Background/Introduction <p>Gears are commonly used machine elements in rotating machinery which play a key role in power/motion transmission applications. The gear element mounted on the power transmission system is prone to fail due to various reasons such as manufacturing defects, improper application, design and maintenance related issues. Vibration signal monitoring and analysis techniques are widely employed to identify and diagnose the existence of faults developed in a geared system under different operating conditions. The extraction of diagnostic features from the vibration signals is quite challenging due to the noise present in the raw signal.</p> Purpose <p>This paper presents the results of numerical simulation and experimental analyses carried out to obtain are liable fault diagnostic information from the two-stage helical gearbox subjected to healthy and faulty operating conditions.</p> Methods <p>The vibration signals acquired from the gearbox were processed using Improved Complementary Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) in conjunction with Time and Envelope Spectrum Kurtosis (TESK) technique.</p> Results <p>The results obtained enable efficient tracking of fault propagation, accurate fault localization, and detection of distinctive spectral patterns.</p> Conclusions <p>The results confirm the reliability and robustness of the proposed diagnostic method.</p>