Smart monitoring of worm gears

Worm gearboxes are commonly used in many various fields of industrial applications such as escalators, presses, conveyors etc. However, heavy industries face important problems about this type of gears due to unexpected failures. These failures, for instance pitting, wear and tooth breakage, cause so many problems such as waste of time, outages and damages on human health. The vibration signal of a gearbox carries the signature of the fault. Hence, vibration measurement and its graphical representation plays an important role for analysis of physical conditions of gearboxes. Although there are many options for vibration measurement systems, cost effective and portable microcontrollers based monitoring system is a good option. This thesis focuses on monitoring condition of worm gearbox via a developed smart device and the detection of pitting damages which have been realistically simulated on a few tooth surfaces. For this purpose, smart monitoring device has been developed via STM32 platform with ARM Cortex M4 microcontroller. Data acquisition system, various vibration analysis techniques, fault detection and visualization have been investigated and integrated in one system. Measurement of angular velocity of shaft, real time monitoring of vibration signal, time synchronous average, fast Fourier transform and statistical metric analysis have been carried out with developed monitoring device. The results of vibration analysis have been visualized on the screen and vibration data has been transferred into memory stick. Moreover, dynamic squared error, image processing based fault detection and support vector machine based pattern recognition method have been applied on vibration data acquired from worm gearbox for better explanation of effects of pitting failures.