AGMA 917-B97 Design Manual For Parallel Shaft Fine-Pitch Gearing

AGMA 917- B97
(Revision of
AGMA 370.01 (1973))
AMERICAN GEAR MANUFACTURERS ASSOCIATION
AGMA 917- B97
Design Manual for Parallel Shaft

Fine- Pitch Gearing
AGMA INFORMATION SHEET

(This Information Sheet is NOT an AGMA Standard)
Design Manual for Parallel Shaft Fine- Pitch Gearing

American
AGMA 917--B97
Gear
[Revision of AGMA 370.01 1973]
Manufacturers
CAUTION NOTICE: AGMA technical publications are subject to constant improvement,
Association
revision, or withdrawal as dictated by experience. Any person who refers to any AGMA
Technical Publication should be sure that the publication is the latest available from the
Association on the subject matter.
[Tables or other self--supporting sections may be quoted or extracted. Credit lines should
read: Extracted from AGMA 917--B97, Design Manual for Parallel Shaft Fine--Pitch
Gearing, with the permission of the publisher, the American Gear Manufacturers Association, 1500 King Street, Suite 201, Alexandria, Virginia 22314.]
Approved September 25, 1997
ABSTRACT
The rewritten Design Manual for Fine--Pitch Parallel Shaft Gearing is a cookbook style manual on how to design
fine--pitch spur and helical gears.
All work has been done with an eye towards computerization of the equations and the graphs.
In addition, the manual contains such specialized subjects as inspection, lubrication, gear load calculation
methods, materials, including a wide variety of plastics.
Published by
American Gear Manufacturers Association

1500 King Street, Suite 201, Alexandria, Virginia 22314
Copyright 1997 by American Gear Manufacturers Association
Reprinted June 1999
All rights reserved.
No part of this publication may be reproduced in any form, in an electronic
retrieval system or otherwise, without prior written permission of the publisher.
Printed in the United States of America

ISBN: 1--55589--694--4
ii
AGMA 917--B97
Contents
Page
Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
1
Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Definitions and symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Theory of involute gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Application considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Design synthesis and analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Design for control of backlash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Gear drawings and specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Gear tooth tolerances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
10
Materials and heat treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
11
Manufacturing methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
12
Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
13
Lubrication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
14
Bearings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
15
Load rating and testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Tables
1
Symbols, terms and units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Profile shift coefficients for 20 profile angle spur gears . . . . . . . . . . . . . . . . . 43
Maximum outside diameter for minimum topland of 0.275/Pnd . . . . . . . . . . . . 44
Figures
1
Basic geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Principal reference planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Planes at a pitch point on a helical tooth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Power transmission by two pulleys and a crossed belt . . . . . . . . . . . . . . . . . . . 8
Point on belt generates involute . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Involute nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Involute action and speed ratio independent of center distance . . . . . . . . . . . . 9
A series of small involute cams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Time = T1: Second pair of teeth just starting engagement . . . . . . . . . . . . . . . 10
10
Time = T2: First pair of teeth just leaving engagement . . . . . . . . . . . . . . . . . . 10
11
Time = T3: One pair of teeth in contact at pitch point . . . . . . . . . . . . . . . . . . . 11
12
Two involute curves showing differences in lengths of corresponding arcs . 12
13
Involute polar angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
14
Helix angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
15
Pitch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
16
Tooth pitches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
17
Principal pitches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
18
Base pitch relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

iii
AGMA 917--B97
Figures (continued)
19
Gear design flowchart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
20
Preliminary design flowchart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
21
Torque split . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
22
Speed decreasing gear system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
23
Gear train with idler gear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
24
External spur gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
25
Internal spur gear and external spur pinion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
26
External helical gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
27
Spur pinion and face gear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
28
Bevel gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
29
Crossed helical gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
30
Worm and wormgear set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
31
Spur rack and pinion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
32
Contact ratio vs. center distance deviation for 20 degree profile angle gears 28
33
Gear tooth as a simple beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
34
Tooth load acting at inscribed parabola . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
35
Shaft alignment deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
36
Standard pitch circle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
37
Transverse pitch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
38
Line of action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
39
Line of contact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
40
Transverse backlash is arc PR -- arc PQ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
41
Undercut teeth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
42
Effect of profile shift (addendum modification) . . . . . . . . . . . . . . . . . . . . . . . . . 40
43
Line of action for external gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
44
Line of action for internal gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
45
Showing angle at which load bears on tooth . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
46
Adjustable center distance gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
47
Spring loaded center distance gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
48
Adjustable split gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
49
Spring loaded split gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
50
Composite gearing with elastic element . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
51
Tapered gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
52
Spring preloaded gearing (for limited rotation) . . . . . . . . . . . . . . . . . . . . . . . . . 57
53
Dual path spring loaded gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
54
Contra--rotating input gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
55
Graphical representations of typical gear errors . . . . . . . . . . . . . . . . . . . . . . . . 63
56
Profile of 10 tooth, 20 DP, 20 PA gear tooth with undercut . . . . . . . . . . . . . . 74
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
iv
AGMA 917--B97
Foreword
[The foreword, footnotes, and annexes, if any, in this document are provided for
informational purposes only and are not to be construed as a part of AGMA Information
Sheet 917--B97, Design Manual for Parallel Shaft Fine--Pitch Gearing.]
Although there is a great deal of information about parallel shaft fine--pitch gearing in the
literature, it is widely scattered and a considerable number of areas are not well covered. As
a result, this manual has been compiled to provide a central source of the best information
available on the design, manufacture and inspection of fine--pitch gearing.
This manual is a revision of 370.01, Design Manual for Fine Pitch Gearing, 1973. Additions
have been made to the design section to broaden the concepts of gear theory and the gear
design process. Omitted from this Manual are wormgears, bevel gearing and face gearing
which appeared in the original design manual. This information is available in other AGMA
Standards.
An important feature of this manual is the bibliography to which the user is referred for
additional data in each area.
Suggestions for improvement of this standard will be welcome. They should be sent to the
American Gear Manufacturers Association, 1500 King Street, Suite 201, Alexandria,
Virginia 22314.
AGMA 917--B97
PERSONNEL of the AGMA Fine Pitch Gearing Committee

Chairman: D. E. Bailey . . . . . . . . . . . . . . Rochester Gear, Inc.
Editor: D. Castor . . . . . . . . . . . . . . . . . . . Eastman Kodak Company
ACTIVE MEMBERS
M.K. Anwar . . . . . . . . . . . . . . . . . . . . . . . .
P. Dean . . . . . . . . . . . . . . . . . . . . . . . . . . .
F. R. Estabrook . . . . . . . . . . . . . . . . . . . . .
I. Laskin . . . . . . . . . . . . . . . . . . . . . . . . . . .
D.A. McCarroll . . . . . . . . . . . . . . . . . . . . .
K. Price . . . . . . . . . . . . . . . . . . . . . . . . . . .
D.Seger . . . . . . . . . . . . . . . . . . . . . . . . . . .
L.J. Smith . . . . . . . . . . . . . . . . . . . . . . . . .
R.E. Smith . . . . . . . . . . . . . . . . . . . . . . . . .
S. Sundaresan . . . . . . . . . . . . . . . . . . . . .
M. Weiby . . . . . . . . . . . . . . . . . . . . . . . . . .
Allied Devices Corporation

Retired
Retired
Consultant
The Gleason Works
Eastman Kodak Company
Perry Technology Corporation
Invincible Gear Company
R.E. Smith & Company, Inc.
Eastman Kodak Company
Bison Gear & Engineering Co.
ASSOCIATE MEMBERS
A.F.H. Basstein . . . . . . . . . . . . . . . . . . . . .
D. Gimpert . . . . . . . . . . . . . . . . . . . . . . . . .
K. Gitchel . . . . . . . . . . . . . . . . . . . . . . . . . .
G.P. Lamb . . . . . . . . . . . . . . . . . . . . . . . . .
J.R. Mihelick . . . . . . . . . . . . . . . . . . . . . . .
G.E. Olson . . . . . . . . . . . . . . . . . . . . . . . . .
C. Sanderson . . . . . . . . . . . . . . . . . . . . . .
D.H. Senkfor . . . . . . . . . . . . . . . . . . . . . . .
M. Shebelski . . . . . . . . . . . . . . . . . . . . . . .
A. Sijtstra . . . . . . . . . . . . . . . . . . . . . . . . . .
Y. Tseytlin . . . . . . . . . . . . . . . . . . . . . . . . .
F.C. Uherek . . . . . . . . . . . . . . . . . . . . . . . .
A. Ulrich . . . . . . . . . . . . . . . . . . . . . . . . . . .
F.M. Young . . . . . . . . . . . . . . . . . . . . . . . .
vi
Crown Gear b.v.

Koepfer America Ltd.
Universal Tech. Systems, Inc. (Deceased)
Lamb & Lamb
Rockwell Automation/Dodge
Olson Engineering Services
Koepfer America Ltd.
Precision Gear Company
Boeing Precision Gear, Inc.
Crown Gear b.v.
Esterline Federal Products Corp.
Flender Corporation
UFE, Inc.
Forest City Gear Company
American Gear Manufacturers

Association --
Design Manual for

Parallel Shaft Fine--Pitch
Gearing
AGMA 917--B97
1.2 Design information

Information of the following subjects is supplied as
required by the design procedure:
-- Analysis of tooth proportions and meshing
conditions;
--
Inspection;
--
Gear tooth tolerances;
--
Gear blank design.
1.3 Additional design related material
1 Scope
This manual provides guidance for the design of
fine--pitch gearing of the following types:
--
Diametral pitch from 20 to 120;
--
Spur and helical (parallel axis);
--
External, internal and rack forms.
The guidance consists of the following:

-- Description of a design procedure in a series
of steps;
-- Design information -- data values, equations
and recommended practices;
--
Additional design related material.
The English system of units is used in this manual.

1.1 Design procedure
The description of the design procedure covers the
following:
--
Establishing proportions of the gears;
--
Selecting detailed gear data;
--
Confirming suitability of the tentative design;
--
Controlling backlash;
-- Meeting contact ratio and other gear mesh

requirements;
--
Specifying gear dimensions and tolerances.
The following background and supplementary

information is also supplied:
--
Manufacturing methods;
--
Gear material and heat treatment;
--
Lubrication;
--
Bearings.
1.4 Annexes
Annex A is a bibliography.
1.5 Limitations
The information in this manual is meant to serve
only as a guide to the designer of fine--pitch gears. It
is not intended that it be the procedure which must
be followed in the design of such gears, nor is it
implied that using the procedures and data will
necessarily result in gears that will meet the
requirements in every application. It remains the
responsibility of the individual designer to properly
evaluate the conditions in the particular application
and to make use of prior experience or proper
testing to confirm the suitability of the design.
1.6 Tooth form (spur and helical gearing,
internal and external)
The tooth form of the spur and helical gearing
considered in this manual is involute. Unless
specifically noted, all external spur and helical
designs resulting from the procedures discussed in
this manual will be conjugate with standard basic
racks. See ANSI/AGMA 1003--G93.
AGMA 917--B97
2 References
The following documents contain provisions which,
through reference in this text, constitute provisions
of the manual. At the time of publication, the
editions were valid. All publications are subject to
revision, and the users of this manual are encouraged to investigate the possibility of applying the
most recent editions of the publications listed:
ANSI/ASME Y14.6--1978 (R1993), Screw Thread

Representation, Engineering Drawing and Related
Documentation Practice (reaffirmation and
redesignation of ANSI/ASME Y14.6--1978).
3 Definitions and symbols

3.1 Definitions
AGMA 203.03, Fine--Pitch On--Center Face Gears

for 20--Degree Involute Spur Pinions.
The terms used, wherever applicable, conform to

ANSI/AGMA 1012--F90.
AGMA 900--F96, Style Manual for the Preparation

of Standards, Information Sheets and Editorial
Manuals.
3.2 Symbols and terms
AGMA 906--A94, Gear Tooth Surface Texture with

Functional Considerations.
AGMA 908--B89, Geometry Factors for Determining the Pitting Resistance and Bending Strength of
Spur, Helical and Herringbone Gear Teeth.
AGMA 910--C90, Formats for Fine--Pitch Gear
Specification Data.
ANSI/AGMA 110.04, Nomenclature of Gear Tooth
Failure Modes.
ANSI/AGMA 1003--G93, Tooth Proportions for
Fine--Pitch Spur and Helical Gearing.
ANSI/AGMA 1012--F90, Gear Nomenclature,
Definitions of Terms with Symbols.
ANSI/AGMA 2000--A88, Gear Classification and
Inspection Handbook.
ANSI/AGMA 2001--C95, Fundamental Rating Factors and Calculation Methods for Involute Spur and
Helical Gear Teeth.
ANSI/AGMA 2002--B88, Tooth Thickness Specification and Measurement.
ANSI/AGMA 2004--B89, Gear Materials and Heat
Treatment Manual.
ANSI/AGMA 2005--B88, Design Manual for Bevel
Gears.
ANSI/AGMA 6034--B92, Practice for Enclosed
Cylindrical Wormgear Speed Reducers and
Gearmotors.
ANSI/AGMA
Lubrication.
9005--D94,
Industrial
ANSI/ASME B46.1--1985, Surface Texture.
Gear
The symbols and terms used throughout this

manual are in basic agreement with the symbols
and terms given in AGMA Information Sheet
900--F96, Style Manual for the Preparation of
Standards and ANSI/AGMA Standard 1012--F90,
Gear Nomenclature, Definitions of Terms with
Symbols. In all cases, the first time that each
symbol is introduced, it is defined and discussed in
detail.
NOTE: The symbols and definitions used in this standard may differ from other AGMA standards. The user
should not assume that familiar symbols can be used
without a careful study of their definitions.
Throughout this manual, the term pinion refers to

the member of the meshing pair with the smaller
number of teeth without regard to which member is
driving. The term gear, as part of a meshing pair,
refers to the member with the larger number of
teeth.
In order to avoid confusion and to achieve consistency, any symbol that is applicable to a specific
member, pinion or gear, is given the subscript P for
pinion and subscript G for gear. The subscript n is
used to distinguish the normal plane from the
transverse plane. The subscript t is used to
distinguish the transverse plane in those situations
when confusion with the normal plane might occur.
Any symbol not specifically designated by a subscript n is assumed to be in the transverse plane.
The subscripts p, i, b and o represent the terms at
the operating diameter, inside diameter, base
diameter and outside diameter respectively.
The symbols and terms, along with the clause
numbers where they are first discussed, are listed in
alphabetical order by symbol in table 1.
AGMA 917--B97
Table 1 -- Symbols, terms and units

Symbol
ah
B
Bb
Bc
BhT
Bn
BnhT
C
Cn
Cs
cG
cmin
cP
DbG
DbP
DG
DiG
DoG
DoP
DP
DrG
DrP
dG
dP
EG
EP
eaG
eaP
F
J
Kv
La
Lr
mF
mG
mp
N
NG
NP
Pd
Pnd
pbn
pb
Terms
Addendum of the cutter
Transverse backlash (backlash in the transverse plane)
Backlash along the line of action
Transverse backlash due to change in center distance
Minimum backlash for spur gears
Normal backlash (backlash in the normal plane)
Minimum backlash for helical gears
Operating center distance
Nominal center distance
Standard center distance
Clearance at the root of the gear
Minimum clearance at the root
Clearance at the root of the pinion
Base diameter of the gear
Base diameter of the pinion
Standard pitch diameter of the gear
Inside diameter of the gear
Outside diameter of the gear
Outside diameter of the pinion
Standard pitch diameter of the pinion
Root diameter of the gear
Root diameter of the pinion
Operating pitch diameter of the gear
Operating pitch diameter of the pinion
Youngs modulus of the gear
Youngs modulus of the pinion
Roll angle at the start of active profile for the gear
Roll angle at the start of active profile for the pinion
Face width
Geometry factor
Dynamic factor
Length of approach
Length of recess
Face contact ratio
Gear ratio
Transverse contact ratio
Number of teeth
Number of teeth on the gear
Number of teeth on the pinion
Transverse diametral pitch
Normal diametral pitch
Normal base pitch
Transverse base pitch
Units
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
psi
psi
degrees
degrees
inches
inches
inches
1/inch
1/inch
1/inch
1/inch
Reference
6.8.1
6.3.9.1
6.3.9.3
6.3.9.1
7.2.1
6.3.9.2
7.2.1
6.3.1
6.5.1
6.2.8
6.8.2
6.8.2
6.8.2
6.2.4
6.2.4
6.2.2
6.5.2
6.5.2
6.5.2
6.2.2
6.8.1
6.8.1
6.3.5
6.3.5
6.9.6
6.9.6
6.7.3
6.7.3
6.9.5
6.9.5
6.9.5
6.7.4
6.7.4
6.7.7
5.3.2
6.7.6.1
5.5.1
5.5.1
5.5.1
6.2.1
4.3
4.2.7
6.2.6
(continued)
AGMA 917--B97
Table 1 (continued)
Symbol
pt
Q
RaG
RaP
RbG
RbP
RG
RiG
RoG
RoP
RP
RrG
RrP
T
tnG
tnP
tinG
titG
tonG
tonP
totG
totP
tpnG
tpnP
tptG
tptP
ttG
ttP
Vcq
Wa
Wn
Wr
Wt
X+min
X - max
XG
Terms
Transverse circular pitch
AGMA quality number (5 to 15)
Radius at the start of active profile of the gear
Radius at the start of active profile of the pinion
Base radius of the gear
Base radius of the pinion
Standard pitch radius of the gear
Inside radius of the gear
Outside radius of the gear
Outside radius of the pinion
Standard pitch radius of the pinion
Root radius of the gear
Root radius of the pinion
Torque
Normal circular tooth thickness at the standard pitch diameter of the
gear
Normal circular tooth thickness at the standard pitch diameter of the
pinion
Normal circular tooth thickness at the inside diameter of the gear
Transverse circular tooth thickness at the inside diameter of the gear
Normal circular tooth thickness at the outside diameter of the gear
Normal circular tooth thickness at the outside diameter of the pinion
Transverse circular tooth thickness at the outside diameter of the
gear
Transverse circular tooth thickness at the outside diameter of the
pinion
Normal circular tooth thickness at the operating pitch diameter of the
gear
Normal circular tooth thickness at the operating pitch diameter of the
pinion
Transverse circular tooth thickness at the operating pitch diameter of
the gear
Transverse circular tooth thickness at the operating pitch diameter of
the pinion
Transverse circular tooth thickness at the standard pitch diameter of
the gear
Transverse circular tooth thickness at the standard pitch diameter of
the pinion
Total composite variation
Axial load (thrust load)
Normal load
Radial load
Tangential load
Minimum required amount of profile shift coefficient
Maximum allowable amount of negative profile shift coefficient
Profile shift coefficient of the gear
Units
1/inch
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
inches
lb--inch
inches
Reference
6.2.5
9.3
6.7.1
6.7.1
6.2.4
6.2.4
6.2.2
6.5.2
6.5.2
6.5.2
6.2.2
6.8.1
6.8.1
6.9.2
6.6.2
inches
6.6.2
inches
inches
inches
inches
inches
6.6.5
6.6.5
6.6.5
6.6.5
6.6.5
inches
6.6.5
inches
6.6.4
inches
6.6.4
inches
6.6.4
inches
6.6.4
inches
6.6.3
inches
6.6.3
inches
lbs
lbs
lbs
lbs
6.6.6
6.9.3
6.9.1
6.9.4
6.9.2
6.4.4.3
6.4.4.4
6.4.4
(continued)
AGMA 917--B97
Table 1 (concluded)
Symbol
XP
C
snG
snP
t
G
P
inG
itG
n
onG
onP
otG
otP
pn
pt
t
b
c
b
iG
oG
oP
p
HCR
LCR
SAP
TCT
TIF
Terms
Profile shift coefficient of the pinion
Involute polar angle
Change in center distance
Tooth thinning coefficient of the gear
Tooth thinning coefficient of the pinion
Change in tooth thickness due to total composite variation
Poissons ratio of the gear
Poissons ratio of the pinion
Normal pressure angle at the inside diameter of the gear
Transverse pressure angle at the inside diameter of the gear
Normal standard pressure angle
Normal pressure angle at the outside diameter of the gear
Normal pressure angle at the outside diameter of the pinion
Transverse pressure angle at the outside diameter of the gear
Transverse pressure angle at the outside diameter of the pinion
Normal operating pressure angle
Transverse operating pressure angle
Transverse standard pressure angle
Constant, value = 3.1415927
Bending stress
Contact stress
Helix angle
Base helix angle
Helix angle at the inside diameter of the gear
Helix angle at the outside diameter of the gear
Helix angle at the outside diameter of the pinion
Operating helix angle
High contact ratio
Low contact ratio
Start of active profile (limit diameter)
Total composite tolerance
True involute form (form diameter)
4 Theory of involute gearing
Clause 4 contains discussions of what an involute is

and why it is the geometric shape of choice for gear
teeth. Simply put, involute gears transmit uniform
rotary motion from one shaft to another shaft. Gear
tooth flanks which transmit uniform rotary motion
are said to be conjugate. Clause 4 also attempts to
bring together the geometric aspects of the involute
gear with the more common mathematical expres-
Units
radians
inches
inches
degrees
degrees
degrees
degrees
degrees
degrees
degrees
degrees
degrees
degrees
psi
psi
degrees
degrees
degrees
degrees
degrees
degrees
Reference
6.4.4
4.1.11
6.3.9.1
6.6.1
6.6.1
6.6.6
6.9.6
6.9.6
6.5.5
6.5.5
6.2.3.2
6.5.5
6.5.5
6.5.5
6.5.5
6.3.8
6.3.6
6.2.3.3
6.9.5
6.9.6
4.1.9.2
6.5.4
6.5.3
6.5.3
6.5.3
6.3.7
6.7.6
6.7.6
6.7.1
8.4.5.1
6.4.3
sions that are used to specify a gear as a machine

element.
Clause 4 presents basic background information
which will lay the foundation for the reader to
proceed with the gear design. Other clauses of this
design manual discuss the selection of gear
parameters such as type of gearing to be used,
number of teeth, diametral pitch and gear ratio.
These parameters are the foundation blocks of a
gear design. Clause 6 explains how to complete the
design, optimize and fully specify the gear system.
AGMA 917--B97
-- Center distance, operating. In parallel axis

gearing, the distance between the gear axes in
the plane of rotation. See figure 1.
4.1 Basic geometric considerations

The following will clarify the concepts needed to
understand the many useful features and
characteristics of involute gearing.
-- Gear ratio. The ratio of the rotational

speeds of the input and output shafts. See 5.3.2.
-- Axis. The axis, perpendicular to the plane of

rotation, and lying in the axial plane, passing
through the geometric center of the gear. See
figure 1.
-- Pitch point. A point on the line of centers

which divides the center distance in the same
proportion as the gear ratio. See 6.3.3.
-- Pitch line. A line lying in a plane of rotation
passing through the pitch point and
perpendicular to the line of centers.
-- Plane of rotation. Any plane perpendicular

to a gear axis. Also known as a transverse
plane. See figure 1.
-- Pitch plane. For a pair of gears, a plane perpendicular to the axial plane and tangent to the
pitch surfaces. A pitch plane in an individual gear
may be any plane tangent to its pitch surface.
The pitch plane of a rack or a crown gear is the
pitch surface. See figure 2.
-- Gear center. The point of intersection of the

gear axis and the plane of rotation. See figure 1.
-- Line of centers. A line that connects the
centers of two meshing gears in the plane of
rotation.
Pitch circle
(operating)
Pinion axis
Pinion
center
Center
distance, C
Pitch point
Gear axis
Line of
centers
Gear
center
Pitch circle
(operating)
Axial plane
90
Plane of rotation
(transverse plane)
Figure 1 -- Basic geometry
-- Axial plane. In a single gear, an axial plane

may be any plane containing the axis and a given
point. See figure 2.
Pitch plane
Transverse
plane
Pitch
cylinder
Axial plane
Figure 2 -- Principal reference planes

-- Transverse plane. A plane perpendicular to
the axial plane and to the pitch plane. In gears
with parallel axes, the transverse plane and the
plane of rotation coincide. See figure 3.
Normal
plane
Transverse
planes
Pitch
plane
Line normal to
tooth surface in
normal plane
Pitch point
Figure 3 -- Planes at a pitch point on a helical

tooth
-- Normal plane. A normal plane is, in general,
normal to a tooth surface at a pitch point and
perpendicular to the pitch plane. See figure 3.
4.1.1 Involute geartooth interaction
If two circles are drawn, one about each of the gear
centers and each with a radius from its center to the
AGMA 917--B97
pitch point, both will be tangent to each other and to

the pitch line. These circles rolling on each other
without slipping will transmit uniform rotary motion
from one shaft to another.
In order to eliminate slipping of two pitch circles,
teeth may be placed on them. The purpose of this
clause is to acquaint the reader with the basic
fundamentals of geartooth interaction. The concepts presented will demonstrate that when one
involute curve drives a mating involute curve, the
two curves touch at a point of contact. Moreover, all
contact takes place along a straight line called the
line of action.
It can also be shown that the force vector between
the teeth is in the direction of the line of action. The
cylindrical nature of spur and helical gears has
fostered the notion that gears transmit force in a
manner similar to friction disks, the teeth being
present only to prevent slippage. This view of gear
tooth interaction has led to confusion for novice gear
designers. Actually gear teeth interact as two cam
surfaces, one driving the other. The interaction of
the cams is accurately modeled as a pair of pulleys
connected by a crossed belt.
4.1.2 Crossed belt theory
As shown in figure 4, power and motion can be
transmitted from one shaft to another using two
pulleys and a crossed belt. (The belt is crossed to
give the same direction of rotation as meshing gears
in the same application.) A point, P, on the belt
generates a curve in space as the pulleys rotate and
the belt moves. The pulley is shown in four selected
positions, 20 degrees apart. The curve generated
by the point on the belt is an involute. The involute
has several unique properties that have made it the
curve of choice for gear teeth.
4.1.3 Involute cams
Figure 5 shows the resulting system if a point on the
same belt is used to generate an involute starting
from the surface of the other pulley. The first useful
property of involute curves is that they transmit
uniform rotary motion. That is, if the input shaft
rotates with uniform rotary motion, its involute curve
driving the involute of the output shaft, the output
shaft will also turn with uniform rotary motion. The
relative speeds of the two shafts will be determined
by the ratio of the diameters of the pulleys.
AGMA 917--B97
Pulley
C2
Belt pulled
at this end
Pulley
C2
Involute
Belt pulled
at this end
Involute
B1
B1
C1
Rb
Base circle disc
Time = T1
Pulley
C1
Rb
Base circle disc
Time = T2
Pulley
Belt pulled
at this end
Involute Involute P
C2
C2
Belt pulled
at this end
B1
C1
Rb
B1
C1
Rb
Base circle disc

Time = T3
Base circle disc

Time = T4
Figure 4 -- Power transmission by two pulleys and a crossed belt

4.1.4 Involute nomenclature
Pulley 1
(Pinion)
Base circle disc 1
Rb1
Involute 1
(Pinion)
Involute 2
(Gear)
Line of action
(belt)
Rb2
Base circle
disc 2
Pulley 2
(Gear)
Figure 5 -- Point on belt generates involute
Figure 6 shows the system being discussed after

the line of centers, the operating pitch circles, and
the pitch line have been constructed. The point at
which the line of centers intersects the line of action
is the pitch point. The circles from which the
involutes emanate (represented by the pulleys) are
the base circles. The shape of an involute curve is
dependent upon the diameter of its base circle.
The operating pitch circles have been drawn
centered on the pulley centers, passing through the
pitch point. Thus, they are tangent to one another at
the pitch point. The ratio of the diameters of the
operating pitch circles is the same as the ratio of the
base circle diameters.
The pitch line has been constructed perpendicular
to the line of centers, passing through the pitch
point. The angle between the pitch line and the line
of action is called the operating pressure angle.
Operating pressure angle, pressure angle and
profile angle will be discussed in more detail later in
this information sheet. See 6.2.3.
Pulley 1
(Pinion)
Base circle 1
Operating pitch
circle, dP
Operating pressure
angle, pt
Pitch line
AGMA 917--B97
R b2
d
= G
R b1
dP
Rb1
Pitch
point
Involute 1
(Pinion)
Line of
action (belt)
Involute
2 (Gear)
Line of
centers
Rb2
Operating pitch
circle, dG
Pulley 2
(Gear)
Base circle 2
Figure 6 -- Involute nomenclature
Involute action is not dependent upon center

distance. Figure 7 shows the effect of changing
center distance. The base circles and involutes are
the same in both examples. When the center
distance is changed, the angle that the line of action
(the belt) makes with the line of centers will change
accordingly. Changing the center distance of the

system changes the diameters of the operating
pitch circles; however, it does not alter their ratio.
Uniform rotary motion is still transmitted and the
speed ratio remains the ratio of the base circles.
Pinion
Pinion
dP1
RbP
RbP
Involute
(Pinion)
pt1
C1
dP2
P
Involute
(Gear)
Line of
Action
(Belt)
RbG
Gear
Involute
(Pinion)
pt2
C2
dG1
R bG
d
d
= G1 = G2
RbP
d P1
dP2
P
Involute
(Gear)
Line of
Action
(Belt)
RbG
dG2
Gear
Figure 7 -- Involute action and speed ratio independent of center distance
AGMA 917--B97
4.1.5 Involute gears

