Welcome to the forums. Please post in English or French.

You are not logged in.

#1 Re: Code_Aster usage » Shells - DKT - Plastsicity - (In)correct Mises » 2021-10-22 21:19:25

tpa

Do you use the same mesh for DKT and Coque3D? Element type for Coque3D should be TRIA7 or QUAD9 ie. 2nd order with a center node. One way to do that in case your mesher cant handle these type (e.g. Salome) is to issue a MODI_MAILLAGE in Code Aster. This should be just after you read the mesh.

#2 Re: Code_Aster usage » Shells - DKT - Plastsicity - (In)correct Mises » 2021-10-22 15:46:42

tpa

Good that it works for DKT modelisation. I do not use Coque3D a lot, as the results require some processing above my level of ability to be handled by PARAVIS.

#3 Re: Code_Aster usage » Shells - DKT - Plastsicity - (In)correct Mises » 2021-10-22 07:19:57

tpa

I have not tried your case but I have one observation.

When postprocessing shell/plate elements you need to extract the result at a certain position across the thickness with post_champ. Something like the below code which is taken from the book (highly recommended) by Jean-Pierre Aubry, "Beginning with Code Aster", p90. You can also look in u2.02.01 section 2.4.2.

stat2=POST_CHAMP(
RESULTAT=stat ,
GROUP_MA= (’panel’ , ) ,
EXTR_COQUE=_F(
NUME_COUCHE=1 ,
NIVE_COUCHE=’SUP’ ,
NOM_CHAM= (’SIGM_ELNO’ , ) ,
) ,);

#4 Re: Code_Aster usage » How to show the node used in a formule » 2021-09-11 09:07:42

tpa

I dont fully understand what you will but it sounds like a task that potentially could be completed with post_releve_t if you alteady have the information in a field (champ) or calc_table.

#5 Re: Code_Aster usage » Calcul sismique » 2021-09-10 14:45:58

tpa

You may find useful inspiration here: webthesis.biblio.polito.it/11838/ Depending on how you use the information giving a reference to Pasquale Mangano would be polite. His work really good demonstrates how code_aster can be used for a complex case.

#6 Re: Code_Aster usage » [SOLVED] Apply variable field of pressure (or any field...) » 2021-03-06 17:55:00

tpa

Why not use AFFE_CHAR_MECA_F for this? Frankly I have not used it myself but I believe it can be used, as its idea is to define loads etc. as function of e.g. location or temperature. See u4.44.01 section 7.6 p48. In the machine translated english version pressure is called NEAR dont be confused by that. The french versions are better. See p59 for beam elements.

#7 Re: Code_Aster usage » Distributed load - POU_D_E element » 2021-02-28 11:30:05

tpa

Don't overlook that in AFFE_CHAR_MECA (u4-44.01) FORCE_POUTRE you can apply distributed forces with reference to the global coordinate system (FX FY FZ MX MY MZ) or in the local (element) coordinate system (N VY VZ MT MFY MFZ). To me the most challenging aspect of beam (poutre) elements is to keep track of the element orientations. This is of course always the case but "especially always" the case when you have some of the less trivial sections (trivial = circular or square) or work in the local (element) coordinate system.

#8 Re: Code_Aster usage » Draw shear-moment diagram » 2021-02-27 15:41:13

tpa

Strictly speaking shear and moment diagrams will be all zeros for trusses. For beam systems I don't know of any other way for visualising than using colourised plots with Salome/ParaVis. I don't think gmsh will do classic force/moment diagrams either. You need a postprocessor which has this functionality.

#9 Re: Code_Aster usage » Distributed load - POU_D_E element » 2021-02-21 15:03:11

tpa

Goto the documentation document u3.11.01 p.5 and realise that you may be right:

#10 Re: Code_Aster usage » [Solved] Thermo-Elastic analysis » 2020-10-10 23:56:17

tpa

Yes, AFFE_VARC it is. Below are some pieces of my .comm file, to inspiration for other:

You will need a function to transfer from thermal analysis time to mechanical analysis time. I just use Tm=Tt:

func = DEFI_FONCTION(identifier=u'7:1',
                     INTERPOL=('LIN', ),
                     NOM_PARA='INST',
                     VALE=(0.0, 0.0, 1.0, 1.0))

Then, for the mechanical analysis, use AFFE_VARC to apply the results from THER_LINEAIRE. This is done in the AFFE_MATERIAU - NOT as I had thought in AFFE_CHAR_MECA.

