© 2017 The Authors. Gears are significant component in a multiplicity of industrial applications such as machine tool and gearboxes. An unforeseen failure of gear may result in significant economic losses. Therefore this research propose fault detection improvement throught series of vibration signal processing techuiques. These techniques have been tested experimentally using vibration data collected from the transmission system of a CS-29 'Category A' helicopter gearbox under different bearing damage severity of the second planetary stage. Results showed successful improvement of bearing fault detection.