Evaluation of Error in Residual Stress Measurements by the Hole Drilling Method

We have developed a specimen configuration with a tailored through-thickness residual stress distribution with the aim of working out conditions for assessing residual stresses by both mechanical and physical methods. The procedure for producing such a specimen is based on nonuniform plastic deformation of an aluminum beam with a rectangular cross section in pure bending. During the loading process, the deformed state was monitored on the end face of the specimen by measuring the normal strain field with a digital image correlation system. The depth of the plastically deformed layer was determined to be 1.3 mm. We obtained a stress state of opposite sign, symmetric with respect to the neutral axis. A theoretical distribution of residual stresses after unloading the plastically deformed specimen was evaluated by numerical finite element calculations with allowance for physicomechanical characteristics, elastoplastic hardening, and the stress–strain curve in true coordinates for uniaxial tension of elementary specimens. Residual stresses nonuniform in the thickness direction have been studied by the hole drilling method in accordance with the ASTM E837 standard on two opposite sides of the specimen: in the tension region during deformation and in the compression region. The strain response during drilling was monitored using three-axis strain gage rosettes. The measured longitudinal residual stress component was compared to its theoretical distribution obtained numerically. In particular, the root mean square error of residual stress measurements relative to the theoretical distribution for an aluminum alloy specimen has been shown to reach 18.7 MPa. It is worth noting that larger measurement errors were obtained at small depths, characterized by relatively low strain, comparable to shot noise.