1、08FTM15AGMA Technical PaperExtending the Benefitsof Elemental GearInspectionBy I. LaskinExtending the Benefits of Elemental Gear InspectionIrving LaskinThe statements and opinions contained herein are those of the author and should not be construed as anofficial action or opinion of the American Gea
2、r Manufacturers Association.AbstractIt may not be widely recognized, that most of the inspection data supplied by inspection equipmentfollowingthepracticesofAGMAStandard2015andsimilarstandards,arenotofElementalaccuracydeviationsbutofsome form of Composite deviations. This paper demonstrates the vali
3、dity of this “Composite” label by firstdefiningthenatureofatrueElementaldeviationandthen,byreferringtoearlierliterature,demonstratinghowthe common inspection practices for Involute, Lead (on helical gears), Pitch, and in some cases, TotalAccumulated Pitch, constitute Composite measurements. The pape
4、r further explains how suchmeasurements often obscure the true nature of the individual deviations. It also contains suggestions as tosomelikelysourceofthedeviationinvariousgearmanufacturingprocessesandhowthatdeviationmayaffectgear performance. It further raises the question of the likely inconsiste
5、ncies of some of these inspectionresults and of inappropriate judgments of gear quality, even to the point of the rejection of otherwisesatisfactory gears. Finally, there are proposals for modifications to inspection software, possibly to someinspection routines, all to extending the benefits of the
6、 basic Elemental inspection process.Copyright 2008American Gear Manufacturers Association500 Montgomery Street, Suite 350Alexandria, Virginia, 22314October, 2008ISBN: 978-1-55589-945-53Extending the Benefits of Elemental Gear InspectionIrving LaskinIntroductionThe gear industry, here in the USA and
7、internation-ally, has adopted two systems of specifying andmeasuringtheaccuracyofgears. Oneiscompositeinspection, which recognizes that the inspectionmeasurements are the result of a combination ofaccuracy deviations, generally derived when thetestgearisengagedwithsomeformofmastergear.The other is g
8、enerally labeled as elemental inspec-tion, or sometimes as analytic inspection.Thisinspection system looks at individual elements ofgear accuracy or, at least, attempts to do so, all aspart of analyzing the complete nature of the gearsaccuracy. This paper deals only with the system ofelemental inspe
9、ction, especially as performed bymodern, computer controlled, inspection equip-ment. Not only is the inspection process, itself,computer controlled but also the processing andreporting of the inspection results. This makespossibleadditionaloralternatedataprocessingwithresults reported in a form that
10、 clarifies or enhancescurrent forms of inspection data reporting.AGMA documentsCurrent USA practice in elemental inspectionclosely follows a pair of AGMA documents. AGMA915-1-A02 1, describes a variety of tangentialmeasurements, many adopted for use in so-calledelemental inspection. The second, ANSI
11、/AGMA2015-1-A01 2, defines the elemental version of agear accuracy classification system based onselected tangential measurements described in theAGMA 915-1 document. This specification ofaccuracy classifications introduces tolerances foreach specified elemental measurement. Thesetolerances, in effe
12、ct, do determine whether the testgearmeetsspecificationsand,uponfailuretodoso,will lead to rejection and scrapping or rework. If thetest gear came from some kind of molding process,the rework could extend into the molding tool.The system of elemental gear accuracy defined in2 lists the following 8 i
13、tems, along with the generalinspection procedures which provide the measure-ment data for each. These will be reviewed ingroups later in this paper.1) Single pitch deviation2) Cumulative pitch deviation3) Profile, total4) Profile slope5) Profile form6) Helix, total7) Helix slope8) Helix formThere is
14、 another set of AGMA documents whichtreats composite inspection radial measurements(see34),butthesemeasurementsarenotdirect-ly connected to the elemental measurements dis-cussed here. There is one exception in 3, dealingwith the subject often called“hidden runout”treatedlater in this paper.Objective
15、sThe objectives of elemental inspection are three-fold:1) to compare the inspected gear to the gearaccuracy specifications, which may have beenstated individually or as a group by accuracyclass.2) to indicate what in the gear manufacturingprocess has caused the departure from idealgear geometry3) to
16、 help identify the effect of the elementalcondition on the performance of the gearIn meeting all these objectives, elemental inspec-tionhasdemonstrateditsbenefittothegeardesign-er, the gear manufacturer, and the user of the gear.The objective of this paper is to examine the ele-mental inspection pro
17、cess as defined by the refer-enced AGMA documents, to indicate where and inwhat way it may be improved in meeting the aboveobjectives,andtoproposechangesthatwillprovidesuch improvements. Success in this effort can berecognized as extending the benefits of elementalgear inspection.4DefinitionsAlthoug
