Search

The increasing demands in productivity and reliability in welding have led to the development of newer welding processes. These processes frequently have superior deposition rates; however, they might lead to a wider heat-affected zone (HAZ) which might impair their mechanical properties. Here, a promising process termed the double cold wire gas metal arc welding (DCW-GMAW) is evaluated, offering higher deposition rates via feeding two cold wires into the weld pool to improve productivity. In contrast to similar processes, the HAZ of DCW-GMAW welds are smaller than gas metal arc welding (GMAW) welds. This work reports a preliminary assessment of the feasibility of the process in terms of electrical arc characteristics and stability. High-speed video acquisition was used to analyze the metal transfer across the arc. Finally, standard metallographic techniques and Vickers hardness were used to assess the weld cross sections and microstructures. The results demonstrate the feasibility of DCW-GMAW based on the process stability at different cold wire feed rates while also offering high deposition (high productivity) of weld deposits with desirable hardness due to increased acicular ferrite content.

There has been a significant increase in the use of lightweight materials in many areas of industry and engineering, which were previously dominated by metallic components. Among these materials, polymers and fiber reinforced thermoplastics enable additional advantages including great design flexibility, low manufacturing cost, dimensional stability and most dominantly recyclability. Consequently, new technological solution for joining thermoplastic composites are crucial and attractive engineering subjects. Friction Stir Welding (FSW) proved to be a reliable method for welding metallic and polymeric materials, and further studies regarding thermoplastic based composites are required, especially concerning tool design and process optimization. A short fiber reinforced thermoplastic composite was used as the base material, to study the effect of the welding parameters on the joint strength. As the welding tool plays a fundamental role in this process, three different tool concepts for welding in a butt-joint configuration were tested and analyzed. After selecting the best welding tool, an L4 Taguchi Design of Experiment (DOE) was employed to optimize the welding parameters. For each welding parameters combination, welding temperature and active forces during welding were recorded to study their effect on the frictional heat generated. It was concluded that the traverse and rotational speeds have the main influence on the weld quality, as they are directly responsible for the heat generation. FSW showed promising results regarding joining thermoplastic based composites and a maximum joint efficiency of 58.8% was achieved. However, further investigation and improvement regarding the welding tool is still required.

The miniature uniaxial tensile testing is a simple yet powerful method to characterise the behaviour of a material. Based on this technique, an experimental system was developed aiming the determination of sheet metal mechanical properties, using the Digital Image Correlation (DIC) technique. A tensile-compression loading device was designed with a maximum load capacity of 2.5 kN. The device includes custom-made clamps that guarantee the alignment of the sample with respect to the loading axis. To evaluate the aptitude of miniature samples and the use of Digital Image Correlation (DIC) in this type of scale, some tensile tests were performed using different types of miniaturised geometries and the results were compared with a macro specimen reference. Additionaly, some cyclic tests were also performed, in order to evaluate the possibilities of developed prototype on the tensioncompression loading paths.

The use of friction stir welding (FSW) to join thermoplastics has proven to produce strong welds with good surface quality when compared to conventional welding methods. In this study, a Teflon stationary shoulder was developed to weld 3-mm-thick plates of high molecular weight polyethylene in butt-joint configuration. Different sets of welding parameters were chosen and tested to evaluate their effect on the weld strength. Also, in order to increase joint performance, the temperature generated during welding was measured. For that purpose, thermocouples were located underneath of the weld nugget surface to measure the generated frictional heat for different tool diameters and parameters. Tool diameter and rotational and welding speeds are the most influential parameters regarding the welding temperature; however, all the input parameters had statistically significant effect on the weld quality. Unlike FSW in metals, using this tool, the heat is generated mainly by surface contact of the rotating probe and copper sleeve than the base material. The strongest welded joint was able to withstand 97% of the force that is necessary to fracture the base material, without using an external heating source.

This article focuses on fatigue life assessment of dissimilar lap-joined polymers. The strength of the friction stir welded polypropylene-to-polyethylene specimens under cyclic loading conditions was analysed. The fatigue life of the strongest welds was compared to both the parent material and blind riveting method with single and double-riveted specimens, in order to compare the welded specimens with available commercial joining techniques. The fatigue lives of the friction stir welded and double-riveted specimens are similar, and they are of the same order of magnitude of the highest performing base material. In order to visualize the temperature variation induced by cyclic loading conditions, a thermographic camera was employed to monitor temperature distribution in the failure region during fatigue testing. For the double-riveted and the parent material specimens, thermal fatigue failure occurred due to temperature increase under cyclic loading. For these two types of specimens, thermal failure occurred when the temperature reached approximately 50 °C in the failure region. Both, the friction stir welded and single-riveted specimens behaved differently from the remaining joints without any significant temperature increase during fatigue testing. The friction stir welded joints presented a brittle failure from the retreating side of the weld nugget, which suffers from defect formation when compared to the advancing side. The single-riveted method produced the weakest joint due to the stress concentration on the drilled hole, while adding a secondary rivet in the same overlapped area created strong joints due to the load transfer among the two rivets.

