Problem with CALC_CHAMP

asai · 2021-01-28 22:58:25

Hello, I'm on the first example of Jean-Pierre Aubry's book and I'm getting the following warning:

stat = CALC_CHAMP(reuse=stat,
RESULTAT=stat,
PRECISION=1.E-06,
CONTRAINTE=('SIEF_ELNO', 'SIPO_ELNO', 'SIPM_ELNO'),
FORCE=('REAC_NODA', ),
CRITERE='RELATIF',
INFO=1,)

!----------------------------------------------------------------!
! <A> <CALCULEL4_8> !
! !
! Le résultat stat n'existe pas. !
! !
! !
! Ceci est une alarme. Si vous ne comprenez pas le sens de cette !
! alarme, vous pouvez obtenir des résultats inattendus ! !
!----------------------------------------------------------------!

What does it mean? In the previous command, everything seems correct with stat:

stat = MECA_STATIQUE(MODELE=model,
CHAM_MATER=material,
CARA_ELEM=elemcar,
LIST_INST=liste,
EXCIT=(_F(CHARGE=ground,
TYPE_CHARGE='FIXE_CSTE',),
_F(CHARGE=selfwght,
FONC_MULT=selfw_m,
TYPE_CHARGE='FIXE_CSTE',),
_F(CHARGE=cc,
FONC_MULT=cc_m,
TYPE_CHARGE='FIXE_CSTE',),
_F(CHARGE=cr,
FONC_MULT=cr_m,
TYPE_CHARGE='FIXE_CSTE',),
),
OPTION='SIEF_ELGA',
SOLVEUR=_F(RENUM='AUTO',
NPREC=8,
ELIM_LAGR='LAGR2',
STOP_SINGULIER='OUI',
TYPE_RESOL='AUTO',
ACCELERATION='AUTO',
LOW_RANK_SEUIL=0.0,
PRETRAITEMENTS='AUTO',
POSTTRAITEMENTS='AUTO',
PCENT_PIVOT=20,
RESI_RELA=1.E-06,
GESTION_MEMOIRE='AUTO',
MATR_DISTRIBUEE='NON',
METHODE='MUMPS',),
INFO=1,)

Le système linéaire à résoudre a 270 degrés de liberté:
- 246 sont des degrés de liberté physiques
(ils sont portés par 41 noeuds du maillage)
- 24 sont les couples de paramètres de Lagrange associés
aux 12 relations linéaires dualisées.
La matrice est de taille 270 équations.
Elle contient 2481 termes non nuls si elle est symétrique et 4692 termes non nuls si elle n'est pas symétrique.
Soit un taux de remplissage de 6.436 %.

After the warning, I get this error:

masse = POST_ELEM(TITRE='masse',
MASS_INER=_F(GROUP_MA=('topbeam', 'mast', 'massN'),),
MODELE=model,
RESULTAT=stat,
PRECISION=1.E-06,
NUME_COUCHE=1,
NIVE_COUCHE='MOY',
MODE_FOURIER=0,
GEOMETRIE='INITIALE',
CRITERE='RELATIF',
INFO=1,)

!----------------------------------------------------------------!
! <EXCEPTION> <JEVEUX_26> !
! !
! Objet inexistant dans les bases ouvertes : &&MEDOMP.NUME_ORDRE !
! l'objet n'a pas été créé ou il a été détruit !
! Ce message est un message d'erreur développeur. !
! Contactez le support technique. !
!----------------------------------------------------------------!

Destruction du concept 'masse'.

How can I solve this?

Thanks in advance

Last edited by asai (2021-01-28 23:01:35)

bpaillard · 2021-01-29 17:00:14

Hi,
I'd start from a test case delivered with aster or salome_meca ; there are hundreds of them, and they are functional instantly.

