profil as I want, I mean that it is not kept stationnary!! (consequently, there might be a kind of parasite forcing...). Is there another way to keep my velocity profile "undisturbed"?

My second remarks is the fact, observing the instantaneous vorticity field, i fear that the subharmonic mode (i.e., the second instability mode causing pairing) might be "numerically" excited because I note that pairing occurs before vortices "seperate" from the rest of the flow, even if I force at the fundamental frequency...

anybody can help me?

cheers,

pierre