Dear All,

I am working on the turbulence simulation in an Oscillating Baffled Tube (OBT). That means flow in a periodically baffled tube. Currently, I just did 2-D simulation. I simulated the whole size tube with a sinusoidal velocity inlet profile , say, U = U0sin(wt),and a pressure-outlet. I turn on the standard k-e turbulence model. Finally, for unsteady simulation, I got a result. Though some of the flow patterns look well,but the flow varies wrongly with time marching on. I got a ridiculous result with a wrong period. Even when the flow is very slow. Anyone knows why? I wonder if it is related to the pressure term treatment in N-S eqs. When I increase the inlet velocity greatly, the flow in the tube is still periodic and symetric with a wrong period while in experiment, the flow becomes chaotic and the symmetry lost in this large Reynolds number.

Since the flow is imcompressible, the discretization of the pressure term of the N-S eqs is very important. Anyone knows how the Fluent treat the pressure term? Is it possible for me to get the source code for this part?

Many thanks in advance for any comments and advice.

Ray Jian

I think that your problem was caused by the boundary conditions which you used. Fluent does not accept the mixing boundary conditions such as velocity boundary at inlet and pressure boundary at out let.

Try to change the boundary conditions, may be it will solve your problem.

Good luck,