One large, single involute cam driving another
single involute cam has functional limitations, such
as being limited to less than one revolution. Figure 8
demonstrates how multiple, smaller involute cams,
spaced evenly around the circumference of the
pulleys, may be used to transmit motion in a
continuous manner as the driver is rotated. These
multi--cammed machine elements are usually
referred to by a more common name ... involute
gears.
leaving contact. Figure 11 shows the gears rotated

to a position where only one pair is in contact.
Contact ratio is discussed in 6.7.
Time = T1
Point of contact
(T1=T)
Base pitch
Pitch point
Involute 2
cam surfaces
Line of centers
Point of contact
(T1=T)
Line of action
Base circle
Base circle
Base circle
Direction of
force vector
Base circle
Involute 1
cam surfaces
Figure 9 -- Time = T1: Second pair of teeth

just starting engagement
Figure 8 -- A series of small involute cams

Not every pulley can be used as the basis for a gear.
Certain criteria must be met. The circumference of
the base circle must be divisible into an integer
number of segments, so that an integer number of
teeth can be constructed around it. With very few
exceptions, gears contain integer numbers of teeth
in order to rotate continuously. This division of the
circumference defines another useful gearing concept. The length of each equal segment of the
circumference of the base circle is called the base
pitch (see 4.2.6). Physically, the base pitch is the
distance between successive teeth around the gear
and along the line of action. Two involute gears will
mesh together if they possess equal base pitches.
Time = T2
Point of contact
(T2 = T)
Base pitch
Line of centers
Pitch point
Point of cont
(T2 = T)
Base circle
4.1.6 Contact ratio

Figure 9 shows that the base pitch is the distance
along the line of action between successive involute
flanks. In figure 9, one pair of teeth is in contact and
the next pair is just beginning contact. Figure 10
shows the gears rotated until the initial pair is just
10
Base c
Direction of
force vector
Figure 10 -- Time = T2: First pair of teeth just

leaving engagement
theoretically vary as subsequent tooth pairs roll

through mesh. However, the loading on a given
tooth will vary depending on whether the load is
carried by a single pair or by multiple pairs. Later
discussion will show that the actual forces on the
flanks vary because of frictional effects due to the
rolling and sliding nature of involute action.
Time = T3
Point of contact
(T2 = T)
AGMA 917--B97
Base pitch
Line of centers
Pitch point
Base circle
Base circle
Direction of
force vector
Figure 11 -- Time = T3: One pair of teeth in

contact at pitch point
Contact ratio can be thought of as the average
number of teeth in contact during one mesh cycle. It
is very important to have a contact ratio greater than
one, ensuring that a driving pair comes into mesh
before the previous pair leave mesh. If contact ratio
is less than one, the action between the gears is not
occurring on involute flanks. Contact ratio less than
one can result in non--uniform rotary motion being
transmitted, noise and high stresses in the gears
and other components of the system. Contact ratio
is a key parameter in most gear designs and will be
discussed in depth in subsequent clauses (see
6.7.5).
4.1.7 Properties of involute gear action
Figures 9, 10 and 11 show an involute gear pair in
mesh. The three positions shown demonstrate how
the point of contact between the flanks of the gear
teeth follows the straight line path of the line of
action. It can be shown that the direction of the force
vector between the two flanks is also in the direction
of the line of action.
This force is equal to and in the same direction as
the force that will be transmitted in the crossed belt
that comprised the first model of the system. Note
the transmitted torque in the system does not
4.1.8 Rolling and sliding of involute gears

Figure 12 shows two involute curves divided into
corresponding lengths of arc which will pass
through mesh at the same time. Notice that the
length of the segment along an involute is short near
the base circle and gets progressively longer farther
out the involute. As two external involute teeth enter
mesh, contact is near the root of the tooth on the
driver and near the tip on the driven. The driver
involute section is much shorter than the driven, yet
both pass through mesh in the same amount of
time. There is relative sliding between the tooth
flanks.
As the pair approach the pitch point, (in arc 12) the
lengths of their respective involute arcs approach
equality. Sliding is much less and at the pitch point
pure rolling exists. Near the end of a mesh cycle,
the situation is the reverse of the beginning, short
segments on the driven and long on the driver. The
result, however, is the same ---- significant relative
sliding. Skilled gear designers attempt to minimize
the sliding in gear meshes in order to increase
efficiency and prolong the life of the gears.
4.1.9 Base pitch, diametral pitch and standard
pitch diameter
In general, the spacing of the driving surfaces of the
teeth on a gear is called the pitch (see 4.2). This
spacing may be measured on any circle, but one
selected must be specified. Prior to the introduction
of involute gearing, cog wheels, cycloidal and other
tooth forms were used for clocks and machines.
The nomenclature already in use for these early
forms of gearing, terms such as pitch diameter and
circular pitch, remained after involute gearing
became more popular. Over time, the mathematics
of the base pitch system of gear specification was
almost completely replaced by the mathematically
simpler diametral pitch system that is used today
in the United States. The nomenclature of the
diametral pitch system contains many of the same
terms as the preinvolute systems, a fact that can
11
AGMA 917--B97
provide confusion for novice gear designers. The

remainder of this design manual uses the diametral
pitch system to design gears. The mathematical
link between the two methods (base pitch and
diametral pitch) that describes the involute gear is
the following:
Diametral Pitch =
cosprofile angle
Base Pitch
Position of minimum sliding is the one shown here where arc 12 equals arc 12. For rotation in either
direction, sliding action will increase until the contact point reaches the base circle.
Figure 12 -- Two involute curves showing differences in lengths of corresponding arcs
12
It is clear that both a diametral pitch and a profile

angle (see 4.1.9.1) must be specified in order to
determine a unique base pitch.
The prior discussion of involute theory has been
based upon two dimensional geometries. The
above expression is true for geometry in the
transverse plane (see figures 1 and 2). A spur gear
is a two dimensional profile projected into the third
dimension just as a circle is projected to create a
cylinder. A helical gear (see 4.1.9.2) is a two
dimensional geometry that is rotated about its
center as it is projected, forming a gear in the shape
of a spiral. With the concept of helical gears comes
the concept of a normal plane, a plane perpendicular to the flanks of the involute teeth (see figure 3),
that is not the same plane as the transverse plane.
In spur gearing, the normal plane and the
transverse plane are the same. Normal diametral
pitch is related to transverse diametral pitch by the
expression presented in 6.2.1.
In order to control the cost of gear cutting tooling,
the industry standardized on the list of fine--pitch
normal diametral pitches that appears in 4.3 as well
as a very few standard normal profile angles.
Normal diametral pitch hobs can be used to cut both
spur and helical gears simply and inexpensively. It
is important to understand that the design process
that follows in this manual involves selecting a
normal diametral pitch and normal profile angle,
usually based upon performance criteria. Usually
the specifications that appear on gear drawings
contain only parameters in the normal plane
because they relate to the cutter used to make the
gear. However, the true gear action occurs in the
transverse plane and therefore accurate analysis of
a gear design requires analysis of the motion in the
transverse plane. A thorough analysis clearly
contains the elements of the base pitch system that
is the root of involute gearing as described in the
beginning of this manual.
Dividing the number of teeth in the full circumference of a gear by its transverse diametral pitch
results in a quantity called the standard pitch
diameter (see 6.2.2). The standard pitch circle
intersects the involute at the point where the normal
pressure angle is equal to the normal profile angle of
the cutter used to generate the tooth (see 4.1.9.1).
The tooth thickness (see 6.6) is generally specified
at the standard pitch diameter. The following
clauses of this design manual will explain the
AGMA 917--B97
diametral pitch system and demonstrate its use.

The user should always remember, despite numerous references to circles and diameters, that
involute gears operate along a line of action, not
around a circle!
4.1.9.1 Pressure angle, operating pressure
angle and profile angle
Many terms in common usage possess multiple
meanings. Pressure angle is one of those terms.
Figure 13 shows one usage of pressure angle, as a
variable angle related to an arbitrary point on an
involute curve. The pressure angle in this case is
different for each point on an involute curve. This
meaning is useful only when describing the
geometric aspects of a single involute gear tooth.
Point
Involute
Pressure
angle
A
Polar
angle
B
rb
O
Figure 13 -- Involute polar angle

Figure 6 illustrates that the term operating pressure
angle, pt , is the angle between the line of action
and perpendicular to the line connecting the centers
of two meshed gears.
Figure 7 demonstrates that the operating pressure
angle changes when the center distance changes
for the same pair of base circles.
As was previously noted, a pressure angle must be
specified in conjunction with a diametral pitch in
order to uniquely specify the base pitch of an
involute gear system. Confusion sometimes accompanies this seemingly contradictory concept of
a specified constant pressure angle that changes
with center distance.
13
AGMA 917--B97
The AGMA has attempted to reduce the confusion

by introducing the term profile angle to gearing
nomenclature. In general, the profile angle is the
angle between a line tangent to an involute gear
tooth at the point of intersection between the
involute and the standard pitch diameter and a
radial line emanating from the center of the gear.
This profile angle can be found elsewhere in gear
geometry. Referring again to figure 13, the profile
angle has the same magnitude as the pressure
angle if the point on the involute is the intersection of
the standard pitch diameter and a radial line from
the center of the base circle. The profile angle at the
standard pitch diameter has the same magnitude as
the flank angle of a rack type cutter used to make a
gear by the generating process. See Colbourne [1]
for a description of gear generating.
As was noted previously, a gear generating hob
specification must include both a diametral pitch
and a pressure angle. This usage of pressure angle
is where employing the term profile angle is
encouraged. Profile angle should be used to
describe individual gears, gears not in mesh.
Pressure angle should be used to describe the
angle of the line of action when gears are in mesh.
The pressure angle is equal to the profile angle
when gears are operating in mesh on their standard
center distance. If they are mounted at a center
distance different from standard, then their operating pressure angle is numerically different from the
profile angle to which they were manufactured.
4.1.9.2 Helix angle and helical gears
A helical gear has the appearance of a spur gear
that has been twisted about its axis. In actual
manufacture, helical gears can be cut on hobbing
machines just as are spur gears. The same normal
diametral pitch hobs are used. The hobbing
machine is set up with a specific feed rate in relation
to the index ratio of the work spindle in order to cut
the desired spiral (see Townsend [5] for more
information on the hobbing process).
The helix angle, , of a helical gear is the angle
between the axis of the gear and the element of the
flank that intersects the standard pitch cylinder, see
figure 14.
Helical gears are normally specified by proportions
measured in the plane normal to the gear tooth
14
profile. This is primarily because normal profile

dimensions correspond directly to the hob used to
manufacture the gear. However, during the design
of the gear, much attention is paid to the proportions
of the gear measured in the transverse plane. The
transverse plane is the plane perpendicular to the
axis of the gear.
Helix
Tooth
Helix angle
Axis
Figure 14 -- Helix angle

Clause 6 presents the equations necessary to
calculate gear parameters for both spur and helical
gears. It is clear from the equations presented that
a spur gear is mathematically a helical gear with
zero helix angle.
4.1.10 The involute and the rack
A driving involute can also drive a rack. Note that as
the diameter of a base circle grows, the curvature of
the involute described by it increases in radius. A
rack may be considered a sector of a gear of infinite
radius. Thus the rack teeth of the involute system of
gearing have straight line profile elements,
(straight--sided teeth). This has great advantages
in the manufacture of gears in that many of the types
of tools used to cut and finish gear teeth can employ
straight line elements. Some hobs, rack cutters and
grinding wheels are examples. Such straight--line
elements can be more accurately formed and
inspected than can complex curves. A single hob
can be used to cut gears of any number of teeth, all
of which will mesh properly with each other.
4.1.11 The involute function
Figure 13 shows the geometric features of the
involute curve. Previous portions of clause 4
described the involute as the path of a point on a belt
as it is transported around a pulley system. The
involute is also the curve produced, as a trace on a
fixed background, by a point (knot) in a taut string
unwound from a circle. The circle is the base circle
of the involute, just as was the pulley in the previous

visualization.
The equation of the involute function is:
= inv = tan --
where
...(1)
is pressure angle at a given radius

measured in radians.
4.2 Pitch
Pitch is the distance between a point on one tooth
and the corresponding point on an adjacent tooth. It
is a dimension measured along a curve in the
transverse, normal or axial directions. The use of
the single word pitch without qualification may be
confusing, and for this reason specific designations
are preferred, e.g., transverse circular pitch, normal
base pitch, axial pitch. See figure 15.
AGMA 917--B97
4.2.4 Normal circular pitch

Normal circular pitch is the circular pitch in the
normal plane and also the length of arc along the
normal pitch helix between helical teeth. See figure
16.
4.2.5 Axial pitch
Axial pitch is the linear pitch in an axial plane and in a
pitch surface. In helical gears, axial pitch has the
same value at all diameters. The term axial pitch is
preferred to the term linear pitch.
Transverse
circular pitch
Axial pitch
Pitch
Axis
Normal
circular pitch
Circular pitch
Figure 16 -- Tooth pitches
4.2.6 Base pitch
Figure 15 -- Pitch
4.2.1 Circular pitch
Circular pitch is the arc distance along the pitch
circle between corresponding profiles of adjacent
teeth. See figure 15.
Base pitch in an involute gear is the pitch on the

base circle or along the line of action. Corresponding sides of involute gear teeth are parallel curves
and the base pitch is the constant and fundamental
distance between them along a common normal in a
plane of rotation, (transverse plane). See figures 17
and 18.
Base pitch
4.2.2 Transverse circular pitch

Transverse circular pitch is the circular pitch in the
transverse plane. See figure 16.
Circular pitch
4.2.3 Transverse operating circular pitch

Transverse operating circular pitch is the arc
distance along the operating pitch circle between
corresponding profiles of adjacent teeth. This
parameter is used to calculate backlash because
backlash is defined on the operating pitch
diameters.
Base pitch
Base circle
Base tangent
Figure 17 -- Principal pitches
15
AGMA 917--B97
4.2.7 Normal base pitch

Normal base pitch in an involute helical gear is the
base pitch in the normal plane. It is the normal
distance between parallel helical involute surfaces
on the plane of action in the normal plane, and it is
the length of arc on the normal base helix. It is a
constant distance in any helical involute gear. See
figure 18.
Helical rack
Base
pitch
and racks, have a basic tooth form which, when

cutting a gear on a generating machine, produces
involute gear teeth. Although a hob, for example,
can generate gears having any desired number of
teeth, it can only produce a single normal diametral
pitch and normal profile angle. Since several tools
may be needed to produce a job lot of gears, it is
generally desirable to select a diametral pitch for
which most gear shops are likely to have tooling.
This may avoid the need to purchase special
tooling. Thus the following have become recommended normal diametral pitches, Pnd .
20, 24, 32, 40, 48, 64, 72, 80, 96, 120
4.4 Center distance
Normal
base pitch
Axial base pitch
Figure 18 -- Base pitch relationships
4.2.8 Axial base pitch
Axial base pitch is the base pitch of helical involute
tooth surfaces in an axial plane. See figure 18.
4.3 Diametral pitch
Diametral pitch is not a pitch in the same sense as
the preceding pitches. It represents the size of the
tooth. The larger the numeric value of the diametral
pitch, the smaller the size of the gear tooth.
Diametral pitch is related to circular pitch by the
following:
Diametral pitch =
Circular Pitch
It is customary to discuss the size of a given gear in
terms of its diametral pitch rather than its circular
pitch.
Many fine--pitch gears are produced by means of
generating tooling. Even gears produced by
molding, casting or stamping are intended to have
teeth which are the same as if they were generated.
Gear generating tools, such as hobs, shaper cutters
16
The center distance, the distance between two gear

axes, is a basic dimension for meshing a pair of
gears. It is controlled by the frame (casing or gear
box) and the bearings that support the gears. Two
values are of particular interest: the minimum
operating center distance which can occur when the
sum of all tolerances accumulates to the minimum
functional value, and the maximum operating
center distance which occurs when all tolerances
act in the opposite sense. Items that comprise the
center distance tolerance in designs using rolling
element bearings include such factors as:
-- The maximum and minimum bearing center
distances;
-- The clearances between the outer races and
the bores;
--
Outer race concentricity;
--
Bearing internal clearance;
--
Eccentricity of the inner race;
-- Concentricity between race journals and the

gear bores;
--
Clearance between inner races and shafts.
A common error committed by the novice designer

is to underestimate the buildup of tolerances and
the effect the resulting change in center distance
has on the operation of gearing. This effect can be
significant in fine--pitch gearing. For example, a
total center distance tolerance of 0.010 may be
trivial in a system using 20 diametral pitch gearing
and completely unacceptable in a system using 64
diametral pitch gears. Careful analysis and control
of center distance variation is imperative in fine-pitch gear design.
An outstanding advantage of the involute as a

profile shape is its ability to transmit uniform motion
independently of center distance (see 4.1.1.). It is
the only curve suitable for gearing that has this
characteristic. This gives the designer great
freedom in creating a spur or helical gear design.
Center distance is generally categorized as standard, enlarged or reduced. Enlarged and reduced
center distance designs are achieved by making
design adjustments to tooth proportions. See
clause 6 for a detailed discussion of center distance.
AGMA 917--B97
the designer to a set of gears that meet the load

requirements (bending strength and surface
strength) and the contact ratio requirements of the
system. In the detailed design stage, the tooth
geometry is determined in order to optimize the gear
system for its application. The specific tooth
modifications will vary depending upon the type of
gear system being designed. At this level of the
design, it is helpful to consult an experienced gear
designer and/or the gear design references listed at
the end of this information sheet.
5.2 Limitations
5 Application considerations
5.1 Principal gear functions
Gears are used to transmit power and/or motion
from one shaft to another. If their principal function
is to transmit power, they are called power gears. If
their principal function is to transmit smooth motion,
they are called smooth motion gears. If their
principal function requires minimal backlash, they
are called zero backlash gears. Since there are
usually major differences in the basic requirements
of each of these systems, the design emphasis will
usually be different for each system. It is, however,
quite possible that each of these systems may have
some design requirements that are common to
each other (example: must be quiet and have
adequate load capacity).
A gear design flow chart is shown in figure 19. This
flowchart should help guide the designer through
the various steps of the design process. It is not
intended to be a detailed road map for every gear
application, but rather a general overview of some
of the important things to consider when designing
gear systems. As shown in the flowchart, at the first
step of the design process the functional requirements of the system should be determined. This will
establish a perspective as one proceeds through
the rest of the steps. Many design decisions will
have to be made as one proceeds through the
different phases of design. Realizing the functional
requirements of the system early will help in making
those decisions. After determining the type of gear
system that is required, one proceeds down through
the column on the flowchart. The steps up to and
including the preliminary design analysis help guide
The information in this design manual is meant to

serve only as a guide to the designer of fine--pitch
gear systems. It is not implied that using the
procedures will necessarily result in gears that will
meet the requirements in every application. It
remains the responsibility of the individual designer
to properly evaluate the conditions in the particular
application and to make use of prior experience
and/or proper testing to confirm the suitability of the
design.
5.3 Design parameters
Gear design is a decidedly iterative process of
estimation, followed by calculation of gear parameters, followed by analysis to assure no engineering
rules have been violated. The initial estimate is
usually difficult for the novice gear designer. Flow
charts (figures 19 and 20) attempt to provide insight
into the complexity of how to choose diametral
pitch, number of teeth, and number of stages in the
gear system.
The designer must determine the external boundary conditions before the type of gears (spur,
helical, worm etc.) can be determined. This clause
lists some of the external boundary conditions that
should be considered for gear applications. It is the
responsibility of the designer to consider any other
boundary conditions that may be relevant to the
particular application.
5.3.1 Loads
To determine the loading conditions on the gear
system, consideration should be given to the
following:
-- The power rating of the prime mover, its
overload potential, and the uniformity of its output
torque. The overload potential is especially
important if there is a possibility of a lockup in the
system which could cause the motor to stall.
17
AGMA 917--B97
-- The output load requirement, including the

following:
-- Normal output load;
-- Peak loads and their duration;
--
Frequency of severe loading or stalling;
-- Inertia loads arising from acceleration or

deceleration of the complete system.
Determine
functional
requirements
Design parameters
Loads
Space constraints
Life requirements
Speed ratio
Operating environment
Preliminary
estimate of gears
Bending stress
Contact stress
Pitting resistance
Fatigue strength
Lubrication
Number of teeth
Diametral pitch
Face width
Helix angle
Material/heat treat
Quality class
Is analysis OK
False
True
Gear Assembly
Mounting/alignment
Housing strength/deflection
Environment
Power gear
systems
Smooth motion
gear systems
Position control
gear systems
Detailed design
optimize considering:
Detailed design
Detailed design
Minimum volume
Fatigue strength
Pitting resistance
Noise
Cost
Gear specifications
Data block
Blank design
Transmission error
Contact ratio
Tooth modification
Minimum volume
Fatigue strength
Noise
Cost
Gear specifications
Data block
Blank design
Figure 19 -- Gear design flowchart
18
Total angular backlash

Contact ratio
Transmission error
Tooth modification
Minimum volume
Fatigue strength
Noise
Cost
Gear specifications
Data block
Blank design
Given:
Speed requirements
Center distance
Assume 20 deg. profile angle
at this stage
Torque requirement
AGMA 917--B97
Determine preliminary estimate of:
Determine gear train ratio

from speed requirements
Add another stage to

the gear train if possible
yes
Is stage ratio > 5
no
Calculate operating pitch diameters
from center distance & gear ratio
Number of teeth in pinion

Number of teeth in gear
Diametral pitch
Helix angle
Face width
Refer to clause 6.3.5
Calculate tangential force from torque

and operating pitch diameter
Spur gears: assume geometry factor = 0.4
Helical gears: assume geometry factor = 0.45
Refer to clause 6.9.5
Preliminary estimate of material based on:

Duty cycle
Cost considerations
Lubrication requirements
Noise considerations
Determine failure criteria and the appropriate
mechanical properties based on application
Refer to clause 5.3
If space constraints permit

F = 0.5 DP
can be used for a starting point
Spur gear: Helix angle = 0.0
Helical gear: Initial choice for helix angle = 15 -- 20 deg.
Pd min.
(Coarsest)
Pd max.
Determine limits for diametral pitch Pd
Calculate Pd (coarsest) from

operating pitch diameter such that
pinion gear has min. of 9 teeth
See clause 6.9.5
(Finest)
Calculate Pd (max., finest) using

Lewis Form approximation for
bending stress
Is Pd max < Pd min
Change material
and/or
heat treatment
False
True
Pd min
20 24
Pd max
32
40
48
64
72
Coarsest
Advantages
80
120
96
Finest
Lowest bending stress

Fewest teeth in pinion & gear
Min. change in contact ratio as center distance
deviates from nominal
Disadvantages
Lowest contact ratio at nominal center distance

Addendum modifications should be done to pinions
with fewer than 24 teeth to eliminate undercutting
Lowest tooth mesh frequency
Advantages
Highest contact ratio at nominal center

distance
Disadvantages
Highest bending stress

Max. change in contact ratio as center distance
deviates from normal
Highest tooth mesh frequency
Select a diametral pitch between the maximum and mimimum limits based on the specific
application and the general guidelines previously mentioned
Calculate NP and NG from operating pitch diameter and diametral pitch
refer to clause 6
Figure 20 -- Preliminary design flowchart
19
AGMA 917--B97
-- The additional loads induced by couplings or

other interconnections between the prime
mover, gear train and driven load;
-- Some gear systems have multiple outputs
driven by a common input. In these cases, the
torque required at each output is determined first.
Torque requirements at each stage are then calculated being careful to add together the torques
at the gear interfaces where the torque path
splits. In this way, the torque is determined for
each gear in the path back to the prime mover.
An example of a torque split system can be seen
in figure 21.
From this analysis, a basic design load for the gear
system as well as suitable overload factors can be
determined. Factors to account for desired life and
other safety factors will also result from this
analysis. The actual loads may vary from the
predicted loads due to inefficiencies of the system.
It is recommended that torque measurements be
made to verify actual loads before the design is
finalized. Clause 15 provides more information on
recommended testing procedures.
5.3.2 Gear train ratio

The gear train ratio is the ratio of the angular speed
of the first (driving) shaft to the last (driven) shaft. A
gear train may consist of several pairs of meshing
gears. Each meshing gear pair consists of a driver
and a driven gear. The ratio of the speed of the
driver to the driven gear, of any pair, is called the
stage ratio. Since there must be an integer number
of teeth on each gear, the achievable stage ratios
are limited to ratios of integer numbers. The gear
train ratio is equal to the product of the individual
stage ratios. The required gear ratio will often
influence the selection of the type of gearing, since
each type of gearing has its own stage ratio
limitations.
Torque required
output 1
n
Gear Ratio : (m Gt ) = n 1
L
where:
n1 is rotation speed of first shaft;
nL is rotation speed of last shaft.
mGt = (mGs1) (mGs2) (mGs3) (mGs4)...(mGs(L-- 1) )
-- When designing parallel axis spur or helical

gear systems, a stage ratio limitation of 5:1
should be considered for the following reasons:
--
To minimize the specific sliding ratios;
--
To increase the mesh efficiency;
-- To achieve more balanced strength

between pinion and gear;
--
Torque required
output 2
Motor
pinion
Figure 21 -- Torque split
20
To minimize the size of the gear.
The 5:1 stage ratio limitation is a general rule of

thumb, not an exact design criterion. Addendum
modifications (profile shift modifications) can be
made to the pinion to improve its strength and
reduce its specific sliding ratio. If the gear
designer/engineer requires stage ratios greater
than 5:1, more detailed information can be found
in gear texts such as On the Geometry of
External Spur Gears by T.W. Khiralla [2]. An experienced gear engineer should be consulted if
the required stage ratio is significantly higher
than 5:1.
AGMA 917--B97
Idler gears, which are both driver and driven

gears, do not affect the overall gear train ratio.
They do, however, change the direction of rotation. See figure 23 for an illustration of a gear
system with an idler gear.
5.3.3 Space constraints
Space constraints can also limit the type of gear that
can be used. Some design parameters that can be
considered as space constraints are listed below.
The examples listed are not meant to be limiting or
all inclusive.
-- A speed decreasing gear train should consist

of only speed decreasing gear stages when possible. (The driven gear has more teeth than the
driver.) An example can be seen in figure 22.
-- A speed increasing gear train should consist
of only speed increasing gear stages when possible. (The driven gear has less teeth than the
driver.) Care must be taken when designing
speed increasing drives for maximum recess
action. The driven gear should have a sufficient
number of teeth so that it can be modified (reduced) without causing excessive undercutting.
Reference [2] contains more information about
recess action gearing.
NOTE: It is sometimes necessary to mix speed decreasing and speed increasing gear stages to achieve
an exact ratio. However, this will usually result in larger
gears and/or more gear stages.
60T
20T
--
The dimensions of the design volume;
-- Existing center distance specifications (for

example: several gear systems that can be
assembled into a common gear box to provide
different gear ratios);
-- Making use of the same gear in more than
one location in the gear system;
-- Reverted gear trains, where the output shaft
and input shaft are co--linear;
-- The shaft angle orientation. The possibilities
are parallel, nonparallel and intersecting, and
nonparallel and nonintersecting. Each of these
conditions will determine the type of gearing that
can be used. The type of gearing that is appropriate for each of these conditions is discussed in
5.4.
60T
20T
Driver
n1, rpm ccw
Stage Ratio
cw
m Gs nn1 = N 2
N1
2
where:
n1 is speed of driving shaft, rpm;
n2 is speed of driven shaft, rpm;
N1 is number of teeth in driving gear of stage;
N2 is number of teeth in driven gear of stage.
Driven
n2, rpm ccw
Figure 22 -- Speed decreasing gear system
21
AGMA 917--B97
60T
40T
20T
Driver
n1 rpm ccw
Idler Gear
Gear Train Ratio = 40 60 = 60 = 3

