open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi everybody,

I know this kind oh topic has already been open before, but I didn't find my answer.

I have a model where I inject different discharges, and I got a lot of instability. My flow are on surfaces with high Stricker coefficient (typically road, with KS = 55 or).

I tried to reduce my time step in order to have a courant number less than 1 but it's not enough. I also tried to move FREE SURFACE GRADIENT COMPATIBILITY to 0.5 or 0, it's a little better but I still have oscillations. I also tried to move my VELOCITY DIFFUSIVITY to higher or lower values, but without significant effect.

I succeed to avoid oscillations by reducing my Strickler to 20, but it's note realistic.

Anyone can tell me if I did something wrong ? I attached my cas file


Thank you,

Regards,

Hello,

What about your mesh refinement around the source points?
It should not be too coarse, but too fine may be not good in a different way.

Anyway, if your results are better when decreasing FREE SURFACE GRADIENT COMPATIBILITY to 0.5 or 0, you can also try to use quasi-bubble elements for velocity: DISCRETIZATIONS IN SPACE = 12;11 (DISCRETISATIONS EN ESPACE = 12;11 in French).

Hope this helps,

Chi-Tuan