Keywords: finite element modelling, mechanical joining processes, damage, fracture, self-piercing riveting.

This work deals with the numerical modeling of various mechanical joining techniques (self-piercing rivets, clinching, clinch-riveting, etc.) as well as their mechanical strength. Indeed mechanical joining leads to plastic deformation and residual stresses that play an important role on the final strength of the joint. It is therefore important to take into account their mechanical history, due to the joining process, in order to get a good representation of their mechanical strength for different loading conditions.

Mechanical joints are widely used in the automotive or naval industry and even more in aircraft structures. Despite the importance of these joining techniques, improvements of the structural behaviour of mechanical joint structures are still mainly due to trial and error tests or knowledge-based procedures. This is mainly due to the numerical difficulties relative to these processes: for example self-pierce riveting (SPR) involves multimaterials contact, friction, high plastic deformation, damage and fracture.

We used the finite element software FORGE2005Rwich deals with large deformation of elastoplastic and elastic-viscoplastic behaviour. We use a master-slave algorithm with penalization and a Coulomb friction law to deal with multimaterials contact. Automatic remeshing enables mesh degeneration to be avoided during high plastic deformation. The Lemaitre model [1] is coupled to the mechanical behaviour in order to model damage. When the damage value reaches a critical value within an element, this element is deleted from the mesh, i.e. the kill-element technique [2]. The SPR process as well as other mechanical joining techniques are modelled and compared with experiments, both in terms of final geometry and joining load-displacement curves.

At the end of the SPR process, it is possible to export mechanical fields (stress, strain, damage, etc.) on a 3D mesh in order to perform a 3D structural analysis test [3]. This operation is particularly complex since that, instead of three initial parts, we now have four different parts (the upper sheet has been split into two different parts). This transfer is performed in three stages: first a 3D geometry is extrapolated from the 2D axisymmetric final geometrical configuration of the SPR process. A new 3D mesh is then generated for each part of the SPR joint. Finally, final mechanical fields (residual stresses, damage) of the SPR simulation are imposed as initial fields for the 3D structural analysis simulation. An interpolation technique based on proximity is used to transfer these fields from the former 2D mesh to the new 3D mesh. A 3D shearing test is then performed using Forge3R(Figure 1).

An automation of this technique is envisaged in order to optimize the final mechanical strength of the riveted joint by testing different rivets and different lower dies.

1

J.A Lemaitre, "A course on damage mechanics", Springer-Verlag, 1996.

2

P.O. Bouchard, J. Signorelli, R. Boussetta et L. Fourment, "Damage and Adaptive Remeshing applied to 3D modeling of blanking and Milling", 7th International Conference on Computational Plasticity (COMPLAS), Barcelona (Spain), April 7-10, 2003.

3

P.O. Bouchard and P. Lasne, "Numerical modelling of riveted joint structures - From riveting process modelling down to structural analysis", Int. J. of Forming Processes - Special issue on forming processes - accepted for publication.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £140 +P&P)