1、AGMA INFORMATION SHEET(This Information Sheet is NOT an AGMA Standard)AGMA935-A05AGMA 935-A05AMERICAN GEAR MANUFACTURERS ASSOCIATIONRecommendations Relative to theEvaluation of Radial Composite GearDouble Flank TestersiiRecommendations Relative to the Evaluation of Radial Composite GearDouble Flank
2、TestersAGMA 935-A05CAUTION NOTICE: AGMA technical publications are subject to constant improvement,revision, or withdrawal as dictated by experience. Any person who refers to any AGMAtechnical publication should be sure that the publication is the latest available from theAssociation on the subject
3、matter.Tables or other self-supporting sections may be referenced. Citations should read: SeeAGMA 935-A05, Recommendations Relative to the Evaluation of Radial Composite GearDouble Flank Testers, published by the American Gear Manufacturers Association, 500Montgomery Street, Suite 350, Alexandria, V
4、irginia 22314, http:/www.agma.org.Approved October 16, 2005ABSTRACTThe condition and alignment of gear measuring instruments can greatly influence the measurement of productgears. This information sheet provides guidelines for the alignment of double flank tester elements such ascenters, ways, probe
5、 systems, etc. It also covers the application of artifacts to determine instrument accuracy.Published byAmerican Gear Manufacturers Association500 Montgomery Street, Suite 350, Alexandria, Virginia 22314Copyright 2005 by American Gear Manufacturers AssociationAll rights reserved.No part of this publ
6、ication may be reproduced in any form, in an electronicretrieval system or otherwise, without prior written permission of the publisher.Printed in the United States of AmericaISBN: 1-55589-872-6AmericanNationalStandardAGMA 935-A05AMERICAN GEAR MANUFACTURERS ASSOCIATIONiii AGMA 2005 - All rights rese
7、rvedContentsForeword iv.1 Scope 1.2 References 13 Definition of terms 14 Instrument environment 25 Measurement system condition 3.6Artifacts 7.7 Statistical process control 10.8 Instrument fitness for use 10.9 Measurement process (instrument correlation) 10.Bibliography 11.Figures1 Cross-axis alignm
8、ent of test arbors 3.2 Parallel alignment of test arbors 4.3 Lost motion diagram 44 Apparent size changes as a function of load change 5.5 Relation between the zeroing setup and the part setup 5.6 Testing frequency response at two speeds 6.7 Eccentric discs 7.8 Concentric disc with flats 9.Tables1 S
9、ymbols 1.AGMA 935-A05 AMERICAN GEAR MANUFACTURERS ASSOCIATIONiv AGMA 2005 - All rights reservedForewordThe foreword, footnotes and annexes, if any, in this document are provided forinformational purposes only and are not to be construed as a part of AGMA 935-A05,Recommendations Relative to the Evalu
10、ation of Radial Composite Gear Double FlankTesters.Between 1994 and 1998, AGMA published three standards on calibration of gearmeasuring instruments: ANSI/AGMA 2010-A94, Measuring Instrument Calibration Part I,Involute Measurement, ANSI/AGMA 2113-A97, Measuring Instrument Calibration, GearTooth Alig
11、nment Measurement, and ANSI/AGMA 2114-A98, Measuring InstrumentCalibration, Gear Pitch and Runout Measurements. These standards covered elementalmeasurements specified in the accuracy requirements of ANSI/AGMA 2015-1-A01,Accuracy Classification System - Tangential Measurements for Cylindrical Gears.
12、TheCalibration Committee decided that supplemental information, on measurement systemconditions for calibration, accuracy requirements and uncertainty determination, wasdesirable to have in an Information Sheet, AGMA 931-A02, Calibration of Gear MeasuringInstruments and Their Application to the Insp
13、ection of Product Gears that was published in2002.The material in these AGMA documents were combined and submitted to ISO for thedevelopment of ISO 18653:2003, Gears - Evaluation of instruments for the measurementof gears, and ISO/TR 10064-5:2005, Cylindrical gears - Code of inspection practice - Pa
14、rt5: Recommendations relative to evaluation of gear measuring instruments.The Calibration Committee decided that the similar standardization and information wasneeded for the evaluation methods of double flank testers used for (radial) compositemeasurement of gears. After a study of existing practic
15、es, standards, and literature theinformation contained herein is a consolidation of the most common practices currently inexistence.The first draft of AGMA 935-A05 was made in August, 2003. It was approved by the AGMATechnical Division Executive Committee in October, 2005.Suggestions for improvement
16、 of this information sheet will be welcome. They should besent to the American Gear Manufacturers Association, 500 Montgomery Street, Suite 350,Alexandria, Virginia 22314.AGMA 935-A05AMERICAN GEAR MANUFACTURERS ASSOCIATIONv AGMA 2005 - All rights reservedPERSONNEL of the AGMA Calibration CommitteeCh
17、airman: Robert E. Smith R.E. Smith - flow rate, distribution and velocity of the cooling(heating) medium;- frequency and amplitude of temperature varia-tions of the cooling (heating) medium;- temperature gradients within the cooling (heat-ing) medium;- vibrations;- electrical power supply quality.4.
