Detection of Failure and Measurement of Number of Cycles to Failure of Nickel–Titanium Rotary files in Bending Fatigue Testing

This study aimed to (1) develop a novel method for fabricating simulated root canals, (2) introduce a reliable approach for detecting file fractures, and (3) establish a direct measurement technique for determining the number of cycles to failure (NCF) in bending fatigue tests of rotary files. Bending fatigue tests were conducted using a custom setup consisting of a simulated root canal, a handpiece–controller unit, and a signal extraction module for monitoring motor current and encoder signals. The simulated canals were fabricated to match the geometry of 25.06 ZenFlex rotary files, and included five configurations with varying radii (2, 5, and 8 mm) and angle of curvature (30°, 45°, and 60°). Tests were performed at room and body temperatures. File fracture was identified by an abrupt drop in motor current, and the NCF was directly determined from encoder-derived revolution counts. To verify accuracy, encoder-based revolution counts were compared with those measured with a mechanical counter. In addition, the proposed method was benchmarked against the conventional time-based calculation of NCF. The fabricated canals provided precise confinement of the files, ensuring reproducible curvature with negligible torsional distortion. Fractographic analysis revealed fatigue features, including striations radiating from the crack origin and dimples in the fast-fracture zone, confirming bending fatigue as the dominant failure mechanism. Encoder-based revolution counts closely matched those from the mechanical counter, with discrepancies of less than one revolution, and yielded slightly lower values than the conventional time-based method. Statistical analysis showed no significant difference between the proposed and conventional methods (p > 0.05). Fatigue life was significantly reduced at body temperature compared with room temperature (p < 0.05) and increased with larger radii of curvature (p < 0.05), while angle of curvature had no significant effect (p > 0.05). The proposed simulated root canals improved reproducibility and standardization of in vitro fatigue testing of files. The signal extraction method provided accurate revolution counts and a more realistic estimation of NCF than conventional time-to-failure methods, which overestimated actual values by 1.85 to 6.28% in this study.