0
Technical Briefs

# Integrator Leakage for Limit Cycle Suppression in Servo Mechanisms With Stiction

[+] Author and Article Information
Soo Jeon

Department of Mechanical and Mechatronics Engineering,  University of Waterloo, Waterloo, ON, N2L 3G1 Canadasoojeon@uwaterloo.ca

J. Dyn. Sys., Meas., Control 134(3), 034502 (Apr 04, 2012) (8 pages) doi:10.1115/1.4005503 History: Received November 27, 2010; Revised November 23, 2011; Published April 03, 2012; Online April 04, 2012

## Abstract

It is well known that the positional proportional-integral-derivative (PID) control of a servo mechanism with stiction always leads to a limit cycle. Related to this fact, two basic questions have still remained unanswered. The first question is, if the limit cycle occurs, how large it becomes for a given value of stiction force. The second question, which is of more practical importance, is how we should modify the PID controller to avoid this limit cycle with minimal sacrifice of the servo performance. This paper presents a rigorous analysis to provide particular answers to these two questions, which turn out to be closely related to each other. More specifically, it is shown that, by exploiting algebraic properties of the state trajectory, a simple bisection algorithm can be devised to compute the exact magnitude of the periodic solution for a given value of stiction. Through the geometric analysis of the impact map, this result is then used to find the minimum value of the integrator leakage to avoid the limit cycle. The work in this paper will be useful as a specific reference in designing servo mechanisms with stiction free from limit cycle.

<>
Copyright © 2012 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

## Figures

Figure 1

Block diagram of the system in concern

Figure 2

State trajectory around the sliding surface

Figure 3

Illustration of Facts 2 and 3

Figure 4

State transformation for integrator leakage

Figure 5

Limit cycle magnitude (θ*) for different values of γ and ρ

Figure 6

Simulation result for γ = 1.95 for the case of d1  = 6.15, d2  = 2.57, and c = 100

Figure 7

Simulation result for γ = 2.1 for the case of d1  = 6.15, d2  = 2.57, and c = 100

## Errata

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections