Design, implementation and conduction of rotordynamic measurements on exhaust turbochargers

Abstract

This MSc thesis belongs to a bigger project named ARH-700 "Basistechnologien für Thermische Strömungsmaschinen". As purpose, this project pretends to lay the theoretical and experimental foundations of thermal turbomachinery in order to have a reference point of knowledge and know-how for future investigations on this topic. Despite the huge range of thermal turbomachinery, two restrictions were established as starting point: Output power around 1kW and high rotational speed. Combining those restrictions plus the fact that compressor wheel velocity should not exceed 500 m/s, optimum wheel diameter range became then similar to automotive turbochargers specifications. Particularly, the contribution of this thesis to the main project is the development of a rotordynamic measurement chain and an analysis interface. Indeed, the reason for building a new measurement chain lays on the need of a faster measurement system for rotordynamic tests to be adapted on a cold-gas test bench (meaning cold, a temperature up to 250°C). This test bench has already an embedded acquisition system, which is fast enough for thermodynamic purposes but not for rotordynamic tests. Then, the new system should be built and integrated so both systems could work concurrently complementing each other. Aside, both systems should also be able to work independently. The departure point of the thesis starts with the configuration of an acquisition system, adaptation and calibration of sensors, and the implementation of measuring sequences. Also a graphic user interface (GUI) based on Matlab® is developed. This GUI is able to manage the real-time acquisition, data postprocessing and evaluation of the results. Once those points are completed, the measurement is validated by testing a serial production turbocharger. After the validation, rotordynamic analyses are carried out on the ARH-700 project prototypes. Unfortunately, the manufacturing of the first prototype was delayed and therefore its rotordynamic analysis runs out of this thesis timeframe. However, its adaptation and balancing could be carried out and are included as Appendices. Then, the thesis experimental part is focused on the validation of the measurement chain and on a deep rotordynamic analysis of the automotive turbocharger. Nevertheless, a failure on the turbocharger during a thermodynamical test made impossible to continue the tests. Hence, the experimental part is finally dedicated to compare the rotordynamic response between a fully operational turbocharger and a damaged on, while validating the measurement chain in the process. Simultaneously, some complementary simulations of linear rotordynamics with a simple model are carried out with otordynamikberechnungsprogramm”, calculation software for rotors. This report is structured in five main blocks. Firstly, a summary of the rotordynamic and signal processing theoretical background required for the development of the thesis and the understanding of the analysed data. Secondly, the measurement chain is described in detail stage by stage mentioning all the devices involved. Thirdly, simulations are carried on following a simple rotordynamic model. Fourthly, experimental tests configuration and conduction are explained, as well as the obtained results. Finally, some conclusions are formulated in terms of results comparison and also recommendations to improve the measurement chain for future tests beyond this thesis.