Search

International audience; Wire arc additive manufacturing (WAAM) is emerging as the main additive manufacturing (AM) technology used to produce medium-to-large-sized thin-walled parts (order of magnitude: 1 m) at lower cost. To manufacture a part with this technology, the path planning strategy used is the 2.5D. This strategy consists in slicing a 3D model into different planar layers parallel to each other. The use of this strategy limits the complexity of the topologies achievable in WAAM, especially those with large variations in curvature. It also involves several start/stops of the arc as it passes from one layer to another, which induces transient phenomena in which the control of the supply of energy and matter is complex. In this article, a new manufacturing strategy to minimize the start/stop phases of the arc to one unique cycle is presented. The goal of this strategy, called "Continuous Three-dimensional Path Planning" (CTPP) is to generate a continuous trajectory in spiral form for closed-loop thin parts. An adaptive wire speed coupled with a constant travel speed allow a modulation of the deposition geometry that ensures a continuous supply of energy and material throughout the manufacturing process. Using the 5-axis strategy coupled with CTPP allows the manufacture of closed parts with a procedure to determine the optimum closing area and parts on non-planar substrates useful for adding functionalities to an existing structure. Two geometries based on continuous manufacturing with WAAM technology are presented to validate this approach. The manufacturing of these parts with CTPP and several numerical evaluations have shown the reliability of this strategy and its capacity to produce complex new shapes with a good geometrical restitution, difficult or impossible to reach today using 2.5D with WAAM technology.

International audience; A thermo-mechanical simulation of the wire + arc additive manufacturing (WAAM) process is presented in this work. The simulation consists in the deposition of 5 successive layers of 316 L stainless steel on a 316 L base plate. The thermo-mechanical analysis is solved in two dimensions under plane stress assumption. Nonetheless, the metal addition is taking into account in this numerical analysis. An increment of material is added at each time step. This numerical approach allows reducing the computational time. The temperature and residual stress fields are computed at each time step. Two patterns of deposition strategy are also investigated. It is shown that the longitudinal stress varies mainly along the vertical axis. A sample with 5 overlaid layers has been scanned with neutron diffraction technique in order to measure the final residual stresses. Both numerical and measured residual stresses are in good agreement. The Aster finite element software is employed for the numerical analysis.

International audience

International audience; Welding processes imply rapid solidification, in presence of high thermal gradient and strong fluid flows in the weld pool. This work presents, in the case of welding, an analysis of the coupling between solidification mechanisms and fluid flows at the macroscale and the microscale. An experimental setup was designed in order to observe in situ a fully penetrated weld pool generated on a Cu30Ni plate with a GTAW torch. At the macroscale, observations are carried out by three cameras: two cameras recording in visible light the top and the back side of the whole weld pool, and one infrared camera catching the thermal field on solid part at the back side. At the microscale, a high-speed camera, mounted with a microscope lens, is used to observe dendritic growth and fluid flows at the back side trailing edge of the weld pool. The observations provide a lot of data allowing analyses and discussions on the correlations between solidification mechanisms and fluid flows, with respect to welding parameters. Then, the experimental results are compared to theoretical results obtained from analytical modelling, in order to discuss the possible limitations of models and try to better understand the coupling between physical phenomena.

International audience; In this paper, non-contact measurements are presented to observe droplet oscillations and to measure surface temperature during Pulsed GMAW. The techniques enable to investigate in situ the behavior of mild steel liquid metal surrounded by Ar + 8%CO2 plasma and then to measure temperature dependence of surface tension. Different temperatures are obtained by changing the process parameters. Surface tension values are a little higher than values found in the literature but the results demonstrate the interest of the developed setup. To conclude, a non-dimensional analysis is carried out, based on experimental results, to investigate the relative importance of the different phenomena acting during the transfer.

International audience; The purpose of this article is to investigate the influence of growth kinetic models and nucleation models on the grain structure predicted by cellular automata (CA) during bead on Cu30Ni thin plate. Temperature field is obtained with the computation of two-dimensional finite element model (FE). Based on these temperatures, a CA model simulates the evolution of the envelope of grains along the solidification front. It is shown that for the same temperature field, the grains structure is modified if the growth model changes. The influence of the nucleation law parameters is also discussed from a modelling point of view. Experimental size of the weld pool are compared with thermal results and EBSD maps are compared with grain structure prediction obtained with the CA algorithm.

