Search

International audience; Among the various mechanisms which occur during impact, the strain rate effect plays a significant role on the mechanical response of layered carbon fibre reinforced polymer structure. In this work, the viscoelastic behaviour of the material is studied to introduce a strain-rate dependency. To preserve numerical efficiency the generalised Maxwell model, formulated in the strain-space, is taken as a basis. The non-linear viscoelastic behaviour is introduced by coupling the generalised Maxwell model with a pre-existing intralaminar matrix continuum damage model. The fact that the Maxwell model preserves the explicit scheme of the damage model leads to an efficient material model for impact simulations. This paper proposes a complete framework to implement the strain-rate sensitive damage model in an explicit finite element code (for low-speed impact simulations). For this purpose, the procedure of parameter identification, based on DynamicMechanical Analysis, is given. Furthermore, a challenging experimental procedure on high-speed jack device with a particular attention paid to the consistency of the results is proposed to validate the developed model.

IF=2.035; International audience; Mechanical behaviour of short-fibre reinforced composites (SFRC) can be affected by numerous damage phenomena. This paper focuses on the modelling of fibre/matrix debonding initiated at fibre tips, which is the dominant damage mechanism when a majority of short-fibres are subjected to local tensile loadings. A failure criterion for SFRC is proposed based on a critical amount of voids located in debonded areas. Very accurate mechanical states at break can be predicted for SFRC with complex fibre distribution of orientation, for different kinds of uniaxial and biaxial tensile loadings.

International audience; Crashworthiness performance of automotives by introducing new materials is of high interest based on the global mass’ reduction of cars. Bonded techniques are tested at present with respect to the reduction of the mass/energy’s ratio combined with recycling processes. In this study, a compression test program is performed on a bulk structural adhesive under a very large range of strain rates, thus [0.1; 5000]/s. Uniaxial compression tests at 0.1 and 50/s are achieved using a high-speed hydraulic testing machine. A split Hopkinson bar’ device made of PA66 is experimented so as to characterize soft materials in the upper domain of strain rates, thus [500; 5000]/s. A special mold is used for the preparation of the adhesive plates. Two sheets are prepared (thickness 4 mm with 1 MPa pressure and 6 mm with 4 MPa pressure) in order to consider the effect of the pressure during the curing process. Water jet technique is used to extract cylindrical samples from the plates with accurate dimensions. The identification of the compression behaviour of the structural adhesive is lead with objective to find material parameters of a modified G’sell model for finite element modeling of crashworthiness.

IF=2.01; International audience; A specific damage characterization method using Digital Image Correlation for semi-crystalline polymers is proposed for a wide range of strain rates. This damage measurement is an extension of the SEĖ method [16] which was developed to characterize the behaviour laws at constant strain rates of polymeric materials. This procedure is compared to the well-known damage characterization by loss of stiffness technique under quasi-static loading. In addition, an in-situ tensile test, carried out in a microtomograph, is used to observe the cavitation phenomenon in real time. The different ways used to evaluate the damage evolution are compared and the proposed technique is also suitable for measuring the ductile damage observed in semi-crystalline polymers under dynamic loading.

The use of fabric reinforced polymers in the automotive industry is growing significantly. The high specific stiffness and strength, the ease of shaping as well as the great impact performance of these materials widely encourage their diffusion. The present model increases the predictability of explicit finite element analysis and push the boundaries of the ongoing phenomenological model. Carbon fibre composites made up various preforms were tested by applying different mechanical load up to dynamic loading. This experimental campaign highlighted the physical mechanisms affecting the initial mechanical properties, namely intra- and interlaminar matrix damage, viscoelasticty and fibre failure. The intralaminar behaviour model is based on the explicit formulation of the matrix damage model developed by the ONERA as the given damage formulation correlates with the experimental observation. Coupling with a Maxwell-Wiechert model, the viscoelasticity is included without losing the direct explicit formulation. Additionally, the model is formulated under a total Lagrangian scheme in order to maintain consistency for finite strain. Thus, the material frame-indifference as well as anisotropy are ensured. This allows reorientation of fibres to be taken into account particularly for in-plane shear loading. Moreover, fall within the framework of the total Lagrangian scheme greatly makes the parameter identification easier, as based on the initial configuration. This intralaminar model thus relies upon a physical description of the behaviour of fabric composites and the numerical simulations show a good correlation with the experimental results.

To accurately predict the damage and failure evolution occurring in crashworthiness and sheet metal-forming simulations, a realistic material description is required. The parameters needed to describe such damage models are not easy to obtain and there is no direct test to achieve these parameters. An inverse method to identify the Gurson and Lemaitre damage model parameters is presented. Quasi-static and high-rate tensile tests have been carried out on thin notched specimen made of high-strength steel (THR800) from ARCELOR. The aim was to study the combined effects of strain rate and stress triaxiality on the damage material behaviour. Macroscopic measured responses are the global forces and the width at the bottom of the notch versus the elongation. An image-processing software has been developed to measure the evolution of the width at the bottom of the notch versus the elongation from the video tests. The damage parameters are identified by correlating experimental and numerical macroscopic m...

