lxg表面粗糙度_图文

admin管理员组

文章数量:1530845

2024年2月5日发(作者：)

IntJAdvManufTechnol(2000)16:668–674©2000SpyDepartmentofMechanicalEngineering,UniversityofVictoria,Victoria,CanadaAnon-contactroughnesssensorisdescribedthatissuitedforintegrationwithacomputer-controlledcoordinatemeasuringmachine(CMM).Thesensoremploysafibreopticinterfer-ometer,binationofthesensorandcomputercontrolledCMMallowssurfacetextureassessmenttobemadeduringscheduleddimensionalinspectionsofcomplexcurvedsurfacecomponents,sorsystemwillmeasuresurfaceroughnessparameters,forexampleRa,usingamethodthatreflhtweightsensorheadcanbemountedonatouchprobearmandtheassociatedarticu-latedmountinghead;thissuitableforautomatedsurfacefisoranditscontrolunitareintegratedwiththeCMMcontrolleranditsoperatiods:Automatedinspection;Coordinatemeasuringmachine;Fibreopticsensor;uctionThispaperdescribesanopticalsurfacetexturesensorthatisintegratedwithacomputernumericallycontrolled(CNC)coordinatemeasuringmachine(CMM).Thecombinationofthetwotechnologiespermitsthemeasurementofbothdimen-sionalandroughnessmetricsonapartplacedonthedeckoftheCMM.1.1LiteratureReviewThedemandforincorporatingsensortechnologyintotheproductionenvironmentisbeingdrivenbythesimultaneousneedtominimisemanufacturingcostswhilemaintainingahighstandardofquality[1].Inparticular,newsurfacetextureCorrespondenceandoffprintrequeststo:y,DepartmentofMechanicalEngineering,UniversityofVictoria,POBox3055,VictoriaBC,V8W3P6,Canada.E-mail:cbrȰorshavepredominantlybeennon-contact,employingoptical,ultrasonic,andcapacitancemethods[2–5].Thesensorstypicallymeasureacommonsurfaceroughnessparameter,rkpresentsanopticaltexturesensor,employingfibreoptics,thatisphysusresearchonusingfibreopticsinsurfaceroughnessmeasurementwasperformedbySpurgeonandSlater[6]andbyLinetal.[7].Bothusedabundleofopticalfibrestodeliverlighttothesurfaceandalsotocollecttherefllationwasfoundbetweentheintensityofthereflectedlight(asmeasuredbythephotodetector)r,bothtechniquessufferedfromchangesinthereflectedlightintensityowingtovaryingreflsurementsalsolackedsufficientsensitivityandwereonlysuitableformeasurementsonsmoothsurfacesuptoRa=0.5␮ndAgarwal[8]circumventedbothoftheseproblemsbyusingtwofibreopticbundles,ioofthetworeflectedlightintensitiesremovedtheproblemofsurfacerefltrumentshowedgoodcorrelationwithstylusmeasurements,uptoRa=1␮fibreopticsensorsdescribedabovedonotprovidesurfaceprofiledata;theysimpfibreopticsensordescribedbelowalleviatestheproblemsstatedabove.1.2BackgroundThereisarangeofhigh-value,geometricallycomplex,anddimensionample,turbinebladeassemblies,machinedonmulti-axisCNCmachinetools,mustmeetdimensionalandsurfacefinishspecificationsbeforetheyareapprovedandincorporatedintotheficeroughnessspecification(usuallyRaorRz)ismandatory,owingtotheeffectthatsurfacetexturehasontheefficiencyoftheairflample,Fig.1showsaturbcontrollerusesthe

Fig.1.(a)Shadedimageofturbineassembly;bladesplushub.(b)CADmoogramtomovethetouchprobearoundthfigureillustratesthegeometriccomplexityofthepart;eachbladecanhaveseveralmachiningpatches,correspondingtodifferentcuttingtoolorientations,ponentisthenremovedfromtheCMMtableandaseqfaceroughnessismeasuredusingamanuallyopctlocationofthemeasurementisnotcriticaltotheresult;how-ever,tionefficiencyforthistypeofcomponentcouldbeenhancedbyautomatingceroughnesssensorsystemthatcanbeintegratedwithaCMMAutomatedSurfaceRoughnessMeasurement669wouldbedesirableandshouldhavethefollowingcharacter-istics:Thesensorheadmustbephysicallycompacsorheadmustbelightweightandsuitableforattach-mentonaCMMprobearticulatinghead(hawPH10probehead).Thesensor’soperatingparametersmustbecompatiblewithastylusprofilometerandhaveameasurementrangeof0.10ϽRaϽ20witha0.10␮ponentmustremainfixedinthesameposition,ontheCMM,ghnesssensorprobehead,dataprocessingsoftionofthesensorheadmOperationThedetailsconcerningtheoperationofthesensorheadasaninterferometriccavityhavebeenpreviouslyreported[9,10].Theessentialcomponentoftheoverallsensoristhehead,illustratedinFig.2,whichiscomprisedofasinglemodefibreattachedtoacylindricalgraded-index(GRIN)sfocusestheincominglightwave(␸),fromthefibre,ntfaceofthelensiscoatedwithapartiallyreflectivematerialanddividestheincominglightwaveintotwocomponents:1,having10%oftheoriginalintensity,thatisreflectedfromthelensfrontfaceandwhichtravelsbackdownthefi2,thetransmittedportionthatisfocusedontothetargetsurfaceandthencollectedbytheGRINlensandre-focusedbackdownthefierveningsurfaceprofiledifference,(d2−d1),ismeas-uredfromthephasedifferenceofthetworefl-awayviewofthesensorheaddetailsshowingmountingblockandGRINlens.