20 40
20
Driven
nL rpm ccw
Figure 23 -- Gear train with idler gear

5.3.4 Operating environment
The operating environment may affect the
performance of the gear system. It is extremely
important that the gear system be designed to
operate in the environment in which it will be used.
Some types of gears have advantages over others
in certain operating environments. Some important
considerations of the operating environment are:
-- Enclosed gear system. The gears are
mounted in an enclosed gearbox. They are usually protected from external contamination and
can often be oil lubricated and cooled. This is an
ideal situation for gears to operate;
-- Open gear system. The gears are mounted
on shafts in the open. They often experience outside contamination such as:
--
Moisture;
--
Corrosive environment;
--
Abrasive wear debris;
--
Dirt, dust, etc.
It is important to understand the type of

contamination that can be present so that the
proper lubrication can be specified. Lubrication
is often a necessity, but occasionally, in open
gear system applications, external oils or grease
can actually increase the wear rate by trapping
contamination or wear debris in the gear mesh. It
is often necessary to increase the operating
22
backlash to allow space for the debris. It is

recommended that gear systems operate in an
enclosure if the possibility of outside
contamination exists;
-- Operating temperature of the gears.
Overlooking or underestimating the effect of
temperature on a gear system can have disastrous consequences. Aircraft gearing must often
function from --50F to +300F. In general, it is
good practice to consider the possibility of lubrication and cooling for any gear system.
Elevated temperatures can contribute to premature failure, especially when the gears are made
from plastics.
The mechanical properties of plastics (tensile
strength, flexural modulus, fatigue strength,
etc.), vary with temperature and in some plastics
are significantly lower at elevated temperatures.
When plastic gears are operated at elevated
temperatures, the designer must account for the
effects the elevated temperatures have on the
mechanical properties. In many applications
plastic gears are run without external lubrication.
The higher coefficient of friction of unlubricated
plastic gear mesh combinations, combined with
high speed operation can generate enough heat
to significantly increase the operating mesh
temperature. Additionally, thermal expansion
coefficients of plastics (K10--5), are typically an
order of magnitude higher than of steels
(K10--6). Therefore, adequate backlash must
be provided to compensate for the increased

tooth thickness at elevated temperature. Some
applications require special tooth compensation
when manufactured at ambient temperature to
ensure the tooth form will be correct at the
elevated operating temperature.
The gear designer is advised to fully research the
mechanical properties of any plastic considered
for a gear design, particularly with respect to the
plastics performance at elevated or reduced
temperature.
5.3.5 Life requirements
The life requirements of the gear system will
determine the allowable bending stress (fatigue),
and surface stress (pitting) of each of the gears in
the system. Some factors that can affect the
allowable stresses are:
-- Number of load cycles. This will be used to
determine the bending fatigue and surface
fatigue endurance limits for the material being
used. Note that non--ferrous materials, such as
plastics, bronze, etc., do not have true endurance limits. The designer must know the
allowable stress for the number of cycles
required. Note that the number of load cycles is
not always the same as the number of revolutions; an idler gear is an example;
-- Reliability of the gear system. The higher
the required reliability of the system, the lower
the allowable stresses. Note that the reliability of
the system includes the reliability of the gear
train, which is the product of the reliabilities of the
successive gear stages. The reliabilities of the
stages need not be equal, and in general are not
equal; e.g., the reliability of a very low speed
stage could be higher than that of a very high
speed stage;
-- One--way vs. reversing drive. The endurance limit for a reversing drive is lower than for a
one--way drive because the root of the tooth
experiences complete stress reversals. Complete stress reversals also occur in idler gears,
even when they are driven in one direction,
because opposite tooth flanks transmit the load.
The designer can obtain more information about
the effect of complete stress reversals on fatigue
life by consulting various machine design texts;
AGMA 917--B97
-- Operating temperature. The operating

temperature can affect the life of a gear system
by reducing the tensile strength and fatigue
strength of the material. The relationship between temperature and mechanical properties
need to be known for the specific material being
used. This information should be obtained from
the material supplier when possible. Testing
should be done prior to finalizing the design;
-- Miscellaneous effects. Many other factors
can affect the life of the gear system. For example, the type and accuracy of the mounting could
affect the load distribution across the gear face,
which may increase the bending stress. The
designer must consider any other factors which
may adversely affect the life of the gear system.
5.3.6 Lubrication requirements
The method of lubrication depends on the
environment in which the gears will operate. Some
designs provide an oil--tight enclosure with suitable
shaft seals. This is generally a very desirable way to
provide lubrication for gears. In other designs the
surrounding equipment may be absolutely intolerant of external lubrication in any form. Gears in this
type of design must be designed to operate in a
non--lubricated condition. The lubrication requirements of the gear system need to be understood
before the type of gear and gear material can be
determined. There are three general categories of
gear lubricants:
liquids (oils), semisolids
(grease), and solids (graphite, molybdenum
disulphide, polytetrafluoroethylene (PTFE)). Oil
and grease are applied externally; however, some
solid lubricants such as PTFE can be molded into a
plastic gear at the time of manufacture.
Oil lubrication of gears has the following
advantages over grease:
-- Oil has a greater range of operating speeds
and temperatures;
-- Oil is more effective in conducting heat from
the gear teeth to the housing;
-- Oil will saturate all areas of contact and is
more effective in removing wear debris;
--
Easier to fill and drain a reservoir using oil.
23
AGMA 917--B97
Grease lubrication of gears has the following

advantages over oil:
Pinion
-- Under heavy loads and slow speeds, grease

can provide a lubricating film whereas under the
same conditions oil cannot;
-- Grease has much slower spreading and
evaporation rates than oil;
-- Grease is easier to seal from leakage in
enclosed gear drives.
Gear
Some features of solid lubricants that make

them suitable for special applications are:
-- Solids can be used in open gearing where oil
or grease would collect contaminates;
Rack
-- Solids resist being squeezed out of gear

teeth under high pressure;
--
Figure 24 -- External spur gears
Solids do not evaporate;
-- Solids can be used in applications where oil

or grease contamination can not be tolerated.
Refer to clause 13 of this Design Manual and
ANSI/AGMA 9005--D94 for more information on
lubrication of gear systems.
5.4 Types of gearing
After the external boundary conditions are known,
the selection of the type of gearing can be made.
Each type of gearing has unique features that make
it more suitable for some applications and less
suitable for others. The following paragraphs give a
summary of each and list some advantages and
disadvantages. Several references, such as Analytical Mechanics of Gears, by Earle Buckingham
[3], contain more detailed information about the
various types of gearing. Brief descriptions are
given for each type of gearing in this clause, but only
spur and helical gear systems are covered in this
Design Manual.
5.4.1 Gearing on parallel axes
5.4.1.1 Spur gears
A spur gear is a gear cut from a cylindrical blank,
with teeth parallel to the gear axis. Spur gears can
be either external, with the teeth projecting away
from the center, or internal, with the teeth projecting
toward the center. A set of external spur gears can
be seen in figure 24. An internal spur gear with an
external spur pinion can be seen in figure 25.
24
Figure 25 -- Internal spur gear and external

spur pinion
Advantages of spur gears:

--
Transmit no axial (thrust) force;
--
Relatively low manufacturing cost;
--
Relatively low assembly costs.
Disadvantages of spur gears:

-- Lower load carrying capacity than helical
gears of same size;
-- Relatively low contact ratio (usually less than
2.0);
-- Larger change in mesh stiffness during a
complete mesh cycle;
--
Often noisier than helical gears.
5.4.1.2 Helical gears

A helical gear is a gear cut from a cylindrical blank,
with teeth that are on helices about the axis of the
gear. The twist direction of the helix is designated
as right hand if it twists clockwise when viewed
along its axis. The twist direction is designated as
left hand if it twists counter--clockwise when viewed
along its axis. External helical gear pairs are of
opposite hands, while external -- internal helical
gear pairs are of the same hand. Figure 26 shows a
set of external helical gears.
AGMA 917--B97
Disadvantages of helical gears:

--
Transmit axial (thrust) forces;
--
Somewhat higher manufacturing cost.
5.4.2 Gearing on nonparallel shafts

5.4.2.1 Face gearing
Face gearing consists of a spur or helical pinion and
a disk--like gear. The axes are usually at right
angles and may be intersecting or nonintersecting.
The face gear resembles a rack wrapped into a
circle. Figure 27 shows a spur pinion and face gear
with right angle intersecting axes. There are some
practical limitations to the allowable stage ratios
when using face gears. Generally ratios less than
1.5:1 should not be attempted with face gears. This
type of gearing is not included in this Design
Manual. For further information, see AGMA
203.03, Fine--Pitch On--Center Face Gears for
20--Degree Involute Spur Pinions.
Offset
Pinion on center
Pinion off center
Figure 27 -- Spur pinion and face gear

Figure 26 -- External helical gears
Advantages of helical gears:
-- Higher load carrying capacity than spur
gears of same size;
--
Higher contact ratio can be obtained;
-- Less change in mesh stiffness during a

complete mesh cycle;
-- Generally quieter than spur gears if the
active face width is greater than one axial pitch.
5.4.2.2 Bevel gearing

Bevel gearing is conical in form. Bevel gears can
carry higher loads than crossed helical gears and
face gears of the same size. The center distance
and shaft alignments in two directions must be very
accurate to ensure proper load distribution across
the tooth. The following tooth element shapes can
be used:
-- Straight tooth elements which if extended,
would pass through the intersection point of the
axes (see figure 28a);
25
AGMA 917--B97
-- ZEROLR curved tooth elements which follow the same general direction as straight teeth
(see figure 28b).
-- Spiral tooth elements, which are curved
and oblique (see figure 28c).
These types of gearing are not included in this

Design Manual. For further information, see
ANSI/AGMA 2005--B88, Design Manual for Bevel
Gears.
5.4.3 Gearing on nonparallel, nonintersecting
shafts
5.4.3.1 Crossed--axis helical gearing (formerly
called spiral gearing)
Straight bevel gears
Skew bevel gears
(a) -- Straight tooth bevel elements
Spiral bevel
gears
Crossed helical gears are helical gears mounted on

axes which are skewed. The teeth may be of the
same or opposite hands, and the helix angles can
be equal or unequal. The use of crossed helical
gears should be limited to low torque applications
because the teeth make point contact with each
other. It is recommended that an experienced gear
designer be consulted when designing gear systems using crossed helical gears. Crossed--axes
gearing is not included in this Design Manual.
Figure 29 shows a set of crossed helical gears.
ZEROL bevel
gears
(b) -- Curved tooth bevel elements

Figure 29 -- Crossed helical gears
Left--hand
5.4.3.2 Worm gearing
Right--hand
Spiral bevel gears
Spiral bevel pinions
Left--hand
Right--hand
(c) -- Spiral tooth bevel elements

Figure 28 -- Bevel gearnig
26
This type of gearing uses a worm to drive a

wormgear. The worm is cylindrical in form with the
teeth shaped like screw threads. The worm can
have one or more teeth (also called starts). The
designer should be aware that the torque ratio is not
equal to the reduction ratio in worm gearing, as it is
in parallel axis spur and helical gear systems. This
is because the pitch diameters are not in the same
ratio as the number of teeth. The efficiencies of
worm drives are dependent upon the coefficient of
friction between the worm and wormgear as well as
the lead angle of the worm (helix angle of wormgear
for 90 degree shaft angles). This type of gearing is
not covered by this Design Manual. Information on

fine--pitch wormgearing may be found in ANSI/
AGMA 6034--B92. Figure 30 shows a worm and
wormgear set.
Cylindrical
worm
Enveloping
wormgear
Cylindrical
(non--enveloping)
wormgear
Figure 30 -- Worm and wormgear set

5.4.3.3 Rotary to linear motion gearing
This type of gearing is made up of a rack and pinion
and is used to convert rotary motion into linear
motion, or vice--versa. When two gears mesh, the
smaller of the two is referred to as the pinion, and
the larger is usually referred to as the gear. Any
gear meshed with a rack is considered smaller than
the rack, since the rack is part of a gear with an
infinite number of teeth (it has an infinite pitch
diameter). Therefore, it is common to speak of a
rack and pinion. A spur rack, which has its teeth at
right angles to the direction of its motion, is used
with a spur pinion (see figure 31). A helical rack,
which has its teeth oblique to the direction of its
motion, is used with a helical pinion.
Figure 31-- Spur rack and pinion
AGMA 917--B97
5.5 Preliminary estimate of gears

After the external boundary conditions are known
and the type of gearing has been determined, a
preliminary estimate of the size of the gears can be
made. The amount of power or torque that a gear
can transmit is a function of the size of the gears,
their type and the materials from which they are
made. The size of each gear is defined by its
diametral pitch, number of teeth and face width.
The preliminary design is only the first step of the
total design process. A design should not be
considered complete until it has been analyzed and
tested. In some design problems, a preliminary
design is needed in order to establish some size
requirements for the complete equipment. Such a
preliminary design may not have the benefit of
adequate knowledge of loads or other constraints
imposed by the final design of the equipment. In
these cases, the preliminary design must be
reviewed and revised in light of new information.
5.5.1 Number of teeth
The number of teeth selected for each member
(pinion and gear) will have a considerable effect on
the performance of the gears in service. The
following should be considered when selecting the
numbers of teeth:
-- Integers. The number of teeth in each gear
must be an integer. This restricts the actual ratio
that a given pair of gears may have;
-- Undercutting. Gears with less than 24
teeth (for 20 degree profile angles) usually
require addendum modifications (profile shift) to
avoid undercut or undesirable contact close to
the base circle. Undercutting reduces the tooths
strength and often reduces the contact ratio. See
clause 6 for additional information on calculating
contact ratio for gears with undercut. Contact
close to the base circle increases the ratio of sliding to rolling which reduces the efficiency of the
mesh;
-- Hunting tooth action. In some gear applications, it may be desirable to provide hunting
tooth action; i.e., using gears whose tooth
numbers have a greatest common factor of 1.
This technique minimizes the number of times a
specific pair of teeth mesh with each other. This
is done to reduce repeatability of individual
meshes and promote uniform wear.
27
AGMA 917--B97
lowers the highest point of single tooth

contact so that when a single tooth is
carrying the load it occurs lower on the tooth
flank. This reduces the bending moment on
the tooth which reduces the deflection.
5.5.2 Diametral pitch

There are several factors which need to be
considered when selecting the diametral pitch for
gear systems. Some of these factors are listed
below:
NOTE: These comparisons are valid for gears of a
constant base diameter (i.e., if the diametral pitch is
doubled, the number of teeth is doubled so that the
base circle diameter remains the same).
--
Load carrying capacity (beam strength).
It is important to remember that a gear tooth acts

as a cantilever beam. It has a decidedly non-uniform cross section and its load point varies,
but when loaded it deflects like a cantilever
beam;
-- The bending stress in the root area of the
tooth is approximately proportional to diametral pitch (e.g., if the diametral pitch is
doubled, the bending stress is doubled). The
reason it is not exactly proportional to
diametral pitch is that a change in contact
ratio also changes the radius of the highest
point of single tooth contact;
-- Even though the mesh stiffness
equation shows that the mesh stiffness is
independent of diametral pitch, the mesh
stiffness actually increases slightly as the
diametral pitch is increased. This occurs
because the contact ratio is slightly higher for
the finer pitch gears. The higher contact ratio
--
Contact ratio.
-- The contact ratio increases as the
diametral pitch increases (at the nominal
center distance);
-- The tolerance on the center distance, to
maintain an acceptable contact ratio, must
decrease as the diametral pitch increases to
maintain the same contact ratio. Figure 32
also shows contact ratio as a function of
center distance deviation for various
diametral pitches.
-- To fit the design envelope (pitch diameters

of gears).
After the number of teeth have been chosen to
give the required ratio, the diametral pitch can be
determined by knowing the allowable size of the
pitch diameters. The designer must ensure that
the gears have adequate load carrying capacity
when determining diametral pitch in this way.
The gear industry has standardized on several
diametral pitches. Although it is sometimes
necessary to design gears with nonstandard diametral pitches, it is usually recommended that
standard diametral pitches be used when
possible because tooling and inspection gears
are readily available.
2.0
Transverse contact ratio
1:1 Ratio
5:1 Ratio
120 DP
80 DP
40 DP
20 DP
1.0
Center distance deviation
NOTE: The numbers on the chart were generated for a specific center distance.
Figure 32 -- Contact ratio vs. center distance deviation for 20 degree profile angle gears
28
The standard fine--pitch normal diametral pitches

are:
AGMA 917--B97
contact ratio be at least 2.0 to achieve good

helical overlap.
5.5.3 Face width
-- The transverse contact ratio is defined

the same as for spur gears, using
transverse dimensions;
Face width is the length of the teeth in the axial

plane. The following factors should be considered
when determining the face width of gears:
-- The face contact ratio is a measurement of the helical overlap and is equal to
the face width divided by the axial pitch.
20, 24, 32, 40, 48, 64, 72, 80, 96, 120
NOTE: The following comparisons are made assuming there is uniform load distribution across the face
width.
-- Load carrying capacity. The bending

stress is inversely proportional to face width.
Therefore, when everything else is held
constant, as the face width is doubled the bending stress is halved. Note that when calculating
load carrying capacity the active face width
should be used. The active face width is the
actual face overlap between the pinion and gear.
Axial misalignment of the gears and rounding or
chamfering of the tooth may reduce the active
face width to a value less than the minimum face
width of the pair. Usually the face width of one of
the gears is made wider to allow for axial
misalignment and still maintain full face contact.
The increase in face width is usually made to the
pinion because it requires less additional material. For additional information on the load rating of
gear teeth, refer to ANSI/AGMA 2001--C95,
Fundamental Rating Factors and Calculation
Methods for Involute Spur and Helical Gear
Teeth;
-- Face width--to--diameter ratio. To achieve
more uniform tooth contact along the face, the ratio of face width to diameter should usually be
held to below 2.0. If the alignment of the pinion
and gear teeth cannot be closely controlled, the
ratio should be held well below this limiting value.
The alignment is influenced by gear accuracy,
housing accuracy, shaft deflections, gear blank
rigidity and the accuracy and rigidity of the gear
mounting features;
--
Contact ratio.
-- Spur Gears: Face width does not affect
the contact ratio of spur gears;
-- Helical Gears: The total contact ratio is
the sum of the transverse contact ratio and
the face contact ratio. If face contact ratio is
less than 1, no advantage is gained from
helical action. It is suggested that the face
The helical overlap is proportional to the face

width for a given diametral pitch and helix angle
(axial pitch). In some applications, it is advantageous to design face width and helix angle
combinations which result in integer face contact
ratios. The resulting contact lines are of constant
length, yielding more uniform mesh stiffness.
Uniform mesh stiffness may reduce vibration
which would result in smoother operation of the
gears.
5.5.4 Helix angle
The following factors should be considered before
choosing a helix angle for a gear system:
-- The higher the helix angle the higher the axial
component (thrust) that is transmitted;
-- The maximum strength of a helical gear occurs between a helix angle of 10 to 20 degrees.
See AGMA 908--B89;
-- The higher the helix angle, the higher the
face contact ratio, for a given face width;
-- The higher the helix angle, the larger the
pitch diameter of the gear, for a given number of
teeth and normal diametral pitch;
-- Index gearing for machine tools. Consult
gear manufacturer for manufacturability;
-- Availability of cutter and helix guide for gear
shaping.
5.5.5 Profile angle
The fine--pitch gear industry has standardized on 20
degree profile angles for most applications. The
use of 14.5 degree profile angle gears has declined
in the past several years due in large part to the
increased
strength
and
reduced
sliding
characteristic of 20 degree profile angle gears.
The contact ratio is higher for 14.5 degree profile
angle gears than for 20 degree profile angle gears of
the same diametral pitch and numbers of teeth. In
some applications this increase in contact ratio
29
AGMA 917--B97
could allow enough additional center distance

deviation to allow for assembly tolerances.
5.5.6 Material and heat treatment
In order to make a preliminary estimate of the size
required for a set of gearing, it is necessary to
establish the general type of material to be used,
such as steel, aluminum, bronze or plastics. In
power gearing, where strength and pitting resistance is of major importance, steel is often selected
because it can be heat treated. In position control
gearing, wear resistance is of prime importance.
The material combinations are chosen to minimize
tooth wear. In some applications there is an added
constraint of operation without lubrication. In these
applications the selection of the materials is very
critical to the reliability of the system. Depending on
the required loads, plastics may be a better choice
than steel when the gears must operate without
external lubricants. For power gear systems, or any
gear system that is transmitting high loads, the heat
treatment is vital to the performance of the system.
The following four general ranges of tooth hardness
may be considered when making a preliminary
design:
--
low through hardened (150 -- 210 BHN);
--
medium through hardened (210 -- 420 BHN);
--
surface hardened (50 -- 62 Rc);
-- through hardened to 40--62 Rc when employing alloy or tool steels, martensitic stainless,
etc.
NOTE: For very fine--pitch gears, care should be taken
so that the case depth from the heat treatment does not
go through the entire tooth. See also clause 10.
5.5.7 Manufacturing method

There are many processes available to manufacture gears. There are many factors which collectively determine the manufacturing process that
should be used. Some of these factors are:
-- Process must be capable of meeting design
quality parameters.
-- Process must make parts for acceptable
cost.
-- Process must be capable of meeting the
schedule demands of the part.
It is important to consider the process to be used
early in the design process because each process
30
has unique design requirements. Refer to clause 11

for more information about the various manufacturing processes.
It is important for the gear designer to understand
that correct gear blank design is essential to gear
quality. A precision gear blank is required to make a
precision gear. See clause 8 for a detailed
discussion of gear blanks.
5.6 Engineering analysis of preliminary design
Once the preliminary design has been determined,
it is necessary to analyze the design to ensure that it
will meet the load capacity and contact ratio
requirements. Refer to clause 6 for the equations
for calculating bending stress, contact stress and
contact ratio. If the analysis of the preliminary
design reveals that it would not be adequate for the
application, the designer should revise the original
design. The cycle of design followed by analysis
should be repeated until the preliminary analysis
verifies that the design is adequate.
This clause contains a brief synopsis of some of the
types of engineering analysis that should be done at
the preliminary design stage. The primary objective
of the preliminary design analysis is to ensure that
the gears have adequate strength for the application. More detailed analysis will be done later during
the detailed design stage (refer to clause 6). At the
detailed design stage the specific tooth geometry
and meshing parameters will be determined to
optimize the design for its particular application.
The reader is also referred to the publications in
annex A for additional information.
5.6.1 Bending stress
Bending stress can be determined using a variety of
techniques. Each method has some advantages
and disadvantages.
-- Lewis form factor. This method treats a
gear tooth as a stubby cantilever beam. Figure
33 maps the geometry of the gear tooth to that of
a cantilever beam. This method usually predicts
conservative (higher) than actual bending stress
for unmodified gears. This method evaluates a
normalized measure called the Lewis form
equation based on the dimensions of the beam.
Machine design text books give a detailed explanation of this method. This method does not
account for load sharing and evaluates stresses
at the root by applying the load at the tip of the
gear tooth.
AGMA 917--B97
W
W
l
t
rf
a
x
(b)
(a)
Figure 33 -- Gear tooth as a simple beam

-- AGMA geometry factor method. This
method is also based on construction. However,
the load is applied at the highest point of single
tooth contact for spur gears and inscribes a parabola inside the gear tooth. Figure 34 illustrates
the construction. The point where the parabola is
tangent to the tooth profile is called the critical
point and is where the maximum stress occurs.
This method evaluates a normalized measure
called the geometry factor which accounts for the
stress concentration at the root of the gear tooth
and the load sharing between the mating gears.
For helical gears, the load is applied at the tip of
the gear tooth. Clause 6 uses this technique to
evaluate bending stresses at the root of the gear
tooth.
C
L Gear tooth
Vertex
Load
Inscribed
parabola
Critical
point
Figure 34 -- Tooth load acting at inscribed

parabola
-- Finite element model. This method uses
the finite element method to model the actual
gear tooth geometry. This method will provide

more accurate results than the Lewis form factor,
especially for modified gear tooth geometry. The
effects of thin rim gears can also be analyzed using this method. The tooth model is usually
loaded at the highest point of single tooth contact
(HPSTC).
-- Boundary element model. This method,
like the finite element method, will provide more
accurate results than the Lewis form factor,
especially for modified gear tooth geometry.
5.6.2 Contact stress
Contact stress occurs when two bodies come in
contact under a force. The involutes of external
gear pairs can be approximated as a cylinder on a
cylinder. The involutes of internal--external gear
pairs can be approximated as a cylinder inside a
cylinder. The Hertzian (contact) stress can be
calculated at any point of contact by knowing the
radius of curvature of each involute at the point of
contact.
The two most common places to calculate contact
stresses are:
-- Pitch diameter. At the pitch diameter the
contact stress is less than the maximum, but it is
the point where pitting usually occurs. Pitting is a
surface fatigue failure due to repetitions of high
contact stress. Pitting occurs near the pitch diameter because the relative sliding between the
pinion and gear changes direction as the contact
passes through the pitch point. This change in
31
AGMA 917--B97
sliding creates frictional subsurface shear

stresses which can eventually remove material
and form the surface cavities known as pitting;
-- Lowest point of single tooth contact.
This is the point where the contact stress is at its
maximum. It occurs here because the radii of
curvature of the involutes are at their most
opposite extremes, when the tooth is under its
maximum load (when there is only one pair of
teeth carrying the load). For external gear pairs,
the contact on the driver is near its root, (relatively small radius of curvature) and the contact on
the driven gear is near its tip (relatively large radius of curvature). If the contact stresses exceed
the surface endurance strength of the material,
surface failure will result.
5.6.3 Contact ratio
Contact ratio can be visualized as the average
number of tooth pairs in contact during the mesh
cycle. For example, if the contact ratio for a gear
mesh is 1.75, then two tooth pairs will be in contact
75% of the time and only one tooth pair will be in
contact 25% of the time. In general, gear meshes
with more tooth pairs in contact exhibit smoother
operation. For most applications, it is recommended that the contact ratio be at least 1.4.
Contact ratio is a function of center distance.
Contact ratio decreases as the center distance
increases from the nominal center distance. It is
important to determine the allowable deviation from
the nominal center distance that will maintain an
acceptable contact ratio for the application. Finer
diametral pitch gears have a smaller allowable
center distance deviation. It is important to consider
the expected center distance tolerance stack up, of
the entire system, early in the design process. The
designer should then design the system so the
contact ratio is acceptable throughout the complete
center distance tolerance zone.
5.6.4 Fatigue strength
Many machine design texts, such as Mechanical
Engineering Design by J.E. Shigley [6], contain
information on how to determine fatigue strengths,
or endurance limits, for ferrous materials. It is
important to realize that the analytical approaches
do not yield absolutely accurate results. The results
should only be used as a guide, as something that
32
indicates what is important and what is not important, in designing to avoid fatigue failure. Many
materials, such as plastics, aluminum and bronze
do not have true endurance limits so they cannot be
designed to have infinite life. Fatigue curve
information must be obtained from the material
supplier and verified by testing in order to design the
gears to meet life requirements. It is extremely
important to confirm the design by conducting a
testing program on the materials that will be used.
Heat treatment of ferrous materials will increase
fatigue strength. Refer to clause 15 for more
information on load rating and testing procedures.
5.6.5 Surface durability
Surface durability, also known as pitting resistance,
is the capacity to resist the kind of failure which
results from repeated surface or subsurface
stresses. See ANSI/AGMA 2001--C95 for more
information on surface durability ratings for spur and
helical gears. These rating methods assume the
design provides adequate lubrication. Inadequate
lubrication can lead to other modes of surface
failure (wear), which are not covered by these rating
methods. The load rating procedure in ANSI/AGMA
2001--C95 is not suitable for every fine--pitch
application. The rating procedure is based primarily
on experience with coarse pitch gears. As with
fatigue strength, available data on material
properties are limited to the more traditional gear
materials. Heat treatment of ferrous materials will
increase surface durability.
Material property
information should be obtained from the material
supplier and testing should be done to confirm the
design.
5.6.6 Gear system assembly
Before the detailed design can be performed, the
designer must consider the method of mounting to
be used. Proper installation of the gear system is
essential for achieving good performance. Some
items of consideration should be:
-- Mounting of gears on the shafts. There
are several methods used to mount gears to
shafts. Many designs offer various degrees of
precision, cost, reliability and ease of assembly.
The designs can be classified into two main
types: removable fastenings and permanent
fastenings.
Removable fastenings
pinning
clamping
set screws
key way
spline
taper and screw
Permanent fastenings:
press fit
shrink fit
molded assembly
cementing compounds
staking
pegging
riveting
spinning
For more information on mounting gears refer to

