Spatially resolved acoustic spectroscopy (SRAS) technology development

In this thesis the progress made in developing the spatially resolved acoustic spectroscopy (SRAS) technology for industrial deployment is presented. SRAS is a noncontact laser ultrasound materials characterisation instrument which uses surface acoustic waves (SAWs) to image and determine surface microstructure on metallic samples. SRAS offers advantages over existing techniques for determining microstructure, primarily speed, large sample sizes and full sample coverage. The technique is developed in four key areas: instrumentation, signal analysis, machine vision for auto-focus and modelling generation of laser-ultrasound. A novel optomechanical system is presented which allows a compact SRAS instrument to be designed and implemented, and which is suitable for industrial manufacture, assembly and application. The capability of this new instrument is thoroughly assessed, with determinations of the noise floor, spatial resolution, and limitations on SAW acoustic wavelength presented. A review of sophisticated signal analysis techniques, suitable for application to SRAS signals, is undertaken and the performance of each technique quantified by both Monte-Carlo simulation and fitting of experimental data. To enable scanning of non-planar sample surfaces and the use of low cost motion platforms, a means by which the optimum distance between the optical SRAS instrument and a sample surface can be maintained is described. Based on characterising changes in the imaged grating used to generate SAWs, a method is presented capable of providing absolute position corrections from single images. In addition a range of focal measures are reviewed for application to SRAS for automation of scan configuration. It has been shown previously that SRAS is capable of determining crystallographic orientation on cubic and hexagonal materials, however some discrepancies in this method exist due to insensitivity to the generation mechanism. In this thesis a numerical model for the laser generation of SAWs on arbitrarily anisotropic materials is presented, and used to generate a family of SAW velocity surfaces for cubic nickel. The existing model is combined with the generation model to create a hybrid for use in SRAS orientation determination