Você está na página 1de 348

2005 ASM International. All Rights Reserved. www.asminternational.

Gear Materials, Properties, and Manufacture (#05125G)

Edited by

J.R. Davis
Davis & Associates

ASM International
Materials Park, OH 44073-0002
2005 ASM International. All Rights Reserved. www.asminternational.org
Gear Materials, Properties, and Manufacture (#05125G)

Copyright 2005
ASM International
All rights reserved

No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise, without the written permission of the copyright

First printing, September 2005

Great care is taken in the compilation and production of this Volume, but it should be made clear that NO WAR-
THIS PUBLICATION. Although this information is believed to be accurate by ASM, ASM cannot guarantee
that favorable results will be obtained from the use of this publication alone. This publication is intended for use
by persons having technical skill, at their sole discretion and risk. Since the conditions of product or material use
are outside of ASMs control, ASM assumes no liability or obligation in connection with any use of this infor-
mation. No claim of any kind, whether as to products or information in this publication, and whether or not based
on negligence, shall be greater in amount than the purchase price of this product or publication in respect of
FROM THE NEGLIGENCE OF SUCH PARTY. As with any material, evaluation of the material under end-use
conditions prior to specication is essential. Therefore, specic testing under actual conditions is recommended.

Nothing contained in this book shall be construed as a grant of any right of manufacture, sale, use, or reproduc-
tion, in connection with any method, process, apparatus, product, composition, or system, whether or not cov-
ered by letters patent, copyright, or trademark, and nothing contained in this book shall be construed as a defense
against any alleged infringement of letters patent, copyright, or trademark, or as a defense against liability for
such infringement.

Comments, criticisms, and suggestions are invited, and should be forwarded to ASM International.

Prepared under the direction of the ASM International Technical Books Committee (2004-2005),
Yip-Wah Chung, FASM, Chair.

ASM International staff who worked on this project include Scott Henry, Senior Manager of Product and
Service Development; Madrid Tramble, Senior Production Coordinator, and Kathryn Muldoon, Production

Library of Congress Cataloging-in-Publication Data

Gear materials, properties, and manufacture / edited by J.R. Davis

p. cm.
Includes bibliographical references and index
ISBN 0-87170-815-9
1. GearingManufacture. I. Davis, J.R. (Joseph R.)

TJ184.G34715 2005
621.833dc22 2005050105


ASM International
Materials Park, OH 44073-0002

Printed in the United States of America

2005 ASM International. All Rights Reserved. www.asminternational.org
Gear Materials, Properties, and Manufacture (#05125G)


Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

Introduction to Gear Technology

Chapter 1 Basic Understanding of Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Gear Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Types of Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Proper Gear Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Basic Applied Stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Gear Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Gear Manufacturing Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Heat Treating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Through Hardening and Case Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

Chapter 2 Gear Tribology and Lubrication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Gear Tribology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Lubrication Regimes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Classication of Gear Tooth Failure Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Lubrication-Related Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Elastohydrodynamic Lubrication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Lubricant Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Oil Lubricant Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Selection of Gear Lubricant Viscosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Application of Gear Lubricants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Case History: Failure of a 24-Unit Speed-Increaser Gearbox . . . . . . . . . . . . . . . . 35

Gear Materials
Chapter 3 Ferrous and Nonferrous Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Wrought Gear Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Surface-Hardening Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Through-Hardening Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Gear Steel Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Processing Characteristics of Gear Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Bending Fatigue Strength of Carburized Steels . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Other Properties of Interest for Carburized Steels . . . . . . . . . . . . . . . . . . . . . . . . 62

2005 ASM International. All Rights Reserved. www.asminternational.org
Gear Materials, Properties, and Manufacture (#05125G)

Other Ferrous Alloys for Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Nonferrous Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

Chapter 4 Plastics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
General Characteristics of Plastic Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Classication of Plastics for Gear Applications . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Metals versus Plastics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Plastic Gear Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Plastic Gear Manufacture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

Gear Manufacture
Chapter 5 Machining, Grinding, and Finishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Machining Processes for Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Selection of Machining Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Cutter Material and Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Speed and Feed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Cutting Fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Comparison of Steels for Gear Cutting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Computer Numerical Control (CNC) Milling and Hobbing Machines . . . . . . . . 113
Grinding of Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Honing of Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Lapping of Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Supernishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

Chapter 6 Casting, Forming, and Forging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

Casting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Forming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Forging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Chapter 7 Powder Metallurgy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

Capabilities and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Gear Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Gear Tolerances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Gear Design and Tooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Gear Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Quality Control and Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
Examples of P/M Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Heat Treatment of Gears

Chapter 8 Through Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Through-Hardening Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Through-Hardened Gear Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Hardness Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Distortion of Through-Hardened Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Case History: Design and Manufacture of a Rack . . . . . . . . . . . . . . . . . . . . . . . 158

Chapter 9 Carburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

Gas Carburizing: Processing and Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Quenching and Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166

2005 ASM International. All Rights Reserved. www.asminternational.org
Gear Materials, Properties, and Manufacture (#05125G)

Tempering of Carburized and Quenched Gears . . . . . . . . . . . . . . . . . . . . . . . . . 168

Recarburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Cold Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Selection of Materials for Carburized Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Carbon Content and Case Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Core Hardness of Gear Teeth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Problems Associated with Carburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Heat Treat Distortion of Carburized and Hardened Gears . . . . . . . . . . . . . . . . . 188
Grinding Stock Allowance on Tooth Flanks to Compensate for Distortion . . . . 201
Shot Peening of Carburized and Hardened Gears . . . . . . . . . . . . . . . . . . . . . . . . 208
Applications and Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
Vacuum Carburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212

Chapter 10 Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

The Gas Nitriding Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
Controlled Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

Chapter 11 Carbonitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

Applicable Steels and Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
Depth of Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
Hardenability of Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
Quenching and Tempering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247

Chapter 12 Induction and Flame Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

Induction Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
Flame Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253

Failure Analysis, Fatigue Life Prediction, and Mechanical Testing

Chapter 13 Gear Failure Modes and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Classication of Gear Failure Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Fatigue Failures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
Wear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Scufng . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
Stress Rupture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Causes of Gear Failures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Conducting the Failure Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Examples of Gear Failure Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283

Chapter 14 Fatigue and Life Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293

Gear Tooth Contact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Gear Tooth Surface Durability and Breakage . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
Life Determined by Contact Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
Life Determined by Bending Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
Gear Tooth Failure by Breakage after Pitting . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
Flaws and Gear Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
Bores and Shafts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

2005 ASM International. All Rights Reserved. www.asminternational.org
Gear Materials, Properties, and Manufacture (#05125G)

Chapter 15 Mechanical Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311

Common Modes of Gear Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
Stress Calculations for Test Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
Specimen Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
Tests Simulating Gear Action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
Gear Power-Circulating (PC), or Four-Square, Tests . . . . . . . . . . . . . . . . . . . . . 323

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329

2005 ASM International. All Rights Reserved. www.asminternational.org
Gear Materials, Properties, and Manufacture (#05125G)


Gears, because of their unique contribution to the operation of so many machines and mechanical
devices, have received special attention from the technical community for more than two millennia.
New developments in gear technology, particularly from the materials point-of-view, have also been
covered in detail by ASM International for many years. Numerous forums, conference proceedings,
books, and articles have been devoted to the understanding of gear performance by examining gear
tribology, failure modes, the metallurgy of ferrous gear materials, heat treatment, gear manufactur-
ing methods, and testing. All of these important technical aspects of gear technology are brought
together in the present offering, Gear Materials, Properties, and Manufacture.
Chapter 1, Basic Understanding of Gears, discusses the various types of gears used, important
gear nomenclature, and applied stresses and strength requirements associated with gears. It also pro-
vides an overview of several important topics that are covered in greater detail in subsequent chap-
ters, namely, gear materials, gear manufacture, and heat treatment. Gear tribology and lubrication is
covered in Chapter 2. Lubrication-related failures (pitting, wear, and scufng), elastohydrodynamic
lubrication, lubricant selection, and gear lubricant application are among the subjects described.
Chapters 3 and 4 describe both metallic (ferrous and nonferrous alloys) and plastic gear materials,
respectively. Emphasis in Chapter 3 has been placed on the properties of carburized steels, the mate-
rial of choice for high-performance power transmission gearing. The increasing use of plastics for
both motion-carrying and power transmission applications is covered in Chapter 4.
Chapters 5, 6, and 7 address methods for manufacturing gears including metal removal processes
(machining, grinding, and nishing), casting, forming, and forging (including recent advances in
near-net shape forging of gears), and powder metallurgy processing. Injection molding, another
important method for the manufacture of plastic gears, is covered in Chapter 4.
The heat treatment of gears is reviewed in Chapters 8 through 12. Both through hardening and sur-
face hardening methods are reviewed. Again, emphasis has been placed on carburizing, the most
common heat treatment applied to gear steels. It should be noted that some of the material presented
in these chapters was adapted, with the kind permission of the author, from Heat Treatment of Gears:
A Practical Guide for Engineers, by A.K. Rakhit (ASM International, 2000). Dr. Rakhits book is an
excellent resource for those seeking a more in-depth reference guide to gear heat treatment.
Failure analysis, fatigue life prediction, and mechanical testing are examined in Chapters 13, 14,
and 15, respectively. In Chapter 13, Gear Failure Modes and Analysis, emphasis has been placed
on two of the most common types of gear failurebending fatigue and contact fatigue. Bending
fatigue of carburized steels is also discussed in depth in Chapter 3.
In summary, this book is intended for gear metallurgists and materials specialists, manufacturing
engineers, lubrication technologists, and analysts concerned with gear failures who seek a better
understanding of gear performance and gear life. It supplements other gear texts that emphasize the
design, geometry, and theory of gears.
Joseph R. Davis
Davis & Associates
Chagrin Falls, Ohio
ASM International is the society for materials
engineers and scientists, a worldwide network
dedicated to advancing industry, technology, and
applications of metals and materials.

ASM International, Materials Park, Ohio, USA

This publication is copyright ASM International. All rights reserved.
Publication title Product code
Gear Materials, Properties, and 05125G
To order products from ASM International:
Online Visit www.asminternational.org/bookstore

Telephone 1-800-336-5152 (US) or 1-440-338-5151 (Outside US)

Fax 1-440-338-4634
Customer Service, ASM International
9639 Kinsman Rd, Materials Park, Ohio 44073, USA
Email CustomerService@asminternational.org
American Technical Publishers Ltd.
27-29 Knowl Piece, Wilbury Way, Hitchin Hertfordshire SG4 0SX,
In Europe United Kingdom
Telephone: 01462 437933 (account holders), 01462 431525 (credit card)
Neutrino Inc.
In Japan Takahashi Bldg., 44-3 Fuda 1-chome, Chofu-Shi, Tokyo 182 Japan
Telephone: 81 (0) 424 84 5550

Terms of Use. This publication is being made available in PDF format as a benefit to members and
customers of ASM International. You may download and print a copy of this publication for your
personal use only. Other use and distribution is prohibited without the express written permission of
ASM International.
No warranties, express or implied, including, without limitation, warranties of merchantability or
fitness for a particular purpose, are given in connection with this publication. Although this
information is believed to be accurate by ASM, ASM cannot guarantee that favorable results will be
obtained from the use of this publication alone. This publication is intended for use by persons having
technical skill, at their sole discretion and risk. Since the conditions of product or material use are
outside of ASM's control, ASM assumes no liability or obligation in connection with any use of this
information. As with any material, evaluation of the material under end-use conditions prior to
specification is essential. Therefore, specific testing under actual conditions is recommended.
Nothing contained in this publication shall be construed as a grant of any right of manufacture, sale,
use, or reproduction, in connection with any method, process, apparatus, product, composition, or
system, whether or not covered by letters patent, copyright, or trademark, and nothing contained in this
publication shall be construed as a defense against any alleged infringement of letters patent,
copyright, or trademark, or as a defense against liability for such infringement.
Gear Materials, Properties, and Manufacture Copyright 2005 ASM International
J.R. Davis, editor, p1-18 All rights reserved.
DOI:10.1361/gmpm2005p001 www.asminternational.org


Basic Understanding of Gears

GEARS are machine elements that transmit Aerospace gears

rotary motion and power by the successive Gears in the oil and gas industry
engagements of teeth on their periphery. They Gears for large mills that make cement,
constitute an economical method for such trans- grind iron ore, make rubber, or roll steel
mission, particularly if power levels or accuracy
Gears range in size from recently developed
requirements are high. Gears have been in use
micrometer-sized gears for electric motors no
for more than three thousand years and they are
bigger than a grain of sand to gears as large as
an important element in all manner of machin-
30 m (100 ft) in diameter. Gear materials range
ery used in current times. Application areas for
from lightweight plastics to ultrahigh-strength
gears are diverse and includeto name a few:
heat-treated steels.
Small, low-cost gears for toys
Gears for ofce equipment
Bicycle gears Gear Nomenclature
Appliance gears
Machine tool gears Before discussing the various types of gears
Control gears used, this section will review some of the terms
Automotive gears used in the gear industry to describe the design
Transportation gears of gears and gear geometries. Figure 1 shows
Marine gears schematically typical gear nomenclature. It

Fig. 1 Schematic of typical gear tooth nomenclature

2 / Gear Materials, Properties, and Manufacture

should be noted that only the most common dedendum. The depth of the tooth below the
terms are discussed below. More detailed infor- pitch circle (Fig. 1).
mation on gear nomenclature can be found in diametral pitch (DP). A measure of tooth size
various standards published by the American in the English system. In units, it is the num-
Gear Manufacturers Association (AGMA), ber of teeth per inch of pitch diameter. As the
most notably AGMA 1012-F90, Gear Nomen- tooth size increases, the diametral pitch
clature, Denitions of Terms with Symbols. decreases (Fig. 4). Diametral pitches range
active prole. The part of the gear tooth prole from 0.5 to 200. Coarse pitch gears are those
that actually comes in contact with the prole with a diametral pitch of 20. Fine pitch
of the mating gear while in mesh (Fig. 2). gears are those with a diametral pitch of >20.
addendum. The height of the tooth above the Table 1 shows the various tooth dimensions
pitch circle (Fig. 1). for different diametral pitches of spur gears.
backlash. The amount by which the width of a face width. The length of the gear teeth in an
tooth space exceeds the thickness of the axial plane (Fig. 1).
engaging tooth on the operating pitch circle llet radius. The radius of the llet curve at
(Fig. 3). the base of the gear tooth (Fig. 2).
base circle. The circle from which the involute gear. A geometric shape that has teeth uni-
tooth proles are generated (Fig. 1). formly spaced around the circumference. In
bottom land. The surface at the bottom of a general, a gear is made to mesh its teeth with
tooth space adjoining the llet (Fig. 1 and 2). another gear.
center distance. The distance between the gear blank. The workpiece used for the manu-
axes of rotation between two mating gears. facture of a gear, prior to machining the gear
circular pitch. Length of arc of the pitch circle teeth.
between corresponding points on adjacent gear pinion. When two gears mesh together,
teeth (Fig. 1). the smaller of the two is called the pinion; the
circular thickness. The length of arc be- larger is called the gear (Fig. 5).
tween the two sides of a gear tooth at the gear quality numbers. AGMA gear quality
pitch circle. numbers ranging from 3 to 15 identify the

Fig. 2 Nomenclature of gear contact areas and boundary zones

Chapter 1: Basic Understanding of Gears / 3

accuracy level of the tooth element toler- erances for the different quality numbers
ances permissible in the manufacture of a may be obtained from the AGMA standards,
particular gear in terms of its specialized use. which show the type of gear and the permis-
The higher the number, the greater the level sible tolerances and inspection dimensions.
of accuracy. Numbers 3 through 7 are for gear ratio. The ratio of the larger to the smaller
commercial applications such as appliances, number of teeth in a pair of gears.
numbers 8 through 13 are for precision appli- helix angle. The angle between any helix and
cations, and numbers 14 and 15 are for ultra- an element of its cylinder. In helical gears
precision applications. The permissible tol- and worms, it is at the standard pitch circle
unless otherwise specied.
involute gear tooth. A gear tooth whose pro-
le is established by an involute curve out-
ward from the base circle (Fig. 6).
normal section. A section through a gear that
is perpendicular to the tooth at the pitch cir-
cle (Fig. 2).
pitch circle. The circumference of a gear
measured at the point of contact with the
mating gear (Fig. 1 and 7).
pitch diameter. The diameter of the pitch cir-
Fig. 3 Schematic of gear backlash. Source: Ref 1 cle.

Fig. 4 Tooth gage chart (for reference purposes only). Source: Boston Gear, Quincy, MA
4 / Gear Materials, Properties, and Manufacture

pitch line. In a cross section of a rack, the pitch tooth is determined by the size ratio between
line corresponds to the pitch circle in the the base circle and the pitch circle (Fig. 8).
cross section of the gear (Fig. 7). Common pressure angles used by the gear
pitch point. The tangency point of the pitch industry are 14.5, 20, and 25.
circles of two mating gears (Fig. 7). rack. A rack is a gear having a pitch circle of
pitch radius. The radius of the pitch circle in a innite radius. Its teeth lie along a straight
cross section of gear teeth in any plane other line on a plane. The teeth may be at right
than a plane of rotation (Fig. 1). angles to the edge of the rack and mesh with
pressure angle. The angle between a tooth a spur gear (Fig. 5 and 9b), or the teeth on the
prole and a radial line at its pitch point (Fig. rack may be at some other angle and engage
1). The pressure angle of an involute gear a helical gear (Fig. 10b).

Table 1 Gear cutting table showing various

tooth dimensions for different diametral
pitches of spur gears
Thickness Depth to
Diametral Circular of tooth on be cut in Addendum,
pitch pitch, in. pitch line, in. gear(a), in. in.

3 1.0472 0.5236 0.7190 0.3333

4 0.7854 0.3927 0.5393 0.2500
5 0.6283 0.3142 0.4314 0.2000
6 0.5236 0.2618 0.3565 0.1667
8 0.3927 0.1963 0.2696 0.1250
10 0.3142 0.1571 0.2157 0.1000
12 0.2618 0.1309 0.1798 0.0833
16 0.1963 0.0982 0.1348 0.0625
20 0.1571 0.0785 0.1120 0.0500
24 0.1309 0.0654 0.0937 0.0417
32 0.0982 0.0491 0.0708 0.0312
48 0.0654 0.0327 0.0478 0.0208
64 0.0491 0.0245 0.0364 0.0156
(a) Hobbed gears. Source: Boston Gear, Quincy, MA
Fig. 6 Schematic of an involute gear tooth. Source: Ref 1

Line of centers


center Pitch
Pitch line

Fig. 5 The pinion, gear, and rack portions of a spur gear.

Source: Ref 1 Fig. 7 Schematic of pitch nomenclature. Source: Ref 1
Chapter 1: Basic Understanding of Gears / 5

root circle. The root circle coincides with the Types of Gears
bottoms of the tooth spaces (Fig. 1).
tooth thickness. The thickness of the tooth There is a wide variety of types of gears in
measured at the pitch circle (Fig. 1). existence, each serving a range of functions. In
top land. The surface of the top of a tooth (Fig. order to understand gearing, it is desirable to
1 and 2). classify the more important types in some way.
transverse section. A section through a gear One approach is by the relationship of the shaft
perpendicular to the axis of the gear (Fig. 2). axes on which the gears are mounted. As listed
in Table 2, shafts may be parallel, intersecting,
or nonintersecting and nonparallel.

Types of Gears that

Operate on Parallel Shafts
Spur gears (Fig. 5 and 9) are used to trans-
mit motion between parallel shafts or between a
shaft and a rack. The teeth of a spur gear are
radial, uniformly spaced around the outer
periphery, and parallel to the shaft on which the
gear is mounted. Contact between the mating
teeth of a spur gear is in a straight line parallel
to the rotational axes, lying in a plane tangent to
the pitch cylinders of the gears (a pitch cylinder
is the imaginary cylinder in a gear that rolls
without slipping on a pitch cylinder or pitch
plane of another gear).
Helical gears (Fig. 10a) are used to trans-
mit motion between parallel or crossed shafts or
between a shaft and a rack by meshing teeth that
lie along a helix at an angle to the axis of the
shaft. Because of this angle, mating of the teeth
occurs such that two or more teeth of each gear
Fig. 8 Schematic of two common pressure angles. Source: are always in contact. This condition permits
Boston Gear, Quincy MA smoother action than that of spur gears. How-

Fig. 9 Sections of a spur gear (a) and a spur rack (b) Fig. 10 Sections of a helical gear (a) and a helical rack (b)
6 / Gear Materials, Properties, and Manufacture

ever, unlike spur gears, helical gears generate gears are sometimes used in compact designs be-
axial thrust, which causes slight loss of power cause the center distance between the internal
and requires thrust bearings. gear and its mating pinion is much smaller than
Herringbone gears (Fig. 11), sometimes that required for two external gears. A typical
called double helical gears, are used to transmit relation between an internal gear and a mating
motion between parallel shafts. In herringbone pinion is shown in Fig. 12.
gears, tooth engagement is progressive, and two
or more teeth share the load at all times. Because Types of Gears that
they have right-hand and left-hand helixes, her- Operate on Intersecting Shafts
ringbone gears are usually not subject to end Bevel gears transmit rotary motion between
thrust. Herringbone gears can be operated at two nonparallel shafts. These shafts are usually
higher pitch-line velocities than spur gears. at 90 to each other.
Internal gears are used to transmit motion Straight bevel gears (Fig. 13a) have straight
between parallel shafts. Their tooth forms are teeth that, if extended inward, would intersect at
similar to those of spur and helical gears except the axis of the gear. Thus, the action between
that the teeth point inward toward the center of mating teeth resembles that of two cones rolling
the gear. Common applications for internal gears on each other (see Fig. 14 for angles and termi-
include rear drives for heavy vehicles, planetary nology). The use of straight bevel gears is gener-
gears, and speed-reducing devices. Internal ally limited to drives that operate at low speeds
and where noise is not important.
Spiral bevel gears (Fig. 13b) have teeth that
Table 2 Types of gears in common use are curved and oblique. The inclination of the
teeth results in gradual engagement and contin-
Parallel axes
uous line contact or overlapping action; that is,
Spur external
Spur internal
more than one tooth will be in contact at all
Helical external times. Because of this continuous engagement,
Helical internal the load is transmitted more smoothly from the
Intersecting axes driving to the driven gear than with straight
Straight bevel bevel gears. Spiral bevel gears also have greater
Zerol bevel load-carrying capacity than their straight coun-
Spiral bevel
Face gear terparts. Spiral bevel gears are usually preferred
Nonintersecting and nonparallel axes
to straight bevel gears when speeds are greater
than 300 m/min (1000 sfm), and particularly for
Crossed helical
Single-enveloping worm very small gears.
Double-enveloping worm Zerol bevel gears (Fig. 13c) are curved-
Hypoid tooth bevel gears with zero spiral angle. They
differ from spiral bevel gears in that the teeth

Fig. 11 A typical one-piece herringbone gear. The opposed

helixes permit multiple-tooth engagement and elim- Fig. 12 Section of a spur-type internal gear (a) and relation
inate end thrust. of internal gear with mating pinion (b)
Chapter 1: Basic Understanding of Gears / 7

are not oblique. They are used in the same way not ordinarily thought of as bevel gears, but
as spiral bevel gears, and they have somewhat functionally they are more akin to bevel gears
greater tooth strength than straight bevel gears. than to any other type.
Face gears have teeth cut on the end face of A spur pinion and a face gear are mounted
a gear, as the term face gear implies. They are (like bevel gears) on shafts that intersect and
have a shaft angle (usually 90). The pinion
bearings carry mostly radial load, while the gear
bearings have both thrust and radial load. The
mounting distance of the pinion from the pitch-
cone apex is not critical, as it is in bevel or
hypoid gears. Figure 15 shows the terminology
used with face gears.
The pinion that goes with a face gear is usu-
ally made spur, but it can be made helical if nec-
essary. The formulas for determining the
dimensions of a pinion to run with a face gear
are no different from those for the dimensions of
a pinion to run with a mating gear on parallel
axes. The pressure angles and pitches used are
similar to spur gear (or helical gear) practice.
The gear must be nished with a shaper-
cutter that is almost the same size as the pinion.
Equipment for grinding face gears is not avail-
able. The teeth can be lapped, and they can be
shaved without too much difculty, although
ordinarily shaving is not used.
The face gear tooth changes shape from one
end of the tooth to the other. The face width of
the gear is limited at the outside end by the
radius at which the tooth becomes pointed. At
the inside end, the limit is the radius at which

Fig. 13 Three types of bevel gears and a hypoid gear

Fig. 15 Face gear terminology. (a) Cross-sectional view

showing gear and pinion positions. (b) Relationship
Fig. 14 Angles and terminology for straight bevel gears of gear teeth to gear axis
8 / Gear Materials, Properties, and Manufacture

undercut becomes excessive. Due to practical worm gear set, both members are throated,
considerations, it is usually desirable to make and both members wrap around each other. A
the face width somewhat short of these limits. double-enveloping worm gear set is shown in
The pinion to go with a face gear is usually Fig. 16. Worm gear sets are used where the ratio
made with a 20 pressure angle. of the speed of the driving member to the speed
of the driven member is large, and for a compact
Types of Gears that Operate on right-angle drive.
Nonparallel and Nonintersecting Shafts Crossed-helical gears are essentially non-
Worm gear sets are usually right-angle enveloping worm gears, that is, both members
drives consisting of a worm gear (or worm are cylindrical (Fig. 17). The action between
wheel) and a worm. A single-enveloping worm mating teeth has a wedging effect, which results
gear set has a cylindrical worm, but the gear is in sliding on tooth anks. These gears have low
throated (that is, the gear blank has a smaller load-carrying capacity, but are useful where
diameter in the center than at the ends of the shafts must rotate at an angle to each other.
cylinder, the concave shape increasing the area Hypoid gears (Fig. 13d) are similar to spi-
of contact between them) so that it tends to wrap ral bevel gears in general appearance. The
around the worm. In the double-enveloping important difference is that the pinion axis of
the hypoid pair of gears is offset somewhat from
the gear axis. This feature provides many design
advantages. In operation, hypoid gears run even
more smoothly and quietly than spiral bevel
gears and are somewhat stronger.
Spiroid gears consist of a tapered pinion
that somewhat resembles a worm (Fig. 18) and a
gear member that is a face gear with teeth curved
in a lengthwise direction; the inclination to the
tooth is like a helix angle, but not a true helical
spiral. The combination of a high gear ratio in
compact arrangements, low cost when mass pro-
duced, and good load-carrying capacity makes
these gears attractive in many applications.
Fig. 16 Mating of worm gear (worm wheel) and worm in a
double-enveloping worm gear set
Proper Gear Selection

The rst step in designing a set of gears is to

select the correct type. In many cases, the geo-
metric arrangement of the apparatus that needs
a gear drive will considerably affect the selec-

Fig. 17 Mating crossed-axes helical gears Fig. 18 Spiroid gear design

Chapter 1: Basic Understanding of Gears / 9

tion. If the gears must be on parallel axes, then (10,000 sfm). In general, though, spur gears are
spur or helical gears are appropriate. Bevel and not used much above 20 m/s (4000 sfm).
worm gears can be used if the axes are at right
angles, but they are not feasible with parallel
axes. If the axes are nonintersecting and non- Basic Applied Stresses (Ref 2)
parallel, then crossed-helical gears, hypoid
gears, worm gears, or Spiroid gears can be used. The loads applied to one tooth by the action of
Worm gears, though, are seldom used if the axes its mating tooth are at any moment of time a line
are not at right angles to each other. contact at the most; or, at the least, a point con-
There are no dogmatic rules that tell the de- tact (more detailed information on gear tooth
signer which gear to use. The choice is often contact can be found in Chapter 14, Fatigue and
made after weighing the advantages and disad- Life Prediction). As the loads are increased, the
vantages of two or three types of gears. Some line may lengthen or even broaden, or the point
generalizations, though, can be made about gear may expand to a rounded area.
selection. The basic stresses applied to a gear tooth
External helical gears are generally used include the six types listed in Fig. 19; often, a
when both high speeds and high horsepowers combination of two or three types are applied at
are involved. External helical gears have been a time. Commonly they are tensile, compres-
built to carry as much as 45,000 kW (60,000 hp) sive, shear (slide), rolling, rolling-slide, and tor-
of power on a single pinion and gear. Larger sion. Each type of gear tooth will have its own
helical gears could also be designed and built. It characteristic stress patterns.
is doubtful if any other type of gear could be Spur Gear. As the contacting tooth moves
built and used successfully to carry this much up the prole of the loaded tooth, a sliding-
power on a single mesh. rolling action takes place at the prole interface.
Bevel and Hypoid Gears. Bevel gears are At the pitchline, the stresses are pure rolling.
ordinarily used on right-angle drives when high Above the pitchline, the rolling-sliding action
efciency is needed. These gears can usually be again takes over, but the sliding will be in the
designed to operate with 98% or better ef- opposite direction. Keep in mind that the action
ciency. Hypoid gears do not have as good ef- on the prole of the contacting tooth is exactly
ciency as bevel gears, but hypoid gears can the same as the loaded tooth except in reverse
carry more power in the same space, provided
the speeds are not too high.
Worm gears are ordinarily used on right-
angle drives when very high ratios (single-
thread worm and gear) are needed. They are
also widely used in low-to-medium ratios
(multiple-thread worm and gear) as packaged
speed reducers. Single-thread worms and worm
gears are used to provide the mechanical index-
ing accuracy on many machine tools. The criti-
cal function of indexing hobbing machines and
gear shapers is nearly always done by worm
gear drive. Worm gears seldom operate at ef-
ciencies above 90%.
Spur gears are relatively simple in design
and in the machinery used to manufacture and
check them. Most designers prefer to use them
wherever design requirements permit.
Spur gears are ordinarily thought of as slow-
speed gears, while helical gears are thought of as
high-speed gears. If noise is not a serious design
problem, spur gears can be used at almost any
speed that can be handled by other types of gears.
Aircraft gas-turbine precision spur gears some- Fig. 19 Basic stresses that are applied to gear teeth. Often,
two or three are simultaneously applied to a specic
times operate at pitch-line speeds above 50 m/s area. Source: Ref 2
10 / Gear Materials, Properties, and Manufacture

order (see Fig. 20). The sliding action of two because it involves rolling surfaces. It is a shear
surfaces, when lubricated properly, will have no stress running parallel to the surface at a distance
problem. However, surface disparities, insuf- from 0.18 to 0.3 mm (0.007 to 0.012 in.) below
cient lubrication, improper surface hardness, the surface. The distance below the surface given
higher temperatures, and abrasive or adhesive above is the average depth for a normal loading
foreign particles will contribute to a breakdown condition. The actual depth of maximum shear
during a sliding contact. At the same time, there could be deeper, depending on the radius of cur-
is a tensile stress at the root radius of the loaded vature of the mating surfaces and the tangential
side of the tooth and a compressive stress at the forces being applied. In one instance there has
root radius of the opposite side. been evidence of rolling loads above the shear
Helical Gear. The helical gear tooth re- strength as deep as 0.86 mm (0.034 in.). The sub-
ceives the same contact action as the spur gear; surface shear stress is most often the originator of
i.e., a rolling-sliding action from the lowest point initial line pitting along the pitchline of gear
of active prole up to the pitchline, rolling over teeth, line pitting low on the prole due to tooth
the pitchline, then sliding-rolling from the pitch- tip interference, line pitting along the tooth tip
line over the addendum. An additional stress is due to the same tooth tip interference, and sub-
being applied to the helical tooth; a lateral slid- surface rolling contact fatigue. The subject of
ing action is applied at all contact levels, includ- rolling contact fatigue is discussed more fully in
ing the pitchline. The force component at 90 to Chapters 3, Ferrous and Nonferrous Alloys
the direction of rotation increases as the helical and 13, Gear Failure Modes and Analysis.
angle increases. Resultants of this side thrust are Straight Bevel Gear. The straight bevel
often overlooked (see Fig. 21). The web between gear undergoes the same stresses as discussed
the center shaft hub and the outer gear rim is con- above, including a very slight helical action lat-
stantly undergoing a cycle of bending stress; it is erally. The larger sliding action component is
not uncommon for a relatively thin web to fail in parallel to the axis of the gears and tends to push
bending fatigue. If the hub of the gear faces
against a thrust bearing, the bearing itself is
under a constant thrust load. The shaft carrying
the gear undergoes a continual rotation bending
stress. It is also not uncommon to have such a
shaft fail by rotational bending fatigue. The
above secondary stresses are found only in a gear
of a single-helix pattern. A double-helical gear
or a herringbone gear will not have a side thrust
component of stress; therefore, the entire stress
load will be absorbed by the teeth.
One additional stress that should be discussed
at this time is a stress common to all gearing

Fig. 21 Secondary stresses set up in associated parameters of

Fig. 20 Diagramatic stress areas on basic spur gear tooth. a helical gear due to the side thrust action of the
Source: Ref 2 helix. Source: Ref 2
Chapter 1: Basic Understanding of Gears / 11

the gears apart, causing a higher prole contact, prole, to double, when the load is applied near
and to exert a rotational bending stress in the the tooth tip. Bending strength of the root radius
web of the part as well as in the shaft. is a function of the surface hardness and the phys-
Spiral Bevel Gear. Aside from all the ical condition of the surface, such as smoothness,
stresses applied above, a spiral bevel gear has a sharpness of radius, and/or corrosive pitting.
resultant peculiar to itself. As the rolling-sliding The strength of the core materiali.e., the
stress tends to move in a straight line laterally, basic material under the carburized steel case
the progression of the points along the stress line is generally to be considered as compressive
moves in a bias across the prole of the tooth. As strength rather than tensile strength. It measures
long as at least two teeth are in contact, the the ability to withstand surface pressures that
resulting load per unit area is well within reason- may crush through the case and/or brinell (in-
able limits. However, there are circumstances dent) the surface.
(and it may be only momentary) when there is a Torsional strength of a pinion shank or of a
1-to-1 tooth contact. This very narrow line con- shaft is a bit more complex. The maximum ten-
tact may be accepting an extremely high load per sile stress is at the surface in a direction 45 from
unit area, and a line of pitting will result early in the central axis or longitudinal direction. The
the life of the tooth. Careful attention should be maximum shear stress, also at the surface, is lon-
given to the design characteristics of these parts, gitudinal (parallel to the central axis) and trans-
such as spiral angle and pressure angle. verse (90 across the central axis) (see Fig. 22).
Hypoid Gear. The hypoid gearing has the The strength at the surface is a function of sur-
same applied stresses as those discussed for the face hardness; therefore, surface-originated tor-
spiral bevel, but sliding becomes the more pre- sional tensile failures of carburized parts are rare
dominant factor. This predominance increases unless a specic type of stress raiser is present at
as the axis of the pinion is placed farther from the surface. This is not so with through hardened
the central axis of the gear, and is maximum or non-heat treated parts since the strength is uni-
when the set becomes a high-ratio hypoid. form throughout the part. Under this condition,
torsional tensile failure is expected to originate
at the surface. In most instances of torsional fail-
Strength (Ref 2) ure of carburized or induction hardened shafts
and pinion shanks, the initial fracture is along the
The strength of any component is measured by
the amount of stress that can be tolerated before
permanent strain (deformation) takes place.
Strain, or deection under load, is a constant
for steel regardless of hardness or heat treatment.
The amount of deection under load of a thin
gear web or the shank of a pinion cannot be
changed by heat treatment or by use of a stronger
material. Hookes law is the same: A change of
deection can be accomplished only by a change
of design.
Bending strength of a gear tooth is the amount
of load per unit area acceptable at the root radius
to the point of permanent deformation. Perm-
anent deformation of a carburized tooth is usu-
ally accompanied by a crack at the root radius,
whereas with a noncarburized tooth, actual bend-
ing may occur. The root radius is mentioned as
the point of deformation because it is the area of
greatest stress concentration in tension. Also,
stress (load per unit area) calculations assume
that the load is applied at the pitchline or the mid-
height of the tooth. Actually, the realistic stress at Fig. 22 Free-body diagram of maximum tensile and shear
stress orientation on a surface element of a shaft in a
the root radius varies from approximately one- torsional mode. Both maximums are at the surface. Stress is con-
half, when the load is applied low on the active sidered to be zero at the central axis. Source: Ref 2
12 / Gear Materials, Properties, and Manufacture

shear plane (i.e., longitudinal or transverse), and specic application areas for commonly used
not at the 45 angle. This means that the shear gear steels. Many high-performance gears are
strength of the subsurface material is the control- carburized. Some special-purpose steel gears are
ling factor. Shear strength is considered to be case hardened by either carbonitriding or nitrid-
only about 60% of the tensile strength. The area ing. Other special-purpose gears, such as those
most vulnerable to the origin of torsional shear used in chemical or food-processing equip-
failure of a shaft is the transition zone between ment, are made of stainless steels or nickel-base
the case and the core of either a carburized or an alloys because of their corrosion resistance, their
induction hardened part. The maximum applied ability to satisfy sanitary standards, or both.
stress often exceeds the shear strength of the Gears intended for operation at elevated temper-
material at this area and initiates a start of sub- atures may be made of tool steels or elevated-
surface failure. temperature alloys.
Most gears are made of carbon and low-alloy
steels, including carburizing steels and the lim-
Gear Materials ited number of low-alloy steels that respond
favorably to nitriding. In general, the steels
A variety of cast irons, powder-metallurgy selected for gear applications must satisfy two
materials, nonferrous alloys, and plastics are basic sets of requirements that are not always
used in gears, but steels, because of their high compatiblethose involving fabrication and
strength-to-weight ratio and relatively low cost, processing and those involving service. Fabri-
are the most widely used gear materials for cation and processing requirements include
heavy duty, power transmission applications. machinability, forgeability, and response to heat
Consequently, steel gears receive primary con- treatment as it affects fabrication and process-
sideration in this chapter. ing. Service requirements are related to the abil-
Among the through-hardening steels in wide ity of the gear to perform satisfactorily under the
use are 1040, 1060, 4140, and 4340. These steels conditions of loading for which it was designed
can also be effectively case hardened by induc- and thus encompass all mechanical-property
tion heating. Among the carburizing steels used requirements, including fatigue strength and
in gears are 1018, 1524, 4026, 4118, 4320, 4620, response to heat treatment (see the section
4820, 8620, and 9310 (AMS 6260). Table 3 lists Selection Guidelines presented below).

Table 3 Recommended steels for various applications and gear types

Typical industrial application Gear design type Typical material choice

Automotive Hypoid, spiral/straight bevel 4118, 4140, 4027, 4028, 4620, 8620, 8622, 8626
Heavy truck Hypoid, spiral/straight bevel 4817, 4820, 8625, 8822
Industrial Helical, spur rack and pinion, worm 1045, 1050, 4140, 4142, 4150, 4320, 4340, 4620
Tractor-accessory Crossed-axis helical, helical 1045, 1144, 4118, 4140
Heavy truck Crossed-axis helical, spur, worm 1020, 1117, 4140, 4145, 5140, 8620
Earth moving Spiral/straight bevel, zerol 1045, 4140, 4150, 4340, 4620, 4820, 8620, 9310
Farming Face, internal, spiral/straight bevel, spur 4118, 4320, 4817, 4820, 8620, 8822
Mining, paper/steel mill Helical, herringbone, miter, spur, spur rack and pinion 1020, 1045, 4140, 4150, 4320, 4340, 4620, 9310
Automotive Spur 1045, 1050
Automotive Helical, spur 4027, 4028, 4118, 8620
Heavy Truck Helical, spur 4027, 4028, 4620, 4817, 5120, 8620, 8622, 9310
Marine Helical, helical conical, spiral bevel 8620, 8622
Off highway Helical, internal, spiral/straight bevel, spur 1118, 5130, 5140, 5150, 8620, 8822, 9310
Tractor Herringbone, internal, spur 4118, 4140, 8822
Source: Ref 3
Chapter 1: Basic Understanding of Gears / 13

Because resistance to fatigue failure is partly surface strength are necessary to handle contact
dependent upon the cleanness of the steel and stress and wear and to prevent spalling.
upon the nature of allowable inclusions, melting Numerous factors inuence fatigue strength,
practice may also be a factor in steel selection including:
and may warrant selection of a steel produced
by vacuum melting or electroslag rening. The Hardness distribution, as a function of case
hardness, case depth, core hardness
mill form from which a steel gear is machined is
another factor that may affect its performance. Microstructure, as a function of retained au-
stenite percentage, grain size, carbides (size,
Many heavy-duty steel gears are machined from type, distribution), nonmartensitic phases
forged blanks that have been processed to pro-
vide favorable grain ow consistent with load Defect control, as a function of residual
compressive stress, surface nish, geome-
pattern rather than being machined from blanks try, intergranular toughness
cut from mill-rolled bar.
Selection Guidelines. In all gears the choice More detailed information on the factors that
of material must be made only after careful con- inuence the properties of gear steels can be
sideration of the performance demanded by the found in Chapter 3, Ferrous and Nonferrous
end-use application and total manufactured cost, Alloys.
taking into consideration such issues as machin-
ing economics. Key design considerations re-
quire an analysis of the type of applied load, Gear Manufacturing Methods
whether gradual or instantaneous, and the de-
sired mechanical properties, such as bending fa- Gears can be made by a variety of manufac-
tigue strength or wear resistance, all of which de- turing processes. This section will briey
ne core strength and heat treating requirements. review these processes. For the manufacture of
Different areas in the gear tooth prole see metal gears, the reader should consult Chapters
different service demands. Consideration must 5, Machining, Grinding, and Finishing, 6,
be given to the forces that will act on the gear Casting, Forming and Forging, and 7, Pow-
teeth, with tooth bending and contact stress, der Metallurgy. Methods for making plastic
resistance to scoring and wear, and fatigue issues gears are described in Chapter 4, Plastics.
being paramount. For example, in the root area,
good surface hardness and high residual com- Metal Removal Processes
pressive stress are desired to improve endurance, As shown in Fig. 23, gear blanks can be
or bending fatigue life. At the pitch diameter, a shaped by a number of cutting (machining) and
combination of high hardness and adequate sub- nishing processes. Often blanks are processed

Fig. 23 Examples of various gear manufacturing processes. Source: Ref 3

14 / Gear Materials, Properties, and Manufacture

by a series of rough cutting and nishing opera- face being honed with lighter force than is typi-
tions. Gears made by machining/nishing have cal of grinding.
the highest AGMA quality tolerance levels than Lapping is a polishing operation that uses
gears made by competing processes (see Table 1 abrasive pastes to nish the surfaces of gear
in Chapter 6). Machining and nishing pro- teeth. Generally a toothed, cast iron lap is rolled
cesses make up about 60% of gear manufactur- with the gear being nished.
ing costs (Fig. 24). Milling is a machining operation which
Broaching is a machining operation which removes the metal between two gear teeth by
rapidly forms a desired contour in a workpiece by passing a rotating cutting wheel across the gear
moving a cutter, called a broach, entirely past the blank.
workpiece. The broach has a long series of cut- Shaping is a gear cutting method in which
ting teeth that gradually increase in height. The the cutting tool is shaped like a pinion. The
broach can be made in many different shapes to shaper cuts while traversing across the face width
produce a variety of contours. The last few teeth and rolling with the gear blank at the same time.
of the broach are designed to nish the cut rather Shaving is a nishing operation that uses a
than to remove considerably more metal. serrated gear-shaped or rack-shaped cutter to
Broaches are often used to cut internal gear teeth, shave off small amounts of metal as the gear and
racks, and gear segments on small gears, and usu- cutter are meshed at an angle to one another.
ally are designed to cut all teeth at the same time. The crossed axes create a sliding motion which
Grinding is a process that shapes the sur- enables the shaving cutter to cut.
face by passes with a rotating abrasive wheel. Skiving is a machining operation in which
Grinding is not a practical way to remove large the cut is made with a form tool with its face so
amounts of metal, so it is used to make very angled that the cutting edge progresses from one
ne-pitch teeth, or to remove heat treat distor- end of the workpiece to the other as the tool
tion from large gears that have been cut and then feeds tangentially past the rotating workpiece.
fully hardened. Many different kinds of grind-
ing operations are used in gear manufacture. Casting, Forming, and Forging Processes
Hobbing. This is a gear cutting method that
uses a tool resembling a worm gear in appear- Casting is a process of pouring or injecting
ance, having helically spaced cutting teeth. In a molten metal into a mold so that the metal solid-
single-pitch hob, the rows of teeth advance ies and hardens into the desired shape. Casting
exactly one pitch as the hob makes one revolu- is often used to make gear blanks that will have
tion. With only one hob, it is possible to cut in- cut teeth. Small gears are frequently cast com-
terchangeable gears of a given pitch of any num- plete with teeth by the die casting process,
ber of teeth within the range of the hobbing which uses a precision mold of tool steel and
machine. low-melting-point alloys for the gears.
Honing is a low-speed nishing process Stamping is a fast inexpensive method of
used chiey to produce uniform high dimen- producing small gears from thin sheets of metal.
sional accuracy and ne nish. In honing, very The metal is sheared by a punching die which
thin layers of stock are removed by simultane- stamps through the sheet stock into a mating
ously rotating and reciprocating a bonded abra- hole.
sive stone or stick that is pressed against the sur- Gear rolling is a process which rapidly
shapes ne gear teeth or worm threads by high-
pressure rolling with a toothed die.
Powder Metallurgy (P/M) Processing. In
its most basic and widely used form, the P/M
process consists of pressing a powder to the
desired shape, followed by heating (sintering) at
an elevated temperature below its melting point.
There are a number of variations of the P/M
process that are applicable to gears (see Fig. 1 in
Chapter 7). The P/M process is suitable for high-
volume production of small gears. It is not eco-
nomical for low-to-medium volume production.
Injection molding is a method of forming a
Fig. 24 Typical gear manufacturing costs. Source: Ref 3 plastic to the desired shape by forcing the heat-
Chapter 1: Basic Understanding of Gears / 15

softened plastic into a relatively cool cavity Lasers are limited to nonreective or semi-
(die) under pressure. It is widely used for high- reective metals. Metals like aluminum and
volume production of thermoplastic resin (e.g., brass that are highly reective are difcult to
acetals and nylons) gears. Often lubricants are cut.
added to the thermoplastic material to further Lasers are limited, like stamping, to at
improve the inherent lubricity of the material. forms such as spur gears.
Forging. The forging process has long been
used to create blanks that will be subsequently Electrical Discharge Machining (EDM).
shaped into gears by metal removal methods. The EDM process uses electricity to melt or
However, it is increasingly being used for the vaporize the material being cut. Many of the
production of near-net shape and net shape attributes and limitations outlined for laser
gears for demanding applications where great machining are also applicable to EDM.
strength and durability are required. The forging Abrasive water jet machining is a hydro-
process is carried out hot with metal preheated dynamic machining process that uses a high-
to a desired temperature under intense pressure velocity stream of water laden with ne abra-
until it lls the die cavity. The resultant grain sive particles as a cutting tool. The process
ow which smoothly follows gear tooth con- typically produces burr-free edges with heat-
tours makes forged gears stronger than those affected zones, can easily handle heat treated
made by other processes. material, and, unlike lasers, can cut through
stacks of material to create multiple parts at the
Alternative or Nontraditional same time, saving time and money. Abrasive
Gear Manufacturing waterjets have been used to:

Although the gear manufacturing processes Cut lapping machine gears made from dif-
discussed above are by far the most prevalent cult-to-machine plastic composites
methods for gear production, there are a number Titanium rack and pinion components for
of other processes that are being used increas- commercial jet pilot seats
ingly by the gear industry. Process phenolics into machinery gear com-
Laser Machining. While sometimes slower ponents
than traditional machining techniques, depend- Cut spring steel into gears with tightly
ing on the material, lasers can cut complex spaced teeth
shapes such as gears with great precision and
very little material waste. This conservation Inspection
comes from the ability of the computerized-
numerically-controlled (CNC) machines con- As noted at the bottom of Fig. 23, inspection
trolling the lasers to reuse cutting paths, getting is an integral part of the gear manufacturing
as many gears from a single sheet as possible. process. As with all manufactured products,
Also, the computer control means that laser gears must be checked to determine whether the
machining is also low maintenance. The setup resulting product meets design specications
and rst runs are closely supervised, but the and requirements. Because of the irregular
actual production runs dont need any real super- shape of gears and the number of factors that
vision due to the CNC programming. must be measured, such inspection is somewhat
Limitations of the laser machining process difcult. Among the factors to be checked are
are as follows: the linear tooth dimensions (thickness, spacing,
depth, and so on), tooth prole, surface rough-
Pieces cut with a laser have heat affected ness, and noise. Several special devices, most of
zones, areas where the metal is heated them automatic or semiautomatic, are used for
beyond a critical transformation point, and this inspection.
recast. These zones are limited, however, to Gear tooth vernier calipers can be used to
the edges of the cutsminimizing, but not measure the thickness of gear teeth on the pitch
eliminating heat distortion and the need for circle. However, inspection is usually done by
further machining. Post production grinding special machines, which in one or a series of
and honing are common. operations check several factors, including
Lasers are limited to cutting metals 19 mm eccentricity, variations in circular pitch, varia-
(0.75 in.). Cutting thicker materials re- tions in pressure angle, llet interference, and
quires too much power. lack of continuous action. The gear is usually
16 / Gear Materials, Properties, and Manufacture

mounted and moved in contact with a master speeds. AGMA has incorporated the critical
gear. The movement of the latter is amplied and dimensions of pitch, concentricity, tooth prole,
recorded on moving charts, as shown in Fig. 25. tooth thickness, and tooth surface nish into a
Noise level is important in many applica- number of standards which the interested reader
tions, not only from the standpoint of noise pol- should refer to. Additional information can also
lution but also as an indicator of probable gear be found in Ref 4 and 5.
life. Therefore, special equipment for its mea-
surement is quite widely used, sometimes inte-
grated into mass-production assembly lines. Heat Treating (Ref 3)
Dimensional variations in gears result in
noise, vibration, operational problems, reduced Heat treating is one of the most important
carrying ability, and reduced life. These prob- steps in the manufacture of precision gearing.
lems are compounded at higher gear-operating Its contribution is vitally important for cost con-
trol, durability, and reliability. As shown in Fig.
24, heat treating represents about 30% of a typ-
ical gear manufacturing cost. If not properly
understood and controlled, it can have a signi-
cant impact on all aspects of the gear manufac-
turing process. This section will briey review
the following heat treating processing steps:
prehardening processes, through hardening and
case hardening processes, applied energy hard-
ening, and post-hardening processes. More
detailed information on these processes can be
found in the following chapters:

Chapter 8, Through Hardening

Chapter 9, Carburizing
Chapter 10, Nitriding
Chapter 11, Carbonitriding
Chapter 12, Induction and Flame Harden-

Prehardening Processes
Several heat treatments are normally per-
formed during the gear manufacturing process
to prepare the part for the intended manufactur-
ing steps. These are essential to the manufacture
of a quality gear.
Annealing consists of heating to and hold-
ing at a suitable temperature followed by cool-
ing at an appropriate rate, primarily intended to
soften the part and improve its machinability.
Supercritical or full annealing involves heating
a part above the upper critical temperature
(Ac3), that is the temperature at which austenite
begins to transform to ferrite during cooling,
and then slowly cooling in the furnace to around
315 C (600 F). Intercritical annealing in-
volves heating the part to a temperature above
the nal transformation temperature (Ac1), the
temperature at which austenite begins to form
during heating. Subcritical annealing heats the
Fig. 25 Typical data obtained on charts generated by auto-
part to just below the Ac1 point followed by a
mated gear-checking machines. (a) Tooth-to-tooth
pitch error. (b) Accumulated pitch error. (c) Spacing error slow cool in the furnace. The rate of softening
Chapter 1: Basic Understanding of Gears / 17

increases rapidly as the annealing temperature austenitic range, typically 815 to 875 C (1500 to
approaches the Ac1 point. 1600 F), followed by quenching and tempering.
Normalizing involves heating the part Case hardening produces a hard, wear
above the upper critical temperature and then air resistant case or surface layer on top of a ductile,
cooling outside the furnace to relieve residual shock resistant interior, or core. The idea behind
stresses in a gear blank and for dimensional sta- case hardening is to keep the core of the gear
bility. Normalizing is often considered from tooth at a level around 30 to 40 HRC to avoid
both a thermal and microstructural standpoint. tooth breakage while hardening the outer sur-
In the thermal sense, normalizing is austenitiz- face to increase pitting resistance. The higher
ing followed by cooling in still or slightly agi- the surface hardness value, the greater the pit-
tated air or nitrogen. In a microstructural sense, ting resistance. Bending strength increases for
normalizing produces a more homogenous surface hardness up to about 50 HRC, after
structure. A normalized part is very machinable which the increase in bending strength is offset
but harder than an annealed part. Normalizing by an increase in notch sensitivity.
also plays a signicant role in the control of Carburizing is the most common of the
dimensional variation during carburizing. case hardening methods. A properly carburized
Stress relieving involves heating to a tem- gear will be able to handle between 30% to 50%
perature below the lower transformation temper- more load than a through hardened gear. Car-
ature, as in tempering, holding long enough to re- burizing steels are typically alloy steels with
duce residual stress and cooling slowly enough, approximately 0.10% to 0.20% carbon. Exam-
usually in air, to minimize the development of ples of commonly carburized gear steels include
new residual stresses. Stress relief heat treating is AISI 1018, 4320, 5120, 8620, and 9310 as well
used to relieve internal stresses locked in the gear as international grades such as 20MnCr5,
as a consequence of a manufacturing step. 16MnCr5, ZF-7B, 20MoCr4, and V2525.
Carburizing can be performed in the tempera-
ture range of 800 to 1090 C (1475 to 2000 F).
Through Hardening and Common industry practice today nds the ma-
Case Hardening jority of carburizing operations taking place at
870 to 1010 C (1600 to 1850 F). Carburizing
Various heat treatment processes are designed case depths can vary over a broad range, 0.13 to
to increase gear hardness. These usually involve 8.25 mm (0.005 to 0.325 in.) being typical. How-
heating and cooling and are typically classied ever, it is common to use the carbonitriding
as through hardening, case hardening, and hard- process for case depths below 0.4 mm (0.015 in.).
ening by applied energy which will be discussed Carbonitriding is a modication of the car-
separately in a subsequent section. burizing process, not a form of nitriding. This
Through or direct hardening refers to heat modication consists of introducing ammonia
treatment methods, which do not produce a case. into the carburizing atmosphere to add nitrogen
Examples of commonly through hardened gear to the carburized case as it is being produced.
steels are AISI 1045, 4130, 4140, 4145, 4340, Examples of gear steels that are commonly car-
and 8640. It is important to note that hardness bonitrided include AISI 1018, 1117, and 12L14.
uniformity should not be assumed throughout Typically, carbonitriding is done at a lower
the gear tooth. Since the outside of a gear is temperature than carburizing, or between 700 to
cooled faster than the inside, there will be a hard- 900 C (1300 to 1650 F), and for a shorter time.
ness gradient developed. The nal hardness is Because nitrogen inhibits the diffusion of car-
dependent on the amount of carbon in the steel; bon, what generally results is a shallower case
the depth of hardness depends on the hardenabil- than is typical for carburized parts. A carboni-
ity of the steel as well as the quench severity. trided case is usually between 0.075 to 0.75 mm
Through hardening can be performed either (0.003 to 0.030 in.) deep.
before or after the gear teeth are cut. When gear Nitriding is another surface treatment pro-
teeth will be cut after the part has been hardened, cess that increases surface hardness. Since rapid
surface hardness and machinability become quenching is not required, dimensional changes
important factors especially in light of the fact are kept to a minimum, which is a major advan-
that machining will remove some or most of the tage. It is not suitable for all gear materials. One
higher hardness material at the surface. The hard- of its limitations is the extremely high surface
ness is achieved by heating the material into the hardness or white layer produced, which has a
18 / Gear Materials, Properties, and Manufacture

more brittle nature than the surface produced by Machining, Vol 16, ASM Handbook, ASM
carburizing. Despite this, nitriding has proved to International, 1989, p 330355
be a viable alternative for numerous applications.
Commonly nitrided gear steels include AISI
4140, 4150, 4340, 7140, 8640, and AMS 6475 REFERENCES
(Nitralloy N).
Nitriding is typically done in the 495 to 565 1. Gear Nomenclature, Denitions of Terms
C (925 to 1050 F) range. Three factors that are with Symbols, ANSI/AGMA 1012-F90,
extremely critical in producing superior and American Gear Manufacturerers Associa-
consistent nitrided cases and predictable dimen- tion, 1990
sional change are steel composition, prior struc- 2. L.E. Alban, Systematic Analysis of Gear
ture, and core hardness. Case depth and case Failures, American Society for Metals,
hardness properties vary not only with the dura- 1985
tion and type of nitriding being performed 3. F.J. Otto and D.H. Herring, Gear Heat
but are also inuenced by these factors. Typi- Treatment: Part I, Heat Treating Progress,
cally case depths are between 0.20 to 0.65 mm June 2002, p 5559
(0.008 to 0.025 in.) and take from 10 to 80 hours 4. D.W. Dudley, Handbook of Practical Gear
to produce. Design, McGraw Hill Book Company,
ACKNOWLEDGMENT 5. D.P. Townsend, Ed., Dudleys Gear Hand-
book: The Design, Manufacture, and Ap-
Portions of this article were adapted from T.J. plication of Gears, 2nd ed., McGraw Hill,
Krenzer and J.W. Coniglio, Gear Manufacture, Inc., 1991
Gear Materials, Properties, and Manufacture Copyright 2005 ASM International
J.R. Davis, editor, p19-38 All rights reserved.
DOI:10.1361/gmpm2005p019 www.asminternational.org


Gear Tribology and Lubrication

GEAR PERFORMANCE is profoundly plication method is as important as the choice of

inuenced by tribology, which can be dened as steel alloy and heat treatment. The interrelation-
the science and technology of interacting sur- ship of the following factors must be considered:
faces in relative motion, or, alternatively, the
science and technology of friction, lubrication, Gear tooth geometry
and wear. This chapter reviews current knowl- Gear tooth motion (kinematics)
edge of the eld of gear tribology and is Gear tooth forces (static and dynamic)
intended for both gear designers and gear oper- Gear tooth material and surface characteris-
ators. Gear tooth failure modes are discussed tics (physical and chemical)
with emphasis on lubrication-related failures. Lubricant characteristics (physical and
Equations for calculating lubricant lm thick- chemical)
ness which determines whether the gears oper- Environmental characteristics (physical and
ate in the boundary, elastohydrodynamic, or chemical)
full-lm lubrication range are given. Also given
is an equation for Bloks ash temperature, Lubrication Regimes
which is used for predicting the risk of scufng.
In addition, recommendations for lubricant Throughout this chapter a number of different
selection, viscosity, and method of application types of lubrication regimes are discussed.
are discussed. Finally, a case history demon- Lubrication regimes are the ranges of operating
strates how the tribological principles discussed conditions for lubricated tribological systems
in the chapter can be applied in a practical man- that can be distinguished by their frictional char-
ner to avoid gear failure. Table 1 lists the acteristics and/or by the manner and amount of
nomenclature used in this chapter and the units separation of the contacting surfaces (Fig. 1). A
required to calculate Eq 1 to 24. brief review of the various types of lubrication
regimes is discussed below. More detailed infor-
mation on this subject can be found in the article
Gear Tribology Lubrication Regimes in Friction, Lubrication,
and Wear Technology, Vol 18, of the ASM Hand-
Because gears are such common machine book (see pages 89 to 97).
components, they may be taken for granted. It is Terminology. When two surfaces are in
not generally appreciated that they are complex contact with each other, the load is carried by
systems requiring knowledge from all the engi- many high points, or asperities, on the surfaces.
neering disciplines for their successful design. During sliding, the total tangential force
Gear design is a process of synthesis in which required to shear these asperity junctions is usu-
gear geometry, materials, heat treatment, manu- ally high, causing unacceptable friction, wear,
facturing methods, and lubrication are selected to and surface damage. To reduce the frictional
meet the requirements of a given application. force and thus allow easier sliding, a lubricant is
The designer must design the gearset with ade- deliberately introduced to separate the asperities
quate strength, wear resistance, and scufng either totally or partially (Fig. 1).
resistance. To do this, he or she must consider The use of liquid or gas lubricants is known as
gear tribology. The choice of lubricant and its ap- uid-lm lubrication. Full-lm lubrication, also
20 / Gear Materials, Properties, and Manufacture

known as thick-lm lubrication, refers to the as boundary lubrication. The exclusive use of
total separation of asperities by a lubricant lm solid lubricants is called solid lubrication.
thickness many times larger than the size of the For highly loaded contacts, lubricant pressure
lubricant molecules. If this condition exists only can cause elastic deformation of the surfaces
partiallythat is, if part of the load is carried by that is on the same order as the lubricant lm
the uid pressure and the rest is borne by con- thickness. When this occurs, the inuence of de-
tacting asperities separated by a molecularly thin formation on lubrication performance becomes
lubricant lmthe term thin-lm lubrication is a signicant parameter. Contacts operating
used. In the most severe form of thin-lm lubri- under this condition are in the regime of elasto-
cation, the entire load is carried by asperities hydrodynamic lubrication.
lubricated by surface lms of molecularly thin Lubricated contacts such as gear teeth and
liquids, gases, or solids; this condition is known rolling bearings have surface deformation com-

Table 1 Nomenclature and units of measure used in this chapter to discuss friction, lubrication,
and wear of gears
Symbol Description Units Symbol Description Units

BM Thermal contact coefcient lbf/(in. s0.5 F) WNr Normal operating load lbf
bH Semiwidth of Hertzian contact band in. wNr Normal unit load lbf/in.
c Constant (see Table 5) hp/(gal/min) Xw Welding factor ...
cM Specic heat per unit mass lbf in./[lb F] X
Load sharing factor ...
d Operating pitch diameter of pinion in. Pressure-viscosity coefcient in.2/lbf
E1 Modulus of elasticity of pinion lbf/in.2 Specic lm thickness ...
E2 Modulus of elasticity of gear lbf/in.2 min Minimum specic lm thickness ...
Er Reduced modulus of elasticity lbf/in.2 M Heat conductivity lbf/[s F]
hmin Minimum lm thickness in. m Mean coefcient of friction ...
Lmin Minimum contact length in. 0 Absolute viscosity reyns (lbf s/in.2)
n Pinion speed rev/min 1 Poissons ratio of pinion ...
P Transmitted power hp 2 Poissons ratio of gear ...
q Oil ow rate gal/min 40 Kinematic viscosity at 40 C (105 F) cSt
S Average root-mean-square (rms) min. 1 Transverse radius of curvature of pinion in.
surface roughness 2 Transverse radius of curvature of gear in.
Tb Bulk temperature F M Density lb/in.3
Tbtest Bulk temperature of test gears F n Normal relative radius of curvature in.
Tc Contact temperature F Root-mean-square composite in.
Tf Flash temperature F surface roughness
Tftest Maximum ash temperature F 1 Root-mean-square surface roughness in.
of test gears of pinion
Ts Scufng temperature F 2 Root-mean-square surface roughness in.
V Operating pitch line velocity ft/min of gear
Ve Entraining velocity in./s b Base helix angle degrees
Vr1 Rolling velocity of pinion in./s 1 Angular velocity of pinion rad/s
Vr2 Rolling velocity of gear in./s 2 Angular velocity of gear rad/s rad/s

Fig. 1 Schematic showing three lubrication regimes

Chapter 2: Gear Tribology and Lubrication / 21

parable to or exceeding the lubricant lm thick- while the term scufng is used in Europe to
ness. Therefore, elastohydrodynamic (EHD) describe the severe form of adhesive wear that
lubrication is extremely important in determin- involves the welding and tearing of the surfaces
ing friction and wear in many mechanical com- of gear teeth. To agree with current usage, the
ponents and will be emphasized throughout this term scufng will be used in this chapter when
chapter (see, for example, the section Elasto- referring to this failure mode. The term scoring
hydrodynamic Lubrication). implies scratching, and it will be used to describe
abrasive wear rather than scufng.

Classication of Lubrication-Related Failure

Gear Tooth Failure Modes
This chapter is concerned with gear tooth
To obtain optimum minimum-weight gear- failures that are inuenced by friction, lubrica-
sets, the gear designer must be aware of the intri- tion, and wear. Pitting or scufng may cause the
cate details of many competing modes of failure. gear teeth to deteriorate and generate dynamic
Erichello (Ref 1) has classied gear failures into forces, which in turn cause the gear teeth to fail
two broad groups: by bending fatigue. In these cases, the bending
failure is secondary and not directly related to
Nonlubrication-related failure
lubrication, whereas pitting or scufng are the
Lubrication-related failure
primary failure modes, and both are denitely
As shown in Table 2, nonlubrication-related inuenced by lubrication. The failure analyst
failures include both overload and bending must discern the difference between primary
fatigue types of failure. Lubrication-related fail- and secondary failure modes because the wrong
ures, the focal point in this chapter, include corrective action is likely to be recommended if
Hertzian fatigue, wear, and scufng. Failure a secondary failure mode is mistaken for the pri-
modes are further discussed in Chapter 13, mary failure mode. For example, increasing the
Gear Failure Modes and Analysis, and in Ref size of the gear teeth to prevent reoccurrence of
2 through 6. the above-mentioned bending failure would
Many gear failures are known by several only make the situation worse by lowering the
names and qualifying terms, such as initial, mod- pitting and scufng resistance. Godfrey (Ref 7)
erate, destructive, and so on. The term scoring gives a good description of lubrication-related
has been used in the past in the United States, failure modes.