International audience; Wire Arc Additive Manufacturing (WAAM) has the possibility to build metallic structures in 3D space. WAAM system is based on welding process to deposit metallic material and on a robot that moves the welding torch to add material at a given position. For large skeleton structures, it was chosen to deposit material point by point. Welding process induces uctuations.To be fully scalable, two main features must be taken into account. First, monitoring of the process is necessary. Local control on the geometry of the deposition must be used to reach the nal shape. Secondly, some deposition strategies must be implemented to manage branch intersections. To reach these two objectives, an adaptive and modular slicer and a process manager have been developed in order to implement this control. It allows us, if an error occurs during the deposition, to change the position of the eector. To obtain the desired geometry, the CAM software have to be able to, (i) do a slicing during the additive process of the part with a variable deposit height in order to take into account variation of the deposition process and (ii) manage the deposition strategy at intersection to output the position of the torch.

International audience; This paper investigates the mechanisms that contribute to the spreading of liquid metal macro-drop deposited during Stationary Pulsed Gas Metal Arc Welding on an initially cold solid workpiece. Surface tension and inertial effects take an important part in the behaviour of the liquid metal macro-drop, but in this configuration the influence of energetic effects can also be significant. The experimental results are discussed in the light of dimensional analysis in order to appreciate the influence of the process parameters and the physical mechanisms involved on the spreading of a macro-drop. A law is established to model forced non-isothermal spreading.

International audience; The relation between fluid flows, solidification mechanisms and welding parameters during GTA weld­ ing of a Cu30Ni plate is investigated. An experimental study, based on in-situ observations of the trailing edge of a fully penetrated weld pool is carried out. The experimental setup includes: (i) a high-speed camera used to observe the solidification front and the fluid flow at microscale, (ii) two cameras placed over and below the plate used to observe the entire weld pool on the top side and back side respectively (iii) and an infrared camera used to monitor the tempe rature field on the back side. Ali these observation devices are employed to do measurements that allow us to explore and better understand the influences of the welding parameters on the solidification and the fluid flow behaviours, as well as between the different physical phenomena being involved.

International audience; Several compositions of metal cored filler wire were manufactured to define the best welding conditions for homogeneous welding, by Gas Metal Arc Welding (GMAW) process, of a modified AISI 444 ferritic stainless steel dedicated to automotive exhaust manifold applications. The patented grade is know under APERAM trade name K44X and has been developed to present improved high temperature fatigue properties. All filler wires investigated contained 19%Cr and 1.8%Mo, equivalent to the base metal K44X chemistry, but various titanium and niobium contents. Chemical analyses and microstructural observations of fusion zones revealed the need of a minimum Ti content of 0.15% to obtain a completely equiaxed grain structure. This structure conferred on the fusion zone a good ductility even in the as-welded state at room temperature. Unfortunately, titanium additions decreased the oxidation resistance at 950°C if no significant Nb complementary alloying was made. The combined high Ti and Nb additions made it possible to obtain for the welded structure, after optimized heat treatment, high temperature tensile strengths and ductility for the fusion zones and assemblies, rather close to those of the base metal. 950°C aging heat treatment was necessary to restore significantly the ductility of the as welded structure. Both fusion zone and base metal presented rather homogenized properties. Finally, with the optimized composition of the cored filler wire - 0.3 Ti minimum (i.e. 0.15% in the fusion zone) and high Nb complementary additions, the properties, including the thermal fatigue strength, of the K44X assemblies are excellent.

International audience; Seven compositions of metal cored filler wires for Gas Metal Arc Welding (GMAW), containing the same weight percent of chromium (Cr) and molybdenum (Mo) as 444 steel, but with different titanium (Ti) and niobium (Nb) contents were investigated. Experimental results pointed out that the filler wire Ti content required to be twice time more than the amount expected in the deposited metal. This was due to the low Ti transfer ratio during arc welding. Moreover, Ti increased the wetting angle and promoted penetration. It was supposed that Ti affected the weld pool surface tension what led to inward Marangoni convection. Columnar to equiaxed grain transiton (CET) was also promoted thanks to the precipitation in the weld pool of Ti rich refractory compounds which act as heterogeneous nucleation sites for equiaxed grains. A minimum 0.3% Ti was required in the filler wire to form a completely equiaxed grain structure in the fusion zone. Niobium in the filler wire did not seem to have any effect on penetration, wetting and grain structure of the fusion zone.