This work concentrates on the characterization of the stress intensity factor range for a compact tension specimen tested under a uniaxial tensile fatigue loading condition. The experimental solution is obtained using a three-dimensional full-field optical technique, digital image correlation. The deformation field is measured and documented for distinct crack lengths. As a relevant fracture parameter, stress intensity factor is thus experimentally measured combined with a computational overdeterministic algorithm for different crack lengths. Moreover, to verify the performance of the proposed fracture model, the cracked compact tension specimen is elasto-statically resolved using advanced discretization techniques, such as the finite element method, the meshless radial point interpolation method and the meshless natural neighbour radial point interpolation method. The finite element method model is thereby analysed with ABAQUS©to enable computation of mode I stress intensity factor results based on strain energy release rate criterion for different crack measurements in addition to strain contours. Likewise, the resolution pattern is repeated for meshless methods, and analogous numerical solutions are thus obtained. Overall, the experimental and numerical stress intensity factor results are compared with an available solution (ASTM E647) exhibiting a reasonable agreement. The novelty of this investigation is the amalgamation of an experimental digital image correlation procedure with a computational overdeterministic algorithm and, most importantly, the meshless formulation performance in the linear elastic fracture mechanics.

An elasto-plastic study of a bi-failure specimen made of DP600 steel under tensile loading state is proposed in this work. As an experimental study, Digital Image Correlation (DIC) technique is used to measure the reaction force response as a function of the acquired displacement field. In addition, the strain and the stress variation on the specimen surface is also captured. The geometrical and material characteristics used to construct the numerical model are in agreement with the experimental specimen. To access the performance of the numerical techniques, the elasto-plastic problem is therefore solved with two advanced discretization formulations, Finite Element Method (FEM) and Radial Point Interpolation Meshless Method (RPIM), to obtain comparable numerical results. Numerically, the adopted elasto-plastic plane stress formulation relies on von-Mises yield criteria based on a return mapping algorithm. An iterative process of pseudo-time stepping scheme is applied and incremental displacement steps are enforced. This study uses the Newton-Raphson non-linear solution algorithm to achieve the non-linear solution. The obtained numerical results, such as force/displacement variation and effective plastic strain, are validated with the experimental DIC solution showing that the supporting computational methodology is valid and accurate.

In this study, using an elasto-plastic constitutive model, the structural response of an Aluminium alloy AA6061-T6 bi-failure specimen is investigated. The mechanical specifications and material model are fully defined. It is intended to evaluate the force variation versus the displacement field during a tensile loading test. Digital Image Correlation (DIC) approach is used to acquire the experimental data. Afterwards, a 2D model is analysed through Finite Element (FE) formulation considering an isotropic elasto-plastic regime and the obtained results are compared to DIC outcome. A reasonable agreement between numerical and experimental results was found. Additionally, the problem is numerically solved using the Natural Neighbour Radial Point Interpolation Meshless method (NNRPIM) formulation assuming an elasto-plastic behaviour for the material. An elasto-plastic return-mapping routine is defined based on von-Mises yield criteria to evaluate the stress tensor in a pseudo-time stepping scheme, which imposes displacements. Taking into account the non-linear system of equations, a non-linear Newton-Raphson algorithm is used to obtain the converged solution. The NNRPIM results have a reasonable agreement compared to the DIC solution. The obtained results imply that the proposed elasto-plastic NNRPIM procedure is feasible and robust.

This study mainly aims to determine the stress intensity factor range (SIF) for a compact tension (CT) specimen under uniaxial tensile fatigue loading state. A 2D full-field optical technique, Digital Image Correlation (DIC), is used to acquire the experimental solution. Therefore, the deformation contour is measured for several crack growth lengths. In addition, SIF is experimentally characterized together with a numerical overdetermined algorithm for different crack lengths. Besides, the strain variation with respect to the notch tip is captured. The innovation of this study is the combination of an experimental DIC procedure with a numerical overdetermined algorithm. Moreover, to assess the performance of the proposed fracture model, the problem is resolved using advanced discretization techniques, such as the Finite Element Method (FEM) and the Meshless Radial Point Interpolation Method (RPIM). Thus, the cracked CT specimen is elasto-statically modeled using above-mentioned numerical approaches. Hence, the FEM model is analyzed with ABAQUS ©, allowing to compute the mode I SIF results for different crack lengths in addition to strain contours. Likewise, the foregoing procedure is repeated for the RPIM analysis and encouraging numerical results are achieved. The SIF is determined with a maximum energy release rate criterion in front of the crack tip in FEM study, while in RPIM study, it is calculated within the same overdetermined algorithm used in the DIC study. Overall, the experimental and numerical SIF results are compared with the reported solution (ASTM E647) exhibiting a reasonable agreement.

This work is focused on the experimental determination of the stress intensity factor (SIF) using thermoelastic stress analysis (TSA) for a compact tension specimen during a fatigue crack growth test. A comparison of the stress field obtained with computational modelling, finite element method, against the experimental data obtained with the thermoelastic stress analysis under mode I loading in a fatigue test is presented. The stress field in front of the crack tip obtained with TSA, was used in William’s expansion, together with an overdetermined algorithm to calculate the SIF under mode I loading. The proposed methodology has a hybrid experimental-numerical nature where the stress intensity factor determination depends on a stress field obtained with an optical technique, TSA. The soundness of the experimentally obtained SIF solution was validated through finite element method computations.

The radial point interpolation method (RPIM), as an advanced discretization approach, is used to analyze concrete structures assuming a non-linear continuum damage model. The theoretical basis of the material, damage characteristics and the computational procedure are presented. The elastic model is extended with a non-local damage formulation for compressive stress state from the Helmholtz free energy function in relation to the damage energy release. The predictive ability of the non-local model will be compared with the local solution, which is inappropriate whenever strong strain softening is encountered. Due to the presence of damage, the Newton-Raphson non-linear algorithm is employed, creating a fictitious pseudo-time stepping scheme. The performance of the proposed model is verified with an experimental benchmark available in the literature.