Johnson, Matthew Christopher (2017-04). Design and Analysis of Axial and Radial Flux Magnetic Gears and Magnetically Geared Machines. Doctoral Dissertation.
Thesis
Over the last two decades, magnetic gears and magnetically geared machines gained interest as a promising technology for use in high torque, low speed applications. Magnetic gears accomplish the same task as mechanical gears, but they do so without mechanical contact between the moving components, instead relying on the modulated interaction between the flux generated by magnets on the rotors. Consequently, magnetic gears offer the potential to combine the compact size and cost effectiveness of mechanically geared systems with the reliability and quieter operation of larger direct drive machines. This work focuses on the development of analysis and design techniques for axial and radial flux magnetic gears and magnetically geared machines. Prototypes of an axial flux magnetic gear, a new compact axial flux magnetically geared machine topology, and a large scale inner stator radial flux magnetically geared machine were constructed and tested to calibrate and validate the analysis tools and investigate the practical considerations associated with the technology. Despite conservative design practices, the largest of these machines achieved a torque density of 82.8 kN?m/m^3. Additionally, a MATLAB-based infrastructure was developed for controlling various simulation models and analyzing their results. Specifically, parametric 2D and 3D finite element analysis (FEA) models were employed for most of the studies, including the designs of the magnetically geared machine prototypes. This system was also used to conduct other simulation studies focused on a plethora of critical design trends and multi-faceted characterizations of the technology's potential. Spurred on by the long simulation times required for FEA models, the later stages of the study describe the development and evaluation of generalized, parametric 2D and 3D magnetic equivalent circuit (MEC) magnetic gear models. These MEC models proved extremely accurate, matching the torque predictions of corresponding FEA models with an average error of less than 2%. The MEC models also achieved simulation speeds up to 300 times faster than those of corresponding FEA models. Collectively, this work provides the tools and methodology for the systematic evaluation of radial and axial flux magnetic gears. It also characterizes design trends for both topologies and validates the results with experimental prototypes.