AGMA 920-A01 - Materials for Plastic Gears

AGMA 920- A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

AGMA 920- 01

Materials for Plastic Gears

AGMA INFORMATION SHEET (This Information Sheet is NOT an AGMA Standard)

Materials for Plastic Gears American AGMA 920--A01 Gear Manufacturers CAUTION NOTICE: AGMA technical publications are subject to constant improvement, revision or withdrawal as dictated by experience. Any person who refers to any AGMA Association technical publication should be sure that the publication is the latest available from the Association on the subject matter.

[Tables or other self--supporting sections may be quoted or extracted. Credit lines should read: Extracted from AGMA 920--A01, Materials for Plastic Gears, with the permission of the publisher, the American Gear Manufacturers Association, 1500 King Street, Suite 201, Alexandria, Virginia 22314.] Approved October 22, 2000

ABSTRACT The purpose of this document is to aid the gear designer in understanding the unique physical, mechanical and thermal behavior of plastic materials. The use of plastic materials for gear applications has grown considerably due to cost and performance issues. Growing markets include the automotive, business machine, and consumer--related industries. Topics covered include general plastic material behavior, gear operating conditions, plastic gear manufacturing, tests for gear related material properties, and typical plastic gear materials. There are no quantitative details on material properties nor any comparative evaluations of plastic types. Such specific information is left to be provided by material suppliers and gear manufacturers. Published by

American Gear Manufacturers Association 1500 King Street, Suite 201, Alexandria, Virginia 22314 Copyright ã 2001 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--778--9

ii

AMERICAN GEAR MANUFACTURERS ASSOCIATION

AGMA 920--A01

Contents Page

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv 1 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2 General nature of plastic materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 Gear operating conditions and related material properties . . . . . . . . . . . . . . . . . 3 4 Gear manufacturing and related material properties . . . . . . . . . . . . . . . . . . . . . . 8 5 Tests for gear related material properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 6 General description of plastic materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 7 Plastic materials widely used for gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 8 Material selection procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Figures 1a 1b 2 3 4 5 6 7

Representative creep behavior of ductile plastic . . . . . . . . . . . . . . . . . . . . . . . . . 2 Representative creep behavior of non--ductile plastic . . . . . . . . . . . . . . . . . . . . . 2 Effect of strain rate and temperature on stress--strain curves . . . . . . . . . . . . . . 2 Typical fatigue curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Effect of temperature on stress vs. strain for acetal (POM) . . . . . . . . . . . . . . . . 4 Effect of moisture on stress vs. strain for nylon 6--6 (PA 6,6) at 23°C . . . . . . . 5 Polymer impact strength as a function of temperature . . . . . . . . . . . . . . . . . . . . 7 Simple gear with three gates (+) on web, small arrows indicate predicted fiber orientation, grayscale indicates advancing flow from gate location . . . . 11 8 ASTM D638 Type 1 tensile specimen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 9 Typical DMA curves normalized at 23°C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 10 Tensile DMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 11 DMA, amorphous and crystalline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 12 Semi--crystalline polymer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 13 Flexural fatigue specimen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 14a Representations of creep -- strain vs. time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 14b Representations of creep -- creep modulus vs. time . . . . . . . . . . . . . . . . . . . . . 20 14c Representations of creep -- isochronous stress vs. strain . . . . . . . . . . . . . . . . 20 15 ASTM D--3702 thrust washer wear and friction test . . . . . . . . . . . . . . . . . . . . . 21 16 Two dimensional representation of crystalline and amorphous thermoplastics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 17 Modulus behavior vs. temperature of crystalline and amorphous resins, neat and glass fiber reinforced . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

Tables 1 2 3

Additives in plastics for molded gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Plastic materials for molded gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Plastic materials for machined gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

iii

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

Foreword [The foreword, footnotes and annexes, if any, in this document are provided for informational purposes only and are not to be construed as a part of AGMA Information Sheet 920--A01, Materials for Plastic Gears.] Plastic materials differ considerably from metals in performance and processing. Many of the important differences, especially those that are critical to gear applications, are not widely recognized. This is partly because many plastic materials specialists are not familiar with gear requirements. Similarly, many gear specialists are not familiar with plastic material characteristics. Hence the need for reference material which will help bridge these gaps. The AGMA Plastics Gearing Committee has brought together technical representatives from plastic material suppliers, gear manufacturers and designers. This document represents their efforts to further this exchange of information. It will not supply answers to many of the questions that arise in the application of plastic materials to gears, but it should encourage inquiry and information exchange. One issue that requires special attention is the availability of plastic material properties in the form most suitable for plastic gear design. This includes properties that are counterparts of those used in the design of metal gears, and those that are special to plastic materials in these applications. To a very large extent, plastic gear designers have access only to property data taken from ASTM tests as reported by material suppliers even though such tests were created to meet other objectives. It was therefore judged essential to include brief descriptions of these tests supplemented by comments on any limitations of such test data when applied to gears. Various industry initiatives are now underway to develop gear specific property data, which will in time supplement the information provided here. The first draft of AGMA 920--A01 was made in 1993. It was approved by the AGMA membership in October, 2000, and approved for publication by the Technical Division Executive Committee on October 22, 2000. 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.

iv

AMERICAN GEAR MANUFACTURERS ASSOCIATION

AGMA 920--A01

PERSONNEL of the AGMA Plastics Gearing Committee Chairman: Clifford M. Denny . . . . . . . . . . . . . . . . . . . . . Consultant Vice Chairman: Edward H. Williams, III . . . . . . . . . . . . LNP Engineering Plastics, Inc.

ACTIVE MEMBERS M.A. Bennick . . . R.M. Casavant . . D.S. Ellis . . . . . . . T. Grula . . . . . . . . J.W. Kelley . . . . . R.R. Kuhr . . . . . . I. Laskin . . . . . . . . D. Michael . . . . . . A. Milano . . . . . . . S.D. Pierson . . . .

RTP Company GW Plastics, Inc. ABA--PGT, Inc. DuPont Company Shell Development Company Enplas, Inc. Consultant CEI Seitz Corporation ABA--PGT, Inc.

M. Schireson . . . D. Sheridan . . . . . L. Siders . . . . . . . Z.P. Smith . . . . . . P.A. Spaziani . . . M. Thompson . . . A.B. Ulrich . . . . . . J.H. Winzeler . . . P. Wyluda . . . . . .

DSM Engineering Plastics, Inc. Ticona Lexmark International, Inc. Ticona Seitz Corporation ABA--PGT, Inc. UFE, Incorporated Winzeler Gear Ticona

G. Martello . . . . . H.S. Oh . . . . . . . . M. Oliveto . . . . . . A.J. Padden . . . . K. Price . . . . . . . . J. Rees . . . . . . . . C. Reese . . . . . . . E. Reiter . . . . . . . J.T. Rill . . . . . . . . . J. Seitz . . . . . . . . . L.J. Smith . . . . . . R.E. Smith . . . . . . P.A Tuschak . . . . B. Ulissi . . . . . . . . G.J. Verros . . . . . M. Wilkinson . . . .

BF Goodrich Siebe Environmental Controls DSM Engineering Plastics, Inc. SPM Minneapolis Eastman Kodak Company ATS Precision Component Div. SPM Minneapolis ATS Precision Component Div. Black & Decker, Inc. Seitz Corporation Consultant R.E. Smith & Co., Inc. E.I. DuPont deNemours & Co. DuPont Performance Lubricants Consultant GW Plastics, Inc.

ASSOCIATE MEMBERS M.K. Anwar . . . . . M. Aube . . . . . . . . D.E. Bailey . . . . . J. Barger . . . . . . . T. Barry . . . . . . . . D. Blakley . . . . . . M. Bogle . . . . . . . B. Butsch . . . . . . . D. Castor . . . . . . . P. Davoli . . . . . . . E. Dornan . . . . . . G.C. Hesser . . . . A.H. LaFord . . . . J. Lay . . . . . . . . . . R.B. Lewis . . . . . . A. Luscher . . . . . . T. Mardis . . . . . . .

Globe Motors GW Plastics, Inc. Rochester Gear, Inc. D.I.G.I.T., Inc. Phillips--Moldex Company Axxicon Components Poly Hi Solidur Co. Lexmark International, Inc. Eastman Kodak Company Politecnico Di Milano Winzeler Gear DuPont Company Static Control Components, Inc. NYE Lubricants, Inc. Lewis Research, Inc. Ohio State University CEI

v

AMERICAN GEAR MANUFACTURERS ASSOCIATION

American Gear Manufacturers Association --

Materials for Plastic Gears

1 Scope This information sheet provides descriptions of plastic materials commonly used in gearing. It relates the general properties of these materials to typical operating conditions of gears. Properties that relate to the manufacturing processes of machining and molding are discussed, including the property of shrink rate in molding. It also describes the types of tests that are customarily used to obtain published values of these properties. It is intended that this information sheet serve only as an introductory guide to the designer of plastic gears when it comes to selecting candidate materials. The designer is advised to look to material suppliers and plastic gear manufacturers for their expert guidance on selecting materials for specific applications. It is also important to recognize that thorough application testing is often needed to confirm the suitability of a material choice. Only a limited number of plastic materials are mentioned here as commonly used for gears. Gears have been made from other plastics as well, but generally because some special material property or commercial consideration was judged essential to a particular application. It is also possible that the suitability of other materials for gears has not yet been recognized. Furthermore, new plastic materials are continually being developed and some, no doubt, will prove themselves as important additions to those discussed in this information sheet.

AGMA 920--A01

2 General nature of plastic materials Although plastic materials are successfully used in place of metals in load carrying applications such as gears, there are important differences between the two types of materials. These differences generally appear in combination and can have a significant effect on plastic gear performance. 2.1 Elastic and viscoelastic behavior Most structural metals behave as essentially elastic materials. Plastics, on the other hand, behave as a combination of elastic and viscous materials, with the balance varying considerably with the type of plastic, its molecular structure, and the type, quantity, and orientation of any additives. This special nature of plastic materials does not interfere with their use in a very wide range of applications which benefit from their many other special properties. It does, however, require a thorough understanding of reported material properties data and their relationship to the specific application. 2.2 Response to load When load is applied to elastic materials such as steel, the resulting deformation is essentially immediate, constant over time, independent of a wide range of temperature, and fully recoverable when the load is removed. When the material has a viscous component combined with the elastic, the initial deformation will increase with time under load (creep deformation) and will depend to a considerable degree on temperature. When the load is removed, there will be some delayed recovery and, possibly some permanent deformation. The time dependent deformation of ductile polymers under constant load is quantified in creep testing. A family of curves resulting from varying the constant load (stress) and recording the increasing creep strain is shown in figure 1a. As the polymer is held under constant stress (load) over time, the creep strain initially increases at a rapid rate (primary creep) and then plateaus to a significantly lower creep strain rate (secondary creep). For nonductile polymers the material will experience creep rupture while deforming under secondary creep (see figure 1b). However, for ductile polymers, the material will experience another increase in creep strain rate

1

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

(tertiary creep) and will creep rupture in tertiary creep. For non--ductile polymers the locus of creep rupture points forms the creep rupture envelope. However, the creep rupture envelope for ductile polymers is the locus of points resulting from the transition from secondary to tertiary creep.

