Development Of A Magnetic Gear Acoustic Optimizer

Abstract

Contactless magnetic gearboxes have been identified by NASA and others to be a potential key enabling technology necessary to realize future fully electric aircraft and spacecraft. These gears are significantly more efficient and reliable compared to mechanical gearboxes. Additionally, mechanical gearboxes are louder than magnetic gearboxes; however, the acoustic noise produced by magnetic gears is typically isolated to a few frequencies as opposed to a wide spectrum. Identifying and eliminating the sources of this noise would significantly improve the attractiveness of magnetic gears to companies. Furthermore, developing and testing magnetic gears can be very time intensive. Magnetic gears typically fall into two broad categories: axial flux gears and radial flux gears. Designing either of these gear types can be tedious due to the relatively large number of components, especially since each gear model must be nearly identical when testing various parameters. In addition to creating these models, simulating magnetic gears usually required importing force vectors to emulate the magnetic forces imposed upon the gear during various modes of operation. Individually adding upwards of one hundred force vectors is extremely inefficient. This research aims to create several automation scripts aimed at resolving these issues.

This research involves the development of a lightweight, modular data analysis tool designed to analyze the acoustics of magnetic gears. The tool will provide a quick analysis of acoustic data and output design optimization suggestions. These optimization suggestions will allow the user to determine how various parameters affect the acoustics that the magnets gear generates. Data used for these calculations will be stored within a local database that can be accessed by the user. The analysis tool also includes functionality for generating plots of the data being analyzed. While base data for this analysis tool will be provided, the tool is designed to use data provided for the user that is specific to the gear design they are testing since different designs could have significantly different input data. Additionally, several companion automation scripts will be developed to allow for the rapid creation and testing of magnetic gear simulation models. These scripts will assist in generating simple magnetic gear models from user-provided parameters. The automation scripts are designed to run directly in 3D modeling software and provide a significant amount of flexibility in their designs. The resulting gear models can serve as a base for more complex gear designs or as a quick method to determine the feasibility of a design. The scripts will allow for the creation of both a radial flux and axial flux magnetic gear. Another significant benefit of these scripts is the consistency of the gear models which will help to test a large number of gear models that slightly vary input parameters.