In comparison with a traditional planetary gearbox, the split torque gearbox (STG) potentially offers lower weight, increased reliability, and improved efficiency. These benefits have driven helicopter OEMs to develop products using STG. However, the effect of multiple gears meshing simultaneously with the central gear and a large number of synchronous components (gears or bearing) in close proximity creates a problem on how to detect the gear fault location in a STG. As of today, only limited research on STG fault detection using vibration sensors and acoustic emission sensors has been conducted.In this paper, an effective methodology on gear fault location detection using AE sensors for STG is presented. The methodology uses wavelet transform to process the AE sensor signals to determine the arrival time of the AE bursts at different locations. By analyzing the arrival times of the AE bursts, the gear fault location can be determined. The parameters of the wavelet transform are optimized by using an ant colony optimization algorithm. Real seeded gear fault experimental tests on a notational STG are conducted. AE sensor signals at the locations of healthy and damaged output driving gears are collected simultaneously to determine the location of the damaged gear. Experimental results have shown the effectiveness of the presented methodology.