AGMA 915- 1- A02

AMERICAN GEAR MANUFACTURERS ASSOCIATION

Inspection Practices - Part 1:

Cylindrical Gears -
Tangential Measurements
AGMA 915- 1- A02

AGMA INFORMATION SHEET

(This Information Sheet is NOT an AGMA Standard)
 American Inspection Practices -- Part 1: Cylindrical Gears -- Tangential
Measurements
Gear
AGMA 915--1--A02
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 As-
sociation on the subject matter.
[Tables or other self--supporting sections may be quoted or extracted. Credit lines should
read: Extracted from AGMA 915--1--A02, Inspection Practices -- Part 1: Cylindrical Gears
-- Tangential Measurements, with the permission of the publisher, the American Gear
Manufacturers Association, 1500 King Street, Suite 201, Alexandria, Virginia 22314.]
Approved April 16, 2002
ABSTRACT
This information sheet provides a code of practice dealing with inspection relevant to tangential element and
composite deviations of cylindrical involute gears (measurements referred to single flank contact) and serves
as a supplement to ANSI/AGMA 2015--1--A01, Accuracy Classification System -- Tangential Measurements for
Cylindrical Gears.
Published by

American Gear Manufacturers Association

1500 King Street, Suite 201, Alexandria, Virginia 22314
Copyright  2002 by American Gear Manufacturers Association
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--798--3

ii
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

Contents
Page
Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
1 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
3 Symbols and corresponding terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
4 Extent of gear inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
5 Identification of deviation position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
6 Measurement of pitch deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
7 Measurement of profile deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
8 Measurement of helix deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
9 Measurement of single flank composite deviations . . . . . . . . . . . . . . . . . . . . . . 26
10 Contact pattern checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Figures
1 Notation and numbering for external gear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Notation and numbering for internal gear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3 Schematic of single probe measuring device . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4 Single pitch deviation, single probe device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
5 Pitch measurement with a pitch comparator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
6 Circular pitch measurement, two probe device . . . . . . . . . . . . . . . . . . . . . . . . . . 8
7 Single pitch deviation, two probe device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
8 Sample table with hypothetical deviation values obtained by pitch
comparator (two probe) device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
9 Sample table with hypothetical deviation values obtained by indexing
(single probe) device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
10 Sample graphic representation of single pitch deviations, fpt . . . . . . . . . . . . . . 10
11 Sample graphic representation of index deviations . . . . . . . . . . . . . . . . . . . . . . 10
12 Base pitch measurement, two probe device . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
13 Schematic of involute inspection device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
14 Profile measuring method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
15 Profile inspection by coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
16 Typical tooth profile measurement charts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
17 Tooth profile and profile diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
18 Mean profile slope deviation, fHαm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
19 Profile inspection by optical projection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
20 Profile inspection by gear tooth caliper method . . . . . . . . . . . . . . . . . . . . . . . . . 18
21 Profile inspection by measurement over pins . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
22 Helix deviation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
23 Graphic charting of helix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
24 Helix diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
25 Traces generated from four tooth flanks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
26 Helix of right hand helical gear with short lead (+ helix angle) . . . . . . . . . . . . . 23
27 Helix of right hand helical gear with long lead (-- helix angle) . . . . . . . . . . . . . 23
28 Helix of left hand helical gear with long lead (-- helix angle) . . . . . . . . . . . . . . . 24
29 Helix of left hand helical gear with short lead (+ helix angle) . . . . . . . . . . . . . . 24
30 Principle of undulation inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
31 Composite gear testing, double and single flank . . . . . . . . . . . . . . . . . . . . . . . . 26
32 Schematic of a single flank measuring device . . . . . . . . . . . . . . . . . . . . . . . . . . 27
33 Individual tooth deviations revealed by single flank testing . . . . . . . . . . . . . . . 27
34 Filtered signal from figure 33 (eccentricity removed) . . . . . . . . . . . . . . . . . . . . 28
35 Angular motion curves from tooth modification . . . . . . . . . . . . . . . . . . . . . . . . . 29

iii
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

36 Effect of contact transfer on the profile component in a tangential

composite deviation diagram (spur gears) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
37 Influence of overlap ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
38 Single flank composite strip chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
39 Single flank composite test, low number of teeth . . . . . . . . . . . . . . . . . . . . . . . . 33
40 Single flank composite test, high number of teeth . . . . . . . . . . . . . . . . . . . . . . . 33
41a Total composite deviation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
41b Long term component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
41c Short term component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
42 Manual interpretation of composite test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
43 Part of tangential composite deviation diagram -- Interpretation example . . . 36
44 Tangential composite deviation diagrams showing influence of mesh
relocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
45 Matching profiles, with tooth alignment mismatch and end relief . . . . . . . . . . . 38
46 Matching helix, with profile mismatch and end relief . . . . . . . . . . . . . . . . . . . . . 38
47 Waviness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
48 Typical specification: approximately 75% contact, excluding extremes of
tooth, which are intentionally relieved . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Tables
1 Symbols and definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

iv
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

Foreword
This Information Sheet, AGMA 915--1--A02, Inspection Practices -- Part 1: Cylindrical
Gears -- Tangential Measurements is provided for informational purposes and is intended
for use with the Standard ANSI/AGMA 2015--1--A01, Accuracy Classification System --
Tangential Measurements for Cylindrical Gears.
AGMA 915--1--A02 replaces AGMA ISO 10064--1, Cylindrical Gears -- Code of Inspection
Practice -- Part 1: Inspection of Corresponding Flanks of Gear Teeth. and the information on
similar subjects as covered in ANSI/AGMA 2000--A88, Gear Classification and Inspection
Handbook -- Tolerances and Measuring Methods for Unassembled Spur and Helical Gears.
The user of this Information Sheet is alerted that differences exist between it and
ANSI/AGMA 2000--A88 and AGMA ISO 10064--1. These include, but are not limited to:
-- Measuring methods refer to an accuracy grade numbering system that is reversed,
such that the smallest number represents the smallest tolerance;
-- Probe direction and measurement requirements for elemental and composite
tolerances may differ from ANSI/AGMA 2000--A88 or AGMA ISO 10064--1;
-- The measurement “profile evaluation range” and “helix evaluation range”, where
the tolerances are applied, are defined for different area than in ANSI/AGMA 2000--A88
or AGMA ISO 10064--1;
-- The measurement of undulations is included;
-- Concepts of “mean measurement trace”, “design trace”, “slope deviation”, “form
deviation”, “gear form filter cutoff”, “tolerance diameter” and “data density” are defined.
Therefore, the user of this information sheet must be very careful when comparing
measurement methods formerly specified using ANSI/AGMA 2000--A88 or AGMA ISO
10064--1.
The first draft of AGMA 915--1--A02 was made in May, 1998. This document was approved
by the Inspection Handbook Committee on January 31, 2002. It was approved by the
Technical Division Executive Committee as an AGMA Information Sheet on April 16, 2002.
Suggestions for improvement of this document will be welcome. They should be sent to the
American Gear Manufacturers Association, 1500 King Street, Suite 201, Alexandria,
Virginia 22314.

v
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

PERSONNEL of the AGMA Inspection and Handbook Committee

Chairman: Edward Lawson . . . . . . . . . . . . . . . . . . . . . . M&M Precision Systems

ACTIVE MEMBERS

W.A. Bradley . . . . Consultant S. Lindley . . . . . . The Falk Corporation

D.R. Choiniere . . Profile Engineering, Inc. M. May . . . . . . . . . The Gleason Works
J. Clatworthy . . . . Gear Metrology, Inc. D.A. McCarroll . . ZF Industries
B.L. Cox . . . . . . . BWXT Y12 LLC D.R. McVittie . . . . Gear Engineers, Inc.
T.C. Glasener . . . Xtek, Incorporated S. Moore . . . . . . . Martin Sprocket & Gear, Inc.
G.G. Grana . . . . . The Gleason Works R.W. Ott . . . . . . . . Caterpillar, Inc.
B. Hofrichter . . . . Arrow Gear Company J.M. Rinaldo . . . . Atlas Copco Comptec, Inc.
T. Klaves . . . . . . . Milwaukee Gear L.J. Smith . . . . . . Consultant
I. Laskin . . . . . . . . Consultant R.E. Smith . . . . . . R.E. Smith & Company, Inc.

ASSOCIATE MEMBERS

M. Antosiewicz . . The Falk Corporation W.E. Lake . . . . . . Mitsubishi Gear Technology Ctr.
M.J. Barron . . . . . Gear Motions, Inc. A.J. Lemanski . . . Penn State University
D. Behling . . . . . . Hamilton Sundstrand Aero. G.A. Luetkemeier Rockwell Automation/Dodge
M.K. Considine . . Considine Associates D. Matzo . . . . . . . Northwest Gears, Inc.
R. Considine . . . . Considine Associates P.A. McNamara . Caterpillar, Inc.
J.S. Cowan . . . . . Eaton Corporation W.J. Michaels . . . Sundstrand Corporation
M.E. Cowan . . . . Process Equipment Company M. Milam . . . . . . . Amarillo Gear Company
B. Cowley . . . . . . Mahr Corporation T. Miller . . . . . . . . The Cincinnati Gear Company
C. Dick . . . . . . . . . The Horsburgh & Scott Co.
M. Nanlawala . . . IIT Research Institute/INFAC
H.D. Dodd . . . . . . Caterpillar, Inc.
M. Octrue . . . . . . Centre Technique Des Ind. Mec.
R. Green . . . . . . . R7 Group, Gear Consultants
T. Okamoto . . . . . Nippon Gear Company, Ltd.
D. Gregory . . . . . Gear Products, Inc.
J.A. Pennell . . . . . Univ. of Newcastle--Upon--Tyne
B. Gudates . . . . . Fairfield Manufacturing Co., Inc.
J.S. Hamilton . . . Regal--Beloit Corporation K.R. Price . . . . . . Eastman Kodak Company
H. Harary . . . . . . . NIST R.S. Ramberg . . . The Gear Works -- Seattle, Inc.
D. Heinrich . . . . . Xtek, Incorporated V.Z. Rychlinski . . Brad Foote Gear Works, Inc.
G. Henriot . . . . . . Consultant D.H. Senkfor . . . . Precision Gear Company
J. Horwell . . . . . . Brown & Sharpe S. Shariff . . . . . . . PMI Food Equipment Group
S. Johnson . . . . . The Gear Works -- Seattle, Inc. E. Storm . . . . . . . Consultant
T. Klemm . . . . . . . Liebherr R.F. Wasilewski . Arrow Gear Company
D.E. Kosal . . . . . . National Broach & Machine Co. F.M. Young . . . . . Forest City Gear Company
J. Koshiol . . . . . . Columbia Gear Corporation P. Zwart . . . . . . . . Caterpillar, Inc.

vi
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

American Gear Manufacturers At the time of publication, the editions indicated were
valid. All standards are subject to revision, and
Association -- parties to agreements based on this document are
encouraged to investigate the possibility of applying

Inspection Practices -- the most recent editions of the standards indicated.

Part 1: Cylindrical
Gears -- Tangential
Measurements
1 Scope
This information sheet constitutes a code of practice
dealing with tangential measurements on flanks of
individual cylindrical involute gears., i.e., with the
measurement of pitch, profile, helix and tangential
composite characteristics.
In providing advice on gear measuring methods and
the analysis of measurement results, it supplements
the standard ANSI/AGMA 2015--1--A01, Accuracy
Classification System -- Tangential Measurements
for Cylindrical Gears.
2 References
The following standards contain provisions which
are referenced in the text of this information sheet.
AGMA 915--3--A99, Inspection Practices -- Gear
Blanks, Shaft Center Distance and Parallelism
ANSI/AGMA 2015--1--A01, Accuracy Classification
System -- Tangential Measurements for Cylindrical
Gears
ISO 53:1998, Cylindrical gears for general and
heavy engineering -- Standard basic rack tooth
profile
ISO 54:1996, Cylindrical gears for general
engineering and for heavy engineering -- Modules
ISO 701:1998, International gear notation --
Symbols for geometrical data
ISO 1122--1:1998, Vocabulary of gear terms -- Part
1: Definitions related to geometry
3 Symbols and corresponding terms
The symbols and terms used throughout this manual
are in basic agreement with the symbols and terms
given in ISO 701:1998, International gear notation --
Symbols for geometrical data. In all cases, the first
time that each symbol is introduced, it is defined and
discussed in detail. See table 1.
NOTE: The symbols and definitions used in this infor-
mation sheet may differ from other AGMA standards.
The user should not assume that familiar symbols can
be used without a careful study of their definitions.

