Magnetic Disk Drive Technology: Heads, Media, Channel, Interfaces, and Integration / Edition 1

by Kanu G. Ashar

Overview

This book will take you from basic academic knowledge about magnetics to a proficient understanding of the most recent advances in the technology. This book offers the latest information about disk technology, including: the fundamentals of magnetics, MIG heads, thin film heads, magnetoresistive heads, thin film media, electrical and mechanical integration of these components into a drive, and how to record writing and reading processes magnetically. You'll also learn about giant magnetoresistance, contact recording, and future disk drive industry developments. MAGNETIC DISK DRIVE TECHNOLOGY is an excellent guide for practitioners, researchers, and recent graduate engineers in the field of magnetics.

Product Details

ISBN-13:	9780780310834
Publisher:	Wiley
Publication date:	09/27/1996
Pages:	368
Product dimensions:	7.20(w) x 10.31(h) x 1.02(d)

About the Author

About the Author Kanu G. Ashar is a consultant and adjunct professor at Santa Clara University, California. Associated with IBM for thirty-four years, he established and managed advanced semiconductor device technology departments at IBM's East Fishkill Laboratory, New York. Later he moved to San Jose and managed magnetic recording groups, which included thin film and MR head projects at IBM Development Laboratory. He accepted an assignment to establish and manage magnetic and optical recording groups at Tokyo Research Laboratory of IBM Japan. He was invited to the Indian Institute of Science, Bangalore, India, as a visiting professor where he taught microprocessors and applications. Dr. Ashar has received several patents and is author and co-author of numerous publications in the fields of semiconductor devices and magnetic recording. He has received three patent plateau awards from IBM, and is recipient of the management Excellence Award from IBM Japan.

Read an Excerpt

Magnetic Disk Drive Technology

Heads, Media, Channel, Interfaces, and Integration

---

By Kanu G. Ashar

John Wiley & Sons

ISBN: 0-7803-1083-7

---

Chapter One

Introduction

1.1 DISK DRIVE INDUSTRY

The progress in magnetic recording technology, particularly that for rigid disk drives, from the mid-1950s to the mid-1990s has been as dramatic as that in semiconductors. Semiconductor memory and rigid disk drives are part of the storage hierarchy of a computer system as shown in Figure 1.1. Within the high-performance memory system, a hierarchy of performance and cost ranges from bipolar cache memory at the one end to dynamic memory solid-state devices at the other, as shown in the figure. A second sector of the system hierarchy is evolving as stand-alone high-performance drives and libraries of disk drive arrays. If solid-state devices had sufficiently low cost and could be organized as reliable nonvolatile storage, there would be no need for the disk drives. With decreasing semiconductor costs, there is a temptation to consider replacement of disk drives with solid-state memories. For this reason, it is interesting to compare the evolution of the two most dynamic technologies of this era.

Cost per bit is a predominant criterion in these technologies. The cost per bit in each of the technologies is driven by the areal density or area per bit. Much has been written in the technical and popular press about semiconductors, but disk technology has received little attention. Table 1.1 shows a density progression in both semiconductor memory (DRAM) chips and disk drives for the period 1980 to 2000. It is remarkable that the bit area of the disk drives had tracked the bit (cell) area of the semiconductor memories closely throughout the period of 1980 to 1990. The ratio of semiconductor bit price to that of disk drive bit has remained approximately 100 during this period. From 1990 through 2000, the density of disk drives is increasing at a faster pace than that of the semiconductors. With these trends, it is highly improbable that the semiconductor memory bit cost could catch up with the disk drive bit cost within the foreseeable future. In the case of semiconductors, memory cell density increases are usually attributed to improvements in photolithography. In disk drive technology, major advances are taking place due to innovations in magnetic head, media, and channel electronics.

