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

You are not logged in.

#1 2017-11-12 07:09:33

konyaro
Member
From: Genève
Registered: 2016-04-02
Posts: 238

[SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

According to doc u4.51.03 :
« On peut également imposer qu'un chargement de Dirichlet soit suiveur dans le cas de la rigidification d'une partie de la structure par l'utilisation de LIAISON_SOLIDE (voir [U4.44.01]) en grandes transformations. »

But when I try to use LIAISON_SOLIDE with TYPE_CHARGE=’SUIV’ and GROT_GDEP I get the following error:

   !----------------------------------------------------------------------------!
   ! <EXCEPTION> <CHARGES_30>                                                   !
   !                                                                            !
   ! Erreur utilisateur :                                                       !
   !   Le chargement contient des relations cinématiques qui sont non-linéaires !
   !   lorsque l'on utilise EXCIT / TYPE_CHARGE='SUIV'.                         !
   !   Le code ne sait pas encore traiter ces relations non linéaires.          !
   !----------------------------------------------------------------------------!

Any ideas are welcome,

Konyaro

Attached: comm, med and message

Last edited by konyaro (2018-11-26 22:13:43)


Attachments:
Charge_suiveuse.zip, Size: 10.82 KiB, Downloads: 159

失敗は成功のもと (L'échec est la base de la réussite)

Offline

#2 2017-11-17 21:10:13

AsterO'dactyle
Administrator
Registered: 2007-11-29
Posts: 312

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

Hello,

Sorry, it's a bug. I didn't find workaround

I emit a ticket in our bugtracker


Code_Asterの開発者

Offline

#3 2017-11-21 20:56:15

konyaro
Member
From: Genève
Registered: 2016-04-02
Posts: 238

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

Thank you AsterO’dactyle !

It would be nice that LIAISON_SOLIDE works with TYPE_CHARGE=’SUIV’. I found no way to apply a torque on a node with large rotations. That means I am not able to simulate 2 gears with a resistive torque.

Since POU_D_T_GD are not stable with large rotations LIAISON_SOLIDE would be the perfect option. It would be so great if it also worked with DYNA_NON_LINE…

Best regards,
Konyaro

Last edited by konyaro (2017-11-21 20:58:54)


失敗は成功のもと (L'échec est la base de la réussite)

Offline

#4 2017-11-28 22:06:08

AsterO'dactyle
Administrator
Registered: 2007-11-29
Posts: 312

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

Hello,

It's really a bug but a limitation of the functionality.
You cannot use LIAISON_SOLIDE in 2D if one of your nodes has a rotational dof DRZ.

But you model is not correct. If you define a rotational DOF to have a solid rotation around the DZ axis, how you link this node with the plane stress model ?
I understand what is your goal and I suggest juste to define a "rotational" boundary condition on all your nodes (see .comm file attached).

Hope it helps !

Best regards


Attachments:
disk.comm, Size: 2.16 KiB, Downloads: 204

Code_Asterの開発者

Offline

#5 2017-12-02 09:38:01

konyaro
Member
From: Genève
Registered: 2016-04-02
Posts: 238

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

Hello  AsterO’dactyle, thank you for your answer.

A POI1-2D_DIS_TR can be linked to a plane stress model with LIAISON_SOLIDE. I can then rotate its DOF DRZ and it works fine, except that it get deformed with large rotations as TYPE_CHARGE=’SUIV’ cannot be activated.

Thank you for your example, it works fine. The problem is that I often need to apply a torque on a free rotational DOF, for instance to simulate a gear as explained in the attached image. I didn’t find other ways than using POU_D_T_GD which are not stable with large rotations. That’s why I tried the LIAISON_SOLIDE with TYPE_CHARGE=’SUIV’.

free rotation

Regards,

Last edited by konyaro (2018-11-25 07:05:11)


Attachments:
Explications.png, Size: 120.26 KiB, Downloads: 657

失敗は成功のもと (L'échec est la base de la réussite)

Offline

#6 2017-12-02 09:42:53

konyaro
Member
From: Genève
Registered: 2016-04-02
Posts: 238

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

Attached the rotation with LIAISON_SOLIDE without TYPE_CHARGE=’SUIV’.