Table 1 -- Symbols and definitions

Where
Symbols Definition1) Units first used
b Facewidth mm Figure 24
D Design pitch diameter mm Eq 4
Db Design base diameter mm Eq 3
d Reference diameter mm Eq 24
db eff Effective base diameter mm 6.5.3
dT Tolerance diameter mm 6.2
Fβ Total helix deviation mm Figure 22
(continued)

1
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

Table 1 (continued)
Where
Symbols Definition1) Units first used
Fis Total single flank composite deviation mm 9.1
Fp Total cumulative pitch deviation mm 6.1
Fps/8 Sector pitch deviation2) mm 6.2
Fr Radial runout mm 9.3.6
Fα Total profile deviation mm Figure 17
fdbm Mean base diameter difference2) mm 6.5.3
fe Eccentricity between gear axis and axis of gear teeth mm Figure 18
ffα Profile form deviation mm Figure 17
ffβ Helix form deviation mm Figure 24
fHα Profile slope deviation2) mm Figure 17
fHαm Mean profile slope deviation2) mm 7.6
fHβ Helix slope deviation2) mm Figure 24
fHβm Mean helix slope deviation2) mm 8.6
fHβmt Mean helix slope deviation, in the transverse plane and tangent to the mm Eq 18
tolerance diameter2)
fid Tooth--to--tooth double flank composite deviation mm 9.3.6
fis Tooth--to--tooth single flank composite deviation mm 9.1
fLm Mean lead difference2) mm 8.7
fpbm Mean normal base pitch deviation2) mm 6.5.3
fpbn Normal base pitch deviation2) mm 6.5
fpt Single pitch deviation2) mm 6.1
fwβ Undulation height (along helix) mm Figure 24
f1, f2 Reading head frequency pulses/sec Figure 32
fα Pressure angle deviation2) degrees 7.5
fαmn Mean normal pressure angle deviation2) degrees 6.5.3
fαmt Mean transverse pressure angle deviation2) degrees 6.5.3
fβ Helix angle deviation2) degrees 8.5
fβm Mean helix angle deviation2) degrees 8.7
gα Length of path of contact mm Figure 36
k Number of pitches in a sector -- -- 5.6
L Left flank -- -- 5.2
L Lead of the design helix mm Eq 17
Leff Effective lead mm 8.7
Lα Profile evaluation range mm Figure 17
Lαc Functional profile length mm Eq 9
Lβ Helix evaluation range mm Figure 24
Lε Base tangent length to start of active profile mm Figure 17
l Left hand helix -- -- 5.3
mn Normal module mm Eq 1
N Pitch number -- -- 5.5
n Number of deviation values included in the mean -- -- Eq 8
pb Base pitch mm Figure 36
pbn Theoretical normal base pitch mm 6.5
pm True position pitch2) mm 6.3.2
(continued)

2
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

Table 1 (concluded)
Where
Symbols Definition1) Units first used
R Right flank -- -- 5.2
r Right hand helix -- -- 5.3
s Undulation measurement bar length mm Figure 30
z Number of teeth -- -- Eq 2
zM Number of teeth in master indexing worm wheel -- -- Eq 24
z1 Driving gear -- -- Figure 32
z2 Driven gear -- -- Figure 32
αTt Transverse pressure angle at the tolerance diameter degrees 6.5.2
αn Normal pressure angle degrees Eq 1
αn eff Effective normal pressure angle degrees 6.5.3
αt Design transverse pressure angle degrees Eq 6
αt eff Effective transverse pressure angle degrees 6.5.3
β Helix angle degrees Eq 5
βb Design base helix angle degrees Eq 2
βeff Effective helix angle at the standard pitch diameter degrees 8.7
βT eff Effective helix angle at the tolerance diameter degrees 8.7
εγ Total contact ratio -- -- 9.3.5
λβ Undulation wave length mm Eq 24
λβx Axial wavelength of undulation mm Figure 24
ξ Involute roll angle degrees Figure 17
I Reference face -- -- 5.2
II Non--reference face -- -- 5.2
NOTE:
1) Symbols used for deviations of individual element measurements from specified values are composed of lower case
letters “f” with subscripts (exceptions include fe, f1 and f2) whereas symbols used for “cumulative” or “total” deviations,
which represent combinations of several individual element deviations,are composed of capital letters “F” also with sub-
scripts. It is necessary to qualify some deviations with an algebraic sign. A deviation is positive when e.g., a dimension
is larger than optimum and negative when smaller than optimum.
2) These deviations can be + (plus) or -- (minus).

4 Extent of gear inspection
It is rarely necessary or economical to measure all
possible deviations on all gears manufactured.
Certain elements may not significantly influence the
function of the gear under consideration. Some
measurements can be substituted for others. Stable
manufacturing processes allow a relatively small
number of samples to be measured and still ensure
that the required quality level is maintained. It is
recommended that specific measuring plans be
negotiated between purchaser and supplier.
4.1 Required inspection information
Certain necessary information should be provided to
the operator(s) of the measuring equipment. The
information required will vary depending on the type
of measurement(s) required. Most measurement
processes require basic gear and blank data,
number of teeth, pitch, pressure angle, helix angle,
tooth size, outside diameter, root diameter, face
width, design profile, design helix, etc. Certain
measuring tasks require additional information. For
example, to measure profile, the profile control
diameter and start of tip break must be provided.
With mechanical measuring equipment, additional
information may be required: base circle diameter
(radius), base helix angle, sine bar setting, etc.
The design engineer or engineering department
should be responsible for supplying this minimum
required inspection information to those performing
the measurements.
4.2 Measurement selection
Inspection may be carried out using a number of
alternate methods. Some measurements may be

3
AGMA 915--1--A02
substituted for others. For example single flank
composite measurement may be substituted for
pitch measurement, or radial composite measure-
ment may replace runout measurement.
A number of factors should be considered when
selecting the measurements, including the quality
level required, size of the gear, manufacturing cost
and most important the application of the product
gear.
4.2.1 Sampling
Gears, like other parts, are manufactured to a certain
level of accuracy dependant on the production
process used. When the process used is proven
capable of producing the required accuracy level
using statistical methods, sampling inspection may
be utilized. Many factors may influence the sample
size and frequency, foremost among these should
be the assurance that the required accuracy level of
the parts is met.
4.2.2 First piece inspection
It may be possible to inspect only the first piece of a
batch to verify that the setup is correct, allowing the
inherent accuracy of the process to assure the
quality of subsequent parts.
5 Identification of deviation position
It is convenient to identify deviations associated with
measurements of gear teeth by specific reference to
individual right flanks, left flanks, pitches or groups of
these.
In the following, conventions are described which
enable positive determination of the location of
deviations.
5.1 Datum axis
Specification of the design profile, design helix, and
design pitch requires definition of an appropriate
reference axis of rotation, called the datum axis. It is
defined by specification of datum surfaces. See
AGMA 915--3--A99.
The datum axis determines tooth geometry, thereby
being the reference for measurements and associat-
ed tolerances. The location and orientation of the
tolerance diameter circle are determined by the
datum axis.
4
AMERICAN GEAR MANUFACTURERS ASSOCIATION
Ideally the surfaces used to construct the datum
axis, the surfaces used to locate the gear for
manufacturing, and the functional surfaces that
define the gear axis of rotation in its final assembly
would all be the same. In practice this is often not the
case. For example, shaft type parts are often
manufactured and inspected using female centers to
define the datum axis. In cases where the inspec-
tion, manufacturing, and/or functional datum sur-
faces are different, these surfaces must be
coincident with each other to a level of accuracy
sufficient to assure the final quality of the gear is
adequately represented during measurement.
The gear being measured should be oriented so that
its datum axis is coincident with the axis of rotation of
the measuring instrument. In the case of mounting
the gear between centers, care must be taken to
assure that the mounting arbor, if used, is in good
condition, and the female centers are clean and
concentric with the datum surfaces of the gear. In the
case of computer controlled measuring instruments,
it may be possible to mount the gear with significant
deviation to the instrument’s axis of rotation. In that
case, the measuring program must be capable of
mathematically correcting the errors resulting from
this off axis mounting condition.
5.2 Right or left flank
It is convenient to choose one face of the gear as the
reference face and to mark it with the letter “I”. The
other non--reference face might be termed face “II”.
For an observer looking at the reference face, so that
the tooth is seen with its tip uppermost, the right flank
is on the right and the left flank is on the left.
Right and left flanks are denoted by the letters “R”
and “L” respectively.
5.3 Right hand or left hand helical gears
The helix of an external or internal helical gear is
referred to as being right hand or left hand. The hand
of helix is denoted by the letters “r” and “l”
respectively.
The helix is right hand (left hand) if, when looking
from one face, the transverse profiles show succes-
sive clockwise (counter--clockwise) displacement
with increasing distance from an observer.
5.4 Numbering of teeth and flanks
Looking at the reference face of a gear, the teeth are
numbered sequentially in the clockwise direction.
The tooth number is followed by the letter R or L,
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

indicating whether it is a right or a left flank. Example: 6 Measurement of pitch deviations

“Flank 29 L”.

5.5 Numbering of pitches 6.1 Pitch deviation

The numbering of individual pitches is related to
tooth numbering as follows: pitch number “N” lies
between the corresponding flanks of teeth numbers
“N--1” and “N”; with a letter R or L it is indicated
whether the pitch lies between right or left flanks. For
example “Pitch 2 L”, (see figures 1 and 2).
5.6 Number of pitches “k”
The subscript “k” of a deviation symbol denotes the
number of consecutive pitches to which the deviation
applies.
In practice, a number is substituted for “k”, for
example Fp3 indicates that a given cumulative pitch
deviation refers to three pitches.
Index, single pitch (fpt), and total cumulative pitch
(Fp) are elemental parameters relating to the accura-
cy of tooth locations around a gear. The following is a
description of the measuring methods and a guide to
the interpretation of data generated by the measur-
ing devices.
6.2 Pitch deviation measurement
Measurements for determining index, single pitch
(fpt), and total cumulative pitch (Fp) are made:
-- relative to the datum axis of the gear;
-- at the tolerance diameter, dT ;
-- In the specified tolerancing direction (within
the transverse plane along the arc of the tolerance
diameter).

