A noninvasive scheme for real-time and online detection of interturn winding faults in single-phase transformers has been proposed. A terminal measurement-based transformer modeling technique is central to the fault detection scheme. First, with the experimentally acquired primary line voltage and line current data of the healthy transformer, a healthy no-load model of the transformer is generated. Next, a healthy estimated indicator value, computed from this model under the given input voltage condition, is compared with the actual indicator value for detecting the presence of an interturn winding fault. Experiments have shown that the proposed scheme is not only capable of detecting incipient levels of interturn winding faults on primary and secondary windings under any load conditions, but also can estimate the fault severity level. In the end, a procedure to determine a suitable tripping threshold to de-energize the transformer under incipient fault has also been outlined.