3D separation of stresses - 40 years on

Over 40 years ago M.M. Frocht analysed the stresses on three-dimensional specimens using photoelastic analysis. Materials and techniques improved through the years, until the late 1960's when finite elements analysis became popular. In the last fifteen years the memory capabilities and speed of computers have increased by several orders of magnitude. This has allowed three-dimensional photoelasticity to re-emerge as a powerful experimental technique. In this paper an automated polariscope is described and is to give oxx, oyy and ozz along arbitrary lines of specimen previously analysed by Frocht in the 1950's. The lines of analysis from the automated polariscope were compared to Frocht's original data and a finite element analysis of the specimen performed as a further comparison. The full field automated polariscope is based on the method of phase stepping and is briefly described. The polariscope is used to calculate the separate Cartesian stresses using Frocht's shear difference method. Regions of background in the image and hence the boundary of the model is first determined and the validity of each boundary point tested using Megnager's theorem. The use of the shear difference method in two orthogonal directions results in full-field images of stress. Subslicing of the model and combination of the data from the two orthogonal planes gives oxx, oyy and ozz. The processing of the photoelastic data and the management of the three-dimensional co-ordinate information is performed on a 386 personal computer.