Rolling contact bearings are widely used in various classes of machines and have a lifespan related to their specific use. The occurrence of small defects within the bearings can lead to failure of the bearings over time and may lead to a major breakdown requiring a significant maintenance period. Major causes of damage to a bearing are excessive load, false brinelling, true brinelling, overheating, failure due to fatigue, contamination, reverse loading, misalignment, loose or tight fits and corrosion, etc. A vibrational analysis technique is used in order to determine the various faults and the extent of any damage sustained. Vibrational analysis includes the use of a Fast Fourier Transform (FFT) algorithm to convert time domain data to frequency domain data along with filtering techniques using the help of MATLAB software. In the current project, vibrational data was collected from self-aligning rolling contact bearings under six different fault conditions, namely a bearing with an inner race fault, a bearing with a cage fault, a bearing with one ball removed, a bearing with two balls removed, a bearing with three balls removed and a bearing with an outer race fault, i.e., a fault on the inner surface of the outer race of the bearing. In addition, healthy bearing conditions and a rotation at speed of 1100 rpm were applied. Methods like FFT and filtering techniques such as a Type 1 Chebyshev filter are proposed in this research for the analysis of faults. Furthermore, the classification of faults includes the use of an artificial neural network (ANN). © Springer Nature Singapore Pte Ltd. 2019.