Engineering stress analysis using PhotoStress and computational techniques

Abstract

The PhotoStress technique is a non-contact full-field experimental method which uses light rays and optical techniques to determine stresses on any structural component under different loading and boundary conditions through surface strain measurements. The presented dissertation concentrates mainly on building up knowledge on the use of the PhotoStress method for experimental stress analysis. The research aims at identifying areas in which the PhotoStress method augments the simulation process. The overall goal of this study is to examine the effectiveness of the PhotoStress technique and to serve as a guide to determine qualitatively and quantitatively strain/stress measurements using the PhotoStress technique. The performance, reliability and accuracy of the PhotoStress coating method was evaluated by applying this technique to benchmark problems whose solutions for stress distribution is known. The benchmark problems include end-loaded cantilever beams, circular discs and annular rings loaded in diametral compression and a flat plate with hole under uniaxial tensile loading. An investigation of the stress/strain state of these two-dimensional benchmark problems was carried out to verify and validate the experimental PhotoStress coating technique by analysing and comparing the PhotoStress' difference in principal strains and stresses with the results obtained from experimental strain gauges, FEM and analytical solutions. Three-dimensional case studies were also investigated using the PhotoStress technique. All components used in the experimentation techniques were machined from stainless steel 304L, whose material specifications were determined from a tensile test. PhotoStress results were comparable to the analytical and FE solutions as well as to the strain gauges' within an average accuracy of 7% absolute difference. Statistical studies have shown that there is no significant difference between all the techniques used to obtain the difference in principal strains/stresses. Hence the PhotoStress technique has been verified and validated. Two different photoelastic coatings, PS-1 and PL-8, were analysed and compared with each other. It was concluded that there exists no significant difference between them. PhotoStress measurements were taken using two different techniques: the compensation technique and the fringe order chart. It was concluded that there exists a significant difference between the two different PhotoStress techniques. The importance of using an experimental approach to validate the FE model or to identify which analytical solution best fits the experimental results is highlighted through the use of the PhotoStress-FEA hybrid method. The slitting technique, separator strain gauges and the PhotoStress-FEA hybrid method were used to obtain the individual principal strains and stresses. The PhotoStress-FEA hybrid method was applied and individual principal strains/stresses were extracted from the FE model. The separator strain gauges give the values of separate principal stresses with a percentage error ofless than 7% with respect to the reference values. The slitting technique is less reliable, having higher percentage error of 19% with respect to the reference values. The PhotoStress coating technique is not suitable on medium-to-high modulus structures under low load conditions.