《AGMA-02FTM9-2002.pdf》由会员分享，可在线阅读，更多相关《AGMA-02FTM9-2002.pdf（10页珍藏版）》请在三一文库上搜索。
1、02FTM9 Gear RollScan for High Speed Gear Measurement by: A. Pommer, FRENCO GmbH TECHNICAL PAPER American Gear Manufacturers Association Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18
2、/2007 12:15:02 MDTNo reproduction or networking permitted without license from IHS -,-,- Gear RollScan for High Speed Gear MeasurementA Revolutionary New Method for a Complete Topographical Inspection of Gears A. Pommer, FRENCO GmbH Thestatementsandopinionscontainedhereinarethoseoftheauthorandshould
3、notbeconstruedasanofficialactionor opinion of the American Gear Manufacturers Association. Inmoderngearproduction,simpleinspectionsystemsforquickandclosebymeasurementshelptoproducehighquality. The Gear RollScan method inspects all flanks of helical gears within about 1 minute. ThegearRollScansystemi
4、ssimilartotheone- -flankgearrolling inspection. Themaster gear,however, hasmeasuring tracks on selected flanks. All other areas of the master gear are clear to create a contact ratio below 1. Thus, a clear allocationbetweentheangleofrotationandthecontactpointcanbemade. Asuitablenumberofteethratiobet
5、weenthe workpieceandthemasterwheelischosen. Everymeasuringtrackonthemasterwheelischosen. Everymeasuringtrack onthemasterwheelwillcontacteveryflankofthespecimenafteraspecifiednumberofrotations. Twomastergearsare in roll contact with the specimen simultaneously. One master gear inspects all left hand
6、flanks, the other all right hand flanks. Thetopographicalmeasurementofallteethrequiresthedeploymentofstatisticalmethodsforevaluatingthedeviations. New parameters have to be derived, describing features of the gear which are critical for function, such as the noise excitation, as function of the topo
7、graphy. With regard to the tolerances, this measuring method has wide- -ranging consequences. Whereas a standard evaluation considers 4 teeth, with the effect that a gears worst tooth is measured more or less by coincidence, the RollScan measuring device always will find the worst tooth Copyright 20
8、02 American Gear Manufacturers Association 1500 King Street, Suite 201 Alexandria, Virginia, 22314 October, 2002 ISBN: 1- -55589- -809- -2 Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04
9、/18/2007 12:15:02 MDTNo reproduction or networking permitted without license from IHS -,-,- 1 Gear RollScan for high speed gear measurement. A revolutionary new method for a complete topographical inspection of gears Andreas Pommer , FRENCO GmbH Rolltester with component between two master gears 1.
10、History Of Development: The idea for a newgear testingprocedure wasborn inCologne.Dr.Guido Hopffrom Fordmade thefirst contact with FRENCO about the development of such a gear tester in 1995. In the following years FRENCO built thefirst fixturesand suitable electronics for fast capturing of measuring
11、 points. The software was written in Cologne. In 1999 the first gauges were installed on the shop floor. The process is protected by two patents 1. The Ford team was honored with the ,Henry Ford technical achievement award 1999. Since 2001 FRENCO is holding the license to bring this technique to the
12、 market. FRENCO is now responsiblefordevelopment,designand production of the complete system. The large potentialofthissystemstillcreatesdemandfornew ideas and solutions. 2. Existing Gear Testing Processes This article is about a new gear testing process, calledRollScanning.Itisusefultoconsider comm
13、on methods first. Existing gear testers could be divided into two groups: Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 12:15:02 MDTNo reproduction or networking permitted with
14、out license from IHS -,-,- 2 Classification of gear testers existingexistingNEW type of contactpoint contactline contactshort line contact examplecoordinatemeasuring machines single or double flank roll tester RollScanning elements of contactball probegearscontact by small tracks on gears precisionv
15、ery highlimited by master gear quality high,duetodiscreet contact areas densitylow, normally only 4teeth high, complete surface is covered high, all areas and all teeth are covered speedslowvery highhigh, 1 2 minutes contacttheoreticallyideal surface contact realistic,gear-to-gear contact realistic,
16、gear-to-gear contact flexibilityvery flexibleindividual master gears required individual master gears required resultsindependentinspection of many gear errors sumerrorinspection withoutaccessto independent errors independentinspection of many gear errors 3. The Process of RollScanning: Single Flank
