Comparative experimental study of different bearings vibration using FFT and statistical feature extraction

In this study, vibration patterns of different types of bearings, such as deep groove ball bearings, cylindrical roller bearings (single row), and radial spherical plain bearings were studied. They're the most commonly used in mechanical parts in rotary machines under the effect of load and different angular speeds: 1488 rpm, 1900 rpm, and 2450 rpm. As well as the two cases, the first without load and the second with a load of 29.43 N, there was a three-disc mounted on the shaft. The vibration monitoring and analysis technique, which is a dependable and accurate assessment method to select a suitable bearing element, the Fast Fourier Transformation analysis was used for processing the vibration signal that was gathered from three different types of bearings for three different running speeds with two load levels. Additionally, several indicative parameters including the root mean square, kurtosis, peak-to-peak, and skewness were extracted from the vibration data using time-domain signal analysis. From experimental results, it was noticed that the deep groove ball bearing gives the lowest amplitude of vibration among all two bearings when the system is without loading effect, whereas the cylindrical roller bearing gives the least amplitude under loading operating conditions; finally, the radial spherical plain bearing was operated at its best at high rotational speeds.