Table 2 Basic failure modes of gear teeth

Nonlubrication-related failures Lubrication-related failures
Overload Bending fatigue Hertzian fatigue Wear Scufng

Brittle fracture Low-cycle fatigue (1000 Pitting Adhesion Scoring

Ductile fracture cycles to failure) Initial Normal Galling
Plastic deformation High-cycle fatigue (>1000 Supercial Running-in Seizing
Cold ow cycles to failure) Destructive Mild Welding
Hot ow Spalling Moderate Smearing
Indentation Micropitting Severe Initial
Rolling Frosting Excessive Moderate
Bruising Gray staining Abrasion Destructive
Peening Peeling Scoring
Brinelling Subcase fatigue (case Scratching
Rippling (sh scaling) crushing) Cutting
Ridging Plowing Plowing
Bending (yielding) Gouging
Tip-to-root interference Corrosion
Fretting corrosion
Electrical discharge damage
Polishing (burnishing)
22 / Gear Materials, Properties, and Manufacture

On the basis of the above considerations, over- surfaces with high strength. Maximum pitting
load and bending fatigue are judged to be unre- resistance is obtained with carburized gear teeth
lated to lubrication and are eliminated from fur- because they have hard surfaces, and carburiz-
ther discussion together with another category, ing induces benecial compressive residual
subcase Hertzian fatigue. Although corrosion, stresses that effectively lower the load stresses.
fretting-corrosion, cavitation, and electrical dis- The drawbacks to using carburized gear teeth
charge damage are inuenced by lubrication, are that they are relatively expensive to produce
they will not be discussed in this chapter because and that they must be nished by grinding. The
these failure modes occur relatively rarely in gear details for obtaining high lubricant specic lm
teeth. Hence, only the following failure modes thickness will be explained later when EHD
are discussed in this chapter: lubrication is discussed, but general recommen-
dations are to use an adequate supply of cool,
Hertzian fatigue (including pitting and clean, and dry lubricant that has adequate vis-
cosity and a high pressure-viscosity coefcient.
Wear (including adhesion, abrasion, and Pitting may initiate at the surface or at a sub-
surface defect, such as a nonmetallic inclusion.
Scufng With gear teeth, pits are most often of the sur-
Hertzian Fatigue face-initiated type because the lubricant lm
Pitting and micropitting are two lubrication- thickness is usually low, resulting in relatively
related gear tooth failure modes caused by high metal-to-metal contact. The interaction
Hertzian fatigue. between asperities or contacts at defects, such
Pitting is a common failure mode for gear as nicks or furrows, creates surface-initiated,
teeth because they are subjected to high Hertzian rather than subsurface-initiated cracks. For high-
contact stresses and many stress cycles. For speed gears with smooth surface nishes, the
example, through-hardened gears are typically lm thickness is greater and subsurface-initiated
designed to withstand contact stresses of pitting, rather than surface-initiated, may pre-
approximately 700 MPa (100 ksi), while the dominate. In these cases, pitting usually starts at
contact stresses on carburized gears may reach a subsurface inclusion, which acts as a point of
2100 MPa (300 ksi). In addition, a given tooth on stress concentration. Cleaner steels, such as
a pinion that is revolving at 3600 rev/min accu- those produced by vacuum melting, prolong the
mulates over 5 106 stress cycles every 24 h. pitting life by reducing the number of inclusions.
Pitting is a fatigue phenomenon (Ref 8) that Contamination from water in the lubricant is
occurs when a fatigue crack initiates either at believed to promote pitting through hydrogen
the surface of the gear tooth or at a small depth embrittlement of the metal, and abrasive parti-
below the surface. The crack usually propagates cles in the lubricant cause pitting by indenting
for a short distance in a direction roughly paral- the tooth surfaces, causing stress concentra-
lel to the tooth surface before turning or branch- tions, and disrupting the lubricant lm. At pres-
ing to the surface. When the cracks have grown ent, the inuence of lubricant additives on pit-
to the extent that they separate a piece of the sur- ting is unresolved.
face material, a pit is formed. If several pits Prevention of Pitting. The following rec-
grow together, the resulting larger pit is often ommendations serve as guidelines for prevent-
referred to as a spall. There is no endurance ing the onset of pitting in gearsets:
limit for Hertzian fatigue, and pitting occurs Reduce contact stresses by reducing loads or
even at low stresses if the gears are operated optimizing gear geometry
long enough. Because there is no endurance Use clean steel, properly heat treated to high
limit, gear teeth must be designed for a suitable hardness, preferably by carburizing
nite lifetime. Use smooth tooth surfaces produced by
To extend the pitting life of a gearset, the careful grinding or honing
designer must keep the contact stress low and Use an adequate amount of cool, clean, and
the material strength and lubricant specic lm dry lubricant of adequate viscosity
thickness high. There are several geometric va-
riables, such as diameter, face width, number of Micropitting. On relatively soft gear tooth
teeth, pressure angle, and so on, that can be opti- surfaces; such as those of through-hardened
mized to lower the contact stress. Steels and gears, Hertzian fatigue forms large pits with
heat treatment are selected to obtain hard tooth dimensions on the order of millimeters. With
Chapter 2: Gear Tribology and Lubrication / 23

surface-hardened gears (for example, carbur- Use carburized steel with proper carbon
ized, nitrided, induction hardened, and ame content in the surface layers
hardened), pitting may occur on a much smaller
scale, typically only 10 mm (400 min.) deep. To Wear
the naked eye, the areas where micropitting has Gearsets are susceptible to wear caused by
occurred appear frosted, and frosting is a pop- adhesion, abrasion, and polishing.
ular term for micropitting. Japanese researchers Adhesion. Adhesive wear is classied as
(Ref 9) have referred to the failure mode as mild if it is conned to the oxide layers of the
gray staining because the light-scattering gear tooth surfaces. If, however, the oxide lay-
properties of micropitting give the gear teeth a ers are disrupted and bare metal is exposed, the
gray appearance. With scanning electron transition to severe adhesive wear usually
microscopy (SEM), it is immediately evident occurs. Severe adhesive wear is termed scufng
that micropitting proceeds by the same fatigue and will be discussed in the section Scufng
process as classical pitting, except the pits are in this chapter. Here we assume that scufng
extremely small. has been avoided through proper design of the
In many cases, micropitting is not destructive gears, selection of the lubricant, and control of
to the gear tooth surface. It sometimes occurs the running-in process.
only in patches and may stop after the tribolog- When new gear units are rst operated, the
ical conditions have been improved by running- contact between the gear teeth is not optimum
in. The micropits may actually be removed by because of unavoidable manufacturing inaccu-
mild polishing wear during running-in, in which racies. If the tribological conditions are favor-
case the micropitting is said to heal. However, able, mild adhesive wear occurs during running-
there have been examples (Ref 911) where in and usually subsides with time, resulting in a
micropitting has escalated into full-scale pit- satisfactory lifetime for the gears. The wear that
ting, leading to the destruction of the gear teeth. occurs during running-in is benecial if it
The specic lm thickness is the most impor- smooths the tooth surfaces (thereby increasing
tant parameter that inuences micropitting. the specic lm thickness) and if it increases the
Damage seems to occur most readily on gear area of contact by removing minor imperfec-
teeth with rough surfaces, especially when they tions through local wear. To ensure that the
are lubricated with low-viscosity lubricants. wear rate remains under control, new gearsets
Gears nished with special grinding wheels to a should be run-in by being operated for at least
mirrorlike nish (Ref 12) have effectively elim- the rst ten hours at one-half load.
inated micropitting. Slow-speed gears are prone The amount of wear considered tolerable
to micropitting because their lm thickness is depends on the expected lifetime for the gears
low. and requirements for control of noise and vibra-
Prevention of Micropitting. To prevent tion. Wear is considered excessive when the
micropitting, the specic lm thickness should tooth proles wear to the extent that high
be maximized by using smooth gear tooth sur- dynamic loads occur or the tooth thickness is
faces, high-viscosity lubricants, and high reduced to the extent that bending fatigue
speeds. Experiments (Ref 10) have shown that becomes possible.
ame-hardened and induction-hardened gears Many gears, because of practical limits on
have less resistance to micropitting than car- lubricant viscosity, speed, and temperature,
burized gears of the same hardness. This is must operate under boundary-lubricated condi-
probably due to the lower carbon content of the tions in which some wear is inevitable. Highly
surface layers of the ame-hardened and loaded, slow speed (<30 m/min, or 100 ft/min),
induction-hardened gears. boundary-lubricated gears are especially prone
The following recommendations serve as to excessive wear. Tests with slow-speed gears
guidelines for preventing the onset of micropit- (Ref 10) have shown that nitrided gears have
ting in gearsets: good wear resistance, whereas carburized and
Use smooth tooth surfaces produced by through-hardened gears have similar but lower
careful grinding or honing wear resistance. Winter and Weiss (Ref 10) con-
Use an adequate amount of cool, clean, and cluded that lubricant viscosity has the greatest
dry lubricant of the highest viscosity per- effect on slow-speed adhesive wear, and that
missible high-viscosity lubricants reduce the wear rate
Use high speeds if possible signicantly. Winter and Weiss (Ref 10) also
24 / Gear Materials, Properties, and Manufacture

found that sulfur-phosphorus additives can be mended lubricant, and a new oil lter should be
detrimental with slow-speed (<3 m/min, or 10 installed.
ft/min) gears, giving very high wear rates. Internally generated particles are usually
A few gear units operate under ideal condi- wear debris from gears or bearings due to Hertz-
tions with smooth tooth surfaces, high pitch line ian fatigue pitting or adhesive and abrasive
speed, and high lubricant lm thickness. For wear. The wear particles are especially abrasive
example, turbine gears that operated almost because they become work-hardened when they
continuously at 9000 m/min (30,000 ft/min) are trapped between the gear teeth. Internally
pitch line speed still had the original machining generated wear debris can be minimized by
marks on their teeth, even after operating for 20 using accurate surface-hardened gear teeth
years. Most gears, however, operate between (with high pitting resistance), smooth surfaces,
the boundary and full-lm lubrication regimes, and high-viscosity lubricants.
under EHD conditions. In the EHD regime, with Breather vents are used on gearboxes to vent
the proper type and viscosity of lubricant, the internal pressure, which may occur when air
wear rate usually reduces during running-in and enters through seals, or when air within the
adhesive wear virtually ceases once running-in gearbox expands (or contracts) during the nor-
is completed. If the lubricant is properly main- mal heating and cooling of the gear unit. The
tained (cool, clean, and dry), the gearset should breather vent should be located in a clean, non-
not suffer an adhesive wear failure. pressurized area and should have a lter to pre-
Prevention of Adhesive Wear. The fol- vent ingress of airborne contaminants. In espe-
lowing recommendations serve as guidelines cially harsh environments, the gearbox can be
for preventing the onset of adhesive wear in completely sealed, and the pressure variation
gearsets: can be accommodated by an expansion chamber
with a exible diaphragm.
Use smooth tooth surfaces All maintenance procedures that involve
If possible, run-in new gearsets by operating opening any part of the gearbox or lubrication
the rst 10 hours at one-half load system must be carefully performed to prevent
Use high speeds if possible. Otherwise, rec- contamination of the gearbox system.
ognize that highly loaded slow-speed gears Abrasive wear due to foreign contaminants,
are boundary lubricated and are especially such as sand or internally generated wear debris,
prone to excessive wear. For these condi- is called three-body abrasion and is a common
tions, specify nitrided gears and the highest occurrence. Two-body abrasion also occurs
permissible lubricant viscosity when hard particles or asperities on one gear
For very slow-speed gears (<3 m/min, or 10 tooth abrade the opposing tooth surface. Unless
ft/min), avoid using lubricants with sulfur- the tooth surfaces of a surface-hardened gear are
phosphorus additives smoothly nished, they will act like les if the
Use an adequate amount of cool, clean, and mating gear is appreciably softer. This is the
dry lubricant of the highest viscosity per- reason that a worm pinion is polished after
missible grinding before it is run with a bronze worm
wheel. Manufacturers of computer disk drives
Abrasion. Abrasive wear on gear teeth is have found that stainless steel pinions mated
usually caused by contamination of the lubri- with anodized aluminum racks have excessively
cant by hard, sharp-edged particles. Contamina- high wear rates. The anodized layer of the alu-
tion enters gearboxes by being built-in, inter- minum rack is extremely thin and brittle, and it
nally generated, ingested through breathers and breaks up and impregnates the relatively soft
seals, or inadvertently added during mainte- stainless steel pinion. The aluminum oxide par-
nance. ticles then act like emery paper and wear the
Many gear manufacturers do not fully appre- teeth of the rack very quickly.
ciate the signicance of clean assembly; it is not The lubrication system should be carefully
uncommon to nd sand, machining chips, maintained and monitored to ensure that the
grinding dust, weld splatter, or other debris in gears receive an adequate amount of cool, clean,
new gearboxes. To remove built-in contamina- and dry lubricant. For circulating-oil systems,
tion, the gearbox lubricant should be drained ne ltration removes contamination. Filters as
and ushed before start-up and again after the ne as 3 mm (120 min.) have signicantly
rst 50 h of operation, relled with the recom- increased gear life. For oil bath gearboxes, the
Chapter 2: Gear Tribology and Lubrication / 25

lubricant should be changed frequently to tures where there is a danger of welding. If the
remove contamination. Under normal operating rate of reaction is too high, and there is a con-
conditions, the lubricant should be changed at tinuous removal of the surface lms caused by
least every 2500 h of operation or every six very ne abrasives in the lubricant, the polish-
months, whichever occurs rst. For critical ing wear may be excessive (Ref 13).
gearboxes, a regular program of lubricant mon- Prevention of Polishing Wear. Polishing
itoring can help prevent gear failures by show- wear can be prevented by using less chemically
ing when maintenance is required. The lubricant active additives. As an alternative to sulfur-
monitoring should include spectrographic and phosphorus additives, antiscuff lubricants are
ferrographic analysis of contamination, along available with dispersions of potassium borate
with analysis of acid number, viscosity, and (Ref 14) that deposit EP lms without chemi-
water content. cally reacting with the metal. Removing the
Prevention of Abrasive Wear. The fol- abrasives in the lubricant by using ne ltration
lowing guidelines should be observed to pre- or frequent oil changes is helpful.
vent abrasive wear in gearsets: In summary, the following guidelines should
be observed to prevent polishing wear in
Remove built-in contamination from new
gearboxes by draining and ushing the
lubricant before start-up and again after the Use less chemically active antiscuff addi-
rst 50 h of operation. Rell with the rec- tives (for example, borate)
ommended lubricant and install a new lter Remove abrasives from the lubricant by
Minimize internally generated wear debris using ne ltration or frequent oil changes
by using surface-hardened gear teeth,
smooth tooth surfaces, and high-viscosity Scufng
Scufng is dened as localized damage
Minimize ingested contamination by main-
caused by solid-phase welding between sliding
taining oil-tight seals and using ltered
surfaces. It is accompanied by the transfer of
breather vents located in clean, nonpressur-
metal from one surface to another due to welding
ized areas
and tearing. It may occur in any sliding and
Minimize contamination that is added dur-
rolling contact where the oil lm is not thick
ing maintenance by using good housekeep-
enough to separate the surfaces. The symptoms
ing procedures
of scufng are microscopically rough, matte,
For circulating-oil systems, use ne ltra-
and torn surfaces. Surface analysis that shows
transfer of metal from one surface to the other is
For oil bath systems, change the lubricant at
proof of scufng.
least every 2500 h or every six months
Scufng can occur in gear teeth when they
Monitor the lubricant with spectrographic
operate in the boundary lubrication regime. If
and ferrographic analysis together with
the lubricant lm is insufcient to prevent sig-
analysis of acid number, viscosity, and
nicant metal-to-metal contact, the oxide layers
water content
that normally protect the gear tooth surfaces
Polishing Wear. If the extreme-pressure may be broken through, and the bare metal sur-
(EP) antiscuff additives in the lubricant are too faces may weld together. The sliding that occurs
chemically reactive, they may cause polishing between gear teeth results in tearing of the
of the gear tooth surfaces until they attain a welded junctions, metal transfer, and cata-
bright mirror nish. Although the polished gear strophic damage.
teeth may look good, polishing wear is undesir- In contrast to pitting and bending fatigue,
able because it generally reduces gear accuracy which only occur after a period of running time,
by wearing the tooth proles away from their scufng may occur immediately upon start-up.
ideal form. Antiscuff additives used in lubri- In fact, gears are most vulnerable to scufng
cants to prevent scufng, such as sulfur and when they are new and their tooth surfaces have
phosphorus, will be covered when scufng is not yet been smoothed by running-in. For this
discussed. They function by forming iron-sul- reason, it is wise to run-in a new gearbox under
de and iron-phosphate lms on areas of the one-half load for at least 10 h to reduce the sur-
gear teeth where high temperatures occur. Ide- face roughness of the teeth before applying full
ally, the additives should react only at tempera- load. The gear teeth can be coated with iron-
26 / Gear Materials, Properties, and Manufacture

manganese phosphate or plated with copper or viscosity increase also helps to reduce the risk
silver to protect them from scufng during the of scufng by increasing the lubricant lm
critical running-in period. thickness and reducing the contact temperature
The basic mechanism of scufng is not generated by metal-to-metal contact.
clearly understood, but by general agreement it Scufng is controlled by the total contact
is believed to be caused by intense frictional temperature, Tc, which consists of the sum of
heating generated by the combination of high the gear bulk temperature, Tb, and the ash tem-
sliding velocity and intense surface pressure. perature, Tf:
Bloks (Ref 15) critical temperature theory is
Tc Tb Tf (Eq 1)
believed to be the best criterion for predicting
scufng. It states that scufng will occur in gear The bulk temperature is the equilibrium tem-
teeth that are sliding under boundary-lubricated perature of the surface of the gear teeth before
conditions when the maximum contact temper- they enter the meshing zone. The ash tempera-
ature of the gear teeth reaches a critical magni- ture is the local and instantaneous temperature
tude. For mineral oils without antiscuff/EP rise that occurs on the gear teeth due to the fric-
additives, each combination of oil and rubbing tional heating as they pass through the meshing
materials has a critical scufng temperature that zone.
is constant, regardless of the operating condi- Anything that reduces either the bulk tempera-
tions (Ref 16). The critical scufng tempera- ture or the ash temperature will reduce the total
tures are not constant for synthetic lubricants contact temperature and lessen the risk of scuf-
and lubricants with antiscuff additives; they ng. Higher viscosity lubricants or smoother
must be determined from tests that closely sim- tooth surfaces help by increasing the specic lm
ulate the operating conditions of the gears. thickness, which in turn reduces the frictional
Today, most antiscuff additives are sulfur- heat and, therefore, the ash temperature. Also,
phosphorus compounds, which form boundary- the lubricant performs the important function of
lubricating lms by chemically reacting with the removing heat from the gear teeth. The lubricant
metal surfaces of the gear teeth at local points of must be supplied to the gear teeth in such a way
high temperature. Antiscuff lms help prevent that it removes heat rapidly and maintains a low
scufng by forming solid lms on the gear tooth bulk temperature. A heat exchanger can be used
surfaces and inhibiting true metal-to-metal con- with a circulating-oil system to cool the lubricant
tact. The lms of iron sulde and iron phosphate before it is sprayed at the gears.
have high melting points, allowing them to Prevention of Scufng. The gear designer
remain as solids on the gear tooth surfaces even at can maximize scufng resistance by optimizing
high contact temperatures. The rate of reaction of the gear geometry so that the gear teeth are as
the antiscuff additives is greatest where the gear small as possible, consistent with bending
tooth contact temperatures are highest. Because strength requirements, to reduce the tempera-
of the rubbing action of the gear teeth, the surface ture rise caused by sliding. Figure 2 shows that
lms are repeatedly scraped off and reformed. In the rolling velocity of the pinion, Vr1, and the
effect, scufng is prevented by substituting mild rolling velocity of the gear, Vr2, linearly increase
corrosion in its place. Occasionally, antiscuff from zero at the interference points to a maxi-
additives (for example, sulfur) are too chemi- mum at each end of the path of contact. The
cally active, causing polishing wear and necessi- sliding velocity is represented by the distance
tating a change to less aggressive additives. between the Vr1 and Vr2 lines. The amount of
Lubricants with antiscuff additives of potassium sliding is proportional to the distance from the
borate do not cause polishing wear because they pitch point, P, and is zero when the gear teeth
deposit glasslike boundary lms without react- contact at the pitch point, and largest at the ends
ing with the metal. of the path. Addendum modication can be
For mineral oils without antiscuff additives, used to balance and minimize the temperature
the critical scufng temperature increases with rise that occurs in the addendum and dedendum
increasing viscosity and ranges from 150 to 300 of the gear teeth. The temperature rise may also
C (300 to 570 F). The increased scufng resis- be reduced by modifying the tooth proles with
tance of high-viscosity lubricants is believed to slight tip and/or root relief to ease the load at the
be due to differences in chemical composition start and end of the engagement path where the
rather than increases in viscosity. However, a sliding velocities are the greatest. Also, the gear
Chapter 2: Gear Tribology and Lubrication / 27

teeth must be accurate and held rigidly in good Cool the gear teeth by supplying an adequate
alignment to minimize tooth loading and, there- amount of cool lubricant. For circulating-oil
fore, the temperature rise. systems, use a heat exchanger to cool the
Gear materials should be chosen with their lubricant
scufng resistance in mind. Nitrided steels, Optimize the gear tooth geometry by using
such as Nitralloy 135M, are generally found to small teeth, addendum modication, and
have the highest resistance to scufng, whereas prole modication
stainless steels are liable to scuff even under Use accurate gear teeth, rigid gear mount-
near-zero loads. The thin oxide layer on stain- ings, and good helix alignment
less steel is hard and brittle and breaks up easily Use nitrided steels for maximum scufng
under sliding loads, exposing the bare metal and resistance. Do not use stainless steel or alu-
thus promoting scufng. Like stainless steel, minum for gears if there is a risk of scufng
anodized aluminum has a low scufng resis-
tance. Hardness does not seem to be a reliable
indication of scufng resistance. Elastohydrodynamic Lubrication
In summary, the following guidelines should
be observed to prevent scufng in gearsets: Gear teeth are subjected to enormous contact
Use smooth tooth surfaces produced by pressures on the order of the ultimate tensile
careful grinding or honing strength of hardened steel, yet they are quite suc-
Protect the gear teeth during the critical run- cessfully lubricated with oil lms that are <1 mm
ning-in period by coating them with iron- (<40 min.) thick. This is possible because lubri-
manganese phosphate or plating them with cants have a fortuitous property that causes their
copper or silver. Run-in new gearsets by viscosity to increase dramatically with increased
operating the rst 10 h at one-half load pressure. Figure 3 depicts the region of contact
Use high-viscosity lubricants with antiscuff between mating gear teeth. It shows the shape of
additives, such as sulfur, phosphorus, or the elastically deformed teeth and the pressure
borate distribution developed within the contact zone.

Fig. 3 Regions of elastohydrodynamic contact between the

mating gear teeth of a gearset. (a) Schematic showing
three distinct regions on pinion and gear tooth surfaces and key
parameters determining oil lm lubrication. (b) Plot of pressure
distribution within contact zone. bH, semiwidth of Hertzian
Fig. 2 Graphical representation of rolling velocities of pinion contact band; h0, central lm thickness; hmin, minimum lm
(Vr1) and gear (Vr2) in relation to sliding velocity thickness
28 / Gear Materials, Properties, and Manufacture

The molecular adsorption of the lubricant onto where n1 is the Poissons ratio of pinion, n2 is
the gear tooth surfaces causes it to be dragged the Poissons ratio of gear, E1 is the modulus of
into the inlet region of the contact, where its pres- elasticity of pinion, and E2 is the modulus of
sure is increased due to the convergence of the elasticity of gear.
tooth surfaces. The viscosity increase of the Figure 4 gives average values of absolute vis-
lubricant caused by the increasing pressure helps cosity, m0, versus bulk temperature for typical
to entrain the lubricant into the contact zone. mineral gear lubricants with a viscosity index
Once it is within the high-pressure Hertzian of 95.
region of the contact, the lubricant cannot escape The pressure-viscosity coefcient, a, ranges
because its viscosity has increased to the extent from 0.5 104 to 2 104 in.2/lbf for typical
where the lubricant is virtually a rigid solid. gear lubricants. Data for pressure-viscosity
The following equation, from Dowson and coefcients versus bulk temperature for typical
Higginson (Ref 17) gives the minimum lm gear lubricants are given in Fig 5.
thickness that occurs near the exit of the contact. The specic lm thickness, , is given by:
The minimum lm thickness, hmin, is obtained
from: hmin
l (Eq 8)
1.63a 0.54
1m0Ve 2 0.7
hmin (Eq 2) where hmin is the minimum lm thickness and s
1XGwNr 2 0.13 E 0.03
is the composite surface roughness given by:
where a is the pressure-viscosity coefcient
(in.2/lbf), m0 is the absolute viscosity (reyns, or s 1s21 s22 2 1>2 (Eq 9)
lbf s/in.2), and Ve is the entraining velocity, in
which: in which s1 is the root-mean-square (rms) sur-
face roughness of the pinion and s2 is the rms
Ve Vr1 Vr2 (Eq 3) surface roughness of the gear.
where Vr1 and Vr2 are the rolling velocities of Load-Sharing Factor. The load-sharing
pinion and gear, respectively, given by: factor, XG, accounts for load sharing between
succeeding pairs of teeth as affected by prole
Vr1 1 r2 (Eq 4a) modication (tip and root relief) and whether
Vr2 w1 r2 (Eq 4b)
the pinion or gear is the driver. Figure 6 gives
plots of the load-sharing factors for unmodied
in which w1 and w2 are the angular velocities of and modied tooth proles.
pinion and gear, respectively, and r1 and r2 are As shown by the exponents in Eq 2, the lm
the transverse radii of curvature of the pinion thickness is essentially determined by the
and gear, respectively. entraining velocity, lubricant viscosity, and
The normal relative radius of curvature, rn, is pressure-viscosity coefcient; the elastic prop-
given by: erties of the gear teeth and the load have rela-
r1r2 tively small inuences. In effect, the relatively
rn (Eq 5) high stiffness of the oil lm makes it insensitive
1r2 ; r1 2cos yb
to load, and an increase in load simply increases
where yb is the base helix angle. Referring back the elastic deformation of the tooth surfaces and
to Eq 2, XG is the load-sharing factor, and wNr is widens the contact area, rather than decreasing
the normal unit load given by: the lm thickness.
Bloks contact temperature theory (Ref
WNr 15) states that scufng will occur in gear teeth
wNr (Eq 6) which are sliding under boundary-lubricated
conditions when the maximum contact temper-
in which WNr is the normal operating load and ature of the gear teeth reaches a critical magni-
Lmin is the minimum contact length. tude. The contact temperature is the sum of two
Er is the reduced modulus of elasticity, given components: the bulk temperature and the ash
by: temperature, as described in Eq 1.
Bloks ash temperature equation as formu-
1 n21 1 n22 1 lated in AGMA 2001-B88, Appendix A (Ref
Er 2 a b (Eq 7)
E1 E2 18) for spur and helical gears is:
Chapter 2: Gear Tribology and Lubrication / 29

0.8mmXwNr1Vr1 2 0.5 1Vr2 2 0.5 s1 s2

Tf (Eq 10) S (Eq 12)
BM 1bH 2 0.5 2

Thermal Contact Coefcient. The ther-

where mm is the mean coefcient of friction, BM mal contact coefcient, BM, is given by:
is the thermal contact coefcient, and bH is the
semiwidth of the Hertzian contact band. BM 1lMrMcM 2 0.5 (Eq 13)
Mean Coefcient of Friction. The follow-
ing equation gives a typical value of 0.06 < mm where lM is the heat conductivity, rM is the den-
< 0.18 for the mean coefcient of friction for sity, and cM is the specic heat per unit mass.
gears operating in the partial EHD regime (l < For typical gear steels, BM is ~43 lbf/(in. s0.5
1). It may give values too low for boundary- F).
lubricated gears, where mm may be greater than Semiwidth of the Hertzian Contact
0.2, or too high for gears in the full-lm regime Band. The semiwidth of the Hertzian contact
(l > 2), where mm may be less than 0.01: band, bH, is dened as:
8XwNr rn 0.5
50 bH a b (Eq 14)
mm 0.06 a b (Eq 11a) pEr
50 S

where: Bulk Temperature. The gear bulk tempera-

ture, Tb, is the equilibrium bulk temperature of
50 the gear teeth before they enter the meshing
a b 3.0 (Eq 11b) zone. In some cases, the bulk temperature may
50 S
be signicantly higher than the temperature of
in which S is the average rms surface roughness the oil supplied to the gear mesh. In a test with
given by: ultrahigh-speed gears (Ref 19), the pinion bulk

Fig. 4 Plot of absolute viscosity versus bulk temperature for selected mineral oil gear lubricants having a viscosity index of 95
30 / Gear Materials, Properties, and Manufacture

temperature was 135 C (275 F), which is 95 For nonantiscuff mineral oils, the mean scuff-
C (171 F) hotter than the oil inlet temperature. ing temperature (50% chance of scufng) in F
For turbine gears at lower speeds, the bulk tem- is given by:
perature rise of the gear teeth over the inlet oil
temperature may range from 11 C (20 F) at Ts 146 59 ln n40 (Eq 15)
3700 m/min (1.2 104 ft/min) pitch line veloc- where n40 is the kinematic viscosity at 40 C
ity to 22 C (40 F) at 4900 m/min (1.6 104 (105 F) in centistokes.
ft/min). At similar speeds, the bulk temperature For mineral oils with low concentrations of
rise of aircraft gears with less oil ow may antiscuff additives, the mean scufng tempera-
range from 22 to 33 C (40 to 60 F). ture in F is given by:
Scufng Temperature. The scufng tem-
perature, Ts, is the contact temperature at which Ts 245 59 ln n40 (Eq 16)
scufng is likely to occur with the chosen com- The scufng temperature determined from
bination of lubricant and gear materials. Forschungsstelle fr Zahnrder und Getriebe-
For mineral oils without antiscuff additives bau (FZG, or Technical Institute for the Study
or for mineral oils with low concentrations of of Gears and Drive Mechanisms) test gears for
antiscuff additives, the scufng temperature is mineral oils without antiscuff additives or with
independent of the operating conditions for a low concentrations of antiscuff additives may
fairly wide range. For these oils, the scufng be extended to different gear steels, heat treat-
temperature may be correlated with the compo- ments, or surface treatments by introducing an
sition of the oil. The viscosity grade is a con- empirical welding factor:
venient index of the composition and, thus, of
the scufng temperature. Ts Tbtest XwTftest (Eq 17)

Fig. 5 Plot of pressure-viscosity coefcient versus bulk temperature for selected mineral oil gear lubricants
Chapter 2: Gear Tribology and Lubrication / 31

where Xw is the welding factor (see Table 3), attention has to be paid to the correlation between
Tbtest is the bulk temperature of the test gears, test conditions and actual or design conditions.
and Tftest is the maximum ash temperature of
the test gears.
Scufng temperatures for synthetic lubri-
Lubricant Selection
cants typically used with carburized gears in the
aerospace industry are shown in Table 4.
The choice of lubricant depends on the type
For mineral oils with high concentrations of
of gearing and enclosure, operating speed and
antiscuff additives (for example, hypoid gear
load, ambient temperature, and method of lubri-
oils), research is still needed to determine
cant application. Most gears are lubricated with
whether the scufng temperature is dependent on
one of the following types:
the materials and operating conditions. Special
Synthetic lubricant
Adhesive open-gear lubricant
Solid lubricant
The optimum lubricant for any application is the
product that is the least expensive, considering
both initial cost and maintenance costs, and
meets the requirements.
Oil is the most widely used lubricant
because it is readily distributed to gears and
bearings and has both good lubricating and
cooling properties. In addition, contamination
may be readily removed by ltering periodi-
cally or draining and replacing the oil. How-
ever, it requires an oil-tight enclosure provided
with adequate shaft seals.
Synthetic lubricants are used for applica-
tions (for example, aircraft gas turbines) where
the oil must operate over a wide temperature
range and have good oxidation stability at high
temperature. Ester and hydrocarbon synthetic
lubricants have high viscosity indices, giving
them good uidity or low viscosities at very low
temperatures and acceptable viscosities at high
temperatures. The volatility of esters is lower
than that of mineral oils of the same viscosity,
Fig. 6 Plot of load-sharing factor (XG) versus pinion roll angle.
(a) Unmodied tooth proles. (b) Modied tooth
proles Table 4 Mean scufng temperature for
synthetic lubricants typically used for operating
carburized gears in aerospace applications
Sub in Table 3 Welding factor for selected gear steels
table head Mean scufng temperature, Ts
is already Material Welding factor, XW
Lubricant C F
not italic
Through-hardened steel 1.00
Phosphated steel 1.25 MIL-L-6081 (grade 1005) 129 264
Copper-plated steel 1.50 MIL-L-7808 205 400
Nitrided steel 1.50 MIL-L-23699 220 425
Carburized steel DERD2487 225 440
Content of austenite < average 1.15 DERD2497 240 465
Content of austenite = average 1.00 DOD-L-85734 260 500
Content of austenite > average 0.85 Mobil SHC624 280 540
Stainless steel 0.45 Dexron II 290 550
Source: Ref 18 Source: Ref 20
32 / Gear Materials, Properties, and Manufacture

thus reducing oil loss at high temperature. are also used. Solid lubricants are expensive to
Despite their long service life, the extra cost of apply and have limited wear lives. However, in
synthetic lubricants generally cannot be justied many applications, such as spacecraft, they are
for oil-bath systems unless there are extreme the only alternative and can provide excellent
temperatures involved, because the oil must be service.
changed frequently to remove contamination.
Grease is suitable only for low-speed, low-
load applications because it does not circulate Oil Lubricant Applications
well, and it is a relatively poor coolant. Grease-
lubricated gears are generally boundary lubri- Of the above-mentioned lubricants, only oil
cated because the grease is either pushed aside will be discussed in greater detail in this chap-
or thrown from the gear teeth. Contamination ter. Oil should be used as the lubricant unless
from wear particles or other debris is usually the operating conditions preclude its use. Gen-
trapped in the grease and requires costly main- erally, the simplest and least expensive lubrica-
tenance to eliminate. Grease is often used to tion system for gears is a totally enclosed oil
avoid leakage from enclosures that are not oil bath of mineral oil.
tight. However, if all the factors are considered, Spur, Helical, and Bevel Gears. The lubri-
it is usually found that an oil lubricant is more cation requirements of spur, helical, straight-
economical and reliable than a grease lubricant bevel, and spiral-bevel gears are essentially the
for gear lubrication. same. For this class of gears, the magnitudes of
Open-gear lubricants are viscous adhesive the loads and sliding speeds are similar, and
semiuids used on large low-speed open gears, requirements for viscosity and antiscuff proper-
such as those used in iron ore and cement mills, ties are virtually identical. Many industrial spur
antenna drives, bridge drives, cranes, and so on. and helical gear units are lubricated with rust and
Gears in these applications run slowly, and they oxidation inhibited (R & O) mineral oils. The
are therefore boundary lubricated. The lubricant low-viscosity R & O oils, commonly called tur-
must bond strongly to resist being thrown off the bine oils, are used in many high-speed gear units
gear teeth. However, the squeezing and sliding where the gear tooth loads are relatively low.
action of gear teeth tends to push the lubricant Mineral oils without antiscuff additives are suit-
into the roots of the gear teeth where it is rela- able for high-speed lightly loaded gears where
tively ineffective. These lubricants are applied the high entraining velocity of the gear teeth
by hand brushing or by automatic systems that develops thick EHD oil lms. In these cases, the
deliver an intermittent spray. Some open-gear most important property of the lubricant is vis-
lubricants are thinned with a quick-evaporating cosity. Antiscuff/EP additives are unnecessary
solvent/diluent to make them easier to apply. because the gear teeth are separated, eliminating
Open-gear lubricants share the disadvantages of metal-to-metal contact and the scufng mode of
grease lubrication, and they are especially costly failure. Slower speed gears, especially carbur-
(in addition to being messy) to maintain. For ized gears, tend to be more heavily loaded. These
these reasons, the trend is away from open gears gears generally require higher viscosity lubri-
and toward the use of enclosed oil-lubricated cants with antiscuff additives.
gearboxes whenever possible. Hypoid gears, such as those used for automo-
Solid lubricants, usually in the form of tive axles, are especially prone to scufng
bonded dry lms, are used when the following because they are heavily loaded and have high
conditions are encountered: sliding velocities. For these reasons, hypoid
gear oils have the higher concentrations of anti-
The temperature is too high or too low for an scuff additives.
oil or grease For critical applications, the contact tempera-
Leakage cannot be tolerated ture should be calculated with Bloks equation
The gears must operate in a vacuum (see Eq 1) (Ref 15) and compared to the scuff-
ing temperature of the lubricant. This quantita-
These lubricants are usually molybdenum disul- tive method is effective for selecting a lubricant
de (MoS2) or graphite in an inorganic binder, with adequate scufng resistance.
which is applied to the gear teeth and cured to Worm gears have high sliding velocity,
form a dry lm coating. Polytetrauoroethylene which generates signicant frictional losses.
(PTFE) and tungsten disulde (WS2) coatings Fortunately, their tooth loads are relatively
Chapter 2: Gear Tribology and Lubrication / 33

light, and they are successfully lubricated with where 40 is the lubricant kinematic viscosity at
mineral oils that are compounded with lubricity 40 C (105 F) (in cSt) and V is the operating
additives. These oils contain 3 to 10% fatty pitch line velocity (in ft/min) given by:
oil or low-acid tallow. The polar molecules of
the additive form surface lms by physical V 0.262 d # n (Eq 19)
adsorption or by reaction with the surface ox-
where d is the operating pitch diameter of pinion
ide to form a metallic soap that acts as a low
(in inches) and n is the pinion speed (rev/min).
shear strength lm, improving the lubricity,
Caution must be used when using AGMA
or friction-reducing property.
recommendations for viscosity. For example,
there was an application where two gear drives
were considered to be high speed. The pinion
Selection of Gear Lubricant Viscosity
speed was 3625 rev/min, qualifying the gear
units as high-speed gear drives per AGMA
The recommendations of AGMA 250.04
421.06. The gear drives were supplied with oil
(Ref 21) should be followed when selecting
having the recommended viscosity (per AGMA
lubricants for enclosed gear drives that operate
421.06) of ISO 68. However, because the pinion
at pitch line velocities 1500 m/min (5000
was relatively small, its pitch line velocity was
ft/min). AGMA 421.06 (Ref 22) should be con-
only 1000 m/min (3000 ft/min). This qualies
sulted for high-speed drives (>1500 m/min, or
the gear drives as slow speed per AGMA
5000 ft/min).
250.04, which recommends a viscosity of ISO
Viscosity is one of the most important lubri-
150. Both gear drives failed within weeks of
cant properties, and the higher the viscosity, the
start-up because of pitting fatigue. The empiri-
greater the protection against the various gear
cal equation (Eq 18) for this application gives:
tooth failures. However, the viscosity must be
limited to avoid excessive heat generation and 7000
power loss from churning and shearing of the n40 128 cSt (Eq 20)
13000 2 0.5
lubricant by high-speed gears or bearings. The
operating temperature of the gear drive deter-
This indicates that the viscosity per AGMA
mines the operating viscosity of the lubricant. If
421.06 (68 cSt, or 6.8 105 m2/s) is much too
the lubricant is too viscous, excessive heat is
low, and the viscosity per AGMA 250.04 (150
generated. The heat raises the lubricant temper-
cSt, or 1.5 104 m2/s) is appropriate. Hence, def-
ature and reduces its viscosity, reaching a point
initions of high-speed versus slow-speed gear
of diminishing returns where increasing the
drives must be carefully considered, and pitch
starting viscosity of the lubricant leads to a
line velocity is generally a better index than shaft
higher operating temperature and a higher oxi-
speed. The gear drives were rebuilt with new
dation rate, without a signicant gain in operat-
gearsets and the ISO VG 68 oil was replaced with
ing viscosity.
ISO VG 150. The gear drives then operated with-
Gear drives operating in cold climates must
out overheating, and the pitting was eliminated.
have a lubricant that circulates freely and does
For critical applications, the specic lm
not cause high starting torques. A candidate
thickness, , should be calculated with Dowson
gear lubricant should have a pour point at least
and Higginsons equation (see Eq 8) (Ref 16).
5 C (9 F) lower than the expected minimum
The specic lm thickness is a useful measure of
ambient start-up temperature. Typical pour
the lubrication regime. It can be used with Fig. 7
points for mineral gear oils are 7 C (20 F),
as an approximate guide to the probability of
whereas synthetic gear lubricants have signi-
wear-related surface distress. Figure 7 is based
cantly lower pour points of about 40 C (40
on the data of Wellauer and Holloway (Ref 23),
F). Pour point depressants are used to tailor
which were obtained from several hundred labo-
pour points of mineral lubricants for automotive
ratory tests and eld applications of gear drives.
hypoid gears to be as low as 40 C (40 F).
The pitch line speed of the gears is a good
index of the required viscosity. An empirical Application of Gear Lubricants
equation for determining required viscosity is:
The method of applying the lubricant to the
7000 gear teeth depends primarily on the pitch line
n40 (Eq 18)
1V2 0.5 velocity.
34 / Gear Materials, Properties, and Manufacture

Splash lubrication systems are the sim- removes heat best by being directed at the out-
plest, but they are limited to a pitch line veloc- going side of the gear mesh where the oil jets
ity of ~1000 m/min (~3000 ft/min). The gears can strike the hot drive-side of the gear teeth.
should dip into the oil bath for about twice the For very high-speed gears (Ref 19) (>5300
tooth depth to provide adequate splash for pin- m/min, or 16,000 ft/min), there is a danger that
ions and bearings and to reduce losses due to the amount of oil carried to the incoming side of
churning. The gear housing should have troughs the gear mesh may be inadequate, and it is pru-
to capture the oil owing down the housing dent to add a supplementary ow at the incom-
walls, channeling it to the bearings. ing side of the gear mesh. Generally, about two-
The range of splash lubrication can be thirds of the oil ow should be supplied to the
extended to ~1500 m/min (~5000 ft/min) by outgoing side of the mesh for cooling, and one-
using bafes and oil pans to reduce churning. third of the ow directed at the incoming side
However, at velocities >1000 m/min (>3000 for lubrication. The placement of the oil jets is a
ft/min), providing auxiliary cooling with fans crucial factor when pitch line velocities are
and improving heat transfer by adding ns to the >6100 m/min (>20,000 ft/min). At speeds this
housing are usually necessary. high, experiments are required to nd the opti-
Pressure-Fed Systems. Above 1500 m/min mum number and location for the oil jets.
(5000 ft/min), most gears are lubricated by a In pressure-fed systems, the following
pressure-fed system. For gearboxes with parameters must be considered to ensure ade-
antifriction bearings, spraying the oil at the gear quate lubrication and cooling of the gear mesh:
mesh only and relying on splash to lubricate the
bearings is permissible up to a maximum pitch Quantity of ow
line velocity of 2100 m/min (7000 ft/min). Jet size
Above this speed and for gear drives with jour- Feed pressure
nal bearings, both the gears and bearings should Number of jets
be pressure-fed.
The oil jets should be placed on the incoming There are general guidelines, based on experi-
side of the gear mesh for pitch line velocities ence and experimentation, for specifying these
2600 m/min (8000 ft/min). Above 2600 parameters, but each application must be evalu-
m/min (8000 ft/min), more oil is needed for ated independently based on its particular oper-
cooling than for lubricating, and the oil ow ating conditions and requirements.

Fig. 7 Probability of wear distress as a function of specic lm thickness and pitch line velocity (in ft/min). Source: Ref 18
Chapter 2: Gear Tribology and Lubrication / 35

An empirical equation used to calculate the by the minimum recommended jet diameter of
quantity of oil ow, q, is: 0.8 mm (0.03 in.).
Jet Quantity. The number of jets should be
q P>c (Eq 21) sufcient to provide complete lubrication cov-
where q is the oil ow rate (in gal/min), P is the erage of the face width. More than one jet for
transmitted power (in hp), and c is taken from each gear mesh is advisable because of the pos-
Table 5. sibility of clogging. The upper limit on the num-
For a typical industrial application transmit- ber of jets is determined by the ow rate and jet
ting 150 kW (200 hp), where weight is not crit- diameter; too many jets for a given ow rate
ical, the designer might choose a value of c = will result in a jet diameter less than the mini-
200 hp/(gal/min), resulting in a copious ow of mum recommended.
4 L/min (1 gal/min). On the other hand, for a
high-efciency aviation application transmit-
ting 150 kW (200 hp), where weight is critical, Case History: Failure of a
a value of c = 800 hp/(gal/min) might be chosen, 24-Unit Speed-Increaser Gearbox
resulting in a lean ow of 1 L/min (0.25
gal/min). Some applications may require differ- This case history demonstrates many of the
ent ow rates than those given by Table 5. For principles of gear lubrication described in ear-
instance, wide-face, high-speed gearing may lier sections of this chapter. Lubricant viscosity,
require a higher ow rate to ensure uniform EHD lm thickness, contact temperature, and
cooling and full-face coverage. design considerations are addressed.
The proper jet size, feed pressure, and num- Background. In an industrial application,
ber of jets must be determined to maintain the 24 speed-increaser gearboxes were used to
proper ow rate, jet velocity, and full-face cov- transmit 258 kW (346 hp) and increase speed
erage. from 55 to 375 rev/min. The gears were parallel
Jet Size. The diameter of a jet can be calcu- shaft, single helical, carburized, and ground.
lated for a given ow rate and pressure based on The splash lubrication system used a mineral oil
the viscosity of the oil at the operating tempera- without antiscuff additives with ISO 100 vis-
ture (Ref 24). There are practical limitations on cosity. After about 250 h of operation, two gear-
jet size, and the minimum recommended size is boxes failed by bending fatigue. The gear tooth
0.8 mm (0.03 in.). If a jet smaller than this is proles were so badly worn that determining the
used, contaminants in the oil may clog it. Typi- primary failure mode was impossible. Three
cal jet diameters range from 0.8 to 3.0 mm (0.03 other gearboxes with less service were selected
to 0.12 in.). for inspection. One had logged 15 h, and the
Feed Pressure. The feed pressure deter- other two had operated for 65 h each. Upon dis-
mines the jet velocity, which in turn determines assembly, no broken teeth were found, but all
the amount of oil that penetrates the gear mesh. three gearboxes had scuffed gear teeth. The pri-
Typical feed pressures range from 20 to 100 mary failure mode was scufng, and the earlier
psig. Industrial application feed pressures are bending fatigue failures were caused by
typically 30 psig, and high-speed aerospace dynamic loads generated by the worn gear teeth.
applications are typically 100 psig. In general, Subsequent inspection of the remaining gear-
the higher the pressure, the greater the cooling boxes revealed that all had scufng damage,
(Ref 25), but the higher the pressure, the smaller which probably had occurred immediately upon
the jet diameter. Therefore, pressure is limited start-up because the loads were not reduced dur-
ing run-in.
Analysis. Fortunately, a prototype gearbox
had been run at one-half load for about 50 h.
Table 5 Recommended values of constant c When these gears were inspected, no signs of
based on oil ow and gear mesh specications distress were seen on any of the gear teeth. The
c, hp/ Flow tooth proles were smooth with surface rough-
(gal/min) conditions Applications ness, Rq, estimated to be 0.5 m (20 in.), and
200 Copious General industrial the contact pattern indicated 100% face contact.
400 Adequate Typical aviation This gearbox was reassembled and run under
800 Lean Lightweight, high-efciency aviation
1000 Starved Only for unusual conditions one-half load until its oil sump temperature
reached equilibrium at 95 C (200 F). For this
36 / Gear Materials, Properties, and Manufacture

application, the ambient temperature was in the Ts 146 59 ln 1100 2 418 F (Eq 23)
range of 10 to 50 C (50 to 125 F). The center
distance of the gears was 405 mm (16 in.) and Figure 9 shows a plot of the contact tempera-
the pitch line velocity was 120 m/min (400 ture versus position on the pinion tooth. The
ft/min). Referring to AGMA 250.04 (Ref 21), maximum contact temperature occurs high on
the recommended viscosity for these conditions the pinion tooth near the highest point of single-
is ISO 150 or ISO 220. tooth contact (HPSTC), where Tc is 226 C (439
Using the empirical equation (Eq 18), the fol- F). The program predicts that the probability of
lowing is obtained: scufng is 63%. This is considered to be a high
risk of scufng. The relatively high-temperature
7000 peak near the tip of the pinion tooth was caused
n40 350 cSt (Eq 22)
14002 0.5 by the geometry of the gears.
The designer selected a long-addendum tooth
Hence, the empirical equation recommends a for the pinion. Long-addendum pinions perform
viscosity close to ISO 320. It is apparent that the well in speed reducers, where they increase the
viscosity that was originally supplied (ISO VG amount of recess action and decrease the
100) was too low. amount of approach action of the gear mesh.
The EHD lm thickness was calculated with Because recess action is much smoother than
a special computer program (Ref 26). The gear approach action, long-addendum pinions give
bulk temperature was assumed to be 110 C speed reducers smooth meshing characteristics.
(230 F), which is 17 C (30 F) hotter than the When operated as a speed increaser, however,
measured oil sump temperature. The following the approach and recess portions of the gear
data for the ISO VG 100 lubricant was obtained mesh reverse, making a long-addendum pinion
from Fig. 4 and 5: m0 = 6.6 cP = 0.96 106 rough running and vulnerable to scufng.
reyns and a = 1.02 104 in.2/lbf. To explore the possibilities for reducing the
Figure 8 shows a plot of the lm thickness scufng risk, new gear tooth geometry was pro-
versus position on the pinion tooth. The mini- posed with the pinion and gear addenda
mum lm thickness occurs low on the pinion designed to minimize the ash temperature rise.
tooth near the lowest point of single-tooth con- The new gearset, analyzed with the program,
tact (LPSTC), where hmin is 0.053 mm (2.1 min.). assumed the lubricant was a mineral oil with
The specic lm thickness, based on Rq of 0.5 antiscuff additives, with a viscosity of ISO 220,
mm (20 min.) for both proles, is l = 0.073. Fig- and with the following properties: m0 = 10 cP =
ure 7 shows that the gears operate in the bound- 1.45 106 reyns, a = 1.09 104 in.2/lbf, and
ary lubrication regime. The program predicts
Ts 245 59 ln 1220 2 563 F (Eq 24)
that the probability of wear is >95%.
The contact temperature was also calculated Figure 10 shows that the lm thickness
with the program. The scufng temperature for increases to hmin of 0.068 mm (2.7 min.), and the
the ISO VG 100 lubricant was calculated with specic lm thickness increases to l = 0.097.
the equation for nonantiscuff mineral oils: Figure 7 shows that the gears still operate in the

Fig. 8 Plot of lm thickness versus pinion roll angle for gear Fig. 9 Plot of contact temperature versus pinion roll angle for
tooth geometry of a scuffed gearset. Minimum specic gear tooth geometry of scuffed gearset. Maximum Tc,
lm thickness, lmin, 0.073; probability of wear, >95% 226 C (439 F); scufng probability, 63%
Chapter 2: Gear Tribology and Lubrication / 37

boundary lubrication regime; however, the recognize that the lubricant is an important
probability of wear is reduced to 94%. Figure 11 component of a gearbox and appreciate that the
shows that the optimized gear geometry tribology of gearing requires the consideration
reduced the maximum contact temperature to a and control of many interrelated factors.
Tc of 150 C (302 F). The combination of
reduced contact temperature and the increased
scufng resistance provided by the higher vis- ACKNOWLEDGMENT
cosity mineral oil with antiscuff additives
This chapter was adapted from R. Errichello,
reduces the scufng probability to <5%.
Friction, Lubrication, and Wear of Gears, Fric-
Summary. Typical of many gear failures,
tion, Lubrication, and Wear Technology, Vol
this case history shows that several factors con-
18, ASM Handbook, ASM International, 1992,
tributed to the failures:
p 535545
The lubricant viscosity was too low
No antiscuff additives were used
A gearbox designed as a speed reducer was
used as a speed increaser 1. R. Errichello, The Lubrication of Gears,
The gear teeth were not provided with a Lubrication Engineering, JanApril, 1990
coating or plating to ease running-in 2. E.E. Shipley, Gear Failures, Mach. Des., 7
The gears were not run-in properly under Dec 1967, p 152162
reduced loads 3. D.W. Dudley, Gear Wear, Wear Control
Gear failures, as exemplied by this case his- Handbook, American Society of Mechani-
tory, can be avoided if designers and operators cal Engineers
4. P.M. Ku, Gear Failure ModesImpor-
tance of Lubrication and Mechanics, ASLE
Trans., Vol 19 (No. 3), 1975, p 239249
5. D.J. Wulpi, Understanding How Compo-
nents Fail, American Society for Metals,
6. L.E. Alban, Failures of Gears, Failure
Analysis and Prevention, Vol 11, 9th ed.,
Metals Handbook, ASM International,
1986, p 586601
7. D. Godfrey, Recognition and Solution of
Some Common Wear Problems Related to
Lubrication and Hydraulic Fluids, Lubr.
Eng., Feb 1987, p 111114
Fig. 10 Plot of lm thickness versus pinion roll angle for gear
8. W.E. Littman, The Mechanism of Contact
tooth geometry that was optimized for maximum
scufng resistance. lmin, 0.097; probability of wear, 94% Fatigue, Interdisciplinary Approach to the
Lubrication of Concentrated Contacts, SP-
237, National Aeronautics and Space
Administration, 1970, p 309377
9. T. Ueno et al., Surface Durability of Case-
Carburized GearsOn a Phenomenon of
Grey Staining of Tooth Surface, Paper No.
80-C2/DET-27, American Society of
Mechanical Engineers, 1980, p 18
10. H. Winter and T. Weiss, Some Factors
Inuencing the Pitting, Micropitting
(Frosted Areas) and Slow Speed Wear of
Surface-Hardened Gears, Paper No. 80-
C2/DET-89, American Society of Mechan-
Fig. 11 Plot of contact temperature versus pinion roll angle ical Engineers, 1980, p 17
for gear tooth geometry that was optimized for max-
imum scufng resistance. Maximum Tf, 150 C (302 F); scufng
11. E.E. Shipley, Failure Analysis of Coarse-
probability, <5% Pitch, Hardened, and Ground Gears, Paper
38 / Gear Materials, Properties, and Manufacture

No. P229.26, American Gear Manufactur- No. 250.04, American Gear Manufacturers
ers Association, 1982, p 124 Association, Sept 1981
12. S. Tanaka et al. Appreciable Increases in 22. Practice for High Speed Helical and Her-
Surface Durability of Gear Pairs with Mir- ringbone Gear Units, No. 421.06, Ameri-
ror-Like Finish, Paper No. 84-DET-223, can Gear Manufacturers Association, Jan
American Society of Mechanical Engi- 1969
neers, 1984, p 18 23. E.J. Wellauer and G.A. Holloway, Applica-
13. A. Milburn, R. Errichello, and D. Godfrey, tion of EHD Oil Film Theory to Industrial
Polishing Wear, Paper No. 90 FTM 5, Gear Drives, J. Eng. Ind. (Trans. ASME),
American Gear Manufacturers Associa- Vol 98, Series B (No. 2), May 1976, p
tion, 1990, p 113 626634
14. J.H. Admas and D. Godfrey, Borate Gear 24. R.J. Drago, Fundamentals of Gear Design,
Lubricant-EP Film Analysis and Perfor- Butterworths, 1998
mance, Lubr. Eng., Vol 37 (No. 1), Jan 25. L. Akin and D. Townsend, Study of Lubri-
1981, p 1621 cant Jet Flow Phenomena in Spur Gears,
15. H. Blok, Les Temperatures de Surface TMX-71572, National Aeronautics and
dans les Conditions de Graissage sons Pres- Space Administration, Oct 1974
sion Extreme, Second World Petroleum 26. SCORING+, computer program, GEAR-
Congress (Paris), June 1937 TECH Software, Inc., Copyright 1985
16. H. Blok, The Postulate about the Constancy 1992
of Scoring Temperatures, Interdisciplinary
Approach to the Lubrication of Concen-
trated Contacts, SP-237, National Aero- SELECTED REFERENCES
nautics and Space Administration, 1970, p
153248 R.H. Boehringer, Grease, Friction, Lubri-
17. D. Dowson, Elastohydrodynamics, Paper cation, and Wear Technology, Vol 18,
No. 10, Proc. Inst. Mech. Eng., Vol 182, ASM Handbook, ASM International, 1992,
PT3A, 1967, p 151167 p 123131
18. Fundamental Rating Factors and Calcula- H.S. Cheng, Lubrication Regimes, Fric-
tion Methods for Involute Spur and Helical tion, Lubrication, and Wear Technology,
Gear Teeth, 2001-B88, American Gear Vol 18, ASM Handbook, ASM Interna-
Manufacturers Association, 1988 tional, 1992, p 8997
19. M. Akazawa, T. Tejima, and T. Narita, R.S. Fein, Liquid Lubricants, Friction,
Full Scale Test of High Speed, High Pow- Lubrication, and Wear Technology, Vol
ered Gear UnitHelical Gears of 25,000 18, ASM Handbook, ASM International,
PS at 200 m/s PLV, Paper No. 80- 1992, p 8188
C2/DET-4, American Society of Mechani- S.Q.A. Rizvi, Lubricant Additives and
cal Engineers, 1980 Their Functions, Friction, Lubrication, and
20. R.J. Drago, Comparative Load Capacity Wear Technology, Vol 18, ASM Handbook,
Evaluation of CBN-Finished Gears, Paper ASM International, 1992, p 98112
No. 88 FTM 8, American Gear Manufac- H.E. Sliney, Solid Lubricants, Friction,
turers Association, Oct 1988 Lubrication, and Wear Technology, Vol
21. AGMA Standard SpecicationLubrica- 18, ASM Handbook, ASM International,
tion of Industrial Enclosed Gear Drives, 1992, p 113122
Gear Materials, Properties, and Manufacture Copyright 2005 ASM International
J.R. Davis, editor, p39-76 All rights reserved.
DOI:10.1361/gmpm2005p039 www.asminternational.org


Ferrous and Nonferrous Alloys

GEAR MATERIALS can be broadly classi- tent generally not exceeding 0.25% C. Table 1
ed into two groups: nonmetallic and metallic lists compositions of surface-hardening steels.
materials. Nonmetallics are the plastics, both Through-hardening steels may be compara-
thermoplastic and thermosetting, used for gear- tively shallow hardening or deep hardening,
ing. These materials are described in Chapter 4, depending on their chemical composition and
Plastics. Metallic gear materials can be fur- method of hardening. Through-hardening steels
ther subdivided into ferrous, or iron-base alloys, include plain carbon and alloy steels with car-
and nonferrous alloys. The most commonly bon content ranging from 0.30 to approximately
used ferrous alloys are the wrought surface- 0.55% C. Table 2 lists the compositions of
hardening and through-hardening carbon and through-hardening gear steels. Table 3 lists the
alloy steels. In fact, these steels are the most mechanical properties of both surface- and
widely used of all gear materials and will be the through-hardening steels in various conditions.
emphasis of this chapter. Other ferrous alloys The steels selected for gear applications must
used for gears are cast irons, cast steels, powder satisfy two basic sets of requirements that are
metallurgy (P/M) irons and steels, stainless not always compatiblethose involving fabri-
steels, tool steels, and maraging steels. cation and processing and those involving ser-
Although a number of nonferrous alloys have vice. Fabrication and processing requirements
been used for gears, by far the most commonly include machinability, forgeability, and re-
employed are copper-base alloys. Die cast alu- sponse to heat treatment as it affects fabrication
minum-, zinc-, and magnesium-base alloys are and processing. Service requirements are re-
also sometimes used. More recently titanium lated to the ability of the gear to perform satis-
alloy Ti-6Al-4V has been used in some special- factorily under the conditions of loading for
ized applications. which it was designed and thus encompass all
mechanical-property requirements, including
fatigue strength, response to heat treatment, and
Wrought Gear Steels resistance to wear.
Because resistance to fatigue failure is partly
Wrought steel is the generic term applied to dependent upon the cleanness of the steel and
carbon and alloy steels which are mechanically upon the nature of allowable inclusions, melting
worked into form for specic applications. The practice may also be a factor in steel selection
standard wrought steel forms are round bar and may warrant selection of a steel produced
stock, at stock, and forgings. Forgings reduce by vacuum induction melting followed by vac-
machining time and are available in a wide uum arc remelting (VIM/VAR) or by elec-
range of sizes and grades. troslag rening. The mill form from which a
In general, there are two types of wrought steel gear is machined is another factor that may
gear steels: surface-hardening and through- effect its performance. Many heavy-duty steel
hardening grades. The surface-hardened steels gears are machined from forged blanks that
are hardened to a relatively thin case depth and have been processed to provide favorable grain
include carburizing, nitriding, and carbonitrid- ow consistent with load pattern rather than
ing steels. Surface-hardening steels include being machined from blanks cut from mill-
plain carbon and alloy steels with carbon con- rolled bar.
40 / Gear Materials, Properties, and Manufacture

Surface-Hardening Steels A case-hardened gear provides maximum sur-

face hardness and wear resistance and at the same
General Properties. Carburizing or nitrid- time provides interior toughness to resist shock.
ing grades of steel are usually specied where In general, case-hardened gears can withstand
maximum wear resistance is required for bear- higher loads than through-hardened gears, al-
ing surfaces. Carburized case-hardened gears though the latter are quieter and less expensive
are best suited for heavy-duty service, for exam- because of the simpler heat treatment required
ple, transmission gears, and offer high resis- (Ref 1).
tance to wear, pitting, and fatigue. Surfaces Selection Factors. The following factors
must be sufciently hard to resist wear and of must be considered when selecting a case-
sufcient depth to prevent case crushing. A hardened gear (Ref 1):
rough rule for case depth is that it shall not
exceed one-sixth of the base thickness of the High tooth pressures will crack a thin case.
tooth. Figure 1 shows the typical structure of a Too soft a core will not provide proper back-
case-hardened gear steel. ing for a hard case.

