A Detection of the Quality of Sound Gear based on Digital Signal Processing Techniques

Despite the fact that products in the industries are made by the same formula, the imperfections and dissimilarities between the produced products are influenced by noises and mechanical connections, moreover the variations of temperature with time during the production contribute to these imperfections. Previously and even to date the produced sound gears in many industries were tested by using the human ears to determine the defect gears and good gears, other methods like pitch detection method and faults detection method are applied. However those method have been found to be unreliable and therefore accelerated to bad sorting between good gears and bad gears. In this paper we proposed a powerful approach which used Digital signal processing techniques to determine the quality of the sound Gears. We integrated Correlation technique, Normalization and alignment methods. The techniques were applied to the extracted sound signals. The final extracted features (amplitudes, frequencies and phases) were classified and hence used to determine the good (non-defect) and bad (defect) gears. The entire process provided the perfect and higher accurate results and suggest that the tool can be used to identify good and bad sound gears