Sản xuất ngược là một khái niệm bắt nguồn từ kỹ thuật ngược (Reserve Engineering), là kỹ thuật tái hiện lại một chi tiết hay bộ phận có sẵn không phải qua thiết kế từ đầu mà thông qua một thiết bị số hóa biên dạng bề mặt. Sản xuất ngược ngày nay được ứng dụng rất rộng rãi trên nhiều lĩnh vực, nhiều ngành nghề, đặc biệt là trong công nghệ chế tạo ô tô. Nắm bắt được thị hiếu của người tiêu dùng, nhiều loại xe đã được ra đời một cách nhanh chóng với nhiều kiểu dáng mẫu mã khác nhau. Mỗi lần thay đổi công nghệ như vậy sẽ rất tốn kém, ảnh hưởng rất lớn đến chi phí trong sản xuất. Dó đó nhà sản sản xuất chỉ việc số hóa một chiếc xe, từ đó chỉnh sửa trên các phần mềm CAD thì có thể cho ra đời một mẫu xe mới. Các lĩnh vực ứng dụng chính của thiết kế ngược bao gồm: + Thiết kế chế tạo khuôn mẫu (khuôn nhựa, khuôn đúc , ..) + Gia công CNC (dữ liệu mô hình CAD đầu vào ) + Thiết kế, sản xuất hàng tiêu dùng (điện thoại, đồ gia dụng ) + Công nghiệp ô tô, hàng không, y tế và giáo dục, ... + Sao chép, phục hồi, sản xuất phụ tùng đơn chiếc không còn sản xuất. + Ngoài việc phục vụ thiết kế chế tạo, quy trình thiết kế ngược còn được sử dụng để kiểm tra, đánh giá độ chính xác giữa sản phẩm gia công so với nguyên mẫu. + Tạo các mẫu mã mới so với hình dáng ban đầu.
Trang 1j ou rn a l h om epa g e :w w w e l s e v i e r c o m / l o c a t e / p r e c i s i o n
Review
micro-CMMs
J.D Claverley∗, R.K Leach
Engineering Measurement Division, National Physical Laboratory, Hampton Road, Teddington, Middlesex TW11 0LW, United Kingdom
a r t i c l e i n f o
Article history:
Received 31 July 2013
Received in revised form 9 April 2014
Accepted 23 June 2014
Available online 1 July 2014
Keywords:
Micro-CMM verification
Specification standards
Calibrated test lengths
a b s t r a c t
Theperformanceverificationofmicro-CMMsisnowofintenseinterestbecauseoftheircapabilityto performlengthmeasurementsinthreedimensionstohighaccuracywithlowuncertainties.Currently, verificationofmicro-CMMsiscompletedinthespiritofexistingspecificationstandards,becausestrict adherencetothesestandardsisoftendifficult.Thisreviewaimstopresentanddiscussverification tech-niquesavailableformicro-CMMs:specificationstandards,existingcalibratedtestlengthsandtraceability routesthatcanbeassociatedwithmicro-CMMs.Threespecificationstandardsusedinthetestingof CMMswillbeconsidered.Inaddition,awiderangeofcalibratedtestlengthsarereported,andany advantagesanddisadvantagesassociatedwiththeirusearediscussed.Itisconcludedthatmicro-CMMs cannotyetbeverifiedinaccordancewithexistingspecificationstandards.Suggestionsaremadeforfuture standardisationworkrequiredtorectifytheseissues
CrownCopyright©2014PublishedbyElsevierInc.Allrightsreserved
Contents
1 Introduction 2
2 VerificationofCMMs 2
2.1 ISO10360 3
2.2 ASMEB89.4 4
2.3 VDI/VDE2617Part12.1 4
2.4 Discussion 5
3 Calibratedtestlengthssuitableformicro-CMMs 5
3.1 1Dartefactsandballbars 5
3.1.1 Gaugeblocks 6
3.1.2 METASminiatureballbars 7
3.1.3 A*STARmini-spherebeam 7
3.2 Ballplatesand2Dartefacts 7
3.2.1 CarlZeissminiatureballplatefortheF25 7
3.2.2 METASballplate 8
3.2.3 Krugerplates–column,ballandhole 9
3.2.4 PTBmicro-ballplates–smoothandroughened 9
3.2.5 Siliconmicro-machineddimensionalcalibrationartefact 9
3.3 Otherartefacts 9
3.3.1 Aztecstandard 9
3.3.2 Micro-holestandard 9
3.3.3 Polytecstepheightstandard 9
3.3.4 Calottecube,calotteplateandotherXCTartefacts 11
3.3.5 Micro-contourstandard 12
∗ Corresponding author Tel.: +44 2089436242.
