open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear all,

I run a simulation in a canale 40 x 1 m, the D50 = 3 mm, I choose the Nikuradse law Set the FRICTION COEFFICIENT =0.007, ROUGHNESS COEFFICIENT OF BOUNDARIES =1.125e-4. The canal slope is 0.4%.

For the velocity profile, I choose: velocity proportional to square root of depth.

But the velocity in inlet is always enenven distributions.


anybody could give some suggestions to it?

Here is the figure and .cas

Hello

first, the mesh is to be improved for the upper corner.
secondly, the velocity profile 4 means that the profile will be proprtional to the water depth square root, so if you have an uneven bathymetry you will have an uneven velocity.
moreover, it seems that these are not the only problems in your case. be sure that you are not simulating a case with an supercritical inlet. furthermore, after the mesh being improved, run the case in its simplist condition (without bed friction; without lateral friction, ...) and see if you can retrieve nice velocities and then go gradually with complexity.

I hope that this is helpful
with my best regards

Riadh

Dear Riadh,

Thank you for your information, I have testing it without friction, the results are more or less the same.

I will remesh the geometry, and I use Blue kenue to do the mesh, the alwys have a little bit problem in the corner.

I think this is the telemac problem, or maybe I missed something. because I also try it without any slope, it is works ok.

but when I add a relative high slope e.g 0.4%. then it is performance bad in the inlet.

No matter which mesh has been used.

I also try all kinds of inlet boundary set, it is also doesn't work.

Hi
How is the bottom on the entry?
could you post your file in a zip?

regards

Dear C.coulet,


Here is the Geometry, boundary conditions and **.cas files




Thank you so much

Hi

I'm not sure but I think the water height is very small.
You impose directly the discharge, so it could create an initial small oscillation.
Did you try without the lateral friction, eventually with another bottom friction law.
I'm not sure but for Nikuradse, it seems to me that the fiction coefficient is the roughness height.
Then 9mm for 7cm of water depth, the ration seems to be high...

regards

I also tried with strickler coefficent with range between 45-55 (Law 3).



also conside without lateral friction, The simulation results are no much difference. Maybe I should try the inlet with velocity.

For the Nikurase friction coefficient. I used it based on the function = 3*D50

I also tried the range with 0.001-0.009. It is no much difference.

I also tried with the input in the inlet is velocity, it is the same.

High mesh density is also tried, no effects at all.