Precision Gearing, Theory and Practice by
Michalec [7];
-- Shaft alignment.
A deviation in the
alignment of shafts is composed of two components: in--plane deviation and out--of--plane
deviation. The in--plane deviation is measured in
the common plane of axes and out--of--plane
deviation is measured in the plane (skew plane)
perpendicular to the common plane of axes. Figure 35 shows the in--plane and out--of--plane
deviations for two shafts A and B;
Shaft A
Shaft B
AGMA 917--B97
-- Gear--box housing. The housing must be

designed to remain sufficiently rigid during operating conditions. It is possible to encounter a
resonant condition, where the natural frequency
of the complete system assembly coincides with
an operating frequency.
It is beyond the scope of this Information Sheet to
provide detailed information on vibration control
and analysis. It is suggested that the designer
consult someone experienced in this field or a
text on the subject.
5.7 Application considerations
The preliminary design step helped to determine the
estimated values for parameters such as number of
teeth, diametral pitch, pressure angle and helix
angle. The detailed design step, that will be
discussed in clause 6, will optimize the specific
tooth geometry in order to meet the functional
requirements. As mentioned previously, the three
main types of gear systems are power gears,
smooth motion gears and zero backlash gears.
This clause will discuss some of the specific
requirements of each of these systems. It is
important to keep in mind that many applications
have requirements that overlap into more than one
of these classifications.
5.7.1 Design emphasis for power gear systems
Plane of axes
In--plane deviation
Shaft B
Shaft A
Out--of--plane
deviation
Plane of axes
Figure 35 -- Shaft alignment deviations
-- Couplings. The coupling must have some
degree of flexibility to accommodate the types of
misalignment mentioned previously. It must be
able to transmit torque, yet limit the forces on machine components such as shafts and bearings
that result from misalignment. However, it is important to understand the effects the coupling
has on smooth motion when misalignment is
present;
As the name suggests, the primary function of

power gear systems is to transmit power (speed
and torque). This does not mean that the gears
transmit only heavy loads, but that greater importance is given to transmitting power than to
transmitting uniform motion. The design process
for power gears emphasizes adequate size, suitable materials, appropriate heat treat procedures
and proper lubrication to maximize gear life.
Designers usually evaluate the resistance of the
gears to pitting and bending fatigue. It is important
to note that the designers must also recognize
life--cycle costs, noise and other related parameters
during the design process.
5.7.2 Design emphasis for smooth motion
gears
In addition to the requirements of power gear
systems, smooth motion gears are designed to
have low transmission error. Transmission error is
defined as the deviation of the position of the driven
gear, for a given angular position of the driving gear,
from the position that the driven gear would occupy
if the gears were geometrically perfect and infinitely
33
AGMA 917--B97
stiff. Transmission error is usually measured as a

linear error along the line of action.
-- Runout in the gear blank mounting arbor or

tool arbor or both;
Generally, transmission error in gear meshes has

the following two components:
--
Variations in the hob;
--
Tooth--spacing error in the shaper cutter;
-- Once per revolution of the gear(or pinion)

variation;
-- Position errors in the gear generators

indexing gear train;
--
-- Effect of approximating the involute profile

with generated straight cuts;
Once per tooth variation.
The once per revolution variation is due to accumulated pitch variation. The once per tooth variations
are due to variations in profile, pitch, tooth thickness
and tooth alignment. The above two components
affect the positional accuracy of the drive train. In
addition, the once per tooth component is related to
the noise and vibration characteristics of the gear
train.
The main sources of transmission error can be
classified into the two following categories:
-- Variations
mounting;
during
manufacturing
and
-- Deflections of gear teeth, shafts, bearings

and housing.
Parameters such as transverse contact ratio and
face contact ratio influence load sharing between
the gear teeth and thus affect the deflections of the
gear teeth. Deflections of shafts, bearings and
housings can cause an uneven load distribution
across the face of the gear tooth. Calculating
transmission error based on the above gear parameters is beyond the scope of this design manual.
However, this clause will discuss those variations
that contribute to transmission error.
Total transmission error for a gear train can be
calculated by assuming that the total transmission
error is the sum of the transmission error of each
mesh. However, such a technique yields very high
transmission error values. Multi--mesh gear trains
should be analyzed using statistical techniques to
determine a more realistic value.
The sources of transmission error are discussed in
5.7.2.1 and 5.7.2.2, and techniques for minimizing
transmission error are briefly discussed in 5.7.2.3.
5.7.2.1 Manufacturing variation
Some of the possible sources of these errors are:
-- Eccentric mounting of the gear blank or the
generating tool or both;
34
--
Vibration and chatter of the machine tool;
-- Deflection due to the work mass and cutting

forces;
--
Deformations of the gear blank;
--
Non--homogeneous gear blank material;
--
Differential temperature effects;
--
Slippage of the blank on the arbor;
-- Errors in the dressing of the wheel profile for

profile ground gears.
5.7.2.2 Assembly variation
Shaft misalignment causes uneven load distribution
and higher tooth fillet stresses in gears. Uneven
load distribution results in larger transmission error.
Misalignment between shafts can cause the shaft
couplings to transmit non--uniform motion. Another
contributor of transmission error comes from
mounting runout that causes the gear true center to
be displaced from the center of rotation. Some of
the sources of runout are:
--
Clearance between gear bore and shaft;
--
Runout of the shaft;
-- Eccentricity of the rotating race in the ball

bearing.
5.7.2.3 Minimizing transmission error
Minimizing the once per revolution component of
transmission error is best achieved by minimizing
accumulated pitch error. Minimizing the once per
tooth component is achieved by controlling tooth
form. Gear meshes with larger contact ratios exhibit
lower tooth to tooth transmission error because they
have more tooth pairs sharing the load. Hence it is
recommended that gear designers maximize contact ratio when designing gears for smooth motion.
Tooth modifications (tip relief and root relief) and
tooth lead modifications (crowning) are techniques
that have been applied successfully under certain
conditions to minimize transmission error. When
designed for a specific load condition, these are
suitable for only that load condition. It is recommended that an experienced gear designer be
consulted when designing gears with profile and
lead modifications.
5.7.3 Zero backlash gears
In addition to the requirements of power gear
systems, zero backlash gear meshes are designed
to have low backlash. Backlash is necessary to
prevent tight mesh and interference due to
manufacturing and assembly variations in a gear. In
some applications, backlash needs to be controlled
to achieve accurate angular positioning of machine
components. Many techniques have been developed to control backlash in a gear mesh. Clause 7
of this Design Manual deals with those techniques
for controlling backlash in a gear mesh.
6 Design synthesis and analysis

6.1 Introduction
Clause 5 introduced the gear designer to basic
considerations for the early conceptualization of a
gear design. The selection of type of gearing and
preliminary values for number of teeth, diametral
pitch, helix angle and number of stages may be very
simple or exceedingly complex. The function of the
design and the size of the design space influence
the options available to the designer. Clause 5
presented general guidelines and the most basic
procedures but could not offer a complete cookbook
approach because the number of options available
at the preliminary stage is far too large. Therefore, it
is assumed that first estimates of number of teeth,
diametral pitch, pressure angle, helix angle and
number of stages will be made using clause 5 as
well as other sound mechanical engineering
practices.
Clause 6 presents procedures by which the designer can determine, refine and analyze the details that
comprise a complete gear design. This optimization
of the design may result in only minute modifications
to the original concept or it may identify the need for
major redesign. In either case, the proper function
and life of the gear system depends upon the
optimization being done with knowledge and
thoroughness.
AGMA 917--B97
The information presented in clause 6 is rigorous

mathematically and may be somewhat intimidating
to the novice gear designer. It may be beneficial to
remember that an involute gear tooth is a combination of four geometric sections: top land, flanks,
trochoid or root fillet and root circle. Within certain
limits, these sections are independent of one
another and each can be varied without necessarily
affecting the others. For example, within certain
limits, the tip diameter can be made larger or smaller
without changing the tooth thickness. Standards,
such as ANSI/AGMA 1003--G93, Tooth Proportions
for Fine--Pitch Spur and Helical Gearing, define
relationships between the tooth sections based
upon the concept of a basic rack. However, without
violating the precepts of the standard, the designer
has freedom to optimize the design by specifying
gear parameters.
The specifics of tip diameter, tooth thickness and
root diameter are what gear optimization is all
about. Clause 6 presents the procedure for
accomplishing that optimization. The procedure is
presented in a logical order for making the calculations, i.e., the inputs needed for a given calculation
have been calculated previously.
Most gear
practitioners use a form of spreadsheet to perform
these calculations, and the sequence of clause 6
will make constructing a spreadsheet a relatively
simple undertaking.
6.2 Standard gear parameters
6.2.1 Transverse diametral pitch (Pd )
This is a measure of the size of a gear tooth in the
plane of rotation. It is the number of gear teeth in
one inch of standard pitch diameter. The larger this
value, the smaller is the size of the gear tooth.
Transverse diametral pitch is given by:
Pd = Pnd cos
...(2)
6.2.2 Standard pitch diameter for pinion and

gear (DP, DG )
This diameter is equal to the ratio of the number of
teeth to the transverse diametral pitch. The circle
with this diameter is called the standard pitch circle.
Figure 36 illustrates the standard pitch circle.
DP =
NP
N
= P
P nd cos
Pd
NG
N
DG =
= G
P nd cos
Pd
...(3)
35
AGMA 917--B97
6.2.4 Base diameter (DbP, DbG )

Base diameter is the diameter of the base circle
associated with the involute profile (figure 36). It is
equal to the standard pitch circle diameter multiplied
by the cosine of the transverse pressure angle at
the standard pitch diameter.
Standard
pitch radius
Base radius
Outside radius
Root radius
DbP = DP cos t
DbG = DG cos t
...(5)
A discussion on the base diameter and
involutometry can be found in clause 4.
6.2.5 Transverse circular pitch (pt )
This is the circular arc distance measured along the
standard pitch circle between two corresponding
points of adjacent teeth. It is equal to the
circumference of the standard pitch circle divided by
number of teeth. It is also equal to the distance
between corresponding points of two adjacent teeth
in the cutter at its pitch line.
pt =
...(6)
Pd
Figure 36 -- Standard pitch circle

6.2.3 Profile angle and pressure angle
The term profile angle is used in conjunction with
the cutter used to manufacture the gear by a
generating process. The term pressure angle is
used in conjunction with the operating
considerations of the meshing gears. See 4.1.9.1.
6.2.6 Transverse base pitch (pb )
6.2.3.1 Profile angle
pb = pt cos t
...(7)
Figure 37 illustrates the transverse standard
circular pitch and transverse base pitch.
The profile angle of a rack--type cutter (hob) is the

angle between a line perpendicular to a pitch line of
the rack and the profile of a rack tooth.
For a pinion--type cutter (shaper cutter), at the point
of intersection of the pitch circle and the tooth
profile, the angle between a line normal to the pitch
circle and a line tangent to a tooth profile is defined
as the profile angle.
6.2.3.2 Normal standard pressure angle (n )
Normal standard pressure angle is the profile angle
of the cutter used to manufacture the gear by the
generation process.
6.2.3.3 Transverse standard pressure angle (t )
Transverse standard pressure angle is the angle
between a line in the transverse plane tangent to the
involute at the standard pitch diameter and a line
passing through the center of the gear and the
tangency point. The transverse standard pressure
angle is given by:
tan n
t = tan1 cos
36
...(4)
This is the circular arc distance measured along the

base circle diameter between two corresponding
points of the adjacent teeth. It is equal to the base
circle circumference divided by the number of teeth.
Transverse
circular
pitch
Transverse
base
pitch
B
C
Base radius
Standard
pitch radius
Figure 37 -- Transverse pitch

6.2.7 Normal base pitch (pbn )
Normal base pitch in an involute helical gear is the
base pitch in the normal plane. In order for two
helical gears to mesh without interference, the two
gears must have the same normal base pitch.
Normal base pitch is the normal distance between
parallel helical involute surfaces on the plane of
action in the normal plane or in the length of the arc

on the normal base helix.
pbn = pt cos cos n
6.2.8 Standard center distance (Cs )
...(8)
DP + DG
2
6.3 Gear mesh related parameters
at the pitch point. At all other contact points along

the line of action, both sliding and rolling occur. In
figure 38, point P is the pitch point.
6.3.4 Line of contact
Standard center distance is the sum of the standard

pitch circle radii of the pinion and the gear and is
given by:
Cs =
AGMA 917--B97
...(9)
Line of contact is the line or curve along which the

two mating tooth surfaces are tangent to each other
at any instant of time. For spur gears, the line of
contact is parallel to the axis of rotation whereas in
helical gears, the line of contact is inclined to the
axis of rotation. Figure 39 shows the line of contact.
6.3.1 Operating center distance (C)

Operating center distance (C) is the distance
between the axis of the pinion and the axis of the
gear. The operating center distance is controlled by
the frame (casing or gear box) and the bearings that
support the gears. Clause 5 presented a brief
discussion on the geometric parameters that influence center distance tolerance. One of the main
advantages of the involute profile is its ability to
transmit uniform rotary motion independently of the
changes in center distance.
Tangent plane
Helical line of contact
6.3.2 Line of action

The line tangent to base circles of the pinion and the
gear in the transverse plane is called the line of
action or the base tangent line. It is the locus of
points in the transverse plane where the pinion tooth
makes contact with the gear tooth. Figure 38
illustrates the line of action.
Base circle
(pinion)
Base circle
(gear)
Operating
pressure
angle
P
Line of centers
Line of action
Driver
Driven
Figure 38 -- Line of action

6.3.3 Pitch point
Pitch point is a function of the actual operating
center distance and not a function of the standard
center distance. It is the point where the line joining
the center of the pinion to the center of the gear
intersects the line of action. Pure rolling occurs only
Spur line of contact

Figure 39 -- Line of contact
6.3.5 Operating pitch diameter (dP, dG )
The diameter of the circle in the transverse plane
passing through the pitch point with its center at the
center of the gear or pinion is called the operating
pitch diameter.
dP =
2N PC
NP + NG
(external gears)
...(10a)
2N GC
dG =
NP + NG
2N PC
NG NP
2N GC
dG =
NG NP
dP =
(internal gears)
...(10b)
The operating pitch radii of the pinion (rP ) and of the

gear (rG ) are equal to one half of the corresponding
operating pitch diameter. The operating pitch
diameter is used to calculate the tangential load on
the gear. For a given mesh, the operating pitch
diameters of the pinion and the gear are tangent to
each other. Note that the standard pitch diameters
37
AGMA 917--B97
of the pinion and the gear are not necessarily

tangent to each other and they are tangent to each
other only when they are equal to the corresponding
operating pitch diameters.
6.3.6 Operating transverse pressure angle (pt )
The angle between the line of action and the line
joining the center of the pinion and the center of the
gear is called the operating pressure angle. This is
equal to the arc cosine of the ratio between the base
diameter and the operating pitch diameter. Figure
38 shows the operating transverse pressure angle.
pt=cos 1
Dd (external and internal gears)

bP
P
....(11)
6.3.7 Operating helix angle (p )

The operating helix angle is used to calculate the
component of the transmitted force in the axial
direction. The operating helix angle is the helix
angle at the operating pitch radius and is given by:
p = tan1
d Dtan
P
(external and internal gears)
...(12)
6.3.8 Operating normal pressure angle (pn )

This is the operating pressure angle in the normal
plane and is given by:
pn = tan - 1 (tan pt cos p )
...(13)
The operating normal pressure angle describes the

direction of the force transmitted between the
contacting gear tooth surfaces. In a given design,
an increase in the operating pressure angle leads to
an increase in the separating force.
Backlash can be observed by holding one of the

mating gears fixed and moving the other gear from a
contact on one side of a tooth to a contact on the
other side. One can measure backlash in a gear
mesh in the following three ways:
--
Transverse backlash;
--
Normal backlash;
--
Backlash along the line of action.
Even though the values for the backlash measured

in the above three ways are different, they are all
related to each other mathematically.
6.3.9.1 Transverse backlash
Transverse backlash (B) is defined as the difference
between the transverse circular tooth space of the
gear measured at its operating pitch circle and the
transverse circular tooth thickness of the pinion
measured at its operating pitch circle. In figure 40,
arc PQ is the transverse circular tooth thickness of
the pinion at the operating pitch circle and arc PR is
the transverse circular tooth space of the gear. The
difference between the arcs PR and PQ is the
transverse backlash for the given mesh. Backlash
is usually achieved by decreasing the circular tooth
thickness of the two mating gears. During machining, the cutter is fed deeper into the gear blank to
decrease gear circular tooth thickness. Transverse
backlash for both external and internal gears is
given by the following equation:
B=
d
d P
t ptP t ptG = G t ptP t ptG
NP
NG
...(14)
Pinion
Operating pitch
circle (pinion)
6.3.9 Backlash
Backlash is a property of the gear mesh and not of
an individual gear. It is the clearance between the
meshing teeth. In most applications, it is not
detrimental to have backlash in a gear mesh. On
the other hand, backlash is necessary to accommodate manufacturing variations in gears and the
relative change in size of gears and their casings
due to the thermal and other environmental effects.
In gears used in position control systems, backlash
may be detrimental. For controlling backlash in
such systems, refer to clause 7 of this manual.
38
Q
R
Operating
pitch circle
(gear)
Gear
Figure 40 -- Transverse backlash is

arc PR -- arc PQ
AGMA 917--B97
Equations for evaluating the tooth thickness at the

operating pitch circle of the pinion (tptP ) and the gear
(tptG ) are given in 6.6.4.
profile. Its shape and size depend upon the

manufacturing method employed.
The amount of backlash should meet the requirements of the application. This amount should
enable gears to run freely when the mesh is
operating under worst case conditions (at the
shortest possible operating center distance, worst
condition of humidity, temperature and geometrical
tooth variations).
In gears that are manufactured by a generation

process, the fillet curve is called a trochoid. During
the generation process, the tip corner of the cutter
tooth generates the trochoid and the flank of the
tooth generates the involute profile. The trochoid is
always tangent to the root circle and in gear teeth
without undercut, tangent to the involute.
Backlash is also a function of center distance. The

amount of transverse backlash (Bc ) due to a change
in center distance (C) is approximated by the
following equation:
6.4.1.2 Formed fillet
Bc = 2 C tan pt (external and internal gears)

...(15)
When two gears mesh at their low points of total
composite variation, the change in center distance
(C) is equal to half the sum of the total composite
variation of the gear and the pinion. For gears with
significant total composite variation, the increase in
the amount of backlash due to this change in center
distance is quite significant. The total transverse
backlash is equal to the sum (B + Bc ).
6.3.9.2 Normal backlash (Bn )
Normal backlash is the difference between the
normal circular tooth space of the gear measured at
its operating pitch circle and the normal circular
tooth thickness of the pinion measured at its
operating pitch circle. The relation between the
normal backlash and the transverse backlash is
given by:
Bn = B cos p (external and internal gears)
...(16)
6.3.9.3 Backlash along line of action (Bb )
Backlash along the line of action is the component
of the transverse backlash measured along the line
of action and is given by:
Bb = B cos pt (external and internal gears)
...(17)
6.4 Tooth fillet based geometric parameters
6.4.1.1 Generated fillet
In some gear manufacturing processes, such as

form milling, form grinding or injection molding, the
fillet shape is transferred directly from the tool. It is
often most convenient to make the fillet portion of
the tool in the shape of a circular arc. This arc is
generally made tangent to the root circle and the
involute profile. The radius of the arc is usually
selected to provide the maximum tooth bending
strength without the risk of interference with the tip
of the mating tooth. In gears that must have
undercut profiles to provide sufficient clearance, it
may not be possible to approximate the undercut
profile with a single arc. In such cases, a series of
arcs can be used, but only after checking for
possible interference.
6.4.2 Undercutting
A gear tooth is said to be an undercut tooth when
any part of the trochoid lies inside a line drawn
tangent to the involute profile at the point of
intersection of the involute and the trochoid. In
figure 41, the gear tooth profile shown is that of an
undercut tooth as the trochoid lies inside a line
drawn tangent to the involute at True Involute Form
(TIF). One needs to be concerned about undercut
when the pinion or the gear has a small number of
teeth. In most applications, some amount of
undercut can be tolerated.
In some cases,
designers intentionally choose gear cutters that
produce an undercut gear tooth to provide relief in
the root of the gear tooth for the grinding or the
shaving process. Severe undercutting can result in
considerable weakening of the tooth and loss of
tooth action.
6.4.1 Tooth fillet
6.4.3 Radius at true involute form (RtifP, RtifG )
Tooth fillet refers to the portion of the gear tooth

profile that connects the root circle to the involute
TIF represents the point of intersection of the fillet

and the involute. Radius at TIF is the radial distance
39
AGMA 917--B97
from the center of the gear to the point of

intersection of the fillet and involute. Figure 41
illustrates the radius at TIF. Non--conjugate action
and interference is seen when tooth contact occurs
below TIF. Khiralla [2] presents a method to
evaluate the radius at TIF for gears manufactured
by hobbing and shaping methods.
Tangent to
involute
at point G
Radius at TIF
Figure 41 -- Undercut teeth
6.4.4 Profile shift (addendum modification)
Figure 42 illustrates another useful characteristic of
involute gear teeth. It is possible, using a cutter of
given diametral pitch and profile angle, to cut the
same number of teeth into gear blanks of various
outside diameters. The resulting gears, despite the
fact that they appear to be of different sizes, all have
the same base pitch and therefore will transmit
uniform rotary motion if run together, providing
contact ratios and clearances are adequate. This
blank diameter modification is called profile shift or
addendum modification.
6.4.4.1 Definitions
The numerical value of profile shift is equal to the
amount by which the nominal center distance of a
X = 0.5
X = 0.0
10 tooth
20 PA
gear pair must be changed in order to accommodate the modified gear. The profile shift coefficient
is the product of the profile shift and the normal
diametral pitch. Unmodified (or standard) gears
have a nominal center distance that is equal to
one--half the sum of their standard pitch diameters.
Standard gears are said to have zero profile shift.
Gears with positive profile shift (called enlarged
gears) result when the generating pitch line (hob
tool) or pitch diameter (shaper cutter) is held at a
larger radius than standard when machining the
gear. Gears with negative profile shift (called
reduced gears) result when this generating radius is
less than standard. Figure 42 shows the change in
the gear tooth form due to a positive profile shift
coefficient of 0.5, resulting in enlarged teeth, for two
profile angles on a 10--tooth pinion. Note that the
enlarged tooth form shows little or no undercut,
which is one of the primary justifications for profile
shift in gear designs.
The addendum of a gear is the radial distance from
the standard pitch diameter to the outside diameter.
Since the outside radius of the gear is usually
increased by the same amount as the profile shift,
the addendum increases for an enlarged gear. For
a time, it was common to refer to enlarged gears
and reduced gears as long and short addendum
gears respectively. Hence, the term addendum
modification is used to describe this process of
enlargement and reduction. However, because it is
not necessary to make the outside diameter any
specific dimension, the amount of change in the
outside diameter may or may not correspond
exactly to the amount of the profile shift.
X = 0.5
X = 0.0
10 tooth
14.5 PA
Figure 42 -- Effect of profile shift (addendum modification)
40
The cutter may be fed in or held out with respect to

the gear blank in order to cut less or greater tooth
thickness. Some gear experts have proposed that
an additional term called rack shift be introduced to
define the amount of cutter feed in or hold out
associated with profile shift coupled with tooth
thickness change for backlash alteration. Since this
proposal was not adopted at the time of this writing,
this Design Manual will use the most rigorous and
simple definition for profile shift. Modification of
tooth thickness is considered a separate topic (see
6.6).
6.4.4.2 Reasons for using profile shift
Avoiding undercut. Pinions with small number of
teeth are usually enlarged to avoid undercut. For 20
degree profile angle gears, gears with fewer than 24
teeth should be designed with positive profile shift.
However, note that there is a maximum limit to
which the gear tooth can be enlarged before the top
land thickness (tooth thickness of the gear at its
outside diameter) becomes zero. The top land
thickness should be at least 0.275/Pnd .
Balancing bending strength. When the shape of
the trochoid of the pinion is considerably different
from that of the gear, the pinion tooth may be
weaker than the gear tooth. In using this to balance
the bending strength of the two mating members,
positive profile shift is applied to the pinion and
negative profile shift to the gear. This makes the
pinion tooth stronger and gear tooth weaker resulting in balanced bending strength. In some cases,
designers balance bending fatigue strength instead
of the bending strength because the pinion tooth
experiences more load cycles than the gear tooth.
Balanced specific sliding. Specific sliding is the
ratio of gear tooth sliding velocity to its rolling
velocity. The amount of specific sliding influences
the amount of wear on the tooth surface of the
pinion and the gear. The extreme specific sliding at
each end of the path of contact should be balanced
in order to minimize wear of the gear teeth. In speed
reducing external gears, balanced specific sliding is
achieved by enlarging the pinion and mating it with
either a reduced or standard gear. In speed
increasing external gear meshes, balancing the
specific sliding by enlarging the pinion usually
results in a lower percentage of recess action (see
6.7.4). It is necessary to compromise between
AGMA 917--B97
balanced sliding and recess action in speed

increasing gear meshes. Khiralla [2] presents a
detailed discussion on the specific sliding
calculations for external gears.
6.4.4.3 Minimum positive profile shift
The minimum amount of positive profile shift is
based upon the following two criteria:
-- When engaged with any mating external
gear or rack, the lowest contact point on the pinion profile should be above any undercutting of
the involute by the generating action of a
standard tool;
-- When engaged with any standard mating
external rack, the lowest contact point on the
pinion profile should be located above an initial
portion of the involute. The initial portion is
avoided because it is often difficult to manufacture accurately. The initial portion corresponds to
the first five degrees of roll.
Based upon the above two criteria, the minimum
value for a positive profile shift coefficient (enlargement) is given by:
X +min =
MAX 0.0, 1.05
N sin tsin t cos t tan 5

2 cos
...(18)
6.4.4.4 Maximum negative profile shift

The teeth of a gear containing a large number of
teeth, mating with a pinion with profile--shifted
enlarged teeth, can be reduced to increase the
percentage of recess action in the mesh. The
amount of negative profile shift should be chosen
carefully and mesh contact below 5 degrees of roll
should be avoided. The maximum value for a
negative profile shift coefficient (tooth reduction) is
given by:
X max =
MIN 0.0, 1.05
N sin t(sin t cos t tan 5)

2 cos
...(19)
Table 2 is the result of charting equations 18 and 19

as a function of number of teeth for 20 degree profile
angle gears. It is clear that when the gear has 24 or
more teeth, the lowest point of contact with a
standard rack is above 5 degrees of roll with no
tooth enlargement required. In fact, in a gear with a
41
AGMA 917--B97
large number of teeth, a significantly high value of

negative profile shift coefficient (tooth reduction)
may be used without contact occurring below 5
degrees of roll. However, use of the maximum
allowable reduction may produce a tooth that has
unacceptably low bending strength. Therefore, it is
recommended that less than 50% of the negative
profile shift allowed in table 2 be used for most
designs. When reducing gears with large numbers
of teeth, engineers normally choose to reduce the
gear by the same amount the pinion was enlarged.
If more than 50% of the reduction in table 2 is
thought to be appropriate, the design must be
carefully analyzed for bending strength, specific
sliding ratio and contact ratio. The amount of
negative profile shift should never exceed that
indicated in table 2. In all cases, tool data and
clearances with mate must be evaluated.
Remember that it is not necessary to reduce gears
with large numbers of teeth. In fact, it is detrimental
to reduce gears in speed increasing drives as it
increases the percentage of approach action (see
6.7.4).
6.5 Additional gear parameters
6.5.1 Nominal center distance
In applications where it is necessary to design for a
fixed value of standard center distance, if one
enlarges the pinion by applying a positive profile
shift, then one has to reduce the gear by applying an
equal but negative profile shift. Such designs are
called Long and Short Addendum Designs (for
enlarged pinions and reduced gears). In applications where there is no need to maintain standard
center distances, one can enlarge the pinion without
reducing the gear. Such gear designs are called
designs with spread centers (enlarged center
distance system).
The nominal center distance for a gear mesh is
given by:
Cn =
NP + NG
X + X G (external gear
+ P
2P nd cos
P nd meshes) ...(20a)
Cn =
NG NP
X X P (internal gear
+ G
2P nd cos
P nd meshes) ...(20b)
Note that for external gears when XP = - XG , the

nominal center distance equals standard center
42
distance. The actual operating center distance

should be greater than or equal to the nominal
center distance. Operating the gears at center
distances smaller than the nominal value could lead
to a tight mesh condition and interference.
6.5.2 Outside and inside diameter
The outside diameter on the external gears and the
inside diameter on the internal gears are normally
related to the dedendum of the cutter when
generation cutters of topping variety (one that cuts
the outside and the inside diameter) are used.
However, when such tooling is not used, the outside
and inside diameter can be made without any
relation to the dedendum of the cutter. The
standard outside diameter and standard inside
diameter are given by:
D oP = DP +
2X P
+ 2
P nd
P nd
D oG = DG +
2X G
+ 2 (for external gears)
P nd
P nd
...(21b)
D iG = DG +
2X G
2 (for internal gears)
P nd P nd
...(21c)
...(21a)
For spur pinions with small numbers of teeth (9 to

12), one would have to design gears with an outside
diameter smaller than the value given by the above
equations because the opposite involute profiles of
a gear tooth intersect inside of, or too close to, this
value (low top land thickness). In such cases, the
outside diameter must be reduced to a size such
that there is sufficient top land. The top land
thickness of the gear should be at least 0.275/Pnd .
Table 3 gives the maximum outside diameter for
pinions with fewer than 15 teeth for 20 degree profile
angle gears. For 20 degree profile angle spur
gears, table 3 shows that maximum outside diameter for pinions with fewer than 13 teeth is lower than
the standard value. 20 degree profile angle spur
gears with 13 teeth or more can be designed with
standard value for the outside diameter.
The standard value for the addendum coefficient of
the cutting tools is 1.2 + (0.002)(Pnd ). In order to
maintain sufficient clearance at the root of the gear
tooth, designers are not encouraged to choose
outside diameters greater than the standard value
given by equations (21 a, b and c).
AGMA 917--B97
Table 2 -- Profile shift coefficients for 20 profile angle spur gears1)

Number
of teeth
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Profile shift coefficient2)

Minimum
Maximum
enlargement
reduction
0.6501
0.0000
0.6057
0.0000
0.5613
0.0000
0.5168
0.0000
0.4724
0.0000
0.4280
0.0000
0.3836
0.0000
0.3391
0.0000
0.2947
0.0000
0.2503
0.0000
0.2058
0.0000
0.1614
0.0000
0.1170
0.0000
0.0725
0.0000
0.0281
0.0000
0.0000
--0.0163
0.0000
--0.0607
0.0000
--0.1052
0.0000
--0.1496
0.0000
--0.1940
0.0000
--0.2385
0.0000
--0.2829
0.0000
--0.3273
0.0000
--0.3718
0.0000
--0.4162
0.0000
--0.4606
0.0000
--0.5050
0.0000
--0.5495
0.0000
--0.5939
0.0000
--0.6383
0.0000
--0.6828
0.0000
--0.7272
0.0000
--0.7716
0.0000
--0.8160
0.0000
--0.8605
0.0000
--0.9049
0.0000
--0.9493
0.0000
--0.9938
0.0000
--1.0382
0.0000
--1.0826
Number
of teeth
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
Profile shift coefficient

