No description available.

A research report submitted in fulfilment of the requirements for the degree of Master’s in Engineering to the Faculty of Engineering and the Built Environment, School of Mechanical, Industrial and Aeronautical Engineering, University of the Witwatersrand, Johannesburg, 2018 / The automotive industry is experiencing a major transition with regards to powertrain where the traditional internal combustion engine is being replaced by hybrid or all-electric solutions. As a result of this transition, many powertrain engineering firms, such as Integral Powertrain, are required to adapt or develop tools to improve the quality and time to respond to the rapid increase in new electric motor enquiries. This paper details the development of the foundations of a software tool (AIEED) used to improve the time and quality of response to new motor enquiries. The foundations of the AIEED tool are built into a Microsoft Excel document which contains the necessary logic required to convert customer requirements data into geometrical design data. A series of macros are included in the tool which allow for the automated generation of a first-level computer-aided design (CAD) model of the motor, including the following components; stator, windings, rotor, magnets, housing and cover. The CAD models, which serve to improve the quality of response to new motor enquiries and provide a basis for further detailed development, are generated on the CATIA V5 design platform. In addition, the tool makes use of the CAD model to determine the estimate cost of the unit based on average material costs per unit mass defined in the materials database stored in the tool. It is estimated that the AIEED tool reduces the response time to new motor enquires by 75%. Certain assumptions and limitations have been applied such that the foundations of the tool only include support for the following motor specifications: Surface Permanent Magnet Alternating Current cylindrical radial flux motors (external stator and internal rotor). • The units are designed for 3-phase supply. • The winding configuration is limited to concentrated double layer option. • Parallel tooth, arc-bottom, semi-closed stator slot profiles. • Arc-shaped radially magnetised surface magnet rotor configuration option. A simple data verification filter is applied to minimise the potential for invalid inputs. However, a full validation and verification process is recommended for future work. In addition, it is recommended that the tool be expanded to support additional motor types and specifications as well as package the foundations into a single, easy-to-use utility. / XL2019

M.Ing. / Shaft encoders, are devices generally used in speed and position control applications for sensing position of rotational and linearly moving objects. Most conventional shaft encoders operate on the principle of reading encoded information off a disk, which is fitted to the shaft or using resolver units. Problems are often experienced in fitting high-resolution encoder disks or resolvers to small motors or moving objects. The resolution of an encoder system, using an encoded disk, is a function of the maximum number of slots that are cut in the disk. If the disk itself has to be very small, physical constraints limit the accuracy of the encoder. However, any machined surface will have characteristic marks on it and it is believed that these marks can be used to obtain a unique analogue signature for the shaft. This analogue signature can then be used to gain further knowledge about the rotational speed and possibly the position of the shaft. This project was an investigation into the possibility of gaining accurate, absolute positional information for a rotating shaft. An optical source- and detector system is used in the process of observing the irregularities on the shaft surface. This signal data is gathered and processed using a technique, which compensates for variation in shaft speed during the period in which data is sampled. The entire algorithm, which calculates the absolute position of the shaft from the data gathered, was implemented on a digital signal processor board. Results were evaluated and the maximum non-linearity of the test system was found to vary between 1.8% and 3.7% of full-scale deflection. The resolution of the sensor was between 1.5 and 2.5 degrees with hysteresis between 1% and 1.5% of full scale deflection. The test system allows for rotation in any direction. Possible problems that might be encountered with this type of encoder system are pointed out in this report. Future developments and possible improvements are also discussed.

Call number: LD2668 .T4 1960 C42