A new approach to crack fault diagnosis of a gear based on the Teager-Huang transform is presented.This method integrates the empirical mode decomposition (EMD)with the Teager Kaiser energy operator(TKEO)technique.The EMD method is self-adaptive to non-stationary and non-linear signal analyses.The method developed in this paper decomposes original times series data into a series of zero mean amplitude modulation-frequency modulation(AM-FM)intrinsic mode functions(IMFs)by using EMD.TKEO can track the instantaneous amplitude and frequency of the AM-FM component at any moment.Experimental examples were conducted to evaluate the effectiveness of the proposed approach.The results show that the performance of the Teager-Huang transform is better tan that of the Hilbert-Huang transform for gear fault monitoring and can effectively diagnose the gear crack faults.