S. Larose

On the potential applications of a 3D random finite element model for the simulation of shot peening

Shot peening is a cold-working process that is used mainly to improve the fatigue life of metallic components. Experimental investigation of the mechanisms involved in shot peening is very expensive and complicated. Therefore, the Finite Element (FE) method has been recognized as an effective mean for characterizing the shot peening process and several types of FE models have been developed to evaluate the effects of shot peening parameters. However, in most of the existing FE models, the shot peening sequence and impact location were defined a priori. It is therefore the purpose of this study to consider the random property of the shot peening process. A novel 3D FE model with multiple randomly distributed shots was developed combining a Matlab program with the ANSYS preprocessor. The explicit solver LS-DYNA has been used to simulate the dynamic impingement process. Several potential applications of this novel model such as: the quantitative relationship of the peening intensity, coverage and roughness with respect to the number of shots have been presented. Moreover, simulations with multiple oblique impacts have been carried out in order to compare with results from normal impingements. Our work shows that such a computing strategy can help understanding and predicting the shot peening results better than conventional FE simulations.

Numerical simulation of the stress peen forming process and experimental validation

Stress peening forming is widely used in the aeronautics industry to induce curvatures in wing skins. Most of the investigations of stress peen forming are empirical and experimental. In this paper, a three step numerical model that can simulate this process was developed. First, an implicit Finite Element Analysis (FEA) with ANSYS where a prebending moment along the spanwise direction of the component was performed. Then, an explicit FEA with LS-DYNA simulating shot impacts on the pre-stressed component was executed in order to obtain the resulting stresses inside the component. Finally, an implicit FEA with ANSYS was performed for calculating the arc heights and the curvature radii of the component in chordwise and spanwise directions. Numerical analysis of the process shows that the prebending moments have an influence not only on the residual stress profiles but also on the curvatures of the deformed component in chordwise and spanwise directions. This model was used to establish a relationship between the prebending moment and the resulting arc heights and residual stress profiles. The numerical strategies developed in this paper supply a useful tool for studying and optimizing the stress peening process.

Prediction and Measurements of Thermal Residual Stresses in AA2024-T3 Friction Stir Welds as a Function of Welding Parameters

Friction Stir Welding (FSW) induces thermal residual stresses resulting in distortions in thin-walled structures. In order to understand and quantify this phenomenon, simulations and experiments of FSW on aluminium alloy (AA) 2024-T3 have been performed using different rotational and welding speeds. A sequentially coupled finite element (FE) model was used to study the residual stresses caused by the thermal cycling induced from FSW. The 3D FE model used temperature-dependent mechanical and thermophysical material properties. The predicted longitudinal stresses peaked at ~300 MPa and had a ‘‘W’’ profile with tensile stress peaks in the weld and compressive stresses outside the weld. In the FE model, the influence of process parameters on residual stress distribution was studied. The application of ‘hot’ welding conditions, i.e. low welding speed and high rotational speed, increased the residual stresses significantly, mainly in the transverse direction. Conversely, ‘cold’ welding conditions resulted in lower residual stresses. The magnitude and distribution of the residual stresses predicted by the FE model were validated by neutron diffraction. The results indicate a good agreement between the measured and predicted residual stresses in AA2024-T3.

Process Window for Friction Stir Lap Welding of Aluminum Alloy 6061

Precipitation-hardenable 6xxx series aluminum alloys are incorporated in many structural components with due consideration of their good combination of properties including a relatively high strength, outstanding extrudability and excellent corrosion resistance. Accordingly, AA6061 has been identified as a very good candidate material for structural lightweighting of transportation vehicles. However, the weldability of aluminum alloy (AA) 6061 by means of conventional technologies such as GMAW and GTAW methods is limited by sensitivity to solidification cracking. In this respect, friction stir welding (FSW) presents a tremendous potential for assembly of aluminum structures for the transportation industry due to the low heat involved that can mitigate crack formation and, thus, translate into improved mechanical performance of the assembly. In this work, FSW of 3.18 mm thick AA6061-T6 sheets in the lap joint configuration was investigated. This configuration is considered to be more challenging for assembly by FSW than the butt joint type due to the orientation of the interface with respect to the welding tools and the necessity to break the oxide layer on two aluminium alloy planar surfaces. Weld trials were performed to examine the influence of the FSW tool geometry and process parameters on the welding defects, microstructure, hardness and bend performance. Unacceptable material expulsion and/or significant thinning in one of the two overlapped sheets were produced under most conditions. A set of FSW tool geometries leading to a viable process operational window under which the risk of defects could be mitigated and/or eliminated was identified in this study.

Limitation of distortion in friction stir welded (FSW) panels using needle peening

Friction stir welding (FSWing) induces residual stresses and distortions in welded structures. Such residual stresses reduce the fatigue life of welded components, while the induced distortions prevent the welding of large or thin components. In the present study, needle peening was used to induce additional residual stresses in 2.3-mm thick (FSWed) aluminum alloy (AA) 2024-T3 sheets. This was done with the objective to counterbalance the welding-induced stresses and thus reduce the overall stresses and distortions. The needle peening process, which stems from shot peening, consists of hammering a surface using cylindrical spherical ended shots sliding back and forth in a treatment head. An instrumented needle peening machine was used to carry out peening on as-received (or bare) and bead-on-plate FSWed AA2024-T3 material. In both cases, the width of the peening area corresponded to that of a typical weld. The influence of the peening process parameters such as needle size, applied power and travel speed on the surface quality and magnitude of the induced distortions were evaluated. The results indicate that, by increasing the needle diameter from 1.2 mm to 2.0 mm, the peening-induced deflection on bare sheet material increased by an average value of 27% while the roughness average, Ra, decreased by an average value of 47%. It was also found that a surface finish qualitatively similar to that of conventional shot peening could be obtained by using appropriate needle peening trajectories. Finally, needle peening with an applied power of 10% was sufficient for eliminating 37% of the welding-induced transverse curvature and 82% of the welding-induced longitudinal curvature.

Finite Element Peen Forming Simulation

Shot peening consists of projecting multiple small particles onto a ductile part in order to induce compressive residual stresses near the surface. Peen forming, a derivative of shot peening, is a process that creates an unbalanced stress state which in turn leads to a deformation to shape thin parts. This versatile and cost-effective process is commonly used to manufacture aluminum wing skins and rocket panels. This paper presents the finite element modelling approach that was developed by the authors to simulate the process. The method relies on shell elements and calculated stress profiles and uses an approximation equation to take into account the incremental nature of the process. Finite element predictions were in good agreement with experimental results for small-scale tests. The method was extended to a hypothetical wing skin model to show its potential applications.