Date of Award

Summer 1999

Document Type

Thesis

Degree Name

Master of Science in Electrical Engineering - (M.S.)

Department

Electrical and Computer Engineering

First Advisor

Timothy Nam Chang

Second Advisor

Barry Cohen

Third Advisor

Andrew Ulrich Meyer

Abstract

In this work, feasibility of using low cost, low resolution sensor for high performance brushless dc (BLDC) motor speed control is investigated. Conventional control, using a tachometer or high resolution encoder, suffers from drawbacks such as high cost, large physical volume, and high sensor processing bandwidth. On the other hand, sensorless BLDC motor, appealing in its hardware simplicity, does not provide sufficient fast performance. Using a standard low resolution sensor, such as a hall sensor or commutation encoder, a compromise between cost and performance can be obtained. However, the use of a low resolution sensor does pose a challenge to the control design: the sensor signal is discrete and speed dependent. Together with the nonlinear drive voltage/speed characteristic of the motor, control of the BLDC motor requires a more advanced algorithm than fixed gain control. This thesis presents a speed dependent control scheme to produce optimal performance. The characteristics of the control scheme is first assessed by numerical simulation, based on the mathematical model of the BLDC motor. This is followed by experimental verification of the BLDC motor. From the available data, it is concluded that speed dependent control provides significant advantages over fixed gain control when low resolution sensor is used.

Share

COinS