open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello everyone,

I have a divergence problem that I can't seem to solve. For a steady flow, the downstream boundary always end up shooting water inside the model at one point (see picture).
I have tried lowering the time step, higher mesh resolutions, different flow discharges, initial conditions with constant elevation, constant depth, and special... with always the same results at the boundary conditions where a rating curve is imposed.

Do anyone have an idea to solve this problem?

Thank you in advance!

Pablo

Hello Pablo

the problem is not that clear with these few explanations and without checking the model's files. Even the picture is not clear enough.
This kind of problem can come from points located at oversonstrained elements, i.e. elements with two different boundary conditions. Moreover, it a wetting/drying phenomenon is also in the vicinity of that point it could increase unstabilities.
It could be also a problem of regime change at the boundary: if the regime changes from sub to super critical, a unique boundary condition becomes not sufficient and the problem becomes ill-posed. To overcome such problem, you have to redefine the boundary (change the bathymetry, change its location, add an extra condition, etc.) or to use a Thompson boundary condition.

I hope that this helps

with my best regards

Riadh Ata

Thank you Riadh!

In this case it was a regime change.

Best Reagrds,
Pablo