SRupture

deformation and creep rupture of polymers needs to be considered. 2.3 Effect of rate of load application Because of the time dependant nature of viscoelastic plastic materials, the strength properties and elasticity modulus are typically greater when the load is applied and removed more rapidly. See figure 2. This characteristic is especially important in gear applications.

Creep rupture envelope

Increasing strain rate

Increasing stress

Stress

Strain

Increasing temperature

0 0

Time

Figure 1a -- Representative creep behavior of ductile plastic

Strain Figure 2 -- Effect of strain rate and temperature on stress--strain curves 2.4 Effect of temperature 2.4.1 Strength and deformation

SRupture

Strain

Creep rupture envelope

Increasing stress

0 0

Time

Figure 1b -- Representative creep behavior of non--ductile plastic Creep deformation appears not to be a factor in gears under continuous operation because the load is applied to gear teeth only for a short time duration. However, for gears run into stalled conditions creep 2

Because a higher temperature reduces the resistance to movement of the polymer chains, the material at high temperatures can be viewed as less viscous (decrease of the viscous component). This decrease in the viscous component of polymers at higher temperatures causes the strength and stiffness properties to decrease with increasing temperatures (see figure 2). Temperature increases of the polymer at critical locations in gears could result from friction, hysteresis, or both in combination. This temperature rise of the gear material at critical locations could, therefore, reduce the load resisting capability of the gear. This condition is a significant factor to consider in gear performance. 2.4.2 Expansion Plastics have higher thermal expansion rates than metals. These high rates can be partially offset by compounding the plastics with various fillers and reinforcements. Thermal expansion must be considered in those applications in which the gears

AMERICAN GEAR MANUFACTURERS ASSOCIATION

2.4.3 Heat aging If a plastic material is subjected to an elevated temperature for an extended period of time, its properties at the end of the period may be degraded from those before the high temperature exposure. 2.5 Effect of moisture A change in moisture content can act in a manner similar to a change in temperature in its effect on strength, deformation and expansion. Materials vary considerably in their moisture absorption, making this influence more significant in some materials than in others.

3 Gear operating conditions and related material properties In order to evaluate a material for a specific gear application, the operating conditions must be recognized along with the related properties of the material. Some of these properties are much more significant in gears made from plastics than in gears made from metals, and require special attention.

apply

in

nearly

103

104

105 No. of cycles

106

107

Figure 3 -- Typical fatigue curve

3.1 General operating conditions These conditions applications.

takes place during a rolling action combined with sliding. In certain types of gear sets (spur, helical and bevel), the relative sliding during each tooth’s engagement cycle varies in magnitude and typically reverses in direction. In other types (worm, crossed helical and hypoid), the relative sliding is more nearly constant. Since many plastic gear applications do not employ the type of lubrication that keeps the contacting tooth surfaces separated by a fluid film, the sliding action often results in significant friction. However, this friction between contacting plastic surfaces is often less than that experienced with many metals similarly employed under non-lubricated conditions.

Increasing stress *

operate over broad ranges of temperature and the structure that controls gear center distance is made from a material of a significantly different expansion rate. See 3.1.3.1.

AGMA 920--A01

all

gear

3.1.1 Repeated load Gear teeth experience repeated loading during successive engagements. Under continuous rotation, the load on an individual tooth is applied and released rapidly. There is also some delay between load cycles while the single tooth is rotating towards its next engagement. The property of the material that resists fatigue failure breakage under this type of load is approximated by the flexural (bending) fatigue limit, see figure 3. Standard tests that report this property, and the limitations of those tests, are described in 5.1. 3.1.2 Rolling and sliding under load Load is transmitted between the curved surfaces of engaging teeth through contact over a relatively small area. This contact between the tooth surfaces

(* NOTE: Linear scale used for stress axis.)

3.1.2.1 Failure due to pitting The repeated contact force on the gear tooth creates repeated shear stresses just below the tooth surface. Under certain conditions, these stresses can cause failure of the gear through the formation of local subsurface cracks which progress into pitting. Normally, when a plastic gear surface is subjected to high loads for a large number of cycles, failure takes place first in the form of excessive wear. See 3.1.2.3. However, if the gear is well lubricated and wear is minimal, failure by pitting may appear. 3.1.2.2 Friction forces and power loss The friction from the relative motion between the contact surfaces has other effects on the operation of gears. The static coefficient of friction between the two contacting materials will influence the starting load in a gear train. The dynamic coefficient of friction will generally have several effects. It

3

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

determines the power loss at that part of the gear train and the overall train efficiency. It can also influence the rise in temperature of the gear teeth and the corresponding loss in strength and stiffness properties. The rise in temperature may also lead to softening of the surface and change in friction properties. Standard tests for static and dynamic coefficients of friction are described in 5.2.

tendency is the percent of water absorption in the type of test described in 5.4.1.

3.1.2.3 Wear

3.1.4 Mechanical property changes under operating conditions

The interaction between the loaded surfaces will often lead to wear. If the wear progresses far enough to modify the gear profiles, it may produce excessive vibration and noise. Further wear may progress to the point that insufficient material remains to support the load. The wear characteristics of the material combination are often expressed by a rate of material removal. Common tests for this characteristic, and their limitations, are described in 5.2. 3.1.3 Dimensional changes under operating conditions Plastic gears commonly operate under conditions that will cause significant dimensional changes in some plastic materials.

3.1.3.3 Chemical conditions An incompatible chemical acting on a plastic material during product manufacture, storage or operation generally leads to an increase in size. An improperly selected lubricant might produce such a result.

3.1.4.1 Thermal conditions As noted in 2.4, strength and modulus properties decrease with increased temperature. Although these properties improve with a decreased temperature, these are typically accompanied by increased brittleness or a reduction in impact strength. See figure 4. These effects can vary considerably with the plastic selected and its additives. 3.1.4.2 Humidity conditions The property changes due to humidity are qualitatively similar to those of temperature. See figure 5. Here also, the degree of change can vary considerably with material selection.

3.1.3.1 Thermal conditions

--40°C 23°C

66°C

Stress

If a gear is operated at a temperature much different than the temperature at which its dimensions were originally specified and measured, its size will be different. Unless there is a compensating change in center distance, the size change will alter the operating backlash and root clearance. If the two mating gears change in a disproportionate manner, the two tooth profiles can become mismatched due to differences in base pitch and axial pitch, resulting in an increase in vibration, noise and dynamic loads. The material property which directly relates to the size change is the thermal coefficient of expansion. The standard tests for these properties are described in 5.3.

85°C 100°C

3.1.3.2 Humidity changes Plastic materials change in size with the level of moisture they contain, giving results similar to thermal size changes. This moisture level is determined by the relative humidity to which the material has been exposed over an extended period of time and to the tendency of the material to absorb moisture. The plastic property used to indicate this 4

0

1

Strain, %

2

3

Figure 4 -- Effect of temperature on stress vs. strain for acetal (POM)

AMERICAN GEAR MANUFACTURERS ASSOCIATION

AGMA 920--A01

Tensile stress

Dry as molded

50% RH

0

0.5

1

1.5 Strain, %

2

2.5

3

Figure 5 -- Effect of moisture on stress vs. strain for nylon 6--6 (PA 6,6) at 23°C 3.1.4.3 Chemical environment Plastics, in contrast to most metals, are generally resistant to a broad variety of chemicals. However, individual plastics may be vulnerable to particular chemicals. The effect of chemical action is generally a reduction in fatigue and impact strength and change in other mechanical properties. See 3.2.6.2. 3.2 Special operating conditions These conditions are encountered less commonly in gear applications, but, when they do, they direct attention to other properties of the plastic material. 3.2.1 Impact loading In some gear applications, the gears may be subject to a suddenly applied load which requires the material to absorb considerable energy associated with the load. This energy tends to be absorbed around design features that, because of their slender shape and reduced size, are most compliant and develop the highest stresses. The contacting gear teeth are commonly most vulnerable. Such loads may appear with sudden starts of the gear train driver, a sudden change or reversal in the driven load, or with sudden braking. The energy to be absorbed is even greater if there has been travel

across a substantial backlash gap before the load impacts the gear teeth. Impact loading tends to be smaller in a gear drive made with plastic gears than in a drive of similar size made with metal gears. The difference can be attributed to the lower inertia and greater compliance of most plastics over most metals. Nevertheless, this type of loading can be severe enough to cause failure in plastic gears. There are significant differences in how otherwise suitable plastic materials respond. This response is indicated, at least to a comparative degree, by some standard tests described in 5.1.4 and 5.1.5. 3.2.2 Short--term overloads In addition to the normal repeated loads encountered in gear applications, there are sometimes higher loads that appear occasionally for brief intervals, or with a much smaller frequency, on any individual tooth. Failure due to this type of load often takes the form of excessive deformation of the gear tooth or, in the case of brittle materials, fracture of the tooth. The material property that needs to be considered in evaluating the risk of failure by excessive deformation is the stress--strain curve. The standard test,

5

AGMA 920--A01

along with its limitations, that measures this set of properties is described in 5.1. 3.2.3 Long--term loads Loads that remain applied to an individual tooth for an extended period of time, whether greater or smaller than the normal repeated load, can also cause failure by permanent deformation. See figures 1a and 1b. While this type of failure is not a factor in most structural metals except at very high temperatures, it can appear in many plastics at typical operating temperatures. The deformation that continues to increase without an increase in load is called creep. The relative degree of creep in a plastic material is reported in a variety of data. Each shows how some characteristic changes over extended time. Plots of strain versus time, for various stress levels, most closely indicate the relative effect of creep in gears. Description and evaluation of the test to collect this data is given in 5.1.8. 3.2.4 High temperature Gears are sometimes required to operate at temperatures well above 23°C, the temperature at which their properties are typically measured. The higher temperature is generally the ambient temperature of the application, but may, in whole or in part, result from heating due to friction, hysteresis or both. 3.2.4.1 Various strength properties Just about all of the material properties noted above change with temperature. Often the best indicator of gear material performance at an elevated temperature is the stress--strain curves measured at or close to the operating temperature (see figure 4). The tests used to report material properties at the standard temperature can also be applied at the higher temperature, as noted in clause 5. 3.2.4.2 (DTUL)