MatMech = AFFE_MATERIAU(identifier=u'13:1',
                        AFFE=(_F(GROUP_MA=('MAT1', ),
                                 MATER=(MAT1, )),
                              _F(GROUP_MA=('MAT2', ),
                                 MATER=(MAT2, ))),
                        AFFE_VARC=_F(EVOL=JumpTher,       <- JumpTher are the result from THER_LINEAIRE
                                     FONC_INST=func,                      <- The function mentioned above
                                     NOM_VARC='TEMP',
                                     TOUT='OUI',
                                     VALE_REF=80.9),
                        MODELE=Mech)

reslin = MECA_STATIQUE(identifier=u'14:1',
                       CHAM_MATER=MatMech,
                       EXCIT=_F(CHARGE=load),
                       LIST_INST=listr,                                     <- Here I use the same list of times as for THER_LINEAIRE but it could be a second list
                       MODELE=Mech)

reslin = CALC_CHAMP(identifier=u'15:1',
                    reuse=reslin,
                    CONTRAINTE=('SIGM_NOEU', ),
                    CRITERES=('SIEQ_NOEU', ),
                    LIST_INST=listr,                                      <- Here I use the same list of times as for THER_LINEAIRE but this could be a third list
                    RESULTAT=reslin)

Attached file is an animation of the displacements colored with Von Mises stresses.

:-)

#11 Re: Code_Aster usage » [Solved] Thermo-Elastic analysis » 2020-10-10 21:38:44

tpa

According to "code-aster.org/V2/UPLOAD/DOC/Formations/06-thermal_analysis.pdf" the missing link could be AFFE_VARC.
Will try that.

#12 Code_Aster usage » [Solved] Thermo-Elastic analysis » 2020-10-10 15:04:58

tpa
Replies: 3

Hello.

I am trying to make a dynamic thermal simulation with code_aster where I use THER_LINEAIRE to generate data for the temperature distribution in my structure of two materials. The two have different thermal expansions and my goal is to investigate the thermal stresses during a transient.

I have the thermal analysis running and the results seem reasonable (development over time).

Next step is to transfer the temperature distribution at each INSTANT to MECA_STATIQUE to calculate the stresses.

Sounds sort of simple ... I just can't find out how to apply the temperature distribution to the MECA_STATIQUE analysis.

Are there any good reference examples or cases to learn from - or do any of you have some good tips/experiences to share?

I work in SalomeMECA 2018 which has this version of code_aster:

              -- CODE_ASTER -- VERSION:  EXPLOITATION (stable) --

                       Modified 13.6.0 version 07/08/2018
                    the revision 3c51e6c237ea - branch  v13
                        Copyright EDF R & D 1991 - 2020

#13 Re: Code_Aster usage » Data for elasto-plastic analysis of steel » 2020-09-10 06:54:29

tpa

Thank you for all your help. From the references I have been given here I have learned about the Ramberg-Osgood equation which seems to be a commonly applied model. I also learned that life is simpler when the plastic work curve is generated on a epsilon(stress) function rather than stress(epsilon) and third lesson is that even conditions after the necking occurs can apparently be handled and fourth lesson is to be picky with regards to parameters as they depend on the state of the material - e.g. cold drawing and heat treatment conditions. Fiifth and foremost lesson is that You, Mario and William are very helpful.

#14 Re: Code_Aster usage » Data for elasto-plastic analysis of steel » 2020-09-08 07:06:54

tpa

Thank you very much. Especially the second article gives some food to thoughts, considering in which condition the material is. I also appreciate the link to the digitizing program. Best regards.

#15 Re: Code_Aster usage » Data for elasto-plastic analysis of steel » 2020-09-07 20:18:47

tpa

Hi Mario. Thank you for your response. I have not so much experience with elasto-plastic analysis of stainless steel so my intention was to see if I could recreate the behavior reported in this article: doi.org/10.1016/j.net.2020.07.014 with AISI304 or AISI316, Both seem to be of the L quality, eg. 1.4307 and 1.4404 resp.

In the past I have used code-aster for plastic analysis of copper stuctures with some succes but then I had to calibrate the plasticity curves for the specific material with experiments.

Especially considering the much more common interest for plastic analysis of (stainless) steels, I was thinking that somewhere there might be tabulated values for standardized conditions as exist for engineering yield and tensile stresses.

#16 Code_Aster usage » Data for elasto-plastic analysis of steel » 2020-09-06 21:30:45

tpa
Replies: 9

Hello.
Is there any recommended sources to find recommended values to use for true stress-strain curves (TRACTION) or fitting parameters for such curves to use with ECRO_PUIS or ECRO_LINE. I am mainly considering stainless steels.
I will appreciate any proposals.

#17 Re: Salome-Meca installation » Salome-Meca on a terminal » 2020-08-02 08:25:07

tpa

Also - as far as I have found the MIT-SHM extension is based on shared memory, requiring the client and server to run on the same machine. In my case I have the client (SalomeMECA) on the far machine and the Xserver (Xming) on my terminal. Strange.

#18 Re: Salome-Meca installation » Salome-Meca on a terminal » 2020-07-31 22:45:38

tpa

Hi,
Thank you for the proposal.
I just tried the method on a fresh install. Some things change, other things seem unaffected.

