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

You are not logged in.

#1 2017-02-16 16:19:02

bbd
Member
From: Vandoeuvre, F
Registered: 2016-09-30
Posts: 189
Website

[RESOLU] problem with chained calculations (reuse) and liaison_solid

I'm trying to chain 2 static calculations (stat_non_line)
at the end of the 1rst computation I'd like to link 2 groups of nodes with liaison_solide and then start the 2nd static run.
But the structure comes back in its initial state at the start of the 2nd run athough  I used :
    reuse=preload,
    ETAT_INIT=_F(EVOL_NOLI= preload,),
it seems the liaison_solide is active but with the nodes in their initial state.

The chained calulcation ran well with no error, but the result is wrong. At least not what I expected.
See attached the files and the sketch of the test.

Last edited by bbd (2017-02-20 16:16:38)


Attachments:
restart.rar, Size: 505.78 KiB, Downloads: 41

Offline

#2 2017-02-17 09:52:33

bbd
Member
From: Vandoeuvre, F
Registered: 2016-09-30
Posts: 189
Website

Re: [RESOLU] problem with chained calculations (reuse) and liaison_solid

j'ai l'impression que mon problème est identique au post : http://web-code-aster.org/forum2/viewtopic.php?id=14299
qui n'avait pas trouvé de solution.

the issue doesn't come from liaison_solide. I substitute ddl_impo (on top and bottom faces) for liaison_solide and the chaining is wrong either. see the video attached.
everything happens like ETAT_INIT is not recognized.


Attachments:
restart.ogv, Size: 130.32 KiB, Downloads: 41

Offline

#3 2017-02-17 11:27:45

bbd
Member
From: Vandoeuvre, F
Registered: 2016-09-30
Posts: 189
Website

Re: [RESOLU] problem with chained calculations (reuse) and liaison_solid

ok same pb as http://code-aster.org/forum2/viewtopic.php?id=21255
I begin to understand what's going on.

Offline

#4 2017-02-17 11:27:56

jeanpierreaubry
Ex-Guru
From: nantes (france)
Registered: 2009-03-12
Posts: 2,994
Website

Re: [RESOLU] problem with chained calculations (reuse) and liaison_solid

hello

everything happens like ETAT_INIT is not recognized.

i do not believe this is the point
in the second instance the boundary conditions of  the first one have vanished
it is possible to to put a multiplier on a DDL_IMPO
but i do not think it is possible to tell the code that this condition does not exist anymore at a given instant

i solved quite many similar problems
but this was with
a set of fix boundary conditions common to the two instances
a load with force

jean pierre aubry

Offline

#5 2017-02-17 13:25:06

RichardS
Member
From: Munich, Germany
Registered: 2010-09-28
Posts: 528
Website

Re: [RESOLU] problem with chained calculations (reuse) and liaison_solid

Hello bbd,
I think the problem is that LIAISON_SOLID is (by default) defined with the initial mesh geometry as the reference, not the deformed geometry.
In newer versions (V13), LIAISON_SOLID works also for large rotations, thus taking into account the updated geometry (basically you have to use TYPE_CHARGE='SUIV').
I did not try it, but it could (should) work.

Best,
Richard

Last edited by RichardS (2017-02-17 13:26:50)


Richard Szoeke-Schuller
Product Management
www.simscale.com
We are hiring! https://simscale-jobs.personio.de/?language=en#all

Offline

#6 2017-02-17 13:51:58

bbd
Member
From: Vandoeuvre, F
Registered: 2016-09-30
Posts: 189
Website

Re: [RESOLU] problem with chained calculations (reuse) and liaison_solid

Exactly Richard that's the reason.
I understood it reading the other post, but I did not know how to ensure continuity of the boundary conditions.
Thanks for the trick with charge suiveuse

Offline

#7 2017-02-17 16:23:41

bbd
Member
From: Vandoeuvre, F
Registered: 2016-09-30
Posts: 189
Website

Re: [RESOLU] problem with chained calculations (reuse) and liaison_solid

before coming back to my test with LIAISON_SOLIDE I solved the problem with DDL_IMPO instead.

As you mentioned, TYPE_CHARGE='SUIV' was necessary to ensure the continuity of the boundary conditions (then DZ=-0.005 imposed in 1st stage becomes DZ=0 in the 2nd stage)

and also TYPE_CHARGE='DIDI' to ensure the continuity of the imposed displacement in the 2nd stage. The explanation given in u4.51.03 is not clear but it works.

See attached the description of the test, the files and the video of the result.
Hope it helps other users.


Attachments:
restart.rar, Size: 543.63 KiB, Downloads: 43

Offline

#8 2017-02-17 17:03:10

bbd
Member
From: Vandoeuvre, F
Registered: 2016-09-30
Posts: 189
Website

Re: [RESOLU] problem with chained calculations (reuse) and liaison_solid

with the LIAISON_SOLIDE the load must be of DIDI type either.
Let's see the test attached.


Attachments:
restart.rar, Size: 603.32 KiB, Downloads: 38

Offline

Board footer