Deflection temperature under load

This property, previously known as heat distortion temperature (HDT), is a measure of the temperature at which the flexural modulus falls below a predetermined value associated with a particular stress. It serves mostly as a relative indicator of material serviceability as a gear operating at elevated temperatures. The test is described in 5.3.3. 6

AMERICAN GEAR MANUFACTURERS ASSOCIATION

3.2.4.3 UL temperature index This is a commercially used indicator of the effect of heat aging described in 2.4.3. It suggests the highest temperature, above 50°C, at which the plastic material is to be used. The temperature index is the temperature at which the specific property will decrease to one--half its original value after exposure for a long time at that temperature. There are separate ratings for mechanical properties with impact and without impact. Index values can be found in the UL “Yellow Card”. The test is described in 5.3.4. 3.2.4.4 Coefficient of thermal conductivity Where high temperature results substantially from heat generated by friction or hysteresis, the coefficient of thermal conductivity becomes an important property. A higher coefficient indicates that such heat will be more rapidly conducted away and the teeth will see less of a temperature increase. See 5.3.2. 3.2.4.5 Coefficient of thermal expansion As described in 3.1.3.1, this property plays a role in the design of gears which are to operate over a wide range of temperatures. Very often, the design process can accommodate the expansion rate of a material selected on the basis of its other essential properties. In special applications, this property can become a controlling factor in material selection. Large differences in expansion coefficients between the housing and the gears can introduce excessive variations in backlash and in depth of engagement. Large differences between the two gears can cause excessive mismatch of gear pitch, enough to generate high dynamic loads, vibration and noise. See 5.3.1. 3.2.4.6 Electrical conductivity Plastic materials are generally considered to be electrically insulating. However, when gears are required to conduct static electricity to some electrical ground, it becomes necessary to use a plastic which has been modified to supply the conductivity. The static electricity may be generated elsewhere in the product or in the gears themselves. The body of the gear, or some feature molded integrally with the gear, may sometimes be used to conduct system electrical current. This will also require conductivity in the plastic. It is not recommended that high system electrical current be transmitted through the contacting tooth surfaces because of the risk of damage to these surfaces and risk of being a source of ignition.

AMERICAN GEAR MANUFACTURERS ASSOCIATION

AGMA 920--A01

3.2.5 Low temperature

3.2.6.2 Chemical conditions

Operation at temperatures much below the usual material testing temperature also has an effect on gear material performance. As the temperature decreases, strength properties usually improve, especially those related to stress--strain testing and creep testing. This is true also for impact resistance as seen in figure 6. However, as the temperature decreases, a region is encountered where impact resistance severely decreases, called the ductile-brittle transition temperature (DBTT). Therefore, when choosing a polymer for gears required to perform under sudden changes in load at low temperatures, the DBTT of each considered polymer should be evaluated and compared.

Although most plastics are more resistant than most metals to many kinds of chemical environments, there are important exceptions in combinations of material and chemically active substances. Exposure of the gears to these chemicals may come from the outside environment, from process material in the gear driven equipment, and, in some cases, even from the lubricant applied to the gears. Description of tests for chemical resistance is beyond the scope of this document.

3.2.6 Other environmental conditions There are environmental conditions other than temperature extremes which can also affect material performance. 3.2.6.1 High humidity

Falling weight impact strength

The water absorbed by extended exposure to a high humidity environment not only affects dimensions, as noted in 3.1.3.2, but also affects strength properties. Testing for these properties under high humidity is the same except that material test specimens are first conditioned to the desired moisture content.

3.3 Vibration and noise When the application is specially concerned with limiting vibration and noise, and design factors such as adequate backlash have been provided, material selection may be influenced by an additional set of properties. 3.3.1 Modulus of elasticity One approach to reducing vibration and noise is to introduce greater compliance into the gears without introducing resonance or excessive loss of load capacity or wear properties. Gear teeth that deflect more readily can reduce the dynamic excitation that originates in imperfect tooth geometry. One of the properties that indicate the relative compliance of the material is the modulus of elasticity. This property is generally determined as part of yield strength testing, as described in 5.1.1, 5.1.2 and 5.1.3.

ductile--brittle transition temperature (DBTT) Decreasing temperature Figure 6 -- Polymer impact strength as a function of temperature

7

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

3.3.2 Hardness durometer

4.1.1 Stock material

When selection of material for increased compliance leads to materials of a rubbery character, the hardness durometer becomes the preferred indicator. Testing for this property is described in 5.5.1 and 5.5.2.

The machining of gears starts with the stock material and its fabricated form. Not every material, especially a material with desired additives, is commercially available in a form which suits the machining process or which results in a gear with the desired performance.

3.3.3 Internal damping Another approach to limiting vibration and noise in a gear train is to increase the degree of internal damping in the gear materials. This serves as a means of absorbing dynamic energy. See 5.1.3.

Sometimes a molded blank is used in place of stock material. Machining gear teeth into a molded blank will give satisfactory results only if the material and its processing have been carefully selected to avoid the difficulties described below. 4.1.1.1 Skin--core effects

4 Gear manufacturing and related material properties When materials are selected for a gear application, the concern for performance must be coordinated with manufacturing considerations. These manufacturing considerations may rule out materials not suited, economically or otherwise, to the planned manufacturing process. Certain manufacturing related characteristics or measurable properties of the material may determine its suitability. CAUTION: Safety is an important consideration in the manufacture of gears from plastic materials. Information on safe handling and processing is available in the Material Safety Data Sheet provided by the material supplier. Also, see 4.1.2.6 and 4.2.4.3.

4.1 Manufacture by machining Machining may be selected over molding as the plastic gear manufacturing process for several reasons: -- the quantities may be too small to justify the tooling cost for molding; -- the required accuracy or some special design feature (such as very thick section) may be too difficult for molding; -- the desired plastic material may not be suited to precision molding. Machining may also be selected as a means for obtaining sample gears for testing before the design is approved for mold tooling. Such samples will be useful as long as consideration is given to the potential differences between a machined and molded gear. 8

Some of the properties of the material in stock form can vary considerably between the outer surface material, or skin, which has formed through more rapid cooling, than the inner material, or core. Machining invariably leaves the gear teeth made from the core material, which generally has different strength, wear resistance, and chemical resistance. In addition, machining may expose voids in the stock. 4.1.1.2 Reinforcements in extruded stock Non--reinforced extruded stock is widely used for machining gears. However, reinforced stock is not as widely used for machined gears. If the application requires the greater strength that comes from the addition of reinforcements, such stock may not be suitable. While it may be possible to extrude the plastic with reinforcements included, the resulting direction of the reinforcement fibers will generally be random and not in the radial direction required for the reinforcement to contribute to the bending strength of the gear tooth. Further disadvantages of machining most reinforced stock relate to the potential accelerated tool wear due to the abrasive qualities of some reinforcements and to the hazards associated with the fine reinforcement particles produced. 4.1.1.3 Reinforcement in laminated stock Gears are also machined from blanks cut from laminated sheets or plates. The reinforcing layers in the laminated material are often in the form of a woven fabric. If the weave of this fabric is too coarse in comparison with the size of the machined gear tooth, the full beneficial effect on gear tooth strength will be lost. The fabric may serve the additional purpose of retaining lubricant used in the application.

AMERICAN GEAR MANUFACTURERS ASSOCIATION

If the fabric is made from an abrasive or hazardous material, the disadvantages noted in 4.1.1.2 will apply. 4.1.2 Machinability Machining of plastic gears is performed by most of the same processes used in the machining of metal gears. The economics of the machining process and the level of accuracy attainable may depend on a variety of material properties. 4.1.2.1 Cutting rate The cutting rate is often limited by thermal considerations. This is especially true for a combination of large energy release in machining, poor thermal conductivity, and low resistance to heat before melting or otherwise deteriorating. The factor of a high coefficient of thermal expansion may also limit the cutting rate. Greater expansion due to heating tends to reduce the accuracy of the machined gear. If a cutting lubricant is to be used, it must be carefully selected to avoid averse effects on the plastic. Such effects may not be readily apparent and the best way to avoid them is by consultation with the material supplier. 4.1.2.2 Tool wear The rate of tool wear and the frequency of resharpening can also affect the economics of the machining process and the accuracy when a large quantity of gears is involved. Some additives to the material may increase the rate of tool wear while others may reduce it. Maintaining tool sharpness is essential in the machining of plastics. A relatively low modulus material will deflect under a dull tool. This may make light, finishing cuts difficult if not impossible. Such deflection also interferes with machining accuracy. Furthermore, dull tools generate added heat with its accompanying problems. 4.1.2.3 Distortion The selection of materials and how they have been processed into their stock form may influence the distortion in a machined gear. Such distortion results from the release of internal stresses. These stresses may have been in the material before machining, to be released when part of the material was removed, or they may have been introduced by the heat and forces of the machining. The distortion may be evident immediately after the gear has been ma-

AGMA 920--A01

chined and cooled, or it may not appear to its full extent until many hours later. Annealing the material, either before or after machining, depending on the cause, often reduces the distortion without adding much to the processing cost. Another source of distortion is excessive chucking or clamping pressure. The plastic material may deflect in the area to be machined with resulting distortion after the pressure is released. Similarly, the surface being machined may deflect under the cutting forces being applied by the tool. It may be necessary to allow for the surface springback after passage of the tool in order to achieve the desired dimension. 4.1.2.4 Burrs Plastic materials vary in the extent to which burrs are formed during the gear machining process. Materials which result in particularly tenacious burrs, even when the tool is properly maintained, should be avoided unless there is provision in the machining set--up to minimize or prevent burrs. The possibility of burrs arises where the cutting tool is removing unsupported material. This might be at the end of its cut across the gear blank or, for a non--topping cutter, at the tooth tips. In each case, a thin layer of unsupported material may either break free without producing a burr or, for some plastics, simply bend away into the open space and remain firmly attached to the gear tooth in the form of a burr. Such burrs cannot always be removed by conventional deburring operations such as filing or wire-brushing. Test machining of sample material may be the best way to determine whether burr formation will be a serious problem. Annealing before machining may help and should be included in the testing where it is practical to add that process. A change in cutting rate may also reduce burr formation, but if it takes a major drop in the production rate to solve the burr problem, the material may be disqualified for economic reasons. 4.1.2.5 Finish The finish of machined tooth surfaces may vary with the nature of the plastic material. The presence of some additives, acting as internal lubricants, may help the machined finish, unless they are improperly dispersed. Fibrous materials, such as glass or carbon fiber, or hard granular materials, when used as additives in molded gear blanks, generally

9

AGMA 920--A01

interfere with a good finish. As an example, some kinds of filler particles may be removed from the surface during machining to leave openings or surface roughness. 4.1.2.6 Sharp corners and machining grooves Sharp corners are to be avoided as potential sources of stress concentration and cracking. Where corner shape is transferred directly from the tool shape, the tool tip radius should be as generous as possible. Similarly, sharp grooves from a pointed tool should be avoided, especially on surfaces like gear tooth fillets that will be highly stressed in operation. See ANSI/AGMA 1006--A97, Tooth Proportions for Plastic Gears. 4.1.2.7 Safety concerns in machining Machining may introduce safety concerns in addition to those that generally apply to the selected material. Excessive rates of material removal may overheat the plastic, releasing unsafe chemical products. Flammability may also be an issue. A machining process which creates fine particles or other unsafe products must be confined to prevent inhalation by operators. An otherwise preferred material may require safety precautions not readily available. In that case, it will have to give way to another choice which will not require such precautions. 4.2 Injection molding process To successfully injection mold a thermoplastic gear, one needs to ensure that all components of the injection molding process are understood. These components are: --

gear design;

-- mold design, including gate and runner location and size, cooling line layout; -- the molding process is one of heat removal; -- cooling is critical to successfully process to correct dimensions; --

material selection and processing thereof;

-- injection molding machine and condition, including auxiliary equipment such as dryers and fixtures; --

gear inspection capabilities.

