A complete reassessment of standard residual stress uncertainty analyses using neutron diffraction emphasizing the influence of grain size

The determination of residual stress in engineering materials with large grains is a challenge when it comes to using diffraction techniques. Not only are the accuracies of the residual stresses themselves important but also the accurate evaluation of their uncertainties. An austenitic steel three-pass slot weld (NeT- TG4) with varying grain size high-lights the potential problems with the data evaluation when grain size is not taken into account whilst measuring strain. Neutron diffraction results are compared with each other (with combinations of slit and radial oscillating collimator beam defining optics) and with high energy synchrotron radiation results with a spiral slit gauge volume defining system. The impact of the grain size on the determination of residual stresses and their associated uncertainties when using diffraction techniques in engineering components is emphasized and discussed. A simple model to estimate the extra random uncertainty contribution due to the so-called grain size statistics is applied and verified. The benefit of continuous or stepwise oscillation to increase the number of detected grains on the detector is discussed together with how to optimize the time of a measurement. From the data obtained, best practice guidelines will be suggested on dealing with large grains when determining strain and stress with neutron diffraction.