Uncovering Bias and Underestimated Errors in Neutron Diffraction Residual Strain Measurements

This work addresses the inadequacy of one‐standard‐deviation measurement uncertainties in neutron and x‐ray diffraction for capturing the complexity of residual strain fields in additively manufactured components. Measured strains, collected at multiple orientations to describe an underlying strain tensor, may yield solutions that systematically deviate from ground truth when few measurements or an overly simplified model is used; in such cases, uncertainties are often underestimated, and the stability (invariance) and systematic deviation (bias) cannot be assessed without additional data. Here, bias—typically unknowable—is evaluated by enforcing that each measured strain satisfies the strain transformation, and the analysis is extended to a ring‐and‐core reference sample with a known stress state, demonstrating that underestimated uncertainties are a general phenomenon in the neutron diffraction/residual stress community. Recommendations regarding the critical number of measurements needed to evaluate systematic effects will be made based on analyses of both the additive and reference samples.