30R 2L

tip

left right
flank flank
30 1
29 2

30 R = pitch No. 30, right flank

2 L = pitch No. 2, left flank

Figure 1 -- Notation and numbering for external gear

30R
1L
tip

left flank right

flank

2
29
1 30

1 L = pitch No. 1, left flank

30 R = pitch No. 30, right flank

Figure 2 -- Notation and numbering for internal gear

5
AGMA 915--1--A02
Measurements made at different diameters or in
other directions must be adjusted so that they are
equivalent to measurements at the tolerance diame-
ter and in the tolerance direction. This adjustment
must be made before comparison of test results to
tolerances.
Sector pitch deviation (Fps/8) is an optional parame-
ter described in Annex E of ANSI/AGMA
2015--1--A01. Measurements of sector pitch devi-
ation are also expected to conform to the above
specified requirements.
Pitch should be measured on both left and right
flanks. However, if the specific operating direction of
the gear is known, only the loaded flanks need to be
measured.
6.3 Pitch deviation measurement methods
Pitch parameters can be measured by either of two
types of device. The indexing (single probe) device
determines the location of each tooth around a gear,
relative to a datum tooth (the index). The pitch
comparator (two probe) device compares the dis-
tances between adjacent tooth flanks to the distance
5
4 Tolerance
-- Index 2
deviation 1
Dash lines represent + Index
theoretical location deviation
Figure 3 -- Schematic of single probe measuring device
6
AMERICAN GEAR MANUFACTURERS ASSOCIATION
between an initial reference pair of adjacent tooth
flanks.
The various pitch parameters can all be determined
by either measuring device with the application of
suitable calculations. The indexing method is
usually preferred because of its accuracy and
simplicity. However, for large diameter gears, use of
the pitch comparator method may be preferable.
Coordinate measuring machines without a rotating
table can also be used for measurements of pitch
parameters by probe movements that correspond to
the principle of the indexing method.
6.3.1 Indexing pitch measurement method
The indexing (single probe) device uses an angular
indexing apparatus such as an index plate, circle
divider, optical or electronic encoder, or polygon and
auto collimator to precisely rotate the gear by an
angular increment equal to its pitch, or 360°/z (see
figure 3). The degree of its precision must be
consistent with the quality grade and diameter of the
gear.
Index mechanism
diameter, dT
Index readings
AMERICAN GEAR MANUFACTURERS ASSOCIATION
The single probe should be oriented to contact the
tooth flanks at the tolerance diameter, dT, and to
gather measurements in the specified measurement
direction. The single probe is adjusted to indicate
zero while the device is contacting the randomly
selected initial test tooth flank. As the gear is
incrementally rotated around its datum axis, the
single probe moves in and out on a precision slide
and stop, measuring each successive tooth flank
position, relative to the indexing mechanism. This
process is repeated until every tooth has been
measured.
It is common practice to complete this series of
measurements by taking a final measurement on the
initial reference tooth, thereby closing the circle.
Ideally, this would produce a second measurement
value of zero for the first tooth, as was set at the
beginning of the process. Excessive deviation of this
second measurement value from zero indicates a
problem with the measurement.
6.3.1.1 Calculation of index
If the indicator always reads plus material as a plus
reading and the gear is indexed counterclockwise
(teeth are numbered clockwise), then the right flank
measurement values provided by the indexing
(single probe) pitch measurement device can be
used directly as the plus and minus values of index
for each tooth of the gear (see figure 3). Left flank
single probe measurement values must be multi-
plied by –1 to produce plus and minus index values.
Other pitch parameters may then be calculated from
that data.
If a graphical recorder is used, data gathered by the
single probe measurement device will appear in the
form shown in figure 4. This figure shows the
measurement value of the initial measured tooth set
to zero, thereby establishing it as the reference. The
measured values shown for all other teeth then
represent the positional deviations of those teeth
from the initial reference tooth.
6.3.1.2 Calculation of single pitch, fpt
Subtraction of each successive pair of index values
produces the plus and minus values of single pitch
deviation for each adjacent pair of tooth flanks of the
gear. See Clause 5 for specified tooth numbering,
pitch numbering, and flank naming conventions.
The number 1 single pitch deviation value is equal to
the index value of the last tooth subtracted from the
index value of the first tooth. The number 2 single
AGMA 915--1--A02
pitch deviation value is equal to the index value of the
first tooth subtracted from the index value of the
second tooth. Since the index value of the first tooth
is set to zero, the number 2 single pitch deviation
value is equal to the index value of the second tooth.
The number 3 single pitch deviation value is equal to
the index value of the second tooth subtracted from
the index value of the third tooth, and so on.
+
Index deviation
--fpt
0 +fpt
--
1 2 3 4 5 6 7 8 9 10
Tooth number
Figure 4 -- Single pitch deviation, single probe
device
If a graphical recorder is used, data gathered by the
single probe measurement device will appear in the
form shown in figure 4. Single pitch deviation values,
fpt, are shown as the differences between adjacent
index values.
6.3.1.3 Calculation of total cumulative pitch
deviation, Fp
The total cumulative pitch deviation, Fp, is equal to
the difference between the most positive and the
most negative index value for the complete gear.
6.3.1.4 Calculation of sector pitch deviation,
Fps/8
Calculation of the sector pitch deviation, Fps/8, is
presented in Annex E of ANSI/AGMA 2015--1--A01.
6.3.2 Comparator pitch measurement method
The pitch comparator (two probe) device may be
mechanized or hand--held. Measurements made by
the mechanized version are preferred. In either
case, both probes should be oriented to contact
adjacent tooth flanks at the tolerance diameter.
One probe serves to establish a reference position
upon a tooth flank. The second probe is fitted with
either a mechanical or an electronic indicator to
measure variations of its position from the first probe.
The device is adjusted to indicate zero while the
probes are contacting the randomly selected initial
pair of teeth (see figure 5).
7
AGMA 915--1--A02
Figure 5 -- Pitch measurement with a pitch
comparator
The mechanized pitch comparator is a device with a
rotational axis that positions the gear for measure-
ment. The gear must be mounted with its datum axis
coincident with the pitch comparator’s rotational
axis.
The two probes should be oriented to contact the
adjacent tooth flanks within the same transverse
plane, at the tolerance diameter, dT. As the gear is
rotated around its datum axis, the pitch comparator
moves in and out on a precision slide and stop,
measuring each successive adjacent tooth pair.
This process is repeated until every adjacent pair of
teeth has been measured.
The hand--held pitch comparator is a portable device
that lacks a means of referencing the datum axis of
the gear. It is therefore fitted with a positioning stop
that contacts the outside diameter of the gear, which
thereby becomes the reference for pitch measure-
ments. This method requires that special consider-
ation be given to the concentricity of the outside
diameter of the gear with its datum axis.
The two probes must be oriented to contact the
adjacent tooth flanks within a normal plane. The
hand--held pitch comparator is applied successively
to each pair of teeth with each indicator measure-
ment observed and recorded. This process is
repeated until every adjacent pair of teeth has been
measured (see figure 6).
8
AMERICAN GEAR MANUFACTURERS ASSOCIATION
spring
loaded
Tolerance
diameter,
dT
Figure 6 -- Circular pitch measurement, two
probe device
Since the hand--held pitch comparator measures in
the normal plane, the measurements must be
converted to transverse pitch deviations before
being summed to determine index as described in
6.3.2.3.
It is important to understand that the readings
collected from two probe pitch comparators are
relative to a randomly selected tooth pair of unknown
position. They must not be compared to the single
pitch tolerances, until they are adjusted by true
position pitch, pm.
6.3.2.1 Calculation of true position pitch, pm
The true position pitch, pm, is the measurement
value for any perfectly spaced tooth pair, with the
given setup of the pitch comparator. It is equal to the
average value found by summing all the adjacent
tooth pair measurements then dividing the result by
the number of tooth pairs (i.e., the number of teeth).
If a graphical recorder is used, data gathered by the
pitch comparator method will appear in the form
shown in figure 7. This figure shows the measure-
ment value of the initial pair of teeth (1--2) set to zero.
Also shown is the true position pitch, pm, as the
calculated mean of pitch comparator measurement
values.
6.3.2.2 Calculation of single pitch deviation, fpt
Subtraction of the true position pitch, pm, from each
adjacent tooth pair measurement produces the plus
and minus values of single pitch deviation, fpt, for
each tooth pair of the gear. See Clause 5 for
specified tooth numbering, pitch numbering, and
flank naming conventions.
If a graphical recorder is used, data gathered by the
pitch comparator method will appear in the form
shown in figure 7. Single pitch deviation values, fpt,
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

are shown as the deviations of individual pitch 6.3.2.4 Calculation of total cumulative pitch
comparator measurement values to the true position deviation, Fp
pitch, pm.
The total cumulative pitch deviation, Fp, is equal to
the difference between the most positive index value
and the most negative index value for the complete
gear.

6.3.2.5 Calculation of sector pitch deviation,

Pitch comparator readings

+ Fps/8
+fpt
Calculation of the sector pitch deviation, Fps/8, is
0 presented in Annex E of ANSI/AGMA 2015--1--A01.
pm pm
6.4 Relationships of pitch parameters and
-- --fpt measuring methods

The relationships of pitch parameters using different

1--2 2--3 3--4 4--5 5--6 6--7 7--8 8--9 9--10 10--11
measuring methods is illustrated within figures 8
Pairs of adjacent teeth through 11.

Figure 7 -- Single pitch deviation, two probe 6.5 Base pitch measurement
device
The normal base pitch measurement device is a two
probe instrument of similar construction to the
hand--held pitch comparator. However, its measur-
6.3.2.3 Calculation of index ing principles are substantially different from those
described under 6.3.2:
The plus and minus index values for each tooth of the
gear can be produced by successive summation of -- Rather than measuring the relative normal
the single pitch deviation values. See clause 5 for pitch at a given measurement (tolerance) diame-
specified tooth numbering, pitch numbering, and ter, it measures the normal base pitch, pbn, which
flank naming conventions. is the shortest distance between adjacent tooth
flanks (see figure 12).
In all cases, the number one (first) tooth shall be the
datum tooth and its index value set to zero -- This method cannot directly or indirectly
accordingly. reference the datum axis of the gear. The tooth
flank features themselves become the reference.
The index value of the second tooth is equal to the Therefore, observations of index and total cumu-
index value of the first tooth plus the number 2 single lative pitch, Fp, can not be properly made with this
pitch deviation value. Since the index value of the device.
first tooth is set to zero, the index value of the second
-- If the instrument is adjusted to the specified
tooth is equal to number 2 single pitch deviation normal base pitch of a gear prior to commencing
value. The index value of the third tooth is equal to measurements, it can provide an observation of
the index value of the second tooth plus the number normal base pitch deviation, fpbn.
3 single pitch deviation value, and so on.
The normal base pitch parameter provides a local-
At the end of this process, the index value of the first ized composite observation of gear tooth flank
tooth will be found by adding the number 1 single accuracy. It is localized, in that the observation is
pitch deviation value to the index value of the last made only at a single point on the tooth flank. It is
tooth. Ideally, this would produce a second index composite in that it combines the effects of involute
value of zero for the first tooth. Excessive deviation profile, helix, and pitch into a single observation that
from zero, of this calculated index value, for the first directly relates to the gear’s ability to achieve
tooth indicates a problem with the measurement. smooth, conjugate meshing action with its mate.

9
 AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

Tooth numbers of pitches 18:1 1:2 2:3 3:4 4:5 5:6 6:7 7:8 8:9 9:10 10:11 11:12 12:13 13:14 14:15 15:16 16:17 17:18
Pitch number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
2--probe pitch
0 1 --1 1 --1 --3 --5 --4 --4 --5 --6 --4 --3 --3 --1 1 1 0
comparator readings
True position pitch pm
--2
(mean of readings)
Single pitch deviations fpt
2 3 1 3 1 --1 --3 --2 --2 --3 --4 --2 --1 --1 1 3 3 2
(readings -- pm)
Tooth numbers for Index
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
values
Index deviations (calcu-
0 3 4 7 8 7 4 2 0 --3 --7 --9 --10 --11 --10 --7 --4 --2
lated)
Figure 8 -- Sample table with hypothetical deviation values obtained by pitch comparator
(two probe) device
(In practice, integer values are seldom encountered. Maximum value of fpt and minimum and maximum
values for index deviations are shaded.)

1--probe readings, 0 3 4 7 8 7 4 2 0 --3 --7 --9 --10 --11 --10 --7 --4 --2
right flanks
Index deviations 0 3 4 7 8 7 4 2 0 --3 --7 --9 --10 --11 --10 --7 --4 --2
Single pitch deviations 2 3 1 3 1 --1 --3 --2 --2 --3 --4 --2 --1 --1 1 3 3 2
fpt (calculated)
Figure 9 -- Sample table with hypothetical deviation values obtained by indexing
(single probe) device
(In practice, integer values are seldom encountered. Maximum value of fpt and minimum and maximum
values for index deviations are shaded.)