Table 1 Chemical compositions of carburizing steels

The carburizing grades commonly used for gears are described in text.
Composition, %
Steel C Mn Ni Cr Mo Other

Carbon steels
1010 0.080.13 0.300.60 ... ... ... (a), (b)
1015 0.130.18 0.300.60 ... ... ... (a), (b)
1018 0.150.20 0.600.90 ... ... ... (a), (b)
1019 0.150.20 0.701.00 ... ... ... (a), (b)
1020 0.180.23 0.300.60 ... ... ... (a), (b)
1021 0.180.23 0.600.90 ... ... ... (a), (b)
1022 0.180.23 0.701.00 ... ... ... (a), (b)
1025 0.220.28 0.300.60 ... ... ... (a), (b)
1524 0.190.25 1.351.65 ... ... ... (a), (b)
1527 0.220.29 1.201.50 ... ... ... (a), (b)
Resulfurized steels
1117 0.140.20 1.001.30 ... ... ... 0.080.13 S
1118 0.140.20 1.301.60 ... ... ... 0.080.13 S
Alloy steels
3310 0.080.13 0.450.60 3.253.75 1.401.75 ... (b), (c)
4023 0.200.25 0.700.90 ... ... 0.200.30 (b), (c)
4027 0.250.30 0.700.90 ... ... 0.200.30 (b), (c)
4118 0.180.23 0.700.90 ... 0.400.60 0.080.15 (b), (c)
4320 0.170.22 0.450.65 1.652.00 0.400.60 0.200.30 (b), (c)
4615 0.130.18 0.450.65 1.652.00 ... 0.200.30 (b), (c)
4620 0.170.22 0.450.65 1.652.00 ... 0.200.30 (b), (c)
4815 0.130.18 0.400.60 3.253.75 ... 0.200.30 (b), (c)
4820 0.180.23 0.500.70 3.253.75 ... 0.200.30 (b), (c)
5120 0.170.22 0.700.90 ... 0.700.90 ... (b), (c)
5130 0.280.33 0.700.90 ... 0.801.10 ... (b), (c)
8617 0.150.20 0.700.90 0.400.70 0.400.60 0.150.25 (b), (c)
8620 0.180.23 0.700.90 0.400.70 0.400.60 0.150.25 (b), (c)
8720 0.180.23 0.700.90 0.400.70 0.400.60 0.200.30 (b), (c)
8822 0.200.25 0.751.00 0.400.70 0.400.60 0.300.40 (b), (c)
9310 0.080.13 0.450.65 3.003.50 1.001.40 0.080.15 (b), (c)
Special alloys
CBS-600 0.160.22 0.400.70 ... 1.251.65 0.901.10 0.901.25 Si
CBS-1000M 0.100.16 0.400.60 2.753.25 0.901.20 4.005.00 0.400.60 Si
0.150.25 V
Pyrowear Alloy 53 0.10 0.35 2.00 1.00 3.25 1.00 Si, 2.00 Cu, 0.10 V
(a) 0.04 P max, 0.05 S max. (b) 0.150.35 Si. (c) 0.035 P max, 0.04 S max
Chapter 3: Ferrous and Nonferrous Alloys / 41

Compressive stresses in the case improve Carburizing Steels. Carburizing is a very

fatigue durability, and a high case hardness important commercial heat-treating operation
increases wear resistance. that is used to modify the surface chemistry of
If the ratio of case depth to core thickness is components manufactured from ferrous alloys
too small, excessive stresses in subsurface by the processes of carbon absorption and diffu-
layers can produce poor fatigue life. sion. The process is carried out at a temperature
Residual tensile stresses are highest with sufcient to render the steel austenitic (gener-
low core hardness and increase with increas- ally between 850 and 950 C, or 1560 and 1740
ing case depth. These stresses can be F), followed by quenching and tempering to
relieved by tempering. form a high-carbon martensitic structure. The
increase in carbon content of a carburized sur-
Grain size variations have an important effect face layer results in a substantial change in the
on core properties. These variations are inu- properties of the effected volume of material.
enced by the type of steel and the method of heat For example, the carburized case will be harder,
treatment used subsequent to carburizing. Sec- will be more resistant to abrasive wear, and will
tion thickness also inuences core properties. exhibit improved fatigue properties compared
A tough tooth core may not be required in with the uneffected core. These variations in
applications where a gear will not be subjected properties are quite useful in applications where
to impact loading. In these applications, core a hard, wear-resistant surface is needed and
properties are relatively unimportant, provided where a softer, more ductile core is required to
the core is sufciently hard to support the case. prevent catastrophic failure of the component.
Considering the case alone, it is important that Carburizing methods include:
the surface resist wear and fatigue bending,
because bending stresses vary from a maximum Gas carburizing
at the surface to zero near the tooth center. Vacuum carburizing

Table 2 Chemical compositions of through-hardening gear steels

Composition, %
Steel C Mn Ni Cr Mo Other

Carbon steels
1035 0.320.38 0.600.90 ... ... ... (a)(b)
1040 0.370.44 0.600.90 ... ... ... (a)(b)
1045 0.430.50 0.600.90 ... ... ... (a)(b)
Free-cutting (resulfurized) carbon steels
1137 0.320.39 1.351.65 ... ... ... 0.080.13 S
1141 0.370.45 1.351.65 ... ... ... 0.080.13 S
1144 0.400.48 1.351.65 ... ... ... 0.240.33 S
Alloy steels
1340 0.380.43 1.601.90 ... ... ... (c)(d)
3140 0.380.43 0.700.90 1.101.40 0.550.75 ... (c)(d)
4042 0.400.45 0.700.90 ... ... 0.200.40 (c)(d)
4130 0.280.33 0.400.60 ... 0.801.10 0.150.25 (c)(d)
4140 0.380.43 0.751.00 ... 0.801.10 0.150.25 (c)(d)
4142 0.400.45 0.751.00 ... 0.801.10 0.150.25 (c)(d)
4145 0.410.48 0.751.00 ... 0.801.10 0.150.25 (c)(d)
4150 0.480.53 0.751.00 ... 0.801.10 0.150.25 (c)(d)
4340 0.380.43 0.600.80 1.652.00 0.700.90 0.200.30 (c)(d)
4350 0.480.53 0.600.80 1.652.00 0.700.90 0.200.30 (c)(d)
5140 0.380.43 0.700.90 ... 0.700.90 ... (c)(d)
6145 0.410.48 0.700.90 ... 0.801.10 ... (c)(d) 0.15 V min
8640 0.380.43 0.751.00 0.400.70 0.400.60 0.150.25 (c)(d)
8740 0.380.43 0.751.00 0.400.70 0.400.60 0.200.30 (c)(d)
9840 0.380.43 0.700.90 0.851.15 0.700.90 0.080.15 (c)(d)
(a) 0.04 P max, 0.05 S max. (b) 0.100.60 Si. (c) 0.035 P max, 0.04 S max (d) 0.150.35 Si
42 / Gear Materials, Properties, and Manufacture

Plasma carburizing burizing), liquids (salt bath carburizing), or

Salt bath carburizing solid compounds (pack carburizing). All of
Pack carburizing these methods have advantages and limitations,
but gas carburizing is used most often for large-
These methods introduce carbon by the use of scale production because it can be accurately
gas (atmospheric-gas, plasma, and vacuum car- controlled and involves a minimum of special

Table 3 Mechanical properties of selected gear steels

Tensile strength Yield strength
Elongation, Reduction Hardness,
Steel Condition MPa ksi MPa ksi in 50 mm, % in area, % HB

Carbon steel bar(a)

1015 Hot rolled 345 50 190 27.5 28 50 101
Cold drawn 385 56 325 47 18 40 111
1018 Hot rolled 400 58 220 32 25 50 116
Cold drawn 440 64 370 54 15 40 126
1020 Hot rolled 380 55 205 30 25 50 111
Cold drawn 420 61 350 51 15 40 121
1022 Hot rolled 425 62 235 34 23 47 121
Cold drawn 475 69 400 58 15 40 137
1025 Hot rolled 400 58 220 32 25 50 116
Cold drawn 440 64 370 54 15 40 126
1040 Hot rolled 525 76 290 42 18 40 149
Cold drawn 585 85 490 71 12 35 170
1045 Hot rolled 565 82 310 45 16 40 163
Cold drawn 625 91 530 77 12 35 179
Annealed, cold drawn 585 85 505 73 12 45 170
Spheroidized annealed, cold drawn 650 94 500 72.5 10 40 192
1117 Hot rolled 425 62 235 34 23 47 121
Cold drawn 475 69 400 58 15 40 137
1118 Hot rolled 450 65 250 36 23 47 131
Cold drawn 495 72 420 61 15 40 143
Low-alloy steels(b)
4130 Normalized at 870 C (1600 F) 670 97 435 63 25.5 59.5 197
Annealed at 865 C (1585 F) 560 81 460 67 21.5 59.6 217
Water quenched from 855 C (1575 F) 1040 151 979 142 18.1 63.9 302
and tempered at 540 C (1000 F)
4140 Normalized at 870 C (1600 F) 1020 148 655 95 17.7 46.8 302
Annealed at 815 C (1500 F) 655 95 915 60 25.7 56.9 197
Water quenched from 845 C (1550 F) 1075 156 986 143 15.5 56.9 311
and tempered at 540 C (1000 F)
4150 Normalized at 870 C (1600 F) 1160 168 731 106 11.7 30.8 321
Annealed at 830 C (1525 F) 731 106 380 55 20.2 40.2 197
Oil quenched from 830 C (1525 F) 1310 190 1215 176 13.5 47.2 375
and tempered at 540 C (1000 F)
4340 Normalized at 870 C (1600 F) 1282 186 862 125 12.2 36.3 363
Annealed at 810 C (1490 F) 745 108 470 68 22.0 50.0 217
Oil quenched from 800 C (1475 F) 1207 175 1145 166 14.2 45.9 352
and tempered at 540 C (1000 F)
5140 Normalized at 870 C (1600 F) 793 115 470 68 22.7 59.2 229
Annealed at 830 C (1525 F) 570 83 290 42 28.6 57.3 167
Oil quenched from 845 C (1550 F) 972 141 841 122 18.5 58.9 293
and tempered at 540 C (1000 F)
8620 Normalized at 915 C (1675 F) 635 92 360 52 26.3 59.7 183
Annealed at 870 C (1600 F) 540 78 385 56 31.3 62.1 149
8630 Normalized at 870 C (1600 F) 650 94 425 62 23.5 53.5 187
Annealed at 845 C (1550 F) 565 82 370 54 29.0 58.9 156
Water quenched from 845 C (1550 F) 931 135 850 123 18.7 59.6 269
and tempered at 540 C (1000 F)
8740 Normalized at 870 C (1600 F) 931 135 605 88 16.0 47.9 269
Annealed at 815 C (1500 F) 696 101 415 60 22.2 46.4 201
Oil quenched from 830 C (1525 F) 1225 178 1130 164 16.0 53.0 352
and tempered at 540 C (1000 F)
9310 Normalized at 890 C (1630 F) 910 132 570 83 18.8 58.1 269 HRB
Annealed at 845 C (1550 F) 820 119 450 65 17.3 42.1 241 HRB
Aged sheet 6 mm (0.25 in.) 2169 315 2135 310 7.7 35 55.1 HRC
(a) All values are estimated minimum values; type 1100 series steels are rated on the basis of 0.10% max Si or coarse-grain melting practice; the mechanical proper-
ties shown are expected minimums for the sizes ranging from 19 to 31.8 mm (0.75 to 1.25 in.). (b) Most data are for 25 mm (1 in.) diam bar.
Chapter 3: Ferrous and Nonferrous Alloys / 43

handling. More detailed information on various than can be achieved with shallow hardening
carburizing methods can be found in Chapter 9, steels quenched in the same size section.
Carburizing. Nitriding Steels. Nitriding is a surface-
The core carbon content of carburized gears hardening heat treatment that introduces nitro-
is usually within the range of 0.10 to 0.25%. A gen into the surface of steel at a temperature
lower carbon content is usually used to obtain range of 500 to 550 C (930 to 1020 F) while it
maximum ductility, and a higher carbon content is in the ferritic condition. Thus, nitriding is sim-
is used to obtain maximum core strength. Some ilar to carburizing in that surface composition is
representative SAE-AISI carburizing steels altered, but different in that nitrogen is added
used for gears include: into ferrite instead of austenite. Because nitrid-
ing does not involve heating into the austenite
Plain carbon steels: 1015, 1018, 1020, 1022, phase eld and a subsequent quench to form
and 1025 martensite, nitriding can be accomplished with a
Free-machining steels: 1117 and 1118 minimum of distortion and with excellent
Alloy steels: 4020, 4026, 4118, 4320, 4620, dimensional control. Process methods for nitrid-
4820, 5120, 8620, 8720, and 9310 ing include gas, liquid (salt bath), and plasma
Many other standard (SAE-AISI) and propri- (ion) nitriding. Details on these processes can be
etary carburizing steels are also available. found in Chapter 10, Nitriding.
The nickel-bearing carburizing steels are used Nitriding steels can be used in many gear
chiey where exceptional core toughness com- applications where a hard, wear-resistant case,
bined with the highest degree of wear resistance good fatigue strength, low notch sensitivity, and
and greatest surface compressive strength is some degree of corrosion resistance are desired.
required. These steels include the nickel-molyb- In addition, nitriding steels make it possible to
denum steels (4600 and 4800 series) and the surface harden the teeth of large gears having
nickel-chromium-molybdenum steels (4300, thin sections that might be impractical to car-
8600, 8700, and 9300 series). The carbon- burize and quench.
molybdenum steels (4000 series) are used where Nitrided gears are relatively free from wear
exceptional toughness and good resistance to up to the load at which surface failure occurs,
temper embrittlement are required. but at this load they become badly crushed and
Another advantage of the more highly pitted. Thus, nitrided gears are generally not
alloyed steels is the ability of heavy sections to suitable for applications where overloads are
harden more completely. This greater harden- likely to be encountered.
ability promotes better core strength properties Nitrided steels are generally medium-carbon
(quenched-and-tempered) steels that contain
strong nitride-forming elements such as alu-
minum, chromium, vanadium, and molybde-
num. The most signicant hardening is achieved
with a class of alloy steels (Nitralloy steels as
described below) that contain about 1% Al.
When these steels are nitrided, the aluminum
forms AIN particles, which strain the ferrite lat-
tice and create strengthening dislocations.
Table 4 lists chemical compositions of Nitral-
loy gear steels. Nitralloy N, a nickel-bearing
(3.5% Ni) nitriding steel, is a precipitation-
hardening alloy that attains a core strength and
hardness after nitriding that are considerably in
excess of its original properties. Both Nitralloy
N and Nitralloy 135M are outstanding for heavy-
duty gears that are highly stressed. Little change
in tensile strength of nitriding steels occurs if the
tempering temperature used for treating the core
is at or above the nitriding temperature. How-
Fig. 1 Uniform case depth on a 40-tooth gear made from Fe-
ever, because of the increased hardness of the
0.16C-0.6Mn-0.37Si-1.65Cr-3.65Ni steel that was pro-
duced by ion carburizing case, the elongation, ductility, and impact
44 / Gear Materials, Properties, and Manufacture

strength of both alloys are considerably reduced high load-carrying capacity. Other standard
after tempering, though not to the same extent; (SAE-AISI) and proprietary through-hardening
Nitralloy N develops a tougher and softer sur- steels are also available.
face and a stronger core than Nitralloy 135M. When selecting a through-hardening steel, it
Any of the SAE-AISI steels that contain should be considered that a higher carbon and
nitride-forming elements, such as chromium, alloy content is accompanied by greater
vanadium, or molybdenum, can also be nitrided. strength and hardness (but lower ductility) of
The steels most commonly nitrided are 4140, the surface and the core. Fully hardened and
4340, 6140, and 8740. In some applications, the tempered medium-carbon alloy steels possess
0.50% C grades are also used. an excellent combination of strength and tough-
ness at room temperature and at lower tempera-
tures. However, toughness can be substantially
decreased by temper embrittlement by slow
Through-Hardening Steels cooling through the temperature range of 450 to
540 C (850 to 1000 F), or by holding or tem-
By virtue of their higher carbon content, pering in this range. Because of their good hard-
through-hardening steel gears possess greater enability and immunity to temper brittleness,
core strength than carburized gears. They are molybdenum steels have been widely used for
not, however, as ductile or as resistant to surface gears requiring good toughness at room and low
compressive stresses and wear as case-hardened temperatures.
gears. Hardness of gear surfaces may vary from
300 to 575 HB. Through-hardened steels may
also be effectively surface hardened by induc-
tion heating or by ame hardening. Gear Steel Requirements
Typical of the relatively shallow-hardening
carbon steel gear materials are SAE-AISI types Some of the more important requirements for
1035, 1040, 1045, 1050, 1137, 1141, 1144, and gear steels are their:
1340. These steels are water-hardening, but not
deep-hardening types that are suitable for gears Processing characteristics (for example,
requiring only a moderate degree of strength hardenability and machinability)
and impact resistance. Response to heat treatment. This subject is
In general, the more highly alloyed through- addressed in Chapters 8 through 12 which
hardening steels harden more completely when cover through-hardening, carburizing, ni-
quenched in heavy sections. This greater harden- triding, carbonitriding, and induction and
ability provides greater strength than can be ame hardening, respectively.
attained with shallow-hardening steels quenched Resistance to tooth bending fatigueboth
in the same size section. low-cycle (105 cycles to failure) and high-
Typical of the low-alloy, medium-to-deep cycle (>105 cycles to failure) fatigue. Be-
hardening gear materials are (in order of increas- cause carburized steels for high-performance
ing hardenability): 4042, 5140, 8640, 3140, gear applications are subjected to cyclic
4140, 8740, 6145, 9840, and 4340. These steels, loading, this is one of the most important
as well as many other alloy steels with the proper properties or measure of gear performance.
hardenability characteristics and a carbon con- The section on Bending Fatigue Strength of
tent of 0.35 to 0.50%, are suitable for gears Carburized Steels deals with this subject in
requiring medium-to-high wear resistance and detail.

Table 4 Nominal chemical compositions for aluminum-containing low-alloy steels commonly

gas nitrided
Steel Composition, %
SAE AMS Nitralloy C Mn Si Cr Ni Mo Al Se

... ... G 0.35 0.55 0.30 1.2 ... 0.20 1.0 ...
7140 6470 135M 0.42 0.55 0.30 1.6 ... 0.38 1.0 ...
... 6475 N 0.24 0.55 0.30 1.15 3.5 0.25 1.0 ...
... ... EZ 0.35 0.80 0.30 1.25 ... 0.20 1.0 0.20
Chapter 3: Ferrous and Nonferrous Alloys / 45

Resistance to surface-contact (pitting) tance, with other factors such as the thermal
fatigue. This subject is addressed in Chapter conductivity of steel and the rate of surface heat
2, Gear Tribology and Lubrication. removal being held constant. Therefore, the
Resistance to rolling contact fatigue terms Jominy distance (J ) and ideal critical
Resistance to wear. This subject is briey diameter (DI) derived from the Jominy end-
discussed in this chapter, but more detailed quench test can be used.
information on adhesive wear, abrasive The hardenability of steel is governed almost
wear, and scufng of gears can be found in entirely by the chemical composition (carbon
Chapter 2. and alloy content) at the austenitizing tempera-
Their hot hardness ture and the austenite grain size at the moment
Their bending strength and bend ductility of quenching. In some cases, the chemical com-
Their toughness, both impact toughness and position of the austenite may not be the same as
fracture toughness that determined by chemical analysis, because
some carbide may be undissolved at the austen-
Each of these will be described in subsequent itizing temperature. Such carbides would be
sections. reected in the chemical analysis, but because
the carbides are undissolved in the austenite,
neither their carbon nor alloy content can con-
Processing tribute to hardenability. In addition, by nucleat-
Characteristics of Gear Steels ing transformation products, undissolved car-
bides can actively decrease hardenability. This
Hardenability refers to the ability of a steel is especially important in high-carbon (0.50 to
to be transformed partially or completely from 1.10%) and alloy carburizing steels, which may
austenite to martensite at a given depth when contain excess carbides at the austenitizing tem-
cooled under prescribed conditions. This deni- perature. Consequently, such factors as austeni-
tion reects the empirical nature of steel hard- tizing temperature, time at temperature, and
enability, and, as discussed in Ref 2, many types prior microstructure are sometimes very impor-
of experiments have been devised to measure or tant variables when determining the basic hard-
describe the hardenability of various kinds of enability of a specic steel composition. Certain
steel. ingot casting and hot reduction practices may
Martensite is the microstructure usually also develop localized or periodic inhomo-
desired in quenched carbon and low-alloy steels. geneities within a given heat, further complicat-
The cooling rate in a quenched part must be fast ing hardenability measurements. The effects of
enough so that a high percentage of martensite is all these variables are discussed in Ref 2.
produced in critically stressed areas of the part. Table 5 provides a qualitative rating of the
Higher percentages of martensite result in higher hardenability of gear steels. Additional infor-
fatigue and impact properties after tempering. mation on hardness and hardenability can be
Hardenability should not be confused with found in Chapter 9, Carburizing.
hardness as such or with maximum hardness. Machinability. The term machinability is
The maximum attainable hardness of any steel used to indicate the ease or difculty with which
depends solely on carbon content. Also, the a material can be machined to the size, shape, and
maximum hardness values that can be obtained desired surface nish. The terms machinability
with small test specimens under the fastest cool- index and machinability rating are used as quali-
ing rates of water quenching are nearly always tative and relative measures of the machinability
higher than those developed under production of a steel under specied conditions. There are no
heat-treating conditions, because hardenability clear-cut or unambiguous meanings for these
limitations in quenching larger sizes can result terms and no standard or universally accepted
in less than 100% martensite formation. method of measuring machinability. Histori-
Basically, the units of hardenability are those cally, machinability judgments have been based
of cooling ratefor example, degrees per sec- on one or more of the following criteria:
ond. These cooling rates, as related to the Tool life: Measured by the amount of mate-
continuous-cooling-transformation behavior of rial that can be removed by a standard cut-
the steel, determine the hardness and micro- ting tool under standard cutting conditions
structural outcome of a quench. In practice, before tool performance becomes unaccept-
these cooling rates are often expressed as a dis- able or tool wear reaches a specied amount
46 / Gear Materials, Properties, and Manufacture

Cutting speed: Measured by the maximum and poor machinability. With good machinabil-
speed at which a standard tool under stan- ity as a base, a fair rating would add 20 to 30%
dard conditions can continue to provide sat- to the machining cost, and a poor rating would
isfactory performance for a specied period add 40 to 50%.
Power consumption: Measured by the power
required to remove a unit volume of material
under specied machining conditions Bending Fatigue
Comparisons with a standard steel based on
Strength of Carburized Steels
experience in machine shops
Quality of surface nish
Bending fatigue of carburized steel compo-
Feeds resulting from a constant thrust force
nents is a result of cyclic mechanical loading.
Some of the test criteria are best suited to labo- The bending produces stresses, which are ten-
ratory studies intended to elicit information sile at the surface, decrease with increasing dis-
about the effects of small changes in micro- tance into the component, and at some point
structure, composition, or processing history on become compressive. Such loading is a charac-
machinability. Other types of tests are useful for teristic of the roots of gear teeth. Carburizing
studying the effects of geometry changes or cut- produces a high-carbon, high-strength surface
ting tool composition. layer, or high-strength case, on a low-carbon,
Figure 2 is a plot of cutting speed for one- low-strength interior or core and is therefore, an
hour tool life versus the Brinell hardness of var- ideal approach to offset the high surface tensile
ious gear steels. This gure indicates a wide stresses associated with bending. Thus when the
range correlation. For example, at a hardness design of a component maintains operating
range of 175 to 200 HB, the cutting speed for stress gradients below the fatigue strength of the
one-hour tool life varies from 23 to 43 m/min case and core microstructures, excellent bend-
(75 to 140 surface feet per minute). The gure ing fatigue resistance is established.
indicates decreased cutting speed for one-hour There are, however, many alloying and pro-
tool life as the hardness increases. In Table 6, cessing factors that produce various microstruc-
the machinability of some of the more common tures, and therefore variable strength and frac-
gear steels is listed on the basis of good, fair, ture resistance, of the case regions of carburized

Table 5 Hardenability characteristics of commonly used gear steels

Common alloy steel grades Common heat treat practice(a) Hardenability

Noncarburizing grades
1045 T-H, I-H, F-H Low hardenability
4130 T-H Marginal hardenability
4140 T-H, T-H&N, I-H, F-H Fair hardenability
4145 T-H, T-H&N, I-H, F-H Medium hardenability
8640 T-H, T-H&N, I-H, F-H Medium hardenability
4340 T-H, T-H&N, I-H, F-H Good hardenability in heavy sections
4150 I-H, F-H, T-H, TH&N Quench crack sensitive
Good hardenability
4142 I-H, F-H, T-H&N Used when 4140 exhibits
Marginal hardenability
4350 T-H, I-H, F-H Quench crack sensitive, excellent
Hardenability in heavy sections
Carburizing grades
1020 C-H Very low hardenability
4118 C-H Fair core hardenability
4620 C-H Good case hardenability
8620 C-H Fair core hardenability
4320 C-H Good core hardenability
8822 C-H Good core hardenability in heavy sections
3310 C-H Excellent hardenability (in heavy sections) for all three grades
4820 C-H
9310 C-H
(a) C-H = Carburize harden, F-H = Flame harden, I-H = Induction harden, T-H = Through harden, T-H&N = Through harden then nitride. Source: American Gear
Manufacturers Association
Chapter 3: Ferrous and Nonferrous Alloys / 47

steels. When applied surface stresses exceed the presented as plots of maximum stress, S, versus
surface strength, surface fatigue crack initiation the number of cycles, N, to fracture for a speci-
and eventual failure will develop. When the sur- ed stress ratio (R), which is the ratio of mini-
face strength is adequate, depending on the mum (or compressive) stress to maximum ten-
steepness of the applied stress gradients in rela- sile stress (R = min-S/max-S). Figure 3 shows an
tionship to the case/core strength gradient, sub- example of typical S-N plots for a series of car-
surface fatigue cracking may develop. burized alloy steels (Ref 4). The S-N curve con-
Bending fatigue performance of carburized sists of two parts: a straight section with nega-
steels can vary signicantly. One study reported tive slope at low cycles and a horizontal section
values of experimentally measured endurance at high cycles. The horizontal line denes the
limits ranging from 200 to 1930 MPa (29 to 280 fatigue limit or endurance limit, which is taken
ksi), with most values between 700 and 1050 to be the maximum applied stress below which
MPa (100 and 152 ksi) (Ref 3). As will be a material is assumed to be able to withstand an
described subsequently, this wide variation in innite number of stress cycles without failure.
fatigue performance is a result of variations in Pragmatically, the endurance limit is taken as
specimen design and testing, alloying, and pro- the stress at which no failure occurs after a set
cessing interactions that produce large variations number of cycles, typically on the order of 10
in carburized microstructures and the response of million cycles. The low-cycle portion of the S-N
the microstructures to cyclic loading. plot denes various fatigue strengths or the
stresses to which the material can be subjected
Bending Fatigue Testing
for a given number of cycles. The more cycles
Data Presentation and Analysis. Most at a given strength, the better the low-cycle
bending fatigue data for carburized steels are fatigue resistance of a material.

Fig. 2 Cutting speed for one-hour tool life versus Brinell hardness number for various through-hardened carbon and alloy steels
48 / Gear Materials, Properties, and Manufacture

Analysis of bending fatigue behavior of car- strains, while high-cycle fatigue behavior is
burized steels based on S-N curves represents a determined by elastic strains. In particular, duc-
stress-based approach to fatigue and assumes tile materials with microstructures capable of
that the carburized specimens deform nominally sustaining large amounts of plastic deformation
only in an elastic manner (Ref 5). This assump- have better low-cycle fatigue resistance, while
tion is most valid at stresses up to the endurance high-strength materials with high elastic limits
limit and is useful when machine components and high yield strengths have better high-cycle
are designed for high-cycle fatigue. However, as fatigue resistance. Figure 4 shows the results of
maximum applied stresses increase above the strain-based bending fatigue testing of uncarbur-
endurance limit, plastic strain becomes increas- ized and carburized 4027 steel (Ref 6). The more
ingly important during cyclic loading, and ductile, low-strength uncarburized specimens
fatigue is more appropriately analyzed by a show better fatigue resistance at low cycles than
strain-based approach. In this approach, the total do the carburized specimens. The performance
strain range is the sum of the applied elastic and is reversed at high cycles, where the carburized
plastic strains, and the strain amplitude is plotted specimens with their high-strength surfaces
as a function of the strain reversals required for show better fatigue resistance, especially those
failure at the various levels of strain. specimens with the deeper cases.
According to the strain-based approach, low- Specimen Design. Many types of speci-
cycle fatigue behavior is determined by plastic mens have been used to evaluate bending

Table 6 Machinability of commonly used gear materials. Refer also to the text describing this table
Material Remarks

Low-carbon carburizing steel grades(a)

1020 Good machinability, as rolled, as forged, or normalized.
4118 Good machinability, as rolled, or as forged. However, normalized is preferred. Inadequate cooling during
4620 normalizing can result in gummy material, reduced tool life and poor surface nish. Quench and temper as a
8620 prior treatment can aid machinability. The economics of the pretreatments must be considered.
3310 Fair to good machinability if normalized and tempered, annealed or quenched and tempered. Normalizing
4320 without tempering results in reduced machinability.
Medium-carbon through-hardening steel grades(a)
1045 Good machinability if normalized.
4130 Good machinability if annealed, or normalized and tempered to approximately 255 HB or quenched and
4140 tempered to approximately 321 HB. Over 321 HB, machinability is fair. Above 363 HB, machinability is
4142 poor. Inadequate (slack) quench with subsequent low tempering temperature may produce a part which
meets the specied hardness, but produces a mixed microstructure which results in poor machinability.
4145 Remarks for medium carbon alloy steel (above) apply. However, the higher carbon results in lower
4150 machinability. Sulfur additions aid the machinability of these grades. 4340 machinability is good up to 363
4340 HB. The higher carbon level in 4145, 4150, 4345, and 4350 makes them more difcult to machine and
4345 should be specied only for heavy sections. Inadequate (slack) quench can seriously affect machinability in
4350 these steels.
Other gear materials
Gray Irons Gray cast irons have good machinability. Higher strength gray cast irons [above 345 MPa (50 ksi) tensile
strength] have reduced machinability.
Ductile Irons Annealed or normalized ductile cast iron has good machinability. The as cast (not heat treated) ductile iron
has fair machinability. Quenched and tempered ductile iron has good machinability up to 285 HB and fair
machinability up to 352 HB. Above 352 HB, machinability is poor.
Gear Bronzes and Brasses All gear bronzes and brass have good machinability. The very high strength heat treated bronzes [above 760
MPa (110 ksi) tensile strength] have fair machinability.
Austenitic Stainless Steel All austenitic stainless steel grades only have fair machinability. Because of work hardening tendencies, feeds
and speeds must be selected to minimize work hardening.
(a) Coarse grain steels are more machinable than ne grain. However, gear steels are generally used in the ne grain condition since mechanical properties are improved,
and distortion during heat treatment is reduced. Increasingly cleaner steels are now also being specied for gearing. However, if sulfur content is low, less than 0.015%,
machinability may decrease appreciably. Source: American Gear Manufacturers Association
Chapter 3: Ferrous and Nonferrous Alloys / 49

fatigue in carburized steels. Rotating beam, rounding of the corners of the beam section. If
unnotched four-point bend, notched four-point the corners are square, the carbon introduced
bend, and cantilever beam specimens have all into the corner surfaces cannot readily diffuse
been used, and they have in common a maxi- into the interior of the specimen. As a result, the
mum applied surface tensile stress and decreas- corner microstructures may have signicantly
ing tensile stress with increasing distance into elevated levels of retained austenite and coarse
the specimen. Axial fatigue testing of carbur- carbide structuresboth microstructural fea-
ized specimens, which applies the maximum tures that inuence bending fatigue resistance
tensile stresses uniformly over the cross section (Ref 8).
of a specimen, invariably results in subsurface Mean Stress and Stress Ratio. The maxi-
initiation and propagation of fatigue cracks in mum applied surface tensile stress is the testing
the core of carburized specimens, and therefore parameter plotted in S-N curves that characterize
it does not permit evaluation of the resistance of bending fatigue. However, the applied stress
case microstructures to fatigue. ranges between maximum and minimum values
Brugger was the rst to use cantilever bend during a fatigue cycle, and two other parameters,
specimens to evaluate the fracture and fatigue of mean stress and stress ratio, are important for the
carburized steels (Ref 7). Figure 5 shows a can- characterization of fatigue. The mean stress is
tilever bend specimen that has evolved from the algebraic average of the maximum and mini-
the Brugger specimen. The radius between the mum stresses in a cycle, and as discussed, the
change in section simulates the geometry at the stress ratio, R, is the ratio of the minimum stress
root of gear teeth and results in maximum to the maximum stress in a cycle. Thus R values
applied surface stresses just where the cross sec- may range from 1, for fully reversed loading
tion begins to increase, as shown in Fig. 6 (Ref that ranges between equal maximum tensile and
8). An important feature of this specimen is the compressive stresses, to positive values where

Fig. 3 Typical maximum stress (S ) vs. number of cycles (N ) bending fatigue plots for 6 carburized steels. R = 1. Source: Ref 4
50 / Gear Materials, Properties, and Manufacture

the stress is cycled between two tensile values especially effective for parts with sharp notches.
(Ref 5). Much of the bend testing of cantilever In the absence of notches, carburizing is most
specimens described subsequently is performed suitable for parts subjected to fatigue loading at
with R values of 0.1 in order to preserve details low values of mean stress.
of the fracture surface. The testing of actual machine compo-
Figure 7 shows a typical allowable-stress dia- nents is another important approach to the
gram that plots fatigue strength versus mean fatigue evaluation of carburized steels. An
stress for a given material (Ref 9). The diagram example of component testing is the bending
shows that the most severe condition for fatigue fatigue testing of single teeth in gears (Ref 11).
is for fully reversed testing with R = 1.0. As the Gears are fabricated, carburized, and mounted
mean stress increases, the fatigue strength in in a xture so that one tooth at a time is sub-
terms of maximum applied stress increases, but jected to cyclic loading. More recently, identi-
the allowable stress range decreases. Zurn and cally carburized specimens of the same steel
Razim (Ref 10) have examined the effect of were subjected to cantilever bend and single
notch severity and retained austenite on allow- tooth bending fatigue testing (Ref 12). The
able-stress/mean-stress diagrams of carburized
steels, and they conclude that carburizing, rela-
tive to the use of through-hardened steels, is

Fig. 5 Example of a cantilever specimen used to evaluate

bending fatigue of carburized steels. Specimen edges
are rounded and maximum stress is applied at the location
shown in Fig. 6. Dimensions in millimeters

Fig. 4 Strain amplitude vs. reversals to failure for uncarbur-

ized (solid symbols) and carburized (open symbols)
4027 steel (0.80% Mn, 0.28% Si, 0.27% Mo). Source: Ref 6

Fig. 6 Location of maximum stress on the cantilever bend

specimen shown in Fig. 5 as determined by nite ele-
ment modeling Fig. 7 Typical allowable-stress-range diagram. Source: Ref 9
Chapter 3: Ferrous and Nonferrous Alloys / 51

mechanisms of fatigue failure, based on fracture in direct-quenched carburized specimens, inter-

surface examination, were found to be the same, granular cracks are initiated as soon as the
but the single tooth testing showed higher levels applied surface bending stress reaches a level
of fatigue resistance than did the cantilever test- sufcient to exceed the surface compressive
ing, a result attributed to the higher surface residual stress and the cohesive strength of the
compressive stresses that were measured in the prior-austenite grain boundary structures. The
gear tooth specimens. More detailed informa- surface intergranular cracks are shallow, typi-
tion on testing of gears can be found in Chapter cally approximately two to four austenite
15, Mechanical Testing. grains, and are arrested, perhaps because of a
plastic zone smaller than the grain size at the tip
Stages of Fatigue and Fracture
Bending fatigue fractures of carburized steels
consist of well-dened stages of crack initiation,
stable crack propagation, and unstable crack
propagation. The fracture sequence is strongly
inuenced by the gradients in strength, micro-
structure, and residual stress that develop in car-
burized steels. Figure 8 shows a series of SEM
fractographs that characterize the typical frac-
ture sequence of a direct-quenched carburized
steel. The cantilever bend specimen from which
the fractographs of Fig. 8 were taken was a 4320
steel carburized to a 1 mm (0.03937 in.) case
depth at 927 C (1700 F), quenched from 850
C (1560 F) into oil at 65 C (150 F), and tem-
pered at 150 C (300 F) for 1 h. The specimen
was tested in bending fatigue with an R value of
0.1 (Ref 13). Figures 8(a) and (b) show a low-
magnication overview of the initiation, stable
propagation, and unstable fracture surfaces, and
Fig. 8(c) shows the intergranular initiation and
transgranular stable crack propagation zones of
the fracture at a higher magnication.
Intergranular cracking at prior-austenite
grain boundaries is an almost universal fracture
mode in the high-carbon case of direct-
quenched carburized steels (Ref 13). Not only
do the fatigue cracks initiate by intergranular
cracking, but also the unstable crack propagates
largely by intergranular fracture until it reaches
the lower-carbon portion of the case, where duc-
tile fracture becomes the dominant fracture
mode. In fact, sensitivity of the case microstruc-
tures to intergranular fracture makes possible
the quantitative characterization of the size and
shape of the stable fatigue crack, as shown in
Fig. 8(b). The transition from the transgranular
fracture of the stable crack to the largely inter-
granular fracture of the unstable fracture is iden-
tied by the dashed line.
A replica study of carburized specimens sub- Fig. 8 Fatigue fracture in gas-carburized and modied 4320
steel. (a) Overview of initiation, stable crack propaga-
jected to incrementally increasing stresses tion, and unstable crack propagation. (b) Same area as shown in
showed that surface intergranular cracks initi- (a), but with extent of stable crack indicated by dashed line. (c)
Higher magnication of intergranular initiation and transgranu-
ated when the applied stresses exceeded the lar stable crack propagation areas. SEM Fractographs. Source:
endurance limits (Ref 13). Thus, it appears that Ref 13
52 / Gear Materials, Properties, and Manufacture

of the sharp intergranular cracks, and the fact sizes and the stresses at which the cracks become
that strain-induced transformation of retained unstable are used to calculate the fracture tough-
austenite in the plastic zone ahead of the crack ness of the case microstructures of carburized
introduces compressive stresses (Ref 14, 15). steels (Ref 13), the results show good agreement
The fatigue crack then propagates in a trans- with the range of fracture toughness, 15 to 25
granular mode, and when the stable crack MPa 1m , that has been measured from through-
reaches critical size, as dened by the fracture hardened specimens with high-carbon LTT
toughness, unstable fracture occurs. martensitic microstructures (Ref 16). Table 7
The initiation and stable crack zones of car- shows the data used to calculate the various case
burized steels are quite small and are often dif- fracture toughness values in gas-carburized
cult to identify. Figure 9 based on the measure- 4320 steel and the fracture toughness values cal-
ment of critical crack sizes in a number of culated according to three different fracture
direct-quenched carburized 4320 steels, shows toughness equations (Ref 13):
that the size of the unstable cracks ranges from
0.170 to 0.230 mm (0.00669 to 0.00906 in.), and 1.2sa 2ap
that the cracks, therefore, become unstable well KIC
Q (Eq 1)
within the high-carbon portion of the carburized
specimens. The small critical crack sizes are 0s
consistent with the low fracture toughness of c 1.2sa 0.683 a b a d 2ap
high-carbon steel low-temperature-tempered KIC
(LTT) microstructures susceptible to intergranu- Q (Eq 2)
lar fracture (Ref 16). When the critical crack

Msa 2ap
Q (Eq 3)

sa 2
Q f2 0.212 a b

with the aspect ratio of crack depth (a) and

crack length (c) such that:
a 1.65
f2 1 1.464 a b

Fig. 9 Hardness vs. distance from the surface of direct-cooled The unstable crack that proceeds through the
gas-carburized SAE 4320 steel. Superimposed on the high- and medium-carbon martensitic portions
hardness prole is the range of critical depths (vertical dashed
band) at which stable fatigue cracks became unstable in bending
of the case may be arrested when the sensitivity
fatigue of similarly processed steels. Source: Ref 13 to intergranular fracture decreases at a case car-

Table 7 Fracture toughness results for carburized SAE 4320 bending fatigue specimens
Cycles Depth, a, Width, c, KIc, MPa 1m , KIc, MPa 1m , KIc, MPa 1m ,
Max stress, MPa to failure mm mm a/c M Eq 21 Eq 22 Eq 23

1370 6,400 175 388 0.45 0.87 30 29 22

16,900 170 355 0.48 0.86 29 28 21
15,300 200 210 0.95 0.78 18 18 12
1285 17,400 230 300 0.77 0.81 22 22 15
18,700 230 295 0.78 0.81 22 22 15
1235 34,100 210 300 0.70 0.81 22 22 15
1160 21,700 230 295 0.78 0.81 19 19 13
32,300 220 295 0.75 0.81 19 19 14
Equations used to determine the information in this table are dened in text. Source: Ref 13
Chapter 3: Ferrous and Nonferrous Alloys / 53

bon content between 0.5 and 0.6% (Ref 17). The 4320 steel. Auger electron spectroscopy (Fig.
continued application of cyclic loading at this 12) shows that such intergranular fracture sur-
point then may cause a secondary stage of stable faces have higher concentrations of phosphorus
fatigue crack propagation, characterized by and carbon, in the form of cementite, than do
transgranular fracture, resolvable fatigue stria- transgranular fracture surfaces removed from
tions, and secondary cracking (Ref 12). This prior-austenite grain boundaries (Ref 21, 22).
stage of low-cycle, high-strain fatigue is short Thus, the brittle intergranular fracture that
and gives way to ductile overload fracture of the occurs in stressed high-carbon case microstruc-
core. tures of carburized steels is associated with the
combined presence of segregated phosphorus
Intergranular Fracture and Fatigue
and cementite at prior-austenite grain bound-
Intergranular fracture at the prior-austenite aries. These grain boundary structures are pres-
grain boundaries of high-carbon case micro- ent in as-quenched specimens and do not
structures dominates bending fatigue crack ini- require tempering for cementite formation, as is
tiation and unstable crack propagation of direct- typical in the intergranular mode of tempered
quenched carburized steels. The intergranular martensite embrittlement in medium-carbon
cracking may be associated with other micro- steels (Ref 23). This embrittlement, termed
structural features, such as the surface oxides quench embrittlement, is found in quenched
generated by gas carburizing, but it generally steels with carbon contents as low as 0.6% (Ref
extends much deeper into a carburized case than 24). There is evidence that phosphorus segrega-
the oxide layer. Several studies have docu- tion stimulates the formation of the grain
mented bending fatigue crack initiation by in- boundary cementite (Ref 22).
tergranular fracture even in the absence of sur- The higher the phosphorus content of a car-
face oxidation, where, for example, the oxidized burized steel, the lower its bending fatigue resis-
surface has been removed by chemical or elec- tance and case fracture toughness. Figure 13
tropolishing (Ref 18) or no oxidation is present shows S-N curves for a series of gas-carburized
because the specimens were vacuum or plasma and direct quenched modied 4320 steels with
carburized (Ref 19). systematic variations in phosphorus content
Figure 10 shows an example of intergranular from 0.031 to 0.005% (Ref 22). Endurance lim-
fatigue crack initiation in a direct-quenched its and low-cycle fatigue resistance increase
specimen of gas-carburized type 8719 steel (Ref with decreasing phosphorus content, but little
20). There is a shallow zone of surface oxida- difference is noted between the performance of
tion, about 10 m in depth, but the intergranular the 0.005 and 0.017% phosphorus specimens.
cracking extends much deeper into the speci- All of the specimens, even those with the lowest
men. Figure 11 shows extensive intergranular phosphorus content, failed by intergranular ini-
cracking in the unstable crack propagation zone tiation of fatigue cracks.
in the case of a direct-quenched, gas-carburized

Fig. 10 Intergranular bending fatigue crack initiation at the Fig. 11 Intergranular fracture in case unstable crack propa-
surface of a gas-carburized and direct-cooled SAE gation zone in gas-carburized and direct-cooled SAE
8719 steel specimen. Source: Ref 20 4320 steel
54 / Gear Materials, Properties, and Manufacture

Fig. 12 Auger electron spectra from case fracture surfaces of carburized 8620 steel. (a) From transgranular fracture surface. (b) From
intergranular fracture surface. No phosphorus peak is detectable in the spectrum produced from the transgranular fracture,
and a small phosphorus peak, clearly shown by the 10 magnication, is produced from the intergranular fracture surface. These obser-
vations are consistent with other investigations that show that phosphrous segregates to austenite grain boundaries during austenitizing.
Source: Ref 21
Chapter 3: Ferrous and Nonferrous Alloys / 55

The bending endurance limits of gas-carbur- deoxidation, shrouding of liquid steel to prevent
ized specimens in which fatigue is initiated by reoxidation, vacuum degassing, argon blowing,
intergranular fracture typically range between and desulfurization, are designed to substan-
1050 and 1260 MPa (152 and 183 ksi) (Ref 25, tially increase the cleanliness of steels by
26). This range is based on studies of cantilever lowering the number and/or modifying the mor-
bend specimens with good surface nish, phology of inclusions.
rounded specimen corners, nominal amounts of In carburized steels, the very high strength of
surface oxidation, and loading at R = 0.1. Vari- the carburized case makes the plastic zone ahead
ations within this range may be due to variations of surface discontinuities (such as machining
in austenitic grain size, inclusion contents, marks), aws, or cracks very small, and there-
retained austenite content, or residual stresses, fore, in clean steels, distributed inclusions play a
as discussed subsequently. Nevertheless, the smaller role in fracture than, for example, uni-
common mechanism of bending fatigue crack formly distributed and closely spaced grain
initiation of direct-quenched specimens is inter- boundary embrittling structures or surface ox-
granular fracture at embrittled grain boundaries ides. In other words, the high stresses in the plas-
in a microstructure of LTT martensite and tic process zone have a much higher probability
retained austenite. of acting on grain boundaries or surface oxides
Carburized steels with high nickel content do than on widely spaced inclusions.
not appear to be as susceptible to intergranular Although inclusions often play a secondary
cracking as steels with low nickel content (Ref role in the fatigue of gas-carburized steels, espe-
27). Also, major changes in the case microstruc- cially when there are other microstructural
tures of carburized steel, such as those produced causes of fatigue crack initiation, they may be
by reheating, result in bending fatigue crack ini- involved in the fatigue process in several ways.
tiation sites other than embrittled prior-austenite Inclusions in the carburized steel may either
grain boundaries. Microstructural conditions combine with other features that initiate fatigue
that produce fracture initiation other than by cracks, or in the absence of such features, serve
intergranular cracking are also described herein. as the sole source of fatigue crack initiation.
An example of the rst type of effect was
Inclusions and Fatigue demonstrated in a study that examined the
Inclusionsphases formed between metallic effects of systematic variations in sulfur content
elements and nonmetallic elements such as sul- on the bending fatigue resistance of a gas-
fur and oxygenare an important microstruc- carburized low alloy steel (Ref 20). Sulfur com-
tural component of steels. Coarse or high densi- bines with manganese to form manganese sul-
ties of inclusions initiate fracture and lower the de (MnS) inclusions in steel. The MnS
toughness of steels. As a result, modern steel- particles are plastic during hot rolling, and as a
making practices, which incorporate improved result, are elongated in the rolling direction.
This elongation imparts an anisotropy to the
mechanical properties of the steel, which makes
the effect of inclusions a function of the orienta-
tion of the particles relative to the direction of
the applied load.
S-N (stress vs. life) curves for specimens of a
gas-carburized SAE 8219-type steel with three
levels of sulfur are plotted in Fig. 14. The
endurance limit decreases with increasing sulfur
content. A number of the specimens with the
higher sulfur contents showed runouts at 10 mil-
lion cycles at stress levels higher than the
endurance limits shown, but the specimens that
failed at the lower stresses were used to establish
the endurance limits. In these specimens, elon-
gated MnS inclusion particles that happened to
Fig. 13 Effect of phosphorus content on the bending fatigue be close to the specimen surfaces were associ-
of direct-quenched, gas-carburized modied 4320
steel with 0.005, 0.017, and 0.031 wt% phosphorus, as marked. ated with the fatigue fracture initiation. Fatigue
Source: Ref 22 fracture initiation of the direct-quenched, gas-
56 / Gear Materials, Properties, and Manufacture

carburized specimens was still dominated by size of carburized steels correlates strongly with
intergranular fracture, but if suldes were pres- bending fatigue resistance. Generally, the ner
ent at the highly stressed surfaces of the bending the prior-austenite grain size, the better the
fatigue specimens, they apparently provided an fatigue performance. For example, Fig. 16 shows
extra source of stress concentration and reduced a direct relationship of bending fatigue en-
fatigue performance. durance limit on prior-austenitic grain size, plot-
An example of MnS particles associated with ted as the inverse square root of the grain size, for
fatigue crack initiation in a carburized 8219-type several sets of carburized 4320 steels (Ref 28).
steel is shown in Fig. 15. Although fatigue resis- The renement of austenite grain size has
tance is lowered somewhat by the presence of several effects on the case microstructure of car-
increased densities of MnS particles, the de- burized steels. A ner prior-austenite grain size
crease may be outweighed by a gain in machin- produces a ner martensitic microstructure on
ability associated with higher levels of sulfur. quenching, and therefore, raises the strength of
the carburized case. Increases in strength are
Austenitic Grain Size and Fatigue benecial to high-cycle fatigue resistance, as
discussed previously in the section on bending
Effect of Fine Grain Size on Microstruc- fatigue testing.
tures and Properties. Prior-austenite grain Another very important consequence of ne
austenitic grain size is the dilution of the grain
boundary segregation of phosphorus. In fact,
very ne austenitic grain sizes can eliminate the
sensitivity of high-carbon case microstructures
to intergranular fracture. As a result, other mech-
anisms of fatigue crack initiation replace inter-
granular cracking, generally to the benet of
fatigue performance. The high values of en-
durance limits shown for the very ne grain spec-
imens in Fig. 16 were associated with fatigue
crack nucleation at surface oxidation, not at em-
brittled prior-austenite grain boundaries. Al-
though, as discussed below, surface oxides form
on austenite grain boundaries and fatigue cracks
may nucleate on the oxide-covered austenite
Fig. 14 S-N curves determined by bending fatigue of a gas- boundaries, ne-grain specimens show no inter-
carburized SAE 8219-type steel containing 1.40 Mn,
0.61 Cr, 0.30 Ni, 0.20 Mo, and three levels of sulfur. Source: granular fracture below the oxidized surface
Ref 20 layers.

Fig. 15 Elongated manganese sulde particles associated

with bending fatigue crack initiation at the surface of Fig. 16 Endurance limits as a function of prior-austenite
a gas-carburized SAE 8219-type steel. SEM photomicrograph. grain size from various studies of bending fatigue of
Source: Ref 20 gas-carburized 4320 steels. Source: Ref 28
Chapter 3: Ferrous and Nonferrous Alloys / 57

Reheat Treatments to Achieve Fine Grain reduces the amount of retained austenite in the
Size. The most effective way to produce very as-quenched case microstructures. Reduced lev-
ne grains in carburized steels is by slow cooling els of retained austenite raise the strength of
and reheating of carburized parts at temperatures case microstructures, and therefore, also may
below the Acm where austenite and cementite are contribute signicantly to the improved high-
stable. The cementite particles effectively retard cycle fatigue performance of ne-grain, inter-
austenite grain growth and reduce the carbon critically reheated carburized steels.
content of the austenite. Specimens reheated to
Surface Oxidation and Fatigue
above the Acm may show grain renement,
depending on the temperature of heating, but The surface oxidation produced during gas
because all carbides are dissolved, grain size carburizing may or may not signicantly reduce
renement is not as effective as in specimens bending fatigue resistance. The most severe
heated below the Acm. Figure 17 shows aus- effects of such oxidation are associated with a
tenitic grain size as a function of distance from reduction in near-surface case hardenability,
the carburized surface of gas-carburized 4320 which results from the removal of chromium,
steel specimens in the direct-quenched condition manganese, and silicon from solution in the
and after one and three reheating treatments (Ref austenite by the oxide formation (Ref 29, 30).
28). The reheat treatments very effectively The reduced case hardenability can cause non-
reduce the near-surface case grain size where martensitic microstructures, such as ferrite, bai-
carbon content is the highest, and therefore, the nite, and pearlite, to form at the surface of the
greatest density of carbide particles is retained carburized steel. Not only is the surface hard-
during intercritical reheating. ness reduced, but the residual surface stresses
Intercritical-temperature reheat treatments of may become less compressive or even tensile.
carburized steels produce very ne austenitic Figure 18 and 19 show the effects of surface
grain sizes and high endurance limits. Typically oxidation with reduced hardenability on the
the endurance limits are above 1400 MPa (203 bending fatigue and residual stresses of 8620 and
ksi) (Ref 28). However, the benecial effects of 4615 gas-carburized specimens (Ref 31). The
the reheating on bending fatigue resistance are 4615 steel has higher hardenability by virtue of
not due to grain size renement alone. The higher nickel and molybdenum contents and a
reduced carbon content of the austenite when lower sensitivity to surface oxidation by virtue
carbides are retained raises Ms temperatures and of reduced manganese and chromium contents

Fig. 17 Prior-austenite grain size as a function of depth from the surface of gas-carburized 4320 specimens in the as-carburized,
direct-quenched condition and reheated conditions. Source: Ref 28
58 / Gear Materials, Properties, and Manufacture

(Ref 31). As a result of the different chemistries, gas-carburized and reheated specimen of 4320
the 8620 steel formed pearlite in the near-surface steel. The initiation is conned to the oxidized
regions of the case, while the microstructure of zone, and stable transgranular fatigue propaga-
the 4615 steel, despite some oxidation, consisted tion proceeds directly below the oxidized zone
only of plate martensite and retained austenite at with no evidence of intergranular fracture.
the surface (Ref 31). These differences in micro-
structures due to surface oxidation and reduced
case hardenability are consistent with the differ- Retained Austenite and Fatigue
ences in bending fatigue performance and resid- Retained austenite is one of the most impor-
ual stresses shown between the two steels in Fig. tant microstructural components in the case of
18 and 19. This study illustrates the importance carburized steels. The amounts of retained
of steel chemistry on controlling surface oxida- austenite vary widely, depending on carbon and
tion and the associated formation of nonmarten- alloy content, heat-treating conditions, and spe-
sitic microstructures in gas-carburized steels. cial processing steps such as shot peening and
Another approach used to reduce surface oxide subzero cooling. Generally, the higher the car-
formation in steels with low hardenability is to bon and alloy content, the lower the Ms temper-
use more severe quenching with higher cooling ature and the higher the retained austenite con-
rates. tent in the microstructure. The low-temperature
If the hardenability of a steel is sufcient to tempering applied to carburized steels, gener-
prevent the formation of nonmartensitic micro- ally performed at temperatures below 200 C
structures for a given gas carburizing and (400 F), is not high enough to cause the re-
quenching schedule, surface oxidation has a tained austenite to transform, and therefore,
much reduced effect on bending fatigue per- retained austenite remains an important compo-
formance. In direct-quenched specimens, as dis- nent of the microstructure. At higher tempering
cussed previously and demonstrated in Fig. 10, temperatures, retained austenite transforms to
intergranular fracture to depths much deeper cementite and ferrite with attendant decreases in
than the oxidized layers dominates fatigue crack hardness and strength as the martensitic micro-
initiation. However, when the conditions for structure coarsens.
intergranular crack initiation are minimized, as, The role that retained austenite plays in the
for example, by reheating or shot peening, the bending fatigue performance of carburized
surface oxide layers become a major location for steels has been difcult to identify because of
bending fatigue crack initiation. Figure 20 the variable loading conditions that may be ap-
shows crack initiation in the oxidized zone of a plied to carburized machine components and
the complicating effects of other factors, such as

Fig. 18 S-N curves for direct-quenched gas-carburized 4615

and 8620 steels, notched 4-point bend specimens.
Compositions of the steels are given in Table 1. Non-martensitic Fig. 19 Residual stress as a function of depth below the sur-
transformation products were present on the surfaces of the 8620 face of the direct-quenched gas-carburized 4615
steel specimens and absent on the 4615 steel specimens. Source: and 8620 steel specimens described in Fig. 18. E-Polish = elec-
Ref 31 trolytic polish. Source: Ref 31
Chapter 3: Ferrous and Nonferrous Alloys / 59

residual stresses, grain boundary embrittling and hardness that retained austenite contributes
structures, and surface oxidation. With respect to a composite LTT martensite/austenite case
to loading conditions, it appears that higher microstructure. In addition, retained austenite,
amounts of retained austenite are detrimental to at sufciently high applied strains and stresses,
high-cycle fatigue and reduce endurance limits undergoes deformation-induced transformation
(Ref 10, 14) while higher amounts of retained to martensite (Ref 35). The volume expansion
austenite are benecial for low-cycle, high- associated with the strain-induced formation of
strain fatigue (Ref 14, 32, 33). martensite creates compressive stresses (Ref
Reduced retained austenite contents of LTT 21) that lead to reduced rates of fatigue crack
martensite/austenite composite microstructures growth, accounting for the enhanced low-cycle
increase elastic limits and yield strengths (Ref fatigue performance that is observed in carbur-
34), and therefore, benet stress-controlled, ized steels with high amounts of retained
high-cycle fatigue. One of the approaches to austenite in the case (Ref 33).
reducing the retained austenite content in the
Subzero Cooling and Fatigue
case microstructures of carburized steels, is to
reheat carburized specimens to temperatures Cooling carburized steel below room temper-
below the Acm and quench. Invariably such re- ature is a processing approach sometimes used
heating and quenching signicantly increases to reduce the retained austenite content in the
bending fatigue endurance limits compared to case regions of carburized steels. The transfor-
direct-quenched specimens of identically car- mation of austenite to martensite is driven by
burized specimens (Ref 28). The reheating not temperature changes, and the low Ms tempera-
only reduces retained austenite but also renes tures of high-carbon case regions of carburized
the austenitic grain size, renes the martensitic alloy steels limit the temperature range between
structure, and reduces susceptibility to inter- Ms and room temperature over which martensite
granular fractureall features that are known to forms. Therefore, the temperature range for
improve fatigue resistance. Therefore, im- martensite formation and the reduction of re-
proved high-cycle fatigue resistance of reheated tained austenite is extended by cooling below
and quenched specimens is related to a combi- room temperature. The cooling treatments are
nation of microstructural changes, including variously referred to as subzero cooling, refrig-
low retained austenite contents. eration treatments, or deep cooling.
The benet of retained austenite to strain- In addition to the effects of retained austenite
controlled, low-cycle bending fatigue is related on bending fatigue, as discussed, any deforma-
to the improved ductility and reduced strength tion-induced transformation of retained austen-

Fig. 20 Bending fatigue crack initiation in gas-carburized and reheated 4320 steel. The dashed line corresponds to maximum depth
of surface oxidation, and all fracture below dashed line is transgranular. Source: Ref 28
60 / Gear Materials, Properties, and Manufacture

ite during cyclic loading in service, because of decreased. However, the residual stresses in the
the volume expansion that accompanies the austenite phase are measured to be tensile, espe-
transformation of austenite to martensite, may cially at the surface of the carburized specimens.
change the dimensions of a carburized compo- These tensile stresses would then be expected to
nent. Therefore, subzero cooling is one ap- lower the surface tensile stresses applied in bend-
proach to reduce retained austenite in parts that ing to initiate fatigue cracks. Microcrack forma-
require high precision and stable dimensions tion within martensite plates and at plate/austen-
throughout their service life. However, several ite interfaces may be enhanced by the localized
studies show that subzero cooling lowers the residual stresses induced by subzero cooling
bending fatigue resistance of carburized steels. (Ref 41), but they could be minimized by main-
Nevertheless, high-quality, high-performance taining a ne prior-austenite grain size and
aircraft and helicopter gears are routinely sub- applying reheating treatments (Ref 42, 43).
jected to subzero cooling without apparent detri-
mental effects (Ref 36). For example, a com- Residual Stresses,
monly used carburizing steel for aircraft gears Shot Peening, and Fatigue
is 9310, which contains about 3 wt% Ni (see Compressive residual stresses are formed in
Table 1). The high nickel content lowers the Ms the case microstructures of carburized steels as
temperature and increases the amount of aus- a result of transformation and temperature gra-
tenite at room temperature. The austenite con- dients induced by quenching (Ref 44, 45). The
tent can be reduced by subzero cooling, probably magnitude and distribution of the residual
with adverse effects on localized residual stress. stresses, therefore, are complex functions of the
However, the latter adverse effect of subzero temperature gradients induced by quenching
cooling may be offset by ne austenite grain (Ref 46), which in turn are dependent on speci-
size, and high nickel content may improve the men size and geometry, the hardenability of the
fracture toughness and fatigue resistance of car- steel, the carbon gradient, and the case depth.
burized steels (Ref 37) to a level where fatigue The residual stresses as a function of case depth
resistance is not adversely affected by subzero are routinely measured by x-ray diffraction, and
cooling. considerable effort has been applied to model-
In the low-alloy steels commonly used for car- ing residual stress proles in carburized steels
burizing, subzero cooling used to reduce re- as a function of cooling and hardenability (Ref
tained austenite may reduce the bending fatigue 47, 48).
resistance of carburized components. If refriger- Figure 22 shows the range and pattern of com-
ation treatments are applied, parts should be pressive residual stresses typically formed in the
tempered both before and after. Figure 21 shows case regions of direct-cooled carburized steels
an example of the detrimental effects of subzero (Ref 49). The compressive residual stresses off-
cooling on the bending fatigue resistance of car- set the adverse effects of factors such as quench
burized specimens. The data were produced in embrittlement and intergranular fracture to
an experimental study of vacuum-carburized
specimens of 8620 and EX 24 steel that were
deep cooled to 196 C (321 F) in liquid nitro-
gen (Ref 38). The overall fatigue performance in
this study was complicated by high retained
austenite contents and coarse carbide particles at
square specimen corners, but the detrimental
effect of subzero cooling on bending fatigue per-
formance is clearly demonstrated in Fig. 21.
The detrimental effect of subzero cooling on
the bending fatigue of carburized specimens has
been related to changes in residual stress by sev-
eral investigations (Ref 3941). The overall sur-
face residual stresses become increasingly com-
pressive, as measured from the martensite in the Fig. 21 S-N curves of vacuum-carburized 8620 and EX 24
case and as expected from the constraint of the (0.89% Mn, 0.24% Mo, 0.55% Cr) steels. The lower
curves were obtained from specimens subzero cooled to 196
expansion that accompanies the transformation C, and the upper curves were obtained from specimens not sub-
of austenite to martensite as temperature is jected to subzero cooling. Source: Ref 38
Chapter 3: Ferrous and Nonferrous Alloys / 61

which high-carbon microstructures are suscepti- deviation from at is measured by an Almen

ble (Ref 24), and they increase the fracture and gage. This measurement is called arc height.
fatigue resistance of direct-quenched parts to Details of the design and use of Almen gages
levels that provide good engineering perfor- and test strips are given in SAE standards J442,
mance. As discussed earlier, case residual Test Strip, Holder and Gage for Shot Peening,
stresses in carburized steels are adversely modi- and J443, Procedure for Using Shot Peening
ed by subzero cooling, and they are positively Test Strip. Additional information on the shot
modied (made locally more compressive) by peening process can also be found in Chapter 9,
the strain-induced transformation of austenite to Carburizing. Refer to the section Shot Peen-
martensite. Tempering lowers residual compres- ing of Carburized and Hardened Gears and in
sive stresses because of dimensional changes particular Fig. 62 in Chapter 9 which illustrates
that accompany the recovery and coarsening of the shot peening test setup.
the martensitic microstructure during tempering Figure 23 shows the data obtained from the
(Ref 50). fatigue life versus Almen intensity study on car-
The fatigue lives of carburized steel gears can burized 4023 steel. The gears were peened to
be signicantly improved by shot peening, three Almen intensity A levels (0.305, 0.457,
which is a method of cold working in which and 0.609 mm, or 0.012, 0.018, and 0.024 in.)
compressive stresses are induced in the exposed using steel shot in the size of S280, and to one
surface layers of metallic parts by the impinge- Almen intensity C level of 0.008 in. with S280
ment of a stream of shot. The shot, which may shot. As Fig. 23 clearly shows, the mean fatigue
be composed of cast steel, cast iron, or glass life of unpeened gears was about 9800 cycles
beads, is directed at the metal surface at high while optimized shot peening procedures (24A
velocity under controlled conditions. Details on Almen intensity) resulted in a mean fatigue life
the process can be found in the article Shot exceeding 33,000 cycles. The improved fatigue
Peening in Volume 5 of the ASM Handbook.
The effect of shot peening process parame-
ters on fatigue life of carburized helical gears
has been investigated (Ref 51). Low-cycle gear-
tooth bending fatigue lives were evaluated as a
function of Almen intensity parameter. Almen
intensity is an indirect method of monitoring the
aggregate amount of energy transfer or residual
compressive stresses imparted to the workpiece.
The method uses a strip of SAE-AISI 1070
spring steel in the 44 to 50 HRC range. The strip
is mounted to an Almen test block and exposed
to the blast in identical fashion as the critical
area of the workpiece, with the same parame-
ters. After peening, the Almen strip is removed
from the block and the amount of maximum

Fig. 23 Effect of a shot peening parameter (Almen intensity

Fig. 22 Ranges and patterns of residual stresses as a function as described in text) on the low-cycle fatigue of a
of depth for 70 carburized steels. Source: Ref 49 carburized helical gear made from 4023 steel. Source: Ref 51
62 / Gear Materials, Properties, and Manufacture

life for 24A peened gears was attributed to the ture of the rolling element and catastrophic fail-
highest compressive residual stress incurred. ure occurs. Fractured races can result from
fatigue spalling and high hoop stresses.
Extreme cases of spalling are associated with
Other Properties of case crushing or cracking initiated at the case-
Interest for Carburized Steels core interface. If sliding is coupled with contact
loading, surface pits develop. Very high contact
Although much of the recent research on prop- loads cause microstructural changes within
erties of carburized steels has centered around high-carbon martensite that are revealed by var-
bending fatigue (see previous section), there are ious types of etching (Ref 5254). Generally
other properties that effect the service life of car- retained austenite is regarded as a microstruc-
burized components. As will be described forth- tural constituent that is benecial for rolling-
with, these mechanical properties are strongly contact fatigue resistance (Ref 55, 56).
inuenced by core and case microstructure, case One of most effective means for improving
depth, residual stresses, and alloy chemistry. the rolling contact fatigue life of carburized
steels is by utilizing clean steel technology. Sec-
Rolling Contact Fatigue
ondary metallurgical techniques in ladle ren-
Rolling contact fatigue is a surface-pitting- ing and teeming practice have become the key
type failure commonly found in gears (see elements in improved steelmaking practices.
Chapter 13, Gear Failure Modes and Analy- Important developments include degassing and
sis, for details). Rolling contact fatigue differs deoxidation practices, improved temperature
from classic structural fatigue (bending or tor- control, inert gas shrouding, improved refracto-
sional) in that it results from contact or Hertzian ries, and the use of bottom-poured or signi-
stress state. This localized stress state results cantly improved tundish systems, and large
when curved surfaces are in contact under a nor- cross-section continuous casters. Figure 24
mal load. Generally, one surface moves over the demonstrates the improvements in rolling con-
other in a rolling motion. The contact geometry tact fatigue life of carburized steels as steelmak-
and the motion of the rolling elements produces ing practices have improved.
an alternating subsurface shear stress. Subsur-
face plastic strain builds up with increasing
cycles until a crack is generated. The crack then Wear Resistance (Ref 58)
propagates until a pit is formed. Once surface Another important property afforded by the
pitting has initiated, the bearing becomes noisy carburizing process is wear resistance resulting
and rough running. If allowed to continue, frac- from the high hardness of the case. As with bend-