This article describes an image analysis algorithm used to detect profiles during arc welding processes. The new algorithm is an aggregation of image processing (segmentation, filtering), computational geometry (alpha shape) and graph theory (cycle detection). It allows to extract precise geometrical profile entities, whether open or closed contours, that could be used for the monitoring of the welding process. The algorithm is shown to be really efficient and could be used for real-time monitoring of gas metal arc welding process.

International audience; The paper presents an apparatus to measure surface temperature distribution of liquid metals during fusion processes. The apparatus is based on dual wavelength radiation thermometry and is designed to measure temperature from 1500 to 3000 K. The pyrometer is based on standard optical parts and industrial CCD cameras. Uncertainties are analysed on the base of the radiometric equations. To insure relative precision in the measurement, a calibration procedure is conducted on tungsten ribbon with known emissivity in the temperature range. This procedure is also used to discuss the influence of wavelength couples. To conclude, experimental measurements on pure laser melted materials are conducted to assess the validity of the apparatus and methodology.

International audience; Feed rate effect on scallop height in complex surface milling by torus-end mill is rarely studied. In a previous paper, an analytical predictive model of scallop height based on transverse step over distance has been established. However, this model doesn’t take feed rate effect into consideration. In the present work an analytical expression of scallop height, including feed rate effect, is detailed in order to quantify feed rate effect and thus to estimate more precisely the surface quality. Then, an experimental validation is conducted, comparing the presented model predictions with experimental results. Actually, the share of the scallop height due to feed effect is highly dependent on the machining configuration. However, most of time, the feed effect on total scallop height values is far from being negligible.

International audience; A two step modeling for predicting the grain morphology − columnar or equiaxed − of the melting zone in welding is proposed. The first step consists in numerically compute, using the Finite Element Method, the thermal fields around the weld pool boundary, according to the welding parameters. The second step consists in a post-treatment of the thermal field in order to compute the undercooling in the liquid all along the solidification front of the columnar grains, and finally to compute the fraction of equiaxed grains formed in the undercooling zone. The modeling takes into account the effect of the convection phenomena on the undercooling, and is based on an heterogeneous nucleation model specially adapted to steels inoculated with titanium. The nucleation model involves the size distribution of Ti-rich refractory compounds formed in the weld pool, that act as heterogeneous nucleation sites. A method to identify this distribution is also presented. The comparison of the modeling with experimental results of welding with a Gas Tungsten Arc Welding process has proved its ability to predict the columnar to equiaxed transition of a ferritic stainless steel inoculated with Ti.

International audience; A two-scale in situ observation setup has been used to investigate the physical phenomena around the solid/liquid interface in welding. A spot gas tungsten arc welding process is used to observe weld pool solidification phenomenon after arc extinction. The solidification front and the fluid flow in the weld pool have been observed at a microscale by a high speed camera. The whole weld pool and the surrounding base metal have been filmed at a macroscale by an infrared camera to estimate temperature fields around the weld pool and bead shape. A qualitative and quantitative dataset was extracted from these observations that involve useful results for further improvements of the understanding of weld solidification mechanisms and to enhance solidification models.

Pour repondre aux nouveaux defis industriels, les constructeurs travaillent sur le developpement de concepts innovants pour reduire le poids des vehicules par l'emploi de materiaux legers, comme les alliages d'aluminium. Ces innovations imposent une amelioration constante des procedes de fabrication, en particulier en soudage. L'accroissement des vitesses de soudage sur les alliages d'aluminium provoque l'apparition de defauts. La fissuration a chaud est l'un des plus severes. Toutefois, la comprehension de ce phenomene est un probleme complexe faisant intervenir le procede, le materiau et la geometrie de l'assemblage. Les methodes experimentales d'investigation de ce defaut (D. Cartaud, Soudage et techniques connexes 49 (1995) 17-28) sont nombreuses, mais il s'avere souvent difficile d'interpreter les parametres influant sur l'initiation de la fissuration. Quant a la simulation numerique du soudage et l'elaboration d'un critere de fissuration (M. Rappaz, J.-M. Drezet, M. Gremaud, Metall. Mater. Trans. A 30 (1999) 449), les etudes se poursuivent encore tant le sujet est complexe. Dans ce contexte, il semble utile de developper une simulation numerique coherente d'un essai simple de fissuration en vue d'ameliorer le procede de soudage. Ce document presente brievement les parametres influant sur la fissuration a chaud en soudage pour un alliage d'aluminium 6061. Ensuite, une reflexion sur les lois de comportement a ete menee pour traduire au mieux les sollicitations de la zone pâteuse. Nous terminerons sur l'application de cette modelisation a un essai simple de fissuration, avec pour objectif d'evaluer l'effet du materiau et de la structure sur la fissuration a chaud et de proposer des voies d'optimisation du procede.