CAUTION: For informed guidance on how the molding and related part design factors influence material selection, consultation with a molder experienced in gear molding is advised.

10

AMERICAN GEAR MANUFACTURERS ASSOCIATION

4.2.1 Moldability This characteristic of a material relates to how well it will fill the mold cavity without, at the same time, flashing into the very fine gaps at the edges of the cavity. It is primarily determined by the viscosity of the molten plastic during mold filling. The viscosity can be altered by changes in the molding process, for example mold and melt temperatures, but sometimes only at the expense of the quality of the molded part. Quality defects from improper process changes may take the form of voids, sinks, warpage, and internal stresses. Plastic materials can be especially sensitive to overheating. Outgassing can develop from the chemistry of additives such as toughening agents, flame retardants, and some internal lubricants. 4.2.2 Shrink rate (shrinkage) The degree and uniformity of shrinkage of the plastic material is also a factor in its suitability for a molded gear application. The shrinkage is determined first by the molecular structure of the plastic and other additives. In the molding process, it will also be influenced by cross sectional area, cooling rate, fiber orientation, molding temperatures and pressures, and other processing variables. The predictability and consistency of the shrinkage is generally more significant in producing accurate gears than the magnitude of the shrinkage (or shrink rate). Shrinkage in gears is not always uniform, as in photographic size reduction. Molding process and part design, along with material shrinkage properties, may contribute to this non--uniformity. When mold gear cavity design inappropriately assumes that the shrinkage will be purely uniform, the quality of the molded gear will suffer. The directional non--uniformity of the shrink rate is also an important factor as discussed below. 4.2.2.1 Dimensional non--uniformity in gear diameter Directional non--uniformity in shrinkage can be a major factor contributing to eccentric and out--of-round gears, especially in plastics with high aspect ratio fiber reinforcement. In such materials, differences in both the relative amount and direction of fiber orientation will lead to different amounts of shrinkage in the plastics, both radially and axially. The shrinkage will be reduced in the direction of fiber orientation. The effect of such directional properties can sometimes be offset by techniques of mold

AMERICAN GEAR MANUFACTURERS ASSOCIATION

design such as location, size and number of gates. For example, when using multiple gates the radial dimension of the gear will be greatest in the area between the gates due to the formation of weld (knit) lines, which orient the fibers radially, thus reducing the amount of radial shrinkage. This will result in a high spot on the gear. At the area near the gate, the fibers are more randomly oriented, and shrinkage will actually be greater. See figure 7. In a simple gear using a central diaphragm gate, the fibers are all oriented radially, and the shrinkage will be the same

AGMA 920--A01

in all directions. However, for some materials, these directional effects can be so great that the materials must be ruled out if a high accuracy gear is required. 4.2.3 Other directional properties Flow induced orientation can also affect the mechanical, electrical, tribological and thermal properties of a molded gear. This non--uniformity is generally tied to flow direction in the filling of the mold cavity and is most marked in materials with fiber reinforcement.

2

1

3

4

5

shrinkage and the gear will have a high spot in these Key: areas. 1 Weld lines. 4 Gate, (+) on web. 2 Fibers near gate are randomly oriented and 5 Small arrows indicate predicted fiber orientation, shrinkage is anisotropic (non--directional). grayscale indicates advancing flow from gate loca3 Fibers at weld lines, where flow fronts meet, are tion. oriented radially (isotropically). The fibers resist Figure 7 -- Simple gear with three gates

11

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

5 Tests for gear related material properties The various material properties that may affect plastic gear performance, as described in clause 3, are measured in standard commercial tests. These tests are very briefly described below. Further details may be found in the referenced ASTM standards or other industry standards. There are several cautions to be considered in relating these tests and the data they produce to how the material will perform in a gear application. The tests are performed on specimens of standard size and shape, selected to suit the test equipment and generally not representative of the loaded features of a gear. Although the specimen may be manufactured by molding, the standard molding conditions used for specimens are not necessarily representative of the conditions used for molded gears. Correlation of test results from any of the described tests with the actual performance of plastics is dependent upon the similarity between the testing and the actual use conditions. 5.1 Strength properties 5.1.1 Standard test method for tensile properties of plastics -- ASTM D638 This test method covers the determination of the tensile properties of unreinforced and reinforced plastic materials in the form of a standard dumbbell-shaped test specimen (see figure 8) when tested under defined conditions of pretreatment (sample preparation), temperature, humidity, and testing machine speed. Materials suppliers generally list the data on material data sheets using the terms tensile strength, tensile modulus and tensile elongation. Tensile strength is the ultimate strength of the material, either at the

13 mm

yield point (tensile strength at yield) or, if the material does not yield, tensile strength at break. The percent elongation should be reported the same way. Material data sheets do not always indicate which type of strength is being reported. Tensile modulus is the modulus of elasticity calculated taking the slope of the line formed by extending the initial linear portion of the load--extension curve. 5.1.1.1 Significance of test It is important to note that tensile strength is reported as a specific value generated from a specific test. Depending on the material and the particulars of the test, this may be a best case value. In any evaluation of materials, it will be important to look at the tensile strength stress vs. strain curves (if available) at different temperatures to understand completely the material’s behavior. 5.1.1.2 Limitations for gear applications The differences between test conditions and typical gear operating conditions should be considered before applying the reported test data in gear design. Two of these differences are load related: -- The purely tensile load used in the test produces essentially uniform stress across the critical section of the specimen. In a gear, however, the load is a bending load which produces a non--uniform stress across the critical section at the tooth fillet area, with the maximum stress at the fillet surface. If the surface material in a molded plastic gear has different properties than the core material, its relative contribution to tooth bending strength may differ from than that suggested by the reported test data. -- The test load is applied at a relatively low rate, while in a gear the load is generally applied at a much higher rate. Since the properties of plastic materials vary significantly with the load rate, the reported test data may not properly represent the material strength in the gear.

Wc

19 mm

50 mm 76 mm

57 mm 115 mm Figure 8 -- ASTM D638 Type 1 tensile specimen

12

7 mm or less

AMERICAN GEAR MANUFACTURERS ASSOCIATION

Other differences between test specimen and gear may be in the material itself, resulting from differences in relative size at the critical sections and especially from differences in processing. 5.1.2 Standard test methods for flexural properties of plastics -- ASTM D790 These test methods cover the determination of flexural properties of unreinforced and reinforced plastics. They are generally applicable to rigid and semi--rigid materials. Flexural strength for those materials which do not break or do not fail in the outer fibers is reported at 5% strain. Specimens are in the form of rectangular bars which are to be loaded as simply supported beams. 5.1.2.1 Significance of test In this test the stress--state varies across the cross--section from tension to compression. This varying stress--state may give different strength properties than those from a uniform stress tensile test. In fact, flexural strength properties are often reported as higher than tensile strength properties and with greater variation. Because of the higher and more variable flexural strength values, a more conservative tooth design would result from using the tensile strength, even though gear teeth experience flexural deformation. 5.1.2.2 Limitations for gear applications The proportions of the slender flexural test specimens are quite different from those of gear teeth, as is also the simple beam support different from the cantilevered support in gear teeth. Other differences between test conditions and gear operation are noted in 5.1.1.2 for tensile test data. 5.1.3 Dynamic mechanical properties of plastics -- ASTM D4065 There are seven ASTM standards for Dynamic Mechanical Analysis, DMA, of materials in the solid state. They are: D4092 on terminology, D5023 on the three point bending method, D5418 on the dual cantilever beam method, D5024 on the compression method, D5026 on the tension method, and D5279 on the torsion method. ASTM D4065 on determining and reporting dynamic mechanical properties covers all of these methods. These tests are used to determine the elastic and viscous response of plastics to a small harmonic excitation over a wide temperature range. See bibliography for complete references.

AGMA 920--A01

5.1.3.1 Significance of test DMA tests have been extensively used by polymer physicists to understand the internal structure of plastics. The elastic and loss modulii and the tan(δ) as a function of temperature, frequency, and time (if desired) may be determined. Transition temperatures for the material can be determined. The method can also be used to evaluate the effect of processing, conditioning and chemical exposure on materials. It can be used to show the effect of phase separation of multicomponent systems and effect of type, amount, dispersion, and orientation of fillers. (For some materials, the thermal behavior is clearer if the curves are plotted both semilog and linear.) 5.1.3.2 Limitations for gear applications Just as the value of modulus from tensile test can vary considerably depending on test specimen and test conditions used, the values obtained by DMA test methods differ depending on which test method and apparatus is used. However, the storage modulus gives a clear indication of the behavior of the elastic modulus with temperature and frequency. This can be very useful in analyzing a gear set for mesh stiffness and tooth deflection. Therefore, the storage modulus data can be particularly useful for gearing. However, it is often more convenient to normalize the storage modulus data by dividing all of the values on the curve by the value at ambient temperature, 23°C. Thus, a dimensionless shift factor is created which can be used in any equation containing a modulus to provide a temperature adjustment to the calculation, see figure 9. In addition to the temperature effect on elastic modulus, the DMA data provides considerable useful information about the thermal behavior of plastics. In particular, it is a much better indicator of thermal behavior than the HDT test, see 5.3.3. The maximum useful temperature of any thermoplastic would be near the last point of constant slope of the storage modulus curve before its final downturn. (This is sometimes clearer in a semilog plot.) However, exposure time and load must be greatly reduced as the temperature approaches that point. Beyond this point, the material softens to an unusable state. Semi--crystalline plastics are often used in gears operating in the plateau region past the glass transition temperature. However, if a semi-crystalline plastic gear is operated in the region of the glass transition temperature, performance variability should be expected as properties change rapidly with temperature and frequency.

