UPM Institutional Repository

A Sensorless Positioning System For Linear Dc Motor


Citation

Mat Saat, Ezril Hisham (2008) A Sensorless Positioning System For Linear Dc Motor. Masters thesis, Universiti Putra Malaysia.

Abstract

In recent years, linear motor has gained popularity as linear motion drive in industry and factory automation which provides an alternative to conventional rotary motor. This development was encouraged by the advantages offered by linear motors such as flexibility in size and design, clean and silent operation, ease of maintenance and deliver high performance for applications requiring linear motion. However, the main constraint for system designers to consider linear motors in their application is the cost for every complete package of linear motor. Generally, linear motors are expensive than their counterparts due to the number of permanent magnet used in the motors and the sensory technologies used for positioning system. Linear motors provide direct linear motion and normally are designed with specific length which requires more permanent magnets than rotary motor. Therefore, the costs of linear motors increase with the length of the motor itself. For a typical linear motor driver, a positioning sensor, usually attached to the motor, provides feedback positioning signal to the controller. This sensor is expensive and normally has the same length with the motor. The price for this type of sensor increases with the size, which eventually increase the cost of linear motors. For some applications, where positioning is not too critical such as robot end gripper, the high precision and expensive positioning sensor is not necessary. Removing this sensor from the system reduces the overall system cost, thus encouraging more development and application of linear motors. This research and study proposes a sensorless positioning system for linear DC motor (LDM). The ideas to control the position of the LDM are by controlling the current supplied to the motor. The technique used to control the current is by manipulating Pulse Width Modulation (PWM) signal which is generated by a microcontroller circuit based on Atmel AVR ATmega8535 processor. A simple model of LDM was constructed for experimental purposes. A mechanical spring was used in the motor design to absorb the force produced by the motor. The displacement of the spring will then be used to determine the position of the motor. Due to the harmonic oscillation produced in mass-spring system, the position of the motor oscillates before reaching the final desired position. In order to eliminate the harmonic oscillation effect, a few variants pattern of PWM signals are used to drive the motor. These variant patterns are single stage, dual stage, triple stage and quadruple stage. The variant patterns of PWM signals were created by combining multiple values of duty cycle running in a single PWM signal. A mathematical model for constructed LDM has been derived based on damped force oscillation of mass spring system theory. The equation of motion for LDM is then simulated using Matlab software. The time response of the system based on simulation results has been studied and proper adjustment to control parameters has been made to improve the rise time, overshoot percentage and steady state error. The adjusted driving parameters are then transferred to microcontroller unit for actual laboratory experiment. The objective of this research to develop a sensorless positioning system for LDM which include a motor driver and control approach was successfully achieved. Comparison between simulation results and laboratory experiment shows almost identical results. Linear relation between motor position and timing parameters used in control algorithms has been studied and linear equations have been derived. These linear equations will be converted into microprocessor programming as feed-forward control algorithms. Any desired motor position can be fed into the system and microprocessor unit will generates a proper PWM driving signal to the motor. The developed system can be used for any low cost applications which do not require precise positioning.


Download File

[img]
Preview
PDF
FK_2008_17.pdf

Download (483kB)

Additional Metadata

Item Type: Thesis (Masters)
Subject: Electric driving - Automatic control
Subject: Electric machinery - Alternating current - Automatic control
Call Number: FK 2008 17
Chairman Supervisor: Norhisam Misron, PhD
Divisions: Faculty of Engineering
Depositing User: Nurul Hayatie Hashim
Date Deposited: 08 Apr 2010 09:06
Last Modified: 27 May 2013 07:22
URI: http://psasir.upm.edu.my/id/eprint/5364
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item