International audience; An experimental setup is presented in order to obtain experimental data during solidification of a static weld pool after arc extinction with a GTAW process. Several devices have been setup to extract three kinds of measurements: (i) solidification front velocity (ii) fluid flow velocity at the vicinity of the front (iii) temperature field in the solid part. A high-speed camera is used to film the interface during welding at microscopic level and an infra-red in order to take thetemperature field around the weld pool in the solid part. After processing and calibration of the videos, the experimental results are compared to theoritical results founded on an adapted model from the KGT [1] and from the one of Gandin et al. [2]. All the tests are done thin plate of Cu-30wt.%Ni.

Damage in heterogeneous model materials was measured using high-resolution X-ray absorption tomography. The material consisted of an aluminium matrix containing 1% and 4% of spherical ceramic particles acting as nucleation sites for an interface decohesion mechanism of damage. The damage initiation stage was quantified using the global population of particles in the 4% material. A strain path change experiment was then applied to the 1% material. The sample was first deformed in tension in order to create elongated cavities and then compressed at 45 ∘ to rotate and close these cavities. The results of a model based on the Rice and Tracey approach accounting for the presence of particles inside the cavities and calculating their rotation with assuming a linear hardening plastic behaviour of the matrix were compared with the observations. The model was modified to account for the damage initiation phase. It was shown to give a good global prediction of the void volume fraction provided that the physical, mechanical and morphological information are corresponding in the experimental and the model cases. The cavity rotation experiment was also shown to compare well with the calculation although only one cavity was sufficiently opened after compression to allow the comparison.

A phenomenological ductile damage model is developed to take into account the growth and the changes of orientation of defects in a undamaged material matrix under large plastic straining. This constitutive model is based on a skew-symmetric tensor-valued function, originally presented by Wang C C (1970 New theorem for isotropic function part 1 and part 2 Arch. Rational Mech. Anal. 36 166–223) and modified for anisotropic functions by Boehler J P (1978 Lois de comportement anisotrope des milieux continus J. Mecanique 17 153–90). A finite element model of a three-dimensional unit cell containing a tilted ellipsoidal void is used to identify the constitutive parameters of the proposed model. Then, this law of orientation change is compared with the two-dimensional model of Bilby B A and Kolbuszewski M L (1977 The finite deformation of an inhomogeneity in two-dimensionnal slow viscous incompressible flow Proc. R. Soc. A 355 335–53) deduced from the original work of Eshelby J D (1957 The determination of the elastic field of an ellipsoidal innclusion and related problem Proc. R. Soc. A 241 376–96). In the proposed ductile damage model, the radius changes of the void are based on the modified version of the Rice and Tracey void growth law presented by Thomason and adapted for the transformation of an ellipsoidal void of any orientation in the sound matrix. The new proposals are checked with different non-linear finite element analyses.

National audience

International audience; Hot cracking phenomena in welds are still a major issue to choose suitable welding conditions. Solidification cracking in welds depends on microstructure morphology, thermal and mechanical loadings. It is still difficult to discriminate the influence of the different phenomena because they are all connected. To investigate solidification cracking, numerical modelling that integrates most of these phenomena is developed in this paper. Thermal and mechanical fields are computed with a finite element model at the process scale. The thermal field is combined with a cellular automata (CA) to predict microstruc-ture at a meso scale. Based, on this prediction, the intergranular network is extracted. Finally, the pressure in the intergranular fluid network is computed during solidification. The fluid flow model along grain boundaries predicts an eventual nucleation of pore/void in the remaining liquid by investigating the maximum pressure drop in the solidifying microstructure. The numerical model is able to describe some hot cracking phenomena observed during experimental tests.

International audience

Liquid droplet deposition on solid surfaces has an important role in the industrial and research activities. The behaviour of such deposit is influenced by volume and interfacial phenomena and involves a large number of mechanisms such as gravity effect, mass transfer, capillary forces, and wetting. For the case of metal deposition the analysis of the problem is more complex because of the importance of thermal effects, involving steep gradients and phase changes. A unique experimental approach is presented in order to study the evolution of the spreading of a large drop of liquid metal called “macro-drop”. The objective of this work is to supply qualitative and quantitative information during the deposit of liquid metal in relation with process parameters. The overall shape of the macro-drop, especially its spreading and contact angles are studied in detail. The gradual spreading of the macro-drop is mainly governed by mass and heat transfers. The initial rapid spreading is due to kinetic energy of depositing droplets and direct arc heating on the solid target. All experimental results are analysed in the light of process parameters to identify the physical mechanisms involved and appreciate their effects on the behaviour of such a macro-drop.

