Sonic analysis is very similar to vibration analysis. A gas turbine engine produces an acoustic spectrum which reflects the rotating speed and geometric relationships of the various rotating components. Additional acoustic energy is released every time a disparity or fault comes in contact with another surface. The frequency and characteristic of these signals can be used to isolate faulty components. The sonic analysis concept is based on employing narrow band filtering to separate the discrete signals from the noise components to determine signal characteristics relative to amplitude, harmonic, and modulation content. When a part or component begins to wear or go through some other physical change, the character of the acoustic signal is altered. By monitoring the characteristics of all the discrete frequencies produced by the rotating components, it is possible to detect changes in the mechanical condition and to pinpoint the individual component which is deteriorating.

This content is only available via PDF.