Detection of rolling-element bearing signal corrupted by noise of similar frequency using adaptive noise cancellation

The detection of a defective bearing signal corrupted by background machine noise and strong shaft unbalance is illustrated by using adaptive noise cancellation technique. A trial and error method is adopted to determine the optimum convergency factor for the adaptation process. Successful detection of the desired bearing signal depends on the choice of convergency factor and the location of the reference sensor for the measurement of correlated noise. Although the results show a strong potential for its application in detecting spall type bearing signal, a more definitive approach needs to be developed for determining the convergency factor for the adaptation process.