13

AMERICAN GEAR MANUFACTURERS ASSOCIATION

Normalized modulus

AGMA 920--A01

Temperature Figure 9 -- Typical DMA curves normalized at 23°°C 5.1.3.3 Details of test In a DMA test, the test specimen receives a harmonic excitation at a known frequency and very small amplitude. In most test equipment, the harmonic excitation is the displacement and the resulting force is the measured response. In other equipment, the reverse is used. The sample is mounted in a closed chamber with accurate temperature control. The test chamber temperature is increased at a constant rate during the test. Thus, data are obtained over a wide range of temperatures, typically from --50°C to near the melting point of the material. Figure 10 illustrates a tensile DMA geometry with fixed displacement amplitude. Data obtained from the test are the load amplitude, displacement amplitude, and frequency at a considerable number of temperatures. The frequency is often fixed but a number of frequencies may also be used. Since all materials, and especially plastics, are viscoelastic, there is a phase shift between the displacement and the measured load. The phase shift, δ, is also recorded either directly but more often as tan (δ).

14

The measured load is a combination of two loads: an elastic load which is in phase with the displacement; and a viscous load which is 90 degrees out of phase with the displacement. The load amplitude divided by the displacement amplitude is directly proportional to the absolute value of the complex modulus, |E*|, of the material. The proportionality constant depends on the test geometry. The test geometry also determines which modulus is obtained, i.e., tensile, flexural, shear or compression. The equation for the complex modulus is: E* = E¢ + iE¢¢

(1)

where E¢

is the storage modulus (elastic modulus of material); = |E*| cos δ (2)

E¢¢

is loss modulus (related to viscosity of material); = |E*| sin δ (3)

δ

is the phase shift;

i

is proportionality constant depending on test geometry.

AMERICAN GEAR MANUFACTURERS ASSOCIATION

Apply harmonic displacement

AGMA 920--A01

Displacement

1

2

3

4

5

3

4

5

Phase shift

Test specimen

6

7 time

Load

1

2

7 time

Measured load

Figure 10 -- Tensile DMA Temperature, E¢, E¢¢ and tan (δ) data are generally reported in tabular form for a particular frequency. However, a graphical presentation of E¢, E¢¢ and tan (δ) versus temperature is usually included. Sometimes, the graphical presentation will include frequency as a parameter. The glass transition temperature, Tg, will appear as a peak in the loss modulus curve. This peak will be just after the start of a large decrease in storage modulus. For amorphous plastics, the Tg is near the processing temperature and the storage modulus will continue to decrease as the material softens, see figure 11. For semi--crystalline materials, after a large drop at the Tg, the storage modulus will be relatively constant until the onset of crystallite melting which will be evident as the storage modulus curve bends down near the melting point and the tan (δ) increases rapidly, see figure 12. 5.1.4 Izod impact test -- ASTM D256 The Izod test determines the breakage resistance of a notched, cantilever test specimen subjected to flexural impact. The impact is delivered by a pendulum--type hammer. There are methods for

when the notch faces the hammer and faces away from the hammer. There are special considerations for low energy breaks. Also, a method requiring two different notch radii provides a calculation procedure for notch sensitivity. As each specimen is broken, the type of break according to the following categories must be recorded: C

= complete break -- two or more pieces;

H = hinge break -- an incomplete break in which the attached half cannot support itself when the other is held; P = partial break -- an incomplete break in which more than 9.1 mm (90%) of the 10.16 mm sample width has fractured but does not qualify as a hinge break; NB = non--break -- an incomplete break less than 90% of the original width and all other conditions. The energy required to break each specimen divided by its individual thickness is recorded in J/m, except for NB where no energy is recorded. In every case, the data to be reported, in addition to full equipment and specimen identification, shall include the number of failures in each category along with a

15

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

statistical summary for each category except NB. When two notch radii are used and if all breaks are of the same category, the notch sensitivity may be

calculated by dividing the difference in the average energy per unit width for each radius by the difference in the radii.

Tg

Loss modulus, E¢¢

Log storage modulus, E¢ Log tan (δ)

E¢ = |E*| cos (δ) tan (δ) E¢¢ = |E*| sin (δ)

Temperature Figure 11 -- DMA, amorphous and crystalline polymers

Tg

E¢ = |E*| cos (δ) tan (δ)

Loss modulus, E¢¢

Log storage modulus, E¢ Log tan (δ)

E¢¢ = |E*| sin (δ)

Temperature Figure 12 -- DMA, semi--crystalline polymer

16

AMERICAN GEAR MANUFACTURERS ASSOCIATION

5.1.4.1 Significance of test Izod is widely used in the United States for plastics. Charpy (see 5.1.5) is used for metals in the U.S. and for plastics in Europe (although the method is different). With the adoption of ISO methods in the U.S., Charpy is becoming more common. The most widely reported data are for a notched specimen. They are typically used as an indication of the relative notch sensitivity when comparing materials. However, this should only be done when comparing materials that have the same break category. Unfortunately, the break category is rarely reported so such comparisons should be made with great care. Data are sometimes reported for reversed notch or un--notched (not part of a standard) samples. The reversed notch data may be compared to the notched data as a further indication of relative notch sensitivity. Notch sensitivity by multiple notch radii is rarely reported. Although NB stands for non--break, the proper interpretation is “no test” as NB is used for all outcomes that do not fall under the other three categories. Finally, Izod data are often mistakenly interpreted as a measure of toughness. A material with a higher Izod value may or may not have greater toughness. Izod should only be used as a relative indication of notch sensitivity. 5.1.4.2 Limitations for gear applications Plastic gears designed with a full fillet radius, properly molded, and handled without damage generally do not require consideration of notch sensitivity. However, many gears are designed using little or no fillet radius, have flow lines in the tooth root, and are mishandled. Such gears could benefit from a material with an increased notched lzod value. Within a material family, increased Izod values are usually obtained by adding an elastomer to the base material. This will typically lower the modulus of the material and increase the hysteresis. The reduced modulus will increase tooth deflection, reduce contact stress and could result in a poor contact condition. Increased hysteresis can increase operating temperature. Thus, for continuously operated gears, the life may be reduced. However, if gears that operate intermittently with occasional shock loads are failing prematurely, a material modified for increased Izod may give longer life.

AGMA 920--A01

5.1.5 Charpy impact test -- ASTM D256 This test method determines the breakage resistance of a notched, three--point bending test specimen subjected to impact. The impact is delivered by a pendulum--type hammer. The sample is supported on both ends and is struck in the center while the notch faces away from the hammer. Data is obtained and reported as in notched Izod, except only complete breaks are reported. 5.1.5.1 Significance of test Izod is widely used in the United States for plastics. Charpy is used for metal in the U.S. and for plastics in Europe (although the method is different). The data are typically used as an indication of the relative notch sensitivity when comparing materials. Charpy data are often interpreted as a measure of toughness. A material with a higher Charpy value may or may not have greater toughness. Charpy data should only be used as an indication of notch sensitivity. 5.1.5.2 Limitations for gear applications Limitations are essentially identical to the Izod test. See 5.1.4.2. 5.1.6 Shear strength of plastics by punch tool -ASTM D732 This test reports the extent to which the plastic material can resist shear stress. A portion of the material is forced to separate from the rest by sliding in a direction parallel to the applied load. The data are expressed as a shear strength calculated from the peak punch load and the sheared area. 5.1.6.1 Significance of test If applied to similar specimens of various materials, the test indicates the relative strength of the materials under conditions of a shear load. 5.1.6.2 Limitations for gear applications Loading on gear teeth typically subject them to failure by bending rather than by shear. One exception is encountered when a worm is loaded against the teeth of a plastic gear, shearing portions of the engaged teeth. In such cases, the shear strength test data can be used for design as long as proper adjustments are made for differences between test specimen and plastic gear in respect to material processing, cross--section size and rate of loading.

17

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

5.1.7 Standard test method for flexural fatigue by constant--amplitude--of--force -- ASTM D671 This test method covers the determination of the effect of repetitions of the same magnitude of flexural stress on plastics by fixed--cantilever type testing machines designed to provide a constant-amplitude--of--force. The test results provide data on the number of cycles of stress to produce specimen failure by fracture, softening, or reduction in stiffness by heating as a result of internal friction (damping). The test is performed by repeatedly flexing a fixed cantilever specimen with a fully reversing predetermined load, see figure 13. 5.1.7.1 Significance of test Thermoplastics can fail in two different ways. Like metal, they can fail in fatigue due to cumulative damage caused by a repeated stress. In this failure mode, cracks grow continuously with each stress cycle until the effective load bearing area is too highly stressed to support the resulting stress. When this occurs, the cracks will propagate catastrophically and the component will fail.

5.1.7.2 Limitations to gear applications Differences between the flexural fatigue test and typical gear applications limit the suitability of using the test data as gear design data. In the flexural fatigue test, the loading at the critical section is a bending load, which is also the case at the critical section in gear teeth. While the load rate in the test may be more closely matched to the generally high load rate in gears, the test does not allow for the delay between successive load applications typical of gears. An even greater difference is in the type of loading, which is full reversing in the test, but only zero--to--maximum loading in most gear applications. 5.1.7.3 Details of test The testing is to be conducted at 50% RH and 23°C. The mechanical properties of many plastics change rapidly with small changes in temperature. Since heat is generated as a result of the flexing action of the test, the test is conducted without forced cooling to ensure uniformity of test conditions. The temperature of the sample during testing is to be measured and recorded, but it is seldom reported. For most plastics, fatigue failures are frequency dependent. Therefore, data should not be extrapolated to other frequencies unless the frequency response is known. ASTM suggests testing at 30 Hz  5%.

50.8 mm

Unlike metals, thermoplastics have a second failure mode which is due to their viscoelastic nature. A thermoplastic subjected to repeated load cycles at high frequency will generate heat due to internal friction (damping). Since thermoplastics are insulative, heat generated can easily exceed the material’s ability to dissipate it. The resulting increase in temperature leads to material softening, diminishing its ability to resist stress. This failure, often referred to as thermal failure, is characterized by excessive

deflection of the test sample. This type of fatigue failure is said to occur when the apparent modulus of of the material decays to 70% of the original modulus of the specimen at the start of the test.

