open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi everyone,

I am learning Telemac. These days, I got a problem with telemac3d, anybody can give me a hand?

At first, I used BlueKenue to generate mesh. The model ran well, but the results always showed some uncertainties at the edge of domain (the liquid boundary). I thought it might be produced by the rough mesh. Therefore, I generated a new one (higher quality) with SMS.

However, when I ran this new cas with the same parameters in telemac3d, I encountered the problem --- "Exceeding maximum iterations". I have found that the problem was from the keyword "MAXIMUM NUMBER OF ITERATIONS FOR DIFFUSION OF VELOCITIES".

Based on the manual, I have tried reducing the time step. But it seems unuseful.

The attached are some files of my cas.

Many Thanks.

Hello,

Your time step 0.05 is much too small, try larger values like 1, 10 and 30 s and look at the number of iterations of solvers.

You can :

* Change all preconditionnings that are 7 or 14 to 34. 7 and 14 are Crout preconditioning which is rather meant for EBE storage, while you have an edged-based storage. It could be the reason of your problem.

* Remove the keywords : STEERING FILE, MASS-LUMPING FOR DIFFUSION

* choose IMPLICITATION FOR DEPTH : 1.

* choose SOLVER FOR VERTICAL VELOCITY : 1

* SCHEME FOR ADVECTION OF VELOCITIES : 1

An important point is whether you have dry zones in your domain or not.

With best regards,

Jean-Michel Hervouet

Hi Jean-Michel Hervouet,

Thanks a lot. I will try what you advised.

Kindest,
Hao

Hi Jean-Michel Hervouet,

According to your recommend, my work is done, although there seems to still exist some uncertain error of my result

As I plan to use Telemac3d to model the turbulence in a 10x1x0.3 m scale , which keywords I should choose to improve the accuracy, such as PRECONDITIONINGS, SOLVERS, and VERTICAL/HORIZONTAL TURBULENCE MODEL.

Kindest,
Hao