This paper presents a statistical analysis on a fault index computed based on the stator current space vector instantaneous frequency on a real-time platform for online noninvasive gear tooth surface damage fault detection in a single-stage gearbox. This analysis is crucial for defining a threshold beyond which the condition of gear teeth can be considered defective with a significance level less than the critical value. Besides, the efficiency of the fault detection algorithm realized on a real-time platform will be evaluated for both pinion and wheel tooth surface damage faults by using a setup based on a 250-W three-phase squirrel-cage induction machine shaft-connected to a single-stage spur or helical gearboxes.