Investigation of FBG sensor length effects on fault detection and damage location in rotating systems

Bearings are essential load-bearing elements that enable smooth and efficient motion between interacting components in mechatronic systems. The health of these components have a direct impact on the overall performance, safety, and maintenance requirements of rotating machinery. When faults occur in bearings, they often produce distinctive signs such as an increase in vibration, temperature fluctuations, or acoustic noise. Traditional diagnostic techniques therefore make use of accelerometers, thermocouples or acoustic emission sensors. Despite their widespread use, these methods have drawbacks, such as indirect sensing due to bulky transducers and vulnerability to electromagnetic interference. Fiber Bragg grating (FBG) sensors, on the other hand, are compact, immune to EMI, and highly sensitive to strain, making them promising tools for detecting bearing defects. Although such sensors have been previously proposed for bearing monitoring, the ideal grating length for precise fault detection and localization within bearing raceways, particularly when using well-established diagnostic indicators like the ball pass frequencies (BPFO/BPFI) has not been clearly established. In this study, strain data was collected from both healthy and artificially damaged bearings operated at a constant speed, using FBGs of various grating lengths positioned near the fault region in the primary load zone. The findings indicate that while all grating configurations can identify bearing anomalies, the shortest gratings offer the highest accuracy in detecting and pinpointing outer race faults. Overall, the use of FBGs for locating faults provides significant benefits compared to conventional monitoring approaches, contributing valuable insights for prototyping and developing more durable bearing systems.