X-ray diffraction accuracy in the measurement of stresses in post-tensioned strands

The evaluation of the stresses acting in prestressing steels is a key issue in the structural assessment of existing post-tensioned bridges. Several non-destructive techniques have been proposed for this purpose, but many are difficult to apply in situ or provide only qualitative and hard-to-interpret results. Among them, X-ray diffraction (XRD) is a promising method, as it yields quantitative information on the stress state in steel. However, its accuracy and reliability for prestressing strands are still not fully established. This study investigates the performance of the XRD technique in measuring stresses in PT strands by combining laboratory tests and advanced three-dimensional finite element (FE) modelling. Experimental tests were performed on three 7-wire strands with nominal diameters of 9.3, 15.2 and 15.7 mm under service-level tensile loads. XRD measurements were then compared with SG-derived stresses and FE predictions to quantify the systematic underestimation and to identify the main sources of deviation, namely: strand diameter, uncertainty in the measurement location, residual stresses magnitude, and the difference between surface and average cross-sectional stress. A systematic underestimation was observed, with mean absolute percentage errors (MAPE) of 28%, 17%, and 7% for the 15.7 mm, 15.2 mm, and 9.3 mm strands, respectively. Based on these findings, correction formulations are proposed. After correction, the deviation (MAPE) with respect to nominal stress is generally reduced to 5%, 5% and 2% for the 15.7 mm, 15.2 mm, and 9.3 mm strands, respectively. XRD can therefore be reliably employed for stress assessment in PT strands, provided that diameter-dependent calibration, residual stress correction and uncertainties related to the positioning of the instrument and to the measure itself are properly accounted for.