Signal decomposition method based on spectral criteria or more recently data-driven, are the basis of bearing and gear fault diagnosis. In this work, a new signal decomposition method based on a Spectral combination of Hjorth’s parameter is proposed. The effectiveness and sensitivity of such a signal decomposition to bearing fault diagnostics has been investigated based on both synthetic and real bearing vibration signals. The results show good performances in extracting the frequency band which carries the most useful diagnosis information, and sensibility to the different fault severities