E-mail address: james.claverley@npl.co.uk (J.D Claverley).
http://dx.doi.org/10.1016/j.precisioneng.2014.06.006
0141-6359/Crown Copyright © 2014 Published by Elsevier Inc All rights reserved.
Trang 23.4 Discussion 12
4 Conclusions 13
Acknowledgements 13
References 13
Aco-ordinatemeasuringmachine(CMM)isdefinedasa
mea-suringdevicewiththemeanstomoveaprobingsystemandthe
capabilitytodeterminespatialco-ordinatesonaworkpiece
sur-face[1].CMMshavebecomeessentialforindustrialdimensional
metrology,anditis,therefore,alsoessentialthattheiraccuracycan
beestimatedandtheirtraceabilitycanbeconfirmed.The
verifica-tionoftheperformanceofCMMsiswellunderstood,withextensive
specificationstandardsavailableforusersandmanufacturersto
ensureconformity
ArecentlydevelopedgenerationofCMMsarethosespecially
designedforminiaturegeometriesrangingfrombetweena few
micrometres toapproximately 1mm [2] The first recognisable
micro-CMMwasdevelopedandconstructedattheNational
Physi-calLaboratory,NPL,in1999[3,4].AconcurrentprojectatEindhoven
Universityof Technologydesigned andbuilta micro-CMM that
wouldact astheprototypefora commercialmachine [5]
Sev-eralothermicro-CMMsweredevelopedbyNationalMeasurement
Institutesoverthefollowingyears[6–9].Twocustombuilt
CMMs,areshowninFig.1
Sincetheirinitialdevelopment,micro-CMMshavebeensubject
toresearchintheareaofverification, calibrationand
standard-isation[10–13] This research has becomeever more essential
as micro-CMMs have become commercially available and are
nowusedinindustrialenvironments.Twocommerciallyavailable
micro-CMMsareshowninFig.2,theF25fromCarlZeissAG[14]and
theIsara400fromIBSPrecisionEngineeringBV[15].Severalother
micro-CMMsarecommerciallyavailable,including(butnot
lim-itedto):theVideoCheckUAfromWerthMeßtechnikGmbH[16],
theNMMfromSIOSMeßtechnikGmbH[17],andtheUMAPvision
systemfromMitutoyoCorporation[18]
Forthepurposeofthisreview,amicro-CMMisdefinedasa
con-tactingCMM(ofvarioustypesandmakes)whichisusedtomeasure
geometrieswhosedimensionsrangefromafewmicrometresto
approximately1mm[19]
The verification of micro-CMMs is now of intense interest
becauseof theircapability toperform lengthmeasurements in
2.5 dimensions(2.5D) to highaccuracywith low uncertainties
Althoughitiswidelyreportedthatmostmicro-CMMsare3D
instru-ments,limitationsontheorientationofprobestylioftenresultsin
aninabilitytomeasureundercuts.Therefore,althoughthe
micro-CMMplatformsthemselvesareabletomeasure3Dco-ordinatesto
highaccuracywithlowuncertainties,certain3Dgeometriescannot
bemeasured
Atypicalmaximumpermissibleerroroflengthmeasurement
ofamicro-CMMforsizemeasurement(EL,MPE)is±250nmwithan
additionallengthdependentterm,whichtendstobearound2nm
foreverymillimetremeasured.Withsuchprecisemeasurements
beingcommonplaceintheareaofmicro-co-ordinatemetrology,
verificationoftheinstrumenttoagreedinternationalstandardsand
availabilityoftraceablecalibratedtestlengthsisessential
Currently, EL,MPE for micro-CMMs is determined according
to ISO 10360-2:2009, Geometric Product Specification (GPS) –
Acceptance and reverification tests for coordinate measuring
machines (CMMs) – Part 2: CMMs used for measuring linear
dimensions [20], however, significant sections of this standard
arenotapplicabletomicro-CMMsdue totheirsizeand design
Similarly, whencompleting a micro-CMMcalibration, or deter-minationofthemicro-CMM’serrormap,certainproceduresare difficulttocomplete
Thisreviewwillpresentanddiscussverificationtechniques cur-rentlyavailableformicro-CMMs.Thesediscussionswillincludea descriptionoftheexistingandpertinentspecificationstandards, existingcalibratedtestlengthsandtraceabilityroutesthatcanbe associatedwithmicro-CMMs
TheverificationofCMMsisgovernedbyspecificationstandards andgoodpracticeassociatedwithdimensionalmeasurement.An exampleofaninternationalspecificationstandardforthe verifica-tionofCMMswouldbetheseriesISO10360–GeometricProduct Specification(GPS)–Acceptanceandreverificationtestsforcoordinate measuringmachines(CMMs).Goodpracticeindimensional mea-surementisreliantonexperienceandknow-how,andtherefore difficulttoreport.A great deal ofefforthasbeen spentin dis-seminatingCMMgoodpracticeinNPL GoodPracticeGuides41,
