Failure modes and life prediction model for high-speed bearings in a through-flow universal motor

The focus of this study was the empirical modelling of the high-speed bearings life and failure modes of a through-flow universal motor. An approach was used facilitating predictions of the life through-range of various conditions. The model estimates bearing life for the survival probability of 50% - L50. It influences parameters such as bearing temperature, speed factor, equivalent load, grease fill amount, type of oil, type of bearing cage, type of seals, tolerance class, and side of the motor, all of which are considered in the model. Initial empirical data consisted of 4672 test populations, involving 38,021 vacuum cleaner motors. Strict filtering requirements of all the available test data resulted in 170 final populations, consisting of 1385 tested and 638 failed bearings, which were used for building a Weibull database and for developing the models. The paper's key contributions are the empirical models gained with multiple linear regression and the obtained database of tested bearings.