Papers by Author: Paulo Manuel Salgado Tavares de Castro

Abstract: Friction stir welding (FSW) is a solid-state joining process which emerged as an alternative technology to join high strength alloys that were difficult to weld with conventional techniques, [1]. Developments of this technique are being driven by aeronautic, aerospace and railway industries. An advantage of this joining technique is its low heat input when compared with arc welding processes. This feature allows the achievement of high mechanical properties, low distortion and low residual stresses, [2]. Also, since it is a solid-state welding process, hydrogen cracking or heat affected zone (HAZ) softening phenomena are limited. This paper presents a study of fatigue crack growth behaviour of friction stir welded butt joints of AA2024-T3, aluminium commonly used in riveted aeronautic fuselage structures. Crack growth studies are often carried out using uniform thickness joints, ASTM E647 [3]. Nevertheless, for some applications there is a need to join components with different thicknesses, which, under certain limits, can be welded using FSW. Crack growth tests on these joints are not standard. The present study concerns butt joints made using two plates with different thicknesses, 3.8mm and 4.0mm. The joints’ mechanical behaviour was studied performing static (tensile) and fatigue tests. The fatigue crack growth rate of cracks growing in different zones of the welded joint (nugget, heat affected zone - HAZ) and in base material was analysed. The microhardness profile was assessed in order to analyse the influence of the welding process in each weld zone. Further to higher static properties, welded joints present lower crack growth rate when compared with its base material.

Abstract: Residual stresses parallel to the welding direction on a cross-section of a 3 mm thick friction stir butt-welded aluminum alloy AA6082-T6 plate were determined using the contour method. A full contour map of longitudinal residual stresses on a weld cross section was determined in this way, revealing detailed information on the residual stress distribution in the inside of a friction stir weld, especially in the nugget zone. The typical M-shape, usually described for the residual stress distribution in friction stir welds, was found. The maximum residual stresses are below the yield strength of the material in the shoulder region and, outside of the welding region, low tensile and compressive residual stresses are responsible for the necessary stress equilibrium on the plane of interest. A comparison was made with the established incremental hole drilling technique on an equivalent plate for validation and good agreement of both techniques was obtained. The distribution, as well as the magnitude of the residual stresses measured by both techniques, is very similar, thus validating both the experimental and numerical procedures used for the contour method application, presented and discussed in the present paper.

Abstract: From a lean philosophy viewpoint, the huge amount of fasteners used in airframes appears to be the inverse of what the design for manufacturing should seek. New design concepts with integral joints may be produced using welding processes, simplifying and optimizing the design for manufacturing and assembly process; however, crack arrest capability may be compromised. An introductory quality function deployment (QFD) analysis applying the Cross method evaluates the main design goals in the application of new joining processes in lightweight reinforced structures. The analysis shows that new welding process can improve airframe structures provided the structural integrity is ensured.

Abstract: Tensile and fatigue crack growth tests of the 6056 T651 and T6 aluminium alloys were carried out. The fatigue crack propagation tests were performed on compact tension 4mm thick (CT) specimens, under cyclic loading with R ratios 0.1 and 0.5. The resulting data was used to predict the fatigue behaviour of stiffened panels subjected to fatigue loading under similar R ratios. The AA6056-T651 panels were fabricated using High Speed Machining (HSM) starting with 30mm thick plates. AA6056-T651 CT specimens were cut from the panels mentioned above, whereas AA6056-T6 CT specimens were machined from 5mm thick material. It was found that the AA6056-T651 (HSM material) specimens, machined from a 30mm thick plate presented higher rupture and yield stress than the AA6056-T6 material extracted from a 5mm thick plate. When tested at the same R value the AA6056-T6 specimens present higher crack growth rate than the AA6056-T651 specimens.

Abstract: Great interest is dedicated to the wheel/rail contact problem, in particular as a result of some accidents worldwide and also in Portugal. In the present work a three dimensional finite element analysis of the wheel/ rail contact problem was performed using the software ABAQUS. A preliminary study on simpler geometries was carried out, in order to identify the solution strategies giving more accurate solutions. The influence of mesh refinement, friction coefficient, and numerical techniques as Lagrange and penalty functions were analysed in the simpler cases of contact of two cylinders along a generatrix, and the contact of a cylinder and a rigid plane. The corresponding Hertz solutions were programmed using MATLAB in order to compare with the finite element analyses. The numerical procedures giving better results were later applied to the wheel/rail contact problem, using standard rail and wheels profiles used by the Portuguese Railways company (CP). The influence of small geometry variations on the stress analysis results was then studied, and the study of initiation of defects using the Dang Van fatigue criterion was performed.

Abstract: A model to determine Stress Intensity Factors (SIFs) and simulate the fatigue crack growth in stiffened structures taking into consideration residual stresses is presented in this paper. The stress field required to estimate the SIF was calculated using the Finite Element Method (FEM) considering the residual stress as an initial condition. The residual stress field redistribution as a function of crack growth is taken into account using the Abaqus software. Specimens without and with residual stresses, resulting from different welding techniques, were considered for the present study. The residual stress fields can significantly deteriorate or improve the fatigue life of the structure, depending upon the location of the initial crack; consequently these effects should be analyzed and modelled in order to better understand the consequences of the application of the considered manufacturing processes.

Abstract: A study on the mechanical characterization of friction stir welds between aluminium alloys 6061-T6 and 6082-T6 was carried out. For comparison, single alloy joints made from each one of the two alloys were also performed. The work included microstructure examination, microhardness tests, tensile tests and bending tests of all joint types. An approximate finite element model of the joint, taking into account the spatial dependence of the tensile strength properties, was made, modelling a bending test of the weldments.

Abstract: A study on the fatigue behaviour of friction stir butt welds of 3mm thick 6082-T6 aluminium alloy was carried out. Monotonic tensile and cyclic tests of welded joints and base material were performed to understand the influence of the welding process on the static and fatigue properties. Microhardness profiles were measured and fatigue crack growth curves were determined for cracks growing in different locations of the weldments. Friction stir material exhibited lower strength and ductility properties than the base material. However, an enhanced crack propagation resistance is observed.

Abstract: Cold working introduces a compressive stress field around rivet holes, reducing the tendency for fatigue cracks to initiate and grow under cyclic mechanical loading. As it is well known, for the accurate assessment of fatigue lifetimes a detailed knowledge of the residual stress profile is required. Powerful experimental and numerical tools are nowadays available for that purpose. In the present work both types of tools, X-ray diffraction and 3D Finite Element Analysis (FEA), were used in order to evaluate the residual stress profile. A comparison of experimental and numerical data is presented and discussed.