Papers by Author: Gabriel M. Regino

Abstract: Residual stresses in titanium alloy samples that were subjected to shot peening followed by fretting fatigue loading were investigated using a combined experimental and numerical analysis procedure based on the concept of eigenstrain. Fretting fatigue loading was carried out in the pad – on-flat geometry using the Oxford in-line fretting rig. Flat-and-rounded pad shape was used to introduce the contact tractions and internal stress fields typical of the target application in aeroengine design. The specimens were in the shape of bars of 10mm square cross-section shotpeened on all sides. Both the pads and specimens were made from the Ti-6Al-4V alloy. Small remote displacement characteristic of fretting fatigue conditions was applied in the experiments. The residual elastic strains in the middle of the pad-to-sample contact and near the rounded pad edge were measured using synchrotron X-ray diffraction on Station 16.3 at SRS Daresbury. A combination of finite element analysis and the distributed eigenstrain method was used in the simulations. Commercial finite element analysis software, ABAQUS ver 6.41, was used to build the finite element model and to introduce the residual stresses into the model using eigenstrain distributions via a user-defined subroutine. In an unfretted shot peened sample an excellent agreement of residual stress profiles was obtained between the experimental data and model prediction by the variational eigenstrain procedure. In a fretted sample the residual stress change due to fretting was observed, and predicted numerically. A good correlation was found between the FE simulation prediction and the experimental data measured at contact edges.

Abstract: Residual stress can be found in engineering components as a result of non-uniform plastic strain introduced through a variety of manufacturing processes such as rolling, casting, hot forging, cold working, shot-peening, laser shock peening, welding, etc. The numerical simulation of the resulting residual stress field requires the use of sophisticated coupled microstructural and thermomechanical models that rely on profound understanding of the constitutive laws and detailed knowledge of material parameters. In practice this level of understanding is not generally available, leading to the use of simplified models that are unable to reproduce or predict reliably the real residual stress distributions. This leads to the necessity of using increased safety factors and utilising overly conservative designs. A rational approach to the description of residual stress states is proposed that relies on the use of eigenstrain distributions as sources of residual stress. The problem of residual stress evaluation can then be replaced by the problem of determining the underlying eigenstrain distribution. An approach to this problem is proposed based on a simple variational formulation. Some examples of its application are shown, and the difficulties and challenges that may arise are discussed.