Inertia friction welds between nickel superalloy components: Analysis of residual stress by eigenstrain distributions

Abstract

Machining, surface treatment, plastic forming and stretching, welding, and other manufacturing processes introduce residual stresses and distortion into work pieces and engineering components. These phenomena exert a significant influence on the behaviour of components affecting the response to thermal/mechanical in-service loading, e.g. in terms of crack initiation and propagation under the conditions of creep and fatigue, thus ultimately affecting their durability. In the present study, the inertia friction welding process is considered that is used for butt joining of hollow cylindrical components, such as shafts and drums. An inverse eigenstrain framework is used for the interpretation of neutron diffraction measurements in terms of the underlying eigenstrain distributions. Eigenstrain distributions that describe the nature of permanent inelastic deformation are found by minimizing the sum-of-squares measure of the disagreement between model prediction and experimental measurements of residual elastic strains. Experimental data obtained from neutron diffraction measurements are used in an inverse solution scheme in order to determine the underlying eigenstrain (strains permanently ‘locked in’) that give rise to the residual stress state. Once these are found, approximate reconstruction of the complete stress tensor within the entire component becomes straightforward. Eigenstrain distributions are first obtained for reduced size test specimens which have been characterized in detail using neutron diffraction. Subsequently, the eigenstrain distributions are scaled and applied to more complex, full-size real engine components, with scaling factors adjusted to match surface hole drilling measurements.