No abstract

No abstract

The Friction Stir Welding (FSW) process generally induces a gradient of properties and a softer behaviour along the welded joint. To design aeronautical structures welded by FSW in fatigue, it is necessary to study the impact of this localized soft behaviour on the overall structure. In this study, the 2198-T8 hardening structural aluminium alloy is considered. Monotonic and cyclic mechanical tests are performed by combining conventional extensometric measurements with digital image correlation (DIC) to measure the local displacement fields around the welded zone. Based on these experimental data, constitutive equations are proposed and identified, zone by zone, across the welded joint. In parallel, a quantification of T1 (Al 2 CuLi) strengthening precipitates is performed in different regions of the joint with a transmission electron microscope in order to identify a relationship between the microstructure and the mechanical parameters. Finally, once all the material parameters are identified, the model is validated by a 3D finite element analysis representative of FSW samples.

A finite element approach is used to simulate the precipitation of Ni3(Al,Ti) intermetallics in nickel-based superalloys containing a low volume fraction of spherical g’ precipitates, in which precipitation occurs following nucleation and growth mechanisms. Classical differential equations of nucleation and growth are implemented in the software Comsol (formerly Femlab), to compute the number of precipitates per unit volume and their mean size. Another originality of the model is the use of thermodynamic quantities coming from phase diagram computations (Thermo-Calc), like the temperature variation of the equilibrium g’ volume fraction, and the evolution of the concentration of g’ forming elements (Al, Ti) in the matrix with the volume fraction of precipitates. Once adjusted to experiment in the case of isothermal ageings, the model can be used to simulate precipitation during complex thermal histories. Finally, automatic heat treatment optimisation procedures are proposed and tested, which can reduce heat treatment times by a factor of more than five.