Worldwide disk drive sales volume in 1995 was estimated as 70 million drives with revenues of about $25 billion. Comparatively, global semiconductor memory revenues in 1995 were $44.5 billion. The demand for disk drive storage has been steadily increasing due to the following factors:

1. Increasing demand of desktop computers and workstations.

2. Introduction of a variety of notebook, subnotebook, and laptop computers.

3. Replacement of large mainframe computer drives by arrays of small disk drives.

4. Growth of storage-hungry graphics and multimedia applications.

5. Use of removable small drives in noncomputer consumer applications.

6. Massive and yet unforeseen increases in computer storage requirements of network data bases.

Conversion of information from paper to the more accessible computer media has just begun, yet computer storage is still only a small fraction of the total amount of information stored. So, unless unforeseen conditions intervene, the expansion of the disk drive industry will continue at a high rate in the future.

1.2 DISK DRIVE TECHNOLOGY DEVELOPMENT

Introduced in 1957, the IBM Model 350, or random access method of accounting and control (RAMAC), was the first disk drive. This drive, invented at the IBM research laboratory in San Jose, California, consisted of 50 rotating disks mounted on a vertical shaft (See Figure 1.2), each 24 inches in diameter and each having a magnetic medium coating to store data. Access to this data was by a pair of air-bearing supported heads mounted on an access arm that could be moved under servo control to 1 of the 50 disks. The heads were also able to move in and out across the radius of a disk. The storage capacity of the system was 5 MB, and users paid a rental of $130 a month for the system. The disks rotated at 1200 RPM (revolutions per minute), and the data rate of the information was 12.5 kB/s.

The inductive head and magnetic disk material for this drive originated in earlier work on tape recorders and magnetic drums. (Magnetic drums consisted of a rotating drum with a magnetic coating and a series of fixed heads mounted one head per track.) The advantage realized with the new rotating disk technology was the combination of nonvolatile storage of data with very fast access to this data and, at the same time, low cost due to fewer heads. These characteristics filled the technology gap of the data processing industry between inexpensive tape recorders and expensive magnetic cores used at the time for computer memory. One persistent drawback of many of the early magnetic read/write schemes was the requirement of close proximity between the magnetic medium and the read/write head. In tape recorders, for example, the tape is in contact with the tape head; if the tape were to run continuously at a high rate, the tape would soon wear out and possibly the head as well. The key development that led to disk drive technology was the invention of a low-mass, air-bearing slider carrying a magnetic head floating at a precise spacing between the head and the magnetic medium. Data or information is written magnetically along circular tracks on a disk. A bit is the smallest unit of data or information and consists of a "0" or a " 1." The linear bit density is the number of bits written along a distance of 1 inch on one of these tracks. The track density is the number of tracks per inch along a radius of the disk. The areal density is the product of linear bit density and track density. For example, the Travelstar drive announced in 1995 by IBM, which uses magnetoresistive heads, has a linear bit density of 127.2 Kb/in. (50 Kb/cm) and a track density of 7257 T/in. (2857 T/cm), giving an areal density of 923 Mb/[in..sup.2] (143 Mb/[cm.sup.2]). Increasing linear bit density over time has depended on the scaling down of three factors:

1. Reduction of the head to disk spacing commonly known as flying height.

2. Reduction of the gap size of the head.

3. Reduction of the thickness of the disk magnetic media.

Table 1.2 shows these trends in IBM disk drive characteristics from RAMAC (System 350) in 1957 to 3390 in 1989. Apparently, these three parameters have been reduced in values by about two orders of magnitude over the 32-year period. Table 1.2 spans a total period of 38 years. The linear density has improved by a factor of 1300 during this period. This is the result of progress in inductive head designs and improvements in both ferrite medium prior to 1987 and thin film medium after 1987, and also implementation of magnetoresistive heads since 1990.

Increasing track density over time has depended on improvements in the technology of accurate head positioning. These improvements include reduction in track misregistration, innovations in servo techniques, higher-sensitivity heads, and advances in track width control. Table 1.2 shows that the accompanying improvement in track density over this period is a factor of 355. Some detail on the head and slider technology are also shown in the table.