etesurfacetextureprofileobtainedbsor,measuringasurfaceoverasamplinglengthSbetweenthestartpositionAandtheendpositionAЈ,isedsurfaceprofileisgeneratedthatconsistsofasetofsurfaceprofilesamples{(xi,zi):i=1,%,N}.ThefisorspeedcanbeaccuratelysetbytheCMMcontrollerthroughtherange0.1–1.0mmsϪ1;nsordatasamplingrateof1kHzandasamplinglengthof1.4mm,alargerdatasetthanrequired,therefore,thecontrolsoftwarepermitsdatadecimation(byfactorsof10or100)ectioationshipbetweenrelativephaseshift,laserwavelengthandpathlengthdifferenceisgivenbyEq.(1)figureshowsthesensoracquiringtwoconsecutivemeasurementsofsurfaceprofile,atPoint1andPoint2,n,thesurfaceheightchangebetweenthelocationsis(d2−d1).Itisassumedinthisexamplethat(d2−d1)islessthanhalfofthewavelengthofthelaser,or0.4␮ngeinopticalphaseshiftbetweenthetwopositionsis:⌬␾=(2␲␭−1)2(d2−d1)(1)where,(d2−d1)=verticaldistancebetweenthepositionsofpoints1and2⌬␾=phasethelensdifferencecoatingbetweenandthethesurfacewavesreflectedfrom␭=wavelengthofthelaser(800nmor0.8␮m)Theinterferenceintensityversusopticalphaseshiftchange(Eq.(1))iationofintensity,I,forthedisplacement(d2−d1)inationofsurfaceheightfromthesensor’sintensityprofilefortwopointslessthan0.85␮itionedasshownontheintensityfunction,thenthseshiftmeasuredbythesensorelec-tronics,⌬␸1,ore,phaseshiftorsurfaceheightvariationcanbeaccuratelydeterminedprovidedthetotaldistancechange(fromlenstosurfaceandbacktolens)doesnotexceed0.4␮lueisobtainedbyrearrangingEq.(1)andusingthelaserwavelengthof␭=0.8␮imumintensity(Imax)isattainedwhenthereiszeroor2␲retheso-calringofI,betweenthebrightnessfringes(fringe0,fringe1,fringe2,etc.)allowsthedeterminationofsurfaceprofisiderthesituationillustratedinFig.5,wherethesensormovesoverthesurface,frompoint1topoint2,andthesurfaceheightvariesbymorethan0.4␮mmo-datelargersurfaceheightvariations,thesensortracksthenumberofbrightnessfringesthatpassagivenreferencepoint,denotedbyDinthefiore,asthesensormovesfrompoint1topoint2,throughaverticaldropof(d2−d1)=0.85␮m,thesensortracks2fringesor0.80␮ncetoFig.5showsthat“fringe1”and“fringe2”hctronicsensorcountsthefringesthatn,theremainingphase⌬␸2is

