Detection and evaluation of rail surface defects using alternating current field measurement techniques

Reliable and cost-effective inspection of rail tracks is of paramount importance to ensure the safety of rolling stock operations. In this paper alternating current field measurement (ACFM) sensors are used to carry out experiments on artificially induced rail surface defects at various speeds using testing configurations that simulate actual inspection conditions found in the field. From the obtained results it can be clearly seen that the ACFM sensors can detect the artificially induced rail surface defects even when relatively significant lift-off is involved, i.e. ∼5 mm and that the effect of increasing speed on the amplitude of the Bx signal, which is directly related to the depth of the crack, is negligible. However, clustered defects cannot be easily resolved and the overall amplitude is related to the spacing of the defects within a cluster. The order of clustered defects also significantly influences the maximum amplitude of the recorded Bx signal. The validity of the results obtained from the tests on artificially induced defects was verified by conducting further ACFM tests on a rail sample removed from service that contained mild rolling contact fatigue cracks.