Page 3 of 13
Re: turbine modelling
Posted: Thu Jun 21, 2012 12:05 am
by Yvan Fournier
Please look at the files indicated in the message (listing and error*) for both domains.
They should contain more details.
Re: turbine modelling
Posted: Thu Jun 21, 2012 8:36 am
by stage75
This is what gives the file "listing". Apparently the code did not take into account the coupling Saturn if I understand
Code: Select all
Applications accessibles grâce à MPI :
--------------------------------------
0; type : "Code_Saturne 2.0.4" (this instance)
nom de cas : ""
rang racine : 0 ; nb_rangs : 1
1; type : "Code_Saturne 2.0.4"
nom de cas : ""
rang racine : 1 ; nb_rangs : 1
Couplage Code_Saturne :
id de couplage : 0
nom local : "$q���"
nom d'application distante : ""
numéro local : 1
numéro d'application MPI : 1
rang racine MPI : 1
nombre de rangs MPI : 1
Couplage Code_Saturne 0 : initialisation de la communication MPI ...Signal SIGTERM (terminaison) reçu.
--> calcul interrompu par l'environnement.
Pile d'appels :
1: 0x7f8c6d1ff0ec <+0x40ec> (mca_btl_sm.so)
2: 0x7f8c71114f9a <opal_progress+0x5a> (libopen-pal.so.0)
3: 0x7f8c728b1885 <+0x38885> (libmpi.so.0)
4: 0x7f8c728d286e <PMPI_Intercomm_create+0x1de> (libmpi.so.0)
5: 0x7f8c7315514a <fvm_coupling_mpi_intracomm_create+0xfa> (libfvm_coupl.so.0)
6: 0x7f8c741bd3b7 <cs_sat_coupling_all_init+0x907> (libsaturne.so.0)
7: 0x7f8c74088345 <cs_run+0xc5> (libsaturne.so.0)
8: 0x7f8c74088da5 <main+0x1f5> (libsaturne.so.0)
9: 0x7f8c71cf6c4d <__libc_start_main+0xfd> (libc.so.6)
10: 0x40bf29 <> (cs_solver)
Fin de la pile
Re: turbine modelling
Posted: Thu Jun 21, 2012 9:09 am
by stage75
Another question about the rotation, I set the rotation angle "omega-y" in the subroutine "usini1.f90", but I don't know whot I should defiened my rotation axis?
Re: turbine modelling
Posted: Thu Jun 21, 2012 9:14 am
by Yvan Fournier
There is a listing in both domains. You need to check both (as well as error* files).
Also, it is strange that "nom local" does not appear correctly in the coupling.
The fist thing you should check is your "runcase" file. I post an example runcase from another rotor/stator/case (path names and mesh names should be different, but the rest should be similar).
Regards,
Yvan
Re: turbine modelling
Posted: Thu Jun 21, 2012 9:38 am
by stage75
I changed my "runcase" as in "rotor / stator" that you passed me. I added my two meshes and I changed "d" by [d1,d2]. Then I put my two meshes in "MESH" file. (THe is my runcase attached)
So since the two meshes are in the menu "MESH" they necessarily same path!!?.
Re: turbine modelling
Posted: Thu Jun 21, 2012 10:41 am
by Yvan Fournier
Yes, both mesh files must be in the MESH directory of the study structure (the same as for a regular calculation). Only the file names are different for each domain.
Do things improve with this setup ?
Re: turbine modelling
Posted: Thu Jun 21, 2012 10:57 am
by stage75
Yes I still have the same error.
Just to clarify: mesh 1 has a name :'Mesh_1-TURBINE.med" and the second has a name "Mesh_2-DOMAINE.med"
Re: turbine modelling
Posted: Thu Jun 21, 2012 12:27 pm
by stage75
In fact, concerning the boundary conditions, I defiened the boundary conditions of my first mesh in subroutine "usclim.f90" localised in "SRC.1" then I defiened the Boundary conditions of my second mesh in "usclim.f90" in "SRC.2"? is this right??
Re: turbine modelling
Posted: Thu Jun 21, 2012 1:55 pm
by Yvan Fournier
Yes, this seems right.
Do you have the same error in listing.1 and listing.2 ?
Re: turbine modelling
Posted: Thu Jun 21, 2012 2:38 pm
by stage75
Apparently the problem comes from my definition of boundary conditions.
After some modifications on subroutines, i have this errors ( it's a little clearer now)
Listing 1
Code: Select all
Code_Saturne : /home/itouche/Téléchargements/tmp/installer/ncs-2.0.4/src/base/cs_selector.c:242 : Avertissement
Le groupe ou attribut "26" figurant dans le
critère de sélection:
"range[26,40]"
ne correspond à aucune face de bord.
Signal SIGTERM (terminaison) reçu.
Listing 2:
Code: Select all
/home/itouche/Téléchargements/tmp/installer/ncs-2.0.4/src/base/cs_boundary_conditions.c:429: Erreur fatale.
Certaines définitions de conditions aux limites sont incomplètes
ou incorrectes :
Pour plus de détails, vérifiez la fin du "listing" de calcul,
ou visualisez le post traitement dédié aux erreurs.
Pile d'appels :
1: 0x7f229d39478a <cs_boundary_conditions_error+0x82a> (libsaturne.so.0)
Normally to define the boundary conditions (I defined the faces that make up my wall as turbine)
Salome on my faces are numbered (faces :26, 27, 28,......,40),
So i definied these faces in "usclim.f90" on "SRC.1"like :
Code: Select all
! --- Prescribe at boundary faces of color 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39 and 40 a wall for all phases
call getfbr('range[26,40]', nlelt, lstelt)
!==========
do ilelt = 1, nlelt
ifac = lstelt(ilelt)
! Wall: zero flow (zero flux for pressure)
! friction for velocities (+ turbulent variables)
! zero flux for scalars
do iphas = 1, nphas
itypfb(ifac,iphas) = iparoi
enddo
enddo
I did the same with my pond that I have since walls also in the surboutine 'usclim.f90" in SRC.2
It's right ?