Abstract:
Techniques such as vibration monitoring, thermal analysis and oil analysis are well
established as means to have been used to improve reliability of gearboxes and extend
time-to-failure. In this area Acoustic Emission (AE) technology is still in its infancy but
the attention shown by researchers towards this method has increased dramatically
because several studies have shown the AE offers the important advantage of improved
sensitivity over more conventional monitoring tools for the early detection and
prediction of gear failure.
Helical gear lubrication is critically important for maintaining the integrity of operating
gears and the oil also prevents asperity contact at the gear mesh thereby protecting the
gears from a deterioration process and surface failures. In gear systems, there are three
types of lubrication regimes: Dry Running, Boundary Lubrication (BL), Hydrodynamic
Lubrication (HL) and Elastohydro-dynamic Lubrication (EHL). The last regime is
associated with the normal operating running condition of gears.
Acoustic emissions were acquired from gears and analysed for different lubrication
regimes (dry, BL, HL and EHL regimes at different temperatures), and corresponding
specific film thicknesses (λ) levels. The results showed an inverse relationship between
AE signal levels and specific film thickness (λ) of the oil. This relation was used to
determine the lubrication regime from the measured AE signals. For instance, dry
running had the highest AE levels which were attributed to the metal-to-metal contact of
the gear mesh. The BL regime had relatively high AE levels which also attributed to the
level of asperity contact is greater than the oil film thickness. The HL regime was
characterized by the lowest AE levels due to the lubricant oil completely separating the
teeth during gear meshing. Finally, the EHL regime showed intermediate AE levels
compared to the BL and HL regimes because the oil film was less than for the HL
regime but greater than for the BL regime.
It is shown that the application of advanced signal processing methods is necessary for
monitoring helical gears; Kurtosis and Spectral Kurtosis were used to investigate the
AE signatures and found to be effective in de-noising (spectral kurtosis) acquired
signals. Acoustic Emission proved to be a powerful tool to detect the oil regime for both
defective and non-defective conditions.
It is concluded that the experimental findings of this research programme will provide
the foundations for significant advancement in the application of AE for the
determining the lubrication regime present within a helical gearbox and for the detection
of developing gear faults. This should give a new impetus in the field of maintenance
and prevention of human and material catastrophes.
Several papers presenting the findings of this research have been published in
international journals and given at conferences.