Present 1980
Vacuum carbon Original bottom
deoxidation pour
Precipitation Recent bottom
Fatigue life (millions of revolutions)

deoxidation pour
Precipitation Vacuum-arc
100 deoxidation, remelted
plus shrouding


0.0001 0.001 0.01 0.1 1 10
Tool length of inclusion stringers, mm/cm3

Fig. 24 Cleanness and rolling contact fatigue life improvements in carburized steels as steelmaking practices have changed. Source:
Ref 57
Chapter 3: Ferrous and Nonferrous Alloys / 63

ing fatigue, there are a number of microstruc- ing the case carbon content, not by subzero
ture/property relationships that must be consid- quenching after carburizing or by tempering
ered for wear-resistant applications. at temperatures above 200 C (390 F).
Microstructural Features. The indepen- The tempering temperature chosen should
dent variables available for controlling the mi- be as low as possible, but above the surface
crostructure/properties of carburized cases are temperatures anticipated in nishing opera-
those that dene the carburizing alloy (composi- tions and in service.
tion, cleanliness) and those that dene the The content of nonmetallic inclusions
carburizing process (time/temperature/carbon- should be no higher than that needed for
potential carburizing history, time/temperature economical machining.
quenching history, time/temperature tempering Coarse primary carbides can be helpful in
history). These tools provide a considerable resisting abrasive wear. Fine primary car-
degree of control over these microstructural bides can permit more retained austenite at
features: the same hardness level. Experiments should
be conducted to verify any benets presumed
Martensite to be associated with primary carbides.
a. Carbon content of source austenite
b. Plate size (austenite grain size) Hot Hardness
c. Strength Retention of hardness at elevated tempera-
d. Secondary hardening tures (hot hardness) is vitally important in appli-
Primary carbides cations where high local temperature conditions
a. Size can be encountered. Examples include helicop-
b. Volume fraction ter gears, speed reduction gear sets, and turbine
gearing. Figure 25 provides hot hardness data
Retained austenite for several carburized steels. It is evident from
a. Volume fraction these data that higher alloy content is needed to
b. Carbon content assure sufcient hardness at temperatures above
Nonmetallic inclusions 315 C (600 F) encountered in severe service.
and these global features:
Bending Strength and Bend Ductility
Case depth
In service, carburized and hardened steels are
Residual stress distribution
subjected to bending loads and must be able to
which determine the tribological properties of
the case. The combination of properties that is
best for each application must then be decided.
The necessary case depth and case hardness
can be estimated from a Hertzian stress calcula-
tion, but other microstructural objectives can be
specied only qualitatively. For many applica-
tions, the following rules of thumb apply:

Sufcient case depth and case hardness must

be provided to prevent indentation or case
crushing under the anticipated contact loads.
For gears loaded in line contact, a mini-
mum case hardness of 58 HRC frequently is
specied. When high contact loads are ac-
companied by sliding, the near-surface hard-
ness (to a depth of about 50 m, or 2 mils)
may have to be raised to prevent shearing of
surface layers.
Fig. 25 Hot hardness of three carburized steels. The dashed
The retained austenite content should be as line corresponds to a surface hardness of 58 HRC.
high as possible, consistent with the require- Compositions for SAE 9310 and CBS 1000M are listed in Table
1. The nominal composition for steel D is 0.12% C, 0.5% Mn,
ments of the previous rule. The retained aus- 1.1% Si, 1.0% Cr, 2.0% Ni, 2.3% Mo, and 1.2% V. Source: Ref
tenite content should be controlled by adjust- 59
64 / Gear Materials, Properties, and Manufacture

resist design loads and overloads without frac- attempts have been made to simulate service
ture. A variety of laboratory tests have been per- loads, single- as well as multiple-load applica-
formed to dene the resistance of carburized tions, different rates of load application, and to
and hardened steels to failure under bending include realistic details in the test specimen
loads, and to provide information on the contri- such as changes in notch severity or the use of a
bution of alloys in resisting failure. part such as a gear tooth as the specimen. Re-
Bend ductility of several carburized and sults of one such test, the impact fracture test,
hardened steels over a range of test tempera- will be described in this section. Additional test
tures from room temperature to 195 C (320 results are reviewed in Ref 4.
F) is given in Fig. 26. A comparison of the data Impact Fracture Strength. The specimen
for carburized SAE 4817 steel with those for shown in Fig. 27 was used by Cameron and oth-
SAE 4027 (both with about the same amount of ers (Ref 61) for single-blow impact tests as well
retained austenite) shows that the steel alloyed as fatigue tests. This simulated gear specimen
with a substantial amount of nickel exhibited was loaded in a standard pendulum type testing
much greater ductility. machine, with the specimen held in a vertical
Razim (Ref 60) has observed that the static position, similar to an Izod impact test. An
bend test is useful in evaluating the ability of a instrumented tup permitted recording the energy
carburized and hardened surface zone to sustain absorbed by the specimen as a function of time
plastic deformation without cracking. A suit- during the test. For carburized-and-hardened
able measurement can be the initial crack steels, the investigators found that alloy content
strength. His summary of such tests indicates and core carbon content signicantly inuenced
that with low surface carbon contents of about resistance to fracture under impact bending con-
0.6% C, the initial crack strength increases with ditions. In Fig. 28 the chromium-molybdenum
increasing core strength; while with high sur- steels exhibited higher fracture strengths than
face carbon contents (about 1.2% C), the initial the manganese-chromium steels, but the nickel-
crack strength becomes less dependent on core chromium-molybdenum steel, PS55, not only
strength, but drops to a considerably lower level exhibited much higher fracture strength, but
than the crack strength exhibited by the steels fracture strength did not decrease with increas-
with 0.6% C at the surface. ing carbon content.
In another study by Smith and Diesburg (Ref
62), an attempt was made to dene in more
detail the effect of alloy interactions on fracture
Toughness in carburized steels can be mea- strength using the simulated gear-tooth speci-
sured in a variety of ways. In recent years, men shown in Fig. 27. Figure 29 summarizes

Fig. 26 Bend ductility transition curves for carburized and hardened steels. Nominal alloy contents of the steels are listed within
the diagram. Source: Ref 59
Chapter 3: Ferrous and Nonferrous Alloys / 65

Fig. 27 Test specimen used by Cameron for fatigue, bend, and impact fracture tests. Typical results are shown in Fig. 28 and 29.
Source: Ref 61

Fig. 28 Effect of core carbon content and alloy content on impact fracture strength of a series of steels, carburized at 925 C (1700
F), cooled to 840 C (1550 F), oil quenched, and tempered at 150 C (340 F). Source: Ref 61
66 / Gear Materials, Properties, and Manufacture

the effect of molybdenum and nickel on fracture all depths below the surface. Close to the car-
strength for several different groups of steels, burized surface, only small differences could be
each modied by adding molybdenum. Each observed among the steels tested.
group of steels was carburized to have similar
case depths and surface carbon contents, then
the steel specimens were cooled from the car- Improved Elevated-Temperature
burizing temperature to 845 C (1550 F), Performance (Ref 58)
quenched in oil at 65 C (150 F), and then tem- Several secondary hardening carburizing
pered at least 1 h at 170 C (340 F). It is evident alloys have been developed for applications that
that nickel in amounts greater than 0.5% is very require resistance to elevated temperatures,
effective in increasing fracture strength, espe- such as helicopter gearing. (See the Special
cially when molybdenum content is greater alloys listed in Table 1.) These alloys make use
than 0.25%. The investigators also compared of the precipitation of copper and/or M2C and
vacuum-melted steels with air-melted steels MC carbides to provide resistance to softening
and obtained higher fracture strengths with the for temperatures up to 550 C (1020 F).
vacuum-melted steels. Because they contain substantial amounts of
Fracture Toughness Data. Work by Dies- molybdenum and vanadium, these alloys re-
burg (Ref 63) included a study of fracture semble low-carbon versions of tool steels. Some
toughness of several steels carburized to pro- of the alloys are difcult to carburize because of
duce case depths (at 0.5% C) between 0.75 and high silicon and chromium contents; preoxida-
1.00 mm (0.030 and 0.040 in.) thick. Results tion prior to carburizing is necessary to permit
shown in Fig. 30 indicate that the plane-strain carbon penetration. Secondary hardening alloys
fracture toughness (KIc) values for the higher could also be useful in ambient-temperature
nickel steels SAE PS55, PS32, 9310, and 4820 applications in which lubrication is marginal,
are quite similar, much higher than the values because the heat generated by intermittent
observed for the lower nickel SAE 8620 steel at metal-to-metal contact would not readily soften
the underlying metal.
One of the Special Alloys listed in Table 1 is
Pyrowear 53 (UNS K71040) which is used for
helicopter transmission drive gears. The higher
copper content increases core impact strength,
while the increased molybdenum provides im-
proved elevated-temperature properties. The re-
duced chromium levels enhance carburizability,
while the vanadium maintains a ne grain size.
When compared to conventional grades, e.g.,
SAE-AISI 9310, Pyrowear Alloy 53 exhibits
superior temper resistance and high case hot
hardness while maintaining high core impact
strength and fracture toughness (126 to 132
MPa 1m , or 115 to 120 1in. ). Table 8 lists
core mechanical properties for this proprietary
carburizing grade.

Other Ferrous Alloys for Gears

In addition to wrought steels, the following

iron-base alloys are also used for gears:
Fig. 29 Impact fracture strength of carburized steels contain-
ing various combinations of molybdenum and Cast carbon and alloy steels
nickel. Open data points are for vacuum-melted heats; solid data
points are for air-melted heats. Compositions were those of stan-
Gray and ductile cast irons
dard steels to which molybdenum was added in varying P/M irons and steels
amounts, as indicated in the accompanying table. Source: Ref 62 Stainless steels
Chapter 3: Ferrous and Nonferrous Alloys / 67

Tool steels Steel castings can be made from any of the

Maraging steels many types of carbon and alloy steel produced in
wrought form. Those castings produced in any
Each will be discussed in subsequent sections. of the various types of molds and wrought steel
of equivalent chemical composition respond
similarly to heat treatment, have the same weld-
Cast Steels
ability, and have similar physical and mechani-
Steel castings are produced by pouring cal properties. However, cast steels do not
molten steel of the desired composition into a exhibit the effects of directionality on mechani-
mold of the desired conguration and allowing cal properties that are typical of wrought steels.
the steel to solidify. The mold material may be This nondirectional characteristic of cast steel
silica, zircon, chromite sand, olivine sand, mechanical properties may be advantageous
graphite, metal, or ceramic. The choice of mold when service conditions involve multidirec-
material depends on the size, intricacy, dimen- tional loading.
sional accuracy of the casting, and cost. While The cast steels used for gears are generally
the producible size, surface nish, and dimen- modications of standard SAE-AISI designa-
sional accuracy of castings vary widely with the tions. Common through-hardening cast steels
type of mold, the properties of the cast steel are include 1045, 4135, 4140, 8630, 8640, and
not affected signicantly. 4340. Carburizing grades are usually 1020,

Fig. 30 Fracture toughness in carburized steels as a function of distance below the surface. The SAE PS55, 9310, and 8620 steels
were commercial heats; the SAE PS32 and 4820 steels were laboratory heats. The PS32 and 4820 steels were quenched
directly after carburizing at 925 C (1700 F) into 170 C (340 F) oil; other steels were cooled from 925 C to 840 C (1700 to 1550 F)
before quenching into 65 C (150 F) oil. Data are also shown for 9310 steel that was refrigerated after quenching and before tempering.
Source: Ref 63

Table 8 Core mechanical properties of Pyrowear Alloy 53

Test temperature Yield strength Ultimate tensile strength Charpy V-notch impact
Elongation Reduction
C F MPa ksi MPa ksi in 4D, % of area, % J ft lbf

54 65 1103 160 1331 193 18 63 5356 3941

Room temperature 965 140 1172 170 16 66.5 118129 8795
100 212 ... ... ... ... ... ... 140163 103120
177 350 896 130 1207 175 12 46 155157 114116
Source: Carpenter Technology Corporation
68 / Gear Materials, Properties, and Manufacture

8620, and 4320 types. The compositions of loying with various metallic or nonmetallic ele-
some cast steels are selected by the steel pro- ments, and by varying melting, casting, and heat
ducer in order to achieve the specied proper- treating practices. Two types of cast irons are
ties. For example, the ASTM A 148 Grade 120- used for gears: gray cast irons and ductile cast
95 alloy steel rack segment shown in Fig. 31 irons.
must have a minimum tensile strength of 827 Gray iron refers to a broad class of ferrous
MPa (120 ksi) and a minimum yield strength of casting alloys normally characterized by a
655 MPa (95 ksi). Table 9 lists mechanical microstructure of ake graphite in a ferrous
properties for various heat treated cast steels. matrix (Fig. 32). Gray irons are in essence iron-
carbon-silicon alloys that usually contain 2.5 to
Cast Irons 4% C, 1 to 3% Si, and additions of manganese,
The term cast iron, like the term steel, depending on the desired microstructure (as low
identies a large family of ferrous alloys. Cast as 0.1% Mn in ferritic gray irons and as high as
irons primarily are alloys of iron that contain 1.2% Mn in pearlitics). Sulfur and phosphorus
more than 2% C and from 1 to 3% Si. Wide vari- are also present in small amounts as residual
ations in properties can be achieved by varying impurities.
the balance between carbon and silicon, by al- Gray cast irons used for gears are classied in
ASTM specication A 48 by their tensile
strengths in ksi. As shown in Table 10, they
range from class 20 (minimum tensile strength
of 138 MPa or 20 ksi) to class 60 (minimum ten-
sile strength of 414 MPa or 60 ksi). Generally, it
can be assumed that the following properties of
gray cast irons increase with increasing tensile
strength from classes 20 to 60:
All strengths, including strength at elevated
Ability to be machined to a ne nish
Modulus of elasticity
Wear resistance
On the other hand, the following properties
Fig. 31 Gear rack segment made from cast ASTM A 148
Grade 120-95 alloy steel decrease with increasing tensile strength, so that

Table 9 Properties of cast carbon and low-alloy steels

Tensile strength Yield strength Endurance limit
Class(a) and Reduction in Elongation, Hardness, Endurance
heat treatment MPa ksi MPa ksi area % % HB MPa ksi ratio

Carbon steels
60 A 434 63 241 35 54 30 131 207 30 0.48
65 N 469 68 262 38 48 28 131 207 30 0.44
70 N 517 75 290 42 45 27 143 241 35 0.47
80 NT 565 82 331 48 40 23 163 255 37 0.45
85 NT 621 90 379 55 38 20 179 269 39 0.43
100 QT 724 105 517 75 41 19 212 310 45 0.47
Alloy steels(b)
65 NT 469 68 262 38 55 32 137 221 32 0.47
70 NT 510 74 303 44 50 28 143 241 35 0.47
80 NT 593 86 372 54 46 24 170 269 39 0.45
90 NT 655 95 441 64 44 20 192 290 42 0.44
105 NT 758 110 627 91 48 21 217 365 53 0.48
120 QT 883 128 772 112 38 16 262 427 62 0.48
150 QT 1089 158 979 142 30 13 311 510 74 0.47
175 QT 1234 179 1103 160 25 11 352 579 84 0.47
200 QT 1413 205 1172 170 21 8 401 607 88 0.43
(a) Class of steel based on tensile strength, in ksi. A = Annealed, N = Normalized, NT = Normalized and tempered, QT = Quenched and tempered. (b) Below 8% total
alloy content. Source: Ref 64
Chapter 3: Ferrous and Nonferrous Alloys / 69

low-strength irons often perform better than gears of the same pitch diameter and face width.
high-strength irons when these properties are Their bending strength capacity is about one-
important: third that of steel gears of the same normal
diametral pitch. By making cast-iron gears
somewhat coarser in pitch for the same pitch
Resistance to thermal shock
diameter, it is possible to design cast-iron gear
Damping capacity
sets about the same size as steel gear sets for a
Ability to be cast in thin sections
given application.
Gray cast iron has long been used as a gear Gray cast iron has low impact strength and
material (Ref 65). Cast iron is low in cost and it should not be used where severe shock loads
can be cast easily into any desired shape for the occur.
rim, web, and hub of a gear. Cast iron machines Ductile cast iron, also known as nodular
easily. iron or spheroidal-graphite (SG) cast iron, is
Cast iron gears generally show good resis- cast iron in which the graphite is present as tiny
tance to wear and are often less sensitive to spheres (nodules) (see Fig. 33). In ductile iron,
lubrication inadequacies than are steel gears. eutectic graphite separates from the molten iron
Cast iron has good dampening qualities. during solidication in a manner similar to that
Cast-iron gear teeth have about three-quarters in which eutectic graphite separates in gray cast
of the surface load-carrying capacity of steel iron. However, because of additives introduced
in the molten iron before casting, the graphite
grows as spheres, rather than as akes of any of

Fig. 32 Flake graphite in a pearlitic matrix in a class 30 as-

cast gray iron

Table 10 Minimum hardness and tensile

strength requirements for gray cast irons
Tensile strength

ASTM class number Brinell hardness MPa ksi

20 155 140 20
30 180 205 30
35 205 240 35
40 220 275 40 Fig. 33 Spheroidal graphite in an unetched ductile iron
50 250 345 50 matrix shown at 75 (a) and in the etched (picral)
60 285 415 60 condition shown at 300 (b). Etching reveals that the matrix con-
Source: ASTM A 48
sists of ferritic envelopes around the graphite nodules (bulls-eye
structure) surrounded by a pearlitic matrix.
70 / Gear Materials, Properties, and Manufacture

the forms characteristic of gray iron. Cast iron formation bath. The properties are characterized
containing spheroidal graphite is much stronger by very high strength, some ductility and tough-
and has higher elongation than gray iron. The ness, and often an ability to work harden, which
akes in gray cast iron act as stress risers and gives appreciably higher wear resistance than
make the impact strength and fatigue strength that of other ductile irons. As shown in Fig. 34
low. The nodules in ductile iron exhibit less of a and Table 12, austempered ductile irons exhibit
stress-riser effect. Ductile iron has good impact in excess of 5% elongation at tensile strengths
strength and enough ductility to have elonga- exceeding 1000 MPa (145 ksi).
tions in the range of 2 to 15%, depending on the Most applications for austempered ductile
class. The fatigue strength of ductile iron can iron gears are in transportation equipment: auto-
approach that of steel of equal hardness. mobiles, trucks, and railroad and military vehi-
Most of the specications for standard grades cles. The same improved performance and cost
of ductile iron used for gears are based on prop- savings are expected to make these materials
erties. That is, strength and/or hardness is spec- attractive in equipment for other industries,
ied for each grade of ductile iron, and compo- such as mining, earthmoving, agriculture, con-
sition is either loosely specied or made struction, and machine tools.
subordinate to mechanical properties. As shown
in Table 11, the ASTM system for designating
the grade of ductile iron incorporates the num- P/M Irons and Steels
bers indicating tensile strength in ksi, yield Both pressed-and-sintered and powder
strength in ksi, and elongation in percent. For forged iron-base gears are produced for a vari-
example, grade 80-55-06 has a tensile strength ety of applications. Most P/M gears are made
of 522 MPa or 80 ksi, a yield strength of 379 from iron-copper steels or iron-nickel steels.
MPa or 55 ksi, and an elongation of 6%. These steels can also undergo surface hardening
Ductile iron may be used as-cast or it may be treatments such as carburizing and carbonitrid-
heat treated to a wide range of strength and hard- ing. More detailed information on P/M materi-
ness levels. Common heat treatments include als for gears can be found in Chapter 7, Pow-
annealing, normalizing and tempering, quench- der Metallurgy.
ing and tempering, and austempering (see dis- Sintered Steels. The copper-bearing steels
cussion below). Figure 34 compares the strength in the as-sintered condition have a tensile
and ductility of as-cast ductile iron with heat- strength range of 170 to 570 MPa (25 to 82 ksi),
treated ductile irons. The hardness levels in duc- an elongation of 2.0% or less, and an apparent
tile irons can range from less than 160 HB to hardness of 11 to 80 HRB. Heat-treated sintered
specied ranges greater than 300 HB. copper steels have a tensile strength range of
If ductile iron is austenitized and quenched in 450 to 720 MPa (65 to 105 ksi), an elongation
a salt bath or a hot oil transformation bath at a less than 0.5%, and an apparent hardness of 58
temperature of 320 to 550 C (610 to 1020 F) to 60 HRC. The nickel steels in the as-sintered
and held at this temperature, it transforms to a condition have a tensile strength range of 170 to
structure containing mainly bainite with a minor 620 MPa (25 to 90 ksi), an elongation of 10% or
proportion of austenite (Fig. 35). Irons that are less (average values are about 4%), and appar-
transformed in this manner are called austem- ent hardness of 55 to 88 HRB. Heat-treated sin-
pered ductile irons. Austempering generates a tered nickel steels have a tensile strength range
range of structures, depending on the time of of 590 to 1340 MPa (85 to 195 ksi), and an
transformation and the temperature of the trans- elongation of less than 0.5%.

Table 11 Minimum mechanical property requirements for ductile cast irons

Tensile strength Yield strength
ASTM grade Recommended heat treatment Brinell Elongation in 50 mm
designation and microstructure hardness range MPa ksi MPa ksi (2 in.), % min

60-40-18 Annealed ferritic 170 max. 415 60 275 40 18.0

65-45-12 As-cast or annealed ferritic-pearlitic 156217 450 65 310 45 12.0
80-55-06 Normalized ferritic-pearlitic 187255 550 80 380 55 6.0
100-70-03 Quench and tempered pearlitic 241302 690 100 485 70 3.0
120-90-02 Quench and tempered martensitic Range specied 830 120 620 90 2.0
Source: ASTM A 536
Chapter 3: Ferrous and Nonferrous Alloys / 71

Powder forged nickel steels for gears are amount needed to prevent the formation of rust
produced from 4600-series powders. The ten- in unpolluted atmospheres (hence the designa-
sile strength for quenched-and-tempered pow- tion stainless). Few stainless steels contain more
der forged 4600 materials can exceed 1500 MPa than 30% Cr or less than 50% Fe. They achieve
(220 ksi). Powder forged steels have densities their stainless characteristics through the forma-
approaching those of wrought steels. tion of an invisible and adherent chromium-rich
oxide surface lm. Unfortunately, the thin oxide
Stainless Steels layer on stainless steels is hard and brittle and
breaks up easily under sliding loads, exposing
Stainless steels are iron-base alloys that con-
the bare metal and thus promoting wear. Never-
tain a minimum of approximately 11% Cr, the
theless, stainless steels gears are used in some
applications where corrosion resistance is criti-
cal, for example, food processing machinery.
Austenitic (types 303, 304, and 316), marten-
sitic (type 440C), and precipitation-hardening
(17-4PH and 17-7PH) stainless steels have been
used as gear materials. Type 304 (18Cr-8Ni),
which is one of the most widely used stainless
steels, is used for gears in juice extractors and
shing reels.

Tool Steels
Tool steels are normally used for machining
and metal forming operations. The two types of
tool steels that are occasionally used for gears
are high-speed tool steels and hot-work tool
steels. More detailed information on these mate-
rials can be found in the ASM Specialty Hand-
book: Tool Materials.
Fig. 34 Strength and ductility ranges of as-cast and heat- High-speed tool steels are so named pri-
treated ductile irons marily because of their ability to machine mate-
rials at high cutting speeds. They are complex
iron-base alloys of carbon, chromium, vana-
dium, molybdenum, or tungsten, or combina-
tions thereof, and in some cases substantial
amounts of cobalt. The carbon and alloy con-
tents are balanced at levels to give high attain-
able hardening response, high wear resistance,
high resistance to the softening effect of heat,
and good toughness.
Hot-work tool steels (group H steels) have
been developed to withstand the combinations
of heat, pressure, and abrasion associated with
various metalworking operations such as
punching, shearing, and other forming methods.
Group H tool steels usually have medium car-
bon contents (0.35 to 0.45%) and chromium,
tungsten, molybdenum, and vanadium contents
of 6 to 25%.

Maraging Steels
Maraging steels are a special class of high-
Fig. 35 Austempered ductile iron structure consisting of
strength steels that differ from conventional
spheroidal graphite in a matrix of bainitic ferritic
plates (dark) and interplate austenitic (white) steels in that they are hardened by a metallurgi-
72 / Gear Materials, Properties, and Manufacture

cal reaction that does not involve carbon. Aluminum bronzes

Instead, these steels are strengthened by the pre- Silicon bronzes
cipitation of intermetallic compounds at tem-
peratures of about 480 C (900 F). The term Tin bronzes, also referred to as phosphor
maraging is derived from martensite age hard- bronzes, are copper-tin or copper-tin-lead
ening of a low-carbon, iron-nickel lath marten- (leaded tin bronzes) that are deoxidized with
site matrix. Commercial maraging steels are phosphorus. These alloys are tough and have
designed to provide specic levels of yield good corrosion resistance. They possess excel-
strength from 1030 to 2420 MPa (150 to 350 lent wear resistance which permits their use in
ksi) with some having yield strengths as high as gears and worm wheels for severe wear applica-
3450 MPa (500 ksi). These steels typically have tions. The most commonly used alloy in this
very high nickel, cobalt, and molybdenum con- group is C90700, or gear bronze, which con-
tents and very low carbon contents. The use of tains 89% Cu and 11% Sn. Nominal composi-
maraging steels for gears is discussed in Chap- tions and properties of tin bronzes and leaded tin
ter 8, Through Hardening (refer to the case bronzes are listed in Table 13. Composition and
history that describes the design and manufac- mechanical property requirements for tin
ture of a gear rack). bronzes used for gears are covered in ASTM
standard B 427.
Manganese Bronzes. This is the name
given to a family of high-strength brass alloys
Nonferrous Alloys (brasses are alloys which contain zinc as the prin-
cipal alloying element). They are characterized
Copper-base alloys account for much of the
by high strength and hardness and they are the
nonferrous gear materials used. Die cast alu-
toughest materials in the bronze family. They
minum-, zinc-, and magnesium-base alloys for
achieve mechanical properties through alloying
gears are also produced but these materials have
without heat treatment. These bronzes have the
largely been replaced by plastics for low-load
same strength and ductility as annealed cast
steels. Table 13 lists compositions and properties
for cast manganese bronzes. Table 14 lists the
Bronzes composition and properties of a wrought man-
A family of four bronzes are widely used for ganese bronze containing 36.5% Zn. Cast and
power transmission gearing. Most of these are wrought manganese bronzes have good wear
used in worm gears and are mated against steel resistance but do not possess the same degree of
gears. The four gear bronzes are: corrosion resistance, wearability, or bearing
quality as the tin bronzes or aluminum bronzes.
Tin bronzes Aluminum bronzes are similar to the man-
Manganese bronzes ganese bronzes in toughness, but are lighter in

Table 12 ASTM standard A 897-90 and A 897M-90 mechanical property requirements

of austempered ductile iron
Tensile (min) Yield (min) Impact(a)
Grade MPa ksi MPa ksi Elongation, % J ft lbf Hardness, HB(c)

125-80-10 ... 125 ... 80 10 ... 75 269321

850-550-10 850 ... 550 ... 10 100 ... 269321
150-100-7 ... 150 ... 100 7 ... 60 302363
1050-700-7 1050 ... 700 ... 7 80 ... 302363
175-125-4 ... 175 ... 125 4 ... 45 341444
1200-850-4 1200 ... 850 ... 4 60 ... 341444
200-155-1 ... 200 ... 155 1 ... 25 388477
1400-1100-1 1400 ... 1100 ... 1 35 ... 388477
230-185 ... 230 ... 185 (b) ... (b) 444555
1600-1300 1600 ... 1300 ... (b) (b) ... 444555
(a) Unnotched Charpy bars tested at 72 7 F (22 4 C). The values in the table are a minimum for the average of the highest three test values of four tested samples.
(b) Elongation and impact requirements are not specied. Although grades 200-155-1, 1400-1100-1, 230-185, and 1600-1300 are primarily used for gear and wear resis-
tance applications, grades 200-155-1 and 1400-1100-1 have applications where some sacrice in wear resistance is acceptable in order to provide a limited amount of
ductility and toughness. (c) Hardness is not mandatory and is shown for information only.
Chapter 3: Ferrous and Nonferrous Alloys / 73

weight and attain higher mechanical properties Silicon bronzes, also referred to as silicon
through heat treatment (refer to Table 13). As brasses, are commonly used in lightly loaded
the strength of aluminum bronzes is increased, gearing for electrical applications because of
their ductility decreases. Bearing characteristics their low cost and nonmagnetic properties. As
of aluminum bronzes are better than manganese shown in Table 13, these alloys contain 14% Zn
bronzes but inferior to tin bronzes. Aluminum with silicon additions of 3 to 4%.
bronzes are available in both cast and wrought Other copper-base alloys used for gear-
(Table 14) forms. ing include yellow brass, an alloy containing

Table 13 Compositions and properties of cast gear bronzes

Typical mechanical properties, as cast (heat treated)

Tensile strength Yield strength Brinell hardness

UNS Nominal Elongation in Machinability Casting
designation composition, % MPa ksi MPa ksi 50 mm (2 in.), % 500 kg 3000 kg rating(a) types(b)

Manganese bronzes
C86100 67Cu, 21Zn, 3Fe, 655 95 345 50 20 ... 180 30 C, I, P, S
5Al, 4Mn
C86200 64Cu, 26Zn, 3Fe, 655 95 331 48 20 ... 180 30 C, T, D, I, P, S
4Al, 3Mn
C86300 63Cu, 25Zn, 3Fe, 793 115 572 83 15 ... 225 8 C, T, I, P, S
6Al, 3Mn
C86400 59Cu, 1Pb, 40Zn 448 65 172 25 20 90 105 65 C, D, M, P, S
C86500 58Cu, 0.5Sn, 39.5Zn, 490 71 193 28 30 100 130 26 C, T, I, P, S
1Fe, 1Al
Silicon bronzes
C87400 83Cu, 14Zn, 3Si 379 55 165 24 30 70 100 50 C, D, I, M, P, S
C87500 82Cu, 14Zn, 4Si 462 67 207 30 21 115 134 50 C, D, I, M, P, S
Tin bronzes
C90700 89Cu, 11Sn 303 44 152 22 20 80 ... 20 C, T, I, M, S
(379) (55) (207) (30) (16) (102) ....
C91600 88Cu, 10.5Sn, 1.5Ni 303 44 152 22 16 85 ... 20 C, T, M, S
(414) (60) (221) (32) (16) (106) ...
C91700 86.5Cu, 12Sn, 1.5Ni 303 44 152 22 16 85 ... 20 C, T, I, M, S
(414) (60) (221) (32) (16) (106) ...
Leaded tin bronzes
C92500 87Cu, 11Sn, 1Pb, 303 44 138 20 20 80 ... 30 C, T, M, S
C92700 88Cu, 10Sn, 2Pb 290 42 145 21 20 77 ... 45 C, T, S
C92900 84Cu, 10Sn, 2.5Pb, 324 47 179 26 20 80 ... 40 C, T, M, S
3.5Ni (324) (47) (179) (26) (20) (80) ...
Aluminum bronzes
C95200 88Cu, 3Fe, 9Al 552 80 186 27 35 ... 125 50 C, T, M, P, S
C95300 89Cu, 1Fe, 10Al 517 75 186 27 25 ... 140 55 C, T, M, P, S
(586) (85) (290) (42) (15) ... (174)
C95400 85Cu, 4Fe, 11Al 586 85 241 35 18 ... 170 60 C, T, M, P
(724) (105) (372) (54) (8) ... (195)
C95500 81Cu, 4Ni, 4Fe, 689 100 303 44 12 ... 192 50 C, T, M, P, S
11Al (827) (120) (469) (68) (10) ... (230)
(a) Free-cutting brass = 100. (b) C, centrifugal; T, continuous; D, die; I, investment; M, permanent mold; P, plaster; S, sand

Table 14 Compositions and properties of wrought gear bronzes

Tensile strength Yield strength
Elongation in 50 mm Hardness HB
UNS designation Nominal composition, % MPa ksi MPa ksi (2 in.), % min and HRB

C62300 87 Cu, 3 Fe, 10 Al 620 90 310 45 25 180 HB (1000 kgf)

C62400 86 Cu, 3 Fe, 11 Al 655 95 345 50 12 200 HB (3000 kgf)
C63000 82 Cu, 3 Fe, 10 Al, 5 Ni 620 90 310 45 17 100 HRB
C64200 91.2 Cu, 7 Al 640 93 415 60 26 90 HRB
C67300 60.5 Cu, 36.5 Zn, 1.2 Al, 485 70 275 40 25 70 HRB
2.8 Mn, 1 Si
Note: Typical mechanical properties vary with form, temper, and section size considerations
74 / Gear Materials, Properties, and Manufacture

nominally 65% Cu and 35% Zn, and copper Handbook: Carbon and Alloy Steels, J.R.
alloys produced by P/M processing (refer to Davis, Ed., ASM International, 1996, p
Table 3 in Chapter 7, Powder Metallurgy). 138166
3. R.E. Cohen, P.J. Haagensen, D.K. Matlock,
and G. Krauss, Assessment of Bending
Other Nonferrous Alloys
Fatigue Limits for Carburized Steel, Tech-
The following alloys are also used for gears: nical Paper 910140, SAE International,
High-strength wrought aluminum alloys
4. D.V. Doane, Carburized SteelUpdate on
such as 2024 (Al-4.4Cu-1.5Mg-0.6Mn),
a Mature Composite, Carburizing: Pro-
6061 (Al-1.0Mg-0.6Si-0.30Cu-0.20Cr), and
cessing and Performance, G. Krauss, Ed.,
7075 (Al-5.6Zn-2.5Mg-1.6Cu-0.23Cr)
ASM International, 1989, p 169190
Die cast aluminum-silicon alloys such as
5. H.O. Fuchs and R.I. Stephens, Metal
360.0 and A360 (Al-9.5Si-0.5Mg), 383.0
Fatigue in Engineering, John Wiley &
(Al-10.5Si-2.5Cu), 384.0 (Al-11.2Si-3.8Cu),
Sons, 1980
and 413.0 (Al-12Si)
6. R.W. Landgraf and R.H. Richman, Fatigue
Die cast zinc-aluminum alloys such as ZA-8
Behavior of Carburized Steel, Fatigue of
(Zn-8Al-1Cu-0.02Mg), ZA-12 (Zn-11Al-
Composite Materials, STP 569, American
1Cu-0.025Mg), and ZA-27 (Zn-27Al-2Cu-
Society for Testing and Materials (ASTM),
1975, p 130144
Die cast magnesium-aluminum alloys such
7. H. Brugger and G. Krauss, Inuence of
as AZ91A and AZ91B (Mg-9Al) and
Ductility on the Behavior of Carburizing
AM60A and AM60B (Mg-6Al)
Steel during Static and Dynamic Bend
Properties of these die cast alloys can be found Testing, Arch. Eisenhuttenwes., Vol 32,
in Properties and Selection: Nonferrous Alloys 1961, p 529539
and Special-Purpose Materials, Vol 2, of the 8. R.E. Cohen, D.K. Matlock, and G. Krauss,
ASM Handbook. Specimen Edge Effects on Bending Fatigue
Because of their high strength-to-weight ratio, of Carburized Steel, J. Mater. Eng. Per-
titanium alloys, such as Ti-6Al-4V, have also form., Vol 1 (No. 5), 1992, p 695703
been used for gears. The poor tribological prop- 9. D.H. Breen and E.M. Wene, Fatigue in
erties of titanium alloys, however, limits their Machines and StructuresGround Vehi-
use to low-load applications. Wear-resistant cles, Fatigue and Microstructure, Ameri-
coatings produced by ion implantation or diffu- can Society for Metals, 1979, p 5799
sion treatments (oxidizing and nitriding) can 10. Z. Zurn and C. Razim, On the Fatigue
increase the wear resistance of titanium alloys. Strength of Case Hardened Parts, Carbur-
izing: Processing and Performance,
G. Krauss, Ed., ASM International, 1989,
11. M.B. Slane, R. Buenneke, C. Dunham, M.
Portions of this chapter were adapted from: Semenek, M. Shea, and J. Tripp, Gear Sin-
gle Tooth Bending Fatigue, Technical
Gas Carburizing, Surface Hardening of Paper 821042, SAE International, 1982
Steels: Understanding the Basics, J.R. 12. D. Medlin, G. Krauss, D.K. Matlock, K.
Davis, Ed., ASM International, 2002, Burris, and M. Slane, Comparison of Sin-
p 1789 gle Gear Tooth and Cantilever Beam Bend
Steels for Gears, ASM Specialty Handbook: Fatigue Testing of Carburized Steel,
Carbon and Alloy Steels, J.R. Davis, Ed., Technical Paper 950212, SAE Interna-
ASM International, 1996, p 591598 tional, 1995
13. R.S. Hyde, R.E. Cohen, D.K. Matlock, and
G. Krauss, Bending Fatigue Crack Char-
REFERENCES acterization and Fracture Toughness of Gas
Carburized SAE 4320 Steel, Technical
1. N.P. Chirons, Ed., Gear Design and Appli- Paper 920534, SAE International, 1992
cation, McGraw-Hill, 1967, p 269274 14. M.A. Zaccone, J.B. Kelley, and G. Krauss,
2. Hardenability of Steels, ASM Specialty Strain Hardening and Fatigue of Simulated
Chapter 3: Ferrous and Nonferrous Alloys / 75

Case Microstructures in Carburized Steel, Carburizing: Processing and Performance,

Carburizing: Processing and Performance, G. Krauss, Ed., ASM International, 1989,
G. Krauss, Ed., ASM International, 1989, p 303310
p 249265 28. R.S. Hyde, D.K. Matlock, and G. Krauss,
15. M.M. Shea, Impact Properties of Selected The Effect of Reheat Treatments on
Gear Steels, Technical Paper 780772, Fatigue and Fracture of Carburized Steel,
SAE International, 1978 Technical Paper 940788, SAE Interna-
16. G. Krauss, The Relationship of Microstruc- tional, 1994
ture to Fracture Morphology and Tough- 29. S. Gunnarson, Structure Anomalies in the
ness of Hardened Hypereutectoid Steels, Surface Zone of Gas-Carburized Case-
Case Hardened Steels: Microstructure and Hardened Steel, Met. Treat. Drop Forg.,
Residual Stress Effects, TMS/AIME, 1984, Vol 30, 1963, p 219229
p 3356 30. T. Naito, H. Ueda, and M. Kikuchi, Fatigue
17. R.S. Hyde, Quench Embrittlement and Behavior of Carburized Steel with Internal
Intergranular Oxide Embrittlement: Effects Oxides and Nonmartensitic Microstruc-
on Bending Fatigue Initiation of Gas- tures Near the Surface, Metall. Trans. A,
Carburized Steel, Ph.D. dissertation, Col- Vol 15, 1984, p 14311436
orado School of Mines, 1994 31. W.E. Dowling, Jr., W.T. Donlon, W.B.
18. L. Magnusson and T. Ericsson, Initiation Copple, and C.V. Darragh, Fatigue Behav-
and Propagation of Fatigue Cracks in Car- ior of Two Carburized Low Alloy Steels,
burized Steel, Heat Treatment 79, The 1995 Carburizing and Nitriding with
Metals Society, London, 1979, p 202206 Atmospheres, J. Grosch, J. Morral, and M.
19. J.L. Pacheco and G. Krauss, Microstructure Schneider, Ed., ASM International, 1995,
and High Bending Fatigue Strength in Car- p 5560
burized Steel, J. Heat Treat., Vol 7 (No. 2), 32. C. Razim, Uber den Einuss von Restau-
1989, p 7786 stenit auf des Festigkeitsverhalten Einsatz-
20. K.A. Erven, D.K. Matlock, and G. Krauss, geharteter Probenkorper bei Schwin-gen-
Effect of Sulfur on Bending Fatigue of Car- der Bean-spruchung, Harterei-Tech. Mitt.,
burized Steel, J. Heat Treat., Vol 9, 1991, Vol 23, 1968, p 18
p 2735 33. R.H. Richman and R.W. Landgraf, Some
21. G. Krauss, The Microstructure and Frac- Effects of Retained Austenite on the
ture of a Carburized Steel, Metall. Trans. A, Fatigue Resistance of Carburized Steel,
Vol 9, 1978, p 15271535 Metall. Trans. A, Vol 6, 1975, p 955964
22. R.S. Hyde, G. Krauss, and D.K. Matlock, 34. M. Zaccone and G. Krauss, Elastic Limit
Phosphorus and Carbon Segregation: and Microplastic Response of Hardened
Effects on Fatigue and Fracture of Gas- Steels, Metall. Trans. A, Vol 24, 1993,
Carburized Modied 4320 Steel, Metall. p 22632277
Trans. A, Vol 25, 1994, p 12291240 35. G.B. Olson, Transformation Plasticity and
23. G. Krauss, Steels: Heat Treatment and Pro- the Stability of Plastic Flow, Deformation,
cessing Principles, ASM International, Processing, and Structure, G. Krauss, Ed.,
1990 ASM International, 1984, p 391424
24. G. Krauss, Heat Treated Martensitic Steels: 36. Final Report, Advanced Rotorcraft Trans-
Microstructural Systems for Advanced mission Program, National Aeronautics
Manufacture, ISIJ Int., Vol 35 (No. 4), and Space Administration, Cleveland, Ohio
1995, p 349359 37. R.J. Johnson, The Role of Nickel in Car-
25. K.A. Erven, D.K. Matlock, and G. Krauss, burizing Steels, Met. Eng. Quart., Vol 15
Bending Fatigue and Microstructure of (No. 3), 1975, p 18
Gas-Carburized Alloy Steels, Mat. Sci. 38. K.D. Jones and G. Krauss, Effects of High-
Forum, Vol 102104, 1992, p 183198 Carbon Specimen Corners on Microstruc-
26. R.E. Cohen, G. Krauss, and D.K. Matlock, ture and Fatigue of Partial Pressure Carbur-
Bending Fatigue Performance of Carbur- ized Steels, Heat Treatment 79, The
ized 4320 Steel, Technical Paper 930963, Metals Society, London, 1979, p 188193
SAE International, 1993 39. C. Kim, D.E. Diesburg, and R.M. Buck,
27. B. Thoden and J. Grosch, Crack Resistance Inuence of Sub-Zero and Shot-Peening
of Carburized Steel under Bend Stress, Treatment on Impact and Fatigue Fracture
76 / Gear Materials, Properties, and Manufacture

Properties of Case-Hardened Steels, J. 52. H. Swahn, P.C. Becker, and O. Vingsbo,

Heat Treat., Vol 2 (No. 1), 1981, p 4353 Martensite Decay during Rolling Contact
40. M.A. Panhans and R.A. Fournelle, High Fatigue in Ball Bearings, Metall. Trans. A,
Cycle Fatigue Resistance of AISI E9310 Vol 7, 1976, p 10991110
Carburized Steel with Two Different Lev- 53. J.A. Martin, S.F. Borgese, and A.D. Eber-
els of Surface Retained Austenite and Sur- hardt, Microstructural Alterations of
face Residual Stress, J. Heat Treat., Vol 2 Roller-Bearing Steel Undergoing Cyclic
(No. 1), 1981, p 5561 Stresses, J. Basic Eng., 1966, p 555567
41. J. Grosch and O. Schwarz, Retained 54. V. Bhargava, G.T. Hahn, and C.A. Rubin,
Austenite and Residual Stress Distribution Rolling Contact Deformation, Etching
in Deep Cooled Carburized Microstruc- Effects, and Failure of High Strength Bear-
tures, 1995 Carburizing and Nitriding with ing Steel, Metall. Trans. A, Vol 21, 1990, p
Atmospheres, J. Grosch, J. Morral, and 19211931
M. Schneider, Ed., ASM International, 55. C.A. Stickels, Rolling Contact Fatigue
1995, p 7176 Tests of 52100 Bearing Steel Using a Mod-
42. M.G. Mendiratta, J. Sasser, and G. Krauss, ied NASA Ball Test Rig, Wear, Vol 98,
Effect of Dissolved Carbon on Microcrack- 1984, p 199210
ing in Martensite of an Fe-1.39 pct C Alloy, 56. L. Kiessling, Rolling-Contact Fatigue of
Metall. Trans., Vol 3, 1972, p 351353 Carburized and Carbonitrided Steels, Heat
43. R.P. Brobst and G. Krauss, The Effect of Treat. Met., Vol 7 (No. 4), 1980, p 97101
Austenite Grain Size on Microcracking in 57. C.V. Darragh, Engineered Gear Steels: A
Martensite of an Fe-1.22C Alloy, Metall. Review, Gear Technology, November/
Trans., Vol 5, 1975, p 457462 December, 2002, p 3340
44. L.J. Ebert, The Role of Residual Stresses in 58. C.A. Stickels, Carburizing, Friction, Lubri-
the Mechanical Performance of Case Car- cation, and Wear Technology, Vol 18, ASM
burized Steel, Metall. Trans. A, Vol 9, Handbook, ASM International, 1992,
1978, p 15371551 p 873877
45. D.P. Koistinen, The Distribution of Resid- 59. Modern Carburized Nickel Alloy Steels,
ual Stresses in Carburized Steels and Their Reference Book Series 11,005, Nickel
Origin, Trans. ASM, Vol 50, 1938, p 227 Development Institute, 1989
241 60. C. Razim, Some Facts and Considerations
46. B. Liscic, State of the Art in Quenching, of Trends in Gear Steels for the Automotive
Quenching and Carburizing, The Institute Industry, Alloys for the Eighties, Climax
of Materials, London, 1993, p 132 Molybdenum Company, 1980, p 9
47. A.K. Hellier, M.B. McGirr, S.H. Alger, and 61. T.B. Cameron, D.E. Diesburg, and C. Kim,
M. Stefulji, Computer Simulation of Resid- Fatigue and Overload Fracture of a Carbur-
ual Stresses during Quenching, Quenching ized Case, J. Metals, May 1983
and Carburizing, The Institute of Materi- 62. Y.E. Smith and D.E. Diesburg, Fracture
als, London, 1993, p 127138 Resistance in Carburizing Steels, Parts I, II,
48. G. Totten, Ed., Quenching and Distortion III, Met. Prog., May, June, July 1979
Control, ASM International, 1992 63. D.E. Diesburg, High-Hardenability Car-
49. G. Parrish and G.S. Harper, Production Gas burizing Steels for Rock Bits, Micon
Carburizing, Pergamon Press Inc., 1985 78: Optimization of Processing, Proper-
50. G. Krauss, Microstructure, Residual ties, and Service Performance Through
Stresses and Fatigue of Carburized Steels, Microstructural Control, STP 672, ASTM,
Quenching and Carburizing, The Institute 1979
of Materials, London, 1993, p 205225 64. Steel Castings Handbook, 6th ed., M. Blair
51. R.P. Garibay and N.S. Chang, Improved and T.L. Stevens, Ed., Steel Founders
Fatigue Life of a Carburized Gear by Society of America and ASM International,
Shot Peening Parameter Optimization, Car- 1995, p 188
burizing: Processing and Performance, 65. Dudleys Gear Handbook, 2nd ed.,
G. Krauss, Ed., ASM International, 1989, D.P. Townsend, Ed., McGraw-Hill, Inc.,
p 283289 p 827
Gear Materials, Properties, and Manufacture Copyright 2005 ASM International
J.R. Davis, editor, p77-88 All rights reserved.
DOI:10.1361/gmpm2005p077 www.asminternational.org



PLASTIC GEARS are continuing to displace Limitations. There are also limitations of
metal gears in applications ranging from auto- plastic gears relative to metal gears. These
motive components to ofce automation equip- include:
ment. In the past, plastic gears were not consid-
ered to be able to transmit power, were limited as Maximum load-carrying capacity lower
to top operating speed, and were not considered than metal gears
able to transmit motion to a high degree of accu- Reduced ability to operate at elevated tem-
racy. However, improved materials with higher perature. Operation is generally limited to
load capacities, advances in mold design and less than 120 C (250 F). Cold temperature
molding technology, and the development of re- applications are also limited.
liable engineering data have led to the successful Ambient temperature and temperatures at
and increasing use of plastics for both motion- tooth contact surface must be limited
carrying gears and power transmission gears. Plastic gears cannot be molded to the same
accuracy as high-precision machined/n-
ished metal gears
General Characteristics of Plastic gears are subject to greater dimen-
Plastic Gears (Ref 1) sional instabilities due to their greater coef-
cient of thermal expansion and moisture
Advantages. Among the characteristics re- absorption
sponsible for the large increase in plastic gear Initial high mold costs in developing correct
usage, the following are the most signicant: tooth form and dimensions
Relative low cost (particularly for high- Can be negatively affected by certain chem-
volume injection molded gears) icals and even some lubricants
Ease and speed of manufacture Improper molding tools and process can
produce residual internal stresses at the
Wide range of congurations and complex
tooth roots resulting in overstressing and/or
shapes possible
distortion with aging
Elimination of machining and nishing
Cost of plastics are based on petrochemical
pricing and thus are more volatile in com-
Capability of fabrication with metal inserts
parison to metals
and integral designs
Lower density (light weight and low inertia)
Ability to dampen moderate shock and
impact Classication of
Ability to operate with minimal or no lubri- Plastics for Gear Applications
Low coefcient of friction One very important classication scheme for
Smooth, quiet operation plastics is based on their response to heat. There
Lower critical tolerances than with metal are two types of response possible, leading to
gears, due in part to their greater resilience the classication of plastics as thermoplastics or
Resistance to corrosion thermosets.
78 / Gear Materials, Properties, and Manufacture

Thermoplastics are materials that repeat- ing plastics, which are viscoelastic, do not re-
edly soften, or melt, when heated, and harden, spond to mechanical stress in the linear, elastic
or freeze, when cooled (Ref 2). Heating permits manner for which most designers have devel-
the intertwined molecular chains to slide rela- oped an intuitive feel. In many applications, en-
tive to each other. At some higher temperature, gineering plastics exhibit a more complex prop-
the sliding is free enough for the materials to erty mix than do metals. Because engineering
behave as a liquid and may be used to ll molds. plastics, as a family, are much younger than en-
The temperature at which this degree of soften- gineering metals, their database is not yet com-
ing occurs varies with the type and grade of plete. In addition, their rapid evolution makes
plastic. Cooling restores the intermolecular the material selection process more difcult.
bonds and the material behaves as a solid. How- A typical stress-strain curve for an engineer-
ever, these solid materials, to a varying degree, ing metal is shown in Fig. 1. The salient features
retain some aspects of a liquid in the form of of metal behavior are that the slope of the stress-
viscoelastic behavior (see the section Metals strain curve is a constant up to the proportional
versus Plastics in this chapter). Thermoplastics limit and is known as the elastic modulus, and
have greater toughness, or resistance to impact that the elastic modulus remains a constant over
loads, than do thermosetting plastics. However, a wide range of strain rates and temperatures. A
their upper temperature limit is only about large number of design equations have been
120 C (250 F). Above this temperature, ther- derived for use in the design of structural mem-
moplastics lose about 50% of their rated room- bers such as beams, plates, and columns. All of
temperature strength. these equations use the elastic modulus as a fun-
Thermosets are plastics that undergo chem- damental measure of the response of the mate-
ical change during processing to become per- rial to stress.
manently infusible (Ref 2). If excessive heat is However, if the designer attempts to use these
applied to a thermoset material after the chemi- equations for plastics part design using a modu-
cal change has taken place, the plastic is de- lus value taken from a plastics property speci-
graded rather than melted. Before the process- cations sheet, serious design errors will be made
ing, the molecular structure of the thermoset in many cases. The implicit assumption behind
plastic is similar to that of a thermoplastic mate- the design equations is that the material behaves
rial. Heating permits the relative sliding of the in a linear, elastic manner. As already noted,
molecules, the material takes on the properties this assumption is not true for plastics.
of a liquid and can be used to ll molds. How-
ever, while still subjected to heat in the mold in
a curing process, the intertwined molecules
develop cross-links to form an irreversible net-
work which prevents further relative sliding.
The resulting solid plastic behaves much
more like an elastic material similar to metals.
The viscoelastic behavior is much reduced from
that of most thermoplastics. At the same time,
these thermoset materials tend to be much less
resistant to impact loading, and tend to be used
only with some reinforcing medium in even
moderate impact applications. On the other
hand, thermosets tend to maintain their strength
properties at much higher temperatures than
most thermoplastics. Very little degradation of
mechanical properties occurs at temperatures up
to 230 C (450 F).

Metals versus Plastics

Designing with plastics is often a more com- Fig. 1 Typical stress-strain curve for an engineering metal.
Note high degree of linearity below the proportional
plex task than designing with metals. Engineer- limit.
Chapter 4: Plastics / 79

When comparing the mechanical behavior processors should always be consulted when
of metals to that of plastics, as shown in Fig. 1 choosing a plastic for a gear application.
and 2, it can be seen that plastic, unlike metal,
has no true proportional limit and that its stress- Additives
strain curve is not linear. Furthermore, the
A wide variety of organic and inorganic
shape of the plastic stress-strain curve is greatly
materials are added to plastics either to reduce
affected by the rate at which stress is applied,
costs or property improvement. Some of these
and stress-strain behavior is greatly affected by
properties are (Ref 2):
Clearly, the mechanical response of plastic is Strength
more complex than that of metal. Because a Modulus (rigidity)
plastic behaves in a manner that seems to com- Impact resistance
bine elastic-solid and highly viscous liquid Wear resistance
behaviors, it is said to be viscoelastic. Thermal conductivity
Flame retardance
Dimensional control
Plastic Gear Materials Color
Heat stability
This section will briey review some of the Noise reduction
commonly used plastic materials for gear appli- Oxidative stability
cations (many more plastic formulations are Ultraviolet stability
anticipated for gears in the future). These mate- Lubricity
rials can be used in the unmodied condition or Processability
modied with various additives to produce a
wide variety of gear congurations including Fillers are added to reduce costs, improve
spur, worm, face, helical, bevel, internal, and dimensional control (stability) during the mold-
elliptical gears, gear racks, and microgears (for ing operation, enhance conductivity, and im-
computer memory devices). Selection of a ma- prove heat resistance. Common mineral llers
terial depends not only on its properties, but also include mica, talc, carbon powder, and glass
on the manufacturing methods used, part shape beads. The amount of ller added can range
and size, molding or machining characteristics, from 5 to 40%. Organic llers (<1%) are also
shrinkage rates, moisture absorption, and pro- sometimes used as a processing aide. Limita-
cessing variables. Material suppliers and/or tions of adding llers include reduced impact
resistance and increased tool wear.
Reinforcements are usually thought of as
an added material whose purpose is to alter one
or more of the mechanical properties of the
basic material. Glass bers (added in the range
of 5 to >40%) are often used to alter both long-
and short-term mechanical properties of the
basic material resin. Heat and wear resistance
are also improved. Limitations of adding glass
bers include increased distortion during mold-
ing, increased wear of the mating gear material,
and increased tool wear.
Carbon bers in the percentage range of 10 to
>40% are added to enhance heat resistance,
electrical properties, and wear resistance. Limi-
tations of adding carbon bers include higher
costs, increased distortion during molding, and
increased tool wear.
Kevlar (aramid) bers in the percentage
range of 5 to 20% are added to increase heat
Fig. 2 Typical stress-strain curve for an engineering plastic. resistance, strength, and wear resistance and
Note that there is no true proportional limit. to reduce friction. The primary limitation of
80 / Gear Materials, Properties, and Manufacture

adding Kevlar bers is the added cost. Some conditions, exhibit outstanding toughness and
distortion may also occur during molding. wear resistance, low coefcient of friction, and
Lubricants. The following lubricants are excellent electrical properties and chemical
added to plastics for the primary purpose of resistance. Properties of Nylons 6 and 6/6 are
reducing friction and wear: listed in Tables 1 (0.2% moisture content) and 2
(50% relative humidity). Some nylons are
Lubricant Typical percentage range added hygroscopic (absorb moisture) and this has a
Polytetrauoroethylene 120%
negative effect on their strengths (see Fig. 3 and
(PTFE, or Teon) compare Tables 1 and 2). In addition, the dimen-
Silicone 14% sional stability of these resins is poorer than
Graphite 510%
Molybdenum disulde (MoS2) 25%
other engineering plastics. Nylon gears can be
manufactured by either machining or molding.
Acetal polymers have a lower water absorp-
The primary limitation of adding PTFE is added
tion rate than nylon and, therefore, are more sta-
cost. Graphite additions can reduce strength and
ble after molding or machining. The acetals,
impact resistance. There are no signicant lim-
which come in unmodied, toughened, rein-
iting factors for adding MoS2.
forced, and internally lubricated conditions, are
Impact modiers are additives that im-
strong, have good resistance to creep and
prove the impact strength or toughness of plas-
fatigue, have a low coefcient of friction, and
tics with minimal impairment of other proper-
are resistant to abrasion and chemicals. Proper-
ties. They can be added in such large percentages
ties of various acetal resins are listed in Table 3.
(>50%) that they become a component in a
The effect of temperature on the stress-strain
blended material rather than an additive. To be
curve for an unmodied acetal grade is shown in
effective, the added phase must be more impact
Fig. 4. The effect of internal lubricants on the
resistant and have a lower modulus than the par-
sliding wear resistance of acetal resins is shown
ent phase. Commonly used impact modiers in-
in Fig. 5. As this gure indicates, the PTFE-
clude acrylonitrile-butadiene-styrene (ABS),
lubricated grades exhibited far greater wear
methacrylate-butadiene-styrene (MBS), acryl-
resistance than the unmodied grades.
ics, chlorinated polyethylene (CPE), and ethyl-
Other thermoplastics used for gears
ene vinyl acetate (EVA).
Thermoplastic Gear Materials
Polycarbonates are generally reinforced
Most of the plastics used for gear applications with glass bers and lubricated with PTFE.
are thermoplastics. The two most commonly Noted for their low-shrinkage characteris-
used thermoplastic formulations are the poly- tics, they are manufactured by either mold-
amides (nylons) and the acetals. These materials ing or machining.
have been used in gears for more than 50 years in Polyesters are unmodied or reinforced with
hundreds of diverse products including: glass bers. Applications for these molded
resins are similar to the nylons and acetals.
Windshield wipers
Polyesters can also be blended with elas-
Automobile window lifts
tomers for extra tough gears subject to
Automobile seat controls
extreme shock or when mesh noise reduc-
tion is required.
Rotary pumps
Polyurethanes are unmodied, reinforced,
or internally lubricated. These molded resins
Electric garage door openers
are noted for their exibility, ability to
Small power tools
absorb shock, and deaden sound.
Polyphenyline suldes are generally rein-
Copy machines
forced with glass bers (up to 40%) with or
Laser printers
without internal lubricants. These resins
Nylons. This is a family of thermoplastic have higher elevated temperature strength
resins that were the rst plastics used for gear than most other thermoplastics.
applications. The most common grades used are Liquid crystal polymer resins contain min-
Nylon 6, 6/6, and 6/12. These materials, which eral llers or are reinforced with glass bers.
come in unmodied, toughened, and reinforced They are used for extremely thin or small
Chapter 4: Plastics / 81

gears and exhibit both high-temperature re- Polyimides, which are available commer-
sistance and chemical resistance and excel- cially either as thermoplastic or thermoset
lent dimensional stability. resins, are used for high-temperature gear appli-
cations. Finish machined thermoset polyimide
Thermoset Gear Materials grades are unmodied or internally lubricated.
Only a small number of thermoset plastics Thermoplastic polyimide gears can be manufac-
have been used for gears to any substantial tured by either injection molding or machining.
degree. The properties of thermoplastics gener- The thermoplastic grades are available in un-
ally make them more suitable for gears. The dif- modied, glass reinforced (up to 65% glass
ferent type of processing required for some ther- bers), toughened, and lubricated forms. Lubri-
mosets also limits their use. cants for thermoplastic polyimides include

Table 1 Property values for Nylons 6 and 6/6

Dry, as-molded, approximately 0.2% moisture content
Nylon 6/6
Nylon 6 Toughened
Molding and Glass ber Glass ber Glass ber
extrusion reinforced, Molding reinforced, reinforced,
Property compound 3035% compound 3033% Unreinforced 33%

Tensile strength at break, ... 165 (24) 94.5 (13.7) 193 (28) 50 (7.0) 140 (20.3)
MPa (ksi)
Elongation at break, % 30100 2.23.6 1560 2.534 125 46
Tensile yield strength, 80.7 (11.7) ... 55 (8.00) 170 (25) ... ...
MPa (ksi)
Compressive strength, 90110 (1316) 131165 86.2103 165276 ... 103 (15)
rupture or yield, MPa (ksi) (1924) (12.515.0)(a) (2440)
Flexural strength, rupture, 108 (15.7) 240 (35) 114117 283 (41) 59 (8.5) 206 (29.9)
or yield, MPa (ksi) (16.517.0)
Tensile modulus, GPa 2.6 (0.38) 8.6210.0 1.593.79 9.0 (1.3) ... ...
(106 psi) (1.251.45) (0.230.55)
Flexural modulus at 23 C 2.7 (0.39) 9.65 (1.40) 2.83.1 8.9610.0 1.65 (0.240) 7.58 (1.10)
(73 F), GPa (106 psi) (0.410.45) (1.301.45)
Izod impact, 3.2 mm (1/8 in.) 3253 (0.61.0) 117181 2953 85240 907 (17.0) 240 (4.5)
thick specimen, notched, (2.23.4) (0.551.0) (1.64.5)
J/m (ft lbf/in.)
Rockwell hardness R119 M9396 R120 R101119 R100 R107
Coefcient of linear thermal 8083 1680 80 1554 ... ...
expansion, 106/K
Deection temperature under
exural load, C (F)
At 1.82 MPa (0.264 ksi) 6885 (155185) 200215 7588 120250 65 (150) 245 (470)
(392420) (167190) (252490)
At 0.46 MPa (0.066 ksi) 185190 215220 230246 125260 ... 260 (495)
(365375) (420430) (450474) (260500)
Thermal conductivity, W/m 0.243 (1.69) 0.2430.472 0.243 (1.69) 0.2110.483 ... ...
K (Btu in./h ft2 F) (1.693.27) (1.463.35)
Specify gravity 1.121.14 1.351.42 1.131.15 1.151.40 1.08 1.34
Water absorption, 3.2 mm 1.31.9 0.901.2 1.02.8 0.71.1 1.0 0.7
(1/8 in.) thick specimen,
24 h, %
Saturation, % 8.510.0 6.47.0 8.5 5.56.5 ... ...
Dielectric strength, 3.2 mm 0.0157 (400) 15.717.7 0.0236 (600) 14.219.7 ... ...
(1/8 in.) thick specimen, (400450) (360500)
short time, MV/m (V/mil)
(a) Yield
82 / Gear Materials, Properties, and Manufacture

graphite (15 to 40%), PTFE (10%) plus graphite The required accuracy or some special de-
(15%), and MoS2 (up to 15%). sign feature (such as a very thick section)
Phenolics are injection molded resins com- may be too difcult for molding
pounded with various mineral and glass llers, The desired plastic material may not be
glass bers, and lubricants such as PTFE and suited to precision molding
graphite. Phenolics are used in applications re-
quiring dimensional stability improved heat Plastic materials to be machined into gears are
resistance. available in a variety of forms, the most common
Laminated Phenolic. Composite, or lami- of which are nonreinforced extruded circular
nated thermosets, consist of a brous sheet mate- rods or tubes. These rods are extruded by stock
rial (paper, cotton fabric, nylon fabric, or glass shape suppliers in diameters up to 200 mm
fabric) that is impregnated or coated with a liq- (8 in.). Some of the properties of the material in
uid phenolic resin binder and consolidated under extruded stock form can vary between the outer
high pressure and temperatures. By stacking and surface (or skin), which has formed through
consolidating the unidirectional laminations (or more rapid cooling, than the inner material (or
plies) at different angles, the cut gear blanks/ core). Machining invariably leaves the gear teeth
teeth will have equal strength in all directions. made from the core material, which generally
has different strength, wear resistance, and
chemical resistance. In addition, machining may
Plastic Gear Manufacture expose voids in the stock. This is particularly
true for brous materials, such as glass or carbon
Plastic gears are manufactured by either ber, or hard granular materials, when used as
machining or injection molding. Machining of additives in molded gears. These additives may
plastic gears is performed by most of the same be removed from the surface during machining
processes used in the machining of metal gears to leave openings or surface roughness.
(see Chapter 5, Machining, Grinding, and Fin-
ishing). Machining may be selected over mold- Injection Molding
ing as the plastic gear manufacturing process for Injection molding refers to a variety of
several reasons: processes that generally involve forcing or
The quantities may be too small to justify injecting a uid plastic material into a closed
the tooling cost for molding mold. The process has been used to produce

