Examination of the computational model for the layer-removal method for residual-stress measurement

TLDR: In this article, an analytical solution is presented to examine the accuracy of the "layer-removal" method for measuring localized residual stresses, and it is recommended that the method can be used for measuring residual stresses for cases in which the ratio of h/a is less than or equal to unity.

Abstract: An analytical solution is presented to examine the accuracy of the ‘layer-removal’ method for measuring localized residual stresses. In this approach, strips, which may have been cut from a pipe or a plate, have strain-gage rosettes placed on one face and layers removed from the other face. The strain measurements are used to deduce the residual stress in the layers removed. The stress measured is that along the axis of the strip. It may vary rapidly with axial distance, as for example when the strip is taken from a welded part. The present analysis shows that the actual stress distribution may be quite different from that predicted by the computational model normally used in the layer-removal method. It shows that the difference increases as the ratio of the heighth of the strip from which a layer is removed to the half dimensiona of the localized residual-stress zone increases. It is recommended that the layer-removal method can be used for measuring residual stresses for cases in which the ratio ofh/a is less than or equal to unity.