Two options : load them from salome_meca, or find the files in some folder called share/aster/tests. You can use the command 'find . -name *comm' to find all comm files in your aster or salome_meca install directory.

asai · 2021-01-29 18:47:33

Hi,

The test cases are working perfectly. I ran the example forma01a, which has some commands very similar to the ones I'm having trouble, and I didn't get any error.

jonas loenartz · 2021-10-13 15:25:01

I am having the exact same issue. I did the tutorial exactly like it is described in the book, but I got the describes error-message.
Attached to this post is a zip-file containing the mesh, the .comm file and the .mess file. It would help me a lot to find a solution to this error.
Thank you

Jonas Lönartz

Attachments:
frame1_case.zip, Size: 14.64 KiB, Downloads: 180

jeanpierreaubry · 2021-10-14 13:16:41

hello

there is something strange

with your file
i run into trouble with version 15.2, with a null pivot

with the master files (stored on my komputer) used as a check up
it runs pretty well with version 15.2

getting a closer look at your mesh i can spot at least two doubles nodes on the top beam
nodes 4 and 6 sitting at the same position
the same with 5 and 8
at these nodes the elements are NOT connected to each others

jean pierre aubry

Last edited by jeanpierreaubry (2021-10-14 13:17:17)

jonas loenartz · 2021-10-20 12:29:21

Hello M. Aubry,

thank you very much for your reply. I rebuilt my mesh from scratch and used the "coherence" option in gmsh, to eliminate the possibility of duplicate nodes.
The error in the .mess file still persists.

reslin = CALC_CHAMP(reuse=reslin,
                      RESULTAT=reslin,
                      PRECISION=1.E-06,
                      CONTRAINTE=('SIEF_ELNO', 'SIPO_ELNO', 'SIPM_ELNO'),
                      FORCE=('REAC_NODA', ),
                      CRITERE='RELATIF',
                      INFO=1,)
   !-----------------------------------------------------------------!
   ! <A> <CALCULEL4_8>                                               !
   !                                                                 !
   !  The result reslin does not exist.                              !
   !                                                                 !
   !                                                                 !
   ! This is a warning. If you do not understand the meaning of this !
   !  warning, you can obtain unexpected results!                    !
   !-----------------------------------------------------------------!

I am using the standalone-version 14.6 on Ubuntu 20.04 , which is, by my knowledge, the latest stable version.
My next step was, to pin down the source of the error.
I built (as far as possible) the exact same case with salome_meca and tried to solve the problem using aster-study and suddenly it worked!
The results showed the expected behavior, so I my mesh seems to be ok now.

I exported the aster-study .comm file, changed the settings in ASTK and the UNITE-options in the exported .comm file to match, ran the case using the same mesh I used with aster-study, but again, the described error message occurred.

Now I'm wondering if there is a problem with my installation of code aster. According to the setup-log, everything went well.
I ran the complete list of test-cases. Most were ok or issued warnings, but quite a bunch of them ended with errors, which i don't think is supposed to happen.
I again added some case-files including the SM .hdf file, frame2.comm being the comm file created using SM.It would help me alot if someone found the time to help me crack this nut. Thanks in advance.

Jonas Lönartz

PS: @M. Aubry, would it be possible, to share the master files for this case with me, so I can compare or check my installation?

Last edited by jonas loenartz (2021-10-20 15:36:30)

Attachments:
case-files.zip, Size: 85.93 KiB, Downloads: 148

jonas loenartz · 2021-10-20 19:34:02

OK, it definetely is my standalone installation of Code Aster causing the error. I ran the ASTK version shipped with sm_2020 as plugin, loaded the exact same .astk file and ran it and it immedeately worked. At least I now have a working version of ASTK and I can sort out the other issues another time.

Many thanks to anyone who invested his/her time in my problem, it is much appreciated. The aster learning curve is indeed steep, but so far it has been a blast!

#1 2021-01-28 22:58:25

Problem with CALC_CHAMP

#2 2021-01-29 17:00:14

Re: Problem with CALC_CHAMP

#3 2021-01-29 18:47:33

Re: Problem with CALC_CHAMP

#4 2021-10-13 15:25:01

Re: Problem with CALC_CHAMP

#5 2021-10-14 13:16:41

Re: Problem with CALC_CHAMP

#6 2021-10-20 12:29:21

Re: Problem with CALC_CHAMP

#7 2021-10-20 19:34:02

Re: Problem with CALC_CHAMP

Board footer