Table 2 Property values for Nylons 6 and 6/6

Conditioned to 50% relative humidity
Nylon 6 Nylon 6/6

Molding and Glass ber Glass ber

extrusion reinforced, Molding reinforced,
Property compound 3035% compound 3033%

Tensile strength at break, ... 110 (16) 75 (11) 152 (22)

MPa (ksi)
Elongation at break, % 300 ... 150300 57
Tensile yield strength, 51 (7.4) ... 45 (6.5) ...
MPa (ksi)
Flexural strength, rupture 40 (5.8) 145 (21.0) 42 (6.1) 170 (25)
or yield, MPa (ksi)
Tensile modulus, GPa 0.690 (0.100) 5.52 (0.800) 3.45 (0.500)
(106 psi)
Compressive modulus, GPa 1.70 (0.250) ... ... ...
(106 psi)
Flexural modulus at 23 C 0.965 (0.140) 5.526.55 (0.8000.950) 1.28 (0.185) 5.52 (0.800)
(73 F), GPa (106 psi)
Izod impact, 3.2 mm (1/8 in.) 160 (3.0) 197292 (3.75.5) 45112 (0.852.1) 138159 (2.63.0)
thick specimen, notched,
J/m (ft lbf/in.)
Rockwell hardness ... M78 ... ...
Chapter 4: Plastics / 83

plastic gears since the 1950s. American Gear ing of injection molded gears requires the joint
Manufacturers Association (AGMA) quality efforts of the gear design engineer, the mold
levels ranging from 6 to 8 can be obtained (Ref producer, the molder, and the material supplier.
4). The process is particularly well suited for Injection-molding compounds are ther-
high-volume, automated production of thermo- moplastic or thermosetting materials and their
plastic gears. The development of manufactur- composites that are specically formulated for

Fig. 3 Effect of temperature and moisture content of an Fig. 4 Effect of temperature on the stress-strain behavior of an
unmodied nylon. Source: Ref 1 unmodied acetal. Source: Ref 2

Table 3 Mechanical properties of various Delrin acetal resins

Delrin is a registered trademark of DuPont Engineering Polymers
Product description

Unmodied, Enhanced Glass
general purpose crystallinity Toughened reinforced PTFE
Properties 100P 500P 900P 511P 100ST 500T 525GR 520MP 510MP

Tensile strength at yield 67 68 69 72 45 53 151(a) 54 61

MPa (ksi) ASTM D638 (9.7) (9.9) (10.0) (10.4) (6.5) (7.7) (21.9)(a) (7.8) (8.9)
Elongation at yield, % 23 15 11 11 35 15 NA(b) 12 10
Elongation at break, % 80 40 25 33 >150 75 3 14 13
Flexural modulus MPa (ksi) 2790 3100 3240 3300 1130 2250 8000 3100 3160
ASTM D790 (410) (450) (470) (480) (160) (330) (1160) (450) (460)
Izod impact, notched J/m 120 75 69 73 NB 128 96 32 27
(ft-lb/in.) ASTM D256 (2.3) (1.4) (1.3) (1.4) (2.4) (1.8) (0.6) (0.5)
Izod impact, unnotched J/m NB(c) NB(c) 1630 NB(c) NB(c) NB(c) 1100 720 950
(ft-lb/in.) ASTM D256 (30.6) (20.6) (13.5) (17.8)
(a) Values are tensile strength at break since these materials do not yield. (b) Not applicable. (c) No break occurred. Material code: Delrin 100P, 500P, and 900P
are unmodied acetals with high, medium, and low viscosities, respectively; Delrin 511P is a medium viscosity acetal; Delrin 100ST and 500ST have high and me-
dium viscosities, respectively; Delrin 525GR is a medium viscosity, 25% glass-reinforced acetal; and Delrin 520MP and 510MP contain 20% and 10% PTFE,
84 / Gear Materials, Properties, and Manufacture

the injection-molding process. This process injection process conditions is well understood.
requires materials capable of being fed into a Where this understanding is absent, reinforce-
molding machine, transported to accumulate ments may be broken, and polymers may be
pressure, injected through channels, and made either degraded or prematurely cured.
to ow into a small opening in the mold. The Injection Molding Machines. Although
process may cause major changes in both the there are many variations in thermoplastic in-
physical and chemical properties of the molding jection molding machines, the reciprocating-
compound. Because of their resistance to ow, screw machine has become the most common
neither high-molecular-weight resins nor long type, and the following description of its opera-
reinforcing bers, or akes, can be effectively tion illustrates the basic principles of all types.
manipulated through the molding process. Con- The functions performed by the machine
sequently, parts produced from molding com- include heating the plastic until it is able to ow
pounds represent a compromise between opti- readily under pressure, pressurizing this melt to
mal physical properties and the essential ability inject it into a closed mold, holding the mold
to ow under pressure. This compromise is off- closed both during the injection and while the
set by the ability to produce three-dimensional material is solidifying in the mold, and opening
products with holes, ribs, and bosses, often the mold to allow removal of the solid part.
without secondary operations or direct labor. As depicted in Fig. 6, the main components of
While the ow process has the potential to the machine are the hopper, heated extruder bar-
physically change the molding compound, these rel, reciprocating screw, nozzle, and mold
changes are not always negative. Improved clamp. The hopper feeds the unmelted thermo-
homogeneity, better reinforcement wetting, and plastic, usually in pellet form, into the barrel.
higher physical properties may be achieved if The hopper is often equipped with a desiccant-
the response of the molding compound to the type drying system to remove moisture degra-

Fig. 5 Effect of internal lubricants on the sliding wear (acetal against steel) of 27 tooth, 2.12 module (12 pitch), involute thermo-
plastic spur gears. Initial contact stress = 42 MPa 9 (6 ksi). Pitchline velocity = 5.6 m/s (18.4 ft/s). Solid lines show median of
measured wear range. Dotted lines are extrapolated data. Source: Dupont Engineering Polymers
Chapter 4: Plastics / 85

dation. A magnet is placed in the hopper throat mixing action for better material and tempera-
to remove any iron that has accidentally entered ture uniformity; and the metering zone, gener-
the feedstock. ally of constant, shallow-ight depth, provides
In the heated extruder barrel, which contains the nal shear heating and mixing of the melt.
the reciprocating screw, the thermoplastic is After the melt passes through the metering sec-
gradually melted by a combination of shear tion and a check valve at the end of the screw, it
heating (caused by the mechanical working of joins the melt pool in front of the screw. As the
the material as it is conveyed down the barrel) volume of melt in front of the screw increases, it
and heat conduction from the barrel. The ther- forces the screw to the rear of the barrel against
moplastic is gradually conveyed by the rotating an adjustable back pressure. This back pres-
screw from the rear of the barrel to the front. sure is applied hydraulically to the back end of
The barrel is usually heated by band-type elec- the screw. Increasing the pressure increases the
trical resistance heaters tted around its periph- amount of mechanical working of the feedstock.
ery. Because the amount of heat required varies Screw rotation and feedstock melting continue
with position along the barrel, the band heaters until a sufcient amount of melt is available in
are controlled in several zones along the barrel. front of the screw to ll the mold. At this stage,
Some barrels are equipped about halfway down the screw rotation is stopped, and the machine is
with a vent to remove gases from the melting ready for injection. The stages in the operation of
feedstock. A vacuum is usually applied to the a reciprocating screw are shown in Fig. 7.
vent, and a specially designed screw is required The melted thermoplastic is injected into the
to depressurize the melt in the region of the vent mold through the nozzle, under high pressure
to keep it from extruding through the vent. (typically 70 to 205 MPa, or 10 to 30 ksi,
The reciprocating screw usually has three suc- depending on the mold-lling resistance). Injec-
cessive zones, each with specic functions: tion occurs as the screw is hydraulically forced
The feed zone of relatively deep screw ights forward in the barrel. The hydraulic cylinder is
conveys the unmelted plastic from the hopper located at the rear of the screw and barrel. A
throat into the heated barrel, where it begins to check valve at the tip of the screw keeps the
melt; the compression zone of decreasing ight melt from owing back along the screw as the
depth, and thus volume, provides gas removal screw is pushed forward. In contemporary
and melt densication, along with a material machines, the injection rate (determined by the

Fig. 6 The injection end of a reciprocating-screw injection molding machine

86 / Gear Materials, Properties, and Manufacture

forward velocity of the screw) and injection clamping force, or tonnage, required is deter-
pressure are closely controlled throughout the mined by the amount of injection pressure
mold-lling stage. required to ll the mold and the projected area
The mold clamp, which holds the halves of of the part (that is, the area perpendicular to the
the mold closed against the injection pressure of axis of the machine). If the machine has inade-
the melt, opens the mold to allow part removal quate clamping force, the two halves of the tool
after the thermoplastic has solidied. During the will begin to separate during injection, causing
cooling and solidication period, the screw the melt to squirt out, or ash, at the mold
begins to rotate and melt new material for the parting line, potentially causing an incomplete
next shot. mold-ll, or short shot.
The clamping system consists of a xed Contemporary molding machines have two
platen and a movable platen, each of which has or more closing speeds: a high-speed closing,
half of the mold attached. The xed platen requiring only a low force, followed by a slow-
(shown in Fig. 6) has a hole in its center to allow speed closing stage, which generates the high
the injection cylinder nozzle to be placed in con- clamping force to close the mold rmly (stretch-
tact with the sprue bushing of the mold. A mov- ing the tie bars) prior to injection. The clamp
able platen is moved along tie bars by either usually opens slowly at rst, followed by a rapid
hydraulic or mechanical means, or a combina- traverse. These high-speed motions reduce the
tion of both. The amount of clamping force the time of the overall process cycle.
machine can apply, rated in tons, generally Injection Mold Design. Injection molds, in
determines the size of the part that the machine their simplest form, consist of two halves, often
can process. It is this clamping force that over- called the core and cavity (Fig. 8). A hole or
comes the injection pressure of the melt in the sprue bushing conducts the melt from the injec-
mold and keeps the mold halves together. The tion nozzle through the sprue and through a gate
into the mold cavity.
Gates and Runners. Commonly, more than
one gate is used to deliver material into the mold,
and each gate is fed by a runner channel lead-
ing from the sprue to the gate. After the mold is

Fig. 7 Stages in the operation of a reciprocating-screw injec-

tion molding machine Fig. 8 Two-plate injection mold
Chapter 4: Plastics / 87

lled and the melt has solidied, the material in incorporated into the mold to speed the solidi-
the sprue and runners also solidies and must be cation of the plastic. This usually consists of
removed with the part before the next shot. The holes bored in each half of the mold through
sprue and runner material are usually reground which a heat-exchange uid, usually water, can
in a granulator machine and fed back, along with circulate. The mold temperature is usually con-
virgin material, into the injection cylinder for trolled above room temperature. The optimal
reuse. To reduce the amount of material that mold temperature depends on the type of plastic
must be recycled, some molds are equipped with being molded, but typical mold temperatures for
a hot runner manifold (Fig. 9), which keeps the the more common thermoplastics vary from 40
material in the sprue and runners molten to to 120 C (100 to 250 F).
become a part of the next shot. Material Shrinkage. All thermoplastic so-
With larger and more complex parts, it is lidication is accompanied by a volumetric
common to use more than one gate for injecting shrinkage. For crystalline plastics, the shrinkage
material into the mold. The conguration of the is associated with crystallization. For amor-
runners and gates determines the way the mold phous plastics, the shrinkage is generally less
is lled. This can change the amount of injec- and is associated with the glass transition. In
tion pressure required, the location and condi- both cases, the amount of shrinkage depends on
tion of knit lines, the orientation of material various processing parameters, including the
ow (which can affect mechanical properties), mold temperature and rate of cooling of the melt.
and the ability to vent air from the mold. Knit This shrinkage can continue for a period of time
lines are surfaces along which the ow fronts after the part is removed from the mold. It is
meet and are generally weaker. Nonuniform important that the shrinkage be repeatable, so the
mold lling, caused by poor gate locations, can mold can be appropriately sized. For parts
result in overpacking the mold in the regions requiring tight dimensional control or optimal
which ll rst, causing residual stresses and mechanical properties, uniform cooling is essen-
possible warp-age in the part. Computer pro- tial. If the part does not solidify uniformly in the
grams are now available to assist in mold design mold, residual stresses will occur as a result of
and the layout of runners and gates for effective differential shrinkage. Computer programs are
mold lling at minimum pressure. available to assist in the optimal layout of cool-
Mold Cooling. With thermoplastic injec- ing channels for uniform part cooling.
tion molding, a method of cooling is usually Part Removal. The mold is usually designed
such that the part remains on the moving half of
the mold when it is opened. Ejector pins are then
actuated to separate the part from the mold.
These ejector, or knockout, pins are activated
either as a direct result of the moving platen
(mechanical knockouts), or hydraulically.
The geometry of the mold must be such that
the part can be readily removed after the material
has solidied, sometimes by means of a mechan-
ical part remover. This requires careful part de-
sign and selection of the parting surface across
which the two halves of the mold separate. Nei-
ther half of the mold can have undercuts or die
locks that would trap the part or keep it from
being ejected. If the part cannot be designed
without undercuts, then moving cores, slides, or
lifters must be incorporated to eliminate the
trapped condition when the mold is opened.
Mold Venting. Air must be removed as the
mold lls with plastic. This is usually accom-
plished by grinding channels, or vents, into the
parting line of the mold. These vents must be
narrow enough to keep molten plastic from
Fig. 9 Hot runner injection mold escaping, while allowing air to escape. They
88 / Gear Materials, Properties, and Manufacture

should be located adjacent to the last region of Handbook Desk Edition, ASM International,
the mold to ll. Sometimes undersized ejector 1995, p 299307
pins are also used to assist in venting the mold.
Insufcient venting can result in an incom-
pletely lled mold, or in cosmetic defects called REFERENCES
diesel burns, which are caused by the heating of
the air as it is compressed. 1. C.E. Adams, Plastics Gearing: Selection
Mold Materials. For high-volume gear and Application, Marcel Dekker, 1986
production applications, the mold cavity and 2. Materials for Plastic Gears, AGMA 920-
core are usually machined from special mold- A01, American Gear Manufacturers Asso-
making steels. Steel is chosen for its wear resis- ciation, 2001
tance and durability. 3. M.I. Kohan, Polyamides (PA), Engineering
Plastics, Vol 2, Engineered Materials
Handbook, ASM International, 1988,
4. D.P. Townsend, Ed., Dudleys Gear Hand-
A portion of this chapter was adapted from book, Second Edition, McGraw Hill, Inc.
Injection Molding, Engineered Materials 1991, p 17.2
Gear Materials, Properties, and Manufacture Copyright 2005 ASM International
J.R. Davis, editor, p89-127 All rights reserved.
DOI:10.1361/gmpm2005p089 www.asminternational.org


Machining, Grinding, and Finishing

METAL REMOVAL processes for gear cutting is called hobbing. Second, when the cut-
manufacture can be grouped into two general ting tool is shaped like a pinion or a section of a
categories: rack, it will be used in a cutting method called
shaping. Third, in the milling process, the cut-
Rough machining (or gear cutting) opera-
ting tool is a toothed disk with a gear tooth con-
tions such as milling, hobbing, and shaping
tour ground into the sides of the teeth. The fourth
which cut the tooth spaces into prepared
general method uses a tool (or a series of tools)
gear blanks leaving the desired teeth
that wraps around the gear and cuts all teeth at
Finishing (or high-precision machining)
the same time. Methods of this type are termed
operations such as shaving, grinding, hon-
broaching, punching, or shear cutting.
ing, and lapping which improve the accu-
Finishing. As listed earlier, there are several
racy and surface nish of previously pre-
methods available for the improving the quality
pared gear teeth
of gears following the standard rough machin-
Gear Cutting. The methods of gear cutting ing operation. Gear shaving is a chip forming
can be divided broadly into form cutting and nishing operation that removes small amounts
generating processes. The tooth prole in the of metal from the working surfaces of gear
former is obtained by using a form cutting tool. teeth. Its purpose is to correct errors in index,
This may be a multiple-point cutter used in a helix angle, tooth prole, and eccentricity.
milling machine or a broaching machine, or a Shaving is carried out prior to gear heat treat-
single-point tool for use in a shaper. A variation
of the form cutter method is based on transfer-
ring the prole from a templet. This method is
used in the shaperlike cutting of large tooth
forms and in cutting bevel gear teeth on a bevel
gear planer. In all these processes, the work-
piece is held stationary until a tooth is nished;
the piece is then indexed for successive teeth.
In the generating process, the tooth prole is
obtained by a tool that simulates one or more
teeth of an imaginary generating gear. A relative
rolling motion of the tool with the workpiece
generates the tooth surface. This method is used
in the hobbing, shaping, and milling processes
for manufacturing spur and helical gears and in
the face milling and face hobbing of bevel gears.
A wide variety of machines are used to cut
gear teeth. As shown in Fig. 1, there are four
more or less distinct ways to cut material from a
gear blank so as to leave a toothed wheel after
cutting. First, the cutting tool can be threaded
and gashed. If so, it is a hob, and the method of Fig. 1 Outline of methods of producing gear teeth
90 / Gear Materials, Properties, and Manufacture

ment. Grinding is a technique of nish machin- indexing mechanism. Most indexing techniques
ing utilizing a rotating abrasive wheel. Grinding used on modern gear milling machines incorpo-
is an effective means of nishing heat-treated rate numerical control or computer numerical
high-hardness steels (40 HRC and above). Gear control, and the accuracy can rival that of hob-
honing is a particularly effective method of bing machines.
removing nicks and burrs from the active pro- Broaching. Both external and internal gear
les of the teeth after heat treatment. Lapping is teeth, spur or helical, can be broached, but con-
a low-speed low-pressure abrading operation ventional broaching is usually conned to cut-
used to rene the tooth surface and to reduce ting teeth in internal gears. Figure 3 shows pro-
noise levels in gear sets. The mating gear and gressive broach steps in cutting an internal spur
pinion are run together under a controlled light gear. The form of the space between broached
load while a mixture of abrasive compound and gear teeth corresponds to the form of the broach
suitable vehicle are pumped onto the pair. As teeth. The cutting action of any single broach
with grinding and honing, lapping is also car- tooth is similar to that of a single form tool.
ried out after gear heat treatment. Each cross section of the broach has as many

Machining Processes for Gears

Simple gear tooth congurations can be pro-

duced by basic processes such as milling,
broaching, and form tooling. Complex gear
tooth congurations require more sophisticated
processes designed especially for the manufac-
ture of gears.

Processes for Gears Other

Than Bevel Gears
The methods used to cut the teeth of gears
other than bevel gears are milling, broaching,
shear cutting, hobbing, shaping, and rack cut-
ting. In any method, a xture must hold the gear
blank in correct relation to the cutter, and the
setup must be rigid.
Milling produces gear teeth by means of a
form cutter. The usual practice is to mill one
tooth space at a time. After each space is milled,
the gear blank is indexed to the next cutting
position. Fig. 2 Relation of cutter and workpiece when milling teeth in
a spur gear
Peripheral milling can be used for the rough-
ing of teeth in spur and helical gears. Figure 2
shows teeth in a spur gear being cut by periph-
eral milling with a form cutter. End milling can
also be used for cutting teeth in spur or helical
gears and is often used for cutting coarse-pitch
teeth in herringbone gears.
In practice, gear milling is usually conned to
one-of-a-kind replacement gears, small-lot pro-
duction, the roughing and nishing of coarse-
pitch gears, and the nish milling of gears hav-
ing special tooth forms. Although high-quality
gears can be produced by milling, the accuracy
of tooth spacing on older gear milling machines
Fig. 3 Progressive action of broach teeth in cutting teeth of an
was limited by the inherent accuracy of the internal spur gear
Chapter 5: Machining, Grinding, and Finishing / 91

teeth as there are tooth spaces on the gear. The The shear cutting head is mounted in a xed
diameter of each section increases progressively position, and the gear blanks are pushed through
to the major diameter that completes the tooth the head. Cutting tools are fed radially into the
form on the workpiece. Broaching is fast and head, a predetermined amount for each stroke,
accurate, but the cost of tooling is high. There- until the required depth of tooth space is
fore, broaching of gear teeth is best suited to reached. In shear cutting, some space is required
large production runs. for over-travel, although most workpieces with
Shear cutting is a high-production method integral shoulders or anges (such as cluster
for producing teeth in external spur gears. The gears) do have enough clearance between sec-
process is not applicable to helical gears. In tions to allow shear cutting to be used. There-
shear cutting, as in broaching, all tooth spaces fore, this process is best suited to large produc-
are cut simultaneously and progressively (Fig. tion runs.
4). Cutting speeds in shear cutting are similar to Hobbing is a practical method for cutting
those for broaching the same work metal. teeth in spur gears, helical gears, worms, worm
Machines are available for cutting gears up to gears, and many special forms. Conventional
508 mm (20 in.) in diameter, with face width up hobbing machines are not applicable to cutting
to 152 mm (6 in.). bevel and internal gears. Tooling costs for hob-
bing are lower than those for broaching or shear
cutting. Therefore, hobbing is used in low-
quantity production or even for a few pieces. On
the other hand, hobbing is a fast and accurate
method (compared to milling, for example) and
is therefore suitable for medium and high pro-
duction quantities.
Hobbing is a generating process in which
both the cutting tool and the workpiece revolve
in a constant relation as the hob is being fed
across the face width of the gear blank. The hob
is a uted worm with form-relieved teeth that
cut into the gear blank in succession, each in a
slightly different position. Instead of being
formed in one prole cut, as in milling, the gear
teeth are generated progressively by a series of
cuts (Fig. 5). The hobbing of a spur gear is
Fig. 4 Progressive action in shear cutting teeth of an external shown in Fig. 6.
spur gear. Shear cutting operation proceeds from
roughing (a) to intermediate (b) to nishing (c) operations.

Fig. 5 Schematic of hobbing action. Gear tooth is generated

progressively by hob teeth. Fig. 6 Hobbing of a spur gear
92 / Gear Materials, Properties, and Manufacture

Gear shaping is the most versatile of all Rack cutting is done with a cutter in the
gear cutting processes. Although shaping is form of a rack with straight teeth (usually three
most commonly used for cutting teeth in spur to ve). This cutter reciprocates parallel to the
and helical gears, this process is also applicable gear axis when cutting spur gears and parallel to
to cutting herringbone teeth, internal gear teeth the helix angle when cutting helical gears. Metal
(or splines), chain sprockets, ratchets, elliptical is removed by a shaper-like stroke similar to the
gears, face gears, worm gears, and racks. Shap- cutting action in gear shaping. In addition to the
ing cannot be used to cut teeth in bevel gears. reciprocating action of the rack cutter, there is
Because tooling costs are relatively low, shap- synchronized rotation of the gear blank with
ing is practical for any quantity of production. each stroke of the cutter, with a corresponding
Workpiece design often prevents the use of advance of the cutter as in the meshing of a gear
milling cutters or hobs (notably, for cluster and rack. By these combined actions, the true
gears), and shaping is the most practical method involute curve of the gear tooth is developed.
for cutting the teeth. Several gear cutting machines use this principle.
Gear shaping is a generating process that uses Rack cutters are less expensive than hobs. Rack
a toothed disk cutter mounted on a spindle that cutting is especially adapted to the cutting of
moves in axial strokes as it rotates. The work- large gears or gears of coarse pitch or both.
piece is carried on a second spindle. The work- Gears with diametral pitch as coarse as are
piece spindle is synchronized with the cutter commonly cut by the rack method.
spindle and rotates as the tool cuts while it is
being fed gradually into the work. The action Processes for Bevel Gears
between a shaping cutter and a gear blank is The machining of bevel gears is treated as a
illustrated in Fig. 7. Shaping applied to the cut- separate subject because most bevel gears are
ting of a worm (Fig. 8) involves no axial stroke cut in special machines with special cutters.
of the cutter spindle. However, the action of these cutters bears a
close resemblance to one or more of the basic
processesmilling, broaching, or shaping.
Both generating and nongenerating processes
are discussed below.
Milling is not widely used for cutting bevel
gears, because of the accuracy limits of index-
ing devices and because the operation is time
consuming (as many as ve cuts around the gear
may be required for completing one gear).
Straight bevel gears are sometimes roughed by
Fig. 7 Relation of cutter and workpiece in shaping gear teeth milling and then nished by another method.
This two-operation procedure is more common
when the availability of special gear cutting
machines is limited.
Template machining is a low-production,
nongenerating method used to cut the tooth pro-
les of large bevel gears using a bevel gear
planer (Fig. 9). Because the setup can be made
with a minimum of tooling, template machining
is useful when a wide variety of coarse-pitch
gears are required. Template machining uses a
simple, single-point cutting tool guided by a
template several times as large as the gear tooth
to be cut. Under these conditions, high accuracy
in tooth forms is possible.
The necessary equipment is unique. The
setup utilizes two templates; one for each side of
the gear tooth. In theory, a pair of templates
Fig. 8 Relation of cutter and workpiece in generating a worm would be required for each gear ratio, but in
by shaping practice a pair is designed for a small range of
Chapter 5: Machining, Grinding, and Finishing / 93

ratios. A set of 25 pairs of templates encom- Roughing and nishing are sometimes both
passes all 90 shaft angle ratios from 1:1 to 8:1 done by one cutter in the same machine. More
for either 14 or 20 pressure angles. This sys- often, for greater efciency, roughing and n-
tem of templates is based on the use of equal- ishing operations are done in different
addendum tooth proportions for all ratios. The machines.
tooth bearing localization is produced by a The Formate method can be used with the
slight motion of the tool arm as the cutting tool two-cut roughing and nishing method, in
moves along the tooth. The length and position which the gear is roughed to depth and then sin-
of the cut can be controlled by the machine gle-cycle nished, or with a one-cut roughing
operator. and nishing method known as completing. In
To produce a nished gear, ve or six cutting the completing method, the cutter is plunged to
operations are required. The rst is a roughing approximately 0.25 mm (0.010 in.) of whole
operation, made by feeding a slotting tool or a depth, the cutting speed is doubled, and the cut-
corrugated V-tool to full depth. If the rst cut is ter is fed to whole depth, taking a light chip.
made with a slotting tool, a second cut is This nal portion of the cycle removes the built-
required with a V-tool or a corrugated V-tool. up edge on the cutter and produces an accept-
Cuts are made with the template follower rest- able surface nish. The completing method is
ing on a straight guide. After roughing, the tem- used for both face milling and face hobbing.
plates are set up, and the teeth are nished by In the Formate single-cycle nishing method,
making two cuts on each side. Slotting tools are one tooth space is nish cut in one revolution of
specied by point width and depth of cut; corru- the cutter. Stock removal is accomplished by
gated roughing tools, by point width, depth of cutting blades mounted in a circular cutter that
cut, and pressure angle; nishing tools, by point resembles a face milling cutter. Each blade in
width only. To set up a straight bevel gear the cutter is slightly longer and wider than the
planer for template machining, the operator preceding blade; thus, the cutting action is, in
need know only the tooth proportions of the effect, that of a circular broach. A gap between
gear to be cut, plus the template list and the the rst and last cutting blades permits indexing
index gear list furnished with the machine. of the workpiece as each tooth space is com-
Formate cutting and Helixform cut- pleted. The relative positions of cutter and gear
ting are nongenerating methods for cutting spi- during Formate single-cycle cutting are illus-
ral bevel and hypoid gears. Nongenerating trated in Fig. 10.
methods can be used for cutting the gear mem- Helixform cutting, another nongenerating
ber of spiral bevel and hypoid pairs when the method of cutting spiral bevel and hypoid gears,
gear-to-pinion ratio is 2.5:1 or greater. The two is generally similar to Formate cutting. How-
principal methods are Formate cutting and ever, there is one signicant difference in the
Helixform cutting. method and in the nishing operation for the
Formate cutting is applicable to both the gear member.
roughing and single-cycle nishing of gears
with pitch diameters up to 2540 mm (100 in.).

Fig. 9 Cutting teeth in a large straight bevel gear by template Fig. 10 Relative positions of cutter and workpiece in For-
machining in a bevel gear planer mate single-cycle cutting
94 / Gear Materials, Properties, and Manufacture

One turn of the Helixform cutter nishes both the slotted head. This cutter usually has parallels
sides of a tooth space. The cutter has both rota- for changing diameter, and adjusting wedges for
tional and reciprocating motion, and this combi- truing individual blades. The blades are sharp-
nation makes the path of the cutter-blade tips ened in a radial or near-radial plane. The second
tangent to the root plane of the gear. Because the type of inserted-blade cutter has blades that are
cutting edge is a straight line, the surface cut by clamped in the slot. The blades are sharpened by
a Helixform gear cutter is a true helical surface. topping the blades down and repositioning them
The principal advantage of Helixform cutting in the head.
compared to Formate cutting is that the gear For nishing, the three types of cutters can be
produced by Helixform is conjugate to the mat- furnished with all outside blades, all inside
ing pinion, and the resulting contact pattern has blades, or alternate outside and inside blades.
little or no bias. Roughing cutters and completing cutters can
The Cyclex method is also a nongenerat- have either alternate inside and outside blades or
ing method, and the result is the same as that for have end-cutting or bottom-cutting blades alter-
the Formate method. The Cyclex method was nately spaced with inside and outside blades.
developed for the rough and nish cutting of There are four basic cutting methods: com-
gears in one operation and is particularly suit- pleting, single-side, xed-setting, and single-
able when production quantities are not great setting. In each case, the rotating cutting edges
enough to warrant separate Formate roughing of a face mill cutter represent the imaginary gear
and nishing machines for the gear member. surface.
Cyclex machines of the generator type can cut a The completing method is the generation of
wide range of gear sizes and can be used for cut- the part by a circular face mill or face hob cutter
ting both gears and pinions. with alternate inside and outside blades that cut
In the Cyclex method, the gear is roughed and the tooth surfaces on both sides of a tooth space
nished in one chucking from the solid blank. at the same time. With this method, each mem-
The nishing blades of the cutter are set below ber is nished in one operation.
the roughing blades and do not contact the work In the single-side method, the part is n-
during the roughing cycle. Several revolutions ished by a circular face mill cutter with alternate
of the cutter may be necessary for roughing, the inside and outside blades that cut the tooth sur-
number required depending on the pitch of the face on each side of a tooth space in separate
gear. In the nal revolution, after the last rough- operations with independent machine settings.
ing blade has passed through the cut, the work is In the xed-setting method, the part is n-
rapidly advanced, permitting the nishing blades ished by two circular face mill cutters: one with
to make contact and nish the tooth space to size. inside blades only for cutting the convex side of
After the nishing blades have passed through the tooth, and the other with outside blades only
the cut, the work is rapidly withdrawn and for cutting the concave side. The two sides of
indexed, and the cycle is repeated until all teeth the tooth are produced separately in two entirely
are completed. different machine setups. For large production
Face mill cutting machines are used to n- runs, a pair of machines is used. One machine is
ish cut teeth in spiral bevel, Zerol, and hypoid for cutting one side of the tooth, and the other is
gears. Machines and cutters are available for cut- for the other side of the tooth.
ting gears ranging from small instrument gears The single-setting method is a variation of
up to about 2540 mm (100 in.) in diameter. the completing method and is used when the
The three types of face mill cutters are identi- available cutters have point widths too small for
ed by the design of their cutting blades: inte- completing cutting. Both sides are cut with the
gral, segmental, and inserted. All of these can be same machine settings, and the blank is rotated
used for both roughing and nishing. Solid or on its axis to remove the amount of stock neces-
integral-blade cutters are made from a single sary to produce the correct tooth thickness. After
piece of tool steel and are usually less than 152 the rst cut, only one side of the cutter is cutting
mm (6 in.) in diameter. They are used for ne- in the tooth slot. Figure 11 illustrates the action
pitch gears. Segmental cutters are made up of of a face mill when generating pinion teeth by
sections, each having two or more blades. The the xed-setting method (inside blades).
segments are bolted to the cutter head around Face hob cutting is similar to face milling
the periphery. Inserted-blade cutters are of two except that the indexing is superimposed on the
types. The rst type has blades that are bolted to generating cycle. Cutters are arranged with
Chapter 5: Machining, Grinding, and Finishing / 95

blade groupings. As each blade group passes Interlocking cutters, known also as com-
through the cut, the work is being indexed one pleting generators, generate the teeth on straight
pitch (Fig. 12). Gears cut by this method are bevel gears or pinions from a solid blank in one
generally completed in one operation. Most face operation.
hob cutters are of the inserted-blade type. Inte- In this method, two interlocking disk-type
gral and segmental systems are available. cutters rotate on axes inclined to the face of the
mounting cradle, and both cut in the same tooth
space (Fig. 13). The cutting edges present a con-
cave cutting surface that removes more metal at
the ends of the teeth, giving localized tooth con-
tact. The gear blank is held in a work spindle
that rotates in timed relation with the cradle on
which the cutters are mounted. A feed-cam
cycle begins with the workhead and blank mov-
ing into position for rough cutting, without gen-
erating roll, and cutting proceeds until the cut is
just short of full depth. After a rough generating
roll, the work is fed in to full depth, and a fast
up-roll nish generates the tooth sides. At the
Fig. 11 Face mill cutter shown in position to generate a pin- top of the roll, the work backs out, and the cra-
ion by the xed-setting method dle and work spindle roll down again into

Fig. 12 Schematic of the face hob cutting method

96 / Gear Materials, Properties, and Manufacture

roughing position. During this short down-roll, entire cutter into the work. The completing cut-
the blank is indexed. ter contains three kinds of blades: roughing,
Revacycle is a generating process used for seminishing, and nishing. One revolution of
cutting straight bevel gears up to about 255 mm the cutter completes each tooth space, and the
(10 in.) pitch diameter in large production runs. work is indexed in the gap between the last n-
This is the fastest method for producing straight ishing blade and the rst roughing blade. For the
bevel gears of commercial quality. Initial tool- small amount of Revacycle work that is too deep
ing cost is greater for the Revacycle process to be completed in one cut, separate roughing
than for other processes for cutting straight and nishing operations are used. Under these
bevel gears, but the high production rate results conditions, separate cutters and setups are re-
in the lowest cost for mass production. quired for each operation. The cutters and ma-
Most gears produced by the Revacycle chine cycles are similar to those for completing
method are completed in one operation, using cutters, except that the roughing cutters have no
cutters 406, 457, 533, or 635 mm (16, 18, 21, or seminishing or nishing blades. A second cut-
25 in.) in diameter that rotate in a horizontal ter has only seminishing and nishing blades.
plane at a uniform speed (Fig. 14). The cutter Two-tool generators are used for cutting
blades, which extend radially outward from the straight bevel gears by means of two reciprocat-
cutter head, have concave edges that produce ing tools that cut on opposite sides of a tooth
convex proles on the gear teeth. During cut- (Fig. 15). Tooling cost is low for two-tool gen-
ting, the work-piece is held motionless while erators, but production rates are lower than
the cutter is moved by means of a cam in a those for other straight bevel generators, such as
straight line across the face of the gear and par- interlocking cutters and Revacycle machines.
allel to its root line. This motion produces a Two-tool generators are usually used when:
straight tooth bottom while the desired tooth
shape is being produced by the combined effect The gears are beyond the practical size range
of the motion of the cutter and the shapes of the (larger than about 254 mm, or 10 in., pitch
cutter blades. The cutter is actually a circular diameter) of other types of generators
broach in that each successive cutting tooth is Gears have integral hubs or anges that proj-
larger than the one that precedes it along the cir- ect above the root line, thus preventing the
cumference of the cutter. The cutter makes only use of other generators
one revolution per tooth space. A small production quantity or a variety of
Feed is obtained by making cutter blades pro- gear sizes cannot be accommodated by other
gressively longer, rather than by moving the types of machines used for cutting straight
bevel gears
Two-tool generators are used for both rough
and nish cutting. When warranted by produc-

Fig. 13 Relation of interlocking cutters (completing genera-

tors) with the bevel gear being cut Fig. 14 Revacycle cutter in position to cut a bevel gear
Chapter 5: Machining, Grinding, and Finishing / 97

tion quantities, roughing is done in separate Most two-tool generators can produce
machines, which are the same as the generators straight bevel gears with teeth crowned length-
except that the machines used only for roughing wise to localize tooth contact. Crowned teeth
have no generating roll. For small production are produced by means of two angularly
quantities, both roughing and nishing cuts are adjustable guides on the back of each slide. The
made in the generators, the roughing cut being guides ride on a pair of xed rollers (Fig. 16).
made without generating roll. When the guides are in line with each other, the
To make the machine setup for producing a tool stroke is a straight line, and when they are
gear that will operate at right angles to its mat- out of line, the tool is stroked along a curved
ing gear, the operator must have the gear speci- path. The amount of curvature is controlled by
cations and one calculated machine setting setting the two guides. A table with each
called the tooth angle (Fig. 15). The remaining machine lists the guide settings for making the
setup data are taken from tables furnished with tooth contact approximately one-half the face
the machine. When the shaft angle of the two width. The machine settings can be varied to
gears is not a right angle, the ratio of roll, as well shorten or lengthen the tooth contact.
as the data required for checking the roll, must Planing generators are unique because they
be calculated. can cut both straight-tooth and curved-tooth
bevel gears. However, the use of planing gener-
ators is ordinarily restricted to cutting gears
about 889 mm (35 in.) in diameter or larger or to
diametral pitch coarser than 1. Standard
machines can generate straight, Zerol, and spiral
bevel gears. Special heads can be added to stan-
dard machines to permit the cutting of hypoid
Tools have straight cutting edges and are
mounted on a reciprocating slide that is carried
on the face of the cradle and connected to a rotat-
ing crank by a connecting rod (Fig. 17). Tooth
proles are made by rolling the work with the
generating gear. The lengthwise shape of the
teeth is formed by a combination of three

Fig. 15 Angle of straight bevel gear tooth and sections of Stroke of the tool
tools used for two-tool generating Continuous, uniform rotation of the work

Fig. 16 Provision for crowning gear teeth by means of

adjustable guides in two-tool generators Fig. 17 Components of a planing generator
98 / Gear Materials, Properties, and Manufacture

An angular oscillation of the work produced Quantity requirements

by the eccentric shown in Fig. 17 Accuracy requirements
The eccentric motion modies the effect that the Gear-to-pinion ratio
rst two motions have on the shape of the teeth. Cost
The eccentric is timed for the correct tooth
The following sections consider the type of gear
as the major variable and discuss the machining
The spiral angle of the teeth is controlled by
methods best suited to specic conditions.
the angular offset of the tool slide from the
angle of the cradle axis. Continuous rotation of
Machining of Spur Gears
the work is principally for indexing. In effect,
the tool makes a cut on all teeth in succession in Milling, shear cutting, hobbing, and shaping
one generating position, and then the cradle and are the methods most commonly used for cut-
the work roll together a slight distance before ting teeth in spur gears.
another cut is taken on all teeth in the new gen- Form milling, with the cutter ground to the
erating position. Actually, however, rolling is desired shape of the tooth space (Fig. 2), is a
continuous and occurs gradually until all teeth simple means of cutting teeth in spur gears.
are completely generated in the last pass around Tooling cost is low, and the process requires
the gear. only a conventional milling machine, a form
Several passes are required to complete a gear, cutter, and an indexing mechanism. Except for
the number depending on tooth depth and shape. low-quality production, milling is seldom used
Flat gear blanks are usually roughed without a for cutting spur gears. The main disadvantage in
roll, rst using a corrugated tool and then using a the form milling of spur gears is the lack of
single cut with a V-roughing tool. This is fol- accuracy in tooth spacing, which depends on the
lowed by at least two side-cutting operations on accuracy of the indexing mechanism. In addi-
each side of the tooth, including generating cuts tion, form milling is much slower than shear
with roll. A similar sequence is used for cutting cutting or hobbing.
pinions except that roughing is done with roll. One milling cutter is not universal for all num-
Tools for use in planing generators are simple bers of teeth, as are hobs and shaper cutters. To
and inexpensive. Corrugated tools are furnished produce theoretically correct gear teeth, the
with a 14 pressure angle, regardless of the tooth form of the cutter must be designed for the
pressure angle of the gear being cut, but point exact number of teeth. However, if a small
width and depth of cut must be specied when departure in tooth form is acceptable, cutters
the tools are ordered. The operator must know have been standardized for a range of teeth, the
the specications for the gear being cut. A table form being correct for the lowest number of teeth
of settings for specic requirements is supplied in that particular range. Thus, all teeth within the
with the machine. range are provided with sufcient tip relief. The
same form is produced in all tooth spaces within
that range. For reasonably accurate gear cutting,
Selection of Machining Process eight standard involute gear cutters are required
to cut all sizes of gears of a given pitch:
Each gear cutting process discussed in the
preceding sections has a eld of application to Cutter No. Gear tooth range

which it is best adapted. These elds overlap, 1 135 to rack

however, so that many gears can be produced 2 55134
3 3554
satisfactorily by two or more processes. In such 4 2634
cases, the availability of equipment often deter- 5 2125
mines which machining process will be used. 6 1720
7 1416
The type of gear being machined (spur, heli- 8 1213
cal, bevel, or other) is usually the major factor in
the selection of machining process, although one
If a more accurate tooth form is desired for
or more of the following factors usually must be
gears near the higher part of the ranges, seven
considered in the nal choice of the method:
half-number cutters can be obtained. A No. 1
Size of the gear cutter, for example, will cut gears with 80 to 134
Conguration of integral sections (anges or teeth; a No. 2 cutter, 42 to 54 teeth; and a No.
other) 3 cutter, 30 to 34 teeth. For still greater accu-
Chapter 5: Machining, Grinding, and Finishing / 99

racy, cutters of the proper shape for an exact often utilized for cutting two or more gears at
number of teeth can be furnished by most tool one time when the ratio of face width to pitch
manufacturers on short notice. diameter is small.
The tooth form of a single cutter is centered Shaping can produce high accuracy in cut-
with the gear axis so that a symmetrical tooth ting spur gears because shaping is a generating
space is produced. By the use of gang cutters, process. Although seldom as fast as hobbing,
portions of adjacent tooth spaces can be rough shaping is used for a wide range of production
machined simultaneously. Normally, a rough- quantities. Many types of gears can be produced
ing and a nishing cutter are ganged, the nish- to requirements by either shaping or hobbing,
ing cutter being centered with the gear axis. and the availability of equipment determines
Gang cutters and multiple-tooth cutters are spe- which of the two processes is used. However, if
cially designed for the specic application. the workpiece conguration cannot be hobbed,
Shear cutting is faster and more accurate shaping is often the only practical method.
than milling for cutting teeth of almost any Cutting of teeth in cluster gears that must
involute modication in spur gears. Total cut- meet close tolerances is sometimes a problem
ting time is often less than 1 min for gears up to because the method used must frequently be
about 152 mm (6 in.) in diameter. However, restricted to shaping and shaving. The same
tooling cost is high, and shear cutting is there- quality requirements cannot be met by shaping
fore practical only for large-scale production. and shaving as by hobbing and grinding. When
Hobbing is the process most widely used tolerances for cluster gears are closer than can
for cutting teeth in spur gears, usually for one or be met by shaping and shaving, a corrective pro-
more of the following reasons: cedure sometimes must be employed. A cluster
gear can be hobbed to greater precision by sep-
High accuracy in a wide range of gear sizes arating it into a two-piece assembly that is
Flexibility in quantity production rigidly attached using threaded fasteners.
Low cost
Adaptability to work metals having higher- Machining of Helical Gears
than-normal hardness
Milling, hobbing, shaping, and rack cutting
The shape of the workpiece sometimes limits are methods most used for producing teeth in
the use of hobbingfor example, if the teeth to helical gears. Rack cutting is most often used
be cut are close to another portion of the work- for large gears. Identical machining methods are
piece having a diameter larger than the root applicable to conventional helical and crossed-
diameter of the gear. The axial distance between axes helical gears.
the two sections must be large enough to allow Milling is used less than any other method
for hob overtravel at the end of the cut. This because of the difculty in obtaining accuracy
over-travel is about one-half the hob diameter. and productivity. However, for some low pro-
The clearance required between the gear being duction requirements, milling is the most satis-
cut and any ange or other projecting portion of factory method because the tooling cost is low.
the workpiece is, therefore, about one-half the In low-volume production, milling is some-
hob diameter plus additional clearance to allow times used for roughing only, and the gear is n-
for the hob thread angle. ished by hobbing or shaping.
In many cases, it is necessary to cut teeth on The milling of helical gears usually requires
heat-treated gear blanks to avoid the difculties cutters specially designed for the specic gear. In
caused by distortion in heat treating. Hobbing is milling helical teeth, the cutter travels along the
especially suitable for cutting gear teeth in hard- helix angle of the gear. At this setting, the cutter
ened steel (sometimes as hard as 48 HRC). Al- axis of rotation is in the normal plane through the
though hob wear increases rapidly as workpiece center of the gear tooth space. Under these condi-
hardness increases, normal practice should pro- tions, only one point on the nished prole is pro-
duce an acceptable number of parts per hob duced in this normal plane. All the others are pro-
sharpening. Success in hobbing gears at high duced in different planes. Therefore, the form of
hardness depends greatly on maintaining mini- the cutter teeth is not reproduced in the gear. In
mum backlash in the machine and on rigid addition to the setting angle, the diameter of the
mounting of both the hob and the workpiece. cutter affects the gear tooth form and must be
The ability to cut teeth in two or more identi- considered in designing the cutter.
cal spur gears in one setup can also justify use of Hobbing is extensively used for generating
the hobbing method. Inexpensive xturing is the teeth of helical gears for any production vol-
100 / Gear Materials, Properties, and Manufacture

ume. With the exception mentioned below, pro- Rotary cutters such as form milling cutters
cedures for hobbing helical gears are the same as and hobs can be used to cut herringbone teeth
those for spur gears: When hobbing spur gears only when there is a gap wide enough to permit
with a single-thread hob, the blank rotates one cutter runout between the right-hand and left-
tooth space for each rotation of the hob, the rota- hand helixes. Hobbing machines have been
tion being synchronized by means of change built that can cut herringbone teeth in gears up
gears. When hobbing helical gears, the rotation to 5590 mm (220 in.) in diameter.
of the work is retarded or advanced, through the End milling can also be used for machining
action of the machine differential, in relation to teeth in herringbone gears, regardless of whether
the rotation of the hob, and the feed is also held in the gears have center slots. The end mills for cut-
denite relation to the work and the hob. The ting herringbone gears are used in special ma-
decision to advance or retard the workpiece rota- chines. Many large-diameter herringbone gears
tion depends mainly on whether the hob is a are cut by end milling.
right-hand or left-hand type or whether the helix Shaping is also a suitable method for cutting
angle is of right-hand or left-hand conguration. teeth on herringbone gears; those designed with a
The amount by which the workpiece is retarded center slot as well as the continuous herringbone
or advanced depends on the helix angle. In can be shaped. The type of shaper used for cutting
medium-to-high production, it is common to use herringbone gears is similar in principle to the
xtures that allow hobbing of two or more iden- type used for helical gears, except that for her-
tical gears in one loading of the machine. ringbone gears two cutters, one for each helix, are
Gears with integral shanks can usually be operated simultaneously. Both cutters recipro-
hobbed without difculty. The shanks can assist cate, one cutting in one direction to the center of
in xturing and handling for loading and unload- the gear blank and the other cutting to the same
ing. When warranted by high-volume produc- point from the opposite direction when the
tion, hobbing can be done in automatic machines motion is reversed. The cutters not only recipro-
utilizing automatic unloading and loading. cate, but also rotate. Both the gear blank and the
Although hob life decreases as workpiece cutters turn slowly, thus generating the teeth the
hardness increases, helical gears of hardness as same way as in a conventional shaper.
high as 48 HRC are sometimes hobbed. When
hardnesses of 48 HRC or lower can be tolerated, Machining of Internal Gears
the sequence of rough hobbing, heat treating, Broaching, shear cutting, and shaping are the
and nish hobbing is likely to cost less than methods most frequently used for cutting inter-
grinding after heat treatment. nal parts. Milling is seldom used, except for
Shaping is a practical process for generat- some very large gears.
ing teeth of helical gears having helix angles up The broaching of internal spur gear teeth is
to 45. The only difference between shaping restricted to workpieces having congurations
helical gears and spur gears is that the machines that permit the broach to pass completely
used for cutting helical gears must impart addi- through the piece. The action of a broach in cut-
tional rotary motion to the cutter spindle as it ting internal gear teeth is shown in Fig. 3. The
reciprocates. The amount of rotation per stroke broaching of gear teeth, which is similar to other
is controlled by a helix guide. The lead of the types of broaching, is discussed in the section
guide must be the same as the lead of the cutter. Broaching in this chapter. Broaching is an
Workpiece conguration is often the main extremely fast and accurate means of machining
factor in the selection of shaping as a process for internal gear teeth, but tooling cost is high;
cutting helical gears. For example, rotary cut- therefore, broaching is practical only for high-
ters such as hobs cannot be used when the teeth volume production.
being cut are too close to the ange. Shear cutting is applicable to internal gear
teeth. The principle involved is essentially the
Machining of Herringbone Gears
same as that illustrated in Fig. 4 for cutting
Milling, hobbing, and shaping are the meth- external spur gears, except that the cutting edges
ods most often used for cutting herringbone and direction of radial feed are reversed. Unlike
gears. Selection of method depends largely on broaching, shear cutting is not restricted to parts
whether the gear is designed with a gap between where the tool must pass completely through the
the two helixes or whether the herringbone is piece. Shear cutting can be used with no more
continuous. than a 3.2 mm ( in.) relief groove between the
Chapter 5: Machining, Grinding, and Finishing / 101

end of the cut and a shoulder. Because tooling hobs operate, the teeth at the entering end are
cost is high for shear cutting, the process is prac- chamfered to reduce the cutting load. Hobs for
tical only for high-volume production. cutting worms are made to the same tolerance
Shaping is applicable to cutting internal standards as those for cutting spur gears. When
gear teeth. Tooling cost is lower than that for it is necessary to increase the hob diameter to
broaching and shear cutting; therefore, shaping provide more utes, the tolerances are increased
is applicable to low-volume production. Shap- proportionately. The number of utes in a worm
ing is less restricted to specic congurations hob is increased to improve surface nish,
than broaching is, because gear teeth can be cut because the greater the number of utes, the
to within 3.2 mm ( in.) of a shoulder; however, smaller the feed marks.
for adequate chip clearance and to avoid the Shaping. Worms can also be generated by a
danger of striking the shoulder, it is better to shaper-cutter. In this technique, a helical gear
have ample clearance. Shaping is often the only cutter is used in a special machine similar to a
practical method for cutting teeth in large inter- hobber. Both the work and the cutter rotate, and
nal gears because the cost of large broaches or the cutter is rolled axially along the worm, pro-
shear cutting tools is prohibitive. viding true generating action (Fig. 8).

Machining of Worms Machining of Racks

Several types of worms are used for power Milling and shaping are used to cut teeth in
transmission, among them the double-envelop- spur and helical racks.
ing type (Fig. 18), also known as the hour-glass Milling can be used to produce teeth in both
worm because of its shape. Milling, hobbing, spur and helical racks. The milling cutter must
and shaping are used to machine the various have the exact tooth space form. Racks can be
types of worms. cut in any standard milling machine; require-
Milling. For double-thread worms of low ments are essentially the same as those for a
lead angle and commercial accuracy, a duplex conventional milling operation, and the rack to
cutter can be used. Each milling cutter is spe- be cut must be rigidly clamped. Either manual
cially designed for cutting a specic worm. or automatic, indexing mechanisms are avail-
Another technique for machining worms uti- able for milling all sizes of racks in high-volume
lizes the multiple-thread cutter. The cutter is set production.
with its axis parallel to the work axis and is fed The shaping of spur and helical racks
to depth. The work then makes one revolution involves the rolling action of the operating pitch
for completion. The infeed can be made auto- circle of the generating shaper-cutter along the
matic, and because no indexing is required, this corresponding pitch line of the rack. In cutting
method is adaptable to volume production. racks on a gear shaper, the machine is equipped
Hobbing produces the highest-grade worm with a special xture to hold the work. Several
at the lowest machining cost, but hobbing can arrangements are used for imparting a trans-
be used only when production quantities are verse indexing movement to the member carry-
large enough to justify the tooling cost. Because ing the rack. One method employs a face gear
of the large helix angle at which most worm secured on the work spindle that meshes with a
pinion. The latter, by means of change gears,
drives a lead screw, which operates the slide
carrying the rack. Another method is to attach a
pinion to the work spindle, which meshes with a
master rack attached directly to the slide carry-
ing the rack being cut. The rst method is nec-
essary when high ratios are involved in the
drive; the second method needs only the regular
work change gears.

Machining of Bevel Gears

Face mill cutting, face hob cutting, Formate
Fig. 18 Mating of worm gear (worm wheel) and worm in a cutting, Helixform cutting, the Cyclex method,
double-enveloping worm gear set interlocking cutters, Revacycle, two-tool gener-
102 / Gear Materials, Properties, and Manufacture

ators, planing generators, and template machin- equipment and by the same general procedures.
ing are used to cut teeth in straight and spiral (Hypoid gears are by far the most numerous,
bevel gears. The fundamentals of these processes being used in quantities exceeding those of spi-
are discussed in the section Processes for Bevel ral and Zerol gears combined.) Pinions are gen-
Gears in this chapter. Choice of method de- erally cut by some type of generator. Gears may
pends mainly on the type of gear being cut or may not be cut in generators. When the gear-
(straight or spiral bevel), size, conguration, ac- to-pinion ratio is greater than about 2.5:1, it is
curacy requirements, and production quantities. common practice to cut the gears without gen-
Template machining and planing in a gear erating roll. Therefore, the Formate completing,
generator are more often used for cutting teeth Formate single-cycle, Helixform, and Cyclex
in gears larger than 813 mm (32 in.) outside methods are extensively used for cutting spiral
diameter. bevel gears having a ratio of 2.5:1 or greater.
Straight Bevel Gears. The two-tool gener- Nongenerated gears are less expensive than
ator is widely used for cutting straight bevel their generated counterparts, although there is a
gears and is especially well adapted to the cut- smaller difference in cost between the various
ting of gears in a wide range of sizes (up to methods for cutting spiral bevel gears (of less
about 889 mm, or 35 in., outside diameter) in than 813 mm, or 32 in., in outside diameter)
low-to-medium production quantities because than between the various methods for cutting
tool cost is low. Two-tool generating is also straight bevel gears.
adaptable to the cutting of gears that have pro-
truding portions (such as front hubs) that pre- Machining of Large Gears
clude the use of some processes. There is no one dimension that denes a large
Interlocking cutters provide a means of com- or a small gear. As various sizes are reached,
pleting straight bevel gears in one operation. some methods of manufacture become imprac-
The interlocking-cutter method is faster than tical, and other methods must be used.
two-tool generating, but more costly. Gears Herringbone Gears. The same conditions
without front hubs and less than 406 mm (16 in.) previously discussed in the section Machining
in outside diameter are best adapted to machin- of Herringbone Gears in this chapter also apply
ing with interlocking cutters. to the machining of large herringbone gear
The Revacycle process is the fastest method components.
for cutting straight bevel gears. Gear teeth are Spur and Helical Gears. Milling, hobbing,
often completed at the rate of 1.8 s per tooth. and rack cutting are the methods most com-
This method was primarily designed for cutting monly used for cutting large spur and helical
gears having up to 254 mm (10 in.) pitch diam- gears.
eter at 4:1 ratio with the pinion and having a Milling is the least expensive and the least
maximum face width of 29 mm (1 in.). In the accurate of these three methods; therefore, the
Revacycle method, the cutter must have an accuracy required in the gear will determine
uninterrupted path; therefore, the process can- whether or not milling can be used.
not cut gears that have front hubs. Because of Hobbing is more costly than milling, but
the high tooling cost, Revacycle cutting is eco- produces more accurate gears. Hobs are avail-
nomical only for high-volume production. able for cutting gears well over 2540 mm (100
Cost Versus Quantity (Straight Bevel in.) in outside diameter, provided the diametral
Gears). The two-tool generator, because it has pitch is ner than 1. Because of the size of the
the lowest tooling cost, is the most economical hob required and the limitations of hobbing
method for producing up to approximately 150 machines, it is difcult to hob gears of 1 diame-
pairs of gears, at which point the two-tool gen- tral pitch and coarser.
erator and interlocking-cutter methods are Once it has been decided that hobbing will be
equivalent. For large production runs, the two- used, it must be determined whether a ground
tool generating method becomes prohibitively hob will be required or whether an unground
expensive. The Revacycle and interlocking- hob will provide the required degree of accu-
cutter methods are equivalent in cost at about racy. The two types vary greatly in cost. How-
1200 pairs of gears, beyond which the Revacy- ever, if a ground hob is selected, extreme care
cle method is cheaper. must be used in resharpening; if this is not done,
Spiral Bevel Gears. Spiral, Zerol, and the original accuracy will not be maintained,
hypoid bevel gears are cut in the same type of and errors may occur in the gear tooth form.
Chapter 5: Machining, Grinding, and Finishing / 103

Rack Cutting. For machining large gears maximum accuracy in the shaved gear and max-
that have large teeth (coarser than 1 diametral imum cutter life, a minimum of stock should be
pitch), rack cutting (Fig. 19) is usually the most allowed for removal by shaving; the amount
practical method. Rack cutting may be less depends largely on pitch. As little as 0.008 to
expensive than hobbing, even when teeth are 0.025 mm (0.0003 to 0.001 in.) of stock should
ner than 1 diametral pitch. be left on gears having diametral pitch as ne as
Bevel Gears. Face milling and face hob- 48; 0.08 to 0.13 mm (0.003 to 0.005 in.) is
bing, two-tool generating, and planing generat- allowable for gears having diametral pitch of 2.
ing are the methods most commonly used for Operating Principles. The shaving opera-
cutting large straight and spiral bevel gears. tion is done with cutter and gear at crossed axes;
Large face mill generators that can cut gears helical cutters are used for spur gears, and vice
up to 2540 mm (100 in.) in outside diameter are versa. The action between gear and cutter is a
the fastest and most accurate machines for cut- combination of rolling and sliding. Vertical ser-
ting large bevel gears. rations in the cutter teeth take ne cuts from the
Two-tool generators offer a practical means proles of the gear teeth.
for cutting straight bevel gears having diameters During operation, the tip of the shaving cutter
up to about 889 mm (35 in.), face widths up to must not contact the root llet, or uncontrolled,
152 mm (6 in.), and teeth as coarse as 1 diame- inaccurate involute proles will result. For
tral pitch. Tooling cost for the two-tool generat- gears to be shaved, protuberance-type hobs that
ing method is also low. provide a small undercut at the ank of the tooth
Planing generators can be used for cutting may be preferred. This type of hob avoids the
straight bevel gears, but they are most widely initial tip loading of the shaving cutter.
used for cutting spiral bevel gears ranging from Shaving Cutters. A typical rotary gear-
889 to 1830 mm (35 to 72 in.) in outside diam- shaving cutter is shown in Fig. 20(b). This cutter
eter, with up to 254 mm (10 in.) face width and is serrated on the prole to form the cutting
teeth as coarse as diametral pitch. edges. The depth of the serrations governs total
cutter life in terms of the number of sharpenings
Shaving of Spur and Helical Gears permitted. A shaving cutter is sharpened by
Gear shaving is a nishing operation that regrinding the tooth proles, thus reducing the
removes small amounts of metal from the anks tooth thickness. This causes a reduction in oper-
of gear teeth. It is not intended to salvage gears ating center distance for the same backlash and
that have been carelessly cut, although it can in turn changes the operating pressure angle.
correct small errors in tooth spacing, helix These changes are compensated for by a change
angle, tooth prole, and concentricity. Shaving in addendum after resharpening. Tolerance is an
improves the nish on tooth surfaces and can important consideration in original purchase and
eliminate tooth-end load concentration, reduce
gear noise, and increase load-carrying capacity.
Shaving has been successfully used in nishing
gears of diametral pitches from 180 to 2. Stan-
dard machines and cutters are available for
shaving gears that range in size from 6.4 to 5590
mm ( to 220 in.) pitch diameter.
Leaving excessive stock for shaving will
impair the nal quality of the shaved gear. For

Fig. 20 Shaving of gears. (a) Work gear in mesh with shaving

Fig. 19 Rack-type cutter generating the teeth of a spur gear cutter. (b) Serrated gear-shaving cutter
104 / Gear Materials, Properties, and Manufacture

resharpening. Shaving cutters are manufactured radial and axial planes. In the axial traverse
to standardized tolerances, not unlike those of method of shaping, crowning is done by rocking
master gears. For example, cumulative tooth the worktable as it is reciprocated. In the higher-
spacing error can be held to 0.008 mm (0.0003 production angular traverse method, the cutter
in.) and prole to 0.00064 mm (0.000025 in.). is modied to provide crowning. The amount of
Because the engineering and facilities necessary crown varies, but usually 0.0003 to 0.0005
to produce such accuracy are not available mm/mm (0.0003 to 0.0005 in./in.) of face width
in most gear manufacturing plants, cutters are is sufcient.
ordinarily returned to a tool manufacturer for Speed and Feed. Although cutting speeds
resharpening. are always high, the optimum speed of rotation
Shaving Methods. Shaving is done by two for gear shaving varies considerably with work
basic methods: rack and rotary. metal hardness and composition. Speeds and
In rack shaving, the rack is reciprocated feeds for several steels and hardness ranges are
under the gear, and infeed takes place at the end given in Table 1.
of each stroke. Because racks longer than 508
mm (20 in.) are impractical, 152 mm (6 in.) is
the maximum diameter of gear that can be Cutter Material and Construction
shaved by the rack method.
Rotary Shaving. The several applications of High-speed tool steel is used almost exclu-
rotary shaving include underpass, modied sively as the material for cutting edges of gear
underpass, transverse, axial traverse, and angu- cutting tools. The steels most widely used are
lar traverse. Crown shaving can be incorporated the general-purpose grades such as M2 or M7.
in all of these modications. Grade M3 (higher in carbon and vanadium than
In rotary shaving, the cutter has the approxi- general-purpose grades) is also used in many
mate form of a gear (Fig. 20). The size of gear gear cutting applications and is often preferred
that can be shaved is limited by the machine to M2 and M7 for cutting quenched and tem-
rather than by the cutter. Rotary shaving can be pered alloy steels. The more highly alloyed
any of three types: underpass, modied under- grades of high-speed tool steel such as T15 or
pass, and transverse. M30 are recommended only for conditions
Underpass shaving is used on cluster gears or where greater red hardness is necessary. Such
gears with shoulders. To avoid interference with conditions include hard work metal, inadequate
the adjacent gear or shoulder, the cross-axes supply of cutting uid, or high cutting speeds.
angle is usually 4 to 6. The face of the tool must Cutters are made from these highly alloyed
be wider than the face of the shaved gear. grades only when the general-purpose grades
Because underpass shaving is a one-cycle, (or M3) have proved inadequate.
short-stroke process, it is the fastest method of Carbide cutters are used to hand nish gears
shaving. Disadvantages include relatively short cut by the face mill and face hob methods when
tool life and light stock removal, thus requiring small quantities of high-quality parts are
precise size control of the preshaved gear. needed. For most applications, carbide cutters
Modied underpass shaving is the most are not economical. However, one application
widely used method because it is a rapid one- in which they are a cost-effective tool is the
cycle process. Tool cost is moderate because the Tangear generator method.
cutter need be no wider than the gear and may Tangear Generator. Single-point cemented
be narrower. The high cross-axes angle of 30 to carbide cutting tools are rotated in opposite
60 promotes rapid stock removal and smoother directions on the peripheries of two cutter heads
surface nish. with horizontal and parallel axes on the Tangear
Transverse shaving is the slowest shaving generator as shown in Fig. 21. The workpiece,
method because multiple passes are required. It with its axis vertical, is rotated as it is fed hori-
is a method of handling gears much wider than zontally (typically a short distance of 8 mm, or
the cutter; therefore, cutter cost is moderate for 5
16 in.) between the cutter heads. The motions
gears with wide faces. are synchronized so that the cutters act as
Crown shaving is used to relieve load con- though in mesh with the gear and progressively
centration at the ends of gear teeth caused by the generate the teeth and form cut the root llets.
misalignment of axes in operation. Crowning is The teeth are rough cut as the gear moves to the
a modication of the tooth prole in both the cutters and are nish cut on the back stroke. The
Chapter 5: Machining, Grinding, and Finishing / 105

Table 1 Feeds and speeds for the shaving of carbon and low-alloy steel gears with high-speed
steel tools
Gear Feed per revolution Cutter pitch High-speed steel
tooth size of gear(a) line speed tool material

Material Hardness, HB Condition Module Diametral pitch mm in. m/min sfm ISO AISI

Wrought free-machining carbon steels

Low-carbon 100150 Hot rolled or 256 14 0.3 0.012 185 610 S4, S2 M2, M7
resulfurized annealed 53 510 0.2 0.008
1116 1119 21.5 1119 0.12 0.005
1117 1211 1 and ner 20 and ner 0.07 0.003
1118 1212 150200 Cold drawn 256 14 0.3 0.012 205 675 S4, S2 M2, M7
53 510 0.2 0.008
21.5 1119 0.12 0.005
1 and ner 20 and ner 0.07 0.003
Medium-carbon 175225 Hot rolled, 256 14 0.3 0.012 150 500 S4, S2 M2, M7
resulfurized normalized, 53 510 0.2 0.008
1132 1144 annealed, or 21.5 1119 0.12 0.005
1137 1145 cold drawn 1 and ner 20 and ner 0.07 0.003
1139 1146 325375 Quenched and 256 14 0.3 0.012 84 275 S4, S2 M2, M7
1140 1151 tempered 53 510 0.2 0.008
1141 21.5 1119 0.12 0.005
1 and ner 20 and ner 0.07 0.003
Low-carbon leaded 100150 Hot rolled, 256 14 0.3 0.012 215 700 S4, S2 M2, M7
12L13 normalized, 53 510 0.2 0.008
12L14 annealed, or 21.5 1119 0.12 0.005
12L15 cold drawn 1 and ner 20 and ner 0.07 0.003
200250 Hot rolled, 256 14 0.3 0.012 185 600 S4, S2 M2, M7
normalized, 53 510 0.2 0.008
annealed, or 21.5 1119 0.12 0.005
cold drawn 1 and ner 20 and ner 0.07 0.003

Wrought carbon steels

Low carbon 85125 Hot rolled, 256 14 0.3 0.012 160 525 S4, S2 M2, M7
1005 1015 1023 normalized, 53 510 0.2 0.008
1006 1016 1025 annealed, or 21.5 1119 0.12 0.005
1008 1017 1026 cold drawn 1 and ner 20 and ner 0.07 0.003
1009 1018 1029 225275 Annealed or 256 14 0.3 0.012 115 375 S4, S2 M2, M7
1010 1019 1513 cold drawn 53 510 0.2 0.008
1011 1020 1518 21.5 1119 0.12 0.005
1012 1021 1522 1 and ner 20 and ner 0.07 0.003
1013 1022
Medium carbon 125175 Hot rolled, 256 14 0.3 0.012 135 450 S4, S2 M2, M7
1030 1044 1526 normalized, 53 510 0.2 0.008
1033 1045 1527 annealed, or 21.5 1119 0.12 0.005
1035 1046 1536 cold drawn 1 and ner 20 and ner 0.07 0.003
1037 1049 1541 325375 Quenched and 256 14 0.3 0.012 84 275 S5 M3
1038 1050 1547 tempered 53 510 0.2 0.008
1039 1053 1548 21.5 1119 0.12 0.005
1040 1055 1551 1 and ner 20 and ner 0.07 0.003
1042 1524 1552
1043 1525
Wrought free-machining alloy steels
Medium-carbon 150200 Hot rolled, 256 14 0.3 0.012 150 500 S4, S2 M2, M7
resulfurized normalized, 53 510 0.2 0.008
4140 4145Se annealed, or 21.5 1119 0.12 0.005
4140Se 4147Te cold drawn 1 and ner 20 and ner 0.07 0.003
4142Te 4150 325375 Quenched and 256 14 0.3 0.012 76 250 S5 M3
tempered 53 510 0.2 0.008
21.5 1119 0.12 0.005
1 and ner 20 and ner 0.07 0.003
(a) Feed recommendations apply to conventional (axial-transverse) gear shaving. Feeds should be increased 100% for gears shaved by the diagonal (angular-transverse)
method. Source: Metcut Research Associates Inc.
106 / Gear Materials, Properties, and Manufacture

Table 1 (continued)
Gear Feed per revolution Cutter pitch High-speed steel
tooth size of gear(a) line speed tool material
Material Hardness, HB Condition Module Diametral pitch mm in. m/min sfm ISO AISI

Medium- and 150200 Hot rolled, 256 14 0.3 0.012 160 525 S4, S2 M2, M7
high-carbon normalized, 53 510 0.2 0.008
leaded annealed, or 21.5 1119 0.12 0.005
41L30 41L50 86L20 cold drawn 1 and ner 20 and ner 0.07 0.003
41L40 43L40 86L40 325375 Quenched and 256 14 0.3 0.012 84 275 S5 M3
41L45 51L32 tempered 53 510 0.2 0.008
41L47 52L100 21.5 1119 0.12 0.005
1 and ner 20 and ner 0.07 0.003
Wrought alloy steels
Low carbon 125175 Hot rolled, 256 14 0.3 0.012 145 475 S4, S2 M2, M7
4012 4621 6118 annealed, or 53 510 0.2 0.008
4023 4718 8115 cold drawn 21.5 1119 0.12 0.005
4024 4720 8617 1 and ner 20 and ner 0.07 0.003
4118 4815 8620 325375 Normalized 256 14 0.3 0.012 76 250 S5 M3
4320 4817 8622 or quenched 53 510 0.2 0.008
4419 4820 8822 and tempered 21.5 1119 0.12 0.005
4422 5015 9310 1 and ner 20 and ner 0.07 0.003
4615 5115 94B15
4617 5120 94B17
Medium carbon 175225 Hot rolled, 256 14 0.3 0.012 120 400 S4, S2 M2, M7
1330 4427 6150 annealed, or 53 510 0.2 0.008
1335 4626 81B45 cold drawn 21.5 1119 0.12 0.005
1340 50B40 8625 1 and ner 20 and ner 0.07 0.003
1345 50B44 8627 325375 Normalized or 256 14 0.3 0.012 69 225 S5 M3
4027 5046 8630 quenched and 53 510 0.2 0.008
4028 50B46 8637 tempered 21.5 1119 0.12 0.005
4032 50B50 8640 1 and ner 20 and ner 0.07 0.003
4037 5060 8642
4042 50B60 8645
4047 5130 86B45
4130 5132 8650
4135 5135 8655
4137 5140 8660
4140 5145 8740
4142 5147 8742
4145 5150 9254
4147 5155 9255
4150 5160 9260
4161 51B60 94B30
(a) Feed recommendations apply to conventional (axial-transverse) gear shaving. Feeds should be increased 100% for gears shaved by the diagonal (angular-transverse)
method. Source: Metcut Research Associates Inc.

relationship set between workpiece and cutter operations are made of solid high-speed tool
head velocities determines the helix angle. steel. Rotary cutters (hobs and milling cutters)
Conventional and crowned helical gears can for spur and helical gears can be either solid or
be cut with diameters from 19 to 102 mm ( to inserted blade. Economy is the governing factor;
4 in.), face widths up to 25 mm (1 in.), and helix the two methods of construction are equally sat-
angles from 10 to 40. Gears can be generated at isfactory in terms of producing acceptable gears.
the rate of about 600 per hour, six to ten times as Cutters less than about 75 mm (3 in.) in diameter
fast as hobbing. The time required for changing are invariably solid. As cutters increase in size,
cutting tools is about 30 min, but this needs to the practice of using high-speed tool steel cutting
be done only every 30 to 50 h. Although fast, the edges (blades) as inserts in alloy steel bodies is
process is applicable only for the large-quantity usually more economical. Inserts are normally
production of a limited variety of gears. held by mechanical fasteners.
Construction. Most reciprocating tools Rotary cutters (face mill, rotary broach, and
such as those used in gear shaping and planing other) for bevel gears are usually designed to
Chapter 5: Machining, Grinding, and Finishing / 107

use inserts. In rotary cutters, feed rate is a direct G-Trac generator cutter life is 6 to 175 times
function of the number of cutting faces. There- as long as that for hobs. Special machines are
fore, it is desirable to have as many cutting available for sharpening the cutters, and the cut-
edges as possible to increase the amount of ters can also be sharpened all at one time in a
metal removed during each revolution. How- special xture on a standard surface grinder. A
ever, tooth strength and chip clearance must be G-Trac machine has a 19 kW (25 hp) motor. It
considered when selecting the number of blades is six to ten times faster than hobbing, but the
(cutting edges). Despite the limitations imposed machine is expensive.
by chip clearance and tooth strength, gear cut-
ters are often redesigned for greater efciency.
The G-Trac gear generator is a special Speed and Feed
machine that uses large numbers of single-point
high-speed tool steel cutting tools carried on an In addition to the type of gear being machined
endless chain of tool blocks. One model for and the method of cutting, other factors that
moderate quantities of gears has a single row of inuence the choice of cutting speed are work
cutting tools (Fig. 22). As these are sped around metal composition and hardness, diametral pitch
the track, the work is fed into the cutters and of the work gear, rigidity of the setup, tolerance
rolled as though in mesh with a single tooth sim- and nish requirements, and cutting uid used.
ulated by the cutters. Thus, one tooth space is Hobbing Speed. Nominal speeds and feeds
generated. The work is then withdrawn, indexed for gear hobbing that take into account several
to the next tooth space, fed in again, and so on, of the above variables are listed in Table 2. As
around the gear. carbon content, alloy content, and hardness of
Another G-Trac model for large quantities the steel workpiece increase, recommended cut-
has a number of rows of teeth (typically 14) ter speed decreases; for any given work metal,
around the chain of blocks. The work is fed into
the cutters and revolved continuously as though
in mesh with a simulated rack, until all tooth
spaces are nished. On both models, spur and
helical gears (up to 45 helix angle) as large as
356 mm (14 in.) in diameter can be cut and
crowned and can be stacked to a height of 190
mm (7 in.) times the cosine of the helix angle.

