Abstract
The work described in this paper deals with a study to develop a technique for identification of mechanical abnormality in piston engines such as excessives crank train's bearing clearance, through vibration signals. Excessive bearing clearance in this case will cause impact forces to occur which excite vibrations on the engine structure. However, since in an operating engine there are many vibration sources which overshadow vibration signal due to abnormality, a filtering technique is required in order to eliminate the background signals such that correlation between the vibration signal and the faults can be obtained. This study tried to find a method to determine frequency ranges, which depend on the structure of engine components, where the signal caused by malfunction is dominant over the background vibration. Single cylinder engine was used as the test engine. Results show an increase in filtered vibration signal as the clearance was increased.