Single pitch deviations, fpt

12
10
8
6
4
2
0.001 mm

0
--2
--4
--6
--8
--10
--12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Pitch number
Figure 10 -- Sample graphic representation of single pitch deviations, fpt

Index deviations
12
10
8
6
4
2
0.001 mm

0
--2
--4
--6
--8
--10
--12
1 2 3 4 5 8 6 7
9 10 11 12 13 14 15 16 17 18
Flank number
Figure 11 -- Sample graphic representation of index deviations

10
AMERICAN GEAR MANUFACTURERS ASSOCIATION
pbn
Base circle
Figure 12 -- Base pitch measurement, two
probe device
The theoretical normal base pitch can be calculated
as follows:
p bn = m n π cos α n (1)
where
pbn is the theoretical normal base pitch, mm;
mn is the normal module, mm;
αn is the normal pressure angle, degrees.
6.5.1 Normal base pitch measurement device
The normal base pitch measurement device is
usually a hand--held device, which can either be set
to measure directly the deviations from the theoreti-
cal normal base pitch, with the aid of a suitable gage,
or set to reference a randomly selected initial pair of
adjacent teeth.
The two measurement probes of the device are
oriented to contact adjacent tooth flanks within a
base tangent plane. In practice, this involves rocking
the device through the possible range of contact of
the measuring probe with the tooth flank while
observing the measurement indicator. The ob-
served minimum deviation of the indicator will occur
at the point of contact corresponding with a base
tangent plane. It is important to ensure that the
points of contact of the probes do not lie in zones with
profile or helix modifications, especially when mea-
suring deviations from the theoretical normal base
pitch.
The normal base pitch measurement device is
applied successively to each pair of teeth with each
indicator measurement recorded. This process is
repeated until every adjacent pair of teeth has been
measured.
AGMA 915--1--A02
6.5.2 Calculation of single pitch deviation, fpt,
from normal base pitch measurements
Normal base pitch measurements are inherently
composite observations, combining the influences
of pitch, profile, and helix deviations. It is not
possible to decompose normal base pitch deviations
into observations of those individual constituent
deviations such as single pitch. However, since
normal base pitch is a better indicator of gear quality
than single pitch, this document permits comparison
of normal base pitch deviations to single pitch
tolerances.
Before commencing to calculate single pitch devi-
ations, the direction in which normal base pitch devi-
ation values are reported must be converted from
normal to the tooth surface to along the arc of the
tolerance diameter, dT, circle within the transverse
plane, as required by ANSI/AGMA 2015--1--A01.
The first step is to convert the normal base pitch
values to the transverse plane, which requires divid-
ing each by the cosine of the base helix angle, cos βb.
Then, dividing the results by the cosine of the
transverse pressure angle at the tolerance diameter,
cos αTt, converts the values to a direction along the
arc of the tolerance diameter circle.
As is the case with any pitch comparator (two probe)
measurements, these values must be compared
with the true position pitch, pm, to derive single pitch
values. This method can be applied to measure-
ments made by devices set relative to a randomly
selected tooth pair or relative to the theoretical
normal base pitch.
The true position pitch, pm, is equal to the average
value found by summing all the adjacent tooth pair
measurements, then dividing the result by the
number of tooth pairs (i.e., the number of teeth).
Subtraction of the true position pitch, pm, from each
adjacent tooth pair measurement produces the plus
and minus values of single pitch deviation, fpt, for
each tooth pair of the gear.
6.5.3 Additional calculations for normal base
pitch measurements
When the normal base pitch measurement device is
initially set to the theoretical normal base pitch,
resulting measurements can be used to calculate a
variety of parameters that are useful for controlling
the quality of gear involute profiles.
It is important to understand that these calculations
are based upon the assumption that the helical lead
of the gear, which also affects normal base pitch
11
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

measurements, is correct. Included in these calcu- where

lated parameters are:
db eff is the effective base diameter, mm;
-- normal base pitch deviation, fpbn;
Db is the design base diameter, mm.
-- mean normal base pitch deviation, fpbm;
-- mean base diameter difference, fdbm; 6.5.3.5 Calculation of effective transverse
-- effective base diameter, db eff; pressure angle, αt eff
-- effective transverse pressure angle, αt eff; Effective transverse pressure angle, αt eff, can be
-- effective normal pressure angle, αn eff; calculated as follows:
-- mean transverse pressure angle deviation,
fαmt; α t eff = acos  
d b eff
D
(4)
-- mean normal pressure angle deviation, fαmn.
where
6.5.3.1 Calculation of normal base pitch
αt eff is the effective transverse pressure angle,
deviation, fpbn degrees;
Determination of normal base pitch deviation, fpbn, D is the design pitch diameter, mm.
requires setting of the normal base pitch measure-
ment device to the theoretical normal base pitch, 6.5.3.6 Calculation of effective normal
with the aid of a suitable gage, before measurements pressure angle, αn eff
are taken. Resulting measurement values can then Effective normal pressure angle, αn eff, can be
be used directly as the plus and minus values of calculated as follows:
normal base pitch deviation, fpbn, for each adjacent
tooth pair of the gear. α n eff = atantan α t eff cos β (5)

6.5.3.2 Calculation of mean normal base pitch where

deviation, fpbm αn eff is the effective normal pressure angle,
degrees;
The mean normal base pitch deviation, fpbm, is equal
to the average value found by summing all the β is the helix angle, degrees.
adjacent tooth pair deviations of normal base pitch, 6.5.3.7 Calculation of mean transverse
fpbn, then dividing the result by the number of tooth pressure angle deviation, fαmt
pairs (i.e., the number of teeth).
Mean transverse pressure angle deviation, fαmt, can
6.5.3.3 Calculation of mean base diameter
be calculated as follows:
difference, fdbm
f αmt = α t eff − α t (6)
Mean base diameter difference, fdbm, can be
calculated as follows: where
f pbm z fαmt is the mean transverse pressure angle
f dbm = (2) deviation, degrees;
π cos β b
αt is the design transverse pressure angle, de-
where
grees.
fdbm is the mean base diameter difference, mm;
6.5.3.8 Calculation of mean normal pressure
fpbm is the mean normal base pitch deviation,
angle deviation, fαmn
mm;
z is the number of teeth; Mean normal pressure angle deviation, fαmn, can be
calculated as follows:
βb is the design base helix angle, degrees.
f αmn = α n eff − α n (7)
6.5.3.4 Calculation of effective base diameter,
db eff where
fαmn is the mean normal pressure angle
Effective base diameter, db eff, can be calculated as deviation, degrees;
follows:
αn is the design normal pressure angle,
d b eff = D b + f dbm × 10 −3 (3) degrees.

12
AMERICAN GEAR MANUFACTURERS ASSOCIATION
7 Measurement of profile deviations
7.1 Profile
Profile is the shape of the tooth flank from its root to
its tip. The functional profile is the operating portion,
which is in actual contact during tooth mesh, and
cannot extend below the base cylinder.
Profile deviation is the difference between the
specified and the measured profile of the gear.
Unless modifications are specified, the shape of the
profile in the transverse plane is an involute curve.
ANSI/AGMA 2015--1--A01 specifies the direction of
tolerancing for profile deviation to be within the
transverse plane, tangent to the base circle.
7.2 Profile inspection methods
The standard methods of profile measurement are
with generative, coordinate, or portable involute
measurement instruments.
7.2.1 Generative involute measurement
instruments
Generative involute measuring instruments mea-
sure the deviation of the actual profile from a nominal
involute profile, which is generated by the
instrument. Generating the nominal involute re-
quires a tangential movement of a measurement
probe, within the plane tangent to the base cylinder
of the given gear, together with a rotational move-
ment of the gear mounted on the instrument spindle.
These movements must be synchronized such that
the linear movement of the probe is equal to the
distance along the circumference of the base circle
diameter associated with the rotational movement
(see figure 13).
Spindle
Base
circle
Probe
Figure 13 -- Schematic of involute inspection
device
Generative involute measurement instruments may
employ a master base circle or master involute cam
to generate the nominal involute curve. Such
instruments may include a ratio mechanism, which
relates the actual workpiece base circle to the
AGMA 915--1--A02
master base circle. Generative involute measuring
instruments may use a computer numerical control
electronic drive system to generate the nominal
involute curve.
Profile measurements must be made relative to the
datum axis of rotation of the gear. Refer to 5.1 for
more information concerning the datum axis of
rotation.
The probe tip must be accurately positioned within
the plane tangent to the base cylinder, with its zero
roll position precalibrated (see figure 14). Probe tips
may be chisel point, disk, or spherical, provided that
accurate positioning of the point of contact between
the probe tip and the gear tooth surface is main-
tained within the base tangent plane. Measurement
of extreme profile modifications may be adversely
affected by shifting of the probe contact vector.
Root circle Outside circle
Base circle Pitch circle
Probe Axis
Base tangent plane
Figure 14 -- Profile measuring method
It is often desirable to orient the measurement probe
path of motion normal to the tooth surface.
ANSI/AGMA 2015--1--A01 specifies profile toler-
ances in the transverse plane. If measurements are
made normal to the tooth surface, all values must be
corrected by dividing by the cosine of the base helix
angle, cos βb, before comparison against the
tolerances.
7.2.2 Coordinate measurement inspection
instruments
Involute profile can be inspected by non--generative,
coordinate measurement instruments. Such instru-
ments indicate the tooth profile by a series of points,
storing the coordinates of each point. The deviation
of the actual profile from the nominal is then
determined by comparison of the stored test point
coordinates against calculated coordinates of the
theoretical nominal profile (see figure 15).
Coordinate measurement inspection instruments
may operate in two dimensions (X and Y coordi-
nates) or three dimensions (X, Y, and Z coordinates).
Measurement of an involute profile with two--
dimensional systems requires accurate mounting of
13
AGMA 915--1--A02
the gear with its datum axis perpendicular to the X--Y
plane. Three--dimensional systems require align-
ment of the gear datum axis parallel to one of the
three instrument axes. This may be accomplished
by accurate mounting of the part, or mathematically
adjusting the instrument axes to coincide with the
gear axis. Coordinate measurement inspection
instruments may use spherical measurement probe
tips, which require correction for shifting of the probe
contact vector.
X1
X2
X3
Y1
Y2
Y3
Figure 15 -- Profile inspection by coordinates
7.2.3 Portable involute measurement
instruments
Profile measuring instruments are generally fixed
type machines. Gears to be tested must be brought
to the instrument and accurately mounted, typically
on--axis, between centers or on a table. For very
large gears it may be necessary to employ a portable
involute measuring instrument that can be taken to
the gear. Such instruments may operate on a variety
of generative or non--generative principles. The
portable instrument must be accurately mounted at a
True Plus profile
involute profile (minus pressure angle)
True
involute
Profile control
diameter
Figure 16 -- Typical tooth profile measurement charts
14
AMERICAN GEAR MANUFACTURERS ASSOCIATION
known distance from, and in alignment with, the gear
axis. This requires care in design and manufacture
of the gear blank.
7.3 The profile diagram
Amplified traces of the profile inspection test results
should be presented on charts that are graduated for
rolling path length or degrees of roll. They should
also be labeled for magnification and evaluation
points in conformance with the specification.
An unmodified involute profile with no deviations will
be charted as a straight line. Deviations of the curve
from a straight line represent, in magnified form,
deviations of the actual profile from an unmodified
involute. Profile modifications introduced by the
designer also appear as departures from the straight
line, but they are not considered to be deviations
from the “design profile”.
Excess material on the profile is considered a plus
deviation, while insufficient material is considered a
minus deviation. In addition to identifying the
location and magnitude of the highest point on the
profile or the maximum profile deviation, these
charts are valuable for determining profile character-
istics such as tip break, undercut, and tip or root relief
(see figure 16).
Any point along the profile diagram can be related to
a diameter (radius), a base tangent length and an
involute roll angle.
Figure 17 shows a sample tooth profile and the
relation to the corresponding profile trace, together
with the appropriate terms. Details of terms,
definitions and concepts concerning the profile
trace, are provided in ANSI/AGMA 2015--1--A01.
Tip break
Undercut
Minus profile Undercut &
(plus pressure angle) tip chamfer
Tip break
Undercut
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

fHα

ffα

A
B

C
Lαc
2
3

D
E
+

Fα

Lαc
Lε

A
A
B
tip circle of mating gear
2 1
Q
tip circle C
E

reference circle D F
E ξC
F

root circle
base circle

1 Design profile C--Q Base tangent length to point C

2 Measured profile ξc Involute roll angle to point C
3 Mean profile line Q Start of roll (point of tangency of transverse
base tangent)
A Tip circle point Lαc Profile evaluation range
B Start of tip break (chamfer) Lε Base tangent length to start of active profile
D Start of active profile Fα Total profile deviation
E Profile control diameter ffα Profile form deviation
F Origin of involute fHα Profile slope deviation
B--D Active profile
B--E Usable profile
Figure 17 -- Tooth profile and profile diagram