Minimum
Maximum
enlargement
reduction
0.0000
--1.1271
0.0000
--1.1715
0.0000
--1.2159
0.0000
--1.2603
0.0000
--1.3048
0.0000
--1.3492
0.0000
--1.3936
0.0000
--1.4381
0.0000
--1.4825
0.0000
--1.5269
0.0000
--1.5714
0.0000
--1.6158
0.0000
--1.6602
0.0000
--1.7046
0.0000
--1.7491
0.0000
--1.7935
0.0000
--1.8379
0.0000
--1.8824
0.0000
--1.9268
0.0000
--1.9712
0.0000
--2.0157
0.0000
--2.0601
0.0000
--2.1045
0.0000
--2.1489
0.0000
--2.1934
0.0000
--2.2378
0.0000
--2.2822
0.0000
--2.3267
0.0000
--2.3711
0.0000
--2.4155
0.0000
--2.4599
0.0000
--2.5044
NOTES:
1) This table gives the minimum required positive profile shift coefficient (enlargement) and the maximum allowable reduction for 20 degree profile angle gears.
Equations 18 and 19 may be used to generate similar
tables for gears having profile angles other than 20
degrees.
A thorough understanding of 6.4.4 is
recommended before using this table.
2) Profile shift coefficients are dimensionless. Divide
values by normal diametral pitch to obtain profile shift in
inch units.
43
AGMA 917--B97
Table 3 -- Maximum outside diameter for minimum topland of 0.275/Pnd

Number of teeth
9
10
11
12
13
14
15
Pressure angle = 20 degrees

Profile shift coefficient
Max OD1)
0.6501
12.0144
0.6057
13.0256
0.5613
14.0304
0.5168
15.0296
0.4724
15.9448
0.4280
16.8560
0.3836
17.7672
Comments
<Std
<Std
<Std
<Std
Std
Std
Std
NOTE:
1) Max OD column is dimensionless. Divide values by diametral pitch for inch units.
In internal gears, the inside diameter should not

extend below the base circle. For internal gears
with fewer than 34 teeth, one should either apply
positive profile shift or increase the inside diameter
such that roll angle at the inside diameter is at least
5 degrees.
6.5.3 Helix angle at the outside and inside
diameter
The lead of the helical gear tooth is the same at all
diameters and hence helix angle will be different at
different diameters. The helix angle at the outside
diameter of the pinion is given by:
oP = tan1
D Dtan
oP
...(22a)
The helix angle at the outside or inside diameter of

the gear is given by:
bP
otG = cos1
bG
itG = cos1
bG
oP
oG
iG
The normal pressure angle at the outside or inside

diameter is given by:
onP = tan--1 (tan otP cos oG )
...(25a)
-1
onG = tan (tan otG cos oG ) (external gears)
...(25b)
-1
=
tan
(tan
cos
)
(internal
gears)
inG
itG
iG
...(25c)
D Dtan (internal gears)...(22c)
6.6 Tooth thickness calculations
...(22b)
iG
The base helix angle is the helix angle at the base

diameter and is given by:
44
DD
...(24a)
DD (external gears) ...(24b)
DD (internal gears) ...(24c)
otP = cos1
These parameters are needed to evaluate the tooth

thickness at the outside and the inside diameters of
the external and internal gears respectively.
oG
6.5.4 Base helix angle
b = tan1
The transverse pressure angle at the outside or

inside diameter is given by:
D Dtan (external gears)
oG = tan1
iG = tan1
6.5.5 Pressure angle at outside or inside

diameter
D Dtan = tan D Dtan

bP
bG
...(23)
The tooth thickness is a key dimension in defining

the gear tooth proportions. ANSI/AGMA 2002--B88
gives a detailed description on how one should
specify and measure tooth thickness. Throughout
this document, tooth thickness will be defined as a
circular arc distance. The standard value of tooth
thickness for a spur gear with zero profile shift is
one--half of the standard circular pitch. Usually, the
tooth thickness of a gear is reduced from its
standard value to make allowance for backlash.
The parameter that accounts for this allowance

made for backlash is called the tooth thinning
coefficient.
Tooth thinning coefficient is a normalized measure

of the amount by which the tooth thickness of a
pinion or gear is reduced from its nominal value at
the standard pitch diameter. The tooth thicknesses
are reduced by feeding the cutter deeper into the
gear blank. The tooth thinning coefficients are
considered independent of the profile shift in order
to have the outside diameter independent of tooth
thinning for backlash. For a required normal
backlash (Bn ), the tooth thinning coefficients should
sum to:
DP + DG
2C
2X tan n s nP
t nP = + P
2P nd
P nd
P nd
...(27a)
2X tan n s nG
t nG = + G
(external gears)
Pnd
Pnd
2P nd
...(27b)
6.6.1 Tooth thinning coefficient
s nP + s nG = B n P nd
AGMA 917--B97
...(26)
A positive value for tooth thinning coefficient

indicates that the gear tooth has been thinned for
backlash. Suggested values for tooth thinning
coefficients are 0.015 to 0.08. In most cases, when
pinions have small numbers of teeth, tooth thinning
to allow for backlash is applied only on the gear. In
some cases, one can increase the tooth thickness
of the pinion by choosing a negative value for tooth
thinning coefficient. But in such cases, the gear
designer should be sure that the gear mesh has
positive backlash by using equation 14. When the
sum of the tooth thinning coefficients (snP + snG )
is positive, the gear mesh will have positive
backlash.
When using generating cutters of topping variety,
the outside diameter will be affected if the cutter is
fed deeper into the blank to make allowance for
backlash. In such cases, one can reduce tooth
thickness by side trimming or by increasing the
tooth thickness of the cutter.
6.6.2
Normal circular tooth thickness at
standard pitch diameter
The normal circular tooth thickness for a gear with
zero profile shift and no tooth thinning for backlash
will be equal to one--half of the normal circular pitch.
The normal circular tooth thickness at the standard
pitch diameter for the pinion and the gear are given
by:
2X tan n s nG
t nG = G
(internal gears)
2P nd
P nd
P nd
...(27c)
6.6.3 Transverse circular tooth thickness at
standard pitch diameter
Transverse circular tooth thickness at the standard
pitch diameter is given by:
t tP =
t nP
cos
...(28a)
t tG =
t nG
cos
...(28b)
6.6.4 Transverse circular tooth thickness at

operating pitch diameter
The transverse circular tooth thickness at the
operating pitch diameter is given by:
Dt
tP
t ptP = dP
+ inv t inv pt
..(29a)
...(29b)
...(29c)
t tG
+ inv t inv pt (external gears)
DG
t ptG = d G
t ptG = d G
t tG
inv t + inv pt (internal gears)
DG
The normal circular tooth thickness at the operating

pitch diameter is given by:
tpnP = tptP cos p
...(30)
6.6.5 Circular tooth thickness at outside or
inside diameter
The transverse circular tooth thickness at the
outside or inside diameter (top land) is given by:
Dt
tP
t otP = DoP
totG = D oG
titG = D iG
Dt
tG
Dt
tG
G
+ inv t inv otP
....(31a)
+ inv t inv otG (external gears)
...(31b)
inv t + inv itG (internal gears)
...(31c)
The normal circular tooth thickness at the outside or

inside diameter is given by:
tonP = totP cos oP
...(32a)
45
AGMA 917--B97
total composite variation, then the gear mesh will

interfere during operation. For more details on the
effect of total composite variation on gear tooth
thickness, refer to ANSI/AGMA 2002--B88.
tonG = totG cos oG (for external gears) ...(32b)

tinG = titG cos iG (for internal gears)
...(32c)
When the above equations for the tooth thickness at

the outside or inside diameter give negative values,
it means that the two opposite involutes of a tooth
have intersected below the outside diameter for an
external gear or above the inside diameter for an
internal gear. This situation can occur when gears
with small numbers of teeth are enlarged to avoid
undercut. In such cases, it is recommended that the
outside diameter be decreased such that the tooth
thickness at the outside diameter is at least
0.275/Pnd . For internal gears, one would increase
the inside diameter such that the tooth thickness at
the inside diameter is at least 0.275/Pnd .
6.7 Contact ratio calculations

6.7.1 Start of active profile for external gear
meshes
The start of active profile (SAP) represents a point
near the root of the gear tooth where the tip circle of
the mating gear makes contact. Points A and B in
figure 43 represent the start of active profile for the
pinion and the external gear respectively. The
radius at SAP represents the radial distance from
the center of the pinion or the gear to its SAP.
6.6.6 Effect of total composite variation
R aP =
In coarse pitch gears, the change in circular tooth

thickness due to total composite variation is small
when compared to change in tooth thickness due to
tooth thinning for backlash. However, in fine--pitch
gearing, the effect of total composite variation is
significant. The change in tooth thickness from the
nominal tooth thickness (equation 27) at the
standard pitch diameter due to total composite
variation (Vcq ) is given by:
t = Vcq tan n
R aG =
C sin pt R 2oG R 2bG
C sin pt R 2oP R 2bP
+ R 2bP
...(34a)
+ R 2bG
...(34b)
...(33)
C sin pt R 2oG R 2bG > 0.0
...(35a)
C sin pt R 2oP R 2bP > 0.0
...(35b)
If equation 35a is not satisfied, then RaP = RbP .

If equation 35b is not satisfied, then RaG = RbG .
Base circle
(pinion)
D
A
P
Outside circle
(gear)
B
E
Base circle
(gear)
Line of action
Line of centers
Figure 43 -- Line of action for external gears
46
The above two formulae are valid only when the

following two constraints are satisfied:
If the tooth thinning allowance for backlash is

smaller than this change in tooth thickness due to
Outside circle
(pinion)
These constraints ensure that the tip of one gear

tooth does not contact the mating gear tooth below
its base circle. If the first constraint (equation 35a) is
not satisfied, the tip of the gear mates with the pinion
below the base diameter of the pinion. In such
cases, the designer should increase the profile shift
of the pinion or decrease the outside diameter of the
gear such that these constraints are satisfied.
Since the gear is larger than the pinion, it is very rare
that the second constraint will be violated. If the
second constraint (equation 35b) is not satisfied,
the designer should increase the profile shift of the
gear or decrease the outside diameter of the pinion
such that these constraints are satisfied. When
suggested values for minimum required profile shift
are used, this situation will not occur. See table 2.
6.7.2 Start of active profile for internal gear
meshes
The points where the outside diameter of the pinion
and the inside diameter of the gear intersect the line
of action represent the contact points where the
gear teeth come into contact and leave contact. In
internal gears, non--conjugate contact and interference occur when the inside diameter of the gear
contacts the mating pinion below the base diameter
of the pinion. In order to avoid interference, the
following constraint should be satisfied:
R 2iG R 2bG > RbG RbP tan pt
...(36)
When suggested values for minimum required

profile shift are used, the above constraint will be
satisfied. For internal gears, the radius at the start
of active profile for the pinion and the gear are given
by:
R aP = R 2bP +
2
iG
R 2bG
R bG R bP tan pt
R aG = R 2bG +
2
oP
12
6.7.3 Roll angle at the start of active profile

Roll angle at a point on the involute is a good
measure of how far that point is away from the base
circle. Roll angle at the base diameter is zero
degrees. Roll angle at the start of active profile in
degrees is given by:
12
e aP = 180
e aG = 180
1.0
R aP
R bP
R aG
R bG
...(38a)
1.0
...(38b)

As it is difficult to manufacture gears with accurate
involute profile near the base circle, it is recommended that the roll angle at the start of active
profile for the pinion and the gear be at least 5
degrees. If the profile shift coefficients for the pinion
and the gear are more than the minimum required
given in table 2 and if the pinion and the gear have
standard outside diameters, then the roll angle at
the start of active profile will be greater than 5
degrees.
6.7.4 Approach and recess action
In figure 43, the pinion is shown as the driver and
point A represents the first point of contact. Since
mesh contact between points A and P along the line
of action approaches the pitch point, gear mesh
action between these points is termed approach
action and for similar reasons, gear mesh action
between points P and B is termed recess action.
During approach action, the frictional forces oppose
the rotation of the driven gear, whereas during
recess action, the frictional forces aid the rotation of
the driven gear. It is for this reason that involute
action during recess is favorable. Gear meshes that
have a higher percentage of recess action are more
efficient and have longer life.
The distance between points A and P is called the
length of approach and is given by:
R 2bP
+ R bG R bP tan pt
...(37a)
AGMA 917--B97
...(37b)
L a = 1 D2oG D2bG d2G D 2bG

2
(external gears)
...(39a)
The distance between points B and P is called the

length of recess and is given by:
47
AGMA 917--B97
L r = 1 D2oP D2bP d2P D 2bP

2
(external gears)
6.7.6 Transverse contact ratio
...(39b)
Spur gear meshes have transverse contact ratios

typically between 1.0 and 3.1. Spur gear meshes
with transverse contact ratios between 1.0 and 2.0
are called low contact ratio (LCR) spur gear meshes
and those with contact ratios greater than 2.0 are
called high contact ratio (HCR) spur gear meshes.
When gears can be made and mounted with
sufficient accuracy such that the transmitted load is
shared between tooth pairs in contact, it is advantageous to design spur gear meshes with a contact
ratio greater than 2.0 because they have a greater
load carrying capacity and their action is smoother.
However, an individual tooth in a HCR design is
generally weaker than a tooth in a LCR design.
Thus, gear accuracy sufficiently high to ensure load
sharing between tooth pairs is essential.
Figure 44 shows the internal gear mesh. The pinion

is assumed to be driving the internal gear. The
distance between points A and P is called the length
of approach and is given by:
L a = 1 d2G D2bG D2iG D 2bG

2
(internal gears)
...(40a)
The distance between points B and P is called the

length of recess and is given by:
L r = 1 D2oP D2bP d2P D 2bP

2
(internal gears)
...(40b)
6.7.5 Contact ratio

Contact ratio is the average number of tooth pairs in
contact as the gear teeth come into mesh and leave
mesh. This is also known as total contact ratio.
Contact ratio consists of two parameters: transverse contact ratio and face contact ratio. Transverse contact ratio is applicable to spur and helical
gears, whereas face contact ratio is only applicable
to helical gears. If the total contact ratio for gear
mesh is 1.7, then two tooth pairs will be in contact
70% of the time and only one tooth pair will be in
contact 30% of the time.
6.7.6.1 External gears

The points where the outside diameters of the
pinion and the gear intersect the line of action
represent the contact points where the gear teeth
come into contact and leave contact. The active
length of the line of action is equal to the distance
between these two intersection points. The transverse contact ratio is equal to the length of the active
line of action (distance AB in figure 43) divided by
the transverse base pitch.
Line of action
Transverse operating
pressure angle
Pinion outside circle
B
P
A
D
RbP
RoP
RiG
O
RbG
C
Figure 44 -- Line of action for internal gears
48
The transverse contact ratio is given by:

mp =
D2oP D2bP + D2oG D2bG 2C sin pt

2pb
...(41)
It is important to note that this formula for transverse

contact ratio is valid only for gear meshes where the
radius at the true involute form (RtifP, RtifG ) is smaller
than the radius at start of active profile (RaP, RaG ). It
is not uncommon for designers to make this mistake
and design gear meshes whose contact ratio is near
1.4 based on the above formula but in actual gear
action, the contact ratio is near 0.9 due to loss of
involute profile near the root of the gear tooth. This
situation may occur in gear designs with low
numbers of teeth (< 20) and zero profile shift. When
suggested values for profile shift are used, this
situation will not occur. See table 2.
The transverse contact ratio can also be expressed
in terms of radius at SAP and is given by equation
42.
mp =
C sin pt R2aP R2bP R 2aG R 2bG

pb
...(42)
Equation 42 can be used even when constraints in

equation 35 are not satisfied by substituting
RaP = RbP if equation 35a is not satisfied or RaG = RbG
if equation 35b is not satisfied.
From the above formula, the transverse contact
ratio is a function of center distance. Contact ratio
decreases with an increase in the center distance.
Designers should design gears with transverse
contact ratio of 1.3 or greater and should specify a
tolerance on center distance such that the transverse contact ratio is at least 1.1 at the maximum
center distance. The higher the contact ratio, the
smoother the gear mesh.
6.7.6.2 Internal gears
The points where the outside diameter of the pinion
and the inside diameter of the gear intersect the line
of action represent the contact points where the
gear teeth come into contact and leave contact.
The active length of line of action is equal to the
distance between these two intersection points.
Transverse contact ratio is equal to the length of
active line of action (distance AB in figure 43)
divided by the transverse base pitch.
AGMA 917--B97
When the inequality in equation 36 is satisfied, the

transverse contact ratio is given by:
m p=
RbG R bP tan pt+R2oP R2bP R2iG R2bG

pb
...(43)
6.7.7 Face contact ratio

This is applicable only to helical gears. Face
contact ratio for spur gears is zero. In spur gears,
the lines of contact are parallel to the axis of rotation
whereas in helical gears the lines of contact are
inclined to the axis of rotation. This helical action
leads to a parameter called face contact ratio. Face
contact ratio can be viewed as a measure of
average number of teeth in contact in the axial
plane. Face contact ratio is given by:
mF =
F tan b
pb
...(44)
Face contact ratio is directly proportional to helix

angle and face width. One cannot achieve higher
face contact ratios when gears have very narrow
face width. Gear meshes should be designed with
face contact ratio of at least 1.0. This is because
gear meshes with face contact ratio less than 1.0
perform more like spur gears and hence provide no
advantage. In gears with face contact ratio less
than 1.0, the disadvantage of experiencing an axial
thrust outweighs all the advantages of designing a
helical gear mesh.
Gear meshes with integral face contact ratios (face
contact ratio = 1.0 or 2.0 or an integer value) have
the following important property:
The sum of the length of contact lines for gear
meshes with integral face contact ratios does not
vary when the gears rotate.
Since the gear mesh stiffness is approximately
proportional to sum of the length of contact lines,
gear mesh stiffness does not vary with gear
rotation. Gears with constant mesh stiffness exhibit
smoother motion with low vibration and noise.
Hence it is recommended that gear designers
design helical gears with integral face contact ratio.
The designer should be aware that manufacturing
and/or assembly variations will cause a reduction in
face contact ratio.
6.8 Root diameter and clearance
6.8.1 Root diameter
Root diameter is the diameter of the circle passing
through the deepest point of the gear tooth fillet.
49
AGMA 917--B97
The root diameter depends on the addendum of the

cutter used. The standard value for the addendum
of the cutter in inches is given by:
a h = 1.2 + 0.002
P nd
...(45)
The root diameters are given by:
D rP = DP 2 ah
XP
s nP
+
P nd 2P nd tan n
...(46a)
XG
s nG
+
P nd 2P nd tan n
(for external gears)
...(46b)
D rG = DG 2 ah
XG
s nG
+
P nd 2P nd tan n
(for internal gears)
...(46c)
D rG = DG + 2 ah +
6.8.2 Root clearance
The root clearance is the radial separation between

the root circle of a gear and the outside or inside
circle of the mating gear.
The root clearance at the root of the pinion is given
by:
cP = C
cP =
D rP + D oG
2
DiG DrP
2

...(47a)
C (for internal gears)

...(47b)
The root clearance at the root of the gear is given by:

cG = C
cG =
D oP + D rG
2
DrG DoP
2

...(48a)
C (for internal gears)

...(48b)
The minimum root clearance required for normal

successful operation is equal to the clearance in the
cutter and is given by:
c min = 0.2 + 0.002
P nd
...(49)
If standard outside diameters are used and the

gears are operated at nominal center distance,
sufficient root clearance will be present. However, if
the gears are to be operated at closer than nominal
center distance, the root clearance should be
verified to be equal to or greater than the minimum
value.
50
6.9 Load and stress calculations

6.9.1 Normal load (Wn )
This is the force that acts normal to the gear tooth in
the normal plane along the line of action, as shown
in figure 45. In most cases, gears are designed for a
given value of transmitted torque. As it is easier to
calculate the tangential load from a given value of
torque, normal load is often expressed in terms of
tangential load.
Wn =
Wt
cos pn cos p
...(50)
Taken from [9].

6.9.2 Tangential load (Wt )
The tangential load is the component of the normal
force that acts tangentially to the operating pitch
circle of the gear. It is given by:
W t = 2T
dP
...(51)
6.9.3 Axial load (Wa )

The axial load is the component of the normal force
that acts along the axis of the gear. For spur gears,
the axial load is zero. For helical gears the axial load
is:
Wa = Wt tan p
...(52)
6.9.4 Radial load (Wr )

The radial load is the component of the normal force
that acts toward the axis of the gear. It is the
separating force between the two gears. It is given
by:
Wr = Wt tan pt
...(53)
It is important to note that radial loads on the

bearings should be calculated based on the normal
load along the line of action and not based on the
radial load that tries to separate the two gears.
6.9.5 Calculation of bending stress
The evaluation of bending stress at the root of the
gear tooth is based on the AGMA geometry factor
method presented in ANSI/AGMA 2001--C95. It
gives a detailed discussion of the method which is
based upon inscribing a parabola within a gear
tooth. The bending stress in pounds per square
inch at the fillet of the gear tooth is given by:
b =
Wt P d
F Kv J
...(54)
AGMA 917--B97
Wr
t
Wt
Wa
Tooth element
x
z
Pitch
cylinder
Figure 45 -- Showing angle at which load bears on tooth

where
F
is face width;
Kv
is dynamic factor;
is geometry factor.
The J factor accounts for load sharing and stress

concentration at the root of the gear tooth and Kv
accounts for the dynamic effects.
The dynamic factor is given by:
Kv =
50
50 + V
...(55)
where
V
is the pitch line velocity in feet per minute.
The pitch line velocity for the pinion can be

calculated by multiplying rotational speed in revolutions per second by the circumference of the
operating pitch circle in feet.
Note that ANSI/AGMA 2001--C95 contains a different definition for Kv than that which appears above.
The definition of Kv used in this Design Manual is an
older, less sophisticated approximation of the
dynamic factor and is considered completely
appropriate for a majority of fine--pitch designs.
It is recommended that geometry factor, J, be
determined by AGMA 908--B89, Information Sheet,
Geometry Factors for Determining the Pitting
Resistance and Bending Strength for Spur, Helical
and Herringbone Gear Teeth. It includes tables for
some common tooth forms and the analytical
method for involute gears with generated root fillets.
Computer programs are available commercially to
evaluate geometry factor for spur and helical gears
that have profile shift. One can also use analytical
techniques like Finite Element Method (FEM) and
Boundary Element Method (BEM) to evaluate
stresses at the root of the gear tooth. However, to
get an approximate value of bending stresses at the
51
AGMA 917--B97
tooth fillet, the following geometry factor values are

suggested: 0.35 for low contact ratio spur gears;
0.4 for high contact ratio spur gears or helical gears
with face contact ratio less than 1.0; and a value of
0.45 for helical gears with face contact ratio greater
than 1.0.
c =
0.7979 cos
mp cos pn sin pn
Wt m G+1
Fdp mG
1 2P 1 2G
+
EP
EG
...(57)
The contact stress depends upon the modulus of

elasticity. The gear mesh contact stress is based
upon the Hertzian formula for compressive stress
between two cylinders in contact at the pitch point.
Contact stresses are usually higher for pinions with

small numbers of teeth. If the contact stress is
larger than the allowable calculated contact stress,
increasing the number of teeth on the pinion and
increasing the diametral pitch will reduce the
contact stress for constant pitch diameter. Contact
stresses are low in plastics as they have low
modulus of elasticity. In steel, hardening the
material by heat treatment increases the maximum
allowable contact stress. ANSI/AGMA 2001--C95
gives a detailed description of the allowable contact
stress for various steels.
For spur gears, the contact stress at the pitch point

is given by:
6.10 Special topics

6.10.1 Internal gear tooth tip interference
The bending stresses should not be greater than the

allowable limit for the chosen material. If the
calculated bending stress is greater than the
calculated allowable limit, the designer should
choose a coarser diametral pitch and recalculate all
the parameters starting with those given in 6.2.
6.9.6 Calculation of contact stress
c =
0.7979
cos pt sin pt
W t mG+1
Fd p m G
12P 1 2G
+
EP
EG
...(56)
where
EP and EG are the moduli of elasticity of the
pinion material and the gear material
respectively;
P and G are Poissons ratios of the pinion
material and the gear material
respectively.
In low contact ratio spur gears, maximum contact
stress occurs when contact takes place at the
lowest point of single tooth contact. The above
formula evaluates the contact stress at the pitch
point and not at the lowest point of single tooth
contact. In most designs, error in evaluating the
contact stress at the pitch point is small but when
the contact stress is near the maximum allowable
stress value for the material chosen, one needs to
evaluate the contact stress at the lowest point of
single tooth contact. Details on evaluating contact
stresses at the lowest point of single tooth contact
are beyond the scope of this manual and it is
recommended that the gear designers refer to
ANSI/AGMA 2001--C95 and Colbourne [1].
For helical gears, contact stress is measured at the
pitch point and is given by:
52
An internal gear in engagement with a pinion is

subjected to special requirements that do not apply
to external gears. These requirements concern
avoiding interference between portions of tooth
profiles of the internal gear and those of the mating
pinion or the generation tool. Some of the
interferences can be corrected by increasing the
inside diameter of the gear without significant loss in
contact ratio. Other interferences require changes
in tooth thickness and numbers of teeth in the pinion
and the internal gear. Internal gears are usually
manufactured using pinion--shaped (shaper) cutting tools. Design of pinion--shaped cutting tools
plays an important role in the design of internal
gears. The equations that describe the conditions
for interference are given in Colbourne [1] and they
are beyond the scope of this document. This clause
briefly describes some recommendations useful in
avoiding interference between portions of the tooth
profiles of internal gears.
The recommendations are as follows:
-- No part of the internal gear tooth profile
should extend inside its base diameter. This implies that the inside diameter should be greater
than the base diameter. The gear tooth profile is
not an involute below the base diameter and
hence contact below the base diameter could
lead to non--conjugate action or interference with
the mating pinion. It is recommended that the roll
angle at the inside diameter be at least 5 degrees. In order to have at least 5 degrees of roll at
the inside diameter, 20 degree pressure angle

internal gears with fewer than 36 teeth should be
designed with either positive profile shift
coefficient or with an inside diameter larger than
the standard inside diameter. Similarly, in order
to have at least 5 degrees of roll at the inside
diameter, 14.5 degree pressure angle internal
gears with fewer than 72 teeth should be designed with either positive profile shift coefficient
or an inside diameter larger than the standard
inside diameter.
-- The tip of the internal gear tooth should not
mate with the pinion below or near the pinion
base diameter. This condition is called involute
interference. The roll angle at the start of active
profile on the pinion should be greater than 5 degrees. Increasing the inside diameter or applying
a positive profile shift coefficient for the internal
gear will generally be sufficient to avoid involute
interference. When the internal gear mates with
enlarged pinions, it is recommended that positive
profile shift also be applied to the internal gear.
-- The tips of the internal gear teeth should not
interfere with the tips of the pinion teeth as the
teeth are rotated into and out of their contacting
positions. This phenomenon is called fouling.
When the internal gear has at least 10 teeth more
than the pinion, there is sufficient clearance
between the tip of the pinion and the tip of the internal gears. However, it is recommended that
designers refer to Colbourne [1] and check for
the occurrence of interference when the difference in the number of teeth is smaller than or
close to 10. In some cases when fouling occurs,
it can be avoided by increasing the inside diameter of the internal gear or decreasing the outside
diameter of the pinion or both. But, in most
cases, it may be necessary to make extensive
design changes to avoid fouling.
-- Usually, if the first three recommendations
are satisfied, then the gear mesh can be
assembled axially. There are some instances
when axial assembly may be impossible due to
geometry of the gear blank. The tips of the internal gear teeth must not interfere with the tips of
the pinion teeth as the pinion is assembled radially into engagement with the internal gear. For 20
degree pressure angle gears, when the number
of teeth in the gear exceeds the number of teeth
in the pinion by 17 or more, radial assembly may
be possible. The position of the gear tooth profiles depends on the tooth thickness values,
inside diameter of the internal gear, the outside
AGMA 917--B97
diameter of the pinion and pressure angle. It is

recommended that designers check whether the
gear mesh can be assembled radially when the
difference in the number of teeth is smaller than
or close to 17. This requirement is very important
during machining an internal gear with a pinion-shaped generating cutter. The cutter is fed
radially into the blank and when this requirement
is not satisfied, the cutter removes part of the
involute tooth profile near the inside diameter.
Designers are encouraged to refer to the analytical
equations that define the limits to avoid interference
given in the references. When analytical equations
are not available, individual designs can be tested
for all of these requirements by studying enlarged
scale drawings of the two gears in engagement,
both during assembly and during rotation.
6.10.2 Tip rounds and chamfers
The tips of gear teeth are sometimes rounded or
chamfered. This may be due to the burr removal
process or minimum corner radius (wire radius) in
molds and dies made by EDM process. In some
cases, gear teeth are chamfered to avoid damage
to the tooth surface of the mating gear by a sharp
corner at the tip of the gear tooth. Rounding and
chamfering at the tooth tip lead to a decrease in the
effective outside diameter or an increase in the
effective inside diameter. Any decrease in outside
diameter or increase in inside diameter results in
reduced contact ratio and hence gear designers are
requested to evaluate contact ratio using effective
outside and inside diameters. It is important to note
that this effect is significant in fine--pitch gears.
7 Design for control of backlash