When I start salome_mesa I can activate all of the modules but the graphic window does work (is not updated) in either module and I get this error:

Xlib:    extension "MIT-SHM" missing on display "192.168.1.220:0"

The error is repeated on mouse events in the graphic area. The IP is the terminal that I am on, so the terminal server (Xming) might not be compatible. I do not know if that should be a known issue. It has worked in the past.

When I start salome_gpu everything seems working, except for Paravis module which still cause ungraceful crash as before. So - I can calculate but not view results.

#19 Re: Salome-Meca installation » salome-meca for windows Error: Can not save file » 2020-07-28 18:43:57

tpa

Something to try:

Filename/path like "/home/user/User/Gerardo/Desktop/study1.hdf" is not a valid Windows path. Try another name.
Other reasons for not being able to write to the hdf file can be that a phantom server process is rendering the .hdf file R/O. In that case, save in a different name and afterwards shutdown the session server use the kill_salome.bat command (in the distribution directory).

#20 Salome-Meca installation » Salome-Meca on a terminal » 2020-07-16 16:33:51

tpa
Replies: 3

I have installed Salome-Meca 2018 (and 2019, actually) in a linux system (Linux Mint 18 ~ Unbuntu Xenial (16.04)).

The linux system runs on a virtual machine on a Xen-server. I connect to the machine via an X session from remote.
All seems to work well with the GEOM, SMESH and ASTER modules, but whenever I activate PARAVIS in this configuration Salome Meca crashes quite ungracefully with no error windows or goodbyes. I guess something with the screen handling over network might be part of the cause, but the X session itself survives the crash so one has to dig deeper ...

Thanks for reading. Any help on how to get this working will be highly appreciated.

Edit: I have a .crash file including the core dump but this is rather big, so I will wait to hear if anyone requests it or parts from it before uploading.

#21 Re: Salome-Meca usage » Shear/Moment diagrams in ParaVIS » 2020-07-03 17:17:50

tpa

Thanks Voulet2 I will use your suggestion.

#22 Re: Salome-Meca usage » global Stress results visualization for shell DKT element is wired » 2020-06-26 21:40:01

tpa

In human words: For plate/shell elements CALC_ELEM calculates SIGM_ELNO stresses in layers: top (SUP), middle (MOY) and bottom (INF). You have to extract the layer of interest to a separate result by using the EXTR_COQUE function of POST_CHAMP command, and then also remember this additional result in the IMPR_RESU command if you want to visualize in ParaVIS.

The thread "[SOLVED] Plotting stresses from DKT elements" has a .comm file giving an example.

I think that these extra steps are because ParaVIS is not optimized to understand the complex information provided by relatively simple element types such as the shell plate and beam elements. When it comes to that the old POST_PRO was bit better integrated with the code_aster data structures.

#23 Re: Salome-Meca usage » extract external resultant moments » 2020-06-26 20:29:39

tpa

Yes, something like:

Calculate the fields:
reslin = CALC_CHAMP(
  FORCE=('FORC_NODA'),
  RESULTAT=reslin,
  TOUT='OUI')

Extract the group of nodes to a table:
table = POST_RELEVE_T(
  ACTION=(_F(
      GROUP_NO=('P2', 'P1'),
      INTITULE='1',
      NOM_CHAM='FORC_NODA',
      OPERATION=('EXTRACTION', ),
      RESULTAT=reslin,
      TOUT_CMP='OUI'
    ))
)

And dump the table to a text file on UNIT 8:
IMPR_TABLE(
  FORMAT='TABLEAU',
  TABLE=table,
  UNITE=8
)

- will give a table like the attached file. This was for a small beam model with a force FZ=-1000 on N2 and N1 fully constrained. This load is clear from the table as well as the reaction forces at N1.

#24 Salome-Meca usage » Shear/Moment diagrams in ParaVIS » 2020-06-25 21:32:45

tpa
Replies: 2

Hello, all. I am trying to (re)learn to work with beam (poutre) models. In ParaVIS I can make a colored plot of forces and moments in SIEF_ELNO (esp. VY, VZ, MFY, MFZ). Is it possible to illustrate these as an oldfashioned force/moment diagram plotted along the length og the beam. That is what we did to visualise in the old days, and I recall that the long gone POST_PRO was able to do so. But is is possible also in ParaVIS?
Thank you for sharing your experiences.

#25 Re: Salome-Meca installation » [RESOLU] USB Live Salome Meca » 2012-12-13 20:54:58

tpa

Regarding the USB pendrive option ...

For testing x64 I installed Linux Mint and Gnome, OpenOffice, Salome MECA 2010, Code Saturne 2.0 and Paraview 3.14 on a 16GB USB stick. 12GB root partition and 4GB swap. It works perfectly well. Still 55% free without trying to save space except for deleting the build tree after getting Code Saturne to work.

Newer Salome MECA require more space as they also include PARAVIS module. I stuck to the older version as it has IMO the best compatibility between Code Aster and the POST module og Saturne.

Windows cannot data exchange directly with the linux partitions of the USB stick but there are ways to handle.