nge-countingmethodfordeterminingtheverticalheightbetweentwopointsgreaterthan0.85␮redonthephotodetectorbytheincreaseinintensityfromIЈ1toIЈasurementcorrespondstotheremaining0.05␮mofverticalprofiingthefringetrackingmethod,ore,thesensorprovidestwooutputsignalstothedataacquisitionsystemandsupervisorycontrolsoftware:ingesignalcapableofresolvingsurfacedetailtoonehundredthofaninterferencefringe,ecountisignalsareprocessedbythesoftwaretogenerateasurfaceprofileofthepart,overthesamplinglength,fromwhichsurfaceroughnessamplitudeparameters,ationoftheSensorSystemwithaCMMThemajorintegrationissuesinherentincombiningthefibreopticsensorwiththeCMM(MitutoyoBH10M)arehighlightedinFig.6,whichillustratesthemaincomponentsandthedataflowinter-connections.3.1MountingtheSensorHeadontheCMMThemountingofthesensorheadontheCMM,tsicaldimensionsofthesensorprobeheadare:ensionfitstheprobe-articulatingheadatoneend,binationofsensorandarticulatingheadallowsfliculatingheadallowspositioningin7.5°l,pabilityiscrucialtosuccessfuloperationofthefisofthesensor’sGRINlensmustbemaintainedataperpendiculabjectisacompoundcurvedsurface,sitionofthearticul-atingheadisqualifiedduringthemeasurementprficationofthesensorprobepositionisastandardpro-cedurewhenusinganarticulatingprobeheadonacomputercontrolledCMM.3.2SynchronisationoftheCMMandSensorSystemThecommencementofasurfaceroughnessdataacquisitionscanissynchronisedwiththemotionoftheCMM’chronisationmethodisoutlinedbelowandillustratedinFig.8:TheCMMinspectionprogrampositionsthesensorheadatameasurementlocationabovetheobject’iculatedhead’stwoangularpositionsareadjustedtopositionthesensorheadperpendiculartothesurface,sorheadismovedalongtheevaluationlength(adistanceofapproximately5.00mm)thescan,thepositionfeedbacksensors,oneachaxisoftheCMM,provideaxialpositiondatatotheCMMcontrolunit(seeFig.8).raphofthesensor(mountedontheCMMarm)withthecontrolunitandpersonalcomputerdatainterface.

typrofileoftheCMM,inoneaxis,andthecorrespbleasana-quad-bsignal(digitalpulsesignalsspecifyingmotormovementanddirection)whereeachpulsecorrespondstothesmallestdistanceincrementthattheCMMcanmove(forexample,1␮m).Thissignalisaccessed,forallthemotor-scalecombinations,andinputtothese8illustratestheCMMscalesignalthatisus-quad-bpulsetrainisshownforoneofthemeasurementsegments,8alsoshowstheentiretrapezoidalvelocityprofi-quad-bsignalisemployedtodefineeachofthemeasurementsegments,Ra1toRa6,232synchronisationsignal,triggeredbytheCMMinspectionpartprogram,eCMMarm(movinginoneaxisonly)hasattainedaconstantvelocity,hea-quad-bsignal,mplinglengthisadistanceequivalentto(0.8×␭c).TheRaiscalculatedforfivesamplinglengths(Ra1,Ra2,%,Ra5)andanaveragesurfaceroughnessvalue,Rave,iscalculatedoverthefirstfieragevalueisthencheckedagainstRa6,hepredetermineddistancehasbeenmovedbythesensor,histechnique,thedataacquisiinationofSurfaceProfileAmplitudeParametersThenecessarystepsforprocessingthesensordataandgiredataacquisitionandprocessingsystemhaaacquisitioncardisoftheDIO-96typeandallthedataprocessingalgorithms,outlinedbelow,havebeenimplementedintheLabViewenvironment.4.1SensorControlandDataAcquisitionThesoftwareuserinterfaceallowsthedatacoametersarethetranslationspeedoftheCMMarmovertheobjectsurface,thesamplinglength,startlocation,stand-offdistanceofthesensorheaerfacealsosynchronisesthecommencementoficular,thesoftwareensuresthatnodataiaproducedbythesensorelectroniccontrolunithastwocomponents:alsignalrepresentingthedistancefromthelenstoalsignalthattracksbinationofbothdatasetsrepresenttrolsoftwareconvertseachdatapairintoavalueexpressedinmicrometresbasedontheknownvalueofthelaserwave-length(0.80␮m),andeachvalueaisthenwrittentoadatafileandgraphedonthecontrolsoftwareinterface.4.2SurfaceProfileDataProcessingTwoinitialprocessingstepsareper,thedataisfilteredtoremovehigh-frequencynoiseandsecond;arefer-encelineisdet10(a)illustratestheunfil-teredsensorsignalobtainedwhensornoisecanbeattributedtotheairbearingflutteroftheCMMganseistureprofile