7.1 Design procedure
This clause describes a design procedure to ensure
proper backlash in the operating set of gears. It also
serves to correlate manufacturing, inspection and
assembly requirements, dimensional specification
of tooth thickness, measurement over wires and
test radius without ambiguity or conflict of data.
7.1.1 Place in the complete design process
This procedure is concerned only with the geometry
of the meshing gears. It is used after the selection of
materials and general sizes of the two gears, as
described in clauses 5 and 6. It also follows the
selection of the mounting design and gear manufac-
53
AGMA 917--B97
turing processes, all of which determine the expected dimensional variations in the gear set. The
procedure shows whether the assigned tolerances
ensure that all meshing requirements will be met. If
tolerances cannot be met and changes must be
made to the general design, such as in pitch,
numbers of teeth or center distance, then this
procedure is reapplied to the new design. If the
design is acceptable, the procedure will have
supplied the specifications needed to manufacture
and inspect the gears.
7.1.2 Types of gears covered
The procedure is sufficiently general that, with
appropriate selection of equations, it can be applied
to any set of external spur or helical gears and, with
proper precautions, to sets of external--internal
gears.
7.1.3 Basic rules
This design procedure follows these basic rules:
-- Allowances for backlash and gear and
mounting tolerances are made by reductions in
nominal tooth thickness (tooth thinning);
-- The combined effect of tolerance is found by
direct addition of effects of individual tolerances
(worst--case tolerance analysis).
7.2 Backlash
7.2.1 Minimum backlash, BhT (spur gears) or
BnhT (helical gears)
The maintaining of a desired minimum backlash at
the tightest point of mesh is the first objective of this
design procedure. This minimum backlash is
desirable for one or more of the following reasons:
-- To provide room for the fluid lubricant film
that separates the loaded surfaces of the gear
teeth;
-- To provide a flow channel for the fluid
lubricant trapped in the root areas, especially in
high--speed, wide--face spur gears, thereby relieving the pressure that adds load to the shaft
and bearings and reduces efficiency;
-- To permit the passage of small dirt particles
carried by the lubricant;
-- To provide additional clearance between the
gears as an allowance for possible gear and
mounting dimensional variations and thermal
54
effects not considered elsewhere in the design

procedure.
An interference condition, or negative calculated
backlash, should be avoided. Even when it does not
prevent assembly of the gears, the interference will
cause deflections of the gear teeth and of the shafts,
bearings and other supporting members. Such
deflections often result in unpredictability high
forces, greater than the forces from the design load,
and lead to reduced efficiency and premature
failure.
The value for minimum backlash is used as input
data in the design calculations. It should be
selected after review of the above considerations
and their importance to the application.
7.2.2 Maximum backlash
In most power applications, those with a steady
driving action and a steady load, there is no need for
a maximum backlash requirement. As long as tooth
contact occurs on the same side of the tooth and
adequate contact ratio is maintained, large
amounts of backlash are not harmful, assuming
tooth strength is adequate. However, if the rotation
is reversed during operation or if the driven member
changes from resisting the driving torque to aiding
the driving torque, tooth contact abruptly shifts from
one side of the tooth to the other. The resulting
shock and vibration often increase with greater
backlash. For such operating considerations, a
maximum backlash limit is advisable. The value for
this limit is usually selected from prior experience.
If the gear set is used for accurate position control
and operates in both directions of rotation, the
backlash will contribute to the position error. This is
another condition which calls for a maximum
backlash requirement. The value of this requirement is usually selected after making an analysis of
the cumulative effect of gear variations and backlash and relating it to the specified position
accuracy.
In many designs, a maximum backlash requirement
will translate into very tight gear tolerances. When
such tight tolerances are economically unfeasible, it
may be necessary to consider other methods for
backlash control (see 7.5).
7.3 Maximum backlash computation
AGMA Paper 239.14 [8] provides a complete
procedure for correlating data for manufacturing,
inspection and assembly.
AGMA 917--B97
7.4 Test radius

See ANSI/AGMA 2002--B88 for a detailed description of test radius.
7.5 Other methods of backlash control
Some design applications call for severe limits on
the maximum backlash at the point of loosest mesh.
These limits cannot be met using the foregoing
design procedure even if the tolerances are made
as small as practical. Some applications may even
require zero backlash. To meet such conditions, the
designer can use some special methods of backlash control, of which a number of examples using
spur and helical gears are described here.
7.5.1 Selective assembly
Selective assembly is one way of meeting
maximum backlash requirements without further
reducing center distance and tooth thickness
tolerances. Usually one gear in the gear set is
manufactured in a range of sizes and, at assembly,
the proper size is selected to give the desired
minimum backlash at the point of tightest mesh.
With this method, it is still necessary to limit the total
composite tolerances to achieve small enough
backlash at the point of loosest mesh.
7.5.2 Center distance control
Controlling backlash by change in center distance
can be accomplished either by adjustment or by a
spring forcing the gears together. In either case, the
center distance tolerance and the tooth thickness
tolerances no longer influence the backlash.
However, the total composite tolerances are still
important.
Figure 46 -- Adjustable center distance

gearing
7.5.2.2 Spring loading
In this method, the bearings supporting one of the
gears are free to move under the influence of spring
forces so as to tighten the gear center distance.
Except for this movement, the bearings are fully
constrained (see figure 47). This results in zero
backlash for all rotational positions of the gears.
The spring forces must be kept greater than the
opposing bearing forces developed by the gear
loads and other applied loads. This condition
causes wedging of the gear teeth and increased
gear and bearing friction. Because of the total
composite variations in the gears, there is vibration
of the spring loaded bearings and further friction
losses. This method of backlash control is not
suitable for applications with high loads or high
speeds.
7.5.2.1 Adjustment
In this method, the bearings supporting one of the
gears can be moved to obtain the desired minimum
backlash at the point of tightest mesh. One such
arrangement is shown in figure 46. Another
possible arrangement uses eccentric bushings in
fixed housings. After adjustment, the bushings
must be clamped tight enough to resist the forces
developed by the applied loads. Adjustment has a
relative disadvantage. Due to the total composite
variations of the two gears, the backlash for other
rotated positions of the gears will be greater than the
adjusted tightest mesh value.
Figure 47 -- Spring loaded center distance

gearing
55
AGMA 917--B97
7.5.3 Split gearing

In split gearing, one of the gears in the mating pair is
constructed of two half--gears side--by--side (see
figures 48 and 49). Only one half is directly fastened
to its shaft. Rotation of one half relative to the other
has the effect of changing their combined tooth
thickness and thereby controls the backlash. This
relative rotation may be used in the form of an
adjustment during assembly or it may be left as a
continuing action under spring loading. In either
case, the center distance tolerance and the tooth
thickness tolerances no longer influence the backlash. Split gearing has the disadvantage of its more
complex construction and the added space it needs
and the added inertia it introduces. If the gears must
transmit the same load in both directions, the split
design means the doubling or tripling of the overall
size of the split gear and a corresponding increase
in the face width of the one--piece mating gear.
The spring forces must be kept great enough to

withstand the transmitted gear loads in the direction
that opposes the spring. If the transmitted gear
loads are the same in both directions of rotation, the
resulting total gear tooth contact faces are in excess
of three times those for simple gears. The
combined total composite variations of the gears
require some relative motion of the split gear
sections in order to keep the zero backlash
condition. These extra loads and relative motion all
add to the friction losses and make this method of
backlash control not suitable for applications other
than those with light loads and low speeds.
7.5.3.1 Adjustable
In this method, the two half gears are adjusted and
clamped, as in figure 48, to give the desired
minimum backlash at the point of tightest mesh with
the mating gear. The clamping must be tight
enough to resist the operating gear loads. The total
composite variations of all the gears will be reflected
in some increase in backlash during their rotational
cycle.
Figure 49 -- Spring loaded split gearing
7.5.4 Composite gearing with elastic element

One of the gears of the mating pair is constructed
with an elastic element inserted. This type of
gearing consists of a metallic gear with a plastic
element running through the central portion of all of
the teeth. See figure 50.
When this gear is assembled with its mate of
conventional design, the metallic parts of the teeth
will have normal backlash. However, the oversize
plastic section will take up the backlash for all but
the heaviest loads.
Figure 48 -- Adjustable split gearing

7.5.3.2 Spring loaded
The spring loaded design gives zero backlash for all
rotational positions of the gears. A common
construction for such gearing is shown in figure 49.
56
This type will not transmit angular motion accurately, except under light loads, since the plastic
tends to center the composite gear in the center of
the mating gear tooth space. This centering varies
with the amount of load applied. Consideration
should be given to the increased bearing loads and
friction resulting from the contact of the deformed
plastic.
AGMA 917--B97
Figure 50 -- Composite gearing with elastic element

7.5.5 Tapered gearing
It is possible to manufacture a spur or helical gear
with its tooth thickness tapering slightly from one
side of its face width to the other. If two such gears
are made with matching tapers and assembled so
that the tapered teeth fit each other, as in figure 51,
the gears will run together just as well as gears with
uniform tooth thickness. With
only one side of each tooth. This may be done by

preloading the gear train with a spring or weight
acting on the last driven gear, as in figure 52. Such
preloading is possible only if the total rotation is
limited to the range of the spring or weight
arrangement. Unlike some other spring--loaded
methods, this one does not cause wedging of the
gear teeth. However, it does still add to the system
friction and to the load on the gears.
such tapered gears, it is possible to adjust their

relative axial position so as to control backlash.
Figure 51 -- Tapered gearing

The use of spring loading as a means of continually
adjusting the tapered gears for zero backlash is not
recommended. If a small taper angle is used, there
may be wedging of the teeth and increased friction.
7.5.6 Preload
The effect of backlash in a drive train can be
controlled by keeping the gear teeth contacting on
Figure 52 -- Spring preloaded gearing (for

limited rotation)
7.5.7 Dual path
It is possible to achieve the results of the preload
method in a continuously rotating gear train by the
use of the dual path design. In this design, there are
two similar gear trains, a primary train to transmit
the operating loads and a secondary train to
57
AGMA 917--B97
eliminate the backlash. Both are driven by the same

pinion. Their final gears are both on the output shaft
with the primary final gear rigidly connected and
secondary gear spring connected, as shown in
figure 53. As in the preload method, there is no
wedging of the gear teeth but there is still the added
friction and added gear loading.
8 Gear drawings and specifications

8.1 General
The gear drawing should clearly depict the end
product configuration and quality level without
actually describing manufacturing methods. The
drawing may become part of a contract between
gear manufacturer and buyer. Therefore, no design
detail essential to the operation of the gears should
be omitted or assumed. In fairness to both parties,
the drawing should be specific and complete.
See ANSI Y14.6, Geometric Dimensions and
Tolerances.
8.2 Gear data formats
Figure 53 -- Dual path spring loaded gearing

7.5.8 Contra--rotating inputs
The force necessary to keep the driving gear teeth
in contact at only one side can be introduced by
means of a second driving input, as shown in figure
54. This second input applies a torque opposite to,
but smaller than, the torque of the main driving
input. With the proper controls, this opposing
torque can be adjusted to the level needed at each
point in the driving cycle. This avoids the high
friction and gear forces typical of spring--loading
designs.
Output
Rotation
Rotation
Rotation
See AGMA 910--C90, Formats for Fine--Pitch Gear

Specification Data.
8.3 Surface texture (surface
waviness and lay) of gear teeth
roughness,
See ANSI B46.1, Surface Texture and AGMA

906--A94, Gear Tooth Surface Texture with
Functional Considerations.
8.4 Gear blanks
8.4.1 General
The complete design of a gear requires the design
of the gear blank in addition to the design of the gear
teeth. An acceptable blank design will meet the
following requirements:
-- To provide the necessary size for those gear
tooth features which originate in the gear blank;
-- To support and position the gear teeth so that
they will function as intended;
-- To permit economical manufacture with the
desired accuracy.
8.4.2 Design for blank size
Drive
Input
Contra--Rotating
Torque
(Resisting)
Torque
Figure 54 -- Contra--rotating input gearing
58
8.4.2.1 Outside diameter (or inside diameter on

internal gears)
This diameter is usually determined as part of the
general gear design process described in clause 6.
It may be slightly different in the blank when the gear
teeth are to be manufactured by a process in which
the finished outside diameter is formed by the
cutting tool (called a topping tool). The rough blank
outside diameter should then be made greater (or
the blank inside diameter should then be made

smaller) by a suitable machining allowance.
8.4.2.2 Face width
The minimum gear face width is determined as part
of the general gear design process described in
clause 6. The blank face width may be made
greater for various reasons, such as:
-- To insure adequate overlap of the face
widths of the two meshing gears when axial
positioning is not sufficiently well controlled;
-- To allow for reduction of the effective face
width by rounding or chamfering of the ends of
the teeth;
--
To provide adequate face contact ratio.
8.4.2.3 Other blank dimensions

Other blank dimensions, such as the bore diameter,
journal diameter or mounting ring diameter, are not
part of the gear tooth design. They are usually
determined as part of the design of the complete
assembly. However, such dimensions, both in size
and in accuracy, may influence the operation of the
gears and therefore, require the attention of the
gear designer.
8.4.3 Design for gear function
AGMA 917--B97
means used to keep the blank locked to the shaft,

such as a key, pin, setscrew, clamp or interference
fit, the resulting joint must be strong enough to
withstand the applied torque without loosening.
Clearance between the gear blank bore and the
shaft should be minimized. The same applies to the
fit between the key and its keyway, the pin and its
hole, or any joining device and its restraining
surface. Extra strength and an interference fit may
be needed in applications with reversing or abruptly
changing loads.
8.4.3.2 Position of gear teeth
The blank design should ensure the gear teeth are
positioned concentric and parallel to the axis of
rotation of the assembled gear, at least to an
accuracy consistent with the accuracy of the gear
teeth. The blank features that are to provide this
accurate positioning are referred to as the mounting
surfaces. Attachment to the shaft or other supporting members must also be considered. The mating
surfaces, if not in an interference fit or tightly
clamped, should have minimum clearance. The
method of fastening should be such as to preserve
concentricity and parallelism. A loose fitting bore
fastened to the shaft with a set screw, for example,
may result in the loss of the concentricity
maintained separately in the gear and shaft.
8.4.3.1 Support of gear teeth
8.4.4 Design for manufacturing
The shape and proportions of the gear blank should

be adequate to support the gear teeth against the
transmitted tooth loads. The blank should also be
rigid enough to support any other loads tending to
deform the blank and its gear teeth. Weight,
assembly and manufacturing considerations may
require the body of the blank be reduced from the
full face width. In such cases, adequate depth of the
rim and thickness of the web or cross section of the
spokes should be maintained, or ribs should be
added, to prevent deflections which could harm the
gear tooth action or tooth load distribution. In helical
gears, the tooth forces contain axial components
that may require additional rigidity of the blank.
8.4.4.1 Process requirements
Similar concerns apply to the features of the gear

blank that are attached to the shaft or other
supporting member. Where a gear hub is used for
this purpose, it should be of sufficient thickness and
length. It should be axially constrained to support
the radial and axial tooth forces. Whatever the
The shape and proportions of the gear blank must

be compatible with the manufacturing process.
Where the gear teeth are to be formed by a moving
tool, the blank design must permit access to the
tool, both at the beginning and the end of its forming
stroke. Where the gear teeth are to be formed in a
mold or die, the blank design must meet ejection
and other requirements of the forming process.
8.4.4.2 Location requirements
When the blank manufacture is to be followed by a
separate gear tooth machining or finishing operation, the blank design should contain provision for its
location during these operations. Ideally, the
surfaces used for location during machining will be
the same as the mounting surfaces that are to be
used to locate the gear in assembly. For example,
the bore might serve both these purposes on an
external gear and the outside ring diameter serve
59
AGMA 917--B97
these purposes on an internal gear. Often two

surfaces are needed, both in manufacture and
assembly, one to center the gear blank and the
other to keep it parallel to the rotation axis.
Sometimes the one or two assembly mounting
surfaces are not suitable for location during
manufacture. This may be because the mounting
surfaces are not accessible to simple tooling or
because they will not permit the simultaneous
machining of several blanks for greater economy. In
this case, some features should be added to the
blank design to provide these manufacturing location surfaces. These surfaces should be accurately
located relative to the assembly mounting surfaces.
For example, center holes may be added to the
ends of the gear blank journals and used for location
during the gear manufacturing. The journal diameters are made concentric to the center holes. It is
usually advantageous to consult the gear manufacturer during the blank design process, especially if
the gear blanks are to be supplied in finished form to
the gear manufacturer.
8.4.4.3 Clamping requirements
Clamping of the gear blank during machining of the
gear teeth is another consideration in the blank
design. The blank design should permit secure
clamping without any reduction in the accuracy of
the machined gear and without any permanent
deformation of the gear tooth surfaces. Sometimes, the locating surface can also serve as the
clamping surface as, for example, when the gear is
clamped by gripping its bore with an expanding
arbor. At other times, two separate surfaces are
required, as when the gear is centered by an arbor
through its bore but is clamped across its face
against a shoulder on the fixture. In such a case, if
the face of the gear is not square to its bore, the
blank (or the arbor) may be distorted during
clamping. The machined teeth in the unclamped
gear will no longer be accurately positioned relative
to the bore.
Clamping considerations may require that some
part of the blank be extended beyond its specified
shape, with the excess to be removed after the teeth
are completed. When the gear manufacturer is also
to make the blank, such features can be readily
introduced where required.
When the blank
manufacturer and the gear manufacturer are to be
60
separated, such features should be included in the

blank design.
8.4.5 Blank tolerances
Blank dimensions play an important role in gear
performance, either directly or through their influence on assembly or manufacturing accuracy. The
tolerance selected for each dimension should be
appropriate to the role played by the dimension and
to the overall size, pitch and quality level of the gear.
Suggested procedures for selecting these tolerances are described below. Specific applications
may require tighter or looser tolerances depending
on the relative importance given to the gear
performance versus gear blank manufacturing cost.
8.4.5.1 Outside diameter (or inside diameter on
internal gears)
The tolerance on this dimension has a direct effect
on the depth of engagement and, thereby, on the
contact ratio of the mating gears. The direction of
the tolerance should be to reduce material, i.e.,
negative on an external gear and positive on an
internal gear. This is to help prevent interference of
the tooth tip with the fillet of the mating gear and to
help ensure adequate root clearance.
The
suggested tolerances are:
Outside diameter tolerance =
+ 0.0
0.5( TCT ) + 0.1

P nd
...(62)
Inside diameter tolerance =
+ 0.5( TCT ) + 0.1

P nd
0.0
...(63)
where
Pnd is diametral pitch on spur gears and normal
diametral pitch on helical gears;
TCT is total composite tolerance.
The above discussion and suggested tolerances
apply only to the diameter of the finished gear. If the
blank diameter is modified in the cutting (topping)
operation, the tolerance on the original diameter
should be derived from other requirements, such as
the need to control the amount of material to be
removed.
8.4.5.2 Concentricity of outside diameter (or

inside diameter on internal gears)
The tolerance on the concentricity of the finished
diameter relative to the bore or other cylindrical
mounting surface is similar to the diameter tolerance in one respect. By allowing variations in the
radius, it permits greater reduction in depth of
engagement and, hence, in contact ratio. This
additive effect should be kept relatively small. The
suggested concentricity tolerance is:
Tolerance = 0.7(TCT) + 0.0002
...(64)
This tolerance may be increased if there is an

equivalent reduction in the diameter tolerance. The
tolerance may also be increased if the gear design
has been checked for the resulting minimum
contact ratio. See clause 6.
One special situation arises when the gear cutting
set up will center the blank by its outside (or inside)
diameter instead of by its bore or other mounting
surface. If the blank is made by the same shop that
cuts the gear, this shop will set its own tighter blank
concentricity tolerance.
If not, the gear
manufacturer should be consulted before blanks
are made.
8.4.5.3 Other blank dimensions
There are other blank dimensions whose tolerances
can influence the performance of the finished gear.
These include the diameters of the bore, of the
journals, and of other cylindrical surfaces used for
mounting or manufacture. They also include the
concentricities between pairs of these surfaces and
the squareness between these surfaces and the
lateral mounting or manufacture set--up surfaces.
For each of these, there are various assembly and
manufacturing conditions which determine the
effect of the tolerance and the selection of its value.
8.4.5.3.1 Clearance fit assembly
When a cylindrical surface might be assembled
off--center with respect to a shaft in a clearance fit, a
larger diameter tolerance permits a greater clearance and, therefore, a greater off--center condition.
The clearance fit of a bore on a shaft is an example
of this.
AGMA 917--B97
8.4.5.3.2 Rotating fit assembly

When a cylindrical surface on a blank is to rotate
about a stationary member, the added clearance
permitted by the diameter tolerance does not affect
concentricity in the gear meshing action. It may,
however, increase center distance and reduce tooth
engagement. A bore rotating on a stationary shaft
or a journal rotating in sleeve bearings are examples
of this assembly condition.
8.4.5.3.3 Interference fit assembly
When a cylindrical surface on a blank is to be
assembled to the supporting member with an
interference fit, there is usually no change in
concentricity or center distance. The assembly of a
shaft into the inner race of a ball bearing and the
assembly of a thin rim gear on a shaft--mounted
wheel are examples of this assembly condition. The
diameter tolerance is usually determined by the
requirements of the interference fit. In the case of
the thin rim gear, however, large variation in the
amount of interference may cause a corresponding
variation in the size of the stretched gear and in the
resulting depth of engagement. This may present a
problem in very fine--pitch gears and the diameter
tolerance may need to be reduced further.
8.4.5.3.4 Gear cutting set--up
When the blank might be centered differently in the
gear cutting set--up than in the final assembly, a loss
in concentricity may be introduced. This may occur
in two ways:
-- Different surfaces of the blank are used for
the two location purposes. For example, center
holes in the ends of the journals are used in the
gear cutting set--up while the journal diameters
are used in the final assembly;
-- One surface of the blank is used for both location purposes, but its diameter tolerance permits a wide range of size. An off--center condition
may arise when the size of the location device in
the gear cutting set--up cannot adapt to the blank
size variation. In the case of the bore, if the arbor
used to center the blank is made to fit the minimum bore diameter, it will not necessarily center
the blank with maximum bore diameter. If many
blanks are to be located in one set--up on the
same arbor, variations in the bore diameters may
result in off--center gears.
61
AGMA 917--B97
The tolerances needed to control these manufacturing concentricity variations may need to be
tighter than those needed to meet assembly
requirements. These tolerances are best selected
by the gear cutting shop, either by making its own
blanks or by advising the blank designer.
8.4.5.4.1 Bore
Form tolerances may be required on bores or other
inside diameters used for assembly for the conditions of out--of--roundness, taper, bell mouth, and
barrel shape.
8.4.5.4.2 Journals
8.4.5.3.5 Lateral surface assembly

When the gear is to be assembled using a lateral
surface to keep the gear teeth parallel to the shaft
axis, any non--squareness between the lateral
surface and the gear teeth will affect gear performance. Such variations usually introduce an
eccentric condition into the action of the meshing
gears and some degree of tooth misalignment.
If the same lateral surface used to square the gear
in the assembly is also used to position the gear
during manufacture, the blank will not add any
non--squareness. If different surfaces are used, it is
necessary to introduce a tolerance relating the two
surfaces. This tolerance may be a lateral runout
specification, a parallelism specification or both.
When serving purely as in--process tolerances,
these are best set by the gear manufacturer. When
these tolerances are required in order to meet
special assembly conditions, their values should be
selected after considering lateral surface diameter,
gear face width and the amount of eccentricity and
misalignment that can be tolerated.
8.4.5.4 Geometric form
It may be necessary to establish tolerances on
various features of geometric form, such as roundness and flatness, in order to meet especially
stringent assembly conditions. Such conditions
may arise when the tightness of interference fits and
clamped surfaces must be carefully controlled.
They may also arise when the gear blanks are
sufficiently flexible that the tight fit or clamping of
poorly formed surfaces may distort the gear teeth
and reduce their accuracy after assembly. In this
last case, variations in form are less critical if the
gear blank will receive the same kind of support
during manufacture of the teeth as in the final
assembly. The decision to use a form tolerance and
the selection of its value should be based on an
evaluation of these special assembly conditions.
62
Form tolerances may be required on journals, or

other outside diameters used for assembly, for the
conditions of out--of--roundness and taper.
8.4.5.4.3 Lateral surfaces
Lateral surfaces used for assembly may require a
tolerance on flatness. Usually, convexity is not
permitted and the tolerance is expressed as a limit
on concavity.
8.4.6 Drawings
On the gear drawing, the dimensions and tolerances of the gear blank are customarily applied to
the views in accordance with standard drafting
practice. Only the outside diameter and its
tolerance are usually transferred to the format table.
A separate drawing for the gear blank may be made
if the blank is to be obtained from a source other
than the gear shop. Such a separate drawing
should show the larger outside diameter (or smaller
inside diameter) if it is to be cut during gear cutting.
It should also show any extra stock remaining for
clamping during cutting and to be removed later.
9 Gear tooth tolerances

Clause 5.7 of this information sheet discusses
application considerations of gear system design.
Tolerancing of gear teeth cannot be done correctly
without considering the application of the gears. It is
obvious that the closer a gear is to perfect, the better
it will perform in any application. The difficulty is in
determining the character and amount of variation
from perfection that a given application will tolerate.
For example, if noise of a gearbox is of prime
importance, it can be shown that frequency of
excitation is an important factor. Depending upon
where in the geartrain a gear operates, gears are
toleranced differently. The tooth profile variation
may be more important at the low speed end of the
geartrain, whereas, the accumulated pitch variation
AGMA 917--B97
accuracy of a gear is determined by the accuracy of

the shape of each cam (gear tooth) and by how
accurately the teeth are spaced around the cylinder.
In gearing terms, the shape is called the profile of
the tooth and the spacing is called the pitch. Figure
55 shows gear teeth with errors isolated and
exaggerated for illustration. Actual gear teeth would
exhibit combinations of these errors. Inspection of
gear teeth for pitch, profile and lead is termed
elemental inspection and is described in ANSI/
AGMA 2000--A88, Gear Classification and
Inspection Handbook. Elemental inspection of
coarse pitch gears is done by index, involute profile
and lead inspection machines. These machines
may be completely mechanical or computer
controlled.
is more important at the high speed end of the

geartrain.
It is difficult to generalize the process by which
gears are correctly toleranced. The following
paragraphs attempt to offer insight into the
fundamental concepts of tolerancing. Each gear
must be carefully analyzed for functionality in order
to tolerance it correctly.
9.1 Elemental inspection of gears
Clause 4 of this information sheet described a gear
as a series of involute cams uniformly spaced
around the circumference of a cylinder, usually used
to transfer motion from one shaft to another. Perfect
involute gears will theoretically transmit uniform
rotary motion from one shaft to another. The
C
B
E
A
Base circle
A
B
C
D
E
F
Reference profile (perfect form and position)

Spacing error (index error)
Profile error (involute error)
Lead error (helix angle error)
Radial position error (caused by runout or wobble)
Tooth thickness error
Figure 55 -- Graphical representations of typical gear errors
63
AGMA 917--B97
9.2 Functionality of gears

Variations in pitch, profile and lead in gears mounted
on fixed centers cause variations in the rotational
position of the gears as they roll through mesh. That
is, variations cause the driven member of a mesh to
be in a slightly advanced or slightly retarded position
with respect to where it would be if both gears were
perfect. This advance or retardation is known as
transmission error and can be measured directly, if
the gears have been instrumented with position
measuring devices. Transmission error can result
in audible noise as the gears are run, in addition to
mis--positioning of the gear. In some applications,
gear variations can cause unpredicted loads on
machine components and contribute to premature
failure.
9.3 AGMA quality number
Fine--pitch gears are usually toleranced by specifying the parameters associated with a composite
action test as described in ANSI/AGMA 2000--A88.
Gear designers are encouraged to read and
understand the information contained in this
inspection standard before attempting any gear
design.
9.4 Fundamentals of composite testing
The composite inspection of gears is known by
various names. They include: composite action
test, double flank test, rolling test and various
combinations of these terms. The test is based
upon a fundamental geometric property of involute
gears: if two perfect involute gears (of the same
base pitch) are rolled together in tight mesh, the
center distance remains a constant. The gear
rolling fixture is a device based upon this principle
which measures the variation of tight--mesh center
distance when a master gear is rolled with a work
gear. The master gear variation is assumed to be
negligible compared to the work gear variation.
Total composite variation, tooth--to--tooth variation
and tolerance on test radius are limits specified for
the variation in center distance as the work gear is
rolled with a master gear on a gear rolling fixture.
Note the rolling test is done with the gears in tight
mesh (double flank) and with a center distance that
is allowed to vary. This test does not replicate the
action of most gears used in machines. In most
applications, gears are run single flank, with
64
backlash, on fixed centers. Therefore, the rolling

test must be correlated to the true functionality of
the gears.
The AGMA established a standard technique for
calculating the tolerances of gears. The result is an
AGMA Quality Number which is a concise means of
specifying the tolerances which define the quality of
the individual gear. This standard technique is
described in ANSI/AGMA 2000--A88.
However, the AGMA technique for establishing
tolerances is derived from typical tolerances available from standard manufacturing processes. In
other words, each AGMA Quality Class generally
reflects the capability of a gear machining
operation. Use of this tolerancing technique has
been extended to gear manufacturing processes
that do not involve machining of the teeth such as
injection molding, powder metal pressing and die
casting.
Note that the total composite variation, tooth--to-tooth composite variation and test radius tolerance
may be chosen independently. That is, the total
composite variation may be chosen to be that of one
class and the tooth--to--tooth variation of another
class.
Further, the value for any of these parameters used
on a drawing is not required to be that of any
particular class. The AGMA system is a standardized system available if the designer chooses to use
it to determine the tolerances for a gear.
9.5 Functional tolerancing
The gear designer is advised to be certain that the
tolerances placed on gear drawings are derived
from an analysis of the function of the gears being
toleranced. The AGMA system is based upon
manufacturing considerations, not on functional
considerations.
10 Materials and heat treatment

