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

You are not logged in.

#1 Re: Code_Aster usage » how to write out results after every solved increment » 2020-04-13 22:00:15

Well made it work with try/except, just thought there would be way to do that in GUI.

#2 Re: Code_Aster usage » how to write out results after every solved increment » 2020-04-13 15:09:22

Hi,

I made a simple plate file that returns with non inversible stiffness matrix. It fails at 37% of the total load which is correct.
I just would like to know if there is any good practice to save out the results of the last time step.
I tries operator STOP -> Yes in the convergence setup but not working. The methods above are not 100% clear for me.

Thank you for your help in advance

Regards,

MrSnail

#4 Re: Code_Aster usage » Practice with DKT and COQUE_3d » 2019-12-08 17:19:28

The idea came from that model, maybe it will impress somebody else too:
http: //caelinux.org/wiki/index.php/Contrib:Johnnygeling/springplate

#5 Re: Code_Aster usage » Practice with DKT and COQUE_3d » 2019-12-08 17:16:48

So the correct answer is after 1000hrs spent with try and error:
The dummy mesh can be either DKT or 3D
The displacement field has to be projected to the dummy mesh first
The stress need to be calculated by CREA_CHAMP, than extract stress to the layers, than vM stress can be calculated by CREA_CHAMP again. And THAN that stress field has to be projected separately to the same dummy mesh with PROJ_CHAMP.

I think I have to report a possible bug here. If I have more than one PROJ_CHAMP used in the command list, they are not available in the scroll down menu in IMPR_RESU when I want to select the results. But if I close the hdf file and open it again, I can see and select them.

#6 Re: Code_Aster usage » Il n'y a pas de mailles a projeter. » 2019-11-19 20:07:17

Otherwise when you connect 2 3D objects with LIAISON_MAIL, the master is volume, the slave is the connected surface of the volume. Selecting the slave volumetric will make the slave move as a rigid body. Here is a good example smile
http hmm/feaforall.com/salome-meca-tutorial-define-glued-contact-parts/

#7 Re: Code_Aster usage » Il n'y a pas de mailles a projeter. » 2019-11-19 20:02:12

Hi Mahions

I managed to make it work but still getting some warnings. Unfortunately no more time for it today.
You are combining beam+plate+3D, this is not a beginners topic. It is quite easy to work with 3D in code aster but the others need some more knowledge.

You will have 3 result files. One for beam, One for shell and one for 3D. You can open them together in ParaVis in the pipeline browser. However I was praying to all the Gods I know, but still not able to make the deformed shape of the shell. Anyone with more experience maybe..
If I make a 4th result file containing deformations only, than it looks good. But than I cannot project stresses to the deformed shape. Well, you have a full flexibility to organize your result files IF you know how to. But it is not so clear for me yet. I hope I helped.

#8 Code_Aster usage » Practice with DKT and COQUE_3d » 2019-10-16 17:10:56

MrSnail
Replies: 2

Bonjour,

I raised that topic because I am trying to study how to work with DKT and COQUE elements. Made a nice stress plot with DKT but the same method leads to an unexpected error copying the same part into the COQUE command file. As long as I had only deformation in the result they both worked fine but somehow the solver stops at CALC_CHAMP and returns with error.
Usually the error message gives me a hint what the problem is but I am not able to understand this one. Can anyone enlighten me briefly what did I do wrong?


<EXCEPTION> <DVP_1>                                                                               
                                                                                                   
Program error.                                                                                     
                                                                                                 
Condition not met:                                                                                 
   celd1(debugr1+4+4* (iel1-1)+3).eq.lgcata*nbsp1*ncdyn                                         
File                                                                                               
/home/siavelis/PROJETS/code_aster-windows/build/codeaster-src/bibfor/calculel/w155ch.F90, line 275
                                                                                                   
                                                                                                   
                                                                                                   
Il y a probablement une erreur dans la programmation.                                             
Veuillez contacter votre assistance technique.

#9 Re: Code_Aster usage » Aster doesnt read mesh from GMSH » 2019-02-20 20:27:13

Hi. I think the following nodes are on the top of each other:
33-35
34-36

I decided to rebuild the model for Salome (import py) than I noticed them.

#10 Re: Code_Aster usage » Aster doesnt read mesh from GMSH » 2019-02-14 21:10:45

Hi

I am able to read the mesh but the error message I get is that the groups do not form part of the mesh.
Also there is an error sign when I just import the med into the CA mesh module.
With compatibility problem, I am not sure I would able to read the mesh at all (just happened to me when I opened this topic)
I use GMSH 4.0.4 with Code_Aster_Win stable 13.6, they sure are compatible.
I might have an idea what is wrong but if I am right, not possible to trace it back from the med.
Can we actually have the geo file for GMSH?

#12 Code_Aster usage » Steel SHS section (class4) as multifiber - How? » 2019-02-04 17:54:39

MrSnail
Replies: 1

Dear Community,

I would like to make a model of a cantilever beam (figure 1) with a tip mass, using an SHS section (figure 2) which has bolt holes on all sides along the beam. (So the line where the bolt holes are would not be taken into consideration, only effective cross sectional area. Similarly to a class 4 section)
I would like to use multifiber section, because I have further plans with that model.
(to be more specific, add NL material, shake the support with the time history of El Centro EQ, or shock load, or both)
By now I would be more than happy if I could make CA do the following two things:
-Section properties calculated by MACR_CARA_POUTRE
-Working multifiber model for simple gravity load
I have looked through all the tutorials / tests available on the internet but all the multifiber videos were made for concrete
beams where the Centre Line of the section is ON the cross section. Here the CL is not on the section, if I make a separate node
on the mesh of the section than I get error msg because that node is not on the mesh. If I do not provide a node for CL than invalid input for the macro that computes section properties. (fig 3)
Is there any good practice to get this part done?
I tricked that part with adding the full rectangle section so far with 30% tolerance for area and inertia check (just to see if it works or not) but this is not my way, I would like to ask the way to make it correct.

