Acoustic emission source discrimination using a piezopolymer based sensor

Extracting information about the nature and extent of machine and structural damage from the resultant acoustic emission signals is of considerable interest in developing machine condition monitoring systems. The extent to which acoustic emission source characterization can be practically applied to track machine operation and identify developing faults is examined by looking at both the time and frequency content of signals captured by a broadband acoustic emission sensor. The acoustic emission sensor is a piezopolymer based design developed as a low cost rugged alternative to more common high fidelity sensors such as capacitance types or National Board of Standards (NBS) conical design. The response characteristics of the piezopolymer sensor are compared with a commercial piezoceramic transducer and a laser interferometer using pencil lead fracture and helium jet as sources of acoustic emission. The practical value of the polymer sensor is then evaluated in field tests examining fluid and mechanical generated acoustic emission on a single stage centrifugal pump and a turbocharged diesel engine. The results suggest that the broad frequency response of the polymer based sensor can improve the identification of acoustic emission sources, thus avoiding the need for deconvolution of the raw acoustic emission signal. It is concluded that the piezopolymer sensor design offers a useful alternative to many of the acoustic emission transducers currently on the market for industrial use, with the additional benefit of broad-band response. The ability to use the new sensor to reveal detail in machine operating signatures and recognizable fault patterns in the frequency domain offers the potential to incorporate such features into a condition monitoring system with diagnostic capabilities.