Fig.10.(a)Unfilteredsurfaceprofileacquiredwiththesensorsystem.(b)EffectsofairbearingflutterremovedfromthesurfaceprofileusingsoftwarefinFig.10(b)illustratestheeffectofdatafi.10(b),thedashedlineillustratesthesurfaceprofilebeforefiltering,whereasthesolidlinerepresentsthefilteredprofiitionaldetailevidentinthedashedlineisclearlyaneffectofthesensorsystemanrmore,thishigh-frequencycomponentwouldnot-frequencybandpassfilterwasdesignedusingasetofrepresentativesurfaceprofiher-frequencysensornoisewasclearlydistinguishablefromthesurfaceprofieroughnessamplitudeparametersaredeterminedrelativetoameanreferencelinethatislocatedvertically,withrespecttotheprofile,suchthattheprofileareaenclosedabovethelineisequaltothatbelowit(seeWhitehouse[11]fordetailsonthemeanlinecalculation).4.3CalculationoftheProfileAmplitudeParametersThemostcommonamplitudemeasureisRa(roughnessaverage)defiemeandepartureoftheprofilefromthereferencelineandisgivenbyR=1an͸n͉zi͉(2)i=ewofCMMdeckwithrotarytable,.(2),nrepresentsthenumberofdatmanceoftheSurfaceProfileSensorTheprocedureforperformingautomatedsurfaceroughnessmeasurements,onafreeformsurfacepart,isoutlinedintheflnebladeassembly(similartotheoneshowninFig.1)emblywasmountedonarotarytable,attachedtothedeckoftheCMM,adecanbesequentiallypositionedailsofthesensorpositioningstrategyareasfollows:Foranewcomponent,thestartingpointforeachoftherequiredsurfaceroughnessmeasurementprofilesisdefined.

yTheCMMtouchprobeisused,inconjunctionwiththeinspec-tionsoftware,tocreateapartprogram(forthefibreopticsensor)thathasallofthestartpointsdefioceduredeficesurfaceroughnessmeasurementsoneachbladeusingthestartlocationsdefisorheadispositionedatthestartlocationatthecorrectstand-offdistance(5.0mm)ore,thesurfacedeviation(relativetothesensorheadfrontface)traversesthesensorheadoverthe5.00mmevalu-ationlengthtotheendposition,atwhichpointtheCMMmovestothenextlocationontheblade,itionofthearticulatedprobeheadisrequalifiificationisarytablepositionseachblade,relativetotheCMMgantry,sothatonlyoneCMMaxisisinmotionduringaprofilemeasurement,intainsaconsimumspeedoftheCMMis500mms−1,andthiscanbevaried,usingsoftwarecommands,from0%to100%imumaccelerationis1000mms2,andagainthiscanbevariedfrom0%to100%ofthemaximum,ceroughmandoutputsaTTLhighsignal,viaastandardRS232port,tothedataacquisitioncard,wheneverasurfacefi“acquiresurfaceroughnessdata”comovestwicetherequireddistanceandthesensorsoftwaaccomplishedinthedataacquisitioncarsionThefibre-optic-sensor-generatedprofilesdonothavequitethesamelevelofdetailasthestylusprofitsizeofthelaserisacriticalfactorinthesensoroperationand,unlikethestylustipdiameterwhichisafixedparameter,variesineffectivediameterwithchangingsurfaceprofiensmovescloserto,orfurtherawayfrom,thesurfacethan5mm,nd-offof5mm,thediameterisestimatedtobe20␮m,whichislargerthanthestylustipradiusandresponsiblefortheresultinglossinthefi“convolutioneffect”resultsinsmootherlookingprofilesthanforthestylusprofircontributingfactoristhereflectionofnterferenceconditionismet,thedistancerecordedwillbefromthestrongestreflmple,inasharpgapthestrongestreflarsituationoccursforasharppeakbecausemorelightcanbereflectedbaldforassistancewiththeCMMprogrammingandmountingthefi,ld,i,er,,“Toolconditionmonitoring(TCM)–thestatusofresearchandindustrialapplication”,AnnalsCIRP,44(2),pp.541–562,,“Opticaltechniquesforon-linemeasurementofsurfacefinish”,PrecisionEngineering,3(2),pp.61–83,ff,,“Imageprocessinginaproductionenvironment”,AnnalsCIRP,37(2),pp.579–589,,,“Surfaceroughnessmeasurementbyultrasonicsensingforin-processmonitoring”,ASMEJournalofEngineeringforIndustry,117,pp.439–447,i,S.-,,“Surfaceprofilemeasurementduringturningusingfringe-fieldcapacitiveprolifometery”,ASMEJournalofEngineeringforIndustry,114,pp.234–243,,“In-processindicationofsurfaceroughnessusingafibre-opticstransducer”,Proceedings15thInter-nationalMachineToolDesignandResearch,15,pp.339–347,,,“Measurementofsurfaceroughnesswithalaserbeam”,AustralianConferenceonManufac-turingEngineering,pp.132–133,l,“Surfaceroughnessmeasure-mentwithfibre-optics,technicalbrief”,ASMEJournalofDynamicSystems,Measurement,andControl,105,pp.295–297,y,,“Afibreopticsensorforsurfaceroughnesssmeasurement”,ASMETransactions,JournalofManufacturingScienceandEngineering,120,pp.359–367,y,“Afibreopticinterferometerforthemeasurementofsurfacetopography”,COMADEM95,Kingston,Ontario,ouse,HandbookofSurfaceMetrology,InstituteofPhysicsPublishing,Bristol,1994.

本文标签： 表面粗糙度图文