open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

Is the friction coefficient for Manning used as it's exact definition, or does Telemac convert it to ks? Looking in the friction_calc.f sub-routine both Manning and Strickler have separate cases so would I be right in assuming that Manning's n is not converted to ks?

I ask because in Telemac 2D I think all friciton coefficients are converted to Chezy except for Nikuradse, and was wondering how T3D handled friction. I've had some odd results and I want to see if I can rule out it being a result of ks.

Many thanks!

Hello,

There is no conversion to ks because Nikuradse friction has no reference to depth averaging and needs a specific and more difficult treatment (with assumption of a logarithmic profile at the second plane in the mesh).

With best regards,

Jean-Michel Hervouet

Hello,

Thank-you for your reply! So, to clarify, is Manning converted to Chezy?

Many thanks

I just want to clarify how the velocity profile from the friction is computed. As far as i'm aware:

In this case a friction coefficient is chosen (say, Manning.) Given high roughness the friction regime is hydraulically rough. For rough friction regimes for Manning, the constraint is computed from the friction velocity's relation to the friction coefficient. This is used to compute the constraint for shear forces
. The turbulence model (K-E) then provides the modelling for the flow by integrating the Reynolds equation over the vertical, with the production terms calculated form the shear forces.

If the flow is rough but Nikeradsz is used, the flow is computed from the velocity profile using ks, as seen in the image of the equation attached?

Hello,

That's it. You can look at subroutine tfond.f in telemac-3D sources.

JMH

Following friction I've also been looking into how the drag is calculated in Telemac 3D.

In Hervouet (2007) I found one section on drag under the 2D simulation segment (pg.53.) however I couldn't find anything that expanded on how drag is treated and implemented in the 3D code. As far as I'm aware, the drag is calculated on X and Y (dragfo.f), and then the term is added to the momentum equation (although I can't seem to find the subroutine for that?)

How is it applied 3D? I've looked into source.f but its description refers to the diffusion of tracers which aren't relevant to my project.

Many thanks.

Hello,

I am not sure to understand your question, drag forces are not implemented in 3D. In 3D the problem is very different because friction is applied only at the boundaries, and is imposed as an essential boundary condition stemming from the diffusion. In 2D friction is like an extra force added to all points of the domain.

With best regards,

Jean-Michel Hervouet