10.1 Selection of materials
The application usually determines what material is
required. In power transmissions, the criteria for
rating gears is strength and wear. Therefore, if a
gear is to operate for a particular application, the
material must possess adequate strength and wear
properties. Materials cannot be selected on

strength and wear properties alone. Some applications may require the material to have corrosion
resistance or be light in weight. Because applications vary, no one material can be used for every
application.
For most iron and steels, refer to ANSI/AGMA
2004--B89, Gear Materials and Heat Treatment
Manual.
10.2
Types of steels not
ANSI/AGMA 2004--B89
included in
10.2.1 Extreme temperature draw (ETD) steels

Steels can be purchased based upon physical
properties. Usually, the properties are obtained by
austenitizing, quench and tempering or heavy cold
working followed by stress relieving. The guaranteed physicals of ETD are obtained by subjecting
4100 series steel to an elevated temperature draw.
The microstructure is acicular ferrite and fine
pearlite. The resultant strengths are 150,000 psi
minimum tensile for ETD 150 and 180,000 psi
minimum tensile for ETD 180. These steels have
excellent strength and toughness.
The ETD family of steels is used in fine--pitch gears
to provide the necessary physical properties while
bypassing the usual heat treating process after
cutting the teeth.
10.2.2
Stainless steels not included in
ANSI/AGMA 2004--B89
10.2.2.1 300 series
The steels of the 300 series are not heat treatable,
are non--magnetic, and are used for maximum
corrosion resistance. Types 303, 304 and 316 are
very popular for fine--pitch gears when maximum
corrosion resistance properties are required. Due
to their poor machinability, they are used only when
the application requires. The 300 series austenite
steels are susceptible to galling by contact with
each other or with dissimilar metals.
10.2.2.2 400 series
The steels of the 400 series are magnetic and are
heat treated before or after machining, depending
upon hardness required. Types 416, 420 and 440c
steels are used when higher wear resistance and
load carrying capacity are desired.
The
AGMA 917--B97
machinability of this series of stainless steels can be

improved to some degree by heat treating to Rc 20
to 28.
10.2.2.3 Corrosion resistance
Stainless steels should always be passivated for
maximum corrosion resistance. It is advisable to
passivate stainless steel after all machining operations have been completed to insure the ultimate
anti--corrosion qualities of the metal. This process
is used to remove small particles of foreign matter
which have become imbedded in the material during
machining.
10.2.3 Tool steel
Tool steel blanks are being used for fine--pitch gears
for load, life and wear resistance. There are many
applications when tool steels are superior to other
steels.
10.2.4 Cast iron and nickel cast irons
Cast iron and nickel cast iron blanks for fine--pitch
gears are used when wear resistance and good
machinability are required. Complicated shapes
can be cast more economically than machining the
shapes from bars.
Furnace treatment is the usual method for
hardening.
10.3 Aluminum
10.3.1 Alloyed aluminum
Alloyed aluminum is the most common light metal
for gears. Three prominent grades are 6061--T6,
2024--T4 and 7075. Occasionally, blanks are made
from 356--T6 castings. Centrifugal casting reduces
the problem of porosity.
10.3.2 Operating life
Operating tests have shown that gears made of
these materials, carrying light loads and with
suitable lubrication, can be used in the as-machined condition for a few million cycles of
operating life.
Greater wear life can be obtained by having the
gears hard anodized. Dimensional build--up may be
as small as 0.0001 per surface. The result is a hard
aluminum oxided surface, resistant to salt atmosphere corrosion and wear. Lubrication is important
for all applications involving gears and particularly
65
AGMA 917--B97
here to help prevent the oxide coating from breaking

into small abrasive particles.
10.4 Powder metal (P/M)
Metal powders, when properly mixed, compacted,
sintered and heat treated, develop a high resistance
to impact and wear. Powder metal gears are one of
the many parts produced to take advantage of this
method of manufacturing.
Briefly, the steps
involved in producing a powder metal gear are:
-- Mixing together a base material, alloys and
other additives;
-- Compacting the powder mixture in a single or
multiple steps;
--
Sintering the gear.
After sintering, the powder metal gear is ready for

use. However, secondary operations such as
coining, oil impregnation, heat treating and machining can be performed to enhance the mechanical
and physical properties.
Since the base material, alloys and additives are
fine powders, it is relatively simple to alter the
percent composition of a mixture. By doing so, the
designer can test various powder metal gears and
determine which composition best suits an application. Because of the nature of powder metal, gears
can be made with mechanical properties similar to
ferrous or nonferrous metals. Furthermore, it is
possible to make a powder metal gear having
special attributes such as corrosion resistance or
having special sound dampening characteristics.
10.4.1 Hardness
Hardness checking of powder metal gears is difficult
because of the surface porosity. Errors in hardness
measurement occur when using the standard
indenters. Since the powder metal gear has more
voids than a traditional metal, the indenter penetrates deeper than it should, causing error in the
reading. The Metal Powder Industry Federation
recommends that a chisel shaped stylus 0.050 inch
wide with a 0.0005 inch radius tip be used. This
produces a more accurate measurement because
the stylus spans a long distance rather than an area,
thus avoiding the pores on the surface. Surface
hardened gears need special attention because the
indenter may penetrate the case, thereby yielding
an inaccurate measurement.
66
Due to the structure of powder metal parts, the

hardness is specified as apparent as opposed to
actual. Superficial as well as microhardness testing
is employed.
10.4.2 Fatigue properties
Gears which are cyclically loaded should be rated
based on fatigue strength. Powder metal gears
have relatively poor fatigue strength because of
porosity. One way to improve strength is to increase
part density. This can be accomplished by double
pressing the gear rather than single pressing during
the compacting part of the powder metal process.
10.4.3 Lubricating characteristics
Powder metal gears can be oil impregnated to allow
for self lubrication. Possibly, for certain low speed
applications, grease or an oil bath could be omitted.
Careful design of the die and punch can produce a
gear with an area(s) that is dense and strong and
another area(s) that is self lubricating.
10.4.4 AGMA quality numbers
AGMA quality numbers range from Q5 to as high as
Q9 under special conditions. Inspection is accomplished using standard techniques for any metal
gears except as noted in 10.4.1.
10.5 Plastics
10.5.1 General
The use of plastics as materials for fine--pitch gears
is growing. Improved materials, advances in gear
mold design and molding technology, and the
successful use of plastics gears in many
applications have contributed to this growth.
Information in this Design Manual was obtained
from manufacturers of polymers, composites, standard shapes and compounds, as well as of plastics
gears themselves.
Optimum performance of
plastics gearing requires the proper design and
selection of both materials and manufacturing
processes. To add strength to plastic teeth, a full
fillet radius is often used in the roots of the teeth.
10.5.1.1 Uses of plastics gearing
In the power transmission field most molded gears
are used in the fractional horsepower area. Many
cast and machined gears are used in higher
horsepower applications.
10.5.1.2 Advantages of plastics gears

Some reasons for the wide use of plastics gearing
are as follows:
--
Low cost;
--
Resistance to corrosion;
--
Lightweight (about 1/7 the density of steel);
--
Low inertia;
-- Inherent lubricity and compatibility with commercial lubricants, most chemicals and common
solvents;
--
Potential noise reduction;
--
Reduction of wear in mating gears;
-- Overload protection (acting as a sacrificial

member);
--
Color to aid in assembly;
--
Low maintenance.
10.5.1.3 Molded gearing

Molded gears can be combined with other features
to create multifunctional parts, and can be produced
in large quantities by the molding process.
10.5.1.4 Machined gearing
Many basic polymers are available in various
shapes from which gears can be machined using
essentially standard gear cutting equipment. This
makes it possible to use machined plastics gears in
quantities and sizes not practical for the molding
process.
10.5.1.5 Limitations of plastics gearing
A major limitation of plastics gearing is reduced
tooth strength in comparison with ferrous gearing.
The effects of temperature and moisture on plastics
materials used in gearing are not the same as for
metals. It should be recognized that temperature
and absorption of moisture affect the size and
strength of plastics. In addition, plastics are
relatively poor conductors of heat. A temperature
buildup developed through the working of the gear
may be difficult to dissipate. These limitations may
be overcome through the proper selection of the
basic polymer and the addition of reinforcements,
lubricants or other additives.
NOTE: Care must be exercised when designing plastics gearing to account for dimensional and strength
AGMA 917--B97
changes due to temperature or humidity. The dimensional changes can either be growth or shrinkage and
are generally predictable. To insure proper meshing of
gears at either extreme requires adequate clearance
which should be properly specified. Careful material
selection can enhance the long term life of plastics
gearing, and such materials are used with and without
reinforcements and other additives.
10.5.1.6 Specifications for plastics gears

The proper use of plastics in gearing requires the
adequate specification of the gear characteristics
actually needed. It is highly recommended that all
users of plastics gearing utilize a proper drawing
specification format.
10.5.1.7 Available types of plastics gears
Any gear form which has been cut from metal can
be cut from properly selected plastics materials.
Many of the gear forms can be molded providing a
suitable moldable material is selected. Certain gear
forms, such as a throated wormgear which is
conforming to the mating worm, are usually not
considered practical as one piece moldings.
10.5.2 Plastics materials
Gears are made from various plastics, both thermosetting and thermoplastic, with the latter by far the
most widely used. These materials are available as
unfilled polymers for use in molded and machined
gears. Polymers can be reinforced with glass
fibers, glass beads, milled glass, carbon fibers,
fabrics and mineral reinforcements and fillers.
Reinforcement may improve dimensional accuracy
in some cases, because of the reduction in
shrinkage. It may harm dimensional accuracy in
other cases, where non--uniform fiber orientation or
distribution results in non--uniform shrinkage. Most
machined gears are not made from glass or mineral
reinforced plastics since they are difficult to
machine and the exposed reinforcement on the
machined tooth profile may be abrasive to the
mating gear as well as to the cutting tool.
Lubricants such as polytetrafluoroethylene (PTFE),
silicone, molybdenum disulfide and graphite are
often added to improve the inherent lubricity of
plastics. Additives for stabilization (temperature
and ultra--violet light) and coloring are also used.
The list of candidate materials for plastics gearing
continues to expand.
67
AGMA 917--B97
10.5.2.1 Thermoplastic materials

A thermoplastic (T/P) material is one that will
repeatedly soften when it is heated and will harden
when it is cooled. In molding, it undergoes a
physical change only. The most commonly used
thermoplastic materials for gearing are as follows:
-- Nylon (PA) is a family of thermoplastic polymers. The most widely used of any molded gearing material is nylon 6/6, but nylon 6 and nylon 12
are also used. Some nylons absorb moisture
which may cause dimensional instability. Nylon
may be compounded with various types and
amounts of glass reinforcing materials, mineral
fillers, and such lubricants as PTFE and MoS2
(molybdenum disulfide).
-- Acetal (POM) has a lower water absorbtion
rate than nylon and, therefore, is more stable after molding or machining. Acetal polymers are
used unfilled or filled, with glass and minerals,
with and without lubricants, such as PTFE and
MoS2, as well as one version with fibrous PTFE.
-- Polycarbonate (PC) is generally used with
the addition of glass fiber and/or PTFE lubricant
and is a fine, low shrinkage material for producing
consistently accurate molded gears.
-- Polyester (PBT and O\PET) are both unfilled and with glass fiber, and are finding their
way into more markets as a molded gearing
material in competition with nylon and acetal.
-- Polyethylene,
ultra--high
molecular
weight (UHMW/PE) is a low moisture absorption
material that has high abrasion and wear resistance. Most gears of UHMW/PE are machined
from extruded bar or slab stock. Some parts are
made by injection and compression molding
techniques.
-- Polyphenylene sulfide (PPS), when compounded with 40 percent glass fiber with or
without internal lubricants, has been found in
certain gear applications to have much greater
strength, even at elevated temperatures, than
most materials previously available.
-- Polyurethane (TPU) is generally noted for
its flexibility and, therefore, has the ability to
absorb shock and deaden sound.
-- Polyester elastomer (TPE) is a newcomer
to the gearing field, and has excellent sound
68
deadening qualities and resistance to flex

fatigue, impact and creep, among other
advantageous characteristics.
-- Styrene--acrylonitrile (SAN) is a stable, low
shrinkage material and is used in some lightly
loaded gear applications.
Annex A of ANSI/AGMA 2004--B89, Gear Materials
and Heat Treatment Manual, has a detailed
explanation of these thermoplastics.
10.5.2.2 Thermosetting materials
A thermosetting material is one that, during molding, will undergo a physical change as well as a
chemical reaction called polymerization and cannot
be remolded by reapplication of heat. Common
thermosetting materials used in plastics gearing are
phenolic (PF) and polyimide/thermoset (PI).
-- Phenolics (PF) are invariably compounded
with various fillers such as woodflour, mineral,
glass, sisal, chopped cloth and such lubricants
as PTFE and graphite. Phenolics are generally
used in applications requiring stability, and when
higher temperatures are encountered.
-- Polyimide (PI) is usually 40--65 percent
fiberglass reinforced and has good strength
retention when used at high operating
temperatures.
For more information, see annex A of ANSI/AGMA
2004--B89.
10.5.3 Molded plastics gearing
10.5.3.1 Molding considerations
The molding of gears requires special techniques
and knowledge. Processing techniques vary with
molding materials, gear geometry and size, and
mold design. Gear molding is a specialized molding
service.
10.5.3.2 Molding process
Molded gears are usually produced by injection or
transfer molding methods in which the plastics
molding material is forced under pressure through a
gate(s) into the mold cavity. The teeth of a molded
external gear are formed by the metal ring in the
cavity in which the gear teeth have been reproduced. The teeth of an internal gear are formed by
the teeth in the metal core of the cavity set.
10.5.3.3 Mold design

Accurate molded gears require molds which are
properly designed and accurately manufactured.
Mold design must take into account methods of
gating and other elements of the fill system,
temperature control, and the ejector system. Cavity
teeth must be precisely altered in form and changed
in size to assure that the molded gear, after
shrinkage takes place, assumes the desired configuration. Accurate gear mold cavities can be
produced by grinding, electric discharge machining
(EDM) or electroforming.
10.5.3.4 Shrinkage during molding
Gear molding materials may shrink from approximately 0.001 to 0.035 of an inch per inch from the
cold mold dimension to the cold gear dimension,
depending upon the material selected, the molding
method used, and the cross--sectional mass of the
gear. These same materials, when reinforced with
glass fibers or other high aspect ratio fillers will
shrink less in percentage but may shrink non-uniformly because of flow orientation of the
reinforcements. Therefore, a thorough understanding of the molding process, the selected material,
and the required gear configuration is vital in order
to predict whether the desired accuracy can be
achieved. This will determine the size and the gear
profile of the cavity required.
10.5.3.5 Accuracy of molded gears
Molded gearing accuracy depends upon selection
of a suitable material ---- design of the gear,
determination of the shrinkage of the plastics
material, good mold design, accurate mold construction and use of molding techniques applicable
to the molding of gears.
To obtain greater accuracy in a molded gear, the
mold should be designed to have uniform cross
sections, optimum gating location and adequate
ejection locations. The material selected should
display a highly consistent shrinkage and should be
inherently stable under the environmental and use
conditions.
Usually, the greater the accuracy required in a gear,
the fewer the number of cavities in the mold. For
example, an eight (8) cavity mold might well be used
AGMA 917--B97
for an AGMA quality six (Q6) gear, while a single

cavity might be used for a quality ten (Q10) gear.
10.5.3.6 Mold economics
Most plastics gears are specials, inasmuch as
molds for gears of all pitches and diameters and
face widths are not commonly available. This
means that usually a special mold must be purchased for each gear and, therefore, the molding
process on an economic basis is usually limited to
larger production quantities of gears. A molded
gear from a single cavity mold will be more
expensive than the same gear produced from a
multicavity mold.
10.5.3.7 Use of inserts
Molded--in metallic inserts, including simple bushings, sleeves and shafts, are used to achieve
advantages such as:
more accurate bores,
reduced thermal effects because of the conductivity
of the metal, more closely controlled shrinkage, as
in the case of a solid metal gear blank with molded
plastics teeth, improved dimensional stability and
usually greater load carrying capacity.
When contemplating the use of inserts, consideration must be given to the added cost of the insert,
as well as the cost of inserting it into the mold during
the molding operation, or of assembling it as a
post--molding operation.
10.5.4 Machined plastics gearing
10.5.4.1 Machining techniques
These materials do not require special machines or
cutting tools. Gear shapers and hobbing machines
using properly selected and sharpened cutting tools
at prescribed feed and speed rates will produce
quality gears. Materials manufacturers generally
recommend the use of coolants during machining in
order to avoid stress cracking and melt outs. Climb
cutting hobbing technique is also recommended for
improved surface finish. It may be advisable to
consult with the materials manufacturers for complete fabrication recommendations.
10.5.4.2 Molded blanks
Thermoset and thermoplastic gear blanks produced by the molding process, and cut by
conventional hobbing and shaping machines, offer
alternatives to the expense of a toothed gear mold
69
AGMA 917--B97
for low quantities. Special gear requirements such

as throated wormgears or double enveloping
worms which may not be practical for molding, are
machined from such blanks.
NOTE: Blanks should be designed and molded with
special attention to preventing voids and excessive
distortion after machining.
10.5.4.3 Gear and blank accuracy

End product accuracy is primarily dependent upon
gear blank quality and process controls. Gear tooth
finishing techniques common to metal gears, such
as grinding, honing and lapping, are not usually
performed on plastics gearing.
10.5.5 Other manufacturing methods
In addition to the molded and machined plastics
gearing previously discussed, other manufacturing
methods are also used. As Cast and Forged
plastics gears will not be as accurate as machined
or injection molded gears. Stamped and Pressed
and Sintered plastics gears are being manufactured to accuracies comparable to metal gears
produced by the equivalent process.
10.5.5.1 As cast
This method is used with urethanes and cast nylon
materials. The mold, which is usually machined in
metal, is engineered to provide adequate shrinkage
allowance for the material being cast. The basic
procedure closely resembles cast metal processes.
Secondary operations usually include boring, key
seating and machining of set screw holes.
10.5.5.2 Forgings
This method is used to produce special sprockets
and gears for applications such as snowmobiles.
This process is similar to metal forging. A billet of
the desired material is heated below its melting
point, then pressed in a metal--forming die.
Secondary operations usually include boring, key
seating and machining of set screw holes.
10.5.5.3 Stamping
This method is identical to that used for metal gears
in instrument drives, meters, switches and timers.
Stamped gears are usually finished in one operation
depending on the quality of the die set and the
70
accuracy requirements of the part. The stamping

process is generally limited to gears of narrow face
widths.
10.5.5.4 Pressed and sintered
This method is similar to powder metal technology.
Specially blended powders are compacted in a die,
then sintered. They can be oil impregnated since
the material is porous. Secondary operations will
vary, depending on the complexity of the part, and
the limitations of the tooling.
NOTE: The size limit of injection molded gearing is
usually determined by the part wall thickness. The size
limits of machined gearing is determined by the size of
available extruded or cast shapes and parts. Not all
plastics materials can be used for the manufacturing
processes identified in this section. Discussion between designer and manufacturer will usually result in
considerable time savings and mutually acceptable
parts.
10.5.6 Inspection
Plastics gears must be accurate to wear well,
operate quietly and transmit uniform motion. In
order to meet these objectives, the understanding
of variations, tolerances and inspection is necessary. Refer to ANSI/AGMA 2000--A88, Gear
Handbook -- Gear Classification, Materials and
Measuring Methods for Unassembled Gears for
additional information.
NOTE: Inspection of plastics gears must be done in a
manner which accounts for size variation with temperature and moisture content. It is commonly accepted practice to use about one--half the tight--mesh
load recommended for metal gears when measuring
plastics gears on a center distance measuring instrument. Refer to ANSI/AGMA 2000--A88. Care must be
taken not to distort the gear teeth or mounting arbor of
the work gear by application of excessive tight--mesh
applied load.
11 Manufacturing methods
11.1 General
Many methods are used to produce the finished
gear teeth including: hobbing, shaping, milling,
fly--cutting, broaching, casting, powder metal process, molding, stamping, cold rolling, grinding,
shaving, honing, lapping and burnishing. This
clause reviews the most common methods used to
manufacture fine--pitch gear teeth but is not

intended to be all inclusive.
11.2 Hobbing
The generation of a gear tooth is a continuous
indexing process in which both the cutting tool and
the workpiece rotate in a constant relationship while
the hob is being fed into the work. As the hob is fed
across the work, all the teeth in the work are
completely formed. Hobbing is normally limited to
external gears. Caution should be used by the
designer to allow ample room on each side of the
gear for entry and exit of the cutting tool.
11.3 Shaping
Gear shaping is a generation process where the
workpiece and the toothed circular cutter progressively index in a timed relationship simultaneously
with a reciprocating action of the cutter. This
reciprocating action permits the cutting of gears that
are closely banked against obstructions or other
gears of a cluster. Moreover, the nature of shaping
allows the generation of internal teeth.
11.4 Milling
Milling gear teeth is a form cutting method using a
cutter having a profile matching the gear tooth
space. One tooth is cut, the blank is then indexed,
usually using a dividing head, and the next tooth is
cut, continuing around to complete the gear.
AGMA 917--B97
11.8 Powder metal process

In the powder metal process, metal, which has been
reduced to powder particles, is compressed into a
mold cavity. After compression, the part is heated
or sintered to allow the powder particles to bond or
adhere to each other into a homogeneous structure.
11.9 Molding
Gears are currently produced of thermosetting and
thermoplastic materials, using the compression,
injection and transfer molding processes. Heat and
pressure in combination are used to force the
plastics material into a mold cavity. The large
number of materials available has rapidly increased
the use of this process for all kinds of gears.
11.10 Stamping
Using a punch and die, gears may be produced from
thin sheet stock materials. To improve the sheared
surface a second operation using a trim die is
sometimes incorporated into the production process. Fine blanking is a specialized process to
produce the completed gear essentially burr free in
one operation.
11.11 Cold rolling
Fly--cutting may be a form cutting or generating

method using a single tooth cutter.
Cold rolling is the forming of gear teeth by controlled

displacement of metal. Dies, either in circular or
rack form work in the tooth space of the part, forcing
metal to flow and form teeth. The cold rolling
method can be used either to make a gear complete
or as a finishing process on teeth already cut. The
designer should be aware that special procedures
and material are required for the blank.
11.6 Broaching
11.12 Cold drawing and extrusion
11.5 Fly--cutting
Broaching is a machining method by which successive in--line teeth, or a broach, is pulled or pushed
over or through a gear blank. The broach is
designed to remove successive small amounts of
material in a single stroke. The last several rows of
teeth are used for sizing to the final configuration.
11.7 Casting
Casting of gear teeth is accomplished by pouring or
forcing molten metal into a cavity. The quality of the
gear varies largely with the quality of the cavity and
the process used.
Cold drawing and extrusion of gears is extremely

versatile in that almost any shape desired can be
reproduced. The material is forced through one or
more die blocks with the last block giving the final
shape.
11.13 Grinding
The grinding of gear teeth is accomplished by three
different methods. The first is a continuously
formed grinding wheel resembling a large diameter
worm which is fed into the work similar to the
method described in hobbing. The second incorporates a grinding wheel with the proper tooth space
71
AGMA 917--B97
form dressed on the perimeter of the grinding wheel.

In the third, two disc type wheels are positioned so
their axes are perpendicular to the profiles of a rack
tooth and the working portion of the wheels serve as
the rack profiles. Grinding is usually used as a
finishing process but in many fine--pitch applications, it may be used as the complete process to
manufacture the teeth. Caution should be used by
the designer to allow ample room on each side of
the gear for entry and exit of the grinding wheel.
11.14 Gear shaving
Shaving is a free--cutting gear finishing operation
which removes small amounts of metal from the
working surfaces of the gear teeth. It is used to
improve the accuracy of index, helix angle, tooth
profile and concentricity, and to improve the tooth
surface finish.
11.15 Gear honing
The honing process uses an abrasive impregnated
plastic helical gear--shaped tool. This tool is run in
mesh with a hardened gear to remove nicks and
burrs from teeth. It also improves the surface finish
to some degree.
12 Inspection
The methods of inspection should be considered
when gears are being designed, especially when
the specifications are being prepared. The methods
selected should be the minimum necessary to
establish the quality characteristics required by the
application. Needless inspection can add substantially to the cost of the gears, whether performed by
the manufacturer or the user. Other considerations
in the selection of inspection are the availability of
the measuring instruments and the suitability of the
measuring process to the size of the gear teeth.
The standard and preferred measurements are the
composite check and the test radius measurement.
Other commonly used procedures are the measurement over wires, measurement of runout by
indicator and the comparator method.
It is less common, although sometimes desirable, to
measure involute, tooth alignment and tooth placement (pitch variation or index position) by elemental
methods. It is possible to measure teeth as small as
100 to 120 DP. Tolerances for these elemental
measurements do not exist in AGMA for teeth finer
than 20 DP. Elemental inspection tolerances for
gears finer than 20 DP must be agreed upon
between manufacturer and user.
12.1 Composite action test, double flank check
11.16 Gear lapping

Lapping is a method of correcting minute heat-treatment distortions of involute profile, helix angle,
spacing and eccentricity. It is done by either of two
methods. The first method is running the work in
mesh with a gear--shaped lapping tool and the
second is to run two mating gears together. In both
methods an abrasive lapping compound is the
medium that accomplishes the removal of metal.
11.17 Gear burnishing
Burnishing is a finishing operation whose sole
purpose is to produce a smooth uniform surface. It
cannot be used to correct results of improper
cutting. The workpiece is run in metal to metal
contact under light pressure with a special hardened burnishing gear.
72
See ANSI/AGMA 2000--A88, Gear Classification

and Inspection Handbook for detailed descriptions
of measuring methods, master gear tolerances and
inspection tolerances.
12.2 Measurement of runout by indicator
See ANSI/AGMA 2000--A88. However, when
inspecting wide--face gears, it may be desirable to
check at both ends of the teeth. A marked
difference in the two sets of readings, in magnitude
or angular position of the high point, reveals the
presence of a wobble condition. This condition,
incidentally, detracts more from the quality of the
gear than does uniform across--the--face runout
because it prevents uniform contact across the
entire gear face.
12.3 Measurement over wires
See ANSI/AGMA 2002--B88, Tooth Thickness
Specification and Measurement, for detailed in-
formation on the specification and measurement of

tooth size.
12.4 Optical comparator measurement
12.4.1 Use and limitations
These measurements are made in an optical
comparator which projects an enlarged image of
part or all of a gear outline onto a screen. The image
can be compared to a transparent layout of the
desired outline drawn to the same scale. A variety
of gear quality characteristics can be measured,
including gear size (tooth thickness), runout, pitch
and profile variation. While the comparator method
is useful in checking very fine--pitch gears, it is not
an adequate substitute for the composite check.
The method is limited to spur gears. It is generally
not practical to obtain a suitable image from helical
gears, from either the normal or transverse plane.
Some standard magnifications used on optical
comparators are 31.25 and 62.5 derived from the
fractional scale and 10, 20, 50 and 100 derived from
the decimal scale. The advantage of a larger scale
is often offset by the corresponding difficulty in
projecting a sharp image and by the limitation
imposed by the comparator screen size on the
portion of the gear which can be viewed at one time.
Even with the largest scale, outline variations less
than 0.0005 inch cannot readily be detected by
projection.
A proper check of the gear outline by the projection
method requires that the gear image be in proper
relationship with the layout. Ideally, the gear axis is
the reference feature on the gear. When this is not
practical and the outside diameter is used instead,
the tolerance limits on the size and concentricity of
the gear outside diameter must be taken into
account.
12.4.2 Comparator layouts
Computer Aided Design (CAD) software packages
are frequently used to create the profiles of gears for
use with optical comparators. Figure 56 is an
example of this method. The tooth shown is for a 10
tooth, 20 DP, 20 degree PA gear, with tooth
thickness 0.07854 in. This profile depicts the tooth
as if it were cut with a rack--type cutter with tooth
thickness of 0.07854. The tool had an addendum of
0.062 in and had zero tip radii. The parameters of
AGMA 917--B97
the cutter need to be known because the computer

program computes the path of the elements of the
cutter surface as it is rolled through mesh with the
blank in order to create the surface of the tooth. If a
computer program does not request cutter information as part of the input, the program contains
assumptions with regard to cutter parameters. The
gear designer should determine what cutter is
modeled by the software before proceeding with the
design.
It is recommended that reference features (such as
the gear center) and tooth centerlines be part of the
layout so the user of the layout can properly orient
the chart to the projected image.
The mathematical algorithms which a computer
uses to model the tooth create a series of points that
are plotted to form the picture. Figure 56 contains
tables which show the points plotted in this
example. Note that this picture shows overlap of the
involute profile with the trochoid due to the drawing
method used.
12.4.3 Limitations of CAD gear profile software
The gear designer should carefully evaluate the
quality of the software used to draw gear teeth. The
software used in figure 56 did not calculate the
intersection of the trochoid and involute profile.
Each curve was calculated past the intersection and
plotted. The CAD operator must manually trim the
curves to the proper intersection point if the
application requires. The most accurate tooth
drawing programs calculate all intersections and
trim the curves to the intersection points. These
software packages are recommended if molds are
being made directly from the design software.
Less rigorous software may calculate the involute
properly but not attempt the trochoid. These
programs usually use a radial line from the base
circle to the root circle. Even less rigorous software
uses the equations for approximating a gear tooth
found in text books for drawing board drafting
courses. The use of such software is discouraged.
12.5 Elemental measurements
If equipment is available, it is possible to measure
profile variation, tooth alignment variation, index
variation and pitch variation by the use of a
contacting probe on individual teeth. This is
important when trying to diagnose problems with
gear quality.
73
AGMA 917--B97
Outside diameter
X
Y
0.300
0.000
0.299891 0.008080
0.299640 0.014687
Involute Profile
X
Y
0.299640 .014687
0.290149 .021052
0.281099 .026318
0.272587 .030569
0.264703 .033897
0.257527 .036401
0.251127 .038185
0.245563 .039361
0.240883 .040043
0.237122 .040348
0.234306 .040398
0.232166 .040290
0.231457 .040290
Trochoid
X
Y
0.242635 0.044656
0.234305 0.040398
0.214320 0.034128
0.203145 0.033447
0.196298 0.034428
0.191806 0.035934
0.188707 0.037593
0.186493 0.039280
0.184878 0.040960
0.183696 0.042632
0.182848 0.044313
0.182278 0.046028
Root circle
X
Y
0.182278 0.046028
0.180464 0.052694
0.178799 0.058095
Y
(0 1823 0 0460)
(0.1823,0.0460)
(0.2343, 0.0404)
(0 2 0)
(0.2,0)
(0 3 0)
(0.3,0)
X
0.200 to gear axis
RC
Trochoid
Involute
OD
Figure 56 -- Profile of 10 tooth, 20 DP, 20 PA gear tooth with undercut