15
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

7.4 Evaluation of profile diagrams
Depending on accuracy class specified, it may only
be necessary to measure total profile deviation, Fα.
See ANSI/AGMA 2015--1--A01, clause 4.
It may also be necessary to determine the profile
slope deviation, fHα, and the profile form deviation,
ffα. For this it is necessary to superpose the mean
profile line onto the diagram as shown in figure 17,
also in figure 2 of ANSI/AGMA 2015--1--A01. Allow-
able values of fHα and ffα can be calculated in accor-
dance with ANSI/AGMA 2015--1--A01, clause 7.
7.5 Algebraic signs of fHα and fα
The profile slope deviation, fHα, is termed positive
and the corresponding pressure angle deviation, fα,
is termed negative when the mean profile line rises
towards the tooth--tip end A of the diagram, as shown
in figure 17. In figure 18, both positive and negative
slopes, caused by eccentricity of mounting on the
gear generating machine, are shown.
If the slopes seen in the profile diagrams of mating
gears are equal and have the same sign, the
deviations are mutually compensating. This applies
to both external and internal gears.
A B + E
1 3
--11.1
7.6 Mean profile slope deviation, fHαm
Slope deviations of individual profiles can be caused
by eccentricity due to inaccuracies of manufacturing
or inspection set--up. Such deviations will vary
around the gear. The use of mean profile slope
deviations cancels out the influence of eccentricity
on individual profile traces.
The effect of eccentricity on profile slope, and the
determination of mean profile slope deviation, are
illustrated in figure 18.
Calculating the mean profile slope deviation is a step
towards the correction of manufacturing processes
or other suitable action.
For all practical purposes, it is usually sufficient to
calculate the arithmetic mean of the profile slope
deviations by calculating the average of the devi-
ations measured on three or more corresponding
flanks of equally spaced teeth around the gear
circumference according to the following equation:
f Hαm = 1 
n f Hα1 + f Hα2 + + f Hαn
 (8)
where:
fHαm is the mean profile slope deviation, mm;
fHαn is the individual profile slope deviations, mm;
n is the number of profile slope deviation
values included in the mean.

--

M
+ C
fe

2
--6.6

I φ
-- 2

+ 1
5.7

3
--
fHα

Lαc

f Hαm = 1 (− 11.1 − 6.6 + 5.7) = − 4mm

3
M = axis of rotation of the gear on the machine tool.
I = axis of rotation of the gear on the inspection apparatus.
C = position of tool or profile measuring probe
1, 2, 3 = Positions of the profiles from which the traces were obtained (at 45°, 165°, 285°) and
relevant profile traces
Figure 18 -- Mean profile slope deviation, fHαm

16
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

7.7 Additional calculations for profile 7.7.3 Calculation of effective transverse

measurements pressure angle, αt eff

The mean profile slope deviation, fHαm, can be used Effective transverse pressure angle, αt eff, can be
to calculate a variety of parameters that are useful for calculated as follows:
controlling the quality of gear involute profiles.
Included in these calculated parameters are: α t eff = acos  
d b eff
D
(11)

-- mean base diameter difference, fdbm; where:

-- effective base diameter, db eff; αt eff is the effective transverse pressure angle,
-- effective transverse pressure angle, αt eff; degrees;
-- effective normal pressure angle, αn eff; D is the design pitch diameter, mm.
-- mean transverse pressure angle deviation, 7.7.4 Calculation of effective normal pressure
fαmt; angle, αn eff
-- mean normal pressure angle deviation, fαmn. Effective normal pressure angle, αn eff, can be
calculated as follows:
All of the following equations are based upon the
mean profile slope deviation, fHαm. Alternatively, the α n eff = atan tan α t eff cos β (12)
same formulas could be applied to the case of
where:
individual tooth data. The calculation sequence
would then commence with the entry of the individual αn eff is the effective normal pressure angle,
profile slope deviation, fHα. degrees;
β is the helix angle, degrees.
7.7.1 Calculation of mean base diameter
difference, fdbm 7.7.5 Calculation of mean transverse pressure
angle deviation, fαmt
Mean base diameter difference, fdbm, can be calcu-
Mean transverse pressure angle deviation, fαmt, can
lated as follows:
be calculated as follows:
Db
f dbm = f (9) f αmt = α t eff − α t (13)
L αc Hαm
where:
where:
fαmt is the mean transverse pressure angle
fdbm is the mean base diameter difference, mm; deviation, degrees;
Db is the base diameter, mm; αt is the design transverse pressure angle,
degrees.
Lαc is the functional profile length, mm;
Alternatively, fαmt can be calculated (in degrees) by:
fHαm is the mean profile slope deviation, mm.
A positive mean profile slope deviation (profile trace
rising towards its tooth tip end) implies that the
f αmt = − 1  f Hαm

L αc tan α t × 10 3
180
π
 (14)

effective base diameter is too large, and visa versa.

A positive mean profile slope deviation (profile trace
when fHαm > 0, then fdbm > 0 rising towards its tooth tip end) implies that the
effective pressure angle is too small, and visa versa.
7.7.2 Calculation of effective base diameter,
db eff when fHαm > 0, then fαmt < 0

Effective base diameter, db eff, can be calculated as 7.7.6 Calculation of mean normal pressure
follows: angle deviation, fαmn
Mean normal pressure angle deviation, fαmn, can be
d b eff = D b + f dbm × 10 −3 (10)
calculated as follows:
where: f αmn = α n eff − α n (15)
db eff is the effective base diameter, mm. where:

17
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

fαmn is the mean normal pressure angle an indication of profile accuracy (see figure 20).
deviation, degrees; However, readings give no indication as to which
profile may have an error, since two flanks of a
αn eff is the effective normal pressure angle,
measured tooth are contacted at the same time.
degrees;
This method will not reveal deviations that cancel
αn is the design normal pressure angle, each other, such as those caused by a form cutter,
degrees. which has been offset from a true radial position.
A positive mean profile slope deviation (profile trace
rising towards its tooth tip end) implies that the
effective pressure angle is too small, and visa versa.
when fHαm > 0, then fαmn < 0
7.8 Other profile measuring methods
While not commonly used or recommended, the
following profile measuring methods may prove
valuable when more conventional methods are not
practical or available.
7.8.1 Projection
A shadow of the gear tooth under inspection may be
optically magnified and directly or reflex projected to
permit comparison of the profile to a large scale
layout of a specified profile (see figure 19). This
method is normally applied only to fine pitch gears.
When gears are too large to be mounted in the
projector, a thin wafer (manufactured simultaneous-
ly with the gear), or a mold of a gear tooth form may
be used for projection. This method requires two
known reference surfaces to locate the image both Figure 20 -- Profile inspection by gear--tooth
radially and angularly. caliper method

-- Auxiliary gaging elements. The theoretical

position of wires, rolls, pins, or balls of several
different diameters placed in a tooth space may
be computed and compared to actual measure-
Scale ments (see figure 21). This method has limita-
layout tions similar to those of gear tooth caliper
measurements.
Projection

Figure 19 -- Profile inspection by optical

projection

7.8.2 Indirect profile inspection methods

The following techniques may be employed for
inspection of gear profiles. These methods do not
yield actual measurements of deviation of an in-
spected profile from a nominal.
-- Multiple thickness measurement. The chord-
al tooth thickness and associated addendum
depth for several positions on a tooth may be
computed for a gear tooth caliper. Comparison of Figure 21 -- Profile inspection by measurement
measurements with the computed values will give over pins

18
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

7.8.3 Profile measuring with master gear 8.2 Helix inspection methods

Contact pattern checking with a master gear may be
used to check the profile deviation of gears in place
or when gears are too large to be accommodated by
a profile measuring instrument. The axis of the gear
and master must be parallel. Refer to clause 10 for
more information concerning this method.
8 Measurement of helix deviations
8.1 Helix
Helix is the lengthwise shape of the tooth flank
across the face from one end to the other. The
theoretical helix of a helical gear is contained on the
surface of a cylinder, which is concentric with the
datum axis of rotation of the gear, at the intersection
of that cylinder with the tooth flank. The theoretical
helix of a spur gear is a straight line parallel to its
rotating axis. Helix is restricted to the operating
portion, which is intended to be in contact during
loaded operation, and does not include edge rounds
or chamfers.
The standard methods of helix measurement are
with generative, coordinate, or portable helix
measuring instruments.
8.2.1 Generative helix measuring instruments
The most common instruments used for measure-
ment of helix are generative helix measurement
instruments. Such instruments measure the devi-
ation of the actual helix from a nominal helix, which is
generated by the instrument. Generation of the
nominal helix requires the axial movement of a
measurement probe together with a rotational move-
ment of the gear mounted on the instrument spindle.
These movements must be synchronized according
to the specified lead of the gear (see figure 23).
When measuring spur gears, the rotational move-
ment is eliminated.
Total helix deviation, Fβ
Reference
zero
0

Lead, as a term used for helical gears, is the axial

advance of a helix for one complete turn of the gear.
The lead of a spur gear, therefore, is infinite. The
lead of a helical gear is commonly defined by the
angle between the helix at the standard pitch
diameter and the axis of rotation.

Helix deviation is the difference between the speci-

fied and the measured helix of the gear (see figure
22). ANSI/AGMA 2015--1--A01 specifies the Probe travel
direction of tolerancing for helix deviation to be within
the transverse plane, tangent to the base circle.
Helix angle

Total helix
Measured deviation, Fβ
helix

Figure 23 -- Graphic charting of helix

Generative helix measuring instruments may

Design helix employ a variety of mechanical configurations to
generate the nominal helix. For example, the gear
can be rotated by a master disk driven by a straight
Helical tooth
edge, which in turn is driven by the axial movement
Figure 22 -- Helix deviation of the probe slide. The tangential movement of the

19
AGMA 915--1--A02
straight edge is translated into axial movement of the
probe by a ratio mechanism. Combination instru-
ments also capable of measuring involute profile
often utilize their master base circle mechanisms in
this manner.
Other configurations include master lead bar and
follower mechanisms, and master lead screw and
change gearing mechanisms. Newer generative
helix measuring instruments typically use a comput-
er numeric control drive system to generate the
nominal helix.
Helix measurements must be made relative to the
datum axis of rotation of the gear. Refer to 5.1 for
more information concerning the datum axis of
rotation.
Probe tips most commonly used are spherical or
disk--shaped. The probe tip must be positioned to
contact the tooth surface at the specified tolerance
diameter, dT.
It is often desirable to orient the measurement probe
path of motion normal to the tooth surface.
ANSI/AGMA 2015--1--A01 specifies helix tolerances
in the transverse plane. If measurements are made
normal to the tooth surface, all values must be
corrected by dividing by the cosine of the base helix
angle, cos βb, before comparison against the
tolerances.
8.2.2 Coordinate measurement inspection
instruments
Helix can be inspected by non--generative, coordi-
nate measurement instruments. Such instruments
probe the tooth lengthwise at a series of points,
storing the coordinates of each point. The deviation
of the actual helix from the nominal is then deter-
mined by comparison of the stored test point
coordinates against calculated coordinates of the
theoretical nominal helix.
Coordinate measurement inspection instruments
operate in three dimensions (X, Y, and Z coordi-
nates) to measure helix. The gear axis must be
aligned parallel with one of the three instrument
axes. This may be accomplished by accurate
mounting of the part, or mathematically adjusting
instrument axes to coincide with the gear axis.
Coordinate measurement inspection instruments
commonly use spherical measurement probe tips,
which require correction for shifting of the probe
contact vector.
20
AMERICAN GEAR MANUFACTURERS ASSOCIATION
8.2.3 Portable helix measuring instruments
Helix measuring instruments are generally fixed type
machines, which require that gears to be tested must
be brought to the instrument and accurately
mounted, typically on--axis between centers or on a
table. However, for very large gears it may be
preferable to employ a portable helix measuring
instrument, which can be taken to the gear. The
portable instrument must be accurately mounted at a
known distance from, and in alignment with, the gear
axis. This often requires extra care in design and
manufacture of the gear blank.
8.3 The helix diagram
Amplified traces of helix inspection test results
should be presented on charts that are calibrated for
axial probe travel as well as magnification of
measured deviation. Sometimes trace lengths are
magnified representations of small facewidths, or
reduced representation of large facewidths.
An unmodified helix with no deviations will be
charted as a straight line. Deviations of the curve
from a straight line represent, in magnified form,
deviations of the actual helix from an unmodified
helix. Helix modifications introduced by the designer
also appear as departures from the straight line, but
they are not considered to be deviations from the
design helix.
Excess material on the helix is considered a plus
deviation while insufficient material is considered a
minus deviation. In addition to identifying the
location and magnitude of the helix deviation, these
charts are valuable for determining helix characteris-
tics such as edge rounds, crowning, and end relief.
Relevance to right hand and left hand helices can be
indicated by means of the letters “r” and “l”,
respectively, used either as symbols or as sub-
scripts.
In figure 24, a typical example of a helix diagram
shows the helix deviations of a tooth flank of which
the design helix is an unmodified helix. Had the
design helix been crowned, end relieved or other-
wise modified, traces representing it would be
appropriately formed curves.
Details of terms, definitions and concepts concern-
ing the helix trace are provided in ANSI/AGMA
2015--1--A01.
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

