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

You are not logged in.

#1 2020-04-17 13:53:52

svantevid
Member
Registered: 2020-03-13
Posts: 31

Rotational beam displacement question

From my understanding of rotational displacement application in Aster-study (or in general), they cannot be applied to solid elements, only to 2D and 1D elements. I followed the tutorial from Claws (__https://www.caelinux.org/wiki/index.php/Contrib:Claws/Code_Aster/10_x_cases/torque) and defined a simple beam with a rotational displacement on the node, that is connected to the front face with a solid link. The deformed shape of the beam is presented in the attachement bellow.
The beam has a "strange" deformation of the face, where the load node is attached. Is this perhaps a consequence of this method? Has anyone encountered something similar?


Attachments:
rot_disp_beam.png, Size: 84.56 KiB, Downloads: 79

Offline

#2 2020-04-17 13:55:22

svantevid
Member
Registered: 2020-03-13
Posts: 31

Re: Rotational beam displacement question

I also attached the .comm file.


Attachments:
Stage_1.comm, Size: 1.62 KiB, Downloads: 89

Offline

#3 2020-04-17 19:09:01

sameer21101970
Member
Registered: 2019-09-06
Posts: 263

Re: Rotational beam displacement question

code-aster.org/forum2/viewtopic.php?id=24733

refer above link
define rotation by python

        _F(DRX=1.57,
DRZ = ....NOT DRX...

Your shaft centreline on OZ axis

Last edited by sameer21101970 (2020-04-17 19:17:25)

Offline

#4 2020-04-20 11:07:24

svantevid
Member
Registered: 2020-03-13
Posts: 31

Re: Rotational beam displacement question

sameer21101970 wrote:

code-aster.org/forum2/viewtopic.php?id=24733

refer above link
define rotation by python

        _F(DRX=1.57,
DRZ = ....NOT DRX...

Your shaft centreline on OZ axis

If I understand your comment, I should define the torsion of the beam (rotation in relation to the centre axis of the beam).
But I wan't to apply bending to the beam with a rotational displacement at the free end. I did that and the beam has deflected.

The problem is the strange deformation of the beam face at the free end (the face that is attached to the PO1 node with LIAISON_SOLIDE).

Offline

#5 2020-04-22 05:01:56

konyaro
Member
From: Genève
Registered: 2016-04-02
Posts: 255

Re: Rotational beam displacement question

Hello,
LIAISON_SOLIDE calculates displacements at the beginning of the simulation, which means the first direction of your rotation for each node will give the direction for the whole simulation. That's why you get a strange shape.

In recent versions of CA, LIAION_SOLIDE can be used with TYPE_CHARGE='SUIV' which actualizes the direction of the displacement, cf. https://code-aster.org/forum2/viewtopic … 539#p50539
Unfortunately it does not work in your case because it cannot link rotationnal DOF as your 3D elements don't have them cf. https://code-aster.org/forum2/viewtopic.php?id=23018
It's a pity it is not implemented I agree.

The remaining options are:
1) Use POU_D_T_GD, cf. https://code-aster.org/forum2/viewtopic.php?id=21253, which work well for small rotations. Some new beam elements should overcome these instabilities in the future, cf. https://code-aster.org/forum2/viewtopic … 357#p57357
2) Impose the position of the nodes with AFFE_CHAR_MECA_F, cf. https://code-aster.org/forum2/viewtopic … 420#p55420

Konyaro

Last edited by konyaro (2020-04-22 05:02:51)


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

Offline

#6 2020-04-22 06:00:41

sameer21101970
Member
Registered: 2019-09-06
Posts: 263

Re: Rotational beam displacement question

dear
"But I wan't to apply bending to the beam with a rotational displacement at the free end."

You want to apply Torsion and Bending Moment, Both on same end.

You can do with beam element. In Codeaster i have not tried but other software i hv tested multiple bending, torsion moments & axial loads at multiple location on shaft.

send your med and .comm files

Last edited by sameer21101970 (2020-04-22 06:04:21)

Offline

#7 2020-05-13 09:15:03

svantevid
Member
Registered: 2020-03-13
Posts: 31

Re: Rotational beam displacement question

konyaro wrote:

Hello,
LIAISON_SOLIDE calculates displacements at the beginning of the simulation, which means the first direction of your rotation for each node will give the direction for the whole simulation. That's why you get a strange shape.

In recent versions of CA, LIAION_SOLIDE can be used with TYPE_CHARGE='SUIV' which actualizes the direction of the displacement, cf. https://code-aster.org/forum2/viewtopic … 539#p50539
Unfortunately it does not work in your case because it cannot link rotationnal DOF as your 3D elements don't have them cf. https://code-aster.org/forum2/viewtopic.php?id=23018
It's a pity it is not implemented I agree.

The remaining options are:
1) Use POU_D_T_GD, cf. https://code-aster.org/forum2/viewtopic.php?id=21253, which work well for small rotations. Some new beam elements should overcome these instabilities in the future, cf. https://code-aster.org/forum2/viewtopic … 357#p57357
2) Impose the position of the nodes with AFFE_CHAR_MECA_F, cf. https://code-aster.org/forum2/viewtopic … 420#p55420

Konyaro

Thanks for this lenghty answer, I will look into all the options.

What about trying to define a non-deformable part of the 3D model and linking that to the single node (where I apply the rotational displacement)?

Offline