open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

You seem to give only 16 x 3 values of tracers at the sources, could it be this ?

Regards,

JMH

Wow, that's embarrasing haha!

Thanks a lot!

/schmirre

Hello,

Now I get the same problem in v7p2p0, I get the error message below in a simulation with 18 tracers, 14 sources and 2 boundaries. I got this to work on 7p1r1 with changing MAXVAR and recompile. But I don't get this to work on the later version. Also, as I mensioned somewhere else, in version 7p2 you have to put the f-file in one file directly in the same folder and not in fortran-folder and this is a bit disturbing.

Thanks for all help,
Charlotta

^^^^^^^^^

********************************************
* LECDON: *
* BEFORE CALLING DAMOCLES *
********************************************

*********************************************
ADRESS GREATER THAN NMAX = 300
TOO MANY VALUES OF TYPE : 2 DECLARED.
STOP OF DAMOCLES AT KEY-WORD NUMBER: 84
*********************************************

PLANTE: PROGRAM STOPPED AFTER AN ERROR
RETURNING EXIT CODE: 2

job aborted:

Small update, I changed MAXKEYWORD = 400 to 4000 in declarations_telemac3d.f: in v7p1r1 and recompiled which worked well. I cannot fint MAXKEYWORD in the f-file for 7p2r2.

Greetings from Sweden,
Charlotta

Well, I'm also onto this topic again as I'm trying to run a model with 20 sources and 22 tracers. I am still running v7p1v1.

I got it to work with 19 sources and 19 tracers, but as I move to 20 of each the simulation freezes at:

APPEL DE PREADV
@STREAMLINE SCARACT: THE NUMBER OF TRACEBACK INTERFACE CROSSINGS IGEN < 99.

I did the following changes:

• Changed MAXKEYWORD from 400 to 4000 in declarations_telemac3d.f and recompiled.
• MAXIMUM NUMBER OF SOURCES = 50
• MAXIMUM NUMBER OF TRACERS = 50

Very thankful for your ideas on how to solve this!

More greeting from Sweden,
/Schmirre

Hi,

Just an idea: I see from the error message that you use Advection scheme 1 (method of characteristics). What happens if you change to say MURD scheme (14)?
Furthermore, the method of characteristics is not mass conservative and should therefore be avoided for tracers...

Best regards (also from Sweden )
PL

Hi PL!

I already have Advection scheme 14 for all tracers. What part of the error message tells you it is scheme 1?

/schmirre

"@STREAMLINE SCARACT". If I am not wrong this message occurs when, as it says, a single characteristic path crosses more than 99 boundaries. It is therefore more likely to occur in parallel with many cores and subdomains. I got the same error once and solved it by either switching to scalar mode (1 proc) or using MURD scheme.

But this would mean that you do use the MoC for at least velocities? Otherwise I don't know why you get this error message...

Good luck!
PL

Thank you PL.

I changed the SCHEME FOR ADVECTION OF VELOCITIES also to 14, but the same error appears. It freezes after "APPEL DE PREADV".

Running it in scalar also results in the same error.

I also tried advection scheme 4, explicit + MURD scheme N, without luck..

/schmirre

OK, then I was maybe wrong the the method of characteristics, sorry for that. The problem might then be linked to the number of sources/tracers but I have no experience on the matter unfortunately...

Good luck!
PL