λβx λβx able values of fHβ and ffβ can be calculated in accor-
dance with ANSI/AGMA 2015--1--A01, clause 7.

fwβ
2 3 8.5 Algebraic signs of fHβ and Fβ
Helix slope deviation, fHβ, and the total helix devi-
ation, Fβ, are to be reported with an algebraic sign.

fHβ
Fβ

Deviations are deemed to be positive (fHβ > 0 and

ffβ

Fβ > 0) when helix angles are larger, and negative

when helix angles are smaller, than the design helix
1 angle.
Lβ
I II The helix deviations of spur gears if other than zero
b are indicated by the subscripts “r” and “l”, instead of
an algebraic sign, implying deviations in the sense of
1 Design helix Fβ Total helix right or left hand helices, respectively.
deviation
2 Actual helix trace ffβ Helix form In figure 25, both positive and negative slopes,
deviation caused by eccentricity or wobble of mounting on the
3 Mean helix line fHβ Helix slope gear generating machine, are shown.
deviation
b Facewidth or λβx Axial wavelength
fHβ1 fHβ2 fHβ3 fHβ4
distance between of undulation
chamfers
Lβ Helix evaluation fwβ Undulation height
range
I Reference face II Non--reference
face -- + -- + -- + -- +
Lβ
b

Figure 24 -- Helix diagram

0° (360°) 90° 180° 270°

The helix evaluation range, Lβ, is equal to the length
of trace, reduced at each end by the smaller of two
values: 5% of the helix length of trace, or the length
equal to one module. This reduction is made in order
to ensure that unintentional, slight end reliefs caused
by some machining conditions, are not normally
included in the assessment of the deviation magni-
tudes intended for comparison with stringent toler-
ances. For assessment of the total helix deviation,
Fβ, and the helix form deviation, ffβ, excess material
within the end zones of 5%, which increases the
amount of deviation shall be taken into account.
8.4 Evaluation of helix diagrams
Figure 25 -- Traces generated from four tooth
flanks
If the helix slope deviation, fHβ, (assuming equal
evaluation ranges) of the corresponding flanks of
two mating gears are equal in magnitude and
algebraic sign, the deviations are mutually compen-
sating.
8.6 Mean helix slope deviation, fHβm
For correction of machine tool settings or adaptation
to a mating gear, determination of the mean helix
slope deviation, fHβm, of the gear is useful.

For purpose of gear quality classification, it may be
necessary to measure only “total helix deviation”, Fβ.
See ANSI/AGMA 2015--1--A01, clause 4.
It may also be necessary to determine the “helix
slope deviation”, fHβ, and the “helix form deviation”,
ffβ. For this it is necessary to superpose the “mean
helix line” onto the diagram as shown in figure 24
(also in ANSI/AGMA 2015--1--A01, figure 1). Allow-
If the helix slope deviations are either random or are
fairly consistent, then the mean helix slope deviation
may be used to correct the helix setting of the
machine used to manufacture the gear. In the case
of a matched set of mating gears where one has
been manufactured and inspected, then the mean
helix slope deviation may be used to adjust the
manufacture of the other gear in the set. This will
result in improved contact between the gears without

21
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

the need to make corrections to the previously controlling the quality of gear helices. Included in
finished gear. these calculated parameters are:

If the helix slope deviation, fHβ, varies in a regular -- effective helix angle at the tolerance
pattern around the circumference of a helical gear, diameter, βT eff;
then the datum axis of the gear was probably tilted, -- effective lead, Leff;
offset, or mis--orientated relative to the machine axis
-- effective helix angle at the standard pitch
during either manufacture or inspection. See figure diameter, βeff;
25. Tilting affects spur gears in the same manner,
but offset (eccentricity) does not. -- mean lead difference, fLm;
-- mean helix angle deviation, fβm.
-- Eccentricity: The variation of helix slope
deviation caused by eccentricity (if within speci- All of the following equations are based upon the
fied limits) is not normally detrimental to the mean helix slope deviation, fHβm. Alternatively, the
operation of the gear. same formulas could be applied to the case of
individual tooth data. The calculation sequence
-- Tilting: Variation of helix slope deviation
would then commence with the entry of the individual
caused by mis--orientation of the gear teeth
relative to the datum axis may affect the proper helix slope deviation, fHβ.
functioning of the gear. The helix slope deviations 8.7.1 Required preliminary data
will cause the center of contact pressure to shift
axially back and forth with each revolution. This The following data is required for the additional
may in turn cause premature gear tooth failure calculations for helix measurements.
and/or bearing problems. Therefore, attention
should be drawn to this condition even if the Lead, L, can be calculated as follows:
deviations are within tolerance.
L= Dπ (17)
tan β
The mean helix slope deviation, fHβm, is calculated
by averaging the helix slope deviation, fHβ, observed where:
on the corresponding flanks of three or more teeth
L is the lead of the design helix, mm;
equally spaced around the circumference of the
gear. D is the standard pitch diameter, mm;
β is the helix angle at the standard pitch diam-
f Hβm = 1  
n f Hβ1 + f Hβ2 + + f Hβn (16) eter, degrees.

where: Mean helix slope deviation, in the transverse plane

and tangent to the tolerance diameter, fHβmt, can be
fHβm is the mean helix slope deviation, mm; calculated as follows:
fHβn are the individual helix slope deviations, mm; f Hβm
f Hβmt = (18)
n is the number of helix slope deviation values
included in the mean.
 
Db
dT

A suitable mean value can be obtained from the helix where:

diagrams of corresponding flanks of two diametrical-
ly opposite teeth. However, if the helix slope
deviations vary around the gear, this will not always
be disclosed unless traces of at least three
equispaced flanks are obtained.
8.7 Additional calculations for helix
measurements
The mean helix slope deviation, fHβm, can be used to
calculate a variety of parameters that are useful for
fHβmt is the mean helix slope deviation, in the
transverse plane and tangent to the toler-
ance diameter, mm;
fHβm is the mean helix slope deviation, in the
transverse plane and tangent to the base
diameter (the tolerance direction specified
in ANSI/AGMA 2015--1--A01), mm;
Db is the base diameter, mm;
dT is the tolerance diameter, mm.

22
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

8.7.2 Calculation of effective helix angle at the Desired lead Actual lead
tolerance diameter, βT eff of helix of helix

Effective helix angle at the tolerance diameter, βT eff,

can be calculated as follows: Facewidth

 
−3 L  +  d π 
 Hβmt
f × 10
Lβ T 
β T eff = atan  Path of contact
 L  of measuring pointer

  (19)
where:
βT eff is the effective helix angle at the tolerance
diameter, degrees;

Lβ is the helix evaluation range, mm.

Facewidth
8.7.3 Calculation of effective lead, Leff

Effective lead, Leff, can be calculated as follows:

dT π
L eff = (20)
tan β T eff
Figure 26 -- Helix of right hand helical gear with
where:
short lead (+ helix angle)
Leff is the effective lead.

8.7.4 Calculation of effective helix angle at the

standard pitch diameter, βeff
Actual lead
Effective helix angle at the standard pitch diameter, of helix Desired lead
of helix
βeff, can be calculated as follows:

 
β eff = atan D π
L eff
(21) Facewidth

8.7.5 Calculation of mean lead difference, fLm

Mean lead difference, fLm, can be calculated as Path of contact

follows: of measuring pointer

f Lm = L eff − L (22)

where:

fLm is the mean lead difference, mm

Facewidth
A positive mean lead difference implies that the
effective lead is too long, and visa versa. See figures
26, 27, 28 and 29.

A positive mean helix slope deviation implies that the

effective lead is too short, and visa versa.
Figure 27 -- Helix of right hand helical gear with
when fHβm > 0, then fLm < 0 long lead (-- helix angle)

23
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

Desired lead Actual lead 8.7.6 Calculation of mean helix angle

of helix of helix deviation, fβm
Mean helix angle deviation, fβm, can be calculated as
Facewidth follows:
f βm = β eff − β (23)

Path of contact
of measuring pointer
Facewidth
Figure 28 -- Helix of left hand helical gear with
long lead (-- helix angle)
where:
fβm is the mean helix angle deviation, degrees.
A positive mean helix slope deviation implies that the
effective helix angle is too large, and visa versa.
when fHβm > 0, then fβm > 0
8.8 Undulations
Undulations are helix form deviations having
constant wavelength and almost constant height.
Perturbations of gear production machine transmis-
sion elements are their most common cause,
especially those of:
a) the cutter saddle feed--screw drive, and
b) the worm of the indexing wormgear drive.
The wavelength of undulations caused by a),
measured in direction of helix, is equal to the pitch of
the feed--screw divided by cos β.
Of undulations due to cause b) the wavelength is:

λβ = dπ (24)
Actual lead Desired lead z M sin β
of helix of helix
where:
λβ is the undulation wavelength, mm;
Facewidth d is the reference diameter, mm;
zM is the number of teeth in the master
indexing worm wheel.

Path of contact
The number of undulations generated as a result of
of measuring pointer b), projected into a transverse plane, are equal to the
number of teeth, zM, of the master indexing worm
wheel. These can be sources of objectionable
pure--tone components of noise spectra, also known
as ghost harmonics, at frequencies corresponding to
the rotational speed (revolutions) of the affected
gear multiplied by zM.
Facewidth
The method of application of the undulation measur-
ing attachment of a helix measuring apparatus is
shown in the diagram in figure 30. This is discussed
in the following.
When undulations due to the cause a) or b)
Figure 29 -- Helix of left hand helical gear with mentioned above are to be measured, the appropri-
short lead (+ helix angle) ate wavelength is calculated and the spherical

