open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I thought it was easy to add new variables for graphic printouts but I don't manage to get what I want (which are ZFCL_S%ADR(I)%P%R(J) : the evolution for each sediment class due to suspension).

-Firstly I amended the nomvar_sisyphe.f with:

DO I=1,NSICLA
TEXTE(27+I+(NOMBLAY+1)*NSICLA) = TEXT_ZFCL(I)
MNEMO(27+I+(NOMBLAY+1)*NSICLA) = MNEMO_ZFCL(I)
ENDDO

(TEXT_ZFCL(I) and MNEMO_ZFCL(I) have also been defined)

-> I only got the 4 last variables ZFCL(4->7) with wrong values.

-Secondly I expend the number of private array to 10:

-> My result file is corrupted (not readable by Blukenue).

Is the number of extra variables limited somewhere to 4?

A guideline of how to modify nomvar.f to get ZFCL_S(1:7) would be very nice.

Many thanks in advance.

Alexis

Hello Alexis,

You cannot add a new graphic variable only with nomvar_sisyphe. You have also to change point_sisyphe accordingly. At the end of this subroutine (that you will add to your Fortran file), you will find a series of CALL ADDBLO, which starts with:

CALL ADDBLO(VARSOR, U2D ) ! 01
CALL ADDBLO(VARSOR, V2D ) ! 02

These calls add pointers to graphic variables in the block VARSOR, the order is the same than names and mnemos in nomvar_sisyphe. Then VARSOR is used as a list by the subroutines that write in the resuslts file. So you need to add in VARSOR your new variables, like:

CALL ADDBLO(VARSOR,ZFCL_S%ADR(1)%P)
CALL ADDBLO(VARSOR,ZFCL_S%ADR(2)%P)

so here PRIVE has nothing to see with this. The only problem is that some ADDBLO are done in variable numbers, like NSICLA, so the rank of one variable becomes a complicated formula at the end, and it is highly advisable that you add your variables after others. Just try to understand how it is done for the existing variables (see the compatibility between nomvar_sisyphe and VARSOR in point_sisyphe) and I'm sure it will be OK with yours.

With best regards,

Jean-Michel Hervouet

Jean-Michel,

Albeit the WARNING message ([...] bad values found in objet ZFCL CL(1:7)) during the scan of the file in Bluekenue, I get the variables I want.

Thank you very much again!

Alexis