Atom topic feed | site map | contact | login | Powered by FluxBB | réalisation artaban

You are not logged in.

- Topics: Active | Unanswered

**draf1992****Member**- From: Warsaw
- Registered: 2018-04-17
- Posts: 11

Hello everyone,

I would like to learn how to model a hinge with a rotational spring. I know how to model a single hinge.

To get the hinge I used LIANSON_DDL.

LIAISON_DDL =_F ( NOEUD = ('V2', 'V3'),

DDL = ('DX', 'DX'),

COEF_MULT = (1., -1.),

COEF_IMPO = 0., )LIAISON_DDL =_F ( NOEUD = ('V2', 'V3'),

DDL = ('DY, 'DY'),

COEF_MULT = (1., -1.),

COEF_IMPO = 0., )LIAISON_DDL =_F ( NOEUD = ('V2', 'V3'),

DDL = ('DZ, 'DZ'),

COEF_MULT = (1., -1.),

COEF_IMPO = 0., )

Case 1 and 2 are in attached file. In the first case without rotational spring I received a Singular Matrix. In the second case without rotational spring I received correct results consistent with analytical calculations. I would like to avoid using discrete element SEG2. I tried with a discreet element of POI1. I added the zero 0D element to the V2 and V3 nodes. I'm not sure if this is a good solution.

Boundary conditions in the hinge wrote:

U=Y, QL=QP-F, ML=-k*(Up-Yp), MP=k*(Yp-Up).

U - displacement of the left beam

Y - displacement of the right beamUp - derivative of displacement of the left beam

Yp - derivative of displacement of the right beamk - spring stiffness

Ql - shear force in the left beam

QP - shear force in the left beamM - bending moment

How to add these additional conditions except that the displacements are to be equal for node V2 and V3. For any tips thanks in advance

Best

Rafał

*Last edited by draf1992 (2018-12-14 22:28:00)*

Offline

**jlf****Member**- Registered: 2007-11-22
- Posts: 211

Hello,

You could try a DIS_TR (a K_TR_D_L on a SEG2 with a null size or not) with the same LIAISON_DDL for UX, UY, UZ.

You have 6 values to give for the K_TR_D_L :

- 3 values for displacement, with the LIAISON_DDL the values doesn't matter.

- 3 values for rotation, k

Watch out for the orientation of the SEG2, especially if its size is zero!

JLF

Offline

**draf1992****Member**- From: Warsaw
- Registered: 2018-04-17
- Posts: 11

Hi jlf,

Thank you for your response. This method works for case 1 and case 2. I received correct results consistent with analytical calculations. Both beams were along the X axis and the Y and Z axes in the local system corresponded to the Y and Z axes in the global one. What does the SEG2 orientation look like when the size is zero? My system constructed of beams will be very complex. Therefore, I would like to avoid adding an additional element like DIS_TR and problem with his orientation. Is there another way to achieve the same result? I do not know how to impose bending moment as a function of the derivative of displacement of another node. Then it may be possible to write the function ML = -k * (Up-Yp) and use LIANSON_DDL instead of the additional element.

Best

Rafał

Offline

**jlf****Member**- Registered: 2007-11-22
- Posts: 211

Hello,

If you want to release DDL, the only way is to model a SEG2 between the 2 nodes.

There no way to write relation between Force and Deplacement (duals variables) in commands file, in fact it's a behavior.

The solutions I propose to you, simple or not :

- you write the LIAISON_DDL like you already made

*** Case where the hinge have the same 'k' is in the 3 directions, no torsion in the beams:

- you create a SEG2 between the 2 nodes (length and direction don't matter)

- the sames values for rotation KRX, KRY, KRZ

*** Case where the hinge have KRY=KRZ and a different for KRX

- when you create your beams you made 2 nodes with very slightly different. The X direction for the spring is the V2V3 vector. In this case, if the length is not null you don't need to use ORIENTATION keyword when KRY=KRZ.

- the same values for KRY, KRZ and another for KRX.

JL

Offline

**draf1992****Member**- From: Warsaw
- Registered: 2018-04-17
- Posts: 11

Hello jlf,

Thank you for your response and tips. Is it possible to add new boundary conditions in the form

u(2)-u(3) = 0

Q(2)-Q(3) = 0

MY(2)-k (DRY(2)-DRY(3)) = 0

MY(3)-k (DRY(3)-DRY(2)) = 0

The node number is in brackets. I obtained the first of these conditions thanks to LIANSON_DDL.

Best

Rafał

Offline

**jlf****Member**- Registered: 2007-11-22
- Posts: 211

Hello,

No you can't.

In LIAISON_DDL you can only have relations between Degree of Freedom (DDL=DoF). MY,MZ are not a DoF

Cf U4.44.01 : AFFE_CHAR_MECA documentation in §5.1 of the DoF

JL

Offline

**draf1992****Member**- From: Warsaw
- Registered: 2018-04-17
- Posts: 11

Hello jlf,

Thank you for the explanation.

Best

Rafał

Offline

**Lore****Member**- Registered: 2016-04-01
- Posts: 6

Hi JL,

is it possible to have some kind of reference of the zero-length SEG2 elements?

I was not able to define them, and as they are very useful, I'm very interest in their implementation in models.

Thank you

Lorenzo

Offline

**jlf****Member**- Registered: 2007-11-22
- Posts: 211

Hello,

You can find a lot of examples in the test cases of code_aster :

In command file you search the words : 'ORIENTATION' with 'K_TR_D_L' | 'K_T_D_L' | 'K_TR_L' | 'K_T_L'.

If in AFFE_CARA_ELEM/ORIENTATION you have GROUP_MA with CARA= 'ANGL_NAUT', it's found.

The [V6.02.135] 'ssnl135' is a study on a structure with lot of springs with null size.

JLF

Offline

**Lore****Member**- Registered: 2016-04-01
- Posts: 6

Thank you very much JLF!!

Lorenzo

Offline