24
AMERICAN GEAR MANUFACTURERS ASSOCIATION
location feet of the attachment are set at an odd
number of wavelengths distant from each other.
The amount of the undulations is indicated by a
probe situated midway between the feet as the latter
are slid along the helix.
It can be seen in the figure that the displacement of
the probe, when a peak and next a trough are sensed
by the probe, is equal to twice the height of the
undulation as shown in figure 30. This feature
enhances the sensitivity of the apparatus, which also
plots the results in the form of a diagram.
It should be noted that the undulations would not be
indicated if the feet were spaced at a distance equal
to an even number of wavelengths as shown in figure
30 with s = 4λβ.
8.9. Indirect helix inspection methods
The following indirect methods may be employed for
inspection of gear helix. These methods do not
provide the actual levels of helix deviation. However,
in instances where the measurement methods
+fwβ
fwβ
s s
s = λβ
s = 4 λβ
Figure 30 -- Principle of undulation inspection
AGMA 915--1--A02
outlined in 8.2 are impractical, these indirect meth-
ods may prove valuable.
8.9.1 Helix indication using axial pitch
An indication of helix accuracy may be derived from
inspection of axial pitch on gears with sufficient helix
angle and face width to have multiple axial overlaps.
The measurement must be made parallel to the gear
axis at increments equal to the axial pitch. The
deviation in resulting measurement values is indica-
tive of deviation of helix. Pitch deviations of the
measured teeth can affect axial pitch measure-
ments, and must be considered. The axial pitch
method of helix inspection is especially attractive for
large diameter, wide facewidth gears with large helix
angles.
8.9.2 Helix measuring with a master gear
Contact pattern checking with a master gear may be
used to check the helix deviation of gears in place, or
when gears are too large to be accommodated by a
helix inspection instrument. The axes of the gear
and master must be parallel. Refer to clause 10 for
more information concerning this method.
--fwβ
fwβ
25
AGMA 915--1--A02
9 Measurement of single flank composite
deviations
9.1 Single flank composite
Tangential (single flank) composite measurement
can provide valuable information about the transmis-
sion error of a gear, a pair of gears, or an entire gear
train. Transmission error is the deviation of the
position of a driven gear from the position that the
driven gear would occupy if all the gears involved in
the measurement were geometrically perfect.
ANSI/AGMA 2015--1--A01 provides tolerances for
two characteristics of transmission error for individu-
al product gears measured with a master gear, total
single flank composite deviation, Fis, and tooth--to--
tooth single flank composite deviation, fis.
The following is a description of the measuring
methods and a guide to interpretation of the data
generated during single flank measurement of
individual gears measured with a master gear.
Single flank measurement of a pair of product gears
is also described.
Single flank measurement of more than a single
mated pair of gears is the assessment of the
kinematics of a gear transmission. This is not
considered to be within the scope of this document or
ANSI/AGMA 2015--1--A01.
9.2 Single flank composite measurement
For measurement of single flank composite devi-
ations, two gears are mounted rotatably in mesh at
an appropriate center distance. The gears are
mounted with backlash so that contact occurs only
on one set of corresponding flanks. Rotating
synchronously with each gear is a device capable of
Double flank gear test
Measures variation in center distance
Figure 31 -- Composite gear testing, double and single flank
26
AMERICAN GEAR MANUFACTURERS ASSOCIATION
measuring angular motion. These are typically
rotary optical encoders (gratings and reader head
assemblies). Rotary accelerometers and velocity
transducers have also been used as sensing de-
vices. See figures 31 and 32.
During measurement one gear acts as the driver,
rotating the other gear. During rotation, the angular
positions of the driven gear relative to the driver is
calculated through ratioing of the signals from the
two sensing devices using analog or digital electron-
ics. These relative positions are recorded either on a
strip chart or into digital storage on a computer until a
complete diagram has been generated. To compare
these angular readings to the tolerances provided in
ANSI/AGMA 2015--1--A01 they must first be con-
verted to linear values at the tolerance diameter
specified.
Single flank composite measurements are per-
formed with tooth flank contact maintained, under
very light load, and with low angular velocities. The
results generated reflect the combined elemental
deviations (profile, helix, pitch) of both gears.
Single flank composite deviations of heavily loaded
gears can also be similarly measured. Under these
conditions, recorded deviations are influenced by
load induced tooth deformations, by mesh stiffness
variation, and depending on the speed of rotation by
impact effects, as well as by imperfections of tooth
geometry. ANSI/AGMA 2015--1--A01 does not apply
to this kind of measurement.
9.2.1 Single flank composite deviations
Total single flank composite deviation, Fis, is the
maximum measured transmission error range when
the gear is moved through one complete revolution.
See figure 33.
Single flank gear testing
Measures rotational movements
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

z1 = Driving gear z2 = Driven gear

z1 z2

Optical gratings

Reading heads
f1 pulses/sec
f2 pulses/sec
Multiplier
z1

Divider Phase comparator

z2

z1
 
z 2 f 1 = f 2 pulses∕sec

Figure 32 -- Schematic of a single flank measuring device

One gear revolution

40

30

20
fpt
Amplitude, 0.001 mm

10
Fp
0
Fis
--10

--20

--30

--40
0 1 2 3 4 5 6 7 8 9 10 11 12
Tooth number
Figure 33 -- Individual tooth deviations revealed by single flank testing

27
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

40

30

20

10
Amplitude, 0.001 mm

0
fis
--10

--20

--30

--40
0 1 2 3 4 5 6 7 8 9 10 11 12
Tooth number

Figure 34 -- Filtered signal from figure 33 (eccentricity removed)

Tooth--to--tooth single flank composite deviation, fis
is the value of the greatest measured transmission
error over any one pitch (360/z) after removal of the
long term component (sinusoidal effect of eccentrici-
ty) when the gear is moved through one complete
revolution. See figure 34.
9.3 Single flank measurement with master
gear
Recorded diagrams of single flank composite mea-
surements generally include short period compo-
nents corresponding to successive cycles of tooth
engagement, superposed on long period compo-
nents associated with complete revolutions of each
of the meshing gears.
The diagram in figure 33 represents the record of
single flank composite deviations generated during
one revolution of a pinion having 12 teeth when
meshed with a master gear.
9.3.1 Master gear requirements
For single flank measurement of individual product
gears, a master gear of known accuracy (calibrated)
and specifically designed to mesh with the product
gear to be inspected should be used. Attention must
be paid to the fact that the quality of the master gear
will influence the measurement of product gears. If
the quality of the master is at least 4 accuracy grades
better than the required grade of the product gear,
inaccuracies of the master are usually ignored. If the
quality of the master is less than 4 accuracy grades
better than the required grade of the product gear,
inaccuracies of the master should be taken into
account.
9.3.2 Influence of profile deviations
When using a master gear in the measuring of single
flank composite deviations, the assumption that the
master gear is perfectly accurate implies that the
generated single flank composite deviation diagram
represents only the combined deviations of the tooth
elements of the product gear.
Figure 35 shows schematically, single flank compos-
ite recordings of three consecutive cycles of tooth
engagement of a master gear and product gear.
Each corresponds to a different tooth profile. The
first is unmodified and faultless, the second being
progressively modified from mid--depth towards
each limit of the active profile, and the third with
negative profile slope deviation.

28
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

progressively decreasing trend as contact ap-

A proaches the end of the tooth engagement cycle.
tip
master
test
perfect conjugate In figure 35(C), the saw tooth components of the
gear tooth shape diagram show progressive single flank composite
deviation from zero to a negative value as contact
root
moves from the product gear tooth tip towards the
+ start of active tooth profile. At this point, contact
displacement

1 pitch 1 pitch 1 pitch

abruptly transfers to the following tooth with the
introduction of an equally abrupt positive deviation.
angular

0
It must be noted that diagrams of single flank
angular motion curve composite measurements do not merely reflect
-- influences of profile deviations revealed by mea-
surements made on a few teeth, but may be
B influenced by contact involved in any prominences
tip on the working surfaces of the teeth of the product
master
modified tooth gear.
test shape [profile
gear
barrelling Cα] 9.3.3 Influence of pitch deviations
root
Each single pitch deviation introduces a local
+ 1 pitch 1 pitch 1 pitch tangential component, which will show on the single
displacement

flank composite diagram as a displacement of the

corresponding profile generated component of the
angular

0
diagram.

-- The schematic diagram in figure 33 illustrates the

angular motion curve influence of single pitch deviations, fpt, on the single
flank composite diagram.

tip C Single pitch deviations have a cumulative effect on

master the single flank composite displacement arc as they
test modified tooth pass through the mesh. Their influence is clearly
gear shape [modified
pressure angle]
visible on the single flank composite diagram. This
root enables values of cumulative pitch deviations (e.g.,
when k=2, k=3, etc.) to be determined as the
+ 1 pitch 1 pitch 1 pitch
ordinates of tangents to the apices at appropriate
displacement

numbers of pitches apart. The principle is illustrated

0
in figure 33, in which influences of single pitch
angular

deviation and the approximate total cumulative pitch

-- deviation, Fp , are indicated.
angular motion curve
9.3.4 Influence of helix deviations
Figure 35 -- Angular motion curves from tooth
modification A helix slope deviation that is constant in magnitude
and sign, (i.e., is common to every tooth of a gear)
results in consistent localized bearing in the mesh.
Figure 35(A) shows the straight line diagram gener-
This does not substantially influence the single flank
ated by a product gear and master gear that both
composite deviations of spur gears. The single flank
have fault--free unmodified teeth. In figure 35(B), the
composite deviations of helical gears, however, may
record indicates the influence of tip and root relief in
be adversely effected by a constant helix slope
the form of a modification over the whole profile.
deviation. This is due to the different nature of the
From the start of the tooth engagement cycle with
path of contact of helical gears.
first contact at the tooth tip of the driven product gear,
the deviation value increases progressively to zero When helix slope deviations vary in magnitude
as contact nears mid--depth, then changes to a and/or sign around a product gear, the bearing

29
AGMA 915--1--A02
contact location will vary around the gear. This
condition may adversely effect single flank compos-
ite deviations of both spur and helical gears.
Helix form deviations do not substantially influence
the single flank results of spur gears. Single flank
composite deviations of helical gears, however, may
be adversely effected by helix form deviations.
9.3.5 Influence of contact ratio
A single flank composite deviation diagram gener-
ated from a master--gear and product--gear
combination is composed of successive curves
representing for the most part the profile deviations,
as shown in figure 36. Single--pair and two--pair
tooth engagements and the single flank composite
deviation diagram during a complete cycle of tooth
engagement is clearly illustrated. It can easily be
root
master gear
profile deviation diagrams
product gear
tip
direction of paper feed
tangential composite 1
deviation
master gear
pb = base pitch
gα = length of path of contact
Figure 36 -- Effect of contact transfer on the profile component in a tangential composite deviation
diagram (spur gears)
30
AMERICAN GEAR MANUFACTURERS ASSOCIATION
recognized that the maximum length of the single--
pair tooth contact path is realized when the contact
ratio, εγ, is equal to one. As the contact ratio
increases, this length reduces and when the contact
ratio is equal to or greater than two, there is no
single--pair tooth contact at all.
When the total contact ratio, εγ exceeds two, which is
normally the case for helical gears, the short period
components which represent profile irregularities
are smoothed to some extent because in general,
simultaneous contact takes place on two or more
tooth pairs.
Diagrams in figure 37 with the two cases “A”
(generated from helical gears) and “B” (from spur
gears) illustrate the difference between the ways in
which the influence of the overlapping teeth of the
two types combine.
tip
root
2 3
stylus
profile
component
3
product gear
2
1
pb
pb
gα
AMERICAN GEAR MANUFACTURERS ASSOCIATION
number of tooth pairs in mesh
3 2
number of tooth pairs in mesh
2 1
Figure 37 -- Influence of overlap ratio
Single flank composite results can be very different
from what is expected, especially if these expecta-
tions are derived from consideration of theoretical
contact ratio and an assumption that contact is
perfect over the tooth profiles and facewidth of
helical gears.
Single flank composite deviations can be influenced
by modification of tooth profile and helix (tip relief,
crowning, etc.) introduced to accommodate possible
deformations of shafts, housings and teeth under
load. If under full load the tooth bearing is uniformly
distributed over the working surfaces of the teeth,
such is not likely to be the case under the light load
conditions used during single flank composite mea-
surement where the tooth bearing may be localized.
Given these circumstances, the contact ratio during
measurement is likely much less than elementary
theory would suggest.
9.3.6 Interpretation of results
This section contains information and techniques for
interpreting single flank composite results beginning
with a comparison to methods used for traditional
double flank (radial composite) testing.
Double flank composite data charts are made up
primarily of information related to radial runout, Fr
(long term component), and deviations of tooth--to--
AGMA 915--1--A02
3 2 3 A
B
2
tooth composite of both flanks, fid (short term
component).
Single flank composite data charts are made up
primarily of information related to cumulative pitch,
Fp (long term component) deviations, and deviations
in tooth form of the single flank in contact, fis (short
term component).
9.3.6.1 Traditional interpretation
In ANSI/AGMA 2000--A88 and ISO 1328--2, double
flank composite measurements were toleranced for
total composite variation, and tooth--to--tooth com-
posite variation. They were interpreted from a
recorded chart for one revolution of the product gear
as shown in figure 38. The total composite variation
was defined as the difference between the highest to
lowest point on the chart. The tooth--to--tooth
variation was defined as the greatest change in any
360 degree/z part of the chart. This may be
acceptable for evaluation of the final gear quality
relative to the application for some purposes.
However, traditional double flank testing has some
limitations. For example, it cannot detect cumulative
pitch deviation that occurs without radial runout
deviation. Double flank testing is not considered a
reliable method for determining noise potential.
Double flank testing also provides little information
for diagnoses of tooth--to--tooth deviations.
31
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

