Abstract: Interturn short circuit represents one of the most prevalent faults in brushless DC motors (BLDCMs), making online detection during short-circuit conditions critically important. Current research reports on BLDCM turn-to-turn short-circuit detection remain scarce. Given the significant differences in mechanical structure and drive modes between BLDCMs and other motor types, existing detection methods cannot be directly applied. To address this gap, this paper presents an online turn-to-turn short-circuit detection method for BLDCMs based on zero-sequence voltage characteristics, incorporating a novel characteristic parameter index. The research methodology follows three key steps: 1) establishing an analytical model of windings under turn-to-turn short-circuit conditions and theoretically deriving the evolution characteristics of zero-sequence voltage's fundamental frequency component during faults; 2) proposing innovative fault phase location and severity evaluation indices by combining three-phase current and zero-sequence voltage fundamental frequency components; 3) validating the method's effectiveness through combined Simulink simulation and experimental platform studies. Results demonstrate the method's capability to effectively detect faults, locate affected phases, and assess fault severity. Furthermore, the method exhibits strong robustness against motor transient states, enabling real-time detection of BLDCM interturn short circuits.