Flat vs. Stacked rosettes

Abstract

Special four-element flat rosettes were used to measure residual stresses and stress concentrations. The resulting data indicated that flat strain-gage rosettes are sometimes quite inadequate for measuring principal stresses. Since all equations used to determine principal stresses from strain-gage rosettes assume a single-point measurement for all elements, it was decided to evaluate, experimentally, both flat and stacked rosettes when used in different types of stress fields. The two rosette types were found to perform equally well in constant stress fields. In nonconstant stress fields where the stress gradient was linear and of moderate level, both rosette types again performed satisfactorily. When these rosettes were subjected to nonlinear stress gradients, the data from the stacked rosette were much superior. Even a small nonlinearity in the stress gradient across a flat rosette caused significant errors in the results. A mathematical model is used to evaluate the performance of flat rosettes subjected to theoretical linear and nonlinear stress gradients. The theoretical sets of strain readings were programmed into a computer in the same manner as would be done with sets of actual test data. The theoretical data agreed quite well with the experimental data obtained and also provided information on flat-rosette performance under loading conditions difficult to simulate in the laboratory.