Analysis of a permanent magnet hybrid linear stepping motor

Master Thesis

1998

Permanent link to this Item
Authors
Supervisors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher

University of Cape Town

License
Series
Abstract
Hybrid linear stepping motors (HLSMs) are regarded as an excellent solution to positioning systems that require a high accuracy and rapid acceleration. The advantages such as high efficiency, high throughput, mechanical simplicity, high reliability, precise open-loop operation, low inertia of the system, etc. have made this kind of motor more and more attractive in such applications as factory automation, high speed positioning, computer peripherals, numerically controlled machine tools and welding apparatus. This motor drive is especially suitable for machine tools in which the positioning accuracy and repeatability are the key problems. Using the microprocessor controlled microstepping mode, a smooth operation with standard resolution of a few hundred steps/mm can be obtained. Although rotating hybrid stepping motors are well coveredĀ· by literature, there are only a few papers on HLSMs published so far. The objective of this thesis is to apply the reluctance network approach (RNA) and the finite element method (FEM) to performance calculations of HLSMs. A comparison with the experimental measurements is done to evaluate advantages and disadvantages of these two methods. The static characteristics show that the FEM demonstrates a better correlation with the experimental results than the RNA. The reason is that some assumptions have been made in order to obtain a simplified model in the RNA. The RNA, in general, tends to overestimate the forces. In the case of instantaneous characteristics simulation, both the FEM and RNA compare favourably with the experimental data though the FEM gives more accurate (albeit underestimated) results. However, the RNA is a very efficient approach considering the computation time. The optimisation of the finite element model is very important for obtaining the best possible results. The density of the finite element meshes, the aspect ratio of the elements and reducing the problem in size by using symmetry are vital considerations. The studied HLSM possesses a very small airgap, which made a judicious mesh scheme extremely important. Different methods of force calculations were analysed and it was found that both the Coulomb's approach and the Maxwell stress tensor are efficient and accurate methods to implement. The classical virtual work CVW method is very time-consuming since it requires two solutions. The determination of small incremental displacement is also difficult in the CVW. The experimental investigations of the small HLSM need a very high accuracy, Considerable efforts are needed to minimize all sources of noise and interference. It has been found that amplitude of instantaneous force ripple increases when the phase current amplitude increases. The superimposition of the 3rd harmonic on phase current can effectively reduce the force ripple. The transient tests focused on the start-up and braking characteristics. The start-up tests show that the higher the load is applied the longer the start-up time is required. On the contrary, the braking time can be significantly reduced with an increase in load. It can also be concluded that the higher the step resolution, the less settling time is needed for both start-up and braking processes. The recommendations from this thesis for further research on the subject of the HLSM analysis are that the reluctance network model should be improved to take into account the magnetic saturation and edge effects and the FEM enhanced classical approach should be attempted to take advantages of the high accuracy of the FEM without significant increase in the solution time.
Description

Reference:

Collections