PurposeThis work aims to introduce a rotating machine's defects localisation based on a time reversal method. We are interested in the study of ball bearing defects localisation by the time reversal method. Bearings are usually the most subjected components to wear in industrial rotating machines. Vibration analysis method is commonly used to investigate and extract knowledge about this type of faults as soon as they appear. This time, as a new solution, we introduced the time reversal method to this field.MethodsGenerally, the time reversal method is performed in two steps; a propagation media characterization (also known as recording step) then a localisation or focusing (back propagation step). Bearing, gears and balancing defects are some of the main defects that might occur in rotating machines. Those defects act like exterior actions which excites the machine structure. The effect of these actions follows a precise path to propagate. Thus, the first objective in this study is to characterise experimentally that path between a random position on the structure and some areas that may contain a defect. Then, apply the time reversal process to locate defected components. It has been shown in the literature that complexity of the propagation media or structure improves the efficiency of the time reversal method. Wave reflexion/conversion due to the media characteristic could add information about the source location, which is equivalent to virtual sensors. In this study case, it is due to vibrations transfer between components by contact. We propose performing a characterization step when the machine is at rest because the effect of defects occurs only if the machine is rotating. As a first investigation in this field, the interest is made on bearing defects.ResultsThe number of sensors used in the analysis can be reduced, using the time reversal method to locate defects, to a single accelerometer. The time reversal process sometimes needed combining the measurements of multiple sensors to extract the defect location. The time reversal technique makes it possible to separate the sensors from the active components. Rather than using accelerometers in the bearings, as conventionally done, the method was able to locate the fault in the structure.ConclusionThe time reversal method has a potential to be combined with other techniques to locate accurately any vibration sources in rotating machines, particularly those produced by periodic shocks.