International audience; Hot cracking in welding is a complex phenomena due to coupling between process, metallurgy and mechanical loading. In this paper, a coupled approach that integrates microstructural prediction based on the process simulation and criteria at the microstructural scale is proposed. The criteria is based on some development of Rappaz et al [1] and takes into account the influences of grain morphology, mechanical and thermal fields induced by welding, on hot cracking. To integrate all factors influencing hot cracking, a methodology is developed that couples process simulation, simple microstructural prediction and criteria along columnar grains. The criteria based on the microstructure behavior is able to predict crack onset location in columnar grains zones.

Defects control such as hot cracking or solidification cracking, in aluminum alloys welding is an important industrial issue and must be carefully examined. This phenomenon is a complex problem involving process, material and mechanical loading due to clamping. Several tests have been previously developed in order to characterize the material propensity to hot cracking. The purpose of the present work is to study, using a new hot cracking test and numerical simulation, the relationship between mechanical and metallurgical factors in order to better identify the parameters leading to hot tearing during welding. The originality of the test presented here is that an external stress is applied on the test specimen parallel to the welding direction. The advantage of this test, compared to others like Varestraint test, is its simplicity and therefore its possible for an industrial use. A weld seam is made with a Tungsten Inert Gas (TIG) arc welding on a thin sheet of aluminum alloy (6061). The crack initiation occurs once steady state thermal conditions are reached. The present test enables to distinguish between the structural effects on a global scale and the microstructural effects on a local scale. Microstructure control is made possible by adjusting welding power, welding speed and sample geometries. The grain structure which is characterized by the shape, size and the growth direction and which depends on welding current and speed plays a crucial role in the crack initiation. Microstructural features are observed using high speed camera recording and post mortem micrographs. Mechanical factors are varied by adjusting the welding parameters and the applied initial stress. The relationship between welding parameters, grain morphology, and sensitivity to hot cracking are discussed. Experimental measurements and numerical results will help to better determine global and local conditions at the onset of hot tearing and to compare those conditions using existing hot tearing criteria.

International audience; Dans le but de réduire la consommation d'énergie, les constructeurs travaillent sur l'emploi de matériaux légers, comme les alliages d'aluminium, pour diminuer le poids des véhicules. Ces modifications imposent une amélioration constante des procédés de fabrication, en particulier en soudage. L'accroissement des vitesses de soudage sur les alliages d'aluminium provoque l'apparition de défauts tels que la fissuration à chaud. La compréhension de ce phénomène est un problème complexe faisant intervenir le procédé, le matériau et le chargement mécanique. Les méthodes expérimentales d'investigation de ce défaut sont nombreuses, mais ces essais s'avèrent souvent difficiles à interpréter. L'objectif de ce travail est d'étudier, au travers d'un essai simple, la relation entre la microstructure générée par l'opération de soudage et la sensibilité à la fissuration, afin d'améliorer les modes opératoires de soudage. Un essai original a été développé consistant à appliquer une précontrainte de traction contrôlée dans la direction de soudage sur une tôle d'alliage d'aluminium 6061 de faible épaisseur, avant de réaliser une ligne de fusion avec le procédé TIG. Cet essai permet l'initiation de fissures en régime de soudage établi, au sein de la zone pâteuse où des mécanismes complexes de solidification sont mis en jeu. La morphologie des grains joue un rôle crucial dans l'initiation de la fissure. Elle est fonction de l'intensité et de la vitesse de soudage, ainsi que de la géométrie de l'éprouvette. Elle se caractérise par la forme, la taille mais aussi la direction de croissance des grains qui influencent l'initiation de fissures. Des observations in situ, à l'aide d'une caméra rapide, et post mortem, à partir de macrographies et de micrographies, ont permis d'analyser la microstructure et les fissures éventuelles de chaque éprouvette. Un scénario d'initiation de fissures a ainsi pu être proposé, en fonction de la morphologie des grains. Les relations entre paramètres de soudage, morphologies de la structure de grains générée, et sensibilité à la fissuration à chaud ont été discutées. Enfin, à partir de résultats expérimentaux, des solutions d'optimisation du procédé seront proposées en tenant compte des interactions procédé/ matériau.

