The main aims of this study are to generate a methodology for the optimisation of welding procedures for plasma welding of thin sheet aluminium alloys and to investigate weld quality modelling. Emphasis is focused on the recognition and evaluation of the consecutive stages of welding procedure development and on the formulation of a generalised procedural methodology that is potentially applicable to other processes and materials. The materials under investigation were 1.6 mm thick sheet 6013, 2024 and 7475 alloys, representing the major medium and high strength heat treatable aluminium groups used in the aerospace industry. Initial experimentation generated procedures relating to specimen and equipment pre-weld preparation. Bead on plate and square butt joint trials were performed in the downhand position. The plasma keyhole mode was operated autogenously while filler wire addition was employed for the melt-in mode. Arc monitoring techniques were used to log the arc voltage and welding current values. The effect of background and control parameters on process performance and joint quality was studied and used to generate operating envelopes and reveal optimum welding conditions. Geometrical data from the melt-in joints of all alloys were employed to build joint geometry prediction statistical models. Numerical algorithms, based on the information generated by the statistical models, were used to create joint geometry optimisation techniques. Calorimetric experiments and x-ray examination of joints revealed the relationship between major operating parameters and arc efficiency and the incidence of porosity, suggesting desired welding conditions that were incorporated in the optimisation process. Finally, two software tools for Joint Geometry Prediction and Welding Procedure Optimisation were developed, incorporating the knowledge and information created during the modelling and optimisation stages. These provide the end user with a means of process parameter selection and visualisation of the influence of parameter variation on weld bead geometry.

The VPP A W process has been evaluated and optimised for the joining of high strength aluminium alloys AA-2024 T351, AA-7150 W51 and AA-7150 T651. The initial experimentation was performed with conventional AC-GTAW power supplies with limited results. Failure of the control logic to stabilise the welding process or arc extinction became an issue. Not until a suitable VPP AW power source became available were sound welding conditions developed. Still, a great effort was placed in optimising the process. Factors such as the welding torch, the electrode composition, the plasma nozzle geometry, the pilot arc current, the electrode positive duration and amplitude, the shielding gas quality and the effect of gas backing had to be thoroughly studied to achieve sound welding conditions. Subsequent to the optimisation process, operational envelopes were generated to determine the most favourable welding conditions. Based on those, reliable techniques were developed to produce sound welds on butt welded assemblies. Further on, a special plasma welding torch was developed to weld complex structures such as dual stringer panels. A number of these panels were manufactured and surpassed the damage tolerance requirements of Airbus. Hardness profiles indicated a dip in hardness in the far heat affected zone. Microstructural observation of this area revealed significant coarsening of the precipitates related to the relatively high heat input conditions of the VPPAW process. Studies which involved the reduction of the heat input were carried out. Variables such as the addition of filler wire, the use of helium as shielding gas, the combined effect of both, and the use of an edge preparation were assessed. Temperature measurements were performed at different positions along the HAZ in order to relate the thermal history with the microstructural degradation by means of cooling curves and TTT diagrams. Similar investigations were performed under low heat input welding conditions with DC-GT A W helium shielding. I Variable polarity plasma arc welding of high strength aluminium alloys. Tensile tests were performed on selected specimens in order to asses the effect of temper condition, post weld heat treatment and heat input. The results of which indicated that low heat input conditions improved the tensile properties of the far heat affected zone. With regards to temper condition it was observed that the W51 temper was more advantageous. It was also noticed that VPPA welds offered similar 0.2% proof stress and ultimate tensile strength than that of MIG welds, together with, although still very low, improved elongation properties. This piece of research has involved the welding of high strength aluminium alloys which were considered difficult to weld or even unweldable in some cases. Prior attempts were performed with MIG welding but micro voids acting as initiation sites for cracks limited the damage tolerance of the assemblies. On the contrary VPPAW has proved its ability to produce extremely high quality welds where gas porosity or micro voids were almost inexistent.

