open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi community,

I am trying to use runoff coefficients (given in the mesh with blue kenue). For this, I changed the following files :


- in bief, the fonstr.f (line 97 to line 122 ), in order to read the runoff coefficient in the mesh. See in attachment : New_fonstr.f ;

- the subroutine prosou.f (line 402), which is call by the parameter file. In the subroutine prosou.f we use the variable RUISS define ine the fonstr.f. See in attachment New_prosou.f

Running t2d i have the error as you can see in the attachment : Error1.png

Any help will be greatly appreciate.

Thanks for advance

JP

Hello,

RUISS is not an argument of the subroutine and thus must not have an intent. For the rest look at post #13676.

Regards,

JMH

Actually, I forgot a comma (virgule) in the variables declaration of prosou.f. I corrected it but it still doesn't work. The problem is that the kind of error is not specified. We only have the error message :

"runCas: fail to run" (see in attachment Error2.png)

Thanks for your help

JP

Hello,

You can try DEBUGGER = 1 to see at least in which subroutine it crashes (and try first in scalar mode if yiu are in parallel).

Regards,

JMH

Hello,

Using DEBUGGER = 1, i can see the problem is comming from :

"CALLING PROSOU"

But I don't understand why is comming from prosou.f because in the parameter file i use :

FICHIER FORTRAN = 'HyetogrammeTriangulaire_Ruiss.f'

So, I thought the prosou.f was not call because it is replaced by HyetogrammeTriangulaire_Ruiss.f.

Is the prosou.f used even when using the keyword 'FICHIER FORTRAN' ?
Do I need to add my variable RUISS in the prosou.f ?

Thanks a lot,

JP

Hello,

All the subroutines in your FORTRAN FILE replace the equivalent in the libraries, the psosou that crashes is thus yours as it is in 'HyetogrammeTriangulaire_Ruiss.f'. You can add messages in it to see where and why it crashes.

Regards,

JMH

Hello,

In the HyetogrammeTriangulaire_Ruiss.f, the problem is comming from this lines :

Note that we allocate RUISS in the point_telemac2d.f (In attachment, line 277, and line 1220). In this file we try to use IELMU, IELM0, IELM1 and IELM10. With IELM1 we have the error in attachment : error3

Hello,

This section is for quasi-bubble elements, and IELM1 should be equal to 11, it seems equal to -1000 for some reason. The reason is certainly that the vector FU has been allocated with STATUT=0 (arbitrary size not linked to the mesh, in this case we put a component ELM = -1000 to the vector to trigger messages such as the one you have). If you are sure of what you do you can do before :

FU%ELM=11

so that CHGDIS is not called, or simply remove this section.

REgards,

JMH