Diagnostic approach for an automated filling machine: setupand preliminary results

High production rate machinery is more sensitive to stops and maintenance strategy due to higher loss of production. This paper presents the initial steps of a methodology which aim to realize a condition monitoring system to be embedded on a medicine capsule filler machine. The first required is the gathering and classifying of the machine criticalities, and in this paper a FMEA (Failure Mode and Effects Analysis) is performed, resulting in a cam follower as the main machine fault cause. Subsequently the initial phases of the diagnostic parameter identification process are presented. In this case the parameters chosen are the machine vibration data coming from a triaxial sensor mounted in 4 different locations on the machine. The selection of the best combination of accelerometer-location/accelerometer-axis involves the evaluation of the averaged vibration data and some of their statistical indicator such peak level, RMS and kurtosis. This approach has been applied in a framework of a applied research supported by Emilia Romagna Region, with the cooperation of LIAM (Industrial Laboratory for Packaging Machine Automation) and the company I.M.A. (Industria Macchine Automatiche)