Metodika stanovení modifikací ozubení podle zátěžných podmínek převodovky.

Abstract

The dissertation thesis follows up the issue of creating real contact patterns on gearing in order to reduce acoustic emission. The topic of reducing gear noise with the current trend of keeping the weight of the car transmission as low as possible represents a huge investment in development for manufacturers. This trend is mainly indicated by the strictly prescribed limits by the European Union, which specify the maximum level of emissions gasses and noise for passenger cars. Streamlining of development activities in the design of the ideal gear geometry brings significant financial and time savings to manufacturers. Thanks to this, it is necessary to introduce new procedures not only in the field of gear design, but also in production. However, this carries the risks associated with unexplored issues and its difficult implementation on the problem. The introductory chapter first presents the current trend in the development of powertrain units and then analyses the current knowledge defining the basic procedures leading to the design of silent gearing in terms of tooth geometry. Furthermore, the directions of development of silent gearing in practice are elaborated, because the automobile gearbox is and always will be the result of a number of design, production and technological compromises. The topic of this thesis is based on the needs of the technical development of the ŠKODA AUTO a.s., where on real gears in the serial gearbox are practically verified known things and newly developed theoretical foundations using available technologies. Thanks to this, all new knowledge can be immediately applied in the process of gearbox optimization. The thesis describes in detail the methodology of creating real contact patterns, including processed proposals leading to the possible determination of real coefficients of transverse contact ratio and overlap contact ratio. Furthermore, the thesis deals with possible methodological procedures leading to the determination of the optimal size and types of microgeometric modifications. For this purpose, the information obtained from real contact patterns and from the noise measurement in the cab of a car, which are processed into data files. With using geometric analysis, machine learning in Python software and the application of neural networks, a data files are evaluated. Finally, an accelerated optimization test of gear microgeometry is elaborated in detail, applied directly in the serial gearbox. Its use allows fast detection of parameters leading to a reduction acoustic emission.