Electro-hydraulic actuated systems are widely used in industrial applications due to high torque density, higher speeds and wide bandwidth operation. However, the complexities and the parametric uncertainties of the hydraulic actuated systems pose challenges in establishing analytical mathematical models. Unlike electro-mechanical and pneumatic systems, the nonlinear dynamics due to dead band, hysteresis, nonlinear pressure flow relations, leakages and friction affects the pressure sensitivity and flow gain by altering the system's transient response, which can introduce asymmetric oscillatory behavior and a lag in the system response. The parametric uncertainties make it imperative to have condition monitoring with in-built diagnostics capability. Timely faults detection and isolation can help mitigate catastrophic failures. This paper presents a signal-based fault diagnostic scheme for a gearbox hydraulic actuator leakage detection using the wavelet transform. The novelty of the work is the development of a high fidelity leakage fault detection as low as 0.128 lit/min. The low leakage detection and independence on system model makes this technique quite attractive for industrial applications.