open TELEMAC-MASCARET The mathematically superior suite of solvers

I am sorry but it did not upload the remaining files before...

and the last, the 3D results file (I run a short computation so the file is smaller and I hope that I can upload it!)

Here the 3D result file (zipped) if it finally upload them...



Regards

I am sorry for these posts without content, but the file was always not accepted. I run a simulation of 12 hours saving just the last time step, I hope it will be uploaded. Maybe you should make a "sign" about the conditions of acceptance of the file to share...we could save a lot of useless posts...

About the case, I tried further also with a colleague, expert in TELEMAC-3D and POSTEL-3D, but we always get the same error: NO INTERSECTION WITH THE COMPUTATIONAL DOMAIN. I cannot understand why it happens...

Thank you.
Regards

Hello,

Can you please check my files?
I still get the error that my coordinates are out of the domain, but this is not true.
I attach a zipped file where you can find the files necessary to run POSTEL-3D. In my case (an existent bay in Brazil), I need to run for around 10 days to get steady state, but in this case I just attach results for 6 hours, otherwise the TELEMAC-3D result file would be too large.





Thank you in advance.

Best regards

Hello,

Please find a way to send the files to: This email address is being protected from spambots. You need JavaScript enabled to view it., or send an ftp address where we can download the files.

Regards,

JMH

Hello,

it worked with the latest version v7p0r1 just regarding POSTEL-3D (with Telemac-3d I use v6p3r0).
The problem now is that, no matter which PRINTOUT PERIOD FOR CROSS SECTIONS I use, I got every time in the results just 1 time step and a "zero time step", therefore, all results are equal to zero.

Has it ever happened to anybody?

I tried to change the number of the number of the interval of time steps and also to enable NUMBER OF FIRST RECORD FOR CROSS SECTIONS, but I always have the same result...

Afterwards, I tried to avoid to tell PRINTOUT PERIOD FOR CROSS SECTIONS and NUMBER OF FIRST RECORD FOR CROSS SECTIONS and it worked, I got for each section the same time steps saved for telemac-3D (each day for 10 days).

There is a way to impose less time step to save in Postel?
Also, the results would be affected if I used 24 processors in Telemac and scalar mode in Postel?

I attach the postel cas file and the results of 1 section at 10 days (which look a lot strange...).







Thanks in advance.

Best regards

Hello,

I looked at the first example you sent. The period should not be set to 14400 but to 1 (it corresponds to the records in the results file, which is 2 in your example). Then you get two files, indexed with 1 and 2, every one with a cross section at a different time. A number of variables are 0 (NUX, NUY, NUZ) but the velocities are not. The problem is that we see nothing because of the aspect ratio. You should set a :

DISTORSION BETWEEN VERTICAL AND HORIZONTAL

in your case it must probably be a large value, like 100 or more.

With best regards,

Jean-Michel Hervouet

Thank you for the suggestion. I tried with 10 and it saved me the first and the last time step (at 10 days). I also tried DISTORSION BETWEEN VERTICAL AND HORIZONTAL equal to 100 or 1000, but I always get really strange results... my mesh is not like is shown in postel...







Thank you anyway.. I really don't understand where the mistake is...

Hello,

This becomes a mystery. I tried with the first file you sent, with a distorsion of 100, and I get two files representing the cross section at two different times, and I can see them with Fudaa, the mesh is a regular grid of quadrilaterals, each split into 2 triangles. However Tecplot is not able to show it correctly, I will inquire about that, but there could be a problem. A new feature of Telemac-3D and Postel-3D is that they now put a date in the selafin format, if a post-processor does not look at that it may be a problem.

With best regards,

Jean-Michel Hervouet