18、h the following terms are used in the AGMAdocuments, their definitions are generally by infer-enceratherthanbyreadilylocatedstatements. Thedefinitions given here conform to the generalusageof the terms, inside and outside of gear technology.TheyaregenerallyinagreementwiththeiruseintheAGMA documents,
19、 with the possible exception ofthe use of elemental, as will be noted later.Deviation: the dimensional departure from the ge-ometry of the ideal gear, as defined, directly or byinference, in the gear specifications, including anydesign modifications, such as tip relief or facecrown, introduced by th
20、e gear designer.Elemental: anycomponentofgearaccuracywhichcannot befurther reducedto sub-componentsand,as such, may be present alone or in combinationwith other elemental components. Elemental com-ponentsareoftenassociatedwithasinglesourceinthe manufacturing process and have an individualeffect on g
21、ear performance.Composite: any gear accuracy designation apply-ing to a combination of elemental components.Gear manufactureGear accuracy definitions and measurements areindependent of the method of gear manufacture.However, each manufacturing method mayproduce its own set of typical accuracy deviat
22、ionsrequiring its own set of measurement procedures.For example, wide-faced gears made by thepowder metallurgy (P/M) process tend to have ahollow condition in its face width, namely, a smallerdiameter at its face center with a larger diameternear the end faces. This will be revealed in thehelical in
23、spection trace. Any inspection for profileshould then be made at the face-located largerdiameters because it is these gear sections whichwill interact with a non-crowned mating gear on theparallel shaft. It is the responsibility of the geardesigner to specify such inspection locations.Itis important
24、to recognizethat eachmanufacturingmethodmayalsobringitsown setof typicalaccura-cy deviations. In the measurement and dataanalysis methods in use for one specific accuracydeviation, another type of the accuracy deviationmay insert itself. This could result in a compositemeasurement rather than a true
25、 elementalmeasurement. Such a condition, in the form of“eccentricity of mounting”, is mentioned in 1,clause 7.5, and will be further discussed below.RunoutRunoutisappliedtoavarietyofindividualelementalaccuracy deviations, each associated in some waywith a varying distance of individual gear teeth, o
26、rtooth spaces, from the datum center of the gear.None of these individual elemental deviations arelisted directly as part of elemental inspectionrequirements but their presence often enters intoalmost all of the other inspection procedures. Theeffectofsomekindsofrunoutontheresultsoftheseinspection p
27、rocedures will be given extensive treat-ment in this paper as part of an effort at clarificationand enhancement of the coverage in the AGMAdocuments.Three types of runout will be discussed:1) eccentricity, in which the center of the ring ofgear teeth, which may itself be close to idealshape, is offs
28、et from the datum center of thegear, the one center used for all gear measure-ments and likely to be used in the mounting ofthe gear in its application.2) out-of-round, in which the shape of the ring ofgear teeth is notround, sothat evenif noeccen-tricityispresent,measurementsfromallteethortooth spa
29、ces would not be constant3) “hidden runout,a conditionin whicheccentricityin the gear was present in an earlier step ofmanufacture but, in some later step ofmanufacture,wasmodifiedsoasto“hide”oneofits negative features from some inspectionprocesses.EccentricityRegardless of the method of gear manufa
30、cture,there are likely to be cases in which the inspectedgear has some degree of eccentricity. Examples ofsome of these are given below:In the case of a machined gear, it is commonpracticeto locatethe gearblank byits datumcenterin the form of a through-hole. Manufacturing varia-tion in the size of t
31、he hole, especially when severalgears are stacked on a singe arbor matched to thesize of the smallest possible hole, will introduce ec-centricity between the holes datum axis and the5ring of machined gear, into at least some of thestacked gears.Theremaybeotherexamples. Ifthegearislocatedduring machi
32、ning through some type of tooling withbuilt-in eccentricity, such eccentricity in that toolingwill be transferred to the gear.If the gear is molded from plastic, there are its ownsources of eccentricity. In the mold, the core pinwhich produces the datum center hole may havebeen positioned eccentrica
33、lly to the portion of themold that forms the gear teeth. This is even morelikelyifthetwomoldfeaturesarelocated indifferentmoldsections,onefixedandtheothermovabledur-ing the molding machine operation. This would becommon for the molding of a compound helicalgear. As another example, even without mold
34、construction errors, the gears in a multiple cavitymold may experience different rates of coolingacross angular locations in its face, resulting indifferent local shrink rates and the introduction ofeccentricity.If the gear is made by the P/M process, somesignificant degree of eccentricity is inevit
35、able. Thecompaction tool, which gives the initial shape to thegear, contains punches which must slide axially rel-ative to each other. The core pin, which forms thedatum hole, is surrounded by a punch in the shapeof the gear, which must slide in the die which formsthe gear teeth. For gears with more
36、 complex fea-tures, there may be additional concentric punchesthat participate in the relative sliding. To permit thissliding, there must be some clearance between thevarious enveloping punches. Under the extremelyhigh compaction pressures, these punches arepushed to one side, forcing the core pin t
37、o becomeeccentric to the die, with the resulting gear teethbecoming eccentric to the datum center hole. TocorrectforeccentricityinhigherqualityP/Mgears,itis common to mold the center hole undersize andthen machine it to size in a later operation thatcorrects the eccentricity.The presence of eccentri
38、city may affect gearperformance in a number of ways:1) eccentricity introduces a once-per-rotationvarying center distance with the mating gear, asometimes critical deviation, especially in finepitch gears, resulting in:- variation in backlash between the involuteflanks of the mating gears.Note: Ecce
39、ntricity is sometimes, as inAGMA 2002 5, translated into a variation intooth thickness having the same effect onbacklash, leading to a so-called functionaltooth thickness. This substitution does notconsider the following two other variationconditions.- variation, during meshing, in tooth-tip-to-root
40、-circle clearance and variation in tooth-tip-to-fillet-curve clearance betweenmating gears.- variation, during meshing, in contact ratiobetween mating gears.2) eccentricity introduces a sinusoidal componentin the transmitted motion between the matinggears:- potentially leading to once-per-revolution
41、dynamic forces, especially if the gears arerotating at very high speed in a gear systemwith high mass-inertia components, a rarecondition which, even rarer, may impact thegear noise produced.- for gear systems which have a need forprecise position control at its output, as incomputer printers, where
42、 the transmittedsinusoidalcomponentmayleadtobandinginthe resulting printed image.The conditions listed under item 1 are typicallyresolved by design modifications to remove anycritically harmful effects on gear performance.Those listed under item 2 are rarely encounteredand only in special applicatio
43、ns. It is therefore safeto conclude that some limited eccentricity in amanufactured gear is, by itself, acceptable to thegear designer. What is not acceptable, as will beexplained below, is any misinterpretation of inspec-tion results brought on by the composite presenceof eccentricity.Analysis of i
44、nspection dataInspection data will be analyzed, as follows:1) description of the inspection process and thedata produced;2) identification of any composite components ateach inspection, examples of a source in gearmanufacture, and, most important, theirpotential effect on gear performance;3) proposa
45、ls for separating the composite compo-nents into true elementaldeviations, whetherby6changes to the inspection process or byadditions to the software that is to translate theinspection data to the preferred form.The selected inspection items will start with thoselistedaboveasassociatedwiththetoleran
46、cesofthegear classification system. It will then move toadditional gear accuracy conditions of potentialinterest to thegear designer,manufacturer, oruser.Pitch deviation, cumulative and singleThe measurements for these deviations are madebyindexingthegearaboutitsdatumaxisthroughanangle exactly equal
47、 to the pitch angle (360 degreesdivided by the number of teeth). See figure 1. Ateach indexed position, the measuring probe isinserted in a radial direction to the tolerance or in-spection diameter, providing a probe deflectionreading at the tooth flank. The deflection measure-ment is always made at
48、 the same diameter whichwill be centered on the datum axis. This process isthen repeated for the opposite set of flanks. Thesereadings, typically referenced to the first reading ontooth number one, are then plotted in a bar chart orstepped line, as shown in figure 2. The full range ofreading values
49、for each flank is labeled thecumulative pitch deviation and the single largest in-terval between successive readings is labeled thesingle pitch deviation.Figure 1. Measurement of pitch (or index) deviationsFigure 2. Pitch (or index) measurements on a gear with eccentricity7Thefigureshowstheplotforagearwitheccentricityas its dominant accuracy deviation. For eachflank,theplotfollowsaonce-per-gear-revolutionmathe-matical sine curve. The amplitudes of the two sinecurves are similar but the apparent phase locationsare noticeably diff
