SELECTED METHODS OF EVALUATING RESIDUAL STRESS GRADIENTS MEASURED BY X-RAY DIFFRACTION. TRADITIONAL, FULL TENSOR, AND WAVELET, L. Suominen and D. Carr, pp. 21-30

Residual stresses measured by x-rays in the near surface region of polycrystalline materials are always average over a varying diffracting volume. This volume variation is due to beam tilting during a measurement and the exponential decay of x-rays during penetration. Different methods have been devised to correct this biased data. Perhaps the most used method is described in the publication Residual Stress Measurement by X-Ray Diffraction SAE J784a /9/. This method is based on successive cycles of stress measurement and layer removal. Since the method is based upon extrapolation it fails when the gradient curves. A modified correction method that instead of an extrapolation does a correction based upon piecewise weighted averaging from one observed point to another is proposed in this paper. This new method gives more reliable results in the case presented. Layer removal results are compared to two nondestructive surface methods, which are the new wavelet method /6,8/ and the more traditional full tensor method /5/. Results from the two non-destructive surface methods indicate rough agreement with the layer removal method method but fade in retrieving details. INTRODUCTION The fatigue behavior of many mechanical components can be greatly improved by inducing surface compressive residual stresses. This is especially true in the transportation industry. Considerable cost savings not only in length of service but also in the reduced material requirements for a component can be realized. The predominant method used to measure near surface stresses employs monitoring the crystalline lattice spacing by x-ray diffraction. Like many testing methods, frustration and good results lies in the details. X-rays penetrate only a few μm into a metallic sample, on the order of 4 to 20 μm thick. If the stress field varies over this depth then the results can be greatly skewed. The problem is how to retrieve the true stress field. In this paper an investigation will be taken of a traditional method for correcting /9/ these skewed results and a proposed modification given. The traditional method gains access to subsurface measurements by first removing material. The material removal method must not induce new residual stresses, hence chemical or electrochemical means are usually employed. 1 Stresstech Oy, 16 Ohjelmakaari, 40500 Jyvaskyla, Finland 2 American Stress Technologies, Inc.,267 Kappa Drive, Pittsburgh, PA 15238, USA Copyright(c)JCPDS-International Centre for Diffraction Data 2000,Advances in X-ray Analysis,Vol.43 21