2D & 3D electromagnetic and material loss analysis of an axial flux permanent magnet machine

  • Paul Leslie Chandler

    Student thesis: Doctor of Philosophy (PhD) - CDU

    Abstract

    Advances in computing hardware, software and manufacturing have helped revive the academic and commercial interest in a machine topology known as an axial flux machine (AFM). Traditional linear circuit analysis and basic non-linear Finite Element Analysis (FEA) have provided suitable designs for a wide range of AFM prototypes worldwide. This thesis examines the design of a particular single-sided AFM prototype developed at the Charles Darwin University in 1993 for use in a solar powered racing car. A thorough examination of the machine design shows a reasonable agreement between design specifications and measured outcomes after accounting for a 10% decrease in permanent magnet remanence. The unique property of the AFM whereby mechanical flux weakening is achieved via air gap adjustment is investigated, focusing on the resulting change in several important machine parameters. The effect of an air gap distance increase from 1mm to 4mm is confirmed to reduce the machine constant by 36%. 2D Transient FEA shows that the minimum in cogging torque occurs at a magnet arc width of 128.2°e at a nominal air gap distance of 2mm. 3D FEA confirms the fact that a significant radial flux component exists in both the rotor and stator under stalled conditions. While it is unlikely that radial flux component will significantly impinge upon the efficiency of the CDU solar car motor, lower pole count machines are likely to suffer. The effect of rotor lamination combined with 3-segment per pole magnet segmentation is shown to have the potential for a rotor eddy current loss reduction of 5.5 watts.
    Date of Award2004
    LanguageEnglish
    Awarding Institution

    Cite this

    2D & 3D electromagnetic and material loss analysis of an axial flux permanent magnet machine
    Chandler, P. L. (Author). 2004

    Student thesis: Doctor of Philosophy (PhD) - CDU