18、1.2 Practical guidelinesThe following are practical guidelines for gearmeasurement. However, compliance with theseguidelines does not guarantee measurements to aspecific accuracy.- Artifact temperature. Tooling and artifacts shouldbe left for an adequate period to stabilize to ambi-ent temperature.
19、Artifact temperature ideallyshould be the temperature at which it was cali-brated.- Temperature variation. The instrument manufac-turers temperature variation guidelines for thedesired accuracy should be consulted. If this in-formation is not available, a default mean tem-perature should not change
20、more than 1C perhour, with a maximum change of 3.5 degrees perday.- Temperature cycles. The temperature may cycle+or-2C , centered on the mean temperature,every 5 minutes or faster. The thermal inertia ofmost mechanical systems will allow for rapid cy-clic temperature undulations within these guide-
21、lines for the stated accuracy.If the temperature cycles approach 15 minutes,serious effects on the measuring system areusually noted.Many people use an air conditioner in an attemptto achieve thermal control. The temperaturesensors in these units may be very slow torespond to temperature changes. If
22、 the re-sponse is slower than 5 minutes, serious effectson measurement accuracy may be noted.- Temperature gradient. The temperature shouldbe within 0.5C over the entire area of the instru-ment surface. The best way to do this is with ahigh air flow. Air flow must be uniform throughoutthe room to pr
23、event dead spots and prevent gra-dients. To accomplish this, diffuse the air comingin to the room and if possible design multiple airreturns to further diffuse the air uniformly in theroom. The goal is to have all air moving uniformlyin the room and at the same temperature. Mov-ing air must remove h
24、eat from electronic controls,computers, motors, hydraulics, people, lights,etc., to prevent gradients.- Vibrations. Vibrations caused by instrumentmovements should not be allowed to interferewith measurements.Also, vibrations from the surrounding environ-ment should be observed or measured. If theya
25、re affecting instrument accuracy, vibrationisola-tion of the instrument or a suitable foundationmay be necessary.- Electrical power supply. Power fluctuation maycause some electronic instruments and comput-er numerical control positioning systems to mal-function._2)A more thorough discussion of the
26、effects may be found in such standards as ANSI B89.6.2.AGMA 935-A05AMERICAN GEAR MANUFACTURERS ASSOCIATION3 AGMA 2005 - All rights reserved4.1.3 Shop environmentMany measuring instruments are placed in a shopenvironment where it is difficult to maintain anaccuracy within 5 micrometers. Accumulation
27、of dirtor other contaminants on the ways of the instrumentcan cause inaccuracies as well as premature wear.If an instrument must be used in this kind ofenvironment, care must be taken to avoid certainconditions, such as:- local radiant heat sources such as space heatersor sunlight through nearby win
28、dows that may dis-tort the instrument;- roof vents that allow cold air to drop on the instru-ment;- cooling systems or open windows that cause adraft to hit one side of the instrument.4.2 Effect of temperatureTemperature can have a significant effect on thegeometry of gears, such as effects upon inv
29、oluteprofile slope, helix slope, and tooth thickness.Calculations assume uniform temperature of thegiven test piece; localized temperature variationscannot be conveniently modeled.Although these parameters are not being measureddirectly by the methods used in this document, theywill have an effect o
30、n the results of double flankcomposite measurements. This is especially impor-tant in the plastics industry, where double flankcomposite measurement is used almost exclusively.Plastics have a very high rate of thermal coefficientof expansion. When gears are removed from themolding machine, they may
31、be at a very elevatedtemperature. Therefore, it is important to have theplastic molded gear stabilized to room temperaturebefore measurements are taken.5 Measurement system condition5.1 Axis-alignment check for rolling fixtureThe axes of the product gear and the master gearmust be parallel when chec
32、king spur and helicalgears. The effects of misalignment can be ex-pressed in terms of the vector components in thepitch and the axial planes.The component in the pitch plane (cross axis) willaffect the size reading and apparent functional tooththickness, and is a cause of end contact pattern, alsokn
33、own as cross bearing or end tooth contact.Apparent functional tooth thickness is the tooththickness that a product gear seems to have wheninspected by a master gear of a given tooth thicknesson a specific gear-rolling fixture. The differencebetween the functional tooth thickness of a gear andits app
34、arent functional tooth thickness is affected bymisalignment of the axes of the gear-rolling fixture,and the active face width of the mesh.P=sa sfFa(1)wherePis misalignment in the pitch plane (crossaxis), mm/mm;Fais active face width (width of the narrowermember in the mesh), mm;sais apparent functio
35、nal tooth thickness, mm;sfis functional tooth thickness, mm.It should be noted that the apparent functional tooththickness includes an error due to cross axismisalignment. The alignment in the pitch plane canbe conveniently checked with an alignment fixturesimilar to that shown in figure 1. The vari
36、ation of highpoint readings, as the alignment fixture is movedacross the arbors, should not exceed a value equalto 0.036 times the ratio of the testing radius toleranceto the face width of the product gear.Test arborsDirection of alignmentfixture movementAlignmentfixtureFigure 1 - Cross-axis alignme
37、nt of test arborsAGMA 935-A05 AMERICAN GEAR MANUFACTURERS ASSOCIATION4 AGMA 2005 - All rights reservedThe component in the axial plane (non-parallelism)will affect the center distance setting on the rollingfixture, and is a cause of end contact.a=CLorPtanwhichever is smaller (2)whereais misalignment
38、 in the axial plane (non-parallelism), mm/mm;L is active length of set-up parts (arbors, pitchdiscs, blocks, etc.), mm;C is allowable center distance variation, mm; is pressure angle, degrees.The axial alignment should be checked with agage-block stack. Test arbors of uniform diametershould be place
39、d in the master gear and in theproduct gear axes.The gage-block stack should bemoved axially alongthe arbors and the fixture indicator reading limitsnoted. The variation should not exceed one-twen-tieth (1/20th) of the total composite tolerance (seefigure 2).Test arborsGage blocksDirection of gagebl
40、ock movementFigure 2 - Parallel alignment of test arbors5.2 DiscriminationGood gaging practice requires that the discrimina-tion level of the indicating or recording equipment onthe gear rolling fixture, should be one-tenth (1/10th)of the smallest quantity to be measured (centerdistance tolerance, t
41、otal composite tolerance ortooth-to-tooth composite tolerance).5.3 Lost motionA disc, with calibrated flats, should be run against afixed disc or mandrel. This will introduce smallamounts of bidirectional motion of the measuringslide and probe. A recording of this test will look likefigure 3. The gr
42、aph should show a drop of an amountequal to the calibrated flat depth and return to theoriginal starting level.Lost motionCalibrated displacementProper return(no lost motion)Improper return(lost motion)Figure 3 - Lost motion diagramImproper return, or lost motion, is an indication ofmechanical probl
43、ems in the slide, moving headmechanism, or probe. This will affect the accuracy ofmeasurements, and should be corrected beforeproceeding with instrument calibration.5.4 Load correctionCenter distance measurements are sensitive to loadvariations. Decreasing the load without re-zeroingthe indicator wi
44、ll cause an apparent increase in theindicated gear size. This condition is shown in figure4. In practice, it is not possible to obtain a reading, ator close to zero load, but, since the deflection versusload characteristic is quite linear, it is possible todetermine the no-load value by plotting a s
45、eries ofreadings, and projecting the curve back to zero load.It should be noted that the condition shown in figure 4exists when the roll tester is zeroed to the set-upgage (see figure 5). Thus, for any particularchecking load, it will be observed that zeroing theindicator to the set-up gage will res
46、ult in a zeroingerror equal to an amount Z, which can be correctedfor by offsetting the indicator to a minus Z. Then, itfollows that the true size of the gear will be largerthan indicated by an amount P, which can becorrected by setting the indicator to a plus P. In short,setting the indicator to a
47、value P-Z,withtheset-upgage in place, will result in a true size indication.Experienced inspectors usually know whether or notit is necessary to determine the load correction; i.e.,AGMA 935-A05AMERICAN GEAR MANUFACTURERS ASSOCIATION5 AGMA 2005 - All rights reservedwhen the deflection curves are para
48、llel, as they oftenare, the load correction can be neglected.No load reading2ndmeasurement1stmeasurementLoadIndicationFigure 4 - Apparent size changes as a functionof load changePart set-upZoning set-upLoadChecking loadIndicationZPFigure 5 - Relation between the zeroing set-upand the part set-up5.5
49、Temperature correctionIf all temperature coefficients for gage blocks,arbors, pitch discs, master gear, and product gearare equal, and if all dimensional reference tempera-tures are the same, usually 20C, there is no need fora temperature correction; and conversely, a temper-ature correction is needed if these conditions are notmet. The essential idea of a temperature correctionis to compute the effect of the ambient temperatureupon the set-up parts such as ga