Many of the developments in heads, disks, interface controls, channel electronics, and packaging have been due to the gradual increase in understanding of the technology and development of tools. However, there are several highlights in the technology that must be pointed out.

The earliest recording heads were made of laminated Mu-metal (NiFeMoCu) with gaps formed by copper shims. A major improvement in the mid-1960s was the replacement of Mu-metal by ferrite. The high resistivity of ferrite meant that head lamination was not required. In addition, ferrites are easy to shape and grind. Ferrite surfaces are wear resistant and do not corrode. Many subsequent improvements in ferrites have continued to make them competitive as an alternative head technology. One such advancement is the metal-in-gap (MIG) head. Placement of high-magnetization metal (Sendust) in the gap of ferrite heads made the use of continuous-metal, thin film medium disks possible. Thin film disks resulted in a path to higher linear densities. Continuous progress in film disks through reduction in defects, lowering of medium noise, and control of corrosion have continued to boost areal densities. An important step in head technology came in 1979 with IBM's introduction of thin film heads. This technology allowed heads to be fabricated by the same photolithographic techniques used in the semiconductor industry. This approach led to still higher linear densities and track densities.

The most recent leap in technology came in 1991 with IBM's announcement of the production of the magnetoresistive head. Magnetoresistive (MR) heads are combined with new low-noise, thin-film disk media to provide a new range of capabilities for disk drives. The MR head provides 5 to 10 times higher sensitivity compared to ferrite or thin film heads. Increases in track density are achieved by making dual element heads in which the write element of the head is the usual thin film inductive head and the read element the head is an MR sensor. Another attribute of the MR head is that its sensitivity does not vary with disk rotational velocity; hence, it is possible to provide high bit densities for small disk drives used in laptop computers.

Table 1.2 describes IBM disk file evolution and fairly well portrays the industry progress. However, it should be noted that significant progress was pioneered by several other manufacturers as well. Noteworthy among them:

1. Fujitsu's F 6525 Eagle II file, announced in 1982, had 24.4 Kb/in, linear density and 22.2 Mb/[in..sup.2] areal density, achieved with sputtered medium on a 10.8 in. disk, ferrite head, and head-disk spacing of about 150 nm.

2. Maxtor's XT-8760E, announced in 1987, had 31.5 kb/in. linear density and 43.5 Mb/[in..sup.2] areal density.

3. In 1993, Seagate shipped a 7200 RPM, 95 mm (3.5 in.) disk drive that has 4.17 ms average rotational delay, and an asynchronous data rate of 20 MB/s.

4. Areal technology in 1993 produced a 65 mm (2.5 in.) disk drive with glass substrate disk to reduce head-disk spacing. The linear density of 80 Kb/in, and areal density of 220 Mb/[in..sup.2] were achieved in the drive with an inductive thin film head.

Additional comparative study of disk files from several manufacturers can be found in.

1.3 DISK DRIVE HEAD TECHNOLOGIES

Most dramatic changes in disk drive technology have happened in head development and manufacturing. This is the reason why three chapters in this book are dedicated to heads. Table 1.3 presents a qualitative comparison of four head technologies. A plus (+) refers to a favorable property while a minus (-) refers to a less desirable attribute. The ranking of heads is based on state of the art in the early 1990s and is subject to change depending on future developments.

The linear density of ferrite heads is low because the low magnetization of the ferrite material creates low writing fields. MIG heads have higher magnetization, and thin film heads and MR heads provide still higher write magnetization fields.

The track density limitations of ferrite and MIG heads are due to head material and processing constraints. Some progress is likely in this field with single-crystal ferrites and incorporation of thin film technology. Track definition for thin film and MR heads is controlled by lithographic processes. The higher sensitivity of the MR head allows narrow tracks with acceptable signal levels.