57.2 mm 103.2 mm

Figure 13 -- Flexural fatigue specimen

18

AMERICAN GEAR MANUFACTURERS ASSOCIATION

The results of the test are plotted on a S--N (stress vs. cycles) diagram with the alternating stress amplitude as the ordinate against the common logarithm of the number of cycles required for failure as the abscissa. Fatigue strength is always associated with a number of cycles. If a S--N curve for a material becomes a horizontal line (constant-stress) at very high cycles (>10 million), it is said to have an endurance limit. The endurance limit is the stress level below which the material can be subjected to the fatigue load indefinitely. Not all plastic materials will have an endurance limit. 5.1.8 Tensile, compressive and flexural creep and creep--rupture of plastics – ASTM D2990 This test method determines the time--dependent deformational response (viscoelastic deformation) of plastics subjected to constant loading conditions under specified environmental conditions. Procedures are described for constant tensile, compressive or flexural loadings. However, measurements of creep--rupture require tensile loading, since rupture does not occur in compression or flexure. Therefore, tension is the preferred stress--state for these tests and only the tensile creep/creep--rupture test will be covered in this description. 5.1.8.1 Significance of test The information obtained from these tests can be used in the design process of parts subjected to time--dependent loadings. This design process, know as the “quasi--elastic” design methodology, uses results (formulas) from elastic stress analysis. The material properties (e.g., Poisson’s ratio, Young’s modulus, strength, etc.) in those formulas for elastic analysis are normally the instantaneous (time equals zero) values. However, for the “quasi-elastic” design methodology, the values at the required design time (time greater than zero) are used in the elastic formulas. The “quasi--elastic” design methodology has been shown to yield conservative results.

AGMA 920--A01

condition, creep deformation and creep--rupture of plastics may need to be considered. This is especially true for stalled gears where the ratio of stalled time to cycled time is large. Under these conditions significant creep strain could accumulate, possibly to the point of creep--rupture (tooth breakage) and/or significant tooth creep--deformation (tooth spacing errors). 5.1.8.3 Details of test The test specimens used in the constant tensile creep and creep--rupture testing are the dumbbell shaped specimens described in ASTM D638 (see 5.1.1.1), either Type I or Type II. In addition, specimens described in ASTM D1822 can be used for creep--rupture testing. The test specimens are conditioned at 23° C and 50% RH for not less than 40 hours prior to testing. Additionally, the specimens are pre--conditioned for at least 48 hours in the test environment (temperature, humidity and others) immediately prior to being tested. After conditioning and pre--conditioning in the specified testing environment, a constant tensile load is rapidly applied (loading time not to exceed 5 seconds) to each specimen. The time--dependent elongation of each specimen is periodically measured. At the beginning of the test the measurements of elongation are made within minutes of each other. During the middle of the test the elongation measurements are made within hours of each other. For tests lasting longer than 1000 hours the elongation measurements are made at least monthly. The results of this test are a family of curves presenting elongation (creep--strain) vs. time. There is one curve per specimen for each constant tensile load applied (tensile stress applied). However, the data can be presented in a number of fashions: 1) creep--strain vs. time (see figure 14a); 2) creep modulus (tensile--stress/creep--strain) vs. time (see figure 14b); or 3) tensile stress vs. strain curves (see figure 14c), each for a specified time (isochronous curves).

5.1.8.2 Limitations for gear applications

5.2 Wear and frictional characteristics

Typically creep deformation is not normally a factor in gears under continuous operation because the load is applied to each gear tooth only for short time duration during each gear revolution. Therefore, in the limit, those loadings cannot be considered as time--dependent loadings. However, for gears run continuously at low speeds and high loads, or for gears run at high speeds into and/or held in a stalled

5.2.1 Thrust washer wear test -- ASTM D3702 The wear resistance and frictional characteristics of plastic materials in rubbing contact with another surface are important properties to consider in designing plastic gears. Tribological values from a standardized thrust washer wear test are often used for relative comparisons of thermoplastic materials to assess these characteristics.

19

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

5.2.1.1 Significance of test Increasing stress Strain

This test method is used to determine the equilibrium rate of wear and the coefficient of friction for materials in sliding contact at a variety of pressure-velocity conditions. These data are intended only as initial guides in the material identification stage for a given application. 5.2.1.2 Limitations for gear applications

Time (log scale) Figure 14a -- Representations of creep -strain vs. time

Creep modulus

Increasing stress or strain

Time (log scale) Figure 14b -- Representations of creep -creep modulus vs. time

Time (hr)

1

10

100

1000

10 000

Wear and frictional characteristics of plastics are not material properties, they are system properties. It cannot be overemphasized that the operational conditions of individual applications dramatically affect these properties. Parameters such as mating surfaces, velocities, pressures, ambient temperatures, duty cycles, and type of motion, also affect the relative performance of one material compared to another. The majority of thrust washer wear testing data reported by thermoplastic suppliers are generated using a modified version of ASTM D3702. Changes from the standard may include different size, different counterface composition and/or different operating conditions, and the units for wear factor (K) may be expressed differently. When comparing data provided by different suppliers, it is important to understand the possible differences in the test method used to generate the data, and how these differences could affect the values generated. Because of differences in test methods and technique along with highly variable results, comparing data between different labs is not generally appropriate. See figure 15. Further, the results of this constant contact, unidirectional test cannot be used to predict the wear life or frictional characteristics of an intermittent, rolling--sliding, line contact found in many gearing applications.

Stress

5.2.1.3 Details of test

Strain Figure 14c -- Representations of creep -isochronous stress vs. strain 20

Wear rate and wear (K) factors: Lower values indicate better resistance to material loss due to relative motion contact. A material exhibiting a wear rate or wear (K) factor value that is half of another material indicates the material loss is also half at this specific pressure--velocity (PV) point. This does not mean that this relationship necessarily holds true at higher or lower PV points or under any other different operational conditions.

AMERICAN GEAR MANUFACTURERS ASSOCIATION

AGMA 920--A01

Figure 15 -- ASTM D--3702 thrust washer wear and friction test Coefficients of friction (Cf): Lower Cf values for plastic materials indicate reduced resistance to sliding. Given identical testing conditions and normal forces, Cf values are linear in representing the force needed either to initiate (static) or maintain sliding motion (dynamic). Coefficients of friction vary significantly relative to differences in pressures, velocities, mating surface characteristics, ambient temperatures, duty cycles, rotational direction and other operational conditions. Further, Cf values may change significantly between mating surfaces during the break--in or run--in periods due to material transfer mechanisms and other polymer wear phenomena. Published coefficients of friction numbers are typically numerical averages and may not reflect the magnitude of these running changes in sliding resistance.

plastic materials having coefficients of expansion greater than 1 ¢ 10 --6 mm/mm/°C. The thermal expansion of a plastic is a reversible change in dimensions caused by heating and cooling. Superimposed upon this reversible process are other changes in length, which are essentially non--reversible, due to heat, changes in moisture content, curing, loss of plasticizers or solvents, release of stresses, phase changes and other factors. This test method is intended to determine the CLTE under the exclusion of these factors as far as possible. In general, it will not be possible to exclude them altogether. For these reasons, the test can only be expected to give an approximation of the true thermal expansion. 5.3.1.1 Significance of test

5.3 Thermal properties 5.3.1 Standard test method for coefficient of linear thermal expansion of plastics between --30°C and 30°C -- ASTM D696 This test method covers the determination of the coefficient of linear thermal expansion (CLTE) for

Coefficient of linear thermal expansion can be used to estimate the changes in mesh clearances required to prevent gears from binding or from coming out of mesh due to relative expansion between the gears and their housing.

21

AGMA 920--A01

5.3.1.2 Limitations for gear applications As described in 5.3.1, the change in dimensions of a molded part may not completely agree with the change in length predicted by applying the CLTE to the part dimension and temperature change. Also, the CLTE of a material can and will vary over different temperature ranges and flow orientation. 5.3.1.3 Details of test The results of the testing are generally reported as the coefficient of linear thermal expansion over a range. For anisotropic materials, CLTE is referenced to a coordinate (X and Y) or direction (flow and transverse). 5.3.2 Thermal conductivity test Thermal conductivity is the rate at which a material conducts heat energy along its length or through its thickness. This property is important in applications where the polymeric material is used as a thermal insulator, or where heat dissipation is of concern. 5.3.2.1 Significance of test The thermal conductivity of a thermoplastic material will effect how well the material dissipates heat. Low thermal conductivity materials like thermoplastics will not dissipate heat generated by friction (tooth contact, bearings) as well as a metal would, and can result in a greater temperature rise in the application than expected. Housings made from thermoplastics will also dissipate less heat than a similar metal housing. On the positive side, thermoplastics can be used to insulate other components from external heat sources. 5.3.2.2 Limitations for gear applications Thermal conductivity is generally measured at some reference temperature, and the actual value for thermal conductivity can change as the environmental and/or application temperature changes and use of different measuring techniques. 5.3.2.3 Details of test Thermal conductivity can be measured for plastics as a solid or in the melt, and these values are generally very different. The melt thermal conductivity is used for doing computer aided mold filling analysis. The thermal conductivity of the solid thermoplastic is covered by ASTM Standard F433, Standard Practice for Evaluating Thermal Conduc22

AMERICAN GEAR MANUFACTURERS ASSOCIATION

tivity of Gasket Materials, and is used during the design phase of a project. 5.3.3 Heat distortion (deflection temperature) test -- ASTM D648 Among the various ways to characterize the thermal performance of a plastic material is ASTM D648. This test can be used to compare the thermal performance among various plastics. It is performed by submerging a specimen in a temperature controlled environment, raising the temperature incrementally, and reporting the temperature at which the specimen deforms 0.25 mm, the result reported as the heat deflection temperature (HDT). Either of two (or both) load conditions can be specified, 0.455 MPa or 1.82 MPa, the latter reported as “Heat deflection temperature under load” or HDTUL. 5.3.3.1 Significance of test The test shows the temperature at which a certain amount of deflection takes place at known loads. It is not a direct guide to the temperature performance of the material in application. It may be useful to compare various plastic materials under the same conditions. At best it is only a rough guide to the upper limit of a material’s thermal behavior. 5.3.3.2 Limitations for gear applications There are three major shortcomings in the procedure. -- First, an initial deflection corresponding to the applied stress occurs before the bath is heated. The measurement of 0.25 mm is from this point. The measurement does not correlate with any pure physical or design property of the material. -- Second, the thickness of the sample is variable. The calculations that determine the load required to produce the specified stress level theoretically take thickness into account, but in practice, thicker samples perform better than thin ones. -- Third, the heating rate at which the test is run influences the HDT value. Because plastics have low thermal conductivity, higher heating rates produce higher measured deflection temperatures. Better methods for determining a material’s thermal performance are available, namely, the Dynamic Mechanical Analysis test, where the operator of the test can continuously monitor the flexural modulus of a material as a function of temperature. It is better to determine the required yield strengths and modulii