Fig. 21 Schematic of operation of the Tangear gear cutter Fig. 22 Method of operation of a G-Trac gear generator
Table 2 Feeds and speeds for the hobbing of carbon and low-alloy steel gears with high-speed
steel tools
For hobbing class 9 (per AGMA 390.03) or better gears, it may be necessary to reduce speeds and feeds by 50% and/or take two
cuts. To meet nish requirements, it may be necessary to try both conventional and climb cutting.
Gear tooth size Feed per revolution High-speed steel
of workpiece(a) Hob speed tool material
Hardness, Diametral Number
Material HB Condition Module pitch of cuts mm in. m/min sfm ISO AISI

Wrought free-machining carbon steels

Low-carbon 100150 Hot rolled or 2513 12 2 1.50 0.060 67 220 S4, S2 M2, M7
resulfurized annealed 122.5 310 1 1.50 0.060 70 230
1116 21.5 1119 1 1.50 0.060 73 240
1117 10.5 2048 1 1.25 0.050 76 250
1118 0.5 and ner 48 and ner 1 0.75 0.030 85 280
1119 150200 Cold drawn 2513 12 2 1.50 0.060 70 230 S4, S2 M2, M7
1211 122.5 310 1 1.50 0.060 73 240
1212 21.5 1119 1 1.50 0.060 81 265
10.5 2048 1 1.25 0.050 84 275
0.5 and ner 48 and ner 1 0.75 0.030 90 300
Medium-carbon 175225 Hot rolled, 2513 12 2 1.50 0.060 52 170 S4, S2 M2, M7
resulfurized normalized, 122.5 310 1 1.50 0.060 55 180
1132 1144 annealed, or 21.5 1119 1 1.50 0.060 58 190
1137 1145 cold drawn 10.5 2048 1 1.25 0.050 60 200
1139 1146 0.5 and ner 48 and ner 1 0.75 0.030 67 220
1140 1151 325375 Quenched and 2513 12 2 1.50 0.060 21 70 S5, S11 M3, M42
1141 tempered 122.5 310 1 1.50 0.060 27 90
21.5 1119 1 1.50 0.060 34 110
10.5 2048 1 1.25 0.050 37 120
0.5 and ner 48 and ner 1 0.75 0.030 40 130
Low-carbon 100150 Hot rolled, 2513 12 2 1.65 0.065 78 255 S4, S2 M2, M7
leaded normalized, 122.5 310 1 1.65 0.065 81 265
12L13 annealed, or 21.5 1119 1 1.50 0.060 90 295
12L14 cold drawn 10.5 2048 1 1.25 0.050 95 305
12L15 0.5 and ner 48 and ner 1 0.75 0.030 105 340
200250 Hot rolled, 2513 12 2 1.50 0.060 50 165 S4, S2 M2, M7
normalized, 122.5 310 1 1.50 0.060 58 190
annealed, or 21.5 1119 1 1.50 0.060 69 225
cold drawn 10.5 2048 1 1.25 0.050 73 240
0.5 and ner 48 and ner 1 0.75 0.030 84 275
Wrought carbon steels
Low carbon 85125 Hot rolled, 2513 12 2 1.80 0.070 62 205 S4, S2 M2, M7
1005 1016 1026 normalized, 122.5 310 1 1.80 0.070 75 245
1006 1017 1029 annealed, or 21.5 1119 1 1.50 0.060 81 265
1008 1018 1513 cold drawn 10.5 2048 1 1.25 0.050 85 280
1009 1019 1518 0.5 and ner 48 and ner 1 0.75 0.030 88 290
1010 1020 1522 225275 Annealed or 2513 12 2 1.50 0.060 37 120 S4, S2 M2, M7
1011 1021 cold drawn 122.5 310 1 1.50 0.060 49 160
1012 1022 21.5 1119 1 1.50 0.060 56 185
1013 1023 10.5 2048 1 1.25 0.050 59 195
1015 1025 0.5 and ner 48 and ner 1 0.75 0.030 60 200
Medium 125175 Hot rolled, 2513 12 2 1.80 0.070 50 165 S4, S2 M2, M7
carbon normalized, 122.5 310 1 1.80 0.070 53 175
1030 1044 1526 annealed, or 21.5 1119 1 1.50 0.060 56 185
1033 1045 1527 cold drawn 10.5 2048 1 1.25 0.050 60 200
1035 1046 1536 0.5 and ner 48 and ner 1 0.75 0.030 64 210
1037 1049 1541 325375 Quenched and 2513 12 2 1.15 0.045 21 70 S5, S11 M3, M42
1038 1050 1547 tempered 122.5 310 1 1.15 0.045 24 80
1039 1053 1548 21.5 1119 1 0.65 0.025 30 100
1040 1055 1551 10.5 2048 1 0.65 0.025 32 105
1042 1524 1552 0.5 and ner 48 and ner 1 0.65 0.025 37 120
1043 1525
Wrought free-machining alloy steels
Medium-carbon 150200 Hot rolled, 2513 12 2 1.50 0.060 44 145 S4, S2 M2, M7
resulfurized normalized, 122.5 310 1 1.50 0.060 47 155
4140 annealed, or 21.5 1119 1 1.50 0.060 50 165
4140Se cold drawn 10.5 2048 1 1.25 0.050 55 180
4142Te 0.5 and ner 48 and ner 1 0.75 0.030 60 200
4145Se 325375 Quenched and 2513 12 2 1.15 0.045 20 65 S5, S11 M3, M42
4147Te tempered 122.5 310 1 1.15 0.045 23 75
4150 21.5 1119 1 0.65 0.025 29 95
10.5 2048 1 0.65 0.025 30 100
0.5 and ner 48 and ner 1 0.65 0.025 41 135
(a) Feeds are based on the largest standard recommended hob diameter. When using a smaller hob diameter, the feed must be reduced proportionally. Source: Metcut
Research Associates Inc.
Chapter 5: Machining, Grinding, and Finishing / 109

Table 2 (continued)
Gear tooth size Feed per revolution High-speed steel
of workpiece(a) Hob speed tool material
Hardness, Diametral Number
Material HB Condition Module pitch of cuts mm in. m/min sfm ISO AISI

Medium- and high- 150200 Hot rolled, 2513 12 2 1.50 0.060 52 170 S4, S2 M2, M7
carbon leaded normalized, 122.5 310 1 1.50 0.060 55 180
41L30 43L40 annealed, or 21.5 1119 1 1.50 0.060 58 190
41L40 51L32 cold drawn 10.5 2048 1 1.25 0.050 60 200
41L45 52L100 0.5 and ner 48 and ner 1 0.75 0.030 67 220
41L47 86L20 325375 Quenched and 2513 12 2 1.15 0.045 21 70 S5, S11 M3, M42
41L50 86L4 tempered 122.5 310 1 1.15 0.045 27 90
21.5 1119 1 1.15 0.045 34 110
10.5 2048 1 1.15 0.045 37 120
0.5 and ner 48 and ner 1 0.75 0.030 40 130
Wrought alloy steels
Low carbon 125175 Hot rolled, 2513 12 2 1.50 0.060 49 160 S4, S2 M2, M7
4012 4621 6118 annealed, or 122.5 310 1 1.50 0.060 52 170
4023 4718 8115 cold drawn 21.5 1119 1 1.50 0.060 55 180
4024 4720 8617 10.5 2048 1 1.25 0.050 58 190
4118 4815 8620 0.5 and ner 48 and ner 1 0.75 0.030 64 210
4320 4817 8622 275325 Normalized or 2513 12 2 1.15 0.045 27 90 S4, S2 M2, M7
4419 4820 8822 quenched and 122.5 310 1 1.15 0.045 30 100
4422 5015 9310 tempered 21.5 1119 1 1.15 0.045 34 110
4615 5115 94B15 10.5 2048 1 1.15 0.045 37 120
4617 5120 94B17 0.5 and ner 48 and ner 1 0.75 0.030 40 130
Medium carbon 175225 Hot rolled, 2513 12 2 1.50 0.060 38 125 S4, S2 M2, M7
1330 4427 81B45 annealed, or 122.5 310 1 1.50 0.060 41 135
1335 4626 8625 cold drawn 21.5 1119 1 1.25 0.060 44 145
1340 50B40 8627 10.5 2048 1 1.25 0.050 49 160
1345 50B44 8630 0.5 and ner 48 and ner 1 0.75 0.030 55 180
4027 5046 8637 325375 Normalized or 2513 12 2 1.15 0.045 18 60 S5, S11 M3, M42
4028 50B46 8640 quenched and 122.5 310 1 1.15 0.045 20 65
4032 50B50 8642 tempered 21.5 1119 1 0.65 0.025 23 75
4037 5060 8645 10.5 2048 1 0.65 0.025 24 80
4042 50B60 86B45 0.5 and ner 48 and ner 1 0.65 0.025 29 95
4047 5130 8650
4130 5132 8655
4135 5135 8660
4137 5140 8740
4140 5145 8742
4142 5147 9254
4145 5150 9255
4147 5155 9260
4150 5160 94B30
4161 51B60
4340 6150
(a) Feeds are based on the largest standard recommended hob diameter. When using a smaller hob diameter, the feed must be reduced proportionally. Source: Metcut
Research Associates Inc.

cutting speed increases as diametral pitch and spiral bevel gears with two common types
becomes ner. For the numerical values given of cutters are given in Table 4. As in shaping,
in Table 2. it is assumed that an additive-type speed is not varied with diametral pitch when
cutting uid will be used. Low cutting speeds other conditions are the same. For both the
prolong tool life and eliminate the difculties interlocking cutters and the face mill cutters,
involved in changing tools and matching cuts. speeds are increased for nishing when rough-
Shaping speed is inuenced by variables ing and nishing are done separately. The
similar to those that affect hobbing speeds, with speeds used for cutting straight and spiral bevel
the exception of diametral pitch. Nominal speeds gears with rotating cutters are close to those
and feeds for gear shaping under a variety of con- suggested in Table 4. When straight bevel gears
ditions are given in Table 3. Cutter speed varies are being cut with reciprocating cutters, much
with the composition and hardness of the work lower speeds are required.
metal but not with diametral pitch. The speeds Feeds. Nominal feeds for the hobbing of
listed in Table 3 are generally conservative. steel gears are included in Table 2. Values are
Speeds for Interlocking and Face Mill given in millimeters per revolution (inches per
Cutters. Nominal speeds for cutting straight revolution) of the gear being hobbed and are
110 / Gear Materials, Properties, and Manufacture

Table 3 Feeds and speeds for the shaping of carbon and low-alloy steel gears with high-speed
steel tools
Rotary feed per cutter
stroke (102 mm, Cutter
Gear tooth size or 4 in., pitch speed High-speed steel
diameter cutter)(a) tool material
Hardness, Number Diametral m/
Material HB Condition of cuts(a) Module pitch mm in. min sfm ISO AISI

Wrought free-machining carbon steels

Low-carbon resulfurized 100150 Hot rolled 4 256 14 0.55 0.022 26 85 S4, S2 M2, M7
1116 1119 or annealed 2 5.52.5 510 0.40 0.016
1117 1211 2 21.5 1119 0.28 0.011
1118 1212 2 10.5 2048 0.20 0.008
150200 Cold drawn 4 256 14 0.55 0.022 27 90 S4, S2 M2, M7
2 5.52.5 510 0.40 0.016
2 21.5 1119 0.28 0.011
2 10.5 2048 0.20 0.008
Medium-carbon 175225 Hot rolled, 4 256 14 0.55 0.022 24 80 S4, S2 M2, M7
resulfurized normalized, 2 5.52.5 510 0.40 0.016
1132 1144 annealed, or 2 21.5 1119 0.28 0.011
1137 1145 cold drawn 2 10.5 2048 0.20 0.008
1139 1146 325375 Quenched and 4 256 14 0.30 0.012 12 40 S5, S11 M3, M42
1140 1151 tempered 2 5.52.5 510 0.25 0.010
1141 2 21.5 1119 0.20 0.008
2 10.5 2048 0.20 0.008
Low-carbon leaded 100150 Hot rolled, 4 256 14 0.55 0.022 38 125 S4, S2 M2, M7
12L13 normalized, 2 5.52.5 510 0.40 0.016
12L14 annealed, or 2 21.5 1119 0.28 0.011
12L15 cold drawn 2 10.5 2048 0.20 0.008
200250 Hot rolled, 4 256 14 0.55 0.022 27 90 S4, S2 M2, M7
normalized, 2 5.52.5 510 0.40 0.016
annealed, or 2 21.5 1119 0.28 0.011
cold drawn 2 10.5 2048 0.20 0.008
Wrought carbon steels
Low carbon 85125 Hot rolled, 4 256 14 0.45 0.018 27 90 S4, S2 M2, M7
1005 1015 1023 normalized, 2 5.52.5 510 0.30 0.012
1006 1016 1025 annealed, or 2 21.5 1119 0.25 0.010
1008 1017 1026 cold drawn 2 10.5 2048 0.20 0.008
1009 1018 1029 225275 Annealed or 4 256 14 0.40 0.016 18 60 S4, S2 M2, M7
1010 1019 1513 cold drawn 2 5.52.5 510 0.30 0.012
1011 1020 1518 2 21.5 1119 0.25 0.010
1012 1021 1522 2 10.5 2048 0.20 0.008
1013 1022
Medium carbon 125175 Hot rolled, 4 256 14 0.45 0.018 23 75 S4, S2 M2, M7
1030 1044 1526 normalized, 2 5.52.5 510 0.30 0.012
1033 1045 1527 annealed, or 2 2 1.5 1119 0.25 0.010
1035 1046 1536 cold drawn 2 10.5 2048 0.20 0.008
1037 1049 1541 325375 Quenched and 4 256 14 0.30 0.012 12 40 S5, S11 M3, M42
1038 1050 1547 tempered 2 5.52.5 510 0.25 0.010
1039 1053 1548 2 21.5 1119 0.20 0.008
1040 1055 1551 2 10.5 2048 0.20 0.008
1042 1524 1552
1043 1525
Wrought free-machining alloy steels
Medium-carbon 150200 Hot rolled, 4 256 14 0.55 0.022 24 80 S4, S2 M2, M7
resulfurized normalized, 2 5.52.5 510 0.45 0.018
4140 4145Se annealed, or 2 21.5 1119 0.28 0.011
4140Se 4147Te cold drawn 2 10.5 2048 0.20 0.008
4142Te 4150 325375 Quenched and 4 256 14 0.30 0.012 11 35 S5, S11 M3, M42
tempered 2 5.52.5 510 0.25 0.010
2 21.5 1119 0.20 0.008
2 10.5 2048 0.20 0.008
Medium- and high-carbon 150200 Hot rolled, 4 256 14 0.55 0.022 24 80 S4, S2 M2, M7
leaded normalized, 2 5.52.5 510 0.40 0.016
41L30 41L50 52L100 annealed, or 2 21.5 1119 0.28 0.011
41L40 43L40 86L20 cold drawn 2 10.5 2048 0.20 0.008
41L45 51L32 86L40 325375 Quenched and 4 256 14 0.30 0.012 11 35 S5, S11 M3, M42
41L47 tempered 2 5.52.5 510 0.25 0.010
2 21.5 1119 0.20 0.008
2 10.5 2048 0.20 0.008
(a) For cutting gears of class 9 (AGMA 390.03) or better, the rotary feed should be reduced, and the number of cuts should be increased. Source: Metcut Research Asso-
ciates Inc.
Chapter 5: Machining, Grinding, and Finishing / 111

Table 3 (continued)
Rotary feed per cut-
ter stroke (102 mm, Cutter
Gear tooth size or 4 in., pitch speed High-speed steel
diameter cutter)(a) tool material
Hardness, Number Diametral m/
Material HB Condition of cuts(a) Module pitch mm in. min sfm ISO AISI

Wrought alloy steels

Low carbon 125175 Hot rolled, 4 256 14 0.45 0.018 21 70 S4, S2 M2, M7
4012 4621 6118 annealed, or 2 5.52.5 510 0.30 0.012
4023 4718 8115 cold drawn 2 21.5 1119 0.25 0.010
4024 4720 8617 2 10.5 2048 0.20 0.008
4118 4815 8620 325375 Normalized 4 256 14 0.30 0.012 12 40 S5, S11 M3, M42
4320 4817 8622 or quenched 2 5.52.5 510 0.25 0.010
4419 4820 8822 and tempered 2 21.5 1119 0.20 0.008
4422 5015 9310 2 10.5 2048 0.20 0.008
4615 5115 94B15
4617 5120 94B17
Medium carbon 175225 Hot rolled, 4 256 14 0.45 0.018 18 60 S4, S2 M2, M7
1330 4427 81B45 annealed, or 2 5.52.5 510 0.30 0.012
1335 4626 8625 cold drawn 2 21.5 1119 0.25 0.010
1340 50B40 8627 2 10.5 2048 0.20 0.008
1345 50B44 8630 325375 Normalized 4 256 14 0.30 0.012 11 35 S5, S11 M3, M42
4027 5046 8637 or quenched 2 5.52.5 510 0.25 0.010
4028 50B46 8640 and tempered 2 21.5 1119 0.20 0.008
4032 50B50 8642 2 10.5 2048 0.20 0.008
4037 5060 8645
4042 50B60 86B45
4047 5130 8650
4130 5132 8655
4135 5135 8660
4137 5140 8740
4140 5145 8742
4142 5147 9254
4145 5150 9255
4147 5155 9620
4150 5160 94B30
4161 51B60
4340 6150
(a) For cutting gears of class 9 (AGMA 390.03) or better, the rotary feed should be reduced, and the number of cuts should be increased. Source: Metcut Research Asso-
ciates Inc.

based on the assumption that the gears will be diametral pitch; work metal composition and
ground after hobbing. Hobbing feeds should be hardness have a minor inuence. The same rates
decreased when gears are to be nished by shav- of feed are commonly used for both roughing and
ing. As indicated in Table 2, hobbing feeds are nishing, although the cutting edges of the teeth
sensitive to differences in diametral pitch, but are altered for nishing, as indicated in Table 4.
feeds are not ordinarily changed for differences
in the composition or hardness of the gear steel
when diametral pitch is the same, except for the
Cutting Fluids
coarser-pitch heat-treated gears.
Nominal feeds for cutting gears by shaping
Cutting uids are recommended for all gear
are given in Table 3. These values are based on
cutting, although for some applications (notably
the assumption that the gears will be nished by
the cutting of large gears) the use of a cutting
grinding; a reduction in speed is recommended
uid is impractical. The three cutting uids
when gears are to be nished by shaving. Feeds
most commonly used for gear cutting are:
for shaping vary with diametral pitch and vary
comparatively little with work metal composi- Mineral oil (without additives) having a vis-
tion and hardness. Conditions for a specic cosity of 140 to 220 SUS at 40 C (100 F)
operation may require feeds higher or lower Cutting oil, sulfurized or chlorinated (or
than those given in Table 3. both), usually diluted with mineral oil to a
Feeds for cutting straight and spiral bevel viscosity of 180 to 220 SUS at 40 C (100 F)
gears by rotary cutters are suggested in Table 4. Motor oil (SAE 20 grade), either detergent
The optimum feed is greatly inuenced by or nondetergent type
112 / Gear Materials, Properties, and Manufacture

Table 4 Gear cutting speeds for the rough and nish cutting of carbon and low-alloy steel straight
and spiral bevel gears with high-speed steel tools
Cutting speed(a)
High-speed steel
Roughing Finishing tool material
Material Hardness, HB Condition m/min sfm m/min sfm ISO AISI

Wrought free-machining carbon steels

Low-carbon resulfurized 100150 Hot rolled or annealed 38 125 76 250 S4, S2 M2, M7
1116 1118 1211 150200 Cold drawn 43 140 84 275 S4, S2 M2, M7
1117 1119 1212
Medium-carbon resulfurized 175225 Hot rolled, normalized, 26 85 53 175 S4, S2 M2, M7
1132 1140 1145 annealed, or cold drawn
1137 1141 1146 325375 Quenched and tempered 15 50 30 100 S5 M3
1139 1144 1151
Low-carbon leaded 100150 Hot rolled, normalized, 44 145 88 290 S4, S2 M2, M7
12L13 12L15 annealed, or cold drawn
12L14 200250 Hot rolled, normalized,
annealed, or cold drawn 34 110 69 225 S4, S2 M2, M7
Wrought carbon steels
Low carbon 85125 Hot rolled, normalized, 27 90 58 190 S4, S2 M2, M7
1005 1015 1023 annealed, or cold drawn
1006 1016 1025 225275 Annealed or cold drawn 21 70 44 145 S4, S2 M2, M7
1008 1017 1026
1009 1018 1029
1010 1019 1513
1011 1020 1518
1012 1021 1522
1013 1022
Medium carbon 125175 Hot rolled, normalized, 26 85 53 175 S4, S2 M2, M7
1030 1044 1526 annealed, or cold drawn
1033 1045 1527 325375 Quenched and tempered 15 50 30 100 S5 M3
1035 1046 1536
1037 1049 1541
1038 1050 1547
1039 1053 1548
1040 1055 1551
1042 1524 1552
1043 1525
Wrought free-machining alloy steels
Medium-carbon resulfurized 150200 Hot rolled, normalized, 26 85 53 175 S4, S2 M2, M7
4140 4142Te 4147Te annealed, or cold drawn
4140Se 4145Se 4150 325375 Quenched and tempered 14 45 29 95 S5 M3
Medium- and high-carbon leaded 150200 Hot rolled, normalized, 26 85 53 175 S4, S2 M2, M7
41L30 41L50 52L100 annealed, or cold drawn
41L40 43L40 86L20
41L45 51L32 86L40 325375 Quenched and tempered 14 45 27 90 S5 M3
Wrought alloy steels
Low carbon 125175 Hot rolled, annealed, 26 85 53 175 S4, S2 M2, M7
4012 4621 6118 or cold drawn
4023 4718 8115 325375 Normalized or 15 50 30 100 S5 M3
4024 4720 8617 quenched and tempered
4118 4815 8620
4320 4817 8622
4419 4820 8822
4422 5015 9310
4615 5115 94B15
4617 5120 94B17
Medium carbon 175225 Hot rolled, annealed, 20 65 41 135 S4, S2 M2, M7
1330 4047 4161 or cold drawn
1335 4130 4340 325375 Normalized or 15 50 30 100 S5 M3
1340 4135 4427 quenched and tempered
1345 4137 4626
4027 4140 50B40
4028 4142 50B44
4032 4145 5046
4037 4147 50B46
4042 4150 50B50

(a) Cutting speed recommendations are for use with alternate-tooth milling cutters or gear generators. For planer-type generators, use the recommended cutting speeds
in Table 3. Source: Metcut Research Associates Inc.
Chapter 5: Machining, Grinding, and Finishing / 113

Table 4 (continued)
Cutting speed(a)
High-speed steel
Roughing Finishing tool material
Material Hardness, HB Condition m/min sfm m/min sfm ISO AISI

Medium carbon 175225 Hot rolled, annealed, 20 65 41 135 S4, S2 M2, M7

5060 51B60 86B45 or cold drawn
50B60 6150 8650 325375 Normalized or 15 50 30 100 S5 M3
5130 81B45 8655 quenched and tempered
5132 8625 8660
5135 8627 8740
5140 8630 8742
5145 8637 9254
5147 8640 9255
5150 8642 9260
5155 8645 94B30
(a) Cutting speed recommendations are for use with alternate-tooth milling cutters or gear generators. For planer-type generators, use the recommended cutting speeds
in Table 3. Source: Metcut Research Associates Inc.

Water-soluble oils, which are widely used in nal selection, and in all selections both these
many metal cutting operations, are used to a factors must be considered. When the cost of
lesser extent for gear cutting because the oils re- machining standard steels is compared with
ferred to above are more effective for producing the cost of free-machining counterparts, the use
the surface nishes desired on gear teeth and for of a free-machining grade will almost always
prolonged cutter life. Nevertheless, soluble oils result in lower machining costs. However, when
are sometimes used in gear cutting. One notable changing from one standard grade of a given
application for soluble oil (20 parts water to 1 steel to its free-machining counter-part, the ad-
part oil) is the nish broaching of ne-pitch ditional cost of the free-machining grade must
gears. be considered in the cost comparison. When
The relative advantages and disadvantages of changes in steel composition that result in lower
the three oils mentioned above are matters of machining costs require changes in the method
opinion among different plants producing steel of heat treating, the cost of heat treating must
gears. Most data in speed and feed tables are also be considered.
based on the use of cutting oils with additives A change in microstructure, independent of
(the second of the cutting uids listed above). steel composition, also has a signicant effect on
However, in some plants, mineral oils (including tool life or surface nish or both. Altering the
motor oils) are used when records have proved heat treatment can produce a more machinable
that the use of these oils resulted in longer cutter microstructure that results in increased cutter
life than when prepared cutting oils were used. life.
In some plants, additive cutting oils are used
for roughing operations and mineral oils for n-
ishing when the operations are done separately Computer Numerical Control (CNC)
in different machines. An adequate supply (a Milling and Hobbing Machines
ood) of uid under slight pressure (about 35
kPa, or 5 psi) at the cutting area is extremely In practice, gear milling is usually conned to
importantusually more important than the replacement gears or to the small-lot produc-
composition of the oil. tion, roughing, and nishing of coarse-pitch
gears. The milling of low-production, special
tooth forms is also done. Milling is often used as
Comparison of Steels for Gear Cutting the rst operation prior to nish hobbing or
shaping, especially in large diametral pitch
End use is the main factor in the selection of spur, helical, and straight bevel gears. The gear
steel for a specic gear. However, two or more is then nished to nal size by another method
steels will often serve equally well. Under these selected by the manufacturer.
conditions, the cost of the steel and the cost of Special high-power (56 to 75 kW, or 75 to
processing it are signicant factors in making a 100 hp) CNC machines, called gashers (Fig.
114 / Gear Materials, Properties, and Manufacture

23), are used for the roughing of gears from 6 The power of these machines and their ability
to diametral pitch. These machines employ for two-axis interpolation make coarse-tooth
strong, rigid machine components and are straight bevel gears a natural application. The
extremely powerful. They operate at peak ef- time savings can be signicant; two-tool genera-
ciency with large inserted carbide cutters. The tors can be used to nish the gashed bevel blank.
ability to remove large amounts of metal at sub- Computer numerical control gear hobbing
stantial feed rates makes this process quite eco- machines are available in 305, 508, 914, 1220,
nomical for coarse pitches. The gears can be n- and 1625 mm (12, 20, 36, 48, and 64 in.) capac-
ished on the same machines with high-speed ities. Figure 24 shows a CNC machine hobbing
tool steel formed cutters. The accuracy of index a gear.
on these machines is more than adequate to pro-
duce tooth spacing requirements.
The advent of the CNC-controlled index Grinding of Gears
mechanism of the gasher allows the manufac-
ture of large, coarse-pitch sprockets for roller Gear teeth can be produced entirely by grind-
chains of all sizes. The face capacity of these ing, entirely by cutting, or by rst cutting and
machines permits the stacking of many plate then grinding to the required dimensions. Usu-
sprockets at one time with good quality and reli- ally, gear grinding removes only several hun-
able production. dredths of a millimeter of metal from precut

Fig. 23 Four types of gashers. (a) Duo-axis. (b) Helical machine. (c) Rack machine. (d) Slitting head
Chapter 5: Machining, Grinding, and Finishing / 115

gears to make accurate teeth for critical applica- inspection of the nished teeth detects ne
tions. Teeth made entirely by grinding are usu- cracks, and macroetching with dilute nitic acid
ally only those of ne pitch, for which the total detects grinding burn. Both inspection proce-
amount of metal removed is small. The grinding dures are advised in gear grinding.
of ne-pitch gear teeth from uncut blanks may A carbide network in carburized steels, or the
be less costly than the two-step procedure if presence of sufcient retained austenite or of
there is not much metal to be removed. hydrogen in any hardened steel, can increase the
The two basic methods for the grinding of susceptibility of the steel to grinding cracks.
gear teeth are form grinding (nongenerating) Therefore, it is necessary to keep these condi-
and generation grinding. Many varieties of tions under control. Reducing the heat gener-
machines have been built specically to grind ated in grinding (by decreasing the rate of stock
gears and pinions. removal or by the use of a grinding uid) can
Advantages and Disadvantages. A gear is help to eliminate cracking and burning of the
ground when it is so hard that it cannot be n- ground surface.
ished by other methods or when the required Stock Allowance. When workpieces are
accuracy is greater than can be obtained by prepared for grinding by cutting, the least
other methods. If the hardness of the gear does amount of stock possible should be left to be
not exceed 40 HRC and if its size is within the removed by grinding. This stock should be
capacity of the machine, shaving will produce evenly distributed to reduce the wheel dressings
teeth of an accuracy close to that produced by and operating adjustments needed to maintain
grinding. Gears made of steel harder than 40 wheel contour when using conventional (alu-
HRC are nished by grinding when dimen- minum oxide or silicon carbide) dressable
sional accuracy is required or when correction wheels. With cubic boron nitride (CBN) wheels,
beyond that feasible by honing is required. where stock is removed in one pass, evenly dis-
The major disadvantage of grinding gears is tributed stock results in more uniform grinding
the cost. It is usually more expensive to grind pressure.
gears than to cut them because material is In general, it is not practical to remove more
removed in small increments in grinding. than 0.15 mm (0.006 in.) of stock from each
Ground gears are usually subjected to more face of the teeth of spur gears up to 178 mm (7
inspection than cut gears. Magnetic particle in.) in diameter or 0.20 mm (0.008 in.) from
gears between 178 and 254 mm (7 and 10 in.) in
diameter. These amounts of stock can usually be
removed in three to ve passes with conven-
tional dressable wheels or in one pass with CBN
wheels. However, the number of passes de-
pends on number of teeth, pressure angle, and
required nish. Sometimes nine or ten passes
are needed with conventional wheels.
The recommended stock allowance for grind-
ing straight bevel teeth is 0.08 mm (0.003 in.) at
the root of the tooth and 0.08 to 0.13 mm (0.003
to 0.005 in.) on the face of each tooth. Grinding
time depends on the amount of stock to be re-
moved. For example, removing 0.076 to 0.089
mm (0.0030 to 0.0035 in.) of stock from a gear
of 4 diametral pitch and 32 teeth takes 30 s per
tooth; a gear of 10 diametral pitch and 32 teeth
takes about 20 s per tooth. Figure 25, which
shows the cross sections of gear teeth of various
diametral pitches, can be used to gage the rela-
tive sizes of the 4 and 10 diametral pitch gears.
Grinding stock allowance per tooth side for
spiral bevel and hypoid gears should be 0.10 to
0.13 mm (0.004 to 0.005 in.) for gears coarser
Fig. 24 CNC gear hobbing machine producing a gear than 10 diametral pitch and 0.08 to 0.10 mm
116 / Gear Materials, Properties, and Manufacture

(0.003 to 0.004 in.) for gears of 10 to 20 diame- through-hardened and case-hardened steel gears.
tral pitch. Stock allowance at the root of the For nitrided gears, silicon carbide wheels are
tooth should be about 0.08 mm (0.003 in.) for sometimes used. Grit size, which depends
gears up to 20 diametral pitch. mainly on the type of grinding and the pitch of the
Multiple-Piece Loads. Because gear grind- gear, ranges from 46 to 100, although somewhat
ing is an expensive operation, it is important to ner grits can be used for form grinding. A grit
control the amount of metal to be removed by size of 60 is generally recommended for diame-
grinding, to reduce runout and spacing errors, tral pitch in the range of 1 to 10; 80 for the range
and to try to grind multiple-piece loads. Multi- of 8 to 12; and 100 for gears of pitch ner than 14.
ple-start hobs can cause enough spacing error to With crush-formed wheels, grit sizes as small as
make extra grinding passes necessary. There- 400 are used in grinding gears of diametral pitch
fore, when a hobbed gear is to be ground, it is as ne as 200. For grade (hardness), a range of H
better to use a single-thread hob. The multiple- to M is generally recommended.
start hob will not save enough to pay for the Vitried bond wheels are used because of
extra grinding passes. Single-thread hobs are their rigidity and ability to be diamond dressed
available with nonground class-C tooth forms. with a minimum of diamond wear. Wheel rigid-
Wheel Specications. Aluminum oxide ity is particularly important for dish-shaped
wheels are preferred for the grinding of teeth in wheels of the large sizes used in generation
grinding. Resinoid bond cup wheels are satis-
factory for generating bevel gear teeth.
Vitried bond wheels are used for forming.
They are generally crush trued. The vitried
bond is friable and is rigid enough to support the
force of crush truing without deection.
Steel-core wheels plated with CBN are com-
monly used because their open grain texture
allows for slow traverse across the tooth surface
and rapid stock removal without burning. Sur-
faces are generally nished in one pass. Because
of the hardness of the material, several thousand
parts can be ground before wheel contours need
to be replated. More detailed information on
grinding wheel specications and factors in
selection can be found in Ref 1 and 2.
Speed. Most gears ground with conven-
tional wheels are ground with wheel surface
speeds of 1700 to 2000 m/min (5500 to 6500
sfm). When very small amounts of stock are
being removed, speeds higher than 2000 m/min
(6500 sfm) can be used. When grinding gears
with cup-shaped or dish-shaped wheels, slower
speeds (140 m/min, or 450 sfm, or less) are
common. Wheel speeds up to 3000 m/min
(10,000 sfm) are used with CBN wheels.
Infeed. When using conventional wheels,
pitch of the gear being ground is the main factor
inuencing optimum infeed. For example, in
the plunge grinding of case-hardened or hard-
ened-and-tempered steels (50 to 62 HRC), an
infeed of 0.038 mm/pass (0.0015 in./pass) is
recommended for gears having diametral
pitches of 1 to 4. Infeed is reduced to 0.030
mm/pass (0.0012 in./pass) for pitches of 5 to 8
and to 0.013 mm/pass (0.005 in./pass) for
Fig. 25 Gear teeth of various diametral pitch pitches ner than 8. With CBN wheels, 0.13 to
Chapter 5: Machining, Grinding, and Finishing / 117

0.25 mm (0.005 to 0.010 in.) of stock is Form Grinding

removed in one pass.
Usually, form grinding consists of passing a
Correcting for formed wheel through a tooth space to grind to
Distortion in Heat Treating root depth the left side of one tooth and the right
Most steel gears are heat treated. Because side of the next tooth at the same time (see mod-
some diametral change occurs during heat treat- ications in Fig. 27). When straight teeth are
ment, gear teeth are cut prior to heat treating and ground, as in spur gears, the workpiece is held
ground to nish dimensions after heat treating. in a xed radial position during a pass. For hel-
The magnitude of the dimensional change ical gears, however, the workpiece must be
depends on the type of heat treatment and the
shape of the part. Distortion is greatest when the
entire part is heated and quenched. Induction
hardening and ame hardening cause small
changes because only the tooth area of the part
is heated and quenched.
Usually, nitrided gears do not need to be
ground to correct for distortion. When gears are
properly nitrided, dimensional change is so
small that any correction required can be
accomplished by honing or lapping.
When the entire gear is heat treated, dimen-
sional change is less in gears of nearly uniform
cross section than in those of intricate design or
large variation in cross section. For example,
short, stubby pinions are least likely to lose accu-
racy because of unequal cooling rates during
quenching, while thin-web gears with heavy-
section teeth or heavy rims change the most.
The use of a heat-treating procedure that
allows maximum dimensional control, such as
placing the heated part in a xture for quench-
ing, helps reduce the amount of grinding and
sometimes has eliminated the need for grinding.
However, some dimensional changes occur dur-
ing heat treatment regardless of special tech-
niques. The following example is typical of a
gear that required grinding even though it had
been quenched in a die to control distortion.
Example 1: Grinding Die-Quenched
Gears. The gear teeth for the part illustrated in
Fig. 26 were cut by hobbing, after which they
were carburized and quenched in a die to mini-
mize distortion. However, this practice did not
prevent distortion beyond tolerance, and light-
ening holes caused a scalloping of the pitch line
by approximately 0.10 mm (0.004 in.) total
indicator reading. Therefore, grinding was nec-
essary. No grinding burns or cracks could be
Conditions of the grinding operation are
given in the table with Fig. 26. Damper plates
prevented excessive vibration of the gear during
grinding, thus reducing the likelihood of both Fig. 26 Setup for grinding a thin-web gear. Damper plates
prevent excessive vibration. Dimensions in gure
burns and cracks. given in inches
118 / Gear Materials, Properties, and Manufacture

rotated so that the helix will be followed as the Generation Grinding

wheel passes through the tooth space. A lead bar
There are four methods of generating gear
having the same lead as the helix controls rota-
teeth by grinding. Each method is identied by
tion during grinding.
the type of wheel used: straight, cup-shaped,
When the wheel pass through a tooth space is
dish-shaped, and rack-tooth worm wheels.
completed, the workpiece is indexed to the next
Straight Wheels. In this method, a straight
position, and the procedure is repeated. The
wheel beveled on both sides reciprocates across
indexing mechanism usually consists of an
the periphery of the workpiece as the workpiece
index plate with the same number of spaces at
rolls under it in a direction perpendicular to its
the workpiece. Control of tooth spacing errors
reciprocating motion. The reciprocating action
depends primarily on the accuracy of this plate.
of the wheel is similar to that of a reciprocating
Spur gears are ground with a single straight
gear cutter. This grinding method can be used to
wheel with a periphery that has been formed to
generate the teeth of spur gears.
produce the space between teeth (Fig. 27a), a sin-
Cup-Shaped Wheels. In this method, the
gle wheel with several ribs that grind more than
gear tooth is rolled against a cup-shaped wheel,
one tooth space during each wheel pass (Fig.
the sides of which are beveled to an angle equal
27b), or two single-ribbed wheels (Fig. 27c).
to the pressure angle required on the teeth. Dur-
Helical gears are ground with a formed, single
ing grinding, the wheel moves in a straight line
straight wheel. Bevel gears are form ground with
along the length of the tooth. The sides of the
a cup-shaped wheel. Form grinding can also be
wheel simultaneously grind adjacent sides of
used to nish grind precut threads of worms.
two teeth. A number of mechanical methods are
Wheels for form grinding are usually less than
used to control the generating motion and the
152 mm (6 in.) in diameter; wheel thickness
indexing of the workpiece. Electronic controls
depends on the size of the teeth to be ground.
are used on newer machines.
Wheels of larger diameter can be used, but the
Teeth can be generated by grinding with a
smaller wheels are more common because they
cup-shaped wheel on spur and helical gears up
need less clearance in nishing a gear that is close
to 457 mm (18 in.) in pitch diameter. However,
to a larger gear or to a shoulder. In addition,
when helical gears are being ground, the gener-
smaller wheels are better for grinding internal
ating motion must match the pitch in the plane
of rotation of the gear to be ground. With this
Standard gear grinding machines are avail-
grinding method, it is impractical to generate tip
able for form grinding spur gears ranging from
and root reliefs.
a minimum root diameter of 19 mm ( in.) to an
Dish-Shaped Wheels. This method uses
outside diameter of 914 mm (36 in.). Special
two dish-shaped grinding wheels, which may be
machines have been built to grind spur gears 2.4
inclined 15 or 20 or may be parallel in a verti-
m (8 ft) in diameter. Machines are available that
cal position (0), as shown in Table 5.
are capable of form grinding external helical
The generating motion is sometimes con-
gears up to 457 mm (18 in.) in diameter.
trolled by steel bands that are fastened to a pitch
block of the same diameter as the pitch circle of
the workpiece minus the thickness of the rolling
bands. The shape of the tip and root of the teeth
can be modied by using a cam to change the
normal generating motion.
Usually an index plate similar to that used in
form grinding indexes the workpiece if a pitch
block is used. Large machines use worm gear-
ing for indexing. Gears are ground dry, and
machines incorporate a device depending on a
feeler diamond to compensate for wheel wear.
Machines are available for grinding spur and
helical gears ranging from 25 to 3610 mm (1 to
142 in.) in diameter, with diametral pitch 2 to 17
Fig. 27 Relations of wheel and workpiece in the form grind- by this method.
ing of spur gears (top views). (a) Single-ribbed grind-
ing wheel. (b) Multiribbed grinding wheel. (c) Two single-ribbed
When the wheels are in the 15 or 20 position,
wheels, also known as a straddle wheel as in illustration (a) in Table 5 their active faces
Chapter 5: Machining, Grinding, and Finishing / 119

rotate in planes forming an acute angle. In this The advantages of grinding gears in the 0
position, the wheels simulate the tooth anks of position over the 15 or 20 position are:
a rack that the workpiece engages during the
rolling-generating motion (Fig. 28). The incli- Greater production without sacrice in
nation of the grinding wheels to the vertical in
the pitch plane generally equals the pressure Ability to grind more exacting ank proles
and to make longitudinal modications
angle of the gear to be ground (usually 15 or
20). Under these conditions, the generating cir- The main disadvantage is the lack of uniform
cle coincides with the pitch circle of the gear. transition between the ank and the tooth root,
This method produced a criss-cross pattern of which is objectionable in some gearing applica-
grooves (lay of surface roughness) and has been tions.
widely used because machine settings are sim- Rack-Tooth Worm Wheels. The wheel
ple and the desired pressure angle is accurately used in this method of generating gear teeth is
and simply set by adjusting the inclination crush trued into the shape of a conventional
angles of the wheels. rack-tooth worm or a CBN-plated rack-tooth
The 15 or 20 wheel position is not suitable worm. Grinding a gear by this method is illus-
for making longitudinal modications. Al- trated in Fig. 29.
though the two simultaneous points of contact The workpiece is mounted vertically on an
between the grinding wheels and the respective arbor or in a xture that ts between the centers
tooth anks move along the tooth-form invo- of the machine. For grinding helical gears, the
lutes during each transverse rolling-generating workholding slide of the machine is set at an
motion, the locations of these points relative to angle equal to the helix angle required on the
each other change constantly. gear. The workpiece is rotated by a motor that is
For helical gears, this differential position synchronized with the wheel motor.
applies not only along the involutes but also to The gear can be ground with a small portion
the position in the longitudinal tooth direction. of the length of the grinding worm wheel. Gen-
Therefore, it would be impractical to try to coor- eration takes place as a result of rotating the
dinate modication impulses of the grinding worm wheel and gear in mesh as the wheel is
wheel and the generating or feed motion to traversed axially across the gear face.
make longitudinal modications. A soft, conventional aluminum oxide or sili-
In a newer method, the angle between the two con carbide wheel is used to prevent burning the
wheels is 0, as shown in illustration (b) in hardened teeth, but this results in greater wheel
Table 5. The active surfaces of the wheels are breakdown and loss of correct tooth prole. To
parallel and face each other. Because the grind- continue using a fresh, unbroken part of the
ing pressure angle is 0 and corresponds to the wheel, the work is fed in a direction parallel to
pressure angle on the base circle, the generating the axis of the worm using the setup shown in
motion is obtained by rolling on the base circle. Fig. 30.

Table 5 Grinding of gears with dish-shaped wheels in the 0 position

Details for seven different spur and helical gears that had been carburized and hardened.
Parameter Gear details

Diametral pitch 10 5 4 3 5 3 2.75

Module 2.5 5 6 8 5 8 9.5
Number of teeth 23 44 80 80 120 100 379
Face width, mm 55.9 61.0 119 119 300 180 419
(in.) (2.2) (2.4) (4.7) (4.7) (11.8) (7.1) (16.5)
Tip diameter, mm 71.1 234 503 653 630 828 3350
(in.) (2.8) (9.2) (19.8) (25.7) (24.8) (32.6) (132)
Pressure angle, 15 20 20 20 20 15 15
Helix angle, degree 26 20 10 10 15 0 7
Grinding allowance 0.15 0.18 0.20 0.23 0.18 0.23 0.25
per ank, mm (0.006) (0.007) (0.008) (0.009) (0.007) (0.009) (0.010)
120 / Gear Materials, Properties, and Manufacture

Success in gear grinding with a conventional for each pitch and pressure angle required. In
grinding worm wheel depends greatly on care- diamond dressing, two diamonds are needed to
ful dressing of the wheel to the required contour dress both sides of the thread at the same time.
and on synchronization of the workpiece rota- Gear tooth design can be easily modied by lap-
tion with the rotation of the worm wheel. Spe- ping the tip of the dressing diamond.
cial diamond wheel dressers are required. The Burning the hardened steel is much less of a
accuracy of the diamond and of the dressing problem when using CBN-plated wheels be-
operation determine to a great extent the quality cause of the open texture of the wheel surface.
of the gear teeth produced. The hardness of the CBN material and the stiff-
Wheels are about 50 mm (2 in.) wide, and they ness of the steel core allow for several thousand
can be rough formed by crushing and can be n- pieces to be ground before the worm needs
ished to the proper pressure angle by diamond replating and returning. Modications of the
dressing. Some machines use wheels that have gear tooth design are manufactured into the core
been nish dressed by crushing only. Because a before it is plated.
considerable amount of time is required to crush Spur gears and helical gears up to 508 mm (20
true a new wheel, it is advisable to stock wheels in.) in diameter can be ground in machines using
a threaded wheel. Although ne-pitch gears can
be ground from a solid gear blank, gear teeth with
diametral pitch of 24 or coarser should be cut
prior to grinding.
The advantages of gear grinding with a
threaded wheel are accuracy, speed, and ease of
incorporating changes in tooth design. Machines
are also available that crown the gear teeth during
grinding. Because mechanical grinders are com-
plicated and sensitive, a considerable amount of
operator skill is required. However, with elec-
tronic CNC machines, crowning is a standard
feature. Gear grinding with a threaded wheel is
six to thirty times as fast as the other methods dis-
cussed in this section.

Grinding of Bevel Gears

When accuracy is required, grinding is the
Fig. 28 Schematic of dish-shaped wheels simulating the most economical method for nishing teeth as
motion of a rack. The gear being machined rolls and coarse as 1.5 diametral pitch on straight bevel,
moves. Wheels may or may not be reciprocated.
spiral bevel, and hypoid gears. With good grind-
ing practice, bevel and hypoid gears can be n-
ished with a tooth-to-tooth spacing accurate

Fig. 30 Setup used to grind an aircraft pinion by the worm

Fig. 29 Generation grinding with a rack-tooth worm wheel wheel method
Chapter 5: Machining, Grinding, and Finishing / 121

within 0.005 mm (0.0002 in.), eccentricity with used with several revolutions of the workpiece,
bore or shaft within 0.0064 mm (0.00025 in.), as or CBN-plated wheels with one revolution of
little backlash as 0.025 to 0.051 mm (0.001 to the workpiece.
0.002 in.), and a surface nish of 0.38 to 0.76 An A-46-J-V wheel is the conventional
m (15 to 30 in.). In general, any bevel or wheel commonly used to grind gears ranging
hypoid gear that is cut with a circular cutter can from 2 to 10 diametral pitch. Wheel speed
be ground with a wheel of similar shape and ranges from 1200 to 1400 m/min (3800 to 4500
grinding action, except nongenerated gears that sfm). The wheel is dressed automatically at pre-
are cut with a face mill. determined intervals between grinds.
Nongenerated (Formed) Gears. The gear Cubic boron nitride-plated wheels cover the
teeth are ground by two methods. The rst same range of diametral pitch. Wheel speeds
method involves a cup-shaped wheel that simul- range from 1400 to 2000 m/min (4500 to 6500
taneously grinds the facing sides and root of sfm).
adjacent teeth. There is no generating motion, Generated Gears. Spiral bevel and hypoid
and the wheel produces a curved tooth whose gears are ground with a cup-shaped wheel. The
sides have a straight prole. Three or four revo- conventional grinding wheel is dressed to pro-
lutions of the workpiece are required for the duce a smooth blend of ank and root prole.
complete nish grinding of a bevel gear using The core of the CBN wheel is machined to pro-
conventional dressable wheels. A predeter- duce a smooth blend of the ank and tooth pro-
mined amount of stock is removed with each le. The curved prole of the tooth is generated
revolution of the workpiece. With CBN-plated by relative rolling motion between the grinding
wheels, the entire nishing stock is removed in wheel and the workpiece.
one revolution of the workpiece. The grinding Vitried bond and resinoid bond aluminum
wheel makes line contact with the tooth being oxide cup-shaped wheels are the conventional
ground. This method is applicable for nishing wheels used for spiral bevel and hypoid gears.
hypoid and spiral gears and for special spiral Typical wheel classications are A-60-J-V and
gears that have a 0 mean spiral angle. A-60-M-B. Wheel speed ranges from 1200 to
The second method is called the Waguri 1400 m/min (3800 to 4500 sfm). Cubic boron
method. The grinding wheel is circular and has nitride-plated wheels are also used.
an average point width of 0.25 to 0.50 mm Generating gear teeth with a cup-shaped
(0.010 to 0.020 in.) less than the slot width of wheel produces spiral bevel gears with highly
the workpiece. The wheel axis follows a circu- accurate tooth spacing, concentricity, and pro-
lar path with a diameter equal to the difference le shape. The following is an example of a typ-
between its point width and its slot width. Thus, ical application of this grinding method.
the desired surfaces are ground with line contact Example 2: Grinding Spiral Bevel Gears
(Fig. 31). Conventional dressable wheels can be With a Cup-Shaped Wheel. The spiral bevel
gear illustrated in Fig. 32 was carburized, hard-
ened, and tempered before being ground by the
generation method with a cup-shaped wheel. An
automatic wet-type 3.7 kW (5 hp) machine
designed for grinding spiral bevel, Zerol, and
hypoid gears was used for this application.
Additional manufacturing information is given
in the table with Fig. 32.

Grinding Fluids
The use of a grinding uid is generally rec-
ommended for the grinding of steel gears. A
grinding uid prolongs wheel life between
dressings, ushes away chips, and improves
gear tooth nish. Flooding the work surface
minimizes the possibility of burning the sur-
faces of gear teeth. The complete absence of
grinding burn is essential for high-quality gears.
Fig. 31 Waguri grinding method for nongenerated spiral
Virtually all the grinding uids mentioned in the
bevel and hypoid gears
122 / Gear Materials, Properties, and Manufacture

section Cutting Fluids in this chapter have is more desirable than some other nish de-
been used successfully for the grinding of gears. pends on the service requirements of the gear. In
In many critical gear-grinding applications, some applications, a gear with a ground nish
mineral-base sulfochlorinated or sulfurized oils operates more quietly, even though grinding has
are used. Plain soluble-oil emulsions have not improved the dimensional accuracy of the
proved satisfactory for many gear grinding appli- product. The effect of a ground nish, in com-
cations. Chemical solutions have also been satis- parison with other nishes, on the lubrication of
factory. gears has not been precisely evaluated.
The surface nish of ground gears usually
Surface Finish ranges from 0.38 to 0.80 m (15 to 32 in.).
Grinding produces a nish unlike that pro- Under conditions of unusually good control, a
duced by any other process. Whether this nish surface nish of 0.25 m (10 in.) or better can
be produced. Surface nish sometimes becomes
the major consideration in determining whether
a gear will be ground. Grinding can have the
dual purpose of correcting dimensional change
from heat treating and of producing a ne sur-
face nish.