The following are brief descriptions of the above
variations. See ANSI/AGMA 2000--A88 for detailed
descriptions on measuring methods and
interpretations.
12.5.1 Profile variation
Profile variation is the difference between the
measured and the specified involute profile on the
portion of the tooth flank extending from the
specified form circle to the addendum circle or start
of tip chamfer or tip round.
12.5.2 Tooth alignment variation (lead variation)
Tooth alignment variation is the difference between
the measured lead trace and the specified lead
trace, with the difference measured normal to the
specified lead trace. The tooth alignment traces are
taken along the helix on helical gears and across the
face on spur gears.
12.5.3 Index variation
Index variation is the displacement of any tooth from
its theoretical angular position relative to a datum
tooth about an axis established by a specified
74
surface. For racks, linear position is used in place of

angular.
12.5.4 Pitch variation
Pitch variation is the difference between two
distances: the theoretical distance between corresponding points on adjacent teeth and the
measured distance between those points.
13 Lubrication
13.1 Introduction
To achieve maximum reliability and minimum cost in
a gear design, the lubrication function should be
given the same degree of attention as is given to the
gear selection and design, and should be approached as a system. The lubrication system
consists of the lubricant, its means of application, its
means of retention, the gear materials and the
operating environment.
In gear design, the lubricant may provide any or all
of the following functions:
-- To reduce the friction between the gears and
within the bearings;
--
To minimize wear between contacting parts;
-- To remove wear particles or ingested

contaminants;
--
To protect surfaces from corrosion;
--
To remove heat or to preheat.
13.2 General considerations

The following factors govern the selection of the
lubrication system to be used:
--
Operating temperature range;
--
Tooth loading (K factor);
--
Pitch line velocity;
-- The degree to which the gears are enclosed,

and if enclosed, the degree to which the
enclosure can be sealed;
-Operating life between reapplications of
lubricant;
--
Operating environment.
Methods of lubricating fine--pitch gearing are the

following:
--
Force feed (pumped) systems;
--
Splash systems;
--
Grease packed systems;
-- Intermittent
systems;
oil
or
grease
application
--
Sintered materials (oil impregnated);
--
Plastics materials (lubricant filled);
--
Bonded coatings;
--
Solid lubricants;
--
Non--lubricated systems.
13.3 Lubricating system considerations

The following should be considered when selecting
the lubricating system, the lubricant, or the gear
materials:
-- Temperature range. Determine the limits of
the range of temperature over which the gears
are expected to operate, temperatures to which
they may be subjected during storage or shipment and the magnitude of any unusually high or
low temperature excursions. Once these values
have been established, lists of lubricants and
AGMA 917--B97
their properties may be screened for suitable

candidates.
The maximum temperature that a gear will
achieve is a function of the heat that it will generate as a result of friction losses, the rate at which
it can reject heat to the heat sink and the temperature of the heat sink.
-- Tooth loading. Most lubricants act by separating the contacting rubbing surfaces with a
shear--film that prevents direct material--to--material contact. The degree to which wear of these
surfaces is minimized is a direct function of the
integrity of the lubricating film. Minimum wear is
achieved when the thickness of the film exceeds
that of the asperities on the surfaces of the contacting parts. The pressure to which the lubricant
film is subjected is the Hertzian contact pressure.
High pressures tend to squeeze the film to the
point that it will not adequately separate the
parts.
-- Meshing velocity. The rate of sliding at the
lubricating film is a function of the meshing velocity. In the case of fluid film lubrication, the film
thickness developed by a given load in a given lubricant at a given temperature is a function of the
meshing velocity and gear tooth geometry. Thus,
low speed applications tend to demand higher
viscosity materials in order to obtain an adequate
film thickness.
In the case of bonded films or dry (powdered) lubricants, high speed may preclude adequate lubrication. High speeds combined with high tooth
loading may generate sufficient heat to eliminate
from consideration all lubricating systems that
cannot actively carry away heat.
-- Gearing enclosures. The degree to which
fine--pitch gears can be enclosed dictates the
type of lubricating system that can be applied.
Gear enclosures range from designs that are fully enclosed and can be provided with shaft seals
and static casing seals to open designs wherein
no fluid type lubricants are possible or tolerable.
At times the lubricating system may dictate the
casing design.
13.4 Methods of lubricating fine--pitch gearing
There are three general approaches to the
lubrication of fine--pitch gearing.
75
AGMA 917--B97
13.4.1 Oil (fluid)

Oil may be used as a lubricant under the following
conditions:
-- Special materials bonded to the gear tooth

surfaces.
-- An enclosure adequate to retain the oil can

be provided, including dynamic seals at the
shafts and static seals at the split lines;
In general, these materials are limited to applications in which the tooth contact loads are low (low
K factors) and speeds are low enough so that the
heat resulting from meshing friction does not
damage the materials.
-- In the case of open gearing, where lubricant

fling--off is minimal or may not occur, grease applied manually by brush, spatula or pressure gun
may be considered;
In some cases, the actual number of operating

cycles may be low enough that a properly applied
coating will not have worn through by the end of the
service life of the gear.
-- The range of operating temperatures is

within the service range of the oil selected;
13.5 Characteristics of lubricants
-- The operating environment is tolerant of the

oil that may migrate past the seals or be spilled
during oil changes;
-- The operating speeds and loads are such as
to produce an adequate lubricating film
thickness.
13.4.2 Grease
Grease may be used as a lubricant under the
following conditions:
-- An enclosure adequate to retain the grease
can be provided, including dynamic seals at the
shafts and static seals at the split lines;
-- In the case of open gearing, the speeds are
not so high so as to throw off the grease, buttered--on grease may be considered;
-- The gear may receive re--lubrication when
needed;
-- The range of operating temperatures is within the service range of the grease selected;
-The operating environment is tolerant of the
oil or grease that may get by the seals or be
thrown from open, buttered gearing.
13.4.3 Non--fluid systems
Several methods are available to provide a low
friction contact between the surfaces of meshing
gear teeth. These include:
There are several characteristics that can be

measured for different lubricants, and that can be
compared, making possible an intelligent choice of
a specific lubricant for a given application.
13.5.1 Characteristics of oils
The following are the characteristics of oils for
application with fine--pitch gears:
-- Viscosity. A measure of the internal fluid
friction of an oil; it is the difficulty with which an oil
will flow at a given temperature.
The viscosity of an oil at low temperatures may
be so great as to require excessive power to turn
the gears it lubricates. At high temperatures the
oil may become so fluid that it cannot provide an
adequate film to separate the teeth of meshing
gears. Viscosity may be measured using several
distinctly different systems, i.e., Kinematic,
Saybolt Universal, Engler Degrees, etc. ISO recognizes Kinematic viscosity as the standard
measure of viscosity. Kinematic viscosity is
measured by the rate of flow through capillary
tubes. It is calculated in mm2/s (centistokes) and
is measured at standard Celsius temperatures;
-- Viscosity index. A measure of the change
of the viscosity--temperature characteristic of a
fluid, indicating the rate at which the viscosity of a
fluid will change as temperature increases or
decreases;
Certain plastics running against each other;
-- High temperature stability. A characteristic important in gears operating at elevated temperatures. Every oil has some temperature
above which its design characteristics will begin
to break down. In general, synthetic oils retain
their properties to higher useful temperatures;
-- Certain plastics filled with materials having

lubricating properties;
-- Compatibility with materials of manufacture.

The lubricant must be chemically
-- Certain metallic combinations such as brass

against steel;
--
76
AGMA 917--B97
compatible with metallic and elastomer components, sealants, paints and other applied coatings;
14 Bearings
-- Additives. Additives are blended into many

oils to improve specific characteristics and to
tailor the oil to a given application. Typical
additives are used to improve oxidation resistance, viscosity index, load carrying ability and
anti--foaming characteristics.
The bearings discussed in this clause are of the

rolling element type. However, some of the
information will also apply to journal bearings.
13.5.2 Characteristics of greases

Most greases consist of two principal parts, an oil
and a thickener, generally a metallic soap, which
acts like a sponge to retain the oil. The oil
component may be evaluated by the criteria given
above. A thickener (organic or inorganic) provides
body to a grease. Aluminum, bentone, calcium,
lithium and polyurea are common thickener types
used in modern greases. Each provides its specific
grease with distinctive properties.
The following are the characteristics of the grease
(thickener) component for applications of grease to
fine--pitch gears.
-- Penetration number. A physical test established by the National Lubricating Grease Institute (NLGI) to define the consistency (softness or
hardness) of a grease at 35C (77F). A specialized instrument, known as a penetrometer, is
dropped into a prepared sample of the grease
under carefully prescribed conditions. NLGI
penetration numbers range from 0 through 6
(very soft to very hard).
As the grease is worked at the mesh and near the
gear teeth in service, it is subjected to shear
forces that cause the thickener to release the lubricating fluid. When the shear forces are
removed, the thickener reabsorbs the lubricating
fluid. The degree to which a grease softens while
working will influence its ability to stay near the
areas of the teeth where it will be most beneficial.
Therefore it is necessary to select a grease that
will soften sufficiently to enter the mesh but
remain there under load;
-- Dropping point. The temperature at which
a grease passes from a semi--solid state to a liquid state. Dropping point helps determine the
maximum temperature at which the grease will
lubricate and remain in place without running out.
14.1 General
The bearing application process consists of selecting the type of bearing at each location, finding its
operating speed and loads, both radial and thrust,
and then selecting a size of bearing of adequate
capacity over the required life. In addition, bearing
lubrication must be assured.
14.2 Types of bearings
The selection of bearing type is commonly influenced by the magnitude of the bearing loads in
relation to the space available for the bearing. The
size of the thrust load is especially important.
Although thrust loads are not usually associated
with spur gears, some thrust will develop from slight
misalignment of the shaft, from teeth slightly out of
parallel or having a slight lead, or from shaft
bending.
The descriptions of the types of bearings given
below are only suggestive of their individual features. Further information should be obtained from
bearing manufacturers.
14.2.1 Angular--contact ball bearings
This ball bearing has large load capacity for
combined radial and thrust loads. An individual
bearing must have a thrust load, but will support this
load in one direction only. Where the thrust load
may reverse direction or where its direction is
uncertain, these bearings should be used in opposed pairs. The two bearings may be located at
opposite ends of the shaft or they may be placed
back--to--back with an additional bearing at the far
location. Sometimes two or more bearings are
arranged in tandem at one location to obtain the
necessary load capacity, as when a larger diameter
bearing would not fit into the housing design.
14.2.2 Deep groove radial ball bearing
This ball bearing has a smaller thrust load capacity,
but will support thrust loads in both directions in a
single bearing.
14.2.3 Cylindrical roller bearing
This bearing is normally used for its greater radial
load capacity. Since it has no thrust load capability
77
AGMA 917--B97
of its own, it must be accompanied by another type

of bearing, such as the deep groove radial. The
cylindrical roller bearing is also used when free axial
sliding is needed. In such applications, the axial
location of the shaft is fixed by the other bearing and
the cylindrical roller bearing prevents internally
induced thrust loads caused by assembly errors or
axial thermal expansion of the shaft.
14.2.4 Tapered roller bearing
This roller bearing adds high thrust load capacity to
the high radial load found in cylindrical roller
bearings. Their use in relation to thrust load
direction is subject to the same considerations
discussed under angular--contact bearings.
14.3 Bearing speed
Speed affects the bearing capacity in several ways.
The greater the speed, the greater the number of
stress cycles over the total hours of required life.
Also, the greater the speed, the more likely the
bearing will be harmed by misalignment, improper
bearing fit on the shaft, eccentricity of inner to outer
races and excessive bearing play.
Some of these effects are harmful not only to the
bearings. If they are of significant magnitude, they
will also adversely affect the gears supported by the
bearings.
14.4 Bearing loads
14.4.1 Sources of bearing loads
The common sources of bearing loads are:
-- Reactions to the gear tooth forces. With spur
gears, the tooth forces create reactions at the
bearings which are pure radial loads (except as
noted in 14.2). With helical gears, the forces also
have thrust components which induce thrust
forces on at least one of the bearings;
-- Weights of gears and other components
supported by the bearings;
-- Centrifugal forces due to out--of--balance
resulting from runout and other causes;
-- Reactions to external forces. Such forces
may come from other drive components
mounted on the same shaft as the gears, such as
belt or chain drives, or couplings with some
misalignment present. Other forces may come
78
from driving or driven components such as

motors or impellers.
14.4.2 Calculation of bearing reactions from
other forces
14.4.2.1 Weight forces
Locate each gear or other weight contributing
component in its axial position relative to the
bearings. Treating each weight as a vertical force,
combine these forces with the vertical components
of the gear tooth forces in the vertical plane. Then
continue with the remainder of the calculations.
14.4.2.2 Other drive component forces
For belt and chain drives, treat these forces in the
same manner as spur gear tooth forces after
establishing the appropriate force magnitudes and
directions. Coupling forces require a special
treatment which varies with the type of coupling and
are not discussed here.
14.4.2.3 Centrifugal forces
For each centrifugal force in its axial location along
the shaft, find its own set of bearing reactions, all in a
single plane. Since the directions of unbalance may
be the same for all the centrifugal forces, it is
appropriate to add up all the centrifugal force
reactions at each bearing. Since the centrifugal
force is a rotating force, its most harmful effect will
take place when it lines up with the resultant of all
the other bearing reactions. Therefore, add the
combined centrifugal force reaction to the
corresponding bearing reaction resultant.
14.5 Bearing capacity
Information on the capacity of a specific bearing is
available in the bearing manufacturers design
manual. The manual will also describe the calculation procedure for combining the design radial and
thrust loads, bearing speed and required hours of
operation into a single figure for comparison to the
standard rating of each size of bearing.
14.6 Lubrication
In some applications, a single lubricant can be
shared by both the bearings and gears. The
condition of rolling and sliding between the various
elements of the bearing are similar to the rolling and
sliding in the meshing action of the gear teeth.
When the materials, temperature range and surface
texture are also similar, one lubricant can often

meet both sets of requirements. If freely circulating
oil is used, it must be kept clean enough for bearings
even if the gears are more tolerant of wear particles,
contaminants or other material.
When separate lubricants are used, care must be
exercised to prevent mixing. One way to do this is to
use sealed bearings packed with grease by the
bearing manufacturer. This procedure also insures
that the amount and type of grease in the bearing
best suits the bearing application and is kept
especially clean.
The Anti--Friction Bearing Manufacturers Association standards which establish the ratings for rolling
element bearings do not consider the effect of
inadequate lubrication. If there is only boundary
lubrication, bearing life may result in a lesser
standard rating. To insure the presence of a
lubricant film sufficient to fully separate the rolling
elements under high speed or heavily loaded
conditions, it may be necessary to use a jet stream
of fluid. A spray or splash system is usually suitable
for less critical conditions.
15 Load rating and testing

The preliminary design is only the first stage in the
total design process. A design should not be
considered as final unless it has been confirmed by
test experience. The load rating calculation is often
a valuable step before the testing and, under special
conditions, may be used in place of testing.
15.1 Preliminary design
The preliminary design is generally developed
directly from the performance requirements.
Sometimes, however, it can be taken directly, or
with some modification, from a design used in a
similar application.
15.1.1 Preparation of preliminary design
A new preliminary design can be prepared by
carefully following procedures of clauses 4, 5 and 6.
The selection of the basic design load will require
special care and judgment, backed by experience.
As explained in 5.3.1, this selection should include
suitable factors to give protection for expected
intermittent overloads, desired life expectancy and
AGMA 917--B97
safety. Additional caution is needed when the risks

are greater than normal. Examples of such high risk
applications are those involving human safety, risk
of damage to expensive equipment, anticipated
conditions of unreliable maintenance or inspection
and equipment that will not be accessible for prompt
servicing or replacement. The risks in such
applications are even greater when there is the
possibility of failure without warning.
15.1.2 Revisions of preliminary design
In some design problems, a preliminary design is
needed in order to establish some size requirements as a first step in the design of the complete
equipment. Such a preliminary design may not
have the benefit of adequate knowledge of loads or
other constraints imposed by the final design of the
equipment. In this case, the preliminary design
must be reviewed and revised in light of the added
information. When greater gear capacity must be
provided with minimum change to the rest of the
design, the best solution may be the improvement
of the gear material or its heat treatment.
15.2 Load rating
Load rating is the standard calculation of the load
capacity of a designed set of gears. The rating is
generally expressed as the horsepower that the
gear set can transmit at the specified speed and for
the desired life. This horsepower can then be
compared to the power requirement of the
application.
15.2.1 Strength conditions
The standard AGMA rating procedure considers
two strength conditions as determining the load
capacity of the gears, namely, bending and surface
durability. These are the same as the two used in
the preliminary design procedure as noted in 5.6.
15.2.1.1 Bending strength
Bending strength is the capacity to resist the kind of
failure appearing as a crack in the root fillet of the
gear tooth. Such a crack results from repeated
tensile stress cycles and can progress until all or a
major portion of the tooth breaks away. Major tooth
breakage can result in total failure with little
likelihood of advance warning in the form of
increased vibration or noise. See ANSI/AGMA
110.04, Nomenclature of Gear Tooth Failure
Modes, for additional description of this type of
failure.
79
AGMA 917--B97
The standard calculation for bending strength rating

is described in ANSI/AGMA 2001--C95 for spur and
helical gears.
15.2.1.2 Surface durability
Surface durability, also known as pitting resistance,
is the capacity to resist the kind of failure which
results from repeated surface or subsurface
stresses that are beyond the endurance limit of the
material. This type of failure is characterized by the
removal of material and the formation of surface
cavities. It may be a progressive condition which is
accompanied by increased vibration and noise. In
applications where these disturbances can be
sensed, they can serve as advanced warning of
more serious failure. See ANSI/AGMA 110.04 for
additional description of this type of failure.
The standard calculation for surface durability
rating is described in ANSI/AGMA 2001--C95 for
spur and helical gears.
15.2.2 Data for rating calculations
The data used in the rating calculations fall into the
following categories:
adequate lubrication. For an application in which

other failure modes are possible, especially those
related to inadequate lubrication, this standard
rating procedure may not completely evaluate the
design.
Because of these limitations, load rating may be
unreliable for confirming the final design. Load
rating without testing should be used on a new
design only after prior testing on similar designs has
established its reliability.
15.3 Testing
Testing is the primary method for confirming a final
design. It is used even when the design has been
adequately load rated.
15.3.1 Types of testing
The many types of testing, varying in loads and
operating conditions, can be grouped into the three
classifications described below. Any single type
may be selected, but the test program may extend
to two or all three.
15.3.1.1 Laboratory testing
The use of the load rating calculation permits

evaluation of the design while it is still on paper,
before the system or any of its components are
manufactured. If the calculations show that the
design falls short, it will also suggest what changes
to make.
This refers to testing in which the gear components,

separate from the rest of the system, are tested as
sets on special purpose test equipment. Such
variables as load, speed, gear alignment, lubrication and duration of test are controlled by the test
machine. The main advantages of such testing lie in
the relatively short time it takes to get results and the
very close control of test conditions. This control is
especially important when the tests will also be used
to compare alternative designs. The main disadvantage is the uncertainty as to how the test
machine conditions compare to the conditions to
which the gears will be subjected in the complete
system.
15.2.4 Limitations of load rating
15.3.1.2 Bench testing
The load rating method in ANSI/AGMA 2001--C95

is not suitable for every fine--pitch application. The
rating procedure is based primarily on experience
with coarse pitch gears. It may not be applicable
with other than the coarser range of fine--pitch
gears. Available data on material properties are
limited to the more traditional gear materials.
Corresponding data are lacking for other materials
widely used in fine--pitch gears. Rating methods
consider only the two types of failure described
above. Also, they assume that the design provides
This covers the testing of the gears as an assembly,

either in the form of the complete system or as a
gear drive sub--assembly. Loads may be applied to
the gears by operating the product in a loaded
condition or by applying an artificial load directly to
the output shaft. This testing has the advantage of
subjecting the gears to conditions closer to those
the design must satisfy. Such conditions include
lubrication, gear alignment, and in many cases, the
shocks and vibrations which arise in the operation of
the system. The disadvantage for some gear
--
gear geometry;
--
material properties;
--
operating conditions;
--
gear manufacturing and assembly accuracy.
15.2.3 Advantages of load rating
80
applications is the omission of other actual system

conditions, such as operation in severe
environments, poor servicing and abusive treatment.
15.3.1.3 Field testing
In field testing, the gears are tested as part of the
system being used normally. This is the ultimate in
matching test conditions to actual operation. The
disadvantages are the long time periods, and
frequently, the difficulty in adequately monitoring
the test.
15.3.2 Selection of test loads
Gear loads vary considerably with the application,
not only in magnitude but also in duty cycle and in
dynamic components. Each of these must be
considered when selecting test loads to avoid a test
which is either too lenient or too severe.
15.3.2.1 Test load magnitude
Selection of a suitable test load starts with a basic
load derived from the performance requirements of
the system. This load is increased by a factor of
safety to allow for uncertainties in system requirements or in testing. Such uncertainties may be in
the overload applied to the system, in the gear
materials and manufacturing quality, or in the
testing load accuracy.
The test load may be further increased for accelerated testing. This combination of higher load and
shorter test time is appropriate only for those failure
modes which have this kind of load--time
relationship. The two types of failures described
above, tooth breakage and surface pitting under
repeated loading, do have this relationship. Values
of test loads can be increased to decrease time, this
can be derived from the fatigue stress--cycle curves
of the gear materials. Accelerated testing may give
misleading results if other failure modes are
possible; those which depend on load level only, or
on time only.
15.3.2.2 Test duty cycle
In many applications, the gears do not operate
continuously for long periods of time at constant
load. The product may be used only intermittently
or the gears may be loaded only during a fraction of
the operating cycle. For faithful reproduction of
operating conditions, this load cycle can be copied
AGMA 917--B97
in the testing. However, to reduce the testing time

duration and sometimes to avoid the complication of
load programming, the test is run continuously. This
is satisfactory if the anticipated types of failures are
influenced only by the number of individual tooth
loadings. If the timing of the load cycle can influence
failure, as with temperature effects, a continuous
test should not be used. For example, in some
systems designed to operate intermittently, the time
between load periods will permit the lubricant and
gear material to cool and not reach their critical
temperatures. Such systems may fail a continuous
test and yet perform properly in normal use.
Instead of following a simple on--off pattern, the load
may fluctuate in magnitude during the course of a
single cycle or over the long term use of the system.
If the load varies in a single cycle but one or more
teeth always experience the peak load, then a
continuous test should be based on that peak load.
Otherwise, some kind of load averaging should be
used. For fatigue type failures, the averaging would
be derived from the stress--cycle curves for the gear
materials.
15.3.2.3 Dynamic test loads
In their normal use, the gears may experience
externally induced shock and vibration loads superimposed on their steady loads. Unless the complete
system is being tested in its normal use, it is usually
impractical to duplicate this dynamic load condition
in the tests. When they are not duplicated, these
external dynamic loads are taken into account by
the proper increase in the steady test load.
The treatment of internally induced vibration is
generally quite different. If this vibration results
from internal components only, such as the action of
gear tooth irregularities on the inertias of the gears,
then no increase in the steady load is necessary. In
such cases, the gears under test will by themselves
create duplicate dynamic loads. The situation
becomes more complicated when the dynamic
forces result from the interaction of internal irregularities and external inertias. In setting up such
tests, it is important to preserve as much as possible
the external inertias and stiffnesses of connecting
members that would be present in the use of the
system. An improper match of the testing dynamics
to the product application dynamics could easily
generate very misleading test results.
81
AGMA 917--B97
(This page is intentionally left blank.)
82
AGMA 917--B97
Bibliography
The following documents are either referenced in the text of AGMA 917--B97, Design Manual for Parallel Shaft
Fine--Pitch Gearing or indicated for additional information.
1. Colbourne, J.R., The Geometry of Involute Gears, Springer--Verlag, New York, 1987.
2. Khiralla, T.W.,On The Geometry of External Involute Spur Gears, C/I Leaming, North Hollywood, California,
1976.
3. Buckingham, E., Analytical Mechanics of Gears, Dover Publications, Mineola, New York, 1949.
4. Dudley, D.W., Handbook of Practical Gear Design, McGraw Hill, New York, 1984.
5. Townsend, D.P., Dudleys Gear Handbook, Second Edition, McGraw Hill, New York, 1991.
6. Shigley, J.E. and Mischke, C.R., Mechanical Engineering Design, McGraw Hill, New York, 1989.
7. Michalec, G.W., Precision Gearing: Theory and Practice, John Wiley and Sons, New York, 1966.
8. Smith, L.J., Assured Backlash Control -- the ABC System, AGMA Paper No. 239.14, AGMA Fall Technical
Meeting, 1979.
9. Shigley, J.E. and Mitchell, L.D., Mechanical Engineering Design, 4th edition, McGraw Hill, New York, 1983.
83
AGMA 917--B97
84
PUBLISHED BY
1500 KING STREET, ALEXANDRIA, VIRGINIA 22314

AGMA 917-B97 Design Manual For Parallel Shaft Fine-Pitch Gearing

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

AGMA 917-B97 Design Manual For Parallel Shaft Fine-Pitch Gearing

Uploaded by

Copyright:

Available Formats

AGMA 917- B97

AMERICAN GEAR MANUFACTURERS ASSOCIATION

AGMA 917- B97

Design Manual for Parallel Shaft

AGMA INFORMATION SHEET

Design Manual for Parallel Shaft Fine- Pitch Gearing

Approved September 25, 1997

American Gear Manufacturers Association

Printed in the United States of America

AMERICAN GEAR MANUFACTURERS ASSOCIATION

Definitions and symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Theory of involute gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Design synthesis and analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Design for control of backlash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

Gear drawings and specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

Gear tooth tolerances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

Materials and heat treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

Load rating and testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

Symbols, terms and units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Profile shift coefficients for 20 profile angle spur gears . . . . . . . . . . . . . . . . . 43

Maximum outside diameter for minimum topland of 0.275/Pnd . . . . . . . . . . . . 44

Principal reference planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Planes at a pitch point on a helical tooth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Power transmission by two pulleys and a crossed belt . . . . . . . . . . . . . . . . . . . 8

Point on belt generates involute . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

Involute action and speed ratio independent of center distance . . . . . . . . . . . . 9

A series of small involute cams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Time = T1: Second pair of teeth just starting engagement . . . . . . . . . . . . . . . 10

Time = T2: First pair of teeth just leaving engagement . . . . . . . . . . . . . . . . . . 10

Time = T3: One pair of teeth in contact at pitch point . . . . . . . . . . . . . . . . . . . 11

Two involute curves showing differences in lengths of corresponding arcs . 12

Involute polar angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Base pitch relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

AMERICAN GEAR MANUFACTURERS ASSOCIATION

Gear design flowchart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Preliminary design flowchart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Speed decreasing gear system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Gear train with idler gear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

External spur gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Internal spur gear and external spur pinion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

External helical gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Spur pinion and face gear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Crossed helical gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Worm and wormgear set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Spur rack and pinion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Gear tooth as a simple beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Tooth load acting at inscribed parabola . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Shaft alignment deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Standard pitch circle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Transverse backlash is arc PR -- arc PQ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

Effect of profile shift (addendum modification) . . . . . . . . . . . . . . . . . . . . . . . . . 40

Line of action for external gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

Line of action for internal gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

Showing angle at which load bears on tooth . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

Adjustable center distance gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Spring loaded center distance gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Adjustable split gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Spring loaded split gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Composite gearing with elastic element . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Spring preloaded gearing (for limited rotation) . . . . . . . . . . . . . . . . . . . . . . . . . 57

Dual path spring loaded gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

Contra--rotating input gearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

Graphical representations of typical gear errors . . . . . . . . . . . . . . . . . . . . . . . . 63