Abstract
Inertia friction welding (IFW) is a process used to create joints with high geometrical accuracy and near net shape form. To cope with the complex phenomena occurring during welding, the majority of available studies have analysed the interaction of the workpieces to be joined under simplified conditions, in which the influence of machine assembly tolerances, spindle dynamics and system compliance have been neglected. Among the dimensional properties, the headstock-tailstock concentricity is particularly important to assess the conformity of the weld, for this reason, a novel approach was developed to investigate the physical causes behind the evolution of the radial misalignment between the two workpieces, conventionally referred to as radial runout. First an inverse approach to evaluate the equivalent pressure distribution at the weld interface and the equivalent process loads was implemented starting from the experimental data of radial runout, headstock angular speed and strain extracted with a custom monitoring system during a set of steel welds. The results showed a large variability of the pressure distribution in circumferential direction and non-axisymmetric load components in particular during the conditioning and burnoff phases. Then, the equivalent process loads were used as an input for a Timoshenko beam dynamic representation of the spindle. A good agreement between the model and the experimental data was observed with an average relative error in the radial runout of 0.085. From these results, it was possible to conclude that the lack of axisymmetry in the load components has to be attributed mainly to the misalignment between two workpieces, while the irregular runout to compliance of the system to the non-ideal process loads.
Highlights
The need for high-quality joints in materials difficult to weld, such as nickel-based superalloys, has encouraged the research in solid-state friction-based processes
The loads were used as input for a 3D dynamic beam representation of the machine spindle, the outcomes of which were validated against the experimental data of strain and radial runout extracted with a custom monitoring system [22]
These were converted into equivalent process loads and used as input for an optimisation which adjusted the values of pressure to minimise the error in two objective functions: an FE model of the monitoring system (1.3), in which the strain induced by the equivalent loads were compared to the experimental strain and an analytical static representation of the spindle (1.4), in which the deflection induced by the equivalent loads were compared to the experimental data of runout
Summary
The need for high-quality joints in materials difficult to weld, such as nickel-based superalloys, has encouraged the research in solid-state friction-based processes. Inertia Friction Welding (IFW) has found several applications due to its good repeatability, ease of automation, no requirement for shielding gas, filler material or vacuum, and a lack of bulk melting [1]. IFW is conventionally divided into the three stages, shown, namely: conditioning, burnoff and consolidation. The conditioning phase extends from the moment at which the two workpieces come into contact, conventionally referred to as part contact, to the instant in which the two workpieces are in complete contact, with all the surface imperfections abraded by friction, and the bulk deformation starts to occur. The torque caused by the friction between the two workpieces first increases due to the pro gressive increase in the contact area and drops due to the localised softening and consequent switch from a Coulomb friction (http://creativecommons.org/licenses/by/4.0/).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