International audience; The purpose of this paper is to propose a simplified macroscopic three-dimensional cohesive element dedicated to crash engineering. At the boundary between the CCZM (Continuous Cohesive Zone Method) and the DCZM (Discrete Cohesive Zone Method) approaches, this cohesive element includes an improved constitutive model in order to take into account the viscoplaticity of the adhesive, its different behaviour in tension and compression but also its damage and failure. After presenting the model used, we will detail how to implement the new macroscopic element in the explicit time loop calculation. An example will illustrate and justify the development of such a cohesive element under dynamic loading.

These paper deals with the development of new bonded joint modelling for crash application. A new testing device has been set up on Split Hopkinson bars in order to identify adhesive’s properties in assemblies for high strain rate and for different loading angles. These tests led to the development of a new cohesive element model used for solving nonlinear dynamic problems with an explicit integration time scheme. An example illustrates and justifies the development of such a cohesive element under dynamic loading by a good efficiency and a significant saving in calculation time.

In the recent years, structural bonding takes an important place in assemblies techniques used for the automotive design. The next step to optimize the use of adhesive in car structures is to realize accurate finite element simulations of behaviour until failure of bonded joints. These kinds of calculations are only possible if fine behaviour and failure models are provided into the finite element software. In these works, tests on bulk adhesive specimens are realized to characterize the mechanical properties. 2D and 3D Digital Image Correlation are used to investigate the behaviour and failure of the shear and tensile specimens.

Inside a large displacement, rotation and strain finite element framework, a damage model is proposed to take into account the microvoid volume fraction increase in the constitutive matrix. The corresponding formulations are described. The damage model is applied to compute the porosity evolution during the multi-step hot forging of tubes.

Numerical simulations of the cold stamping process of bars are performed with pore free and porous materials. An isotropic elasto-plastic potential with three material parameters is used to compute the porosity evolution. The damage model takes into account the void nucleation, growth and coalescence. The numerical simulations of the cold stamping process of bars are carried out with ASTRID finite element code for pore free and porous material to enhance the modifications of the dimensions and bulk properties of the forging product occurring with the damage evolution.

Improvement of automotives’ crashworthiness is of high interest for governments prior to objectives focused on the reduction of passengers’ injuries. In these recent years, steel industries have studied the dynamic behaviour of tapered side rails so as to increase the capacities of energy absorption combined with light-weight aspects. On the basis of the global mass’ reduction of cars, bonded techniques have been tested with respect to the reduction of the mass/energy’s ratio. In this study, a test programme has been performed on a structural adhesive under a very large range of strain rates, thus [0.1;5000] /s. For that, a split Hopkinson bars device made of PA66 has been used so as to access to materials responses in the upper domain of strain rates, thus [500;5000] /s. To complete the expected domain of strain rates, compression tests have been done using a high-speed hydraulic machine on the same geometry from 0.1 up to 50 /s. A special mould has been machined and 2 sheets have prepared (thickness: 4 and 6 mm) in order to consider the effect of the pressure during the curing process. Water jet technique has been used to extract cylindrical samples from the sheet with accurate dimensions.

Inside a basic large strain finite element framework, computation of void volume fraction is achieved in taking into account nucleation growth and coalescence in the material matrix. The material constitutive relation involves an elasto-plastic potential and the reference values of matrix parameters may correspond to some forged low carbon steel matrixes. Computations concern patch test of axisymmetric three node element first, collar test proposed in earlier experiments to define limit curves second and pipe bulging process third. It is shown that the three main steps occuring in the most severe processes can be accurately predicted if the initial matrix parameters are well known.

Publisher Summary This chapter discusses a coupled damage-mechanical model for strain rate dependant voided material. The microvoid nucleation, growth, and coalescence are modeled. The ductile fracture is predicted at the end of the damage process. The improvement of the numerical simulation for crashes or forming processes under dynamic loading takes the prediction of the ductile rupture into account. The numerous material damage parameters must be defined to make the damage model available. An inverse method has been developed by coupling an optimizer and the finite element code. An experimental tensile test of the notched specimen is used to determine the variation of the inner radius of the specimen in function of the elongation. This measurement is compared with that obtained by numerical simulation and the material damage parameters are identified by minimizing the gap between these two measurements.

To predict damage evolution occurring under dynamic loading, a damage model is implemented inside the explicit finite element framework. The damage model is based on the description of the growth, the nucleation and the coalescence of the microvoids. The microvoid growth is related to the plastic incompressibility relation. The microvoid nucleation is either controlled by the plastic strain or by the stress. The microvoid coalescence is described by a specific function. This damage process leads to the progressive loss of the stress carrying capacity of the structure. The ductile fracture occurs when the stress carrying capacity of the structure vanishes. The sensitivity of damage volution under dynamic loading in the case of porous strain rate sensitive material is analysed using single tensile tests. The dynamic bending test of a cantilever beam with a U-cross-section is performed. The influence of the strain rate on the deformed shape and on the loss of the structure's stress carrying capacity is shown.