Martin Dazer

In this study the vibration behavior of synchronous belts with representative damages is investigated using a test bench and a design of experiments approach. A laser triangulation sensor is used to record the vibrations of undamaged belts and belts with artificially introduced tooth crack, tooth separation and cord damage. The vibration signals are evaluated in the frequency domain. A physics-based target value derived from the harmonics of the belt run frequency is proposed for damage detection. The subsequent analysis proves an effect of the damages on this target value, with sensitivity to the number of considered frequency components. In addition, the operating conditions affect the target value, with torque showing the greatest effect. In particular, the investigated damages are better detectable for low torque levels. These findings provide a basis for vibration-based condition monitoring of synchronous belts and highlight the importance of considering operating conditions when interpreting damage-related features.

Field monitoring yields heterogeneous operating-load data, including event-based histories and broadband random excitation. Durability validation requires reducing these data to traceable load targets and laboratory test profiles without losing fatigue relevance. We propose the Damage-Based Load Transfer Framework (DBLTF), which preserves domain-specific processing paths but links them through one common damage-based backbone: conditioning, load representation, lifetime modeling, Miner-type accumulation, and statistically interpretable target derivation. Two automotive case studies demonstrate the framework. In the first, fleet rear-axle torque histories are converted into damage per reference distance and into a reference-user spectrum at a prescribed coverage level. In the second, battery-housing vibration measurements are mapped to PSDs and fatigue damage spectra (FDS), consolidated into a conservative envelope, and transformed into a damage-equivalent shaker PSD. The resulting targets provide a transparent, statistically interpretable link between field usage, laboratory qualification, and durability assessment.