Rotating machines are commonly used in industrial applications. Mechanical faults such asrotor unbalance, shaft misalignment, pulley misalignment, structural looseness, and bearingfaults leading to unplanned shutdown based on the severity of these faults. The conditionmonitoring technique based on vibration analysis has the potential to detect and diagnose agreat number of early stage faults. However, some mechanical faults have correlatedvibration features leading to ambiguous diagnosis to identify and distinguish these faults. Inthis paper, a proposed method based on the Principal Component Analysis (PCA) ispresented to produce uncorrelated Principal Components (PCs) to identify the healthy anddifferent faulty cases. A test rig was prepared to simulate a group of mechanical faults suchas rotor unbalance, pulley misalignment, belt damage, combined unbalance with pulleymisalignment, and combined unbalance with belt damage. The conventional vibrationmeasurements were collected for each case and their features were extracted and used toproduce the equivalent PCs. It was found that the produced uncorrelated PCs have thesuperior to distinguish the majority of simulated faults which have correlated vibrationfeatures as presented in the rest of paper.