A vibration dataset was developed to investigate the behaviour of deep groove ball bearings and cylindrical roller bearings across different health states. The dataset provides a benchmark for developing and evaluating bearing fault diagnosis methods under controlled laboratory conditions. Experiments were conducted on the dedicated Drive Test Bench 58 (Antriebsprüfstand 58). Baseline vibration measurements were first acquired from healthy bearings, followed by measurements from bearings with artificially introduced faults. These faults created reproducible damage scenarios, enabling systematic analysis of vibration signatures across different bearing health states. This work was funded by the research projects SEED (Smart Efficient Electric Driveline, FKZ 13FH585KX0) and KI-Werkstatt (FEIH_2486678).
