open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello all,

I want to try modeling energy dissipation downstream of a spillway. The layout of the downstream reach, from the spillway to the river includes the spillway structure, a more or less confined flow on a steep slope (< 15%) and the river itself.

I though the better approach would be to have an upstream boundary downstream of the spillway where I could impose Q and H (supercritical flow with V > 10 m/s). Downstream boundary would be subcritical with imposed H.

Is telemac expected to give exact answers with those kind of boundary?
Is there any particular care to model energy dissipation adequately?
Is there any particular configuration in the steering file I should try first (solver type, friction law...)?
Since boundary conditions are imposed to nodes, should the last node be exactly at the spillway structure wall or a bit "inside" the hydraulic passage?

Thanks for your help

Hello,

I understand that your upstream boundary will be supercritical, and the downstream boundary subcritical with depth imposed. So you will get probably a hydraulic jump in between. It seems to me that in this configuration the conservation of water mass and of momentum between upstream and downstream can give the head loss, it is a classical problem of Saint-Venant equations, at least in 1D on flat bottom. So you might be right, though at first sight we are tempted to think that the complexity of the energy loss is beyond reach.
You can look at the case hydraulic_jump, which is similar to what you want to do.

With best regards,

Jean-Michel Hervouet

Thank you for the quick response.

I'm not sure what you mean by : "So you might be right, though at first sight we are tempted to think that the complexity of the energy loss is beyond reach". Do you mean that the complexity of the hydraulic phenomenon will not be adequately modeled by telemac, or the opposite?

Thanks!

Hello,

No, I think it may work, because in the end in the 1D problem with flat bottom we get the head loss just by considering mass and momentum at boundaries, whatever the complexity of the hydraulic jump, so it seems that a good representation of turbulence in the hydraulic jumps is nor important.

With best regards,

Jean-Michel Hervouet