A novel use of signal processing tools for fault detection in IC engines

Condition Monitoring of Internal combustion engines is becoming more and more crucial as the engine designers try to achieve better fuel efficiency and/or more power while adhering to the stringent pollution control norms practiced today. Most of the condition monitoring practices follows a periodic maintenance model or in case of cheaper engines, none at all. Keeping in mind the importance of continuous monitoring of critical parts of engine like piston rings and cylinder liner there are a number of real time condition monitoring methods in practice. Our work deals with an unconventional use of signal processing tools on Acoustic Emission (AE) from engine cylinder liner to identify or enhance the features associated with various faults commonly found in critical engine components. Experimental data is obtained from a small petrol engine in which various fault conditions are recreated. Signals collected from the engine running with bad piston, worn out piston rings and under lubricant oil level were analyzed.