4. B modeling: 3D claving JOINT_MECA_RUPT#

4.1. Characteristics of modeling#

The simulation is carried out with 3D modeling_ JOINT. The elements are of type TETRA4 for the studs and PENTA6 for the joint elements. The corresponding law of behavior is JOINT_MECA_RUPT, the associated material has the same name. The solid elements are elastic.

4.2. Characteristics of the mesh#

A linear unstructured mesh is produced by extruding the 2D mesh (figure below).

Volume elements (pads): 2761 TETRA4

Attachments: 92 PENTA6

_images/10000201000002C8000002EA93C2332FAE72A31A.png _images/1000020100000310000002E801C5309B3FC04878.png

Drawing 3:3D mesh, on the left the initial state, on the right the state after the key

4.3. Tested sizes and results#

4.3.1. External source comparison#

The first test is performed on the profile values of the joint thickness after claving by comparing it with the results of GEFDYN. The meshes of GEFDYN and Code_Aster are not the same which explains the fairly high comparison accuracy. We note \({\delta }_{n}\) (\(\mathit{V7}\)) the normal joint opening after claving:

Size tested

GEFDYN

tolerance

\({\delta }_{n}\) at height: \(2m\)

6.38e-7

4%

\({\delta }_{n}\) at height: \(5m\)

2.14e-6

5%

\({\delta }_{n}\) at height: \(8m\)

3.88e-6

4%

4.3.2. Non-regression tests#

The values of the normal joint opening \({\delta }_{n}\) (\(\mathit{V7}\)), joint thickness \({\delta }_{\mathit{offset}}\) (\(\mathit{V10}\)), as well as the normal stresses \({\sigma }_{n}\) are tested at various heights (\(2m,4.5m,8.5m\)).