[FMDenaro]

- what about your dy+ and dz+ close to walls?
- the plot of the static pressure is for the periodic part?

[sbaffini]

Sorry, forget about L, i don't know where did it come from... i meant AR=W/H.

The viscosity is what i called mu.

The periodicity has to be set via Text User Interface (yes, between inlet and outlet): gr -> mo -> mp

Once it is set, you can define your pressure gradient in the relative panel (the dp/dx i defined above)

Once you set everything and start by the perturbed RANS solution, an initial development time is required, over which the solution will adapt (slowly, because of the periodicity) to the computational setup. You can see that by monitoring the volume average (over the full domain) of the velocity magnitude squared. Once this quantity starts oscillating around some mean value you are done, you can start collecting the statistics.

[Far]

Originally Posted by FMDenaro

- what about your dy+ and dz+ close to walls?
- the plot of the static pressure is for the periodic part?

dy+ and dz+ are similar to Y+? or there is any difference?

Yes this is the plot of static pressure for the periodic part.

[FMDenaro]

Originally Posted by Far

dy+ and dz+ are similar to Y+? or there is any difference?

Yes this is the plot of static pressure for the periodic part.

y+ is the local position in the direction normal to wall, expressed in wall units, that is y+= u_tau*y/ni = Re_tau* y/h. Not that h is characteristic length used in Re_tau.

dy+ and dz+ are the mesh sizes expressed in wall units. Close to the walls you need 3 -4 cells within y+ and z+ =1.

The plot of the pressure seems to me not to fulfill the periodic condition in streamwise direction