open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear all,
I read the Kenue manual and I have to run my first simulation on a square domain. So I just created a close polygon, and I created a new T3mesh object. I dragged the closed polygon into outline and I generated a mesh with edge=35. I really need to use a resolution of the triangles of about 35 meters for what I am trying to do otherwise I cannot resolve some morphodynamics pattern. The problem is that the mesh has two elongated areas in which the resolution is way smaller and I do not understand why. Things are getting better with edge = 40 or 45 meters but its not what I want. I also tried to give the external polygon the value 35 and drag it in the density field but nothing changes. Any suggestion please?
I attach the picture of the mesh and the workspace with mesh and close polygon.
thanks
V.

Hi V,

If you turn off the "Auto Smooth Mesh" option, the resulting mesh is more regular.

Alternatively...
If you want a "perfectly" regular mesh, create a regular grid and save that as a t3s.

To cover the same domain as you supplied...

1. "File->New->Regular Grid..."
2. Set: X Count=215, Y Count=161, X Delta=35, Y Delta=35
3. push the "Initialize" button
4. Select the "new Regular Grid" object in the WorkSpace and using
"File->Save Copy As..." save the grid to a mesh by selecting "2D T3 Scalar Mesh" as the "Save as Type".

The result is attached.

Martin

Thanks for your answer. But i do not want a regular triangular mesh because I do not want any preferential direction. I want a triangular irregular mesh with an (approximately) constant edge length and with the triangle edges having an as random as possible direction. So turning off auto-smooth is not a good thing because it leaves all the triangles orientated in the same way. I actually reached the result I was looking for by using the attached polygon. I think there is some sort of bug on the mesh generator because this is exactly the same polygon as the one I attached above but its nodes are ordered anticlockwise instead of clockwise. The clockwise polygon leads to a mesh with the two undesired coarser parts as in my previous post. With the anticlockwise polygon the mesh is good. I noticed it gets even better if I add some random soft lines inside (note: this does not help so much for the clockwise polygon because where I no not have soft lines i still have the coarser stripes).
My mesh is done but somebody should maybe take a look to the anticlockwise/clockwise bug if you have time.
thanks again for your answer.
cheers
V.

OK, Thanks for pointing out the problem.
I understand the issue and potentially the solution.

Something for the next version if I get a chance.

Cheers... Martin