The data rate for ferrites and MIG heads is limited by high coil inductance. Thin film heads have considerably lower inductance compared to ferrite heads because of miniaturized geometry. However, as the number of turns for thin film heads increases, the inductance and capacitance of the head circuit result in self-resonance, which eventually limits reading at high data rates. The MR head read sensor is effectively a single-turn read element and therefore does not suffer from these high-frequency reading problems.

Table 1.3 further indicates that over the years a great deal of experience has been built up from earlier technologies like the ferrite head. The MR head, in contrast, is relatively new and has several manufacturing and application challenges that require extra care and often add to the product cost.

Figure 1.3 shows the progress and density constraints of these head technologies. One-sided and two-sided MIG heads are making progress in increasing densities. However, with progress in thin film and MR heads, importance of MIG heads for disk drives is declining. Thin-film head densities are increasing beyond prior expectations, through innovations in slider technology and reduction in inductance factor (discussed in Section 5.8). The MR head characteristics are ideal for disk drives of all sizes. The areal density limit is over several gigabits per [in.sup.2]. For multigigabit per [in.sup.2] areal densities, newer approaches such as GMR (giant magnetoresistance) heads may be required. Ferrite, thin film, and MR heads are discussed in Chapters 4, 5 and 6, respectively, while GMR is described in Chapter 11. Many factors influence these limits. Improvements and innovations in any of these technologies cannot be ruled out. We shall discuss some of the reasons for these limits and point out exploratory efforts under way to extend these limits in these chapters.

1.4 SCOPE OF THE BOOK

There are several books on the subject of magnetic recording. They generally cover a wide range of magnetic recording applications, including tape recording, audio recording, and sometimes magneto-optical recording. To discuss the subject in adequate detail, this book is confined to magnetic recording on rigid disks. We describe the essential elements of rigid disk recording heads, disk media, electronic channels, and the integration of components into a disk drive. There is a long history of developments in these fields, and more innovations are taking place all the time. We offer the reader an up-to-date perspective on the technology, including relevant theory, modeling, current component design practices, design limitations, and potential innovations. Material from publications and patents up to early 1996 is included to keep the book current and useful to professionals engaged in the industry. This book is written primarily for practitioners in the disk drive industry or graduates interested in entering this field. Graduate students of science or engineering should be able to follow the content with little or no difficulty. An attempt has been made to keep the material simple and explanatory rather than detailed or exhaustive. References are provided for those who wish to explore specific topics in greater detail. Illustrations are used profusely to clarify the text and the geometries involved. Mathematics in the book are kept at a graduate student level. Examples and their solutions are given to clarify concepts and to provide help in handling magnetic unit conversions. Every attempt is made to report the latest developments in the field by emphasizing significant, new ideas and concepts. The disk drive technology involves many disciplines such as physics, chemistry, electrical, and mechanical engineering and material sciences. Also, the technology is divided into sectors such as heads, media, electronics, processing, manufacturing, development, research, and marketing. Individuals working in one area often have difficulty understanding the requirements and jargon of the other areas. One of the book's motives in providing this self-study information is to promote understanding and communication among complementing sectors. Here, managers may find helpful technical and general information to broaden their perspective about the technology and the industry.

1.5 OUTLINE OF TOPICS COVERED

A short introductory comment on each chapter is given here. This first chapter explains the importance of rigid disk drives in computers and data processing industry. Historical progress of disk drive technology is examined to get a perspective on future developments in the field. Major components of the technology responsible for dramatic enhancements in the field are briefly reviewed.

(Continues...)

---

Excerpted from Magnetic Disk Drive Technology by Kanu G. Ashar Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---

Table of Contents

Preface.

Acknowledgments.

List of Symbols, Abbreviations, and Formulas.

Introduction.

The Fundamentals of Magnetism (James C. Suits).

Disk Drive Magnetic Recording.

Ferrite and MIG Heads.

Thin Film Heads.

MR Heads.

Thin Film Media.

Recording Channel.

Magnetic Disk Recording Integration.

Head-Disk Interface (Roger F. Hoyt).

Future Trends in Technology.

Index.

Author's Biography.