[tom_cfdstudent]

Hi everybody,
I need some advises to compute a flow in a turbine design. What is quite special is that i'm trying to perform an supersonic oscillating flow at the inlet. So we are in unsteady conditions.

- Here is the Bc's at the inlet:
BC group:inflow/outflow
BC subgroupe: special - oscillatory inflow/outflow (prescribed velocity)

This is a flow oscillating (sinusoidal) around 1888 m/s, with a frequency of 1s, an amplitude of about 20 m/s.

- Concerning the time integration panel:
I'm using a restart file (a steady computation for a flow velocity of 1888 m/s), a transient simulation, implicit integration type.


The problem is that I get a complete unconverged solution: very important mass flows (like 2 tons/s!!). I have several possible solutions:
- too big time step: it seems to be not better
- too much unstable flow: that's why i used a small amplitude
- better to play with the CFL: in an implicit scheme, does the CFL have an influence?

what i don't understand exactly is the process used by CFD++ for an unsteady computation: time and space discretization, what's going on exactly for each time step.

Maybe can u help me, or give some advices or ideas to resolve my problem?

thank you all!

[xiaomi]

hi,i want to know where can i download CFD++ software?thanks