Reconstruction of axisymmetric strain distributions via neutron strain tomography

Abstract

Predicting the behaviour of structural components under a particular set of loading conditions requires knowledge of the residual elastic strain distribution throughout the bulk of these components. Characterising the 3D strain state at any particular point involves the measurement of six independent components which make up the second order strain tensor. Mapping the complete strain distribution throughout large volumes thus presents significant practical challenges. One possible solution to this problem is to reconstruct the 3D variation of strain components using tomographic techniques. The basic principle underpinning this idea is that the multi-component strain tensor can be reconstructed from a redundant set of lower order projection data. Here we demonstrate this fundamental concept for two samples: a shrink fit ‘ring-and-plug’ sample, and a spray-quenched circular cylinder, both possessing axially symmetric internal strain distribution. We present and contrast different approaches to the strain tomography problem. The methods described here can also be readily applied to high-energy X-ray diffraction measurements and represent an important step toward developing the tomographic reconstruction framework for strain tensor distributions of arbitrary complexity. The major benefit of neutron strain tomography is that the incident beam flux is utilised more fully, greatly reducing the data collection times. Using micro-channel plate (MCP) neutron detectors, a spatial resolution of the order of 0.1 mm can be achieved [1].