42and43(CMMmeasurementstrategies,CMMverificationand CMMprobing,respectively)[21–23].Anextensivereviewof clas-sicalCMMverificationtechniquesandcalibratedtestlengthscan
befoundin[24] Forclarity,ISO10360-1,GeometricProductSpecification(GPS) – Acceptance and reverification tests for coordinate measuring machines(CMMs)–Part1:Vocabulary[1],definesacceptancetests andreverificationtestsasfollows
• Acceptancetest–asetofoperationsagreeduponbytheCMM manufacturerandtheusertoverifythattheperformanceofa CMMisasstatedbythemanufacturer,performedwhentheCMM
isinstalledorafteranymajormodification
• Reverificationtest–atesttoverifythattheperformanceofaCMM
isasstatedbytheuserandexecutedaccordingtothesame pro-ceduresasthoseoftheacceptancetest,performedperiodically
asrequired
Also,itisimportanttonotethatathirdtest,theinterimcheck,
isalsoavailablefortheuserofaCMM
• Interimcheck–atestspecifiedbytheuserandexecutedbetween reverificationstomaintainthelevelofconfidenceisthe measure-mentstakenontheCMM,performedbytheuseratanytime OthertermsusedtodescribetestingofaCMMinclude qualifi-cation,verificationandcalibration.Thetermqualificationusually referstotheday-to-daydeterminationoftheeffectiveradiusofthe stylustip.Thetermsverificationandreverificationareessentially interchangeableand describetestscompletedtoverifythatthe performanceofaCMMisasstatedbythemanufacturer.Itshould
benotedthatverificationproceduresareusedtotestthe perfor-manceoftheCMM,whereascalibrationprocedures(oftenreferred
toaserrormapping)areusedtodeterminethemagnitudeofall twenty-onekinematicerrorsourcesoftheCMM
ThetechnicalprocedurefortheacceptancetestsofaCMMare detailedinISO10360-2:2009[20].Thereareproblemswhen try-ingtoapplyexistingacceptancetests,asdefinedinISO10360-2,
tomicro-CMMsduetoseveralfactors.Firstly,thenominalartefact
Trang 3Fig 1. The NPL Small CMM developed in 1999 (left) and the METAS Ultraprecision CMM developed in 2001 (right – courtesy of METAS, CH).
sizessuggestedbythestandardsareusuallylargerthanthe
maxi-mumsizeofartefactthatwouldfitinthemeasurementvolumeof
themicro-CMM.Insomecases,thisisnotamajorissue,asthe
sug-gestedmeasurementlengthsareonlyarecommendationandcan
bealteredtofittheneedsoftheuser.However,thisissuedoes
high-lightthefactthatthestandardswerenotwrittenwithmicro-CMMs
inmind
Secondly,thereareseveralrequirementsoftheverificationtest
that,duetokinematicdesignsofmicro-CMMs,areimpossibleto
complete.Thedesignsofmanymicro-CMMsaresuchthatcertain
orientationsofcalibratedartefactsareinaccessibletoamicro-CMM
probingsystem.Theoverallstyluslengthandtheeffectiveworking
lengthofthestylussystemareshorttoincreaseaccuracywiththe
downsidethatmanymeasurementfeaturesbecomeinaccessible
Finally,there islimited availabilityof calibratedtest lengths
suitableformicro-CMMswhicharealsocalibratedtouncertainties
comparableto,orindeedsignificantlybetterthan,thecapability
ofthemicro-CMMtomeasurelineardimensions.Infact,itis
pos-siblethatonlygaugeblocksmeasuredaccordingtothemeasuring
principleoflightinterference[25],offersufficientlysmall
measure-mentuncertaintytoverifythelengthmeasurementcapabilityof
micro-CMMs
Allofthesepointscombinetoproduceasituationwhere
micro-CMMsarecovered byveryfew specificationstandards, andare
thereforeverifiedaccordingtotestsagreedbetweentheinstrument manufacturer and theuser Toproperly describe this situation, threespecificationstandardsusedinthetestingofCMMshavebeen selectedfordiscussionastotheirapplicabilitytomicro-CMMs 2.1 ISO10360
TheISO10360seriesofstandardsispreparedbyWorkingGroup
10ofTechnicalCommitteeISO/TC213,thecommitteeconcerned withthedimensionalandgeometricalspecificationand verifica-tionofproducts.Thecompletespecificationstandardconsistsof severalparts(sevenatthetimeofwriting,withthreefurtherdraft partsindraftform),thateachdetailaspectsofCMMperformance verification,dependingonthenumberandtypeofaxestheCