Identification of the Onset of Cracking in Gear Teeth Using Acoustic Emission

The development of diagnostic methods for gear tooth faults in aerospace power transmission systems is an active research area being driven largely by the interests of military organisations or large aerospace organisations. In aerospace applications, the potential results of gear failure are serious, ranging from increased asset downtime to, at worst, catastrophic failure with life-threatening consequences. New monitoring techniques which can identify the onset of failure at earlier stages are in demand. Acoustic Emission (AE) is the most sensitive condition monitoring tool and is a passive technique that detects the stress wave emitted by a structure as cracks propagate. In this study a gear test rig that allows the fatigue loading of an individual gear tooth was utilised. The rig allows a full AE analysis of damage signatures in gear teeth without the presence of constant background noise due to rotational and frictional sources. Furthermore this approach allows validation of AE results using crack gauges or strain gauges. Utilising a new approach to AE monitoring a sensor was mounted on the gear and used to continuously capture AE data for a complete fatigue load cycle of data, rather than the traditional approach where discrete signals are captured on a threshold basis. Data was captured every 10th load cycle for the duration of the test. A developed fast fourier transform analysis technique was compared with traditional analytical methods. In this investigation the developed techniques were validated against visual inspection and were shown to be far superior to the traditional approach.