open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi everyone!

Actually I try to use a binary data file into T2D simulation. This binary data file contain the wave propagation results.

But when I run the model, this message error appear:

"Lit: erreur de lecture on voulait lire un enregistrement de 1 valeurs de type : CH sur le canal 24"

I tested without the key word "binary data file 1", but i have still the error.

The manual reference v6p0 said in page 17, "Les données de ce fichier seront à lire sur le canal Fortran de numéro 24". But I don't know how to check this channel.

Anyone had this error before? Have you an idea to solve my problem?

Thanks in advance.

Romain_D

Hello,

It is not clear whether you want to apply the stress due to waves from a file stemming from Tomawac, or if you implement yourself specific Fortran for reading yoour own data in the file, in the latter case channel 24 is correct except if you do coupling with Sisyphe, in which case the channel numbers are changed. The correct implementation would then be :

Beginning of your subroutine, if necessary :

USE DECLARATIONS_TELEMAC2D, ONLY : T2D_FILES,T2DBI1

INTEGER UL


then:

UL=T2D_FILES(T2DBI1)%LU

is the channel number (and T2DBI1 is equal to 24)

The error seems to be on a character string, possibly the title of a serafin file, hence the first record, it could be also a big_endian little_endian mismatch.

Regards,

Jean-Michel Hervouet

Hello JM,

Thanks for your quick answer.

I want to apply the stress due to waves from structured model (with refraction and diffraction effects). So I built a selafin file to input in T2D. I can open it in fudaa and see the stress due to waves.

I checked my fortran file, and I already have:

"USE DECLARATIONS_TELEMAC2D, ONLY : T2D_FILES,T2DBI1"
"INTEGER LU"
but not "INTEGER UL", is that the same think?

I work on V6P1, and the subroutine is PROSOU.

Regards

Romain_D.

Hello,

I tried with

LU=T2D_FILES(T2DBI1)%LU

and INTEGER UL
UL=T2D_FILES(T2DBI1)%LU

But I still have the error message.

Romain_D

Hello,

The problem occurs in subroutine find_in_sel (bief library) when you start to read the file, the subroutine tries to read only one letter of the title (then the title record is skipped) and it crashes there, so probably the file is empty or corrupt when we get to this point.

What you can check is the file T2DBI1 in the temporary folder created when you run the case, it must be your wave file.

LU is the logical unit for listing printouts, so you cannot change it, this is why I have chosen UL in my example.

Regards,

Jean-Michel Hervouet

Hello,

Thanks to spend your time help me.

I can't read the file T2DBI1 because the temporary folder is empty...

I can open my wave file with fudaa, so it's not empty or corrupt (no error message with fudaa).

I check the find_in_sel in bief. Ligne 87 is about TITLE, and the subroutine read only one letter like you said. So I analysed my wave file with anatel_pc:

* NOM DU FICHIER TELEMAC:
wave.res
*
* Titre de l'étude
test

I continue the tests and try to build my wave file differently.

Regards.

Romain_D

Hi everyone.

I finnaly chose to run à TOMAWAC model and input this results on T2D.

I've created a new topic.

Anyway thanks a lot for your help JM.

Regards.

Romain_D