This work was motivated by the industrial and academic needs for process development of automated keyhole plasma arc welding (K-PA W) in thin Ti- 6AI-4V sheet and the control required for the process to be suitable for aerospace components. K-PAW, is a high energy density precision welding process which is a lowcost alternative to laser and electron beam welding. It is potentially capable of fabricating high integrity titanium alloy welds in aero-engine thin panel structures. However, the process has always represented a challenge owing to the complexity of welding torch configurations and the associated large number of process parameters to take into consideration. Three types of weld joint, which represent the welding fabrications in a simplified aero-engine casing component, were manufactured in the work: flat bead-on-plate, flat Tjoint and curved T -joint. A new welding procedure was developed to produce thin sheet T-joints by K-PAW, which has overcome the difficulty of operating the structurally complicated welding torch in limited space. An analytical model was experimentally validated and was employed to identify the process parametric envelopes for valid keyhole welding modes from the numerous possible parameter combinations. Weld joints were characterised in tenns of thermal history, micro-hardness and metallurgical microstructure. Distortion and residual stresses are maJor concerns associated with fusion welding fabrication. While distortions can lead to geometrical inaccuracy, paIticularly in the thin panel stl11ctures, residual stresses can combine with applied stresses to reduce the life of components. In this thesis, the out-of-plane l distortions of welded thin Ti-6AI-4V bead-on-plate and T-joint welds were measured using contact and non-contact coordinate-measuring techniques, from which the effects of welding sequence on the distortions can also be identified. Residual stress distributions in the welded sheet were detennined by synchrotron X-ray and neutron diffraction techniques. The use of synchrotron X-ray methods is relatively newly developed as titanium alloys respond weakly to neutron beams. These not only have provided improved understanding of residual stresses in thin sheet welds but also have greatly contributed to validation of finite element (FE) modelling work undertaken by other researchers.

There are many problems in welding metallurgy for which it is difficult to develop a first principles scientific model due to their complexity. A significant problem faced by today's welding engineers, is the need to relate welding parameters to the quality of the finished weld. This is usually done by experience, and the need for many experimental trials, eventually leading to optimal welding parameters. Important characteristics in the evaluation of line-pipe seam weld quality are the weld bead shape and size, which can have a significant effect on the microstructure and mechanical properties of the weldment through heat flow effects. Properties of the final weld may therefore be difficult to predict, especially quantities such as weld metal toughness, which are known to be dependent on many factors. One approach to such complex problems is to use neural networks. A neural network is an artificial simulation of the brain which models data through a learning process and stores the information as a set of rules akin to knowledge. This research is concerned with the application of neural network techniques to the prediction of the mechanical and physical properties, including the shape of the weld bead, of submerged arc line-pipe steel welds. A limited experimental investigation has been carried out using optical and transmission electron microscopy to establish an understanding of the complex microstructures that result from the welding processes used in the production of line-pipe.

Big gains in productivity are found in tandem and dual tandem pipeline welding but require highly skilled operators who have to control the position of the torch very accurately for long periods. This leads to high demands on the skills and stamina of the operators of mechanised pipeline welding systems. There is a very strong motivation to fully automate the welding process in order to reduce the required skills and to improve consistency. This project focuses on the use of through-the-arc sensing for seam following and contact-tip-workpiecedistance (CTWD) control. A review of literature reveals very little development work on arc sensing for Pulsed Gas Metal Arc Welding (GMAW-P) in narrow grooves. GMAW-P is often used to achieve optimum properties in weld quality and fusion characteristics and also positional welding capability, all of which are important factors for pipeline welding. The use of through-the-arc sensing for narrow groove pipe welding applications poses specific challenges due to the steep groove sidewalls and the use of short arc lengths, producing very different behaviour compared to V-groove arc sensing techniques. Tandem welding is also quite different from single wire techniques with both wires working in close proximity producing mutual interferences in arc signals. An investigation was conducted in order to assess GMAW-P arc signals and it was found that improved consistency, higher sensitivity and less noise was present in voltages in the peak current period (peak voltages) used for torch position control. As a result of this investigation, a CTWD and cross-seam control system was developed and tested for single and tandem GMAW-P, using a 5º narrow groove. The test results have revealed accuracies for both controls of better than 0.2 mm. CTWD control was developed by following the existent welding procedure voltage average and cross-seam control by peak voltage comparison between maximum torch excursions. Experiments were also performed to evaluate the influence of torch oscillation frequency on arc voltage behaviour and sensitivity, along with weld bead characteristics and fusion profiles. The resultant arc signal sensitivity was consistent with the results found in the literature for conventional GMAW. For GMAW-P, although no data was available from the literature for comparison, the results have shown no increase in sensitivity with the increase of oscillation frequency with the welding setup used. Bead profile analysis performed at different sidewall proximities indicated that optimum wire to sidewall proximities can be found between 0 mm and +0.2 mm, measured from the outer edge of the wire to the sidewall corner. Accurate control is required since +1 mm proximity produced poor sidewall fusion and no signal differentiation for control recognition of groove width. This work showed that negative proximities or wire proximity beyond the sidewall produce wire burn back and hence very long arc lengths, resulting in poor depths of penetration and shallower beads, with major undercut defects. In addition, this work has also shown the importance of torch oscillation width control, in order to produce accurate cross-seam control. A method is proposed to achieve torch oscillation width control by a continuous peak voltage comparison between centre and sidewall torch positions, using the optimum values of wire to sidewall proximity found and the resultant peak voltage value. This control will also provide a clear indication of actual groove width. Clearly this data can also be used to implement a system which adapts welding parameters to groove width.

No description available.