5.0

4.0

3.0

2.0

1.0
Total
composite
Amplitude

0.0 Unfiltered
tooth--to--
--1.0 tooth

--2.0

--3.0

--4.0

--5.0
0 1 2 3 4 5 6 7 8 9 10 11 12
Tooth number
Figure 38 -- Single flank composite strip chart

The traditional analysis method used for calculation
of tooth--to--tooth composite deviation may give a
distorted indication of the tooth form that the
machine and tool is producing. This is due to the
influence of the long term component on the
tooth--to--tooth composite deviation.
This distortion is best explained by the case of a gear
that had identical tooth form on all teeth being
measured with a perfect master. In this case the
greatest tooth--to--tooth variation will be along the
part of the long term component curve that has the
greatest slope. This has the effect of distorting the
amplitude of the data relating to that particular tooth.
For the same quality of tooth form and runout, the
tooth--to--tooth composite deviation would be great-
er for a gear with a lower number of teeth than it will
for higher numbers of teeth. See figures 39 and 40
for a comparison.
9.3.6.2 Relationship between tolerances
Because of the relationship between the long term
component and the tooth--to--tooth deviation, toler-
ances have had unrealistic values in some cases. In
previously existing standards, the tooth--to--tooth
composite tolerance has been approximately 1/3 to
1/2 of the total composite tolerance. This is partly in
order to accommodate the distortion of tooth--to--
tooth data, by the long term component, especially
for low numbers of teeth.
In the case of a gear with very little or no runout, there
should be a greater difference between total and
tooth--to--tooth composite deviation tolerances. In
this case the tooth--to--tooth composite tolerance
should be 0.1 to 0.2 times the total composite
tolerance. This more appropriate tolerance ratio is
feasible regardless of the amplitude of the long term
component if the tooth--to--tooth composite devi-
ations are separated from the long term component
prior to analysis.
9.3.6.3 Separation of tooth -- to -- tooth composite
observations
The separation of eccentricity effects from observa-
tions of tooth--to--tooth composite deviations can be
done by different techniques. The preferred method
is by use of a digital computer program that is
capable of fitting and extracting a sine wave
according to the given test data. This would result in
charts as shown in figures 41a, 41b, and 41c.

32
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

5.0

4.0

3.0

2.0

1.0
Amplitude

0.0
Unfiltered
--1.0 tooth--to--tooth
(12 tooth gear)
--2.0

--3.0

--4.0

--5.0
0 1 2 3 4 5 6 7 8 9 10 11 12
Tooth number

Figure 39 -- Single flank composite test, low number of teeth

5.0

4.0

3.0

2.0 Unfiltered tooth--to--

tooth (30 tooth gear)
Amplitude

1.0

0.0

--1.0

--2.0

--3.0

--4.0

--5.0
0 5 10 15 20 25 30
Tooth number

Figure 40 -- Single flank composite test, high number of teeth

33
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

5.0

4.0

Total composite deviation

3.0

(Fis -- Single flank)

2.0

1.0
Amplitude

0.0

--1.0

--2.0

--3.0 30 tooth gear

--4.0

--5.0
0 5 10 15 20 25 30
Tooth number

Figure 41a -- Total composite deviation

5.0

4.0

3.0

2.0
Long term component
Amplitude

1.0

0.0

--1.0

--2.0

--3.0 30 tooth gear

--4.0

--5.0
0 5 10 15 20 25 30
Tooth number
Figure 41b -- Long term component

34
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

5.0

Short term component

(fis -- Single flank)
4.0

3.0

2.0
Amplitude

1.0

0.0

--1.0

--2.0

--3.0 Composite tooth--to--tooth

30 tooth gear
--4.0

--5.0
0 5 10 15 20 25 30
Tooth number

Figure 41c -- Short term component

If this method is not available in the measuring
system, an approximation can be done manually.
This involves drawing an upper and lower envelope
of essentially sinusoidal shape enclosing the mea-
sured data. The vertical distance between these
upper and lower envelope lines is the tooth--to--tooth
composite error, fis. This is shown in figure 42.
9.3.6.4 Additional diagnostics
These techniques are focused on evaluation of the
final gear quality relative to the given application.
However, it may be desirable to carryout additional
analysis for diagnostic purposes, such as noise
potential or manufacturing process monitoring. In
such cases more comprehensive data filtering is
appropriate.
Most situations with long term component deviations
will be in the sinusoidal form, which is caused by
eccentricity, as shown in figures 41a, 41b, 41c and
42. There are cases, however, where long term
deviations will show up in higher orders, such as
shown in figures 33 and 34. This can be caused by
oval shapes, triangular shapes, etc. This is common
in ring gears where heat treat distortions occur at the
location of each bolt hole in the blank. Even the short
term component can have distortions from variations
in the tooth shape.
Analysis of composite test data can be enhanced by
the use of analog or digital filters that segregate long
and short term component deviations at a selected
“cutoff” wavelength. Still more comprehensive
analysis of higher order deviations is possible by use
of Fourier analysis techniques, such as a Fast
Fourier Transform (FFT) analyzer.
9.4 Single flank measurement of product gear
pair
The single flank tooth--to--tooth and total composite
deviations involving a mated pair of product gears
are termed “transmission deviations of a gear pair”.
To fully explore the complete spectrum of the
deviations, it is necessary to continue rotation until
the complete meshing period of both gears has been
explored. The number of revolutions required
corresponds to the number of teeth in the larger
member divided by the largest factor common to
both members.

35
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

5.0

4.0

Total composite variation

3.0

(Fis -- Single flank)

Long term component

2.0

(Fp -- Single flank)

Short term component
1.0 (fis -- Single flank)
Amplitude

0.0

--1.0

--2.0

--3.0 12 tooth gear

--4.0

--5.0
0 1 2 3 4 5 6 7 8 9 10 11 12
Tooth number

Figure 42 -- Manual interpretation of composite test

Analysis is similar to that described in 9.3.6 for a
product gear with a master gear, except that the
deviations should be calculated based on the
complete meshing period of both gears rather than
on a single revolution of the product gear.
9.4.1 Identification and location of defects
The measurement of tangential composite devi-
ations facilitates the identification and location of
defects (nicks or burrs) which may degrade the
quality of transmission. For example, as indicated in
the diagram in figure 43, the presence of a defective
tooth can readily be seen. Furthermore, it is
sometimes possible to carry out corrective mea-
sures while still connected to the measuring appara-
tus, in which case the effectiveness of the
adjustments can be verified without delay.
9.4.2 Selective meshing of gears
In some exceptional cases, involving mated pairs of
gears with equal numbers of teeth or other integer
ratios, special steps can be taken to ensure that
optimum performance is realized. Such gears can
be meshed to best advantage by remeshing the
gears with a phase shift of ninety degrees to find the
quadrant in which single flank composite deviations
are smallest. Following this, the process is repeated
by remeshing the gears with phase shifts less than
ninety degrees in order to find the optimum meshing
phase.

representing one pitch

damaged tooth

Figure 43 -- Part of tangential composite deviation diagram -- Interpretation example

36
AMERICAN GEAR MANUFACTURERS ASSOCIATION
In figure 44 diagrams are shown which were
generated from a pair of gears at the different phases
of mesh indicated. It is quite evident that the single
flank composite deviation diagrams for the left flanks
and right flanks are not the same. It may be
necessary to choose an intermediate meshing
position that provides the best compromise solution
if a high degree of transmission accuracy is needed
for both directions of rotation.
10 Contact pattern checking
Contact checking is used for the inspection of mating
gear sets to determine their operational compatibility
and for the inspection of gears which will not fit into
available measuring machines because of size and
weight limits. This clause explains a quasi--static
method of obtaining and analyzing contact patterns,
and a method for evaluating the observed deviations
from designed contact. Contact checking is com-
monly used on bevel, mill, marine, and high speed
gears.
representing one gear revolution
0°L
90°L
180°L
270°L
90° + 1 tooth (L)
Figure 44 -- Tangential composite deviation diagrams showing influence of mesh relocation
AGMA 915--1--A02
10.1 Control of test conditions
The reproducibility of contact pattern checks is
dependent upon careful control of the test condi-
tions. A small variation, 0.01 mm, in location of the
gears from test to test may have a significant effect
on the results.
Caution should be used when static contact checks
are performed on gears and shafts mounted in
dynamic bearings. The shafts should be located in a
fixture in such a way that they represent the final
operating conditions.
10.1.1 Gear axes parallel
When the gears are tested outside the housing in
which they may be used, or if the assembled centers
are adjustable, the gears are normally mounted with
their axes parallel. This is usually accomplished in
an adjustable testing frame with the line of centers
horizontal, so that a precision level and micrometers
can be used to establish parallel axes at the given
center distance in a common plane. The absolute
value of center distance is not as important as
maintaining the gear axes parallel.
0°R
90°R
180°R
270°R
90° + 1 tooth (R)
37
AGMA 915--1--A02 AMERICAN GEAR MANUFACTURERS ASSOCIATION

10.1.2 Test gears 10.4 Interpretation of results

If test gears are being compared to a master gear,
the master gear must be of known quality, and of
sufficient accuracy to assure that errors in the master
gear will not appreciably affect the results.
Gears may also be tested as matched pairs.
10.1.3 Marking compound
Various marking compounds can be used including
Prussian blue, dye check developer, and proprietary
compounds. It is important that the compound be
controlled carefully, since its viscosity and the
method of application will affect the film thickness,
which is critical to the interpretation of results.
Typical values for carefully applied marking com-
pound thickness are from 0.008 mm to 0.012 mm.
Figure 45 shows a contact pattern obtained with
good profile contact, and some tooth alignment
mismatch. If the marking compound thickness is
0.01 mm, the tooth misalignment shown over the
length of the contact pattern is also 0.01 mm. An
angular correction in helix angle or mounting of 0.01
mm divided by the length of contact should produce
full contact.
Length
of contact

10.1.4 Test load

Usually, the test load is very light. In some gear

testing machines, the test load can be varied and
controlled.

10.1.5 Operator training

Since operator skill is an important factor in applica-

tion of the marking compound and control of the test
load, it is important that uniform procedures be Figure 45 -- Matching profiles, with tooth
established and that operators be trained in these alignment mismatch and end relief
procedures, so that reproducible results may be
obtained. The contact pattern shown in figure 46 shows perfect
10.2 Calibration tooth helix alignment with profile mismatch. Using
the same marking compound calibration as the
Calibration of the thickness of the marking com- example above, the profile mismatch is 0.02 mm,
pound is essential to interpretation of contact pattern since contact extends over only one--half of the
test results. Once an operator has developed a profile.
consistent technique, it is possible to establish the
thickness of the marking compound by shifting the
axes of the gears out of parallel in a vertical direction
in the tangential plane by a known angle; i.e.,
shimming one bearing support and observing the
change in the pattern. This calibration should be
performed regularly to be sure that the marking
compound, test load, and operator technique have
not varied.
10.3 Recording results

Contact patterns are usually recorded by photogra-

phy, sketches, or tapes. Instant developing color film
and digital photography are particularly useful for
recording contact patterns. Tapes are made by
carefully applying transparent mending tape (such
as Scotch tape) over the contact pattern, removing
the tape, and applying the tape with the adhering
pattern to white paper.
Figure 46 -- Matching helix, with profile
mismatch and end relief
Figure 47 shows an undulating contact pattern which
might be caused by periodic error in the generating
machine.

38
AMERICAN GEAR MANUFACTURERS ASSOCIATION AGMA 915--1--A02

tions of approximately 75 percent of contact, exclud-

ing extremes of tooth which are intentionally
relieved.

Figure 47 -- Waviness

10.5 Specifications
Contact pattern acceptability is specified by defining
the area in which contact may not occur, the areas in
which contact should occur, and the percentage of
contact required in the desired area. Figure 48
illustrates a contact pattern which meets a specifica-
Figure 48 -- Typical specification:
approximately 75% contact, excluding
extremes of tooth, which are intentionally
relieved

39
 PUBLISHED BY
AMERICAN GEAR MANUFACTURERS ASSOCIATION
1500 KING STREET, ALEXANDRIA, VIRGINIA 22314