Diagnosing Dimensional Defects and Valve Cracks using Machine Vision and Acoustic Emission

Valve originates from the French word, Soupape, meaning valve and its role is to control the mixture of air and fuel entering the engine and the exhaust fumes. In this research, dimensional defects and valve cracks were diagnosed with the help of machine vision and acoustic emission. After offline transfer of the images to MATLAB software, the images were processed and the dimensional parameters of the valve such as length, stem diameter, large diameter (seat) and valve stem curvature were measured. These parameters were compared with the actual sizes and the percentages of measurement error for these parameters which were estimated as 0.45%, 1.8% and 1.18%, respectively. Because there was no real way to measure the valve stem, the error percentage was not calculated for this parameter. To detect cracks from 60 similar new valves, an acoustic diagram prepared by AE-MAP 1.0 device and the acoustic diagram of 30 new and used valves were compared. The accuracy of the system in crack detection was estimated to be approximately 96.7%.