Induction motors (IMs) are widely used in agricultural, industrial, and commercial applications. Bearing faults are one of the most common causes of breakdown in IMs. Motor current signature analysis (MCSA) is becoming a popular tool for bearing fault detection because of its cost effectiveness and noninvasiveness. However, when a motor operates under low‐load and lower load conditions, MCSA cannot be used to detect the bearing outer raceway fault because of the influence of the supply frequency component spectrum leakage, eccentricity harmonics, and other noises. In this paper, a new method for the fault detection of bearing outer raceway fault in IMs is proposed. The method is based on the analysis of the instantaneous frequency (IF) of the IM stator current using Hilbert transform, and Fast Fourier Transform (FFT) spectrum analysis is used to detect the frequency associated with the bearing outer raceway fault. The proposed method can significantly reduce the negative influence of the supply frequency component spectrum leakage, and thus can enhance the fault features to detect bearing outer raceway fault under low‐load and lower load conditions, compared with the traditional MCSA method. Finally, the experimental results on an IM with bearing outer raceway fault prove the effectiveness of the proposed method. ? 2018 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.