AMERICAN GEAR MANUFACTURERS ASSOCIATION

for the gear design, and then find the temperature at which these values are exceeded, and use that temperature as the upper limit for the material. 5.3.4 U.L. temperature index test (relative thermal index or RTI) -- UL 746B This procedure is used to determine the relative thermal index for a particular material. The relative thermal index (RTI) is the maximum temperature a selected material will retain 50% of its original mechanical and electrical properties after heat aging for 100,000 hours. 5.3.4.1 Significance RTI data is typically used when selecting materials for electrical components that require UL listing. The electrical application may require the material to have an RTI rating at or above the expected operational temperature of the device. 5.3.4.2 Limitations for gear applications RTI testing is a static test that does not take into account the mechanical properties of the material at temperature. Actual strength and stiffness properties of the material at the operational temperature are needed to determine the material’s capability in the application. In spite of this, the RTI of a material is a good indicator of a material’s capability for long term exposure to elemental temperatures. Some plastic applications, such as gears to be used in appliances, may require an Underwriter’s Laboratories relative thermal index. Generally, this is a temperature value assigned to a polymer, based on long term testing, extrapolated to the life time of a product design, at which the part can operate without failing as an electrical insulator. 5.3.4.3 Details of test A relative thermal index of a material is an indication of the ability of a material to retain a particular property (physical, electrical, etc.) when exposed to elevated temperatures for an extended period of time. It is a measure of the material’s thermal endurance. For each material, a number of relative thermal indices can be established, each index related to a specific property of the material. The RTI of a material is established on the basis of either accelerated aging experiments or on a generic basis from field experience with specific families of materials. There may be up to three independent

AGMA 920--A01

RTI’s assigned a single material, namely: electric, mechanical with impact and mechanical without impact. In this method, pertinent properties are measured as a function of time and temperature, and using appropriate mathematical techniques (regression analysis), determine the time to end of useful service at each temperature. End of useful life is defined as the time at which the property being measured has degraded to 50% of its original value. The long term material performance is determined relative to that of a reference or control material, thus the term relative temperature index. The RTI is then published in the UL Components Index, listed along with the various other properties measured for electrical applications. 5.3.5 Brittleness temperature of plastics and elastomers by impact -- ASTM D746 This test method determines, under specified impact conditions, the temperature at which plastics exhibit significant brittle behavior. The temperature that is determined is termed the “brittleness temperature”. The brittleness temperature is the temperature at which 50% of the test specimens fail in a brittle manner under the specified impact conditions of this test. 5.3.5.1 Significance of test Data collected under this method can be used to predict the behavior of plastics or elastomers at low temperatures. Such data can only be used where the conditions of deformation are similar to those specified in the test method. It is useful for specifications, but does not measure the lowest temperature at which a material may be used. 5.3.5.2 Limitations for gear applications Actual low temperature performance may be understated using this test, as it may not measure the lowest temperature at which the material may be used. The test is difficult to run, the specimens bear no relation to the shape of a gear, and the needed attendant support of the gear design would further understate the performance. Data should only be used as a relative indication of performance between material properties which tested identically, and hence for specification work only. When choosing a polymer for gears required to perform under sudden changes in load at low temperatures, the brittleness temperature for each considered polymer should be evaluated and compared.

23

AGMA 920--A01

5.3.5.3 Details of test The test determines temperature at which 50% of test specimens fail when subjected to an impact of a striking edge moving at 2000 mm/s (+/-- 200 mm/s) over a distance (after striking the specimens) of 6.4 mm, the specimen holder having been held in a cooling medium at a known temperature for a period of 3 minutes (+/-- 0.50 min). 5.4 Environmental properties 5.4.1 Standard test for water absorption of plastics -- ASTM D570 This test method covers the determination of the relative rate of water absorption by all types of plastics. 5.4.1.1 Significance of test Moisture absorption will affect the dimensions and physical properties of plastic gears in varying degrees. 5.4.1.2 Limitations for gear applications This test shows only weight change due to water absorption when immersed. Effects of water absorption on dimensions and properties are not quantified by this test. 5.4.1.3 Details of test Depending on the material’s water absorption characteristics versus temperature, the procedure for achieving its “dry” condition varies as described in the ASTM standard. These “dry” samples are then measured and weighed. There are several immersion tests in distilled water that may be conducted. In water at 23°C, these are the 24 hour, the two hour, the repeated, and the long--term immersion tests. Conducted in boiling water are the 2 hour and the 1/2 hour immersion test. 5.4.2 Standard practices for evaluating the resistance of plastics to chemical reagents -ASTM D543 These practices cover the evaluation of all plastic materials for resistance to chemical reagents. Two major tests are described: an immersion test, and a mechanical stress test to reagent exposure. The first is for weight and dimensional changes, and also for 24

AMERICAN GEAR MANUFACTURERS ASSOCIATION

mechanical property changes. The second is for the susceptibility to attack in stressed regions. 5.4.2.1 Significance of test The choice of types and concentration of reagents (including lubricants), temperature of the test, and properties to be reported is necessarily arbitrary. The specification of these conditions provides a basis for standardization and serves as a guide to investigators wishing to compare the relative resistance of various plastics to typical chemical reagents. 5.4.2.2 Limitations for gear applications As with all standard tests, correlation of the test results with the actual performance of plastic gears is dependent upon the similarity between the testing and the actual conditions. It should be noted that this ASTM standard addresses statically loaded specimens, unlike gearing applications which are dynamically loaded. These dynamic loads may contribute to differences in performance and susceptibility to chemical exposure. 5.4.2.3 Details of test The test specimens are conditioned at 23°C and at 50% RH for not less than 40 hours prior to testing. Shape and dimensions of the specimens depend on the test to be performed according to the ASTM standard. Two procedures are followed in the immersion test depending on interest. Procedure I is for weight and dimensional changes. Procedure II is for changes in mechanical property. The mechanical stress test evaluates specimens mounted on strain fixtures and exposed to chemical reagents, either by immersion or by the wet--patch method. These specimens are subsequently compared to unstrained specimens similarly exposed. Exposure times are 7 days for room temperature, of 3 days for elevated temperatures. Another set of specimens are identically strained, but not exposed, and serve as a control. After exposure, mechanical properties of exposed and unexposed specimens are compared. Standard methods are followed for tensile, flexural, or other property evaluation.

AMERICAN GEAR MANUFACTURERS ASSOCIATION

AGMA 920--A01

5.5 Miscellaneous properties

5.5.2.2 Limitations for gear applications

5.5.1 Test method for rubber property -durometer hardness -- ASTM D2240

The Rockwell hardness measurements in plastics do not necessarily relate to the material’s resistance to wear.

This test method is used to determine the indent hardness of elastomeric and plastic materials using the durometer hardness test apparatus.

5.5.2.3 Details of test

5.5.1.1 Significance of test Durometer hardness testing is used to determine the indentation or surface hardness of a material. Durometer hardness is often used in the specification of elastomers with all other physical properties implied by the durometer for that elastomer class. 5.5.1.2 Limitations for gear applications The durometer hardness measurements in plastics do not necessarily relate to the material’s resistance to wear. 5.5.1.3 Details of test In this test, a pointed or blunt indenter of set diameter (sharp point for Shore A, 0.10 mm radius point for Shore D) is applied to the specimen. The spring load for the durometer tester determines the force of penetration. The application force should be high enough to ensure adequate contact of the Shore meter’s presser foot. The Shore hardness number is read from the durometer scale on the test apparatus. The durometer can be determined after immediate application or after a time period agreed upon by specification. The Shore number is inversely related to the distance the indenter penetrates into the test specimen. One Shore point is equal to 0.025 mm of penetration. The higher the Shore number the harder the material. There are other Shore scales, however, the “A” and “D” scales are traditional for elastomer and plastic materials. 5.5.2 Standard test method for Rockwell hardness of plastic and electrical insulating materials -- ASTM D785 This test method is used to determine the indent hardness of plastic materials using the Rockwell hardness test apparatus.

In this test, a rounded indenter of set diameter, (12.7 mm for Rockwell R, 6.35 mm for Rockwell M) is applied to the specimen under a 10 kg minor preload. A 60 kg major load for Rockwell R or a 100 kg major load for Rockwell M is applied for 15 seconds. The load is released. The Rockwell number is determined from the difference in travel of the indenter from the major load to the preload. Each Rockwell division is equal to 0.002 mm of travel. The Rockwell number is the number of divisions traveled (depth of indent) subtracted from 150. The higher the Rockwell Number the harder the material. There are other Rockwell Scales, however the “R” and “M” scales are traditional for plastic material. 5.5.3 Density and specific gravity (relative density) of plastics by displacement – ASTM D792 This test method presents procedures to determine the specific gravity (relative density) and density of solidified plastics as in extruded shapes and molded objects. The specific gravity is the ratio of the mass of a certain volume of solidified plastic to the mass of an equal volume of water or some other reference liquid. Two methods are presented: 1) procedures in water, and 2) procedures in liquids other than water, which can be used if the plastic is lighter than water or undergoes significant absorption of water at 23° C over the testing duration. 5.5.3.1 Significance of test The specific gravity or density can be measured and used to 1) identify a plastic, 2) to track physical changes in a sample, 3) to determine the uniformity among samples, or 4) to indicate the average density of large objects. Differences in density of the same object, or among samples, may be due to changes or differences in crystallinity, loss of plasticizer, absorption of a solvent, and/or differences in thermal history, porosity, and/or composition. Density can also be used to calculate, knowing the volume of a part, expected weight and material costs of that part.

5.5.2.1 Significance of test

5.5.3.2 Limitations for gear applications

Rockwell hardness testing is used to determine the indentation or surface hardness of a material. Rockwell hardness can also be used to determine the degree of cure for a thermoset material.

In practice, when using the specific gravity or density along with an expected gear volume to calculate the expected weight of material in a gear, the calculated weight might not be exact. The lack of exactness

25

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

might be due to differences in crystallinity induced by differences in processing conditions of the gear and sample used to determine the specific gravity. Another factor causing the lack of exactness that is not discussed in 5.5.3.1 is unaccounted for shrinkage of the gear during molding which might change the anticipated volume.

some cases be subject to a source of flame. It is important to note that any UL flammability rating is given at a particular thickness, and that as the sample thickness increases, a better rating can often be obtained. When specifying or requesting a rating, the thickness of the application must be specified.

5.5.4 Flammability

Compliance with these regulatory requirements often requires the addition of additives which may alter the mechanical performance or characteristics of the base material. Because of these performance differences, the development for any gear application shall be done with the material that meets the regulatory requirements.