Honing of Gears
Honing is a low-velocity abrading process
that uses bonded abrasive sticks to remove stock
from metallic and nonmetallic surfaces. As one
of the last operations performed on the surface
of a part, honing generates functional character-
istics specied for a surface, such as geometric
accuracy, dimensional accuracy, and surface
features (roughness, lay pattern, and integrity).
It also reduces or corrects geometric errors
resulting from previous operations.
The most common application of honing is
on internal cylindrical surfaces (Fig. 33a). How-
ever, honing is also used to generate functional
characteristics on external cylindrical surfaces,
at surfaces, truncated spherical surfaces, and
toroidal surfaces (both internal and external). A
characteristic common to all these shapes is that
they can be generated by a simple combination
of motions.
Gear-tooth honing is designed to improve
geometric accuracy and surface conditions of a
hardened gear. The teeth of hardened gears are
honed to remove nicks and burrs, to improve
nish, and to make minor corrections in tooth
shape. Gear teeth are honed on high-speed
machines specially designed for the process
(Fig. 34). The honing tool is like a gear driving
the workpiece at high speed (up to 30 m/min, or
100 sfm) while oscillating so that the teeth slide
axially against the workpiece.
Spur gears and internal or external helical
Fig. 32 Grinding of a spiral bevel gear. Dimensions in gure gears ranging in diametral pitch from 24 to 2.5,
given in inches in outside diameter from 19 to 673 mm ( to
Chapter 5: Machining, Grinding, and Finishing / 123

26 in.), and up to 75 mm (3 in.) in face width Plastic tools are supplied with abrasives of
have been honed on these machines. Finishes of 60-grit to 500-grit size. Size of abrasive, gear
0.75 m (30 in.) are easily achieved, and n- pitch, and desired nish are usually related as:
ishes of 0.075 to 0.10 m (3 to 4 in.) are pos-
sible. Both taper and crown honing can be done. Finish

Tools used in honing gear teeth are of two Grift size Gear pitch m in.
types, a helical gear shape tool made of abrasive 60 16 0.750.89 3035
impregnated plastic, and a metal helical gear 100 1620 0.630.75 2530
with a bonded abrasive coating that is renew- 180 >20 0.380.50 1520
280 >20 0.250.30 1012
able. The plastic tool, which is discarded at the 500 >20 0.0750.10 34
end of its useful life, is widely used. The metal
tool is used mainly for applications in which Honing tools do not load up, and a plastic hon-
plastic tools would be likely to break; also, it is ing tool can wear until its teeth break. Stock
used primarily for ne-pitch gears. removal of 0.025 to 0.050 mm (0.001 to 0.002
in.) measured over pins is the recommended
Methods. The two methods used to hone
gear teeth are the zero-backlash method and the
constant-pressure method. In the zero-backlash
method, which is used for gears made to com-
mercial tolerances, the tool head is locked so
that the distance between the center of the work
gear and the center of the honing tool is xed
throughout the honing cycle. In the constant-
pressure method, which is used for gears pro-
duced to dimensions outside commercial toler-
ance ranges, the tool and the work gear are kept
in pressure-controlled tight mesh.
Applicability. The use of honing for remov-
ing nicks and burrs from hardened gears can
result in a considerable cost saving in compari-
son to the usual method. In the usual method,
the gears are tested against master specimens on
sound test machines. Nicks indicated are
searched for and removed using a hand grinder.
The gear is then retested to make certain the
nick has been removed. When honing is used,
all of these various tests and procedures can be
Fig. 33 (a) Schematic representation of honing. (b) eliminated.
Schematic representation of supernishing
Some shape correction can be achieved in the
removal of 0.050 mm (0.002 in.) of stock by
honing. A helical gear 127 mm (5 in.) in diame-
ter may show lead correction of 0.010 mm
(0.0004 in.), involute prole correction of
0.0075 mm (0.0003 in.), and eccentricity cor-
rection of 0.010 mm (0.0004 in.).
The advisability of using honing for sal-
vaging hardened gears hinges on cost consider-
ations. As the error in tooth shape increases,
honing time increases and tool life decreases.
On the other hand, if the gears represent a large
investment in production time and material,
honing may be the most economical method.
Because honing is not designed for heavy
Fig. 34 Honing of helical gear teeth stock removal or tooth correction, it cannot be
124 / Gear Materials, Properties, and Manufacture

substituted for grinding or shaving of gears. in.) pitch diameter down to the smallest gear that
Rotary shaving usually leaves gear teeth smooth can be manufactured. Production lapping
within 0.25 to 1.00 m (10 to 40 in.). machines can be adjusted to lap gears with shaft
Microhoning is a low-velocity abrading angles of 0 to 180.
process very similar to honing. However, unlike Lapping Media. Optimum grit size varies
conventional honing, microhoning focuses pri- with different types and sizes of gears. A 280-
marily on the improvement of surface nish and grit abrasive is used for spiral bevel gears; a
much less on correction of geometric errors ner abrasive (about 400-grit) is more suitable
(Fig. 33b). As a result, the pressures and ampli- for hypoid gears because the sliding action is
tude of oscillation applied during microhoning greater. As a rule, coarser grit is used for gears
are extremely small. This process is also re- having a coarse pitch, and ner grits are used for
ferred to as microsurfacing, microstoning, and gears having a ne pitch. When compound is
supernishing. brushed on, as in manual operation, a paste-type
vehicle is used. However, in semiautomatic or
automatic lapping, the abrasive should be mixed
Lapping of Gears with a thin oil (such as mineral seal oil) so that
it can be pumped to the workpieces (Fig. 35).
Lapping is the process of nishing work mate- Processing Techniques. It is important to
rials by applying a loose abrasive slurry between roll all mating gears together before lapping to
a work material and a closely tting surface, detect nicks and burrs, which can be removed
called a lapping plate. When loose abrasive is by a small portable hand grinder before the
used to machine the work material, it may slide, gears are lapped. This preliminary rolling also
roll, become embedded, or do all three, depend- inspects tooth contact, which should be in the
ing on the shape of the abrasive grain and the same location for each set of gears and is espe-
composition of the backup surface. cially important in automatic lapping.
Gear lapping corrects the minute errors in During the lapping operation, the pinion
involute prole, helix angle, tooth spacing, and (smaller gear) is used as the driver, and the
concentricity created in the forming or cutting larger gear is the driven member. The driven
or in the heat treatment of the gears. The lapping spindle is also used for applying the necessary
can be done by running a set of gears in mesh or tooth-contact load by adjusting to a slight drag.
by running one gear with a gear-shaped master
lapping tool.
Gear lapping is most often applied to sets of
hardened gears that are required to run silently
in service. Gear lapping is strictly a mating
process and is not intended for stock removal.
Two gears that have been matched by lapping
should be operated as a set, and they should be
replaced as a set, rather than singly.
Gears are lapped in special machines, which
can be arranged for manual, semiautomatic, or
automatic operation. In semiautomatic opera-
tion, loading and unloading are manual; in auto-
matic operation, loading and unloading are done
automatically in accordance with a programmed
Angular, spur, and helical gears can be lapped,
but the process is mainly applied to spiral bevel
gears and hypoid gears. A typical setup used for
lapping hypoid gears is shown in Fig. 35. Gears
of up to 915 mm (36 in.) pitch diameter can be
lapped in semiautomatic or automatic operation.
Manual lapping, in special equipment, is em-
ployed for gears of approximately 2540 mm (100 Fig. 35 Setup for the lapping of hypoid gears
Chapter 5: Machining, Grinding, and Finishing / 125

Running cycles as short as 15 s at about 76 Supernishing

m/min (250 sfm) can frequently produce de-
sired results. However, longer time cycles may Supernishing processes are those that can
be necessary, depending on initial gear-tooth achieve a surface roughness average (Ra) of
nish and service requirements. 0.10 m (4 in.) (Ref 3). As shown in Fig. 36,
Low noise level is the criterion of successful supernishing can be achieved under carefully
gear lapping. Because gear lapping is strictly a controlled conditions by grinding, honing/
mating process and no stock removal is in- microhoning, lapping, polishing /electropolish-
tended, measurements are not made as in most ing, or combinations of nishing processes (for
other lapping processes. Minor corrections in example, grinding, honing, and lapping). In the
tooth bearing shape and position can be ob- past, such processing was rarely carried out on
tained, however. Lapping does improve the n- gears because of the time and expense involved.
ish of gear teeth, but improved nish is seldom Recently, however, there have been a number of
the purpose of gear lapping. studies that show that the reduction of surface
roughness improves the lubricating condition of

Fig. 36 Surface roughness values produced by common manufacturing/production methods. The ranges shown are typical of the
processes listed. Higher or lower values can be obtained under special conditions. Source: Ref 3
126 / Gear Materials, Properties, and Manufacture

Fig. 37 Weibull plot of surface fatigue test results for both carburized and ground and carburized, ground, and supernished gears.
Source: Ref 4

gears and offers the possibility of increasing their gears were about four times longer than those of
surface fatigue lives (Ref 4). the conventionally ground gears.
In order to quantify the effect of surface
roughness on fatigue life, a series of tests were
carried out on two sets of 28-tooth, 8-pitch gears ACKNOWLEDGMENTS
made from consumable-electrode vacuum-
melted (aerospace quality) AISI 9310 steel (Ref This chapter was adapted from:
4). Both sets of gears were case carburized and T.J. Krenzer and J.W. Coniglio, Gear Man-
ground. One set of gears was supernished by ufacture, Machining, Vol 16, Metals Hand-
placing them in a vibrating bath consisting of book, ASM International, 1989, p 330355
water, detergent, abrasive powder, and small Honing, Machining, Vol 16, Metals Hand-
pieces of zinc for several hours (details of the book, ASM International, 1989, p 472491
supernishing treatment can be found in Ref 5). P. Lynah, Lapping, Machining, Vol 16,
Upon removal from the bath the gears had Metals Handbook, ASM International,
highly polished, mirrorlike surfaces. The 1989, p 492505
supernishing treatment removed about 2 to
3 m (79 to 118 in.) of material from the tooth
surfaces. The effects of the supernishing on the REFERENCES
quality of the gear tooth surfaces were deter-
mined from metrology, prolometry, and inter- 1. W.N. Ault, Grinding Equipment and
ferometric microscope inspections. The su- Processes, Machining, Vol 16, Metals
pernishing reduced the roughness average by Handbook, ASM International, 1989, p
about a factor of 5. The mean Ra value before 430452
supernishing was 0.380 m (15.0 in.). After 2. K. Subramanian, Superabrasives, Machin-
supernishing the mean Ra value was lowered ing, Vol 16, Metals Handbook, ASM Inter-
to 0.070 m (2.8 in.). The gears were then national, 1989, p 453471
fatigue tested at a Hertzian contact stress of 1.71 3. M. Field, J.F. Kahles, and W.P. Koster,
GPa (248 ksi) for 300 million cycles or until Surface Finish and Surface Integrity,
surface failure occurred on any one tooth. As Machining, Vol 16, Metals Handbook, 9th
shown in Fig. 37, the lives of the supernished ed., ASM International, 1989, p 1936
Chapter 5: Machining, Grinding, and Finishing / 127

4. T.L. Krantz, M.P. Alanou, H.P. Evans, and the NASA Scientic and Technical Infor-
R.W. Snidle, Surface Fatigue Lives of mation Center (http://www.grc.nasa.gov)
Case-Carburized Gears With an Improved
Surface Finish, J. Tribology (Trans. SELECTED REFERENCES
ASME), October 2001, Vol 123, p
709716 Dudleys Gear Handbook: The Design,
5. R.W. Snidle, H.P. Evans, M.P. Alanou, Manufacture, and Application of Gears,
The Effect of Supernishing on Gear Tooth 2nd ed., D.P. Townsend, Ed., McGraw-Hill
Prole, Report AD-327916, June 1997, Inc., 1991
available from the Defense Technical D.W. Dudley, Handbook of Practical Gear
Information Center (DTIC), the National Design, McGraw-Hill Book Company,
Technical Information Service (NTIS), or 1984
Gear Materials, Properties, and Manufacture Copyright 2005 ASM International
J.R. Davis, editor, p129-137 All rights reserved.
DOI:10.1361/gmpm2005p129 www.asminternational.org


Casting, Forming, and Forging

GEAR MANUFACTURE depends on ma- For example, cast tooth internal gears (Fig. 3)
chinery available, design specications or are produced in several sizes up to 1633 kg
requirements, cost of production, and type of (3600 lb). They are heat treated to strength lev-
material from which the gear is to be made. els of 689 MPa (100 ksi) and machining is not
There are many methods for manufacturing required on these gears. In circumstances where
gears including: machining is necessary, the machining expense
is reduced by casting close to nal shape.
Metal removal processes (hobbing, shaping, Another example of a cast tooth gear is the
milling, shaving, grinding, honing, and lap- pinion gear produced from cast high-manganese
ping) as discussed in Chapter 5 (Hadeld) steel for an electric mining shovel
Various casting processes for both produc- shown in Fig. 4. It was not necessary to machine
tion of gear blanks and near-net shape gears the gear teeth.
Stamping and ne blanking Specic Casting Processes. Most casting
Cold drawing and extrusion processes have been used to produce gear
Powder metallurgy (P/M) processing as dis- blanks or cast tooth gears including sand cast-
cussed in Chapter 7 ing, shell molding, permanent mold casting,
Injection molding (applicable to plastic gear centrifugal casting, investment casting, and die
materials as discussed in Chapter 4)
Gear rolling
Forging for production of gear blanks and
precision-forged near- and net-shape gears
Table 1 Recommended tolerances in terms
Most of the processes listed are suited to gears of AGMA quality numbers for various gear
for low wear requirements, low power transmis- manufacturing processes
sion, and relatively low accuracy of transmitted Highest quality number
motion (Ref 1). When the application involves Process Normal With extra care
higher values of one or more of these character- Sand mold casting 1 3
istics, forged or cut/machined gears are used. Plaster mold casting 3 5
Table 1 lists the tolerances in terms of American Permanent mold casting 3 5
Investment casting 4 6
Gear Manufacturers Association (AGMA) Die casting 5 8
quality numbers for various gear manufacturing Injection and compression molding 4 8
Powder metallurgy 6 9
processes (Ref 2). Stamping 6 9
Extrusion 4 6
Cold drawing 6 9
Milling 6 9
Casting Hobbing 8 13
Shaping 8 13
Broaching 7 12
Although the casting process is used most Grinding 10 15
often to make blanks for gears which will have Shaving 9 14
Honing 10 15
cut teeth (Fig. 1 and 2), there are several varia- Lapping 10 15
tions of the casting process that are used to
Source: Ref 2
make toothed gears with little or no machining.
130 / Gear Materials, Properties, and Manufacture

casting. Cut gears have also been produced from

continuously cast bars. Some of the commonly
employed processes will be briey reviewed
below. More detailed information on these
processes can be found in Casting, Volume 15
of the ASM Handbook.
Sand casting is used primarily to produce
gear blanks. In recent times there has been only
very limited use of gears with teeth made by
sand casting (Ref 4). In some instances gears for
farm machinery, stokers, and some hand-oper-
ated devices have used cast teeth. The draft on
the pattern and the distortion on cooling make it
difcult to obtain much accuracy in cast iron or
cast steel gear teeth. Table 1 shows that sand
cast gears have the lowest AGMA quality levels
of the major gear manufacturing methods.
The shell molding process is particularly
suited to castings for which:

The greater dimensional accuracy attainable

Fig. 1 Cast steel gear blank. Weight: 478 kg (1053 lb). with shell molding (as compared with con-
Source: Ref 3 ventional green sand molding) can reduce

Fig. 2 Large machined cast steel gear. Source: Ref 3

Chapter 6: Casting, Forming, and Forging / 131

the amount of machining required for com- out of materials which are so hard that teeth can-
pletion of the part not be readily produced by machining (Ref 4).
As-cast dimensions may not be critical, but The process can be used with a variety of steels,
smooth surfaces (smoother than can be bronzes, and aluminum alloys. With machinable
obtained by sand casting) are the primary materials, the process is still useful if the gear is
objective. An example of a cast tooth bevel integral with some complicated shape that is
gear with an excellent surface nish pro- very difcult to produce by machining.
duced by shell molding is shown in Fig. 5. Large quantities of small, low-cost gears are
made by the cold chamber die casting process
The investment casting process has also had (die cast gears are usually under 150 mm (6 in.) in
limited use in gear manufacture. Its most appar- diameter and from 10 to 48 diametral pitch, DP).
ent value lies in the making of accurate gear teeth Complicated gear shapes which would be quite
costly to machine can be made quickly and at low
cost by the die casting process. The main disad-
vantage of the process is that the low-melting
point metals suitable for die castingaluminum,
zinc, and copperdo not have high enough hard-
ness for high load-carrying capacity.
Many different types of gears can be die cast,
such as spur, helical worm, clusters, and bevel
gears. Applications for these types of gears
include toys, washing machines, small appli-
ances, hand tools, cameras, business machines,
and similar equipment.

Stamping and Fine Blanking. Stamping is
a metalworking technique that has been com-
pared to using a cookie cutter. In this process, a
sheet of metal is placed between the top and bot-
Fig. 3 Cast tooth internal gears. Weights up to 1633 kg (3600 tom portions of a die; the upper die is pressed
lb). Source: Ref 3 into the lower section and removes or cuts the
gear from the sheet. This is a low-cost, very
efcient method for producing lightweight
gears for no-load to medium-duty applications.
Stamping is restricted by the thickness of the

Fig. 5 Cast tooth bevel gear produced by the shell molding

Fig. 4 Cast tooth pinion gear for an electric mining shovel. process to obtain excellent surfaces and close toler-
Weight: 212 kg (468 lb). Source: Ref 3 ances. Source: Ref 3
132 / Gear Materials, Properties, and Manufacture

workpiece and is used primarily for spur gears duced. As the name implies, a bar is pulled
and other thin, at forms (Ref 5). Stamped gears (drawn) or pushed (extruded) through a series of
range in size from 20 through 120 DP and 0.25 several dies, the last having the nal shape of
to 3 mm (0.010 to 0.125 in.) thick (Ref 6). As the desired tooth form. As the material is run
the pitch becomes ner, the materials specica- through these dies it is actually squeezed into
tion must become thinner. Table 2 shows rec- the shape of the die. Since the material is dis-
ommended stock thicknesses for various pitches placed by pressure, the outside surface is work
that are commonly used and require no special hardened and quite smooth.
care in die maintenance. As shown in Table 1, The bars that are blanks for this process are
tolerances for stamped gears are good and usually 3 to 3.7 m (10 to 12 ft) in length. After
AGMA quality class 9 can be achieved with passing through the dies, they are known as pin-
extra care. ion rods, and often are put into screw machines
A wide range of materials can be processed by that nish the individual gears. Experience has
stamping, including all the low and medium car- shown that it is more economical to slice a seg-
bon steels, brasses, and some aluminum alloys. ment off an extruded bar than to cut an individ-
Nonmetallic materials can also be stamped. ual gear. In some cases it would be impossible
Gears manufactured by this process are used in to produce the desired shape of pinion in any
toys, clock and timer mechanisms, watches, other way. Pinion rods from 16 to 100 DP can
small appliances such as mixers, blenders, toast- be obtained, but as the pitch becomes ner, it
ers, and can openers, as well as larger appliances becomes more difcult to obtain the close toler-
such as washers and dryers. ances that are sometimes desired on ne-pitch
Fine blanking (also known as ne-edge pinions. Any material that has good drawing
blanking) is actually more akin to cold extrusion properties, such as high-carbon steels, brass,
than to a cutting operation such as stamping. bronze, aluminum, and stainless steel, may be
The process takes metal from a sheet like stamp- used for the drawn pinion rod.
ing but differs from it in that it uses two dies and Gears and pinions manufactured by this pro-
forms the workpiece by pressing it into the cess have a large variety of applications and have
desired shape. The metal is extruded into the die been used on watches, electric clocks, spring-
cavities to form the desired shape. Also unlike wound clocks, typewriters, carburetors, magne-
stamping, ne blanking offers the designer a tos, small motors, switch apparatus, taximeters,
limited three-dimensional capability and can cameras, slot machines, all types of mechanical
thus be used to create bevels, multiple gear sets, toys, and many other parts for machinery of all
and other complex forms (Ref 5). Fine blanked kinds.
gears can be found in a wide range of applica- Gear Rolling (Ref 1). Spur and helical
tions including the automotive, appliance, ofce gears, like splines, are roll formed. Millions of
equipment, hydraulic, and medical equipment high-quality gears are produced annually by this
industries. process; many of the gears in automobile trans-
Cold Drawing and Extrusion (Ref 6). This missions are made this way. As indicated in Fig.
process requires the least tool expenditure for 6, the process is basically the same as that by
mass production of spur gear toothed gear ele- which screw threads are roll formed, except that
ments and is extremely versatile in that almost in most cases the teeth cannot be formed in a
any tooth form that can be desired can be pro- single rotation of the forming rolls; the rolls are
gradually fed inward during several revolutions.
Because of the metal ow that occurs, the top
lands of roll-formed teeth are not smooth and
Table 2 Recommended stock thicknesses for perfect in shape; a depressed line between two
stamped gears slight protrusions can often be seen. However,
Thickness because the top land plays no part in gear tooth
Pitch range mm in.
action, if there is sufcient clearance in the mat-
ing gear, this causes no difculty. Where de-
2036 0.511.98 0.0200.078
3860 0.381.57 0.0150.062 sired, a light turning cut is used to provide a
6272 0.251.02 0.0100.040 smooth top land and correct addendum diameter.
7490 0.250.89 0.0100.035 Rolling produces gears 50 times as fast as
92120 0.250.64 0.0100.025
gear cutting and with surfaces as smooth as 0.10
Source: Ref 6
m (4 in.). Not only does rolling usually need
Chapter 6: Casting, Forming, and Forging / 133

no nish operation, but rolling renes the mi- by open-die forging, closed-die forging (Fig. 7),
crostructure of the workpiece. and hot upset forging. During the past thirty-
Production setup usually takes only a set of ve years there has been considerable research
rolling dies and the proper xture to equip the and development aimed at producing near-net
rolling machine. By either the infeed (plunge) or net-shape gears by precision forging. Today
method or the throughfeed method, the rolling precision forged gears requiring little or no n-
dies drive the workpiece between them, forming ish machining are commonly used in the auto-
the teeth by pressure. motive, truck, off-highway, aerospace, railroad,
Limits. Spur gears can be rolled if they have agriculture, and material handling industries, as
18 teeth or more. Fewer teeth cause the work to well as the energy and mining elds.
roll poorly. Helical gears can be rolled with
fewer teeth if the helix angle is great enough. High-Energy Rate Forging
It is usually impractical to roll teeth with One of the rst forging processes for manu-
pressure angle less than 20. Lower angles have facturing near- or net-shape gears was the high-
wide ats at root and crest that need more pres- energy rate forging process which is a closed-
sure in rolling. Lower angles also hinder metal die hot or cold forging process in which the
ow. Although 0.13 mm (0.005 in.) radius l- work metal is deformed at unusually high veloc-
lets can be rolled, 0.25 mm (0.010 in.) is a bet- ities. Ideally, the nal conguration of the forg-
ter minimum. For greater accuracy, gear blanks ing is developed in one blow, or at most, a few
are ground before rolling. Chamfers should be blows. Velocity of the ram, rather than its mass,
30 or less. generates the major forging force.
Steels for gear rolling should not have more It is possible to produce gears with a con-
than 0.13% S and preferably no lead. Blanks toured grain ow that follows the conguration
should not be harder than 28 HRC. of the teeth using high-energy-rate forging. In
the case of spur gears, this is achieved by pan-
caking to cause lateral ow of the metal in a die
Forging containing the desired tooth conguration at its
periphery. Contoured grain increases the load-
Forging has long been used in the manufac- bearing capacity without increasing the tooth
ture of gears. This is particularly true for the size. In addition, the process minimizes the
production of gear blanks which would subse- machining required to produce the nished
quently be cut/machined into the nal desired gear. Although spur gears are the easiest to
conguration. Gear blanks have been produced forge, helical and spiral-bevel gears can also be

Fig. 7 Gear blank that was closed-die forged in four hammer

blows from pancaked stock (not shown), and then
Fig. 6 Method for forming gear teeth and splines by cold trimmed and pierced in one press stroke. Dimensions are given
forming. Source: Ref 1 in inches.
134 / Gear Materials, Properties, and Manufacture

forged if their congurations permit ejection of were forged with 0.51 mm (0.020 in.) of stock
the gear from the die cavity. Gears have been for nish machining.
forged from low-alloy steel, brass, aluminum The die inserts originally used to forge this
alloys, stainless steel, titanium, and some of the gear were made of H11 or H13 tool steel. This
heat-resistant alloys. steel typically softened after producing 20 gears
Gears with a DP of 5 to 20 are commonly because of its temperature rising above the 565
forged with little or no machining allowance. C (1050 F) tempering temperature of H13
The die life decreases signicantly when forg- steel. The use of Alloy 718 (UNS N07718) was
ing ner-pitch gears. found to improve the die insert life.
The forging of 5 DP gears with an involute The automotive ywheel shown in Fig. 9,
tolerance of 0.013 mm (0.0005 in.) and total 272.49 mm (10.728 in.) in diameter over the
composite error of 0.08 mm (0.003 in.) has been teeth and weighing 11 kg (24 lb), was forged
reported. These gears were forged with a tooth- from a machined blank cast from class 40 gray
to-tooth spacing deviation of about 0.025 mm iron (generally considered unforgeable). The
(0.001 in.) and a total accumulated deviation of machined preform, a section of which is shown
0.089 mm (0.0035 in.). Over-the-pins dimen- in Fig. 9, was heated to 955 C (1750 F) and
sions were held to 0.05 mm (0.002 in.) on forged at an energy level of 271,000 J (200,000
these gears, and the total composite error was ft lbf). This part had the smallest tolerance
about 0.20 mm (0.008 in.). specication. The diameter over the teeth and
Holding gear dimensions to extremely close the thickness of the body had a tolerance of
tolerances may eliminate nish machining, but +0.00 mm, 0.18 mm (+0.000 in., 0.007 in.).
the savings may be exceeded by higher die mak- The largest tolerance on the part was 1.02 mm
ing/maintenance costs. Consequently, most (0.040 in.) on the diameter of a recess. Toler-
forged gears have an allowance for machining. ances on the other recesses were 0.18 mm
A surface nish of 0.5 to 1.5 m (20 to 60 (0.007 in.) and +0.48 mm, 0.00 mm (+0.019
in.) on gear teeth is practical. However, even in., 0.000 in.). This gear was forged to the n-
with a 0.5 m (20 in.) nish, local imperfec- ished dimensions.
tions can increase the average to 1.5 m (60 Various gears with teeth as an integral part
in.) or greater. Therefore, it would be difcult have been forged. These have ranged in outside
to maintain a good surface nish on gear teeth diameter from 64 to 267 mm (2.5 to 10.5 in.)
without grinding. and in weight from 0.54 to 11 kg (1.2 to 24 lb).
Typical Gear Forgings. The 4.5 kg (10 lb) Most have been made with 0.13 to 0.51 mm
gear shown in Fig. 8 was forged from 8620 steel (0.005 to 0.020 in.) of stock on the ank of each
billet 75 mm (3 in.) in diameter by 124 mm (4.9 tooth for nish hobbing and grinding. Gears
in.) in length. An energy level of 353,000 J forged with integral teeth normally have longer
(260,000 ft lbf) was needed to forge the gear in
one blow at 1230 C (2250 F). The web on the
gear was forged to nal thickness; the teeth

Fig. 8 Near-net shape cluster gear made by high-energy rate Fig. 9 Near-net shape automotive ywheel made by high-
forging. Dimensions are given in inches. energy rate forging. Dimensions are given in inches.
Chapter 6: Casting, Forming, and Forging / 135

fatigue and wear life than those made from a Near-Net Shape Quality Gears. The ma-
conventionally forged blank on which the teeth jority of forged gears produced today are near-
are hobbed, shaped, or milled. net shape congurations. Gear teeth are forged
with an envelope of material (stock allowance)
Precision Forging around the tooth prole. This envelope is subse-
The term precision forging does not spec-
ify a distinct forging process but rather de-
scribes a philosophical approach to forging. The
goal of this approach is to produce a net shape,
or at least a near-net shape, in the as-forged con-
The term net indicates that no subsequent
machining or nishing of a forged surface is
required. Thus, a net shape forging requires no
further work on any of the forged surfaces,
although secondary operations may be required
to produce minor holes, threads, and other such
details. A near-net shape forging can be either Fig. 10 Material/weight savings using the near-net shape
forging process. The large spur gear weighs 25 kg (55
one in which some but not all of the surfaces are lb) as a blank (left side). As a forged tooth gear (right side) with
net or one in which the surfaces require only 1 mm (0.04 in.) of stock allowance on the tooth prole for nish
machining, it weighs 17 kg (37 lb). Source: Presrite Corporation
minimal machining or nishing. Precision forg-
ing is sometimes described as close-tolerance
forging to emphasize the goal of achieving,
solely through the hot forging operation, the
dimensional and surface nish tolerances
required in the nished part.
In recent years, computer-aided design and
manufacturing (CAD/CAM) techniques have
been applied to various forging processes (Ref
7). This computerized approach is applicable to
precision hot forging of spiral bevel, spur, and
helical gears in conventional presses in that it
allows the die designer to examine the effects of
various process variables (loads, stresses, and
temperature) on the die design.
Precision hot-forged gears have the same
advantages over cut gears as other molded gears
(cast, P/M processed, injection molded) in that
there is little or no material lost (Fig. 10). This is
a cost savings from the standpoint of both the
cost of the material itself and, more importantly,
the cost of machining. In addition, precision
forged gears also have the advantage over cut
gears of increased load-carrying capacity. This
added strength in the form of increased fatigue
strength is due to the difference in grain ow be-
tween gears cut from bar stock and forged gears.
The grain ow in cut gears is determined by the
hot rolled orientation of the bar stock and has no
relationship to gear tooth contour. On forged
specimens, the grain ow follows the tooth con-
tour in every gear tooth. Figure 11 compares
the fatigue properties of cut and forged gears
Fig. 11 Fatigue data for (a) cut gears and (b) near-net shape
(Ref 8). forged gears. Source: Ref 8
136 / Gear Materials, Properties, and Manufacture

quently removed by the forging house or the cus- Near-net shape gears can be produced using
tomer purchasing the forged gears. any carburizing or induction hardening steel in
The manufacturing process begins with steel ve basic congurations: spiral bevel, helical,
bar stock, usually turned and polished to straight bevel, spur gears with a 1 mm (0.04 in.)
improve the surface, and cut to the exact weight. stock allowance, and spur gears with a 0.1 to 0.3
The exact weight is critical because the amount mm (0.004 to 0.012 in.) stock allowance. The
of steel must completely ll the die to produce near-net shape gears can be produced in diame-
the complete gear prole. Prior to forging, bil- ters up to 425 mm (17 in.) with stock allowances
lets are heated between 925 and 1230 C (1700 ranging from 0.1 to 1.5 mm (0.004 to 0.06 in.).
and 2250 F) in an electrical induction furnace The specications for various gear congura-
that is controlled by an optical pyrometer to 14 tions include:
C (25 F).
In a single stroke, standard mechanical forg- Spiral bevel gears can be produced up to 425
ing presses, ranging from 14,235 to 53,375 kN mm (17 in.) in diameter, with 0.5 mm (0.02
(1600 to 6000 tonf), form near-net shape gears in.) minimum stock per ank. A maximum
with the complete allowable stock allowance. spiral angle of 35 and a pitch range of less
The purpose of this rst operation, which forms than 7 DP can be achieved.
a pancake, is to break the scale off of the billet Straight bevel gears have congurations/
and size the outside diameter to just under the properties similar to spiral bevel gears
size of the root diameter in the gear die. Next, an Helical gears can be produced up to 250 mm
operator positions the billet into the nish die. (10 in.) in diameter and up to 40 kg (90 lb)
After forging, a hydraulic knockout system in weight, with a 0.5 mm (0.02 in.) mini-
immediately extracts the gear from the nish die. mum stock per ank. A maximum helix
After the raw forged gear is hydraulically angle of 25 and a pitch range of 4 to 12 DP
ejected from the die, it is placed in a trimming can be achieved.
nest where the hole is punched. It is then allowed Spur gears with a 1 mm (0.04 in.) stock
to cool to ambient temperature, which usually allowance can be produced up to 400 mm
takes up to 24 h. Once cooled, it is ready for n- (16 in.) in diameter and up to 135 kg (300 lb)
ish machining. in weight, with a 1 mm (0.04 in.) minimum

Fig. 12 Examples of near-net shape forged gears. (a) Spiral bevel gear with a 0.5 mm (0.02 in.) stock allowance developed for use
on gears with a DP less than 7. (b) Coarse-pitch (less than 5 DP) spur gear with a stock allowance of 1 to 2 mm (0.04 to
0.80 in.). Source: Presrite Corporation
Chapter 6: Casting, Forming, and Forging / 137

stock allowance per ank. A pitch range of 2. J.G. Bralla, Handbook of Product Design
less than 5 DP can be achieved. This type of for Manufacturing, Vol 16, McGraw-Hill
gear requires a nish process of hobbing, Book Company, 1986, p 355
hobbing and shaving, or hobbing and grind- 3. M. Blair and T.L. Stevens, Ed., Steel Cast-
ing, or skiving. ings Handbook, Sixth Edition, Steel
Spur gears with a 0.1 to 0.3 mm (0.004 to Founders Society and ASM International,
0.012 in.) stock allowance can be produced 1995, p 234
up to 250 mm (10 in.) in diameter and up to 4. D.W. Dudley, Gear-Manufacturing Meth-
150 mm (6 in.) maximum face width, with a ods, Handbook of Practical Gear Design,
0.1 to 0.3 mm (0.004 to 0.012 in.) stock McGraw-Hill Book Company, p 5.86
allowance per ank. A pitch range of 4 to 12 5. C. Cooper, Alternative Gear Manufactur-
DP can be achieved. This type of gear re- ing, Gear Technol., July/Aug 1998, p 916
quires a nish process of grinding or skiv- 6. D.P. Townsend, Ed., Gears Made by Dies,
ing. A net root is possible. Dudleys Gear Handbook: The Design,
Manufacture, and Application of Gears,
Figure 12 shows examples of precision
Second Edition, McGraw Hill Book Com-
forged near-net shape gears.
pany, 1992, p 17.117.21
7. D.J. Kuhlmann and P.S. Raghupathi, Man-
REFERENCES ufacturing of Forged and Extruded Gears,
Gear Technol., July/Aug 1990, p 3645
1. T.T. Krenzer and J.W. Coniglio, Gear 8. T. Russell and L. Danis, Precision Flow
Manufacture, Machining, Vol 16, ASM Forged Gears, Gear Manufacture and Per-
Handbook, ASM International, 1989, formance, American Society for Metals,
p 330355 1974, p 229239
Gear Materials, Properties, and Manufacture Copyright 2005 ASM International
J.R. Davis, editor, p139-153 All rights reserved.
DOI:10.1361/gmpm2005p139 www.asminternational.org


Powder Metallurgy

POWDER METALLURGY (P/M) is a exi- production, compaction, and sintering has

ble metalworking process for the production of resulted in new techniques such as inltration,
gears. The P/M process is capable of producing powder forging, surface rolling, and several n-
close tolerance gears with strengths to 1240 MPa ishing treatments. These developments have
(180 ksi) at economical prices in higher volume enabled the P/M parts to compete successfully
quantities. Spur, helical, bevel, face, spur- with wrought metal counterparts.
helical, and helical-helical gears are produced by Figure 1 summarizes the various P/M manu-
P/M techniques. The process is particularly facturing methods used in gear production. The
attractive when the gear contains depressions, conventional process consisting of compaction
through holes, varying levels, or projections. (pressing) and sintering is the most commonly
The rst application of P/M gears was for oil employed method for producing gears. Selec-
pumps in 1937/1938 (Ref 1). Gear pumps con- tion of a process method depends on factors
stitute a common type of application. Depend- such as dimensional accuracy. The American
ing on application stress levels, different mate- Gear Manufacturers Association (AGMA)
rials and density levels are used. For example, quality classes 5 to 7 can be obtained through
iron-carbon or iron-carbon-copper materials conventional press-and-sinter operations. When
with densities of 6.0 to 6.8 g/cm3 are used in AGMA quality levels above 7 are required, sec-
gear pumps for low-pressure applications such ondary operations such as coining, burnishing,
as engine lubrication and automatic transmis- and grinding typically are employed. See the
sions. For higher stress applications up to 20 section Gear Tolerances in this chapter for
MPa (3 ksi), alloy steels at the minimum density additional information.
of 7.1 g/cm3 are used. Powder metallurgy mate- Performance of any P/M part is inuenced
rial requirements for gear pump applications at primarily by density. Theoretical density of
various stress levels are given in Table 1. steel is around 7.85 g/cm3, and Table 2 summa-
With the considerable progress in P/M tech- rizes density levels obtained from various P/M
nology over the years, the quality and applica- processes for steel. All of these P/M methods
tion of P/M gears has increased. Progress in are employed for gear manufacture, including

Table 1 Material requirements for P/M pump gear applications

Application Designation(a) Density, g/cm3

Output pressure less than 0.69 MPa (100 psi) and light service FC-0208 type; as-sintered 5.86.2
Output pressure less than 6.9 MPa (1000 psi) but subject to high vibration and FC-0208 or FN-0106 6.46.8
heavy service; hardened by heat treating
Medium high pressure 6.98.6 MPa (10001250 psi) for short intervals; hardened FC-0208 type 6.26.5
by heat treating
General-purpose hydraulic pumps up to 10.3 MPa (1500 psi) continuous service FN-0106 type; heat treated 6.87.1
General-purpose hydraulic pumps up to 17.2 MPa (2500 psi) continuous service AISI 4630 type; heat treated 7.27.6
(a) FC-0208 is a copper inltrated steel. FN-0106 is a nickel steel. Source: Ref 1
140 / Gear Materials, Properties, and Manufacture

various surface treatment such as carbonitrid- ders also provides a variety of possibilities for the
ing, shot peening, and steam oxide treatments development of alloy compositions formulated
(Ref 2). for specic mechanical and physical properties.
A wide variety of base metals are available in
powder form, such as brass, bronze, iron, low-
alloy steels, and stainless steel. The selection is Capabilities and Limitations
governed by required mechanical properties,
physical properties, service conditions, and cost. Powder metal gears are used in various appli-
Table 3 lists commonly used P/M materials and cations for appliances, ofce machines, ma-
broad applications of standard grades. Cus- chinery, machine tools, and automotive prod-
tomized mixing and blending of elemental pow- ucts. The P/M process offers various advantages

Fig. 1 P/M gear production process. CIP, cold isostatic pressing; HIP, hot isostatic pressing

Table 2 Density range for P/M steel consolidation methods

P/M method Density range, g/cm3 Notes

Pressed-and-sintered steel 6.97.1 Conventional powder metallurgy

Double processing (press, presinter, restrike, full sinter, 7.27.4 Restrike (or repress) tooling has tolerances closer to
heat treat) nish dimensions than the tooling for the rst strike
Warm compaction (single pressed) 7.27.5 Specially formulated powders to obtain higher green
strength at lower compaction pressure
Warm compaction (double pressed) 7.47.7 ...
Cold densication (press, sinter, cold form, heat treat) 7.67.8 Cold forming improves fatigue and wear resistance
Powder forging (press, sinter, forge, heat treat) 7.67.8+ Limited to spur, bevel, and face gears although trials with
helical gears have been done (Ref 2)
Powder injection molding 7.8+ ...
Copper inltrated steel Near full density Copper inltration also improves machinability
Roll densication 7.67.8 Increases localized densication for wear and fatigue
Chapter 7: Powder Metallurgy / 141

over traditional gear manufacturing for various prole during P/M manufacture, eliminating
applications. subsequent machining operations. Thus, P/M
In traditional methods, gears are manufac- gear manufacturing allows efcient utilization
tured from blanks obtained by machining of of materials for moderate- to high-volume pro-
castings, forgings, and rolled bar stock. These duction. The minimum quantity required to
blanks are machined by various methods such make the P/M process economically competi-
as milling, hobbing, and shaping. For precision tive with machining depends on the size and
and high-speed applications, the gear teeth are complexity of the part and its accuracy and
nished by secondary operations such as shav- other property requirements. While P/M pro-
ing and grinding. Gear production by these cessing is best suited and most widely known
methods has the following disadvantages: for high-volume production quantities, there are
numerous applications where small quantities (a
Processing times are long, particularly with
thousand parts) can still offer cost benets over
difcult-to-machine alloys.
traditional manufacturing methods.
Material utilization is very low.
General process advantages of P/M gear
Machined gear teeth have an unfavorable
manufacture include:
grain ow pattern with ow lines intersect-
ing the gear tooth prole. Economy in mass production
In comparison to the traditional gear manu- Repeatability and uniformity of part features
facture process, the P/M process offers several and dimensions
advantages, particularly the elimination of ma- Production of multilevel gears
chining and scrap losses for the manufacture of Close control of density or, conversely,
various gear types such as helical, bevel (both porosity to suit a particular application
straight and spiral), rack, face, internal and Reduction or elimination of secondary oper-
external spur gears, and also compound gears. ations
Internal congurations (splines, keys, keyways) Improved surface nishes on gear teeth by
also can be formed simultaneously with the gear reducing or eliminating the machining
marks (or scoring) that can be imparted by
conventional methods
Self-lubricating ability from impregnation
of oils and lubricants
Noise reduction from the sound dampening
Table 3 Commonly used metal powder grades qualities of the pore structure of a P/M com-
ponent, high surface nish and tooth form
designation Chemistry Applications consistency also can have an effect on over-
F-0000 Pure Fe Low strength at high all reduction of noise in comparison to cut
density, soft magnetic gear sets of comparable quality
F-0005 Fe + 0.5% C
Moderate strength,
Weight reduction by incorporating weight
medium-carbon steel saving or lightening holes (or similar
FC-0208 Fe + 2% Cu + 0.8% C Higher-strength shapes) in the initial tool design
structural components
FN-0205 Fe + 2% Ni + 0.5% C High strength (heat Limitations of the P/M process for gear
treated), good impact
resistance manufacturing include (Ref 2):
FX-2008 Fe + 20% Cu + 0.8% C Copper-inltrated steel,
high strength,
Not economical in low- and medium-quan-
SS-316L 316 stainless steel Good toughness,
tity production
corrosion resistance Not normally suitable for production of worn
SS-410 410 stainless steel Good hardenability, gears, herringbone gears, and helical gears
abrasion resistance
CZ-2002 Brass Good toughness,
with helix angle exceeding 35 because of
elongation, corrosion various pressing and tooling considerations
CT-1000 Bronze
Structural and bearing
In view of the press capacities required for
applications compaction, the sizes of gears are limited
CNZ-1818 Nickel silver Decorative, tough,
ductile, corrosion
The gear thickness or face width is restricted
by the requirement that the depth of the die
cavity and the stroke of the press must be at
MPIF: Metal Powder Industries Federation
least 2 to 2 times the gear thickness
142 / Gear Materials, Properties, and Manufacture

Gear Forms also is particularly advantageous for gear designs

that have irregular curves, radial projections,
The size of P/M gears range from as small as keyways, or recesses, because these features are
a few millimeters to more than 300 mm in diam- easily obtainable without requiring secondary
eter. For all practical purposes the face width of machining operations. Full tooth face width for
P/M gears is limited to less than 75 mm (3 in.). compound gears and sufcient llet radii at the
The commonly used compaction presses have root diameter for tip clearance can be attained
insufcient press stroke to consolidate the pow- without allowances for cutter clearance.
der to greater than this length (face width). Spur gears are the most common type of
The following types of gears can be produced P/M gear in use today. The P/M process is capa-
by the P/M process: ble of producing spur gears of many tooth forms
and modications. If an electrodischarge ma-
Spur gears chining (EDM) electrode can be ground or cut,
Bevel gears the mating form can be duplicated in P/M dies.
Face gears Compared to wrought metal gears, the face
Hypoid gears width or thickness of a spur gear is restricted in
Helical gears of helix angle not exceeding powder metallurgy in order that the density and
35 mechanical properties are maintained fairly uni-
Combination gears or multiple gears in which form throughout the section. Fine pitch (module
two or more gear forms are combined into one <0.75 mm) spur gear teeth present powder ll-
component are a specialty of the P/M process. ing problems, which are more serious if the face
Some examples of P/M gear design are shown width is large.
in Fig. 2 to 5. True involute forms (Fig. 3a) that Helical gears are also commonly made by
are expensive to make by other methods can be the P/M process. However, in contrast to metal-
easily made by the P/M process. The P/M process cutting processes, the helix angle of P/M gears

Fig. 2 Examples of powder metallurgy gear forms. Courtesy of Carbon City Products
Chapter 7: Powder Metallurgy / 143

is limited to approximately 35. As with spur teeth in the offset position so that a tooth gap
gears, many tooth forms and modications are comes below a tooth crown; this allows full
possible. punch pressure to be exerted on the tooth
Helical gears have not been produced by crowns of the functional gear, enabling a better
powder forging, but recent production tests density distribution (Ref 3).
have been performed on a unique tooling Combination Gears. Powder metallurgy
method for powder forging of helical gears. One gear manufacture allows the production of com-
production trial involved the powder forging of bination gears or multilevel (two-to-four level)
a helical gear with a 31 helix angle and a nor- gears resulting in a single part with two or more
mal pressure angle of 15 for an automotive gear forms. All the gear forms can be positioned
transmission part (Ref 4). Experimental work close to one another, thus making the part very
later carried out by the German Research Soci- compact. This helps in saving space and han-
ety showed the superiority of the nished prod- dling and assembly costs. Combination gears
uct in terms of toughness, tooth root fatigue, and can be produced by either of the two methods:
contact fatigue (Ref 5). During powder forging
of a helical gear, the die is the only helical part Compacting the part in the combined form
of the tool. Compacting separately, assembling them in
Bevel and face gears are readily made by green state and sintering
the P/M process, but differ somewhat from spur
and helical gears in mechanical property exi-
bility. Due to the nature of the compaction Gear Tolerances
process, sections such as face gear teeth cannot
be made to high densities without resorting to Tolerances of P/M gears are determined by
copper inltration or repressing. Most bevel and the compaction tooling, compaction process,
face gear forms can be made providing they are sintering process, and the P/M alloy. Tooth
not undercut, but the maximum mechanical form and dimensions remain relatively stable
properties attainable will be somewhat less than throughout the process. The density variation in
those that can be achieved for spur and helical the compact should not exceed 0.2 g/cm3, other-
gears. wise, dimensional control will be difcult dur-
All face tooth forms must be such that the ing sintering.
green compact does not lock into the punch. For Powder metallurgy gears are produced typi-
this reason undercut gear forms cannot be pro- cally in AGMA 5 to 7 quality levels. Higher
duced. Under the action of the punch, the crown AGMA classication is possible, depending on
area of face gear teeth also normally receives the size of the compacted gear and the additional
less compaction than the root zone. This results secondary operations. Higher AGMA classica-
in an undesirable density variation, which can tion increases the cost of manufacturing due to
be offset by copper inltration or repressing. the secondary (machining) or restrike operations
Another method to offset density variations is that are added to meet the tighter tolerances. In
by putting a nonfunctional gear form with its general, traditional press-and-sinter operations

Fig. 3 Features produced in powder metallurgy gears. (a) True involute forms. (b) Gear teeth straight up to a ange in P/M gear.
Source: Ref 3
144 / Gear Materials, Properties, and Manufacture

are economical for achieving AGMA quality ods, and gear material. Because most P/M gears
levels 5 to 7. It is not advisable to produce higher are pressed in carbide dies, the tooth form and
quality levels through only compaction and sin- dimensions remain relatively stable and con-
tering, as closer tolerances and tool wear will stant throughout the process. Concentricity of
impact costs. Where higher accuracy is required, the gear outside diameter and pitch diameter to
secondary operations such as sizing, shaving, the bore are also controlled by the tooling. If the
burnishing, and grinding are required. Sizing pressing core rod or punches are eccentric to the
(repressing) is the most common method die, the parts will be eccentric. Once the part is
because it improves mechanical properties pressed, the material characteristics and sinter-
through cold working. ing process determine the dimensional tolerance
Because dimensional variation is a function capability of the outside diameter, pitch diame-
of part geometry, material, density, and subse- ter, test radius, and bore diameter.
quent thermal operations, dimensional toler- Because of the number of variables within the
ances may vary by application. P/M process it is not possible to state absolute
Powder metallurgy gear tolerances are deter- dimensional tolerance capabilities. However
mined by the compaction tooling, compaction minimum tolerances that one should expect for
process, sintering techniques, and the alloy. The ferrous-base P/M gears without secondary oper-
tolerances can be divided into functional toler- ations are listed in Table 4 for spur and helical
ances that directly affect gear function and non- gears and in Table 5 for bevel gears. Typical tol-
functional elements that pertain to portions of erances for single press and sinter are listed in
the gear that are more properly termed general Table 6.
machine design considerations. Nonfunctional
elements include such items as hub design
dimensions, fastening designs such as keyways, Gear Design and Tooling
set screw holes, inside diameters for press ts,
and other special body congurations. Properly designed gears and compaction
Functional gear tolerances are controlled by tooling are the key factors in the economical
the compaction tooling, the processing meth- production of P/M gears with consistent quality.

Table 4 Minimum tolerance capabilities for

spur and helical P/M gears
Tooth-to-tooth composite AGMA class 6
Total composite tolerance AGMA class 6
Test radius 0.05 mm + 0.002 diameter
Over-pin measurement 0.10 mm + 0.002 diameter
Lead error 0.001 mm/mm
Perpendicularityface to bore 0.05 mm + 0.001 diameter
Parallelism 0.025 mm + 0.001 diameter
Outside diameter 0.10 mm + 0.002 diameter
Fig. 4 Keyways and integral keys. Source: Ref 3 Prole tolerance 0.008 mm
Tolerances with single press and sinter without secondary operations. Source:
Ref 3

Table 5 Minimum tolerance capabilities for

bevel P/M gears
Tooth-to-tooth composite AGMA class 6
Total composite tolerance AGMA class 6
Outside diameter 0.10 mm + 0.002 diameter
Perpendicularityface to bore 0.05 mm + 0.001 diameter
Tolerances with single press and sinter without secondary operations. Source:
Ref 3
Fig. 5 Weight reducing holes on P/M gear. Source: Ref 3
Chapter 7: Powder Metallurgy / 145

Blended metal powder is usually compacted at completion of the compaction cycle, the upper
room temperature, at pressures between 275 to punch retracts from the die, and the lower punch
690 MPa (40 to 100 ksi) of projected surface initiates an upward motion, ejecting the preform
area. The tooling for a single-level gear includes from the dies.
a die, an upper punch, and a lower punch. If the The shape of the compacted preform is deter-
particular component to be formed requires a mined by the shape of the tooling and the axial
bore or other inside-diameter conguration, a motion of the compaction press. Two main fac-
core rod is also part of the tooling. Multilevel tors inuence part design: the ow behavior and
compound gears or other complex structural characteristics of the metal powders and the
parts may have two or more upper and/or lower movement of the tools within the pressing
punches (Fig. 6). cycle. Metal powders do not ow hydraulically,
To form the gear, the die is lled with powder and the allowance for friction between the pow-
at a ratio of approximately two times the parts der particles themselves and with the moving
thickness; for example, a spur gear that has a tool members must be factored in the nal P/M
thickness of 19 mm (0.75 in.) will be compacted part design. The pressing action from both top
to form a column of 38 mm (1.50 in.) thick. and bottom largely governs the shape, length,
During the compaction cycle of the more com- and dimensional details of the preform. The
monly used type of press, the upper punch clearance tolerance of the moving tool mem-
enters the die while the lower punch, which in bers, relative to the inside diameter or bore, con-
the ll position seals off the bottom of the die, tributes to a larger degree of eccentricity than
remains stationary. As the upper punch travels with machined gears. Details, or features that
downward, the compressive forces cause the die increase the number of tool members compound
assembly to move downward in relationship to the eccentricity. Concentricity can be improved
the lower punch, resulting in the same effect as with the addition of secondary operations, such
if the lower punch were moving upward during as grinding the bore in relationship to the pitch
the compaction stroke. The rate of movement diameter.
and pressure of the upper punch and the motion The compacted preform is sintered typically
of the die are relatively equal to ensure uniform at temperatures of approximately 1120 C
density within the compacted preform. After (2050 F) to obtain tensile properties around
125 to 620 MPa (18 to 90 ksi), depending on the
material composition. The addition of heat
treatment can increase ultimate tensile proper-
ties of a compacted-and-sintered steel compo-
nent to 1240 MPa (180 ksi).
Table 6 Typical tolerances of single pressed
and sintered gears
See Tables 4 and 5 for specic gear forms.
Tolerance (without
secondary operations)
Feature mm in.

Length 0.125 0.005

Flatness on end 0.05 0.002
Concentricity 0.075 0.003(a)
Inside diameter 0.050 0.002(a)
Prole tolerance 0.0075 0.0003
Tooth-to-tooth error 0.025 0.001
Circular tooth error 0.05 0.002
Parallelism on ends 0.04 0.0015
Lead error 0.025 0.001
Total composite error 0.08 0.0032(b)
Outside diameter tolerance
with pitch diameter up to:
25 mm (1 in.) 0.075 0.003(c)
50 mm (2 in.) 0.1 0.004(c)
75 mm (3 in.) 0.125 0.005(c)
100 mm (4 in.) 0.15 0.006(c)
(a) Tolerance depends on feature size. (b) Tolerance depends on gear size
and number of levels. (c) Total tolerance on outside diameter and meas-
urement over wires
Fig. 6 Compaction tooling for compound gear. Source: Ref 6
146 / Gear Materials, Properties, and Manufacture

Design Guidelines nations, which are well within the capability of

the P/M process. Uniform density and high
Design guidelines for P/M gears include the strength is best achieved by limiting the number
following considerations: of section thickness changes (levels) that are
Ejection from the die designed into the preform. The number of levels
Die lling during compaction a part may have is determined by the specic
Clearance type of compaction press and/or the design of
Thickness changes the compaction tooling.
Right-angle intersections Right-Angle Intersections. The congura-
Edge detail tion of the gear should minimize right-angle
intersections. Radii should be incorporated at
Ejection from the Die. Some features must right-angle intersection of section thickness
be eliminated from the nal part (or added changes. These radii improve the integrity of the
through secondary machining operations) when preform. Gear engagement should be designed
they inhibit ejection of the preform from the die. to occur above the radius. Often raised surface
For example, undercuts, reverse angles, details features can be added to the compacted preform
at right angles to the direction of pressing (e.g., to assist in this engagement.
holes, grooves), threads, diamond knurls and Edge Detail. The conguration of the gear
reentrant angles would interfere with smooth should incorporate edge detail (i.e., chamfers on
ejection. top and bottom of the gear teeth and on outside
Die Filling During Compaction. Gear de- and inside diameters) (Fig. 9). This edge detail
sign should allow for the movement of the metal of chamfer has two main purposes. First, it
powders throughout the tool members during increases the density of the teeth. Higher tooth
the compaction cycle. Metal powders do not density results in improved mechanical proper-
ow hydraulically. Therefore, extremely thin- ties, particularly strength. It also reduces the
walled sections, very narrow grooves, and deep adverse effects of burrs. The chamfer detail will
counter-bores should be avoided because the in most cases keep the burr within the overall
metal powders may not completely ll these
details within the tool members.
Tool life can be improved by avoiding nar-
row deep grooves, very sharp edges, complete
spherical proles, and knife-thin tool thick-
nesses. Often design simplication of these fea-
tures allows for a more robust or practical set
of compaction tools without adding machining
Clearance. Hubs or bosses for gears,
sprockets, or cams can be readily produced, but
the designs should ensure the maximum permis-
sible material between the outside diameter of
the hub and the root diameter of gear or sprocket Fig. 7 Hub feature. Source: Ref 6
features (Fig. 7) (Ref 6). Another consideration
for gear design is to maintain sufcient clear-
ance between the inside diameter and the root
diameter, which can range from 0.9 mm (0.035
in.) on small pinions to 7.5 mm (0.30 in.) for
more demanding applications. Design consider-
ation should be given to maximize the clearance
between the gear root diameter and compound
gears, pinions, or features (i.e., cams, etc.) that
are incorporated parallel to the gear teeth.
Thickness Changes. The conguration of
the gear should limit major changes in section
thickness (Fig. 8) for compound gear geome- Fig. 8 Preferred design for section thickness changes. Source:
tries such as gear-pinion and spur-face combi- Ref 6
Chapter 7: Powder Metallurgy / 147

thickness (width) of the gear. These burrs are a requirements, the environment, the type of
result of the t clearance of the tool members lubrication, and the gear and assembly preci-
and, if necessary, can be removed by a vibratory sion. Allowable loads on contact surfaces of
nishing (tumbling) operation. gears are limited by the occurrence of pitting, in
most cases, provided they are not run at very
high speeds to cause scoring (Fig. 10). Even at
Gear Performance light loads, gear teeth experience relatively high
concentrated stresses because contact areas are
In general terms, gear teeth loading is akin to very small. In general, Hertz equations are used
normal loading of two parallel cylinders with a to design for contact stresses, and these equa-
combined sliding and rolling. In rolling contact, tions consider compressive loads on stationary
two elements roll in contact with each other contact surfaces.
without slippage. Both elements rotate at the Spalling damage of wrought and P/M alloys
same velocity. Gear teeth have sliding contact has similar appearance but different crack prop-
over most of their surfaces, with rolling contact agation mechanisms. In wrought alloys, most of
occurring only at the pitch line. The amount of the time is spent in the crack initiation phase
sliding varies with the distance from the pitch (under high-cycle conditions) with relatively
line. When the rolling and sliding are in the quick crack propagation once the crack is big
same direction, less damage occurs than with enough. In P/M alloys, the porosity blunts crack
negative slide (Ref 7). growth to the surface. Until the crack can prop-
Gears can be damaged by several mecha- agate to the surface, a band of subsurface crack-
nisms such as bending fatigue or by surface ing continues. The depth of the band is probably
fatigue mechanisms such as pitting or spalling. related to the depth of the maximum shear
The origin of pitting occurs where the combina- stress.
tion of pressure and negative sliding are the Selecting Gear Materials
highest. On spur gears, this is usually at the low-
est point of single-tooth contact on the deden- Required strength is the rst factor when
dum of the drive gear (Ref 8). selecting a gear material. In spite of the approx-
As with any other material the choice of a imations involved, one can generally arrive at
P/M alloy is governed by the magnitude and the correct material by using the methods out-
nature of the transmitted load, the speed, the life lined in the AGMA standards for rating the
strength of spur, helical, and straight bevel gear
teeth (Ref 911). A good approximation of
compressive load carrying ability can be found
by using the methods outlined in the AGMA
standards for surface durability of spur, helical,
and bevel gear teeth (Ref 1214). In addition,
the allowable contact stress Sac, the elastic mod-
ulus, and Poissons ratio of the P/M material
must be determined.

Fig. 10 Schematic of gear damage mechanisms. Source:

Fig. 9 Edge radii design features. Source: Ref 6 Ref 15
148 / Gear Materials, Properties, and Manufacture

Allowable Contact Stress. Tensile strength, Outside diameter (as a reference dimension
elastic modulus, and Poissons ratio for most fer- for straight bevel gears)
rous P/M materials is given in MPIF standard 35. Root diameter
In the absence of data for allowable contact Pitch apex to back face
stress of ferrous P/M materials, an approxima- Total composite tolerance
tion of Sac is 70 MPa (10 ksi) below the ultimate Tooth-to-tooth tolerance (not appropriate
tensile strength (Sac = UTS = 10 ksi). for straight bevel gears)
Allowable bending stress (Sab) for a P/M Test radius
material can be approximated in the absence of
data at 0.3 of the ultimate tensile strength of fer- Measurement over wires (MOW) and mea-
rous-base P/M materials (Sab = 0.3 UTS). The surement between wires (MBW) (for internal
only caution one should be aware of is that ten- gears) are convenient measurable criteria that
sile data may be typical properties. The designer should also be included as drawing specica-
should consult the supplier or specications for tions. Master gear information (specication
minimum property data. and/or data) should be included on the drawing.
Impact and Fatigue Strength. Most P/M Tip radius detail for spur gears is also necessary.
gears are limited in performance by their impact It is advisable that right-angle (rolling) xtures
and fatigue strength as opposed to their com- be developed to check mesh and backlash.
pressive yield strength and surface durability. Often several ID (inside diameter) pins are
Surface treatments such as shot peening, mate- required to conrm mesh, after/during critical
rial cleanliness, and other factors inuence con- processing operations (minimally, compaction,
tact fatigue. Detailed information on P/M gear sintering, and heat treatment). Typically, these
fatigue testing and evaluation is discussed fur- inspection xtures are specied and purchase at
ther in Ref 15 and 16. Additional information on the time the tooling is manufactured.
fatigue is also contained in Chapter 13, Gear Gear drawing specications are outlined in
Failure Modes and Analysis and Chapter 14, ANSI/AGMA 6008-A98, Specications for
Fatigue and Life Prediction. Powder Metallurgy Gears. This standard de-
nes the minimum detailed information to be
included in the P/M gear specications submit-
ted by the gear purchaser to the gear producer.
Quality Control and Inspection In addition to gear drawing specications, this
information also covers gear tooth geometry
Inspection is one of the most important stages data and gear material specications.
in gear production. Inspection and testing is In production, it is common practice to check
done on dimensional specications, mechanical the face width, outside diameter, over-pin meas-
properties, and surface durability (hardness). urement, total composite error, tooth-to-tooth
The information required to describe a gear for error, perpendicularity of face-to-bore, paral-
the P/M process and subsequent inspection lelism, and inside diameter of the gears. Also
methods are essentially the same as for a the tooth strength and macroscopic hardness are
machine-cut gear. The part drawing should usually checked. The sampling plan and fre-
include: quency of measurement varies and depends on
the criticality of the application.
Number of teeth Gear inspection terms include:
Diametral pitch (normal diametral pitch for
a helical gear) Runout (radial): the total variation of the
Pressure angle (normal pressure angle for a radial distance of the gear teeth from the
helical gear) center or bore
Tooth form for spur, helical, and straight Runout (axial): the total variation of gear
bevel gears teeth along its axis, measured from a refer-
Helix angle and rotation (right or left hand) ence plane perpendicular to its axis. Also
for a helical gear referred to as wobble or face runout
Pitch, face, root, and back angles for straight Lead: the axial advance of a gear tooth in
bevel gears one revolution (360)
Lead variation: the condition in a gear
To assist in the qualication (i.e., inspection) where some teeth vary in lead, plus or minus
the drawing should also include: from the average lead
Chapter 7: Powder Metallurgy / 149

Total composite error: the total variation of mum acceptance criteria ensures 99.85% con-
the center distance in one complete revolu- dence of adequate strength.
tion Surface durability is a specied and mea-
Tooth-to-tooth variation: the variation of the sured minimum Rockwell hardness because this
center distance in a section of its circumfer- macroscopic hardness relates to the compres-
ence (the section of the circumference is an sive yield strength of the materials. The mini-
angle of 360/N where N is the number of mum Rockwell hardness depends on the mate-
gear teeth) rial and can be established either from samples
or from the suppliers historical information.
Mechanical Testing. Mechanical proper- In cases where scoring may be a problem, in
ties are also measured, and in many cases bench particular with heat treated gears, it is appropri-
life and impact tests or eld tests are made to ate to specify and measure a microscopic hard-
check the gear functionality. Because tensile ness. This is accomplished with a diamond
strength of a gear cannot be measured directly, it indenter, and the microscopic or individual par-
is appropriate to establish a minimum tooth ticle hardness of the material is measured. The
breakage strength that can be used as a produc- particle hardness specication is generally
tion acceptance criteria from either sample gears determined from historical data for the pro-
made from a die or gears cut from a P/M slug. ducers material.
One quick and accurate means of measuring In contrast to most metals, P/M materials
tooth strength of a spur gear is shown in Fig. 11. have both a macro- and microhardness as a
By mounting the gear between xed supports result of porosity within the structure. Because
and applying a vertical load, one can measure a the P/M structure consists of individual hard
tooth breakage strength. By testing a number of particles bonded together with interdispersed
gears that perform successfully in the particular voids, the conventional Rockwell indention reg-
application, it is possible to calculate statistical isters a composite hardness. Because sliding
limits for the tooth breakage strength that can be and surface contact involves microscopic con-
used as a future material strength acceptance. tact (individual asperities) the particle or micro-
Helical and bevel gears are usually evaluated in hardness of the P/M material inuences the
torque-type tests and the same type of statistical score resistance of the P/M material more so
analysis applied. The use of statistics is particu- than the macro Rockwell hardness. As a gener-
larly expedient in gear strength testing because ality, when a P/M gear is run against a wrought
proper sampling allows process control to a nor- gear, the P/M particle hardness should be no
mal statistical distribution (i.e., a 3) mini- harder than the hardness of the wrought gear to
prevent wear of the mating gear.

Examples of P/M Gears

Powder metallurgy gears are used in a wide
variety of application areas including business
machines, appliances, farm, lawn, and garden
equipment, automobiles, trucks, and military
vehicles. The examples that follow illustrate
the versatility of the P/M process for gear manu-
Example 1: Drive Gear for Mailhandling
Machine. A at-belt drive gear weighing 490 g
(17.3 oz), which is compacted to a density of
7.45 g/cm3, is shown in Fig. 12. This type of gear
is used in high-speed mailhandling and inserting
machines. This part, consisting of electrolytic
iron preblended with 0.35% C, replaced a two-
piece gear assembly made of American Iron and
Steel Institute (AISI) 1030 steel, thus resulting in
Fig. 11 Spur gear tooth strength test. Source: Ref 17 a total cost savings of 66%.
150 / Gear Materials, Properties, and Manufacture

The powder was rst compacted on a 5340 and subsequently preblended with 0.75% Cu
kN (600 ton) press to a density of 6.8 to 6.9 and 0.35 to 0.45% C. These gears are used in
g/cm3 at a compaction pressure of 414 MPa (30 postage meters for counting applications.
tsi). It was then presintered and repressed to a The powder was compacted in a oating die
density of 7.45 g/cm3 at 965 MPa (70 tsi). After press system to a density of 6.85 to 7.0 g/cm3.
repressing, the part was sintered at 1120 C Sintering was conducted in an endothermic
(2050 F) for 30 min. atmosphere at 1120 C (2050 F). After sinter-
As shown in Fig. 12, considerable machining ing, the parts were carbonitrided to provide a
(turning) of the hub was required to generate the case hardness of 700 DPH for a 0.127 mm
nal shape. In use, this gear rotates and a mag- (0.005 in.) case depth and a core hardness of
netic clutch rides on the machined hub. When 400 DPH. After case hardening, the parts were
the operator trips the cycle control on the deburred and oil impregnated.
machine, the clutch closes around the small hub The original method for manufacturing these
diameter, and the load is transferred to the P/M gears consisted of a two-piece copper-brazed and
gear. Pressing of a large hub ensures a uniform gear-hobbed assembly of AISI 1030 steel. The
density in the cross-sectional area, but requires larger (bottom) gear was rst turned as a disk
additional machining of the P/M blank. with a center hole. Gear teeth were then hobbed.
After machining, the hub area is carboni- The smaller (top) gear was also turned and
trided to a case depth of 0.127 mm (0.005 in.). gear hobbed. The two parts were then press t-
Case hardness is 700 DPH (minimum), while ted and copper brazed. At this point, inspection
core hardness is 550 DPH. To maintain a 0.025 was required to reject and scrap any gears that
mm (0.001 in.) diameter tolerance between the shifted during brazing, thus losing the critical
hub and the magnetic clutch, the hub diameter is tooth-to-tooth relationship required between the
ground to nal dimensions. Typical properties small-diameter gear and the large-diameter
for this P/M part are: gear. The rejection rate due to assembly errors
ran as high as 50%.
Tensile strength, MPa (ksi) 414586 (6085) Figure 13(a) shows the single-piece transfer
Yield strength, MPa (ksi) 365414 (5360) gear manufactured using P/M techniques. The
Elongation, % 12
Transverse-rupture strength, MPa (ksi) 930 (135)
ange section was made thick enough so that
the hub section could be machined as a second-
Example 2: Transfer Gears for Postage ary operation. The normal timing of the oating
Meter Counting Applications. The high- die press mechanism forced powder to transfer
precision, high-strength gears shown in Fig. 13 from the small gear teeth into the ange section,
are made of 4600 alloy powder prealloyed with thus resulting in low density in the small (top)
0.25% Mn, 1.7 to 1.9% Ni, and 0.5 to 0.6% Mo gear teeth. Consequently, tooling was modied
to produce the gear congurations shown in Fig.
13(b), which required no machining of the gear

Fig. 12 Iron-carbon P/M drive gear. (Left) Finished ma- Fig. 13 P/M transfer gear made of high-strength low-alloy
chined, hardened, ground part. The hub diameter is steel. (a) Original P/M processing technique, which
machined oversized (+0.254 mm, or +0.010 in.) to allow for required machining of ange section. (b) Modied P/M tech-
nal grinding. (Right) P/M blank as-pressed nique, which required no additional machining
Chapter 7: Powder Metallurgy / 151

teeth located on the hub section. Typical n- in a steel medium to facilitate sizing. Prior to
ished P/M properties are: sizing, the part is dipped in soluble sizing wax,
which also acts as a long-term storage and rust
Ultimate tensile strength 586793 MPa (85115 ksi) protection medium.
Yield strength 552655 MPa (8095 ksi) Sizing enhances the tooth form and pitch
Elongation <1%
Transverse-rupture strength 11721379 MPa (170200 ksi) diameter tolerances. After sizing, the center
hole is bored for concentricity, and the outside
of the hub is machined to correct any ovality.
Example 3: Eccentric Gear. Figure 14 This part replaced a class 30 gray iron casting
shows an eccentric gear for a washing machine that was machined from a rough blank.
that is made of a blend of 2% copper, 0.5% Example 4: Fourth Reduction Gears. Lawn
graphite, and the balance iron, which is com- and garden tractors require gears that can with-
pacted, sintered, sized, and machined. An inter- stand extremely high wear and heavy loading.
mediate level of graphite was chosen to produce For example, the fourth reduction gear shown in
an as-sintered part that could be sized readily to Fig. 15 is a drive gear in a six-speed transmis-
ensure an accurate gear tooth prole. Lower car- sion. A crush load of 4000 kg (8800 lb) per tooth
bon content also improves machinability. Den- is specied. High tooth density is required of the
sity of this part is 6.5 g/cm3. P/M part. To provide this feature, repressing the
This gear is made from atomized powder to part to a density of 7.3 to 7.5 g/cm3 would seem
obtain optimum compressibility. Because this to be the solution. However, the part is too large
gear is a two-level part, green strength is critical to be compacted on a 4.45 MN (500 ton) press.
to eliminate cracking at juncture. Press motion To overcome this limitation, the part is com-
and tool design also are critical in preventing pacted to three densities: 6.4 to 6.6 g/cm3 in the
hub cracking. hub, 6.6 to 6.8 g/cm3 in the inner ange, and 6.9
Sintering is done in an endothermic atmo- to 7.0 g/cm3 in the tooth area.
sphere for 30 min at 1105 C (2025 F). Dew After presintering, the gear is repressed only
point is monitored to prevent carburization of in the tooth area to a density of 7.3 to 7.5 g/cm3.
the surface layers so as not to affect sizing or Carbon content is kept low to improve com-
machining. After sintering, the part is deburred pressibility during repressing, after which the
teeth are carburized.