National audience

Owing to their low ductility at room temperature, heat resistant cast austenitic stainless steels are very sensitive to weld cracking. Cracks are formed in brittle zones of the heterogeneous microstructure constituted by the carbide rich interdendritic spaces. In order to identify the operating factors affecting the cracking propensity of such steels during welding, a three step method, based on numerical simulation, is presented. First, the macroscopic emperature and stress fields are determined by a finite element calculation, considering a homogeneous material. In a second step, a localisation criterion is defined to identify critical zones, according to the macroscopic fields determined in the first step. Then, the heterogeneous microstructure of the cast steel is modelled in a 'representative cell', and the thermomechanical history of the critical zone previously identified is applied as boundary condition, in order to determine the local stress field. The cracking is then supposed to occur when the maximal principal stress in the interdendritic zone of the representative cell reaches the cleavage stress of the carbides, determined using SEM in situ tensile tests. This method is used to predict cracking during multipass weld repair of bulk samples, under various welding conditions. A comparison is carried out between experiment and simulation to validate the method.

Non renseigné

A numerical model allowing the prediction of the reaction layer size produced during the generation of steel-aluminium assemblies by TIG reactive wetting is developed. The Fe 2 Al 5 growth produced along the interface is governed by the heat transfer in the structure caused by the welding energy deposed by the TIG process. A process modeling coupled with the growth law of the Fe 2 Al 5 phase allows to predict the thickness and the width of the reaction layer., Le développement d’un modèle permettant la prédiction des dimensions de la couche de réaction formée lors de l’élaboration d’assemblages acier-aluminium par mouillage réactif TIG est présenté. La croissance de la phase Fe 2 Al 5 formée à l’interface est régie par les transferts de chaleur au sein de la structure générés par l’apport d’énergie du procédé TIG. Une modélisation du procédé couplée avec la loi de croissance de la phase constituant la couche de réaction permet alors de prédire l’épaisseur et la largeur de cette couche.

International audience; We propose here a model to describe the rupture of woven ply laminated structures. The validity scope of this model depends on the 'diffused damage' phase up to the first intra-laminar macro-crack only (first ply failure model). The model is based on a previous continuum damage approach [Hochard C, Aubourg PA, Charles JP. Modelling of the mechanical behaviour of woven fabric CFRP laminates up to failure. Compos Sci Technol 2001;61:221-30] and a new non-local fibre rupture criterion. This non-local criterion uses a ply scale characteristic volume of ply thickness and in plane area depending on the meso-structure of the ply. In the case of the balanced woven ply studied here, a simple circular area is chosen and identified on a test showing a stress gradient. The local shear damage and inelastic strain evolution laws and the non-local fibre rupture criterion have been integrated in Abaqus. Numerical results compared to tests for woven ply laminated structures show the efficiency of the proposed approach.

A non-linear cumulative damage model for woven ply laminates under static and fatigue loads is developed in this paper. The validity scope of this model depends on the ‘diffuse damage’ phase up to the first intra-laminar macro-crack only (first ply failure model). The model, which is based on a continuum damage approach (CDM) and a non-local fibre rupture criteria, was implemented in a FEM code. Finally, full field strain measurements are used to compare experiments and computations for a plate with an open hole submitted to a fatigue tensile load.

Inexistant

Inexistant

International audience

International audience

Resume Cette Note traite de l'etude experimentale et de la modelisation de la rupture fragile d'un stratifie quasi-isotrope. On met en evidence que la rupture de ce materiau ne peut etre correctement estimee par un critere local en presence de forts gradients de contraintes. Un critere non local est identifie et applique a differentes structures. Les resultats experimentaux sont en bon accord avec ceux issus de la modelisation. Pour citer cet article : N. Lahellec et al., C. R. Mecanique 333 (2005).

Inexistant

International audience; In this work, a three-dimensional elasto-plastic finite element model is used to simulate the rail mechanical response to the action of a rolling wheel. Plastic flows in the near-surface layer are computed as a function of the traction coefficient. The depth and the distortion of the plastically deformed layer are analysed versus this traction coefficient and correlated qualitatively to the microstructural observations of plastic flows in real railheads and existing theoretical studies. The deviations of the numerical results with reality are discussed in terms of the presence of a solid third body layer, neglected in the simulations.