There are several types of flammability testing, the most common being the UL Flammability Class. UL Subject 94 and/or ASTM D635 and ASTM D3801: In this test, specimens are subjected to a specific flame exposure, and the relative ability to maintain combustion after the flame is removed becomes the basis for classification (HB, V--2, V--1, V--0, 5V). In general the more favorable ratings are given to materials that extinguish themselves rapidly and do not drip flaming particles. Each rating is specified on a specific material thickness. Other flammability tests are: Oxygen Index Test, ASTM D2863: This test measures the percentage of oxygen necessary to sustain combustion of the plastic material. The higher the value (more oxygen needed), the lower the combustibility. Since air contains about 21% oxygen, any material with a rating below 21 will probably support combustion in a normal, open air environment. Smoke Density Test, ASTM E662: Often referred to as the NBS Smoke Density Test, a specified area of plastic is exposed to heat under flaming conditions. Smoke measurements are reported as “specific optical density”, a dimensionless unit that represents the optical density of the smoke over a unit path length within a chamber of unit volume produced from a sample of unit surface area. Glow Wire Test, IEC 60695--2--10, 11, 12, 13: This test simulates conditions present when an exposed, current carrying conductor contacts an insulating material during faulty or overloaded operation. The test can be applied at one or more recommended temperatures (550_C, 650_C, 750_C, 850_C, 960_C) and at any sample material thickness. 5.5.4.1 Significance to gearing Most gearing applications do not require a specific flammability rating, since they are usually enclosed and not exposed to a source of flame. The housing around the gear may have a flammability requirement since it has more surface area, and may in 26

5.5.4.2 Limitations for gear applications

6 General description of plastic materials This brief description deals only with characteristics of materials used for gears. 6.1 Classification Plastics are generally classified in two major groups, thermoplastics and thermosets. 6.1.1 Thermoplastics 6.1.1.1 Definition and properties Thermoplastics are materials that repeatedly soften, or melt, when heated, and harden, or freeze, when cooled. Heating permits the intertwined molecular chains to slide relative to each other. At some higher temperature, the sliding is free enough for the material to behave as a liquid and may be used to fill molds. The temperature at which this degree of softening takes place varies with the type and grade of plastic. Cooling restores the intermolecular bonds and the material behaves essentially as a solid. However, these solid materials, to a varying degree, retain some aspects of a liquid in the form of viscoelastic behavior. See 2.1. At the same time, this type of molecular bond generally imparts a greater toughness, or resistance to impact loads. 6.1.1.2 Structure Thermoplastic materials can be further classified by their chemical structure. Many of the physical property differences among plastics can be attributed to their structure. 6.1.1.2.1 Crystalline Thermoplastic materials are divided into two categories or families of plastics: semi--crystalline (general-

AMERICAN GEAR MANUFACTURERS ASSOCIATION

ly referred to as crystalline) and amorphous. Crystalline thermoplastic materials are melt processable plastics that, upon cooling from the melt phase, solidify with distinct crystalline domains. Crystalline materials are easily differentiated from amorphous thermoplastics because they are opaque. Crystalline materials display good chemical resistance, fatigue properties and wear resistance compared to amorphous plastics. Crystalline materials also maintain usable physical properties beyond their glass transition temperature. Crystalline materials (see table 2) have a distinct melt temperature beyond which the plastic is a liquid. See figure 16. The ordering of crystalline plastics makes them stiffer, stronger, but less resistant to impact than their non--crystalline counterparts. See figure 17. 6.1.1.2.2 Amorphous materials Amorphous plastic, like a pane of glass, is more similar to a super cooled liquid than a solidified material. Like glass, amorphous materials are transparent. Amorphous plastics have limited chemical resistance and do not have useable properties beyond their glass transition temperature. Because amorphous plastics do not have specific melt temperatures, the glass transition temperature can be considered the onset of melt. Amorphous plastics do have good creep resistance properties. Crystalline

AGMA 920--A01

Amorphous plastics have low shrink characteristics and are not susceptible to high differential shrinkage allowing for accurately molded, warp free parts. Some amorphous plastics (see table 3) have good impact properties. 6.1.1.2.3 Thermal response -- crystalline vs. amorphous A primary thermal transitional common to all thermoplastic resins is the glass transition temperature, or Tg. For crystalline resins, this is the temperature at which the amorphous regions of the polymer begin to soften, allowing the harder and more ordered crystalline regions to move over each other. Mechanically, crystalline polymers begin to lose a major portion of their modulus through this transition. Amorphous resins, which contain no crystalline regions, very quickly lose all modulus at the Tg, becoming unusable for mechanical purposes. In fact, it is above this thermal point that amorphous materials flow. Crystalline polymers also exhibit a melting point, Tm, the temperature at which the crystalline regions of the material change state. This transition point is typically very sharp and unique for each crystalline polymer, and is the point at which no mechanical structure is evident. Amorphous resins do not exhibit a melting point. Amorphous

Melt

Crystals Amorphous

Solid Figure 16 -- Two dimensional representation of crystalline and amorphous thermoplastics

27

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

Crystalline (glass fiber reinforced) Tg Tg

Modulus

Amorphous (glass fiber reinforced)

Crystalline (unreinforced) Tg

Tg

Amorphous (unreinforced) Tm Temperature Figure 17 -- Modulus behavior vs. temperature of crystalline and amorphous resins, neat and glass fiber reinforced Figure 17 shows the modulus response for neat amorphous and crystalline resins, as well as glass fiber reinforced versions. For both types of polymers, the addition of reinforcing fibers substantially increases the modulus, yet does not affect the inherent thermal transition temperature points. NOTE: Dynamic mechanical analysis (DMA) (see 5.1.3) is an analytical method to determine polymer property changes as a function of temperature. The change in modulus characteristics of a plastic material, as it approaches or exceeds important and unique thermal transitions, can yield important design information about the load bearing capabilities within the temperature range of operation. These data are more useful to the design engineer than the commonly cited ASTM test, deflection temperature under load (DTUL), which determines at what temperature a standard molded specimen deflects 0.25 mm.

6.1.1.3 Molding considerations Thermoplastic materials are particularly well suited for molding. Molded material in the form of sprues and runners, if properly processed, can often be reground and recycled with little or no loss in properties, at least for a limited number of reuse cycles. This is less so for materials with fiber

28

additives. The use of regrind may offer a cost reduction. Further economy is possible with “hot runner” molding systems in which no scrap is produced. Solidification by cooling does make the molded part subject to variation in properties with differences in cooling rate. Very often, the greater rate of cooling at the surfaces which contact the mold produces a harder and stronger “skin”, while the more slowly cooling internal material, or core, is somewhat softer. A similar effect takes place at sharp internal corners of the molded part, such as the sharp fillets at the base of gear teeth. Although this surface also contacts the cooled mold, the correspondingly sharp external corner of the mold is a poorer conductor of heat, and thus produces a weaker skin, than the less sharp surfaces elsewhere in the mold. 6.1.1.4 Design considerations The thermal effects associated with this type of solidification may also affect the design of the gear blank. Best molding results are obtained when the various sections in the blank are approximately uniform in thickness. For many thermoplastic materials, large gears with thick sections cannot be

AMERICAN GEAR MANUFACTURERS ASSOCIATION

molded to meet the same high quality levels of smaller gears.

AGMA 920--A01

Some of these properties are: --

strength;

6.1.2 Thermosets

--

impact resistance;

6.1.2.1 Definition and properties

--

rigidity (modulus);

--

thermal conductivity;

--

flame retardance;

--

dimensional control;

--

color;

--

heat stability;

--

noise reduction;

--

oxidative stability;

--

U.V. stability;

--

lubricity;

--

wear resistance;

--

processability.

Thermosets are plastics that undergo chemical change during processing to become permanently infusible. If excessive heat is applied to a thermoset material after the chemical change has taken place, the plastic is degraded rather than melted. Before the processing, the molecular structure of the thermoset plastic is similar to that of a thermoplastic material. Heating permits the relative sliding of the molecules, the material takes on the properties of a liquid and can be used to fill molds. However, while still subjected to heat in the mold in a curing process, the intertwined molecules develop cross--links to form an irreversible network which prevents further relative sliding. The resulting solid plastic behaves much more like an elastic material similar to metals. The viscoelastic behavior is much reduced from that of most thermoplastics. At the same time, these thermoset materials tend to be much less resistant to impact loading, and tend to be used only with some reinforcing medium in even moderate impact applications. On the other hand, thermosets tend to maintain their strength properties at much higher temperatures than most thermoplastics. 6.1.2.2 Molding considerations Solidification by chemical reaction often requires greater processing time to allow for completion of the curing. This tends to increase processing cost. Some of the molding processes used for thermoset plastics are almost scrapless, but others do leave substantial scrap. 6.1.2.3 Design considerations The thermoset solidification process makes the molded material less sensitive to cooling rates. It therefore may permit the successful molding of parts with varying section thickness and large parts with heavy sections. 6.2 Additives A variety of organic and inorganic materials are added to plastics. A few may be used to reduce cost, but most are used to improve a preferred property. See table 1.

6.3 Available forms 6.3.1 Molding materials For injection molding, which covers just about all the thermoplastic materials, the material is supplied in the form of pellets. If the molded part material is to contain additives, these additives are generally pre--compounded in the pellet material, uniformly distributed in the proper proportion. For molding processes used for thermoset materials, the material may be in the form of pellets or molding preforms. These preforms are of a size and shape to suit the mold, and of a volume to completely fill the mold cavity with a minimum of flash. If the molded part is to contain laminations of reinforcement material, these laminations are introduced into the preforms. 6.3.2 Machining materials (see 4.1) Plastic materials to be machined into gears are available in a variety of forms. They are selected to suit the size of the gear and the machining process used to prepare the gear blank. Most common are extruded circular rods or tubes. For some plastic materials, these rods are extruded by stock shape suppliers in diameters up to 200 mm; for others only up to 150 or 100 mm. To go beyond these diameter limits, materials may be available in the form of cast solid or tubular bars or in the form of disks. For thin gears, the blanks may be stamped or cut from sheets; for thick gears and large diameters, the gear blanks may be cut from plates that are stocked up to 100 mm in thickness.

29

AGMA 920--A01

AMERICAN GEAR MANUFACTURERS ASSOCIATION

Table 1 -- Additives in plastics for molded gears

Type Fillers

Reinforcements

Name Minerals (mica, talc, carbon powder, glass beads, etc.) Organic Glass fiber Carbon fiber Aramid fiber

Percentage range 5--40

Reason for use1) C, DC, E, H

Limiting factors2) WM, RI, TW