Attachments:
Rotation_non_suiveuse.zip, Size: 3.87 KiB, Downloads: 155

失敗は成功のもと (L'échec est la base de la réussite)

Offline

#7 2018-11-26 22:13:23

konyaro
Member
From: Genève
Registered: 2016-04-02
Posts: 238

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

The answer lies in the new error message (2018):

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! <S> Exception user raised but not interceptee.                                 !
! The bases are fermees.                                                         !
! Type of the exception: error                                                   !
!                                                                                !
! Erreur utilisateur :                                                           !
!   Le chargement contient des relations cinématiques LIAISON_SOLIDE qui sont    !
! non-linéaires lorsque l'on utilise EXCIT /                                     !
! TYPE_CHARGE='SUIV'.                                                            !
!   Mais ce cas n'est pas traité car il y a au moins un noeud qui porte le degré !
! de liberté DRZ.                                                                !
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

The LIAISON_SOLIDE with TYPE_CHARGE='SUIV' works only for translations and unfortunately not for rotations.


失敗は成功のもと (L'échec est la base de la réussite)

Offline

#8 2020-04-02 09:03:34

sameer21101970
Member
Registered: 2019-09-06
Posts: 144

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

Mr. Konroyo,

I understand from your above messages, you are doing Gear Simulation under Torque.

I would like to know, You are doing Gear Simulation of 3D Objects or only 2D Objects.

I am trying to simulate 2 Gears - 3D Solid and faced with lot of problems, not working. ASTK just Hangs.

I have gone through u2.04.04, u4.44.11...lot of confusion. It is not straight forward. Gear Meshes will come in contact as gear rotates..so what is the right step.

Your sharing of knowledge on same will helps all , who loves code-aster.

Waiting for your reply.

Offline

#9 2020-04-03 06:32:11

konyaro
Member
From: Genève
Registered: 2016-04-02
Posts: 238

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

Hello,
3D gears can be well simulated with CA. I recommend you the followings:
- use AFFE_CHAR_MECA_F to apply the rotation
- start your simulation without contacts. If some DOF are free add some springs (CREA_POI1 and DIS_T)
- use a penalized contact
- Start with a low COEF_PENA_FROT and then adjust the COEF_PENA_CONT and COEF_PENA_FROT with a convergence study

Konyaro


失敗は成功のもと (L'échec est la base de la réussite)

Offline

#10 2020-04-03 08:29:59

sameer21101970
Member
Registered: 2019-09-06
Posts: 144

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

Thanks for your reply.

I would like to understand your point 2,
"start your simulation without contacts. If some DOF are free add some springs (CREA_POI1 and DIS_T)"

Means, First Not to Define Contact & without Contact Run the Simulation. Then second part DOF ...is not understood...will you please elaborate with some small example.

PETSC Solver is not supported & thus using MUMPS....It is hanging the ASTK 2018.1 for long....with over lac Degree of libretre

Last edited by sameer21101970 (2020-04-03 08:37:17)

Offline

#11 2020-04-03 13:58:10

konyaro
Member
From: Genève
Registered: 2016-04-02
Posts: 238

Re: [SOLVED] LIAISON_SOLIDE with TYPE_CHARGE='SUIV' and GROT_GDEP

Hello,
I guess you know that the K matrix cannot be inverted if there some free DOF.

Attached a small example:
1) the z direction is not constrained, the simulation does not converge.
2) the z direction is not constrained either but some weak springs are added, the simulation converges.

There are many posts concerning these weak springs. It is also explained in doc  U2.04.04 §4.4 and in the testcase SSNA122.

Konyaro


Attachments:
springs.zip, Size: 3.65 KiB, Downloads: 22

失敗は成功のもと (L'échec est la base de la réussite)

Offline