By now if I run this study I get the following error msg (both in Win and Linux):

   ! Program error.                                                              !
   !                                                                             !
   ! Condition not met:                                                          !
   !     .false.                                                                 !
   ! File                                                                        !
   ! /home/siavelis/PROJETS/code_aster-windows/build/codeaster-src/bibfor/debug/ !
   ! contex_param.F90, line 147 

So-far at least I understood parts of the errors but not this one.

Attached model in the next comment, any help/explanation/extra document to read will be highly appreciated.

#13 Re: Code_Aster installation » unable to configure automatically gnu compiler for __main__ product » 2018-12-10 11:42:12

Hi,

Yes I saw that Linux was missing gfortran. I just did not understand why since I had installed gfortran-8, gcc-8 and g++-8 previously and I thought it should have been ok. I am sorry if I made a false alarm but thank you for the answer though.

#14 Re: Code_Aster installation » unable to configure automatically gnu compiler for __main__ product » 2018-12-09 03:17:29

Hi Thomas

I ended up with the same error message, it is uploaded as attachement.
Already installed gcc-8, g++-8 and gfortran-8 too but no progress.

#15 Code_Aster usage » Higway Signal tutorial - a few questions » 2018-11-29 18:11:33

MrSnail
Replies: 4

Dear Community

I am currently studying the Highway Signal example in the Code Aster tutorial from Mr Aubry (thank you for prepaing it, in general I find it very very helpful). I managed to run the calculation with green flag at the end and indeed the deformation values look reasonable. (However, they are not exactly matching with my Staad control calculation but at least the shapes are good, for me it is enough at the moment). However with adding some extra commands I would like to see internal forces, stresses and local coordinate systems, but all the results other than deformation seem to be empty.
I am using Aster 13.4 and some features might work in a bit different way than described in the tutorial, but the command list looks ok and I cannot find the reason for a half-ready result file.
I have the same problem with writing nodal reactions to a file. The result text file does not contain any reaction force.
I uploaded my files and I described my questions in a pdf file, I would highly appreciate (I am actually begging) if somebody could spend a few minutes with them and tell me what did I do wrongly.
I spent already 3-4 evenings with that without any progress.

#16 Re: Code_Aster usage » Aster doesnt read mesh from GMSH » 2018-11-13 20:34:11

And I found the answer for my own question here in this topic:

<AsterStudy> Can't see group in POST_RELEVE_T and POST_ELEM

(I am not allowed to post the link)

Known issues document:
POST_RELEVE_T command
During the edition of the POST_RELEVE_T command, the link with the mesh is not activated until the modifications of the ACTION  factor keyword or the whole command are validated.

#17 Re: Code_Aster usage » Aster doesnt read mesh from GMSH » 2018-11-13 00:12:06

Thank you Mr Aubry, it seems to work with an older GMSH.

However, I would like to ask one more thing as I just came across another interesting situation with the same example.
As it is written in the book I predefined several groups in mesh for beams and nodes and they work very well when I define supports and loads. But when it comes to define GROUP_NO for printing reaction forces into table, the selection box from where I should select node groups is totally blank. Without node group selection, the command does not work.
So I dont understand why is it blank if I predefined my group and I am able to reach/use them in other commands.
If somebody could enlighten me, I would really appreciate that.

#18 Code_Aster usage » Aster doesnt read mesh from GMSH » 2018-10-27 00:06:48

MrSnail
Replies: 12

Hi All,

I am a beginner with CA and started the book called "Beginning with Code_Aster".
I downloaded CAE Linux 2017 (which means Aster 13.4) and GMSH 4.0.4 after.
These versions are a bit newer than what the author used in the book and unfortunately there are some minor changes so not everything goes the same way as written in the book.
I just encountered the following problem:
I made a frame in GMSH 4.0.4 and meshed it. (This is actually chapter 3 in the book) I tried to save the mesh file (what I would like to use with Aster later) in MED format. I assume that this med format is supposed to be the same what Aster reads when both the geometry and mesh is done in Salome. So I decided to try to make Aster read the mesh produced by GMSH instead of typing the command file manually. (theoretically it should work, shouldnt it?)

1., But then I realized that the menu in GMSH->File has been changed and there is no longer "Save As" button.
I can save the mesh with "Save Mesh" but this saves it only in msh format. I dont have the option to change it, and Aster doesnt recognize this msh. Than I tried File->"Export.." where I can choose MED extension but I get error message if I want to import it to Aster.
I tried to import it in 2 different ways.
The first was to go to Mesh module in Salome and File->Import->frame1.med. Aster says: The file is incorrect, some data is missed. (Just produced it with GMSH 1 min ago)
The second way was to launch the Aster Study and read the mesh by the common way, Mesh->Read Mesh and than selecting med file manually. Than Aster says: std::bad_alloc

I was trying to find an answer in both aster and gmsh forums but did not really find anything.
Does anybody know what did I do wrong?
Or would you suggest me to go back to elder versions of these software where the menu is the same as used in the book?
Well I have enclosed the geo/med/msh files here just in case.

Any help/explanation will be appreciated!