MPIF material FN-0205

Carbon, % 0.5
Nickel, % 2
Molybdenum, % 0.5
Density, g/cm3
Tooth 7.37.5
Inner ange 6.66.8
Hub 6.46.6
Fig. 14 Eccentric gear for a washing machine made of iron
Tooth apparent hardness, HRC 5056
alloy powder containing 2% copper and 0.5%
graphite. Center hole inside diameter: approximately 19 mm Tooth crush strength, kN (ton) 40 (4.5)
(0.76 in.); eccentric outside diameter: approximately 57 mm Finished weight, kg (lb) 1.042 (2.300)
(2.28 in.); total thickness: approximately 28.50 mm (1.14 in.);
eccentric thickness: approximately 15.90 mm (0.63 in.); pitch
diameter: approximately 115 mm (4.59 in.) Fig. 15 Fourth reduction gear
152 / Gear Materials, Properties, and Manufacture

Fig. 16 Comparison of material used for a conventionally forged reverse idler gear (top) and the equivalent powder forged part (bot-
tom). Material yield in conventional forging is 31%; that for powder forging is 86%. 1 lb = 453.6 g.

Example 5: Powder Forged Internal Ring

Gear. Powder forging is a process in which un-
sintered, presintered, or sintered powder metal
preforms are hot formed in conned dies. One
of the main economic benets of powder forg-
ing is the reduced amount of machining
required, as illustrated in Fig. 16.
The powder forged internal ring gear shown
in Fig. 17 is used in automatic transmissions for
trucks with a maximum gross vehicle weight of
22,700 kg (50,000 lb). The gear transmits 1355
N m (1000 ft lbf) of torque through the gear
and spline teeth.
Originally, the gear was produced by forging
an AISI 5140M tubing blank. The convention-
ally forged blank required rough machining,
gear tooth shaping, spline machining, core heat
treating, carburizing, and deburring. The only
secondary operations required on the powder Fig. 17 Powder forged internal ring gear used in automatic
forged part are surface grinding, hard turning, transmission for trucks of up to 22,700 kg (50,000 lb)
gross vehicle weight. Courtesy of Precision Forged Products
shot blasting, and vibratory tumbling. Division, Borg Warner Corp.
The powder forged 4618 ring gear is pro-
duced to a minimum density of 7.82 g/cm3. The
part is selectively carburized using a proprietary
process and quench hardened. Minimum sur- ness is 25 HRC (825 MPa, or 120 ksi, ultimate
face hardness is 57 HRC (2070 MPa, or 300 ksi, tensile strength). The internal gear teeth are pro-
ultimate tensile strength), while the core hard- duced to AGMA Class 7 tolerances.
Chapter 7: Powder Metallurgy / 153

ACKNOWLEDGMENTS 11. AGMA Standard for Rating the Strength

of Straight Bevel and Zero Bevel Gear
This chapter was adapted from: Teeth, AGMA 222.02, American Gear
Manufacturers Association, 1964
G.T. Shturtz, Powder Metallurgy Gears,
12. AGMA Standard for Surface Durability of
Powder Metal Technologies and Applica-
Spur Gear Teeth, AGMA 210.02, Ameri-
tions, Vol 7, ASM Handbook, ASM Interna-
can Gear Manufacturers Association, 1965
tional, 1998, p 10581064
13. AGMA Standard for Surface Durability of
Appendix 3: Examples of Powder Metal-
Helical and Herringbone Gear Teeth,
lurgy Parts, Powder Metal Technologies and
AGMA 211.02, American Gear Manufac-
Applications, Vol 7, ASM Handbook, ASM
turers Association, 1969
International, 1998, p 11011108
14. AGMA Standard for Surface Durability
W.B. James, M.J. McDermott, and R.A.
Formulas for Straight Bevel and Zero Bevel
Powell, Powder Forged Steel, Powder Metal
Gear Teeth, AGMA 212.02, American
Technologies and Applications, Vol 7, ASM
Gear Manufacturers Association, 1964
Handbook, ASM International, 1998, p
15. T. Prucher and H. Sanderow, Surface
Fatigue of P/M Alloys, Characterization,
Testing, and Quality Control, Vol 2,
Advances in Powder Metallurgy and Par-
REFERENCES ticulate Materials, Metal Powder Indus-
tries Federation, 1994, p 99111
1. P.W. Lee, Press and Sintered Parts and 16. R. Gnanamoorthy and K. Gopinath, Sur-
Their Applications, Powder Metallurgy face Durability Studies on As-Sintered
Applications Advantages and Limitations, Low Alloy Steel Gears, Advances in Pow-
Erhard Klar, Ed., American Society for der Metallurgy and Particulate Materi-
Metals, 1983, p 82 als1992, Vol 6, Metal Powder Industries
2. S.D.K. Saheb and K. Gopinath, Powder Federation, 1992, p 237250
Metallurgy GearsA Brief Review, Pow- 17. W. Smith, Ferrous Based Powder Metal-
der Metall. Sci. Technol., Vol 1 (No. 4), lurgy Gears, Gear Manufacture and Per-
July 1990, p 4366 formance, American Society for Metals,
3. S.D.K. Saheb and K. Gopinath, Tooling for 1974, p 257269
Powder Metallurgy Gears, Powder Metall.
Sci. Technol., Vol 2 (No. 3), April 1991, p
4. Met. Powder Rep., Vol 43, 1988, p 536
5. W. Konig, G. Rober, K. Vossen, and M. AGMA & MPIF Develop Standards, Infor-
Stromgren, Powder Forging of Helical mation Sheet for Powder Metal Gears, Gear
Gears for Car Manual Gear Boxes Technol., J. Gear Manuf., Sept/Oct 1996
Concept and Properties, Met. Powder Rep., R. Moderow, Gear Inspection and Measure-
Vol 45 (No. 4), 1990, p 269273 ment, Gear Technol., J. Gear Manuf.,
6. Powder Metallurgy Design Manual, Metal July/Aug 1992
Powder Industries Federation, 1994, p 193 MPIF Standard 35, Material Standards for
7. G. Parrish and G.S. Harper, Production P/M Structural Parts, Metal Powder Indus-
Gas Carburizing, Pergamon, 1985, p 20 tries Federation, 1997
8. Gear Design Manufacturing and Inspec- Powder Metallurgy Design Manual, 2nd
tion Manual, SAE No. 841083, Society of ed., Metal Powder Industries Federation,
Automotive Engineers, 1990, p 46 1995
9. AGMA Standard for Rating the Strength Powder Metallurgy Design Solutions, Metal
of Spur Gear Teeth, AGMA 220.02, Powder Industries Federation, 1993
American Gear Manufacturers Associa- H. Sanderow, Powder Metal Gear Design
tion, 1966 and Inspection, Gear Technol., J. Gear
10. AGMA Standard for Rating the Strength Manuf., Sept/Oct 1996
of Helical and Herringbone Gear Teeth, G. Shturtz, The Beginners Guide to Powder
AGMA 221.02, American Gear Manufac- Metal Gears, Gear Technol., J. Gear
turers Association, 1965 Manuf., Sept/Oct 1995
Gear Materials, Properties, and Manufacture Copyright 2005 ASM International
J.R. Davis, editor, p155-162 All rights reserved.
DOI:10.1361/gmpm2005p155 www.asminternational.org


Through Hardening

THE THROUGH-HARDENING PROCESS Four different methods of heat treatment are

is generally used for gears that do not require primarily used for through-hardened gears.
high surface hardness. Typical gear tooth hard- In ascending order of achievable hardness,
ness following through hardening ranges from these methods are annealing, normalizing and
32 to 48 HRC. Most steels used for through- annealing, normalizing and tempering, and
hardened gears have medium carbon content quenching and tempering. Sometimes, hard-
(0.3 to 0.6%) and a relatively low alloy content nesses of through-hardened gears are specied
(up to 3%). The purpose of alloying is to and measured in other scales besides Rockwell,
increase hardenability. The higher the harden- such as Vickers and Brinell. Table 1 shows an
ability, the deeper is the through hardening of approximate relationship among the various
gear teeth. Since strength increases directly commonly used hardness scales.
with hardness, high hardenability is essential
for through hardening steels. High hardenabil-
ity, again, has some adverse effect on material Through-Hardening Processes
ductility and impact resistance. The other draw-
back of through-hardened gears is lower allow- Annealing refers to any heating and cooling
able contact stresses than those of surface-hard- operation that is usually applied to induce soft-
ened gears. This tends to increase the size of ening. There are two types of annealingfull
through-hardened gears for the same torque and process. In full annealing, the steel is heated
capacity compared with those with hardened usually to approximately 38 C (100 F) above
surfaces. the upper critical temperature and held for the
In through hardening, gears are rst heated to desired length of time, followed by very slow
a required temperature and then cooled either in cooling in the furnace. The purposes of full
the furnace or quenched in air, gas, or liquid. annealing are to:
The process may be used before or after the gear
teeth are cut. If applied before cutting the teeth,
Soften the steel and improve ductility and
the hardness usually is governed and limited by
the most feasible machining process. Since
Relieve internal stresses caused by previous
treatment and improve dimensional stability
these gear teeth are cut after heat treatment, no
further nishing operation is needed. On the
Rene the grain structure
other hand, gears that are designed for hardness In process annealing, the steel is heated to a
above the machining limit are rst cut to semi- temperature below or close to the lower critical
nish dimensions and then through hardened. In temperature followed by the desired rate of
case of some minor heat treat distortion, a n- cooling. The purpose here is to soften the steel
ishing operation such as lapping or grinding is partially and to release the internal stresses. In
very often used to improve the quality of these this treatment, grain renement by phase trans-
gearsAmerican Gear Manufacturers Associa- formation is not accomplished as it is in full
tion (AGMA) class 10 and above. For quality up annealing. Process annealing uses temperatures
to class 9, gears are nished cut at least one between 550 and 650 C (1020 and 1200 F).
AGMA class above the requirement prior to Gears with hardness up to 34 HRC are fully
heat treatment. annealed by heating to 800 to 900 C (1475 to
156 / Gear Materials, Properties, and Manufacture

1650 F) and then furnace cooled to a pre- some increase in hardening obtained by normal-
scribed temperature, generally below 315 C izing leads to a more brittle chip and thus im-
(600 F). Typical hardnesses obtained after full proves machinability. Some through-hardened
annealing gears of different materials are shown gears may just require hardness obtained with
in Table 2. normalizing and annealing.
Normalizing and Annealing. In general, Normalizing and Tempering. Normaliz-
the term normalizing refers to the heating of steel ing consists of heating gears to 870 to 980 C
to approximately 38 C (100 F) above the up- (1600 to 1800 F) and then furnace cooling in
per critical temperature, followed by cooling in still or circulated air. This process results in
still air. The normalizing and annealing process higher hardness than annealing, with hardness
is used, either singularly or in a combination, being a function of the grade of steel and gear
as a grain structure homogenizing for alloy tooth size. However, normalizing does not in-
steel gears. The process also is used to reduce crease hardness signicantly more than anneal-
metallurgical nonuniformity such as segregated ing does, regardless of tooth size for plain carbon
alloy microstructures from previous mechanical steels containing up to 0.4% carbon (but it de-
working. A hypoeutectoid steel consisting of a nitely helps to ensure homogeneous microstruc-
structure of ferrite and coarse pearlite may be ture of steels). After normalizing, alloy steel
made easier to machine if the ferrite and cemen- gears are tempered at 540 to 680 C (1000 to
tite are more nely distributed. A very soft steel 1250 F) for uniform hardness and dimensional
has a tendency to tear in machining; therefore, stability.

Table 1 Approximate relation between various hardness-test scales

Brinell(a) C A 30-N 15-N B 30-T 15-T Vickers pyramid Tukon (Knoop)

... 70 86.5 86.0 94.0 ... ... ... 1076 ...

... 65 84.0 82.0 92.0 ... ... ... 820 840
... 63 83.0 80.0 91.5 ... ... ... 763 790
614 60 81.0 77.5 90.0 ... ... ... 695 725
587 58 80.0 75.5 89.3 ... ... ... 655 680
547 55 78.5 73.0 88.0 ... ... ... 598 620
522 54 77.5 71.0 87.0 ... ... ... 562 580
484 50 76.0 68.5 85.5 ... ... ... 513 530
460 48 74.5 66.5 84.5 ... ... ... 485 500
426 45 73.0 64.0 83.0 ... ... ... 446 460
393 52 71.5 61.5 81.5 ... ... ... 413 425
352 38 69.5 57.5 79.5 ... ... ... 373 390
301 33 67.0 53.0 76.5 ... ... ... 323 355
250 24 62.5 45.0 71.5 ... ... ... 257 ...
230 20 60.5 41.5 69.5 ... ... ... 236 ...
200 ... ... ... ... 93 78.0 91.0 210 ...
180 ... ... ... ... 89 75.5 89.5 189 ...
150 ... ... ... ... 80 70.0 86.5 158 ...
100 ... ... ... ... 56 54.0 79.0 105 ...
80 ... ... ... ... 47 47.7 75.7 ... ...
70 ... ... ... ... 34 38.5 71.5 ... ...
(a) Load, 3000 kgf; diam, 10 mm (0.4 in.)

Table 2 Typical Brinell hardness ranges of gears after through hardening

Brinell hardness (HB) Brinell hardness (HB)

Annealed; Maximum Brinell Annealed; Maximum Brinell

normalized Normalized hardness, quench normalized Normalized hardness, quench
Material and annealed and tempered and tempered Material and annealed and tempered and tempered

4130 155200 170215 350 4145 195240 285330 450

8630 155200 170215 350 4150 195240 285330 450
4140 185230 260300 425 4340 210255 300340 480
4142 185230 260300 425 HP 9-4-30 200240 (a) 520
8640 185230 260300 425 Maraging steel 200240 (a) 485
(a) Process generally is not used with these types of materials.
Chapter 8: Through Hardening / 157

Quench and Temper. The quench and tem- It is necessary to develop mechanical prop-
per process involves heating the gears to form erties (core properties) in gears that will not
austenite at 800 to 900 C (1475 to 1650 F), fol- be altered by any subsequent heat treatment
lowed by quenching in a suitable media such as such as nitriding or induction hardening.
oil. The rapid cooling causes the gears to become
harder and stronger by the formation of marten- Typical hardness ranges achieved for different
site. Hardened gears then are tempered at a tem- materials after through hardening by different
perature, generally below 690 C (1275 F), to processes are listed in Table 2.
achieve the desired mechanical properties.
Tempering lowers both the hardness and Through-Hardened Gear Design
strength of quenched steels but improves materi-
als properties such as ductility, toughness, and After nalizing a design, a gear designer
impact resistance. The tempering temperature needs to specify the following information on a
must be carefully selected based on the specied through-hardened gear drawing. This informa-
hardness range, the quenched hardness of the tion will help to minimize confusion for all
part, and the material. Normally, the optimum involved with gear manufacturing and material
tempering temperature is the highest tempera- procurement:
ture possible while maintaining the specied
hardness range. It is to be remembered that hard- Grade of steel with Aerospace Material
ness after tempering varies inversely with the Specication (AMS), if applicable
tempering temperature used. After tempering, AMS specication for material cleanliness,
parts usually are air cooled at room temperature. if required
Some steels can become brittle and unsuit- Hardnesses on tooth surface and at the core
able for service if tempered in the temperature AGMA gear quality level
range of 430 to 650 C (800 to 1200 F). This Each hardness callout should have at least a range
phenomenon is called temper brittleness and of 4 points in HRC scale or 40 points in Brinell
generally is considered to be caused by segrega- hardness (HB). Also, specify a tempering tem-
tion of alloying elements or precipitation of perature range on the drawing. This allows gear
compounds at ferrite and austenite grain bound- manufacturing engineers to select a particular
aries. If the gear materials under consideration tempering temperature for a specied hardness.
must be tempered in this range, investigation to
determine their susceptibility to temper brittle-
ness is needed. Molybdenum content of 0.25 to Hardness Measurement
0.50% has been shown to eliminate temper brit-
tleness in most steels. (Note: Temper brittleness The hardness of through-hardened gears gen-
should not be confused with the tempering erally is measured either on the gear tooth end
embrittlement phenomenon that sometimes re- face or rim section. This is the hardness that is
sults from tempering at a lower temperature used for gear rating purposes. Sometimes,
range, such as 260 to 320 C, or 500 to 600 F.) achieving specied hardness on the tooth end
The major factors of the quench and temper face may not necessarily assure the desired hard-
process that inuence hardness and material ness at the roots of teeth because of the grade of
strength are: steel, tooth size, and heat treat practice. If gear
tooth root hardness is critical to a design, then it
Material chemistry and hardenability
should be specied and measured on a sample
Quench severity
(coupon) processed with the gears. However,
Section size
needless increase of material cost by selecting a
Time at temper temperature
higher grade of steel should be avoided.
Of the four commonly used through-harden-
ing processes, the quench and temper method is
the most commonly used. This is particularly Distortion of Through-Hardened Gears
true when:
All steel gears experience distortion during
The hardness and mechanical properties a heat treat process. It is a physical phenome-
required for a given application cannot be non and cannot be eliminated from any heat
achieved by any of the other three processes. treat operation, although distortion of through-
158 / Gear Materials, Properties, and Manufacture

hardened gears is not as severe as in other in power transmission applications. The other
processes such as carburizing and nitriding. three processes are only employed to either
Still, through-hardened gears, particularly improve machinability or to enhance homoge-
quench and tempered class, experience enough neous grain structure of the gear steel. Use of
distortion that will eventually lower the quality through-hardened gears is limited because of the
level of gears after heat treatment. This necessi- low surface hardness that results in low gear pit-
tates a nishing operation for higher quality. In ting life and low power density gearbox com-
general, some materials expand after a through- pared with the one made with case-hardened
hardening operation while others contract. This gears. Also, for similar torque capacity, through-
requires a suitable stock allowance to be pro- hardened gears are larger with higher pitch line
vided on teeth for nish machining before heat velocity. This increases dynamic problems sub-
treatment of gears that are likely to distort. The stantially in a gearbox. However, in a bending
allowance needs to include expansion or con- strength limited design, through-hardened gears
traction of material and also distortion of tooth sometimes are successfully used, particularly for
geometry. For materials with predictable and large gears (over 508 mm, or 20 in., outer diam-
uniform distortion, gears could be cut to include eter) that normally exhibit high distortion if a
the distortion so that no nishing operation is case hardening process is used. An example for
required, possibly up to AGMA class 10 gear such an application is the internal ring gear of an
tooth quality. The majority of steels listed in epicyclic gearbox. These gears are usually de-
Table 2 expand during the through-hardening signed with hardness in the range of 32 to
process, whereas a few materials, such as marag- 34 HRC that can be nish cut after hardening,
ing steel, are found to contract. The amount of thus eliminating costly nishing operations.
expansion or contraction depends on alloy con- Through-hardened gears also are found to be ef-
tent, quality of steel, and conguration of gears. fective in applications susceptible to gear scuff-
In this regard, knowledge of distortion charac- ing. It is claimed that prole conformance of
teristics is helpful in optimizing the manufactur- through-hardened gears, because of their low
ing process of gears. When gears are made from surface hardness, reduces sliding friction and
a material without any previous heat treat distor- thereby helps to increase scufng resistance.
tion data, an experimental investigation is bene- Overall, through-hardened gears are used in
cial to establish the distortion characteristics of gearboxes that require large gears that cannot be
the material. With such data, cost-effective man- economically case hardened, such as large ma-
ufacturing methods can be established. An rine propulsion gears and railway power trans-
investigation of this nature carried out by an mission gears.
aerospace company to determine the heat treat
distortion characteristics of a through-hardened
gear rack for an aerospace application is dis- Case History: Design
cussed at the end of this chapter. Table 3 shows a
and Manufacture of a Rack
comparative distortion rating of some preferred
through-hardening materials for gears.
As explained in the section Distortion of
Through-Hardened Gears, all steel gears dis-
Applications tort after any type of heat treat process. Carbur-
izing imparts the highest distortion, while
Of the four different through-hardening through hardening imparts the least distortion.
processes described, those gears hardened by Even then, distorted gears require a nishing
quenching and tempering have some limited use operation for higher tooth quality. Sometimes,

Table 3 Distortion ratings of through-hardened gears

AMS AMS Distortion
Material specication quality Hardness (HRC) rating

AISI 4340 6414 2300 48/50 Good(a)

Maraging 250 6520 2300 49/52 Predictable and good(a)
Maraging 300 6521/6514 2300 52/56 Predictable and good(a)
AISI 4140 6382 2300 48/50 Good(a)
(a) Within one AGMA class of gear quality.
Chapter 8: Through Hardening / 159

for through-hardened gears, the knowledge of out. The dimensions and conguration of the
distortion characteristics may be included in the rack are shown in Fig. 1.
design of gear cutting tools such that gears after
heat treatment meet the desired quality. Such a Material Selection
case history is presented here. Table 4 lists the chemical compositions of
The project was to develop a low-cost, various materials considered in this case his-
high-bending strength (minimum of 250 ksi, or tory. The positive and negative attributes of
1720 MPa, ultimate tensile strength) corrosion- each and the associated heat treat process con-
resistant rack. For this application, the quality sidered before rack material selection are
required was rack teeth of AGMA class 9. To described subsequently.
minimize manufacturing cost, it was decided Option 1: Use of Quench-Hardening
not to consider any post-heat-treat nishing Steels. The following steels were considered:
operation. To meet these criteria, selection of a
proper material and a process was vital, for AISI 4340
which the following investigation was carried 300M

Fig. 1 Rack dimensions (given in inches) for preliminary tests. DP, diametral pitch; PA, pressure angle

Table 4 Chemical compositions of steels considered for racks

Material C Mn Si P S Ni Cr Mo V Co W Cu

Quench-hardening steels
AISI 4340 ... 0.60/0.80 0.20/0.35 ... ... 1.65/2.00 0.70/0.90 0.20/0.30 ... ... ... ...
HP 9-4-30 0.30 0.20 0.01 0.005 0.0007 7.50 1.00 1.00 0.08 4.0
H-11 0.35 ... ... ... ... ... 5.00 2.50 0.40 ... ... ...
300M 0.42 0.68 1.61 0.005 0.006 1.77 0.85 0.40 0.09 ... ... ...
Precipitation-hardening steels
17-4 PH ... ... ... ... ... 4.00 16.50 ... ... ... ... 4.00
13-8 Mo ... ... ... ... ... 8.00 12.80 2.30 ... ... ... ...
Age-hardening steels
Maraging 0.026 0.10 0.11 ... ... 18.5 ... 4.30 ... 7.0 ... ...
160 / Gear Materials, Properties, and Manufacture

HP 9-4-30 Process Selection

H-11 Heat Treat Distortion of Racks Made of
An excellent survey was made from published C-250. Although published literature indicated
literature to determine the various properties of distortion of maraging C-250 material is pre-
each, with the following results and conclusion: dictable, it was still necessary to nd out how
Results. High heat treat distortion: much the distortion would be for a rack tooth
used in this application. To determine these
Grinding of rack is necessary after heat
characteristics, a preliminary investigation was
treatment to attain the required accuracy of
teeth undertaken with racks made from readily avail-
able C-250 of shorter lengths (305 mm, or 12
High cost of material
in., long); longer lengths were not commercially
Poor corrosion resistance; additional
available at the time of this investigation. To
process needed to make racks corrosion
resistant expedite the program further, a standard shaper
cutter was used to cut the teeth. The racks were
Conclusion. None of these materials was then heat treated.
found suitable for the application. Heat Treatment of Racks. Some prelimi-
Option 2: Use of Precipitation Hardening nary experiments were conducted to select a
Steels. Steels considered: suitable heat treat furnace. The following fur-
17-4 PH naces were considered:
13-8 Mo Partial vacuum furnace
Results were as follows: Full vacuum furnace
Attainable mechanical properties were not at Air furnace
specied strength level Heat treatment in the air furnace was not accept-
Heat treat distortion was not predictable able due to:
Sensitive to grind burns
Problems with alloy segregation for any Oxidation of racks
post-heat treat nishing, in section size Scale removal resulted in size change
needed Both full and partial vacuum furnaces produced
Conclusion. Materials were not suitable. oxidation-free racks.
Option 3: Use of an Age-Hardening Steel Conclusion. Full vacuum furnace that
(Maraging C-250). This maraging steel has ensures oxidation-free parts was selected to
high nickel (18% or more), very low carbon determine heat distortion.
(under 0.03%), and is capable of developing Twelve racks were selected for heat treat-
very high tensile and yield strengths by means ment. All critical dimensions of the racks were
of an aging process. It is sold in the martensitic inspected and recorded before heat treatment.
state, which, because of the low carbon, is soft The heat treat procedure consisted of the
enough to be readily machinable. Heating to following steps:
approximately 480 C (900 F) for aging, and
cooling in the furnace, causes a change in mate- Vapor degrease racks
rial microstructure that increases the hardness Wipe racks with a cleaning chemical such as
up to 52 HRC. This meets the required tensile
strength. Select any two racks
Results were as follows: Hold the racks together back to back with
nickel-plated boltsprocessed horizontally
Published literature indicated distortion of on a at base in the furnace
maraging C-250 steel is predictable. Heat treat racks along with one tensile test
The material is available as forged, as well bar in each production lot, at 480 6 C (900
as in bar form, to AMS 6412. Sheet or plate 10 F) for 4.5 h at this temperature
stock available to AMS 6420 was not Furnace cool racks
acceptable due to nonuniform distribution of
Inspection consisted of:
mechanical properties.
Conclusion. Maraging C-250 forgings met Critical rack dimensions after heat treatment
the design requirements and were selected for Mechanical properties of material such as
this application. hardness and tensile strength
Chapter 8: Through Hardening / 161

Results. From the experimental results and special cutter. An investigation was then carried
inspection, the following conclusions were out with full-length racks.
made: Manufacturing method for full-length
racks consisted of the following steps:
Contraction rate of maraging steel was
between 0.0005 and 0.0007 mm/mm Forge blanks
(in./in.) and found to be linear and consistent Solution anneal to 35 HRC
in each lot Machine (mill and drill), leaving ground
Parts remained at after the hardening stock on sliding surfaces only
process Shape rack teeth to nal dimensions
Pitch and accumulative pitch errors were rough, seminish, and nish
within acceptable limit Straighten to 0.16 mm/m (0.002 in./ft)
Teeth perpendicularity (lead errors) were Inspect
between 0.013 and 0.025 mm (0.0005 and Heat treat (age) to 50 HRCtwo pieces
0.001 in.) bolted back to back
Pitch dimensions measured over a pin were Clean
held to 0.038 to 0.076 mm (0.0015 to 0.003 Grind sliding surfaces locating from the
in.) pitch line of rack teeth
Recommendations included: Inspect all rack dimensions before and after
heat treatment
Design of rack to include 0.076 mm (0.003
Test results are shown in Tables 5 and 6.
in.) tolerance for pitch dimension over the
Analysis of the results indicates:
Tooth perpendicularity (lead) error to 0.025 Part remained at after heat treatment within
mm (0.001 in.) 0.25 mm (0.010 in.)
A shaper cutter to be developed to include Pitch and accumulative pitch errors were
0.0006 mm/mm (in./in.) contraction rate of within the specied tolerance
rack tooth geometry with the expectation Tooth perpendicularity (lead) error was
that this might eliminate nish processing of within 0.025 mm (0.001 in.)
racks after heat treatment Measurements of pitch line over the pin
were within the new tolerance
New Shaper Cutter. With the proposed
contraction rate, a shaper cutter was designed Conclusions. Racks met the required quality
and manufactured by a cutter manufacturing level (AGMA class 9). In this case, the modied
company. Figure 2 shows the dimensions of this cutter and superior heat treat facilities made it

Fig. 2 Modied shaper cutter dimensions

162 / Gear Materials, Properties, and Manufacture

Table 5 Experimental results: full-sized racks dimensions before heat treat

Rack 1 Rack 2
Item Drawing dimension Left ank Right ank Left ank Right ank

Linear pitch, mm (in.) 9.9746 (0.3927) 9.972010 9.9779.9949 9.97719.9949 9.97719.9924

(0.39270.3936) (0.39360.3927) (0.39280.3935) (0.39280.3934)
Spacing error, mm (in.) 0.0330 (0.0013) max 0.0178 (0.0007) max 0.0229 (0.0009) max 0.0203 (0.0008) max 0.0178 (0.0007) max
Accumulated spacing 0.0762 (0.003) max, 0.1778 (0.007) max 0.1778 (0.007) max 0.0152 (0.006) max 0.0152 (0.006) max
error, mm (in.) over any 12 teeth
Tooth perpendicularity 0.0127 (0.0005) max 00.0533 (00.0021) 00.0533 (00.0021) 00.0406 (00.0016) 00.0406 (00.0016)
(lead error), mm (in.)
Dimension over pin, 20.119320.1574 20.294620.7518 20.294620.7518 20.269220.5994 20.269220.5994
mm (in.) (0.79210.7936) (0.7990.817) (0.7990.817) (0.7980.811) (0.7980.811)
Surface nish, rms 63 45 45 45 45
rms, root mean square

Table 6 Experimental results: full-sized racks dimensions after heat treat and grinding back face
Rack 1 Rack 2
Item Drawing dimension Left ank Right ank Left ank Right ank

Linear pitch, mm (in.) 9.9746 (0.3927) 9.96709.9873 9.97209.990 9.95939.9873 9.96709.9924

(0.39240.3932) (0.39260.3933) (0.39210.3932) (0.39240.3934)
Spacing error, mm (in.) 0.0330 (0.0013) max 0.0152 (0.0006) max 0.0152 (0.0006) max 0.0152 (0.0006) max 0.0178 (0.0007) max
Accumulated spacing 0.0762 (0.003) max, 0.1778 (0.007) max 0.1778 (0.007) max 0.1143 (0.0045) max 0.1143 (0.0045) max
error, mm (in.) over any 12 teeth
Tooth perpendicularity 0.0127 (0.0005) max 00.0533 (00.0021) 00.0533 (00.0021) 0.00250.0279 0.00250.0279
(lead error), mm (in.) (0.00010.0011) (0.00010.0011)
Dimension over pin, mm 20.119320.1574 20.101620.1422 20.101620.1422 20.127020.1955 20.127020.1955
(in.) (0.79210.7936) (0.79140.7930) (0.79140.7930) (0.79240.7951) (0.79240.7951)
Surface nish, rms 63 45 45 45 45
rms: root mean square

possible to manufacture the racks to AGMA ACKNOWLEDGMENT

class 9 without any subsequent nishing opera-
tion such as grinding. This chapter was adapted from A.K. Rakhit,
Experiments of this nature are denitely use- Through-Hardening Gears, Heat Treatment of
ful in determining the distortion of heat treated Gears: A Practical Guide for Engineers, ASM
gears and planning for subsequent nishing International, 2000, p 2132
operation, if needed. When the quantity of gears
to be produced is limited, and the allocated pro- SELECTED REFERENCES
duction development time is short, there may
not be many choices other than to nish the gear C.E. Bates, G.E. Totten, and R.L. Brennan,
after heat treatment. For low distortion, honing Quenching of Steel, Heat Treating, Vol 4,
is useful, while grinding is necessary for large ASM Handbook, ASM International, 1991, p
distortion. However, grinding after through 67120
hardening is not recommended. Some manufac- B.L. Bramtt and A.K. Hingwe, Annealing
turers, instead of honing a gear and the mating of Steel, Heat Treating, Vol 4, ASM Hand-
pinion individually, lap the gear and pinion book, ASM International, 1991, p 4255
together with a slurry of ne abrasive com- T. Ruglic, Normalizing of Steel, Heat Treat-
pound in the mesh until the desired quality is ing, Vol 4, ASM Handbook, ASM Interna-
obtained. tional, 1991, p 3541
In general, gears designed and manufactured M. Schmidt and K. Rohrbach, Heat Treating
to AGMA class 7 and below do not require any of Maraging Steels, Heat Treating, Vol 4,
such process development due to the fact that ASM Handbook, ASM International, 1991, p
the hobbing or shaping process can produce 219228
gears to AGMA class 8 and above. It is expected M. Wisti and M. Hingwe, Tempering of
that gears so produced will meet AGMA class 7 Steel, Heat Treating, Vol 4, ASM Hand-
after heat treatment. book, ASM International, 1991, p 121136
Gear Materials, Properties, and Manufacture Copyright 2005 ASM International
J.R. Davis, editor, p163-226 All rights reserved.
DOI:10.1361/gmpm2005p163 www.asminternational.org



CARBURIZING is a process in which Liquid carburizing, employing fused baths

austenitized ferrous metal is brought into con- of carburizing salts
tact with an environment of sufcient carbon Gas or atmosphere carburizing, employing
potential to cause absorption of carbon at the suitable hydrocarbon gases
surface and, by diffusion, create a carbon con- Vacuum carburizing, employing a rough
centration gradient between the surface and vacuum and a partial pressure of hydrocar-
interior of the metal. The depth of penetration of bon gas
carbon is dependent on temperature, time at Plasma or ion carburizing, employing both
temperature, and the composition of the carbur- vacuum and glow-discharge technology
izing agent. As a rough approximation, a car- (carbon-bearing ions are introduced to the
burized depth of approximately 0.76 to 1.3 mm surface for subsequent diffusion below the
(0.030 to 0.050 in.) on a 6 diametral pitch (DP) surface)
gear tooth can be obtained in about 4 h at 930 C
Today, gas (atmosphere) carburizing is consid-
(1700 F) with a carburizing agent, which may
ered the de facto standard by which all other
be solid, liquid, or gas.
surface hardening techniques are measured and
The primary objective of carburizing and
will be the emphasis of this chapter. However,
hardening gears is to secure a hard case and a
more recently introduced carburizing methods
relatively soft but tough core. For this process,
such as vacuum carburizing are rapidly gaining
low-carbon steels (up to a maximum of approx-
acceptance in the gear industry. The nal sec-
imately 0.30% carbon), either with or without
tion of this chapter reviews vacuum carburizing
alloying elements (nickel, chromium, man-
and compares the attributes of conventional gas
ganese, molybdenum), normally are used. After
carburizing and vacuum carburizing. Figure 1
case carburizing, the gear teeth will have high
compares the use (market breakdown) of vari-
carbon at the surface graduating into the low-
ous carburizing methods.
carbon core.
Sometimes to prevent through hardening of
the tooth tip, carbon penetration through tip of
tooth needs to be controlled. This is accom- Gas Carburizing:
plished by plating or spraying the outside diam-
eter of gear before cutting the teeth with some
Processing and Equipment
material that prevents the passage of the carbur-
As shown in Fig. 1(a), it is estimated that
izing agent. However, the most widely used
90% of gear carburizing is performed in a car-
method is copper plating. Several proprietary
bonaceous gas atmosphere. In this type of car-
solutions and pastes, which are quite effective in
burizing, carbon is induced into the ferrous base
preventing carburization, also are available.
material heated in the gaseous atmosphere with
There are ve general methods of carburiz-
a carbon potential that allows the surface to
ing, depending on the form of the carburizing
absorb carbon. The most commonly used
medium. These methods are:
medium is endothermic (commonly known as
Solid or pack carburizing, employing solid endo) gas produced by reacting natural gas
carburizing material (mainly methane, CH4) with air (1:2.5 to 2.7
164 / Gear Materials, Properties, and Manufacture

ratio) over a heated catalyst. The important Recently, nitrogen-methanol has been used
chemical reactions that take place can be for supplying the carburizing atmosphere. This
expressed as: type of system offers a number of benets over
the conventional endo gas generator. Here, N2
2CH4 O2 S 2CO 4H2 (Eq 1) plays the most important function to keep air out
2CO S C CO2 (Eq 2) of the furnace and prevents the gears from being
oxidized. Typically, N2 is more than 90% of the
Varying the ratio of methane to air alters the nitrogen-methanol atmosphere. Carburizing in
composition of endo and the chemical reactions nitrogen-methanol systems also ensures accu-
slightly. rate carbon potential for improved carburized
Free carbon resulting from chemical reaction case properties. Because of higher cost with
is then dissolved in the austenite that is formed nitrogen methanol system, most of the gears are
when gears are heated above 720 C (1330 F) still carburized in endo gas.
and precipitates as iron carbide (Fe3C).
Carburizing Temperature
The penetration of carbon into the steel
depends on the carburizing temperature, the
time at temperature, and the carburizing agent.
Since the solubility of carbon is greatest above
the Ac3 temperature, carburization takes place
most readily above this temperature. Further-
more, the higher the temperature is, the greater
the rate of carbon penetration will be, since the
rate of diffusion is greater. It is thus customary
to select a temperature approximately 55 C
(100 F) above the Ac3 point. Again, the time at
the carburizing temperature is the most inuen-
tial factor in the control of the depth of carbon
penetration as illustrated in Fig. 2.
Temperatures as low as 790 C (1450 F) and
as high as 985 C (1800 F) have been used for
carburizing gears, although it should be kept in
mind that the life of a furnace deteriorates rap-
idly above 955 C (1750 F). With the desired
amount of carbon absorbed into the tooth sur-

Fig. 1 North American carburizing market. (a) Market in Fig. 2 Depth of carbon penetration for different times and dif-
2000. (b) Anticipated market in 2010. Source: Ref 1 ferent temperatures in gas carburizing a gear steel
Chapter 9: Carburizing / 165

face, gears are quenched in a suitable medium content, and diffusion of carbon into steel. The
(generally oil) to obtain the required case hard- gas mixture may be adjusted to provide either a
ness. Quenching may be performed either carburizing or neutral atmosphere, making it
directly from the carburizing temperature or possible to diffuse the carbon in the case without
from a somewhat lower temperature. In some the further addition of surface carbon.
instances, parts after carburizing are completely Atmospheric Furnaces. Two types of
cooled to room temperature, reheated to the atmospheric furnaces are used: batch and con-
austenitizing temperatures, and then quenched. tinuous. The fundamental difference between
As already mentioned, the depth of case is these, aside from size, is the method by which
dependent on time and temperature selected the work is handled. With a batch furnace, as the
during carburizing. The following equation name implies, the workload is charged and dis-
generally satises the relationship between the charged as a single unit or batch. With continu-
case depth and carburizing time: ous furnaces, the load is fed on a continuous
basis at the charged end and is received at the
d 2t p (Eq 3) discharge end after processing. In a batch fur-
nace, changes in composition of carburizing
where d is the total case depth in inches, t is the agent take place in both environment and work-
carburizing time in hours at temperature with pieces until equilibrium is reached. In continu-
saturated austenite at the surface, and is the ous furnaces, the heating chamber is divided
proportionality factor of material that varies into zones. These zones may be separated by
with the carburizing temperature. internal doors. The atmospheres in each zone
For low carbon and alloy steels, the value of can be controlled to different carbon potentials.
is found to be approximately 0.025 for gas Coupled with different temperatures in each
carburizing at approximately 930 C (1700 F). zone, atmosphere control provides the main
Thus, Eq 3 can be rewritten for most alloy steels means of controlling the carburizing and diffu-
as: sion portion of the carburizing cycle. The chief
advantage of a batch furnace is its adaptability
d 0.025 2t p (Eq 4) to a variety of cycles. Each batch usually con-
Another relationship used to determine d is: sists of several individual gears. Very large
gears may be carburized one at a time; small
gears may be loaded several hundred to a batch.
31.6 2t
d (Eq 5) One of the major disadvantages of this furnace
6700 is that gears are transferred from ambient tem-
10 a b
T 460 perature into a furnace usually operating at the
where T is the carburizing temperature in F. carburizing temperature, causing thermal shock
Both Eq (4) and (5) are used in industry and pro- that may lead to uncontrolled distortion.
vide satisfactory results. Vacuum and Fluidized Bed Furnaces.
Besides time and temperature, the quality of Besides atmospheric-type furnaces, carburizing
case also depends on the type of carburizing fur- also may be performed in vacuum furnaces or
nace and equipment used. A proper selection is uidized bed furnaces. In general, heat transfer
thus essential for successful carburizing and characteristics of these furnaces are superior to
hardening. atmospheric-type furnaces, assuring better case
properties and uniformity of case. But the oper-
Furnaces and ating cost of these furnaces is higher than
atmospheric furnaces. Hence, such furnaces
Equipment for Gas Carburizing have more limited use.
There are three basic types of furnaces used Carburizing in a vacuum furnace is a rela-
for gas carburizing: atmospheric, vacuum, and tively new process. Any carburizing done in an
uidized bed. Each of these has its own advan- atmospheric furnace also can be accomplished
tages and disadvantages, although a great major- in a vacuum furnace. However, vacuum carbur-
ity of gears are gas carburized in atmospheric izing offers some signicant benets in time,
furnaces because gas carburizing seems to offer cost, and quality. A major reduction of time is in
acceptable control of case depth, surface carbon the heating-up phase of the process, where the
166 / Gear Materials, Properties, and Manufacture

parts are loaded into a cold furnace; parts and Regardless of the type of quenchant used, good
furnace are then heated up together. Further- circulation within the quench bath is extremely
more, the process allows a stepped heat-up important to promote uniform cooling of gears.
mode, which has also been found benecial. As Sometimes, for some materials, the desired
opposed to conventional carburizing in atmo- properties of the case can be developed without
spheric furnaces, where the carburizing medium resorting to a liquid quench, in which case air
is adjusted to generate a near eutectoid surface cooling, furnace cooling, or gas cooling may be
composition, vacuum carburizing is based on a appropriate. Any of these three cooling media
supersaturated carbon reaction. can be used when parts are to be reheated for
Selecting the most suitable alloy for vacuum hardening. If intermediate operations such as
carburizing is very important for its success. straightening or machining are needed prior to
Correct quench rates, commensurate with the hardening, furnace or gas cooling is preferred.
alloy selection, offer a less drastic quench mode Both are done under a protective atmosphere
for fully hardened case and core, which atmo- that keeps the gears clean and free of oxide scale
spheric carburizing and quenching does not and prevents decarburization of the surface.
allow. This also results in reduced distortion of
parts. The major advantage of vacuum carburiz- Direct Quenching
ing is that it offers better control of case depth Most gas-carburized gears are quenched
even at the root llet of the gear tooth. directly after carburizing. Furnace temperature
Carburizing in a uidized bed furnace is very usually is reduced to normal austenitizing tem-
similar to vacuum carburizing, in that supersat- perature (approximately 790 C, or 1450 F)
uration through direct reaction with the carbur- prior to quenching. In certain cases, quenching
izing media (natural gas, methane or propane) directly from the carburizing temperature also is
can take place at the face to control the surface acceptable provided this does not induce ther-
carbon level, together with the controlled case mal cracks in gears. On the other hand, some-
depth. times gears made of some high-alloy steels
(alloy content above 5%), are rst cooled in air
to room temperature after carburizing and then
Quenching and Hardening reheated and quenched for low distortion.
Nevertheless, direct quenching has gained
After carburizing, gears are quenched in a wider acceptance, primarily because of econ-
cooling medium for hardening. Quenching omy and simplicity of the procedure. To mini-
develops a martensitic or a bainitic case with mize carbide network in the case, carburizing
core microstructures other than a mixture of above the Acm temperature is suggested. Direct
proeutectoid ferrite and pearlite. Thus, the quenching reduces the amount of energy used
selection of a proper quenchant is of utmost for heating and eliminates or avoids some of the
importance, and the cooling rate, ideally, should equipment and operating expense of the harden-
be just fast enough to produce the desired core ing operation. Labor costs are reduced, and nicks
structure but not so fast that the case cracks or and other part damage are minimized because
that an undue amount of austenite is retained. the parts are handled less frequently. The use of
For industrial and automotive gears, however, ne-grain (ASTM 5 and above) steels, which
quenching conditions often are chosen solely on exhibit a relatively uniform response to heat
the basis of developing required surface hard- treatment, and the development of equipment
ness, especially in applications where the core and techniques for improved condence in car-
properties are known to have little or no effect bon control have led to wider acceptance of
on product performance. direct quenching as a means of hardening car-
Depending on part size and shape, and on burized gears made from a great variety of steels.
transformation characteristics of the steel, gears The xtures required for adequately support-
may be quenched in water, oil, or any propri- ing and separating individual gears during car-
etary uids. Most often oil is used because it is burizing also promote uniform direction and
a suitable quenchant for most carburizing velocity of the quenchant movement relative to
grades of steel, especially for relatively ne- each part on the xture. This leads to more con-
pitch gears. Small DP gears may require a more sistent metallurgical and dimensional quality.
drastic quench, particularly if densely packed, Furthermore, a single direct-quench operation
and often are susceptible to quench cracks. minimizes distortion by bringing about crystal-
Chapter 9: Carburizing / 167

lographic phase changes during only one heating ing. The original carburizing grade steels, which
and one cooling cycle. Each such phase change were high in nickel, required a reheat operation
results in volume change of grain microstructure after pack carburizing to produce a uniform
and increase in internal stress that may produce microstructure in the case. Reheating was also
substantial dimensional change of a gear. the only effective method of reducing the surface
Both horizontal-batch and pusher-type con- carbon content below saturation. Later, although
tinuous furnaces are well suited for direct low-alloy steels were introduced for gears, pack
quenching. Continuous furnaces sometimes are carburizing or crude gas-carburizing techniques
designed to remove one part at a time from still required gear reheating to control surface
the reduced-temperature section of the furnace carbon and microstructure. Today, the modern
for press quenching with a xture to control carburizing equipment is capable of producing
distortion. the desired microstructure in the case of both
The degree of distortion in some gears, for high- and low-alloy steels. Even then, certain
example, varies with case depth and the amount types of gear steels are still reheated before
of retained austenite in the case, as well as with quenching to ensure the quality of case micro-
alloy and process variables. Good control of structure and low distortion.
carbon-gradient shape, case depth, and surface Gears requiring individual quenching in a x-
carbon content are essential for direct quench- ture sometimes are reheated as a practical means
ing of dimensionally sensitive gears. In case the of conning the tedious one-at-a-time hardening
temperature of gears is reduced prior to quench- operation to a few simple hardening furnaces,
ing to minimize thermal shock, carbon content while a larger, more-expensive continuous car-
near the surface must be held to below satura- burizing furnace is permitted to operate at its
tion; otherwise, carbides will precipitate. A maximum capacity for direct quench. Some-
grain-boundary network of carbides in the case times, the total cost including labor and fuel can
is usually considered to be detrimental to gear be lower for a carburize, cool, and reheat proce-
life, although slow cooling after carburizing and dure than for direct quenching. Also, this tech-
then reheating before quench is one way to nique of carburizing, followed by slow cooling,
avoid or minimize the development of a carbide machining certain areas to remove the case, and
network. then hardening the entire gear sometimes is used
In the case of severe quench sometimes when selected areas must be free of a carburized
required to obtain high core hardness, the shape hardened case.
of the gear section being quenched is of great Furthermore, reheating of gears occasionally
importance since a combination of thick and is specied for grain renement. However,
thin sections (for example, annulus of epicyclic there is considerable disagreement over the
gearbox) may lead to cracking. Cracking results advantage of a reheated microstructure over a
due to a difference in the rate of cooling of thick direct-quenched microstructure, and whether
and thin sections. The transformation of thicker the former is preferred because of tradition, or
sections will take place when the thin sections because of a real need, is unknown. The amount
are at a lower temperature. The expansion in the of retained austenite is usually lower, or at least
thick sections on transformation will set up very is less visible in microstructures of reheated
high stress concentrations, which may cause gears. Again, the effect of retained austenite on
warpage or cracking. If proper core hardness gear performance is controversial, but clearly it
cannot be achieved for a certain gear tooth size, is not always detrimental. Also, if reheating is
an alternate material needs to be investigated. used, there is always the danger of greater ther-
mal distortion. The choice between direct
Reheating of Carburized quenching, and reheating and quenching should
Gears and Quenching often be decided on the basis of a specic appli-
Originally, only high-alloy steels were car- cation. Reheating generally is recommended
burized, because of the relatively unsophisti- only after high-temperature carburizing (above
cated steel-producing techniques required for 930 C, or 1700 F).
alloy additions to yield a steel with uniform Surface Hardness
response to heat treatment. At that time, carbur-
izing was done only by pack carburizing, which
Variations after Quenching
has a different set of material-handling and car- Variation in the surface hardness of gears
bon-control techniques than does gas carburiz- within a lot is a problem that is often encoun-
168 / Gear Materials, Properties, and Manufacture

tered when many small gears are heat treated in addition, tempering temperature has a signi-
the same basket. This variation is due to the cant effect on core hardness, as illustrated in Fig.
parts being too densely loaded, especially at the 4. Furthermore, higher tempering temperatures
center of the load. This restricts the ow of reduce both case hardness and case depth. In
quenchant in such a manner that gears near the applications where gears are required to main-
baskets perimeter may attain full surface hard- tain high compressive and bending strengths at
ness while those in the center do not. If it is not an elevated temperature, carburizing steels that
possible to space out the gears for economic rea- are least affected by tempering temperature are
sons, at least divider screens should be inserted preferred.
to layer the load. To enhance the effect of tempering, it should
Another possible cause of gear surface hard- follow soon after the quench but not until the
ness variation in a lot is insufcient quench bath gears can be comfortably touched with bare
agitation. The obvious solution is to speed up hands. Tempering too early can cause serious
the circulation system of the quench tank in problems by interrupting the martensitic trans-
order to move more quenchant through the load formation. To the other extreme, too long a
faster. Approximately 230 to 260 L/min (60 to delay before tempering might create a major
70 gal/min) of gear is considered an ideal rate. distortion problem and even cracking of the
This establishes the rate of cooling for ideal gears.
martensitic transformation of most gear steels. Tempering is a necessary nishing treatment
after hardening. However, it also involves heat-
ing and cooling. This may again generate new
Tempering of stresses in the gears being processed. Fortu-
Carburized and Quenched Gears nately, the inuence of these new stresses on
geometric shapes of gears is very small due to
Tempering is a process of reheating quench- low temperature levels involved. Nevertheless,
hardened gears to a temperature below the trans- uniform heating and cooling is advisable during
formation range of steel and holding at this tem- tempering to keep distortion-causing stresses at
perature to reduce thermal stresses induced a minimum.
during quenching and improve dimensional sta- Some controversy still exists concerning the
bility. Normal tempering temperature for car- value of tempering carburized and quenched
burized and quenched gears varies between 115 gears. For critical applications, experience has
and 175 C (240 and 350 F). The surface hard- proved that tempering is denitely benecial.
ness of quenched gears decreases as the temper- Carburized and hardened gears used in aero-
ing temperature increases as shown in Fig. 3. In space applications invariably need to be tem-

Fig. 3 Variation of tooth surface hardness with tempering tem- Fig. 4 Inuence of tempering temperature on core hardness
perature of carburized and hardened AISI 8620H gears of some high alloy steels
Chapter 9: Carburizing / 169

pered. The reasoning is that tempering is not which retained austenite in the case is trans-
harmful and provides some benet to resist formed to martensite. The percentage of trans-
cracking or chipping of gears under edge load- formation is related to temperature rather than to
ing. However, in thousands of other less critical time at a temperaturelower temperatures yield
applications, it is difcult or sometimes impos- higher levels of transformation. Multiple treat-
sible to prove the need of a tempering operation ments produce diminishing improvement. Al-
for carburized and hardened gears. most all of the signicant transformation is
achieved by the rst cold treatment.
The specic amount of martensitic transfor-
Recarburizing mation achieved by a given subzero treatment is
extremely difcult to predict. The degree of
Occasionally after carburizing and harden- reluctance to transform at a given temperature is
ing, gears in a certain lot are found to have inuenced by:
lower surface carbon and case depth even when
all the furnace carburizing parameters are kept The amount of retained austenite at the start
the same. To salvage these under-carburized of cold treatment
gears by recarburizing sometimes creates a The elapsed time between quenching and
number of potential problems. For example: cold treating
Every time a gear is heated, there is more Any intermediate thermal treatment, such as
If a hardened gear is charged into a hot fur- The general level of residual compressive
stress in the part
nace, it might crack.
If the carburized case depth is shallow, car- Any cold working of the material, such as
straightening of long slender pinions after
burizing the second time will increase case
carburizing and quenching
carbon content.
Temperatures in the range 75 to 100 C (100
This also can result in excessive retained au-
to 150 F) are routinely used in cold treating.
stenite or an undesirable carbide network. Thus,
Equipment for cold treatment up to 75 C
all scenarios are to be considered before recar-
(100 F) can be as simple as dry ice mixed with
burizing gears.
kerosene, trichloroethylene, or alcohol in a
bucket. Temperatures down to 100 C (150
F) is reached with relatively simple mechanical
Cold Treatment
refrigeration. Liquid nitrogen can be used for
chilling to any temperature down to 195 C
The presence of retained austenite in a heat
(320 F) but is seldom used.
treated case can be the source of dimensional
instability, excessive residual stress, or cracking,
all of which may cause service problems with a
carburized gear. Retained austenite is more Selection of
prominent at high surface carbon and alloy con- Materials for Carburized Gears
tents, and in cases where gears have been direct
quenched from high carburizing temperatures There is a wide variety of carburizing grade
rather than cooled, reheated, and quenched. materials that offer different mechanical proper-
High-carbon and high-alloy contents in steels ties (Table 1). Heat treat data for some of these
depress the temperatures at which martensitic steels are given in Table 2. In general, a material
transformation begins and ends. In some in- that can attain a tooth surface hardness around
stances, the transformation temperature may be 60 HRC and a core hardness between 32 and 48
well below room temperature, which favors the HRC after carburizing and hardening is se-
retention of large amounts of austenite. lected. The choice determines the surface case
One way of reducing the amount of retained and core hardnesses. A low tooth surface hard-
austenite in the case microstructure is to cold ness reduces the pitting life of gears. On the
treat a gear following quenching. Cold treatment other hand, low core hardness reduces bending
is basically a continuation of quenching, during fatigue life. As already discussed, the surface
170 / Gear Materials, Properties, and Manufacture

hardness of a gear tooth is strictly dependent on velop a much deeper martensitic structure when
the percentage of carbon at the surface, whereas similarly treated.
the core hardness is related to the hardenability As explained previously, the higher the sur-
of the material and is determined by the alloying face carbon is, the higher the surface hardness
elements in the steel. It is thus essential to have will be until surface carbon reaches around
a clear understanding of the terms hardness and 0.6%. To increase hardenability of steels the
hardenability of gear materials. Metallurgi- chemical compositions of steels need to be
cally, these two terms are quite different from altered to slow down martensitic transformation.
one another as discussed subsequently. This is achieved by alloying the steels. All alloy-
ing elements, except cobalt, increase hardenabil-
Hardness and Hardenability ity of steels. The common method to determine
Hardness is a surface property, whereas hard- hardenability is by the Jominy, or end-quench,
enability of steels refers to depth and hardness test where a test bar of the steel, 25 mm (1 in.) in
distribution induced by quenching. Steels with diameter by 100 mm (4 in.) long, normalized and
low hardenability can be hardened to a rela- machined to remove the decarburized surface, is
tively shallow depth. In these types of steels, heated to the hardening temperature for 30 min-
austenite-martensite transformation takes place utes. It is then quickly transferred to a xture that
in the area close to the surface only. The center holds the bar in a vertical position, and a jet of
of heat treated section remains soft or trans- water under controlled conditions is directed
forms to a structure softer than martensite, such immediately against the bottom end only (Fig.
as pearlite. Steels with high hardenability de- 5a). The end of the bar is thus cooled very

Table 1 Chemical compositions of frequently used carburizing grade gear steels

Composition, wt%

Material C Mn P S Si Ni Cr Mo V W Co

AISI grades
3310 0.080.13 0.450.60 0.025 max. 0.025 max. 0.200.35 3.253.75 1.401.75 ... ... ... ...
3310H 0.070.13 0.300.70 0.035 max. 0.040 max. 0.200.35 3.203.80 1.301.80 ... ... ... ...
4023 0.200.25 0.700.90 0.035 max. 0.040 max. ... ... ... 0.200.30 ... ... ...
4027 0.250.30 0.700.90 0.035 max. 0.040 max. ... ... ... 0.200.30 ... ... ...
4118 0.180.23 0.700.90 0.035 max. 0.040 max. 0.200.35 ... 0.400.60 0.080.15 ... ... ...
4118H 0.170.23 0.601.00 0.035 max. 0.040 max. 0.200.35 ... 0.300.70 0.080.15 ... ... ...
4140 0.380.43 0.751.00 0.035 max. 0.040 max. 0.200.35 ... 0.801.10 0.150.25 ... ... ...
4320 0.170.22 0.450.64 0.035 max. 0.040 max. 0.200.35 1.652.00 0.400.60 0.200.30 ... ... ...
4330M 0.300.33 0.96 0.008 max. 0.006 0.30 1.82 0.88 0.44 0.09 ... ...
4620 0.170.22 0.450.65 0.035 max. 0.040 max. 0.200.35 1.652.00 ... 0.200.30 ... ... ...
4815 0.130.18 0.400.60 0.035 max. 0.040 max. 0.200.35 3.253.75 ... 0.200.30 ... ... ...
4820 0.180.23 0.500.70 0.035 max. 0.040 max. 0.200.35 3.253.75 ... 0.200.30 ... ... ...
8617 0.150.20 0.700.90 0.035 max. 0.040 max. 0.200.35 0.400.70 0.400.60 0.150.25 ... ... ...
8617H 0.140.20 0.600.95 0.035 max. 0.040 max. 0.200.35 0.350.75 0.350.65 0.150.25 ... ... ...
8620 0.180.23 0.700.90 0.035 max. 0.040 max. 0.200.35 0.400.70 0.400.60 0.150.25 ... ... ...
8620H 0.170.23 0.600.95 0.035 max. 0.040 max. 0.200.35 0.350.75 0.350.65 0.150.25 ... ... ...
8720 0.180.23 0.700.90 0.035 max. 0.040 max. 0.200.35 0.400.70 0.400.60 0.200.30 ... ... ...
8822H 0.190.25 0.701.05 0.035 max. 0.040 max. 0.200.35 0.350.75 0.350.65 0.300.40 ... ... ...
9310 0.080.13 0.450.65 0.025 max. 0.025 max. 0.200.35 3.003.50 1.001.40 0.080.15 ... ... ...

Special quality grades

HP 9-4-30 0.30 0.20 0.005 0.007 0.01 7.50 1.00 1.00 0.08 . . . 4.0
Pyrowear 0.10 0.37 ... ... ... 2.15 1.05 3.30 0.12 ... ...
17CrNiMo6 0.140.19 0.400.80 0.125 max. 0.0050.020 0.150.35 1.401.70 1.501.80 0.250.35 0.06 max. ... ...
VASCO X2M 0.15 0.30 0.15 max. 0.01 max. 0.90 ... 5.0 1.40 0.45 . . . 1.35
M50NiL 0.15 0.25 ... ... ... 3.5 3.5 4.0 1.0 ... ...
EX 30 0.130.18 0.700.90 0.04 0.04 0.200.35 0.701.00 0.450.65 0.450.60 ... ... ...
EX 55 0.150.20 0.201.00 0.04 0.04 0.200.35 1.652.00 0.450.65 0.450.80 ... ... ...
Table 2 Heat treating data for some gear steels
Normalizing Annealing Hardening Carburizing Reheat Ms
temperature temperature temperature temperature temperature temperature(a)


3310 900950 16501750 860 1575 775800 14251475 900930 16501700 790815 14501500 350 655
3140 815930 15001700 790840 14501550 815840 15001550 ... ... 310 590
4028 870930 16001700 830860 15251575 ... 870930 16001700 790815 14501500 400 750
4047 840950 15501750 830860 15251575 800840 14751550 ... ... ... ...
4130 870930 16001700 790840 14501550 840900 15501650 ... ... 365 685
4140 870930 16001700 790840 14501550 830885 15251625 ... ... 315 595
4320 870930 16001700 860 1575 ... 900930 16501700 775800 14251475 380 720
3440 870930 16001700 590660 11001225 800830 14751525 ... ... 285 545
4620 930980 17001800 860 1575 ... 900930 16501700 800830 14751525 290 555
4640 870930 16001700 790840 14501550 790840 14501550 ... ... 320 605
4820 900950 16501750 875 1575 ... 900930 16501700 790815 14501500 365 685
5145 870930 16001700 790840 14501550 800830 14751525 ... ... ... ...
6120 930980 17001800 870 1600 ... 930 1700 775800 14251475 400 760
6150 900950 16501750 840900 15501650 840900 15501650 ... ... 285 545
8620 870930 16001700 860 1575 ... 930 1700 775840 14251550 395 745
9310 900950 16501750 860 1575 ... 900930 16501700 775840 14251550 345 650
EX 24 900950 16501750 870 1600 870 1600 900930 16501700 815840 15001550 445 830
EX 29 900950 16501750 870 1600 870 1600 900930 16501700 815840 15001550 445 830
EX 30 900950 16501750 840 1550 870 1600 900930 16501700 815840 15001550 445 830
EX 55 900950 16501750 830 1525 870 1600 900930 16501700 815840 15001550 420 790
(a) Ms is the temperature at which martensite forms.
Chapter 9: Carburizing / 171
172 / Gear Materials, Properties, and Manufacture

quickly while cross sections remote from the end rically opposite ats approximately 6 mm ( in.)
are cooled more slowly. The rate of cooling is wide are ground along the length of the bar and
dependent on the distance from the quenched Rockwell hardness measurements are made at
end. After the cooling is completed, two diamet- intervals of 1.6 mm (1 16 in.), on one or both the
at surfaces so prepared. The relationship of
hardness to distance from the quenched end is an
indication of the hardenability of steel (Fig. 5b
and 6).
Carbon content in alloy steels also plays an
important role in developing hardenability. Fig-
ures 7 and 8 show hardenability curves for some
steels with different content of carbon. The
degree of hardness at the quenched end (sur-
face) depends primarily on the carbon content,
but the hardness at any point away from this end
depends on the alloy content in the steel as well
as carbon. Deep-hardening steels produce atter
hardenability curves (Fig. 6). Commercially
available H-band steels assure high harden-
ability. Producers of such steels usually indicate
minimum and maximum hardnesses that are
expected at any depth from the quenched end of
the bar.
Effect of Common Alloying Elements on
Hardness and Hardenability. There are ve
fundamental factors that inuence hardenability
of steel:
Mean composition of the austenite
Homogeneity of the austenite
Grain size of the austenite
Nonmetallic inclusions in the austenite
Undissolved carbides and nitrides in the
It has been found that the effect of dissolved
elements that combine with carbon in prefer-
ence to dissolving in ferrite have the greatest
inuence in increasing hardenability if they are
dissolved in the austenite before quenching. A
carbide-forming element that is not dissolved in
the austenite has no effect on hardenability
except that as a carbide it may restrict grain
growth, thus reducing the hardenability of the
steel. Undissolved carbides reduce both the
alloy and carbon content of the austenite. Since
undissolved carbides restrict grain growth, in
some instances, quenching from higher temper-
ature that increases grain size promotes deeper
hardening. In summary, the factors that increase
hardenability are:
Dissolved elements in austenite (except
Fig. 5 Jominy end-quench apparatus (a) and method for pre- Coarse grains of austenite