17、 Roll Test RollScanningisbasedontheknownsingleflankroll test. Main difference to the well known double flank roll test is the use of a fixed center distance. With one driving and the other gear being driven a single flank contact is achieved. The deviations of the sample will be found as an angular
18、error between both axes. So two rotary encoders are necessary one for each gear. As this method is a realistic copy of the real transmission a large contact ratio is typical.Thismeanstheresultsshowaninterference of a number of tooth flanks. For helical gears a contact ratio up to 3 is typical. plane
19、 of action Typical single flank test. The section in the ,plane of action shows contact between two gears along 3 lines. Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 12:15:02
20、MDTNo reproduction or networking permitted without license from IHS -,-,- 3 Solution As the main solution the contact ratio needs to be reduced. This is achieved with so called tracks on thegearsurface.Toavoidinterferencebetweentwo tracks,theyare separatedby anumber of ,transport teeth between every
21、 track-flank. Two different types of tracks are used: Lead Track Profile Track The surface beside the tracks is lowered about 0.2mm to always keep contact on the tracks. Also all ,separation teeth are lowered by the same amount on their complete surface. As a result we have now only a short part of
22、the complete ,line of contact left over. It is nearly a point contact now. plane of action with leadtrack: contact is reduced to a short line at one flank The output signal of a single flank test is usually the angular transmission error. If we set contact ratio below 1.0 in some areas tooth flank c
23、ontact is lost. These zones are part of the design and will be ignored by the software. transmission error rotation Raw signal for calibration and for a component; every track takes 12 points for each flank, an active zone is separated by an unused garbage zone because contact ratio is smaller than
24、1.0. Because we know the location of the tracks we can calculate the exact position of the contact point all thetime.Everydetecteddeviationcouldbelocated. So it is possible toscan thegear surface.Therefore we call this process RollScanning. To cover a flank not only in two sections many different tr
25、acks are used. A typical rollscan master is designed with about 7 lead tracks covering the whole flank from majordia.downtoformdiameter.Inaddition,upto3 profile tracks are used to check involute profile in three sections. After one rotation every tooth is checked by a different track, so only one se
26、ction is available. The master gear is usually designed with a different number of teeth than the component. So different flanks will touch for every new rotation. Increasingthenumberofrotationswillresultinmore tracks per flank. If the number of rotations is larger than the number of teeth of the co
27、mponent, all flanks are covered by all tracks. The result is a complete grid of points for all flanks. Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 12:15:02 MDTNo reproduction
28、 or networking permitted without license from IHS -,-,- 4 lead track points profile track points To check both sides of a gear the same time with single flank testing twomaster gearsare used.One forleftandoneforrightflank2.Allgearsaredriven by one motor . Brakes force the gears to stay in direct con
29、tact all the time. At the moment the number of points is limited, finally about 370 points for each flank are stored. The whole process takes less than two minutes, for a small gear about one minute. Remember this will include the check for all teeth. 4. Multiple Inspection Reports The storage of al
30、l points allows the visualization of many different types of inspection reports. First of all the classic report with involute and lead for 4 teethispossible.Nearlyallgearmakersare usedto work with these reports. Torespect habitsof ourusers weshow layoutsvery similartothosefromCMMs.Ofcoursetheuserca
31、n choose which teeth he wants to see in his report. But this will use only 1% of the stored information simultaneously. So FRENCO, together with our customers, have created totally different layouts for reports. Copyright American Gear Manufacturers Association Provided by IHS under license with AGM
32、A Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 12:15:02 MDTNo reproduction or networking permitted without license from IHS -,-,- 5 Themainobjectivewastobringmoreinformationto the user. It is possible to show all profile or all lead graphsbesideeachothertofindthemo
33、st interestingflanksasfastaspossible.Itispossibleto calculate a single flank test result. It is possible to choose lead and involute graphs by ,walking down to index error graphics. Statistical Review The increased number of data creates a problem. The operator will need more time to check all resul
34、ts. As an example, for a gear with 35 teeth we have 70 profiles, each with up to four criteria like sumerror(F),formerror(ff),directionerror(fH) and crowning (c). The same amount is available for lead. Also for spacing we find some values per tooth. Our solution is to use statistical methods to supp
35、orttheoperators.Foreveryvaluethe distribution is shown. Now the attention is reduced to the mean and the maximum values. We call this sheet TOLANY report. The mean value is shown by the vertical line. If some values for a criteria are outside their tolerance the graph will becomeyellow.Ifthemeanvalu
36、eisoutsideitwillbe red. This distribution will also form the base for an automatic assessment. If the gauges are part of a connected production chain, no operator could be asked to have a look at the results every minute. The system has to decide by itself. From other systems with a high density we
37、learned to see tolerance limits under a new light. If you take very more and more points it is no question to findpoints outside the limit. Statistical methods must be used to come to a decision. From this sheet we can learn something else. To have values from all tooth is a very big advantage. Beca
38、use now we can be sure to find the extreme values. By checking only 4 teeth we create an uncertainty. Uncertainties are not what we want in a measuring system. We find a significant variation for the mean value and for the range if we use only 4 of 33 measurements to calculate our results. Even if w
39、e take these 4 measurements with the highest precision, this uncertainty can not be avoided. Systems performing complete checks will show lower uncertainty. Uncertainties RollScanning is a highly precise inspection system for the shop floor. The inspection of the involute profile depends on a succes
40、sful compensation of the master gear errors.Larger deviations of the component can lead to decreasing precision. To compare absolute results you need a reference like a CMM. This can easily create differences of index and runout due to different clamping conditions. If you ensure comparable clamping
41、, uncertainty is 3 micronsorbetter.Repeatingthesame measurement is normally better than 1 micron. The reason is a rigid design and the simple kinematic process with only three rotating axes. Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employ
42、ees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 12:15:02 MDTNo reproduction or networking permitted without license from IHS -,-,- 6 -0,02 -0,01 0 0,01 0,02 1471013161922252831 value mean (t4) 4 teeth check vs all teeth check example T=33 1 2 3 4 Measurement: -0.006 to +0.020 mean value
43、 complete : 0.0068mean value (T=4): 0.0033 to 0.009 range complete : 0.026range (T=4):0.014 to 0.024 5. Topography Themostinterestingandmostdifferentreportisthe so called topography. Only here it is possible to show 100% of the information. And only here you could be sure to find all the truth about
44、 your gears surface. Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 12:15:02 MDTNo reproduction or networking permitted without license from IHS -,-,- 7 The picture contains a 3
45、-dimensional grid to visualize the surface. Colored dots mark all points withina5micronrangetothehighestpointtoforma contactpattern.Acoloredlineshowsthepathofthe contact line along the flank, highlighted sections on this line accentuate the part where this tooth will have contact, while for the rest
46、 of the line the tooth beside is in contact. The vertical position of the whole grid shows the influence of spacing and runoutonthedisplayedtooth.Thisthree dimensional graphic is very easy to understand evenit containssomuchinformation.Both,experts and beginners have learned to work with it soon. Ex
47、periences With Topographic Inspections. A topographic inspection is very useful which we learned from our customer Volkswagon in Kassel. ,Whydidthisgearfail?Askthisquestiontoyourself if you look at the classic report below. You will not find the reason here, because the error was not detected by the
48、 CMM machines. But the assembled gearbox did not pass the final noise check.Thereasonisobviousifyoutakealookatthe topography. The final grinding process left over materialinonecorneroftheflank. RollScanningwill find such errors always, also if they are at one tooth only. 6. Conclusion Applications for RollScanning: The system depends on special master gears. These gears could be replaced fast to adjust the gauge for different components. But they are designed for one component only. For economical reasons