Browse Prior Art Database

Method for Enhancing Servo Charastristics by Adding PES Linearization Function And Its Optimization Procedure in Manufacuring Test Process

IP.com Disclosure Number: IPCOM000015917D
Original Publication Date: 2002-Jul-22
Included in the Prior Art Database: 2003-Jun-21

Publishing Venue

IBM

Abstract

1. Introduction

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 31% of the total text.

Page 1 of 10

  Method for Enhancing Servo Charastristics by Adding PES Linearization Function And Its Optimization Procedure in Manufacuring Test Process

1. Introduction

      One of the method used in detecting the head position is to use the MPES (Master Position Error Signal) and SPES (Slave Position Error Signal) which are defined in the following formula :

MPES = {([A] - [B]) / ([A] + [B])} * H + 80 h
(1)

SPES = {([C] - [D]) / ([C] + [D])} * H
(2)

where [X] stands for the amplitude of burst A/B/C/D written on the media, and H for the gain of the PES calculation.

      MPES and SPES values change dramatically at track center and track border respectively, while both are less sensitive at middle of track center and track border where MPES and SPES are switched to use whichever it is more sensitive for position detection. This kind of "unfairness" in the sensitivity of the PES may lead to critical problems such as off track writing.

      Thus as a solution, we propose a method using the "N segment" partitioning,and the optimization procedure of the parameters used in this method.

2. Basic Idea

      As for the PES linearization method, we assume that 1 track = 256 PES is divided into the following 9 "segments":

Segment 0 : [0x00, 0x10) PES
Segment 1 : [0x10, 0x30) PES
Segment 2 : [0x30, 0x50) PES
Segment 3 : [0x50, 0x70) PES
Segment 4 : [0x70, 0x90) PES
Segment 5 : [0x90, 0xB0) PES
Segment 6 : [0xB0, 0xD0) PES
Segment 7 : [0xD0, 0xF0) PES
Segment 8 : [0xF0, 0x100) PES

      The basic idea of the PES linearization is to equalize or at least relax the large difference in PES slopes at various track locations.

      By dividing a single track into 9 segments and to multiply a simple "slope value" for each segment individually will be not so accurate, but a convinient way to relax the PES slope difference.

3. PES Linearlizing Method

      The actual lineariizing method is explained in this section. The signal error feedback routine in the SA (Servo Assist) will be fed a modified PES value rather than the raw PES value that is calculated from the burst patterns A,B,C, and D written in the servo area. To be more precise, let PESraw be the calculated raw PES value (which was actually fed to the feedback routine in past products) and PESmod be the modified (linearized) value that will be fed instead. The drive has a table for each segment, each servo zone, and each head consisting of :

- slope value

1

Page 2 of 10

- offset value

By applying this table, the SA modifies the PESraw in according to the following formula :

           PESmod = offset value[PESraw] + slope value [PESraw] * (PESraw - offset value[PES raw]) (3)

      where offset[PESraw] and slope[PESraw] is the offset value and the slope value saved in the table for segment which PESraw belongs to.

      Observing each segments, the compensation will only be a multiplication, but when viewed in whole, the compensation depends on the segment the PESraw belongs to, and different parameters are used for compensation, which satisfies the ability to linearize the PES...