code_saturne User's Forum

Hello,

I am dealing with a 1D mesh in a duct filled with a fluid governed by Pressure BCs at the left side and the right side of the geometry (see the pictures). Top and Bottom boundaries are moving in the (OY) direction with an oscillatory imposed displacement. In this context, the simulations are performed using the ALE framework.

I did the simulation with a refined mesh which works well but I get checkerboard effect passing to the 1D mesh. I let you see the pictures of the pressure field for the two simulations at the same time step and using the same scale. The arak coefficient stayed at the default one. Do you have any proposal in order to tackle this problem ?

Thank you in advance.

Best regards,

So_nuc

Hello,

Did you adspt the time step for the CFL number? If the time step is too small the tme-dependent term I'm the Rhie&Chow filter may become too small (at least I think this is the explanation) and we may observe checkerboarding.

Otherwise we need to check your other setup/model options.

Best regsrds,

Yvan

Hello Yvan,

Thank you very much for your fast reply and your help ! It indeed works with a greater time step value: the Courant number at a reference time step equalled about 1E-5 before and now equals about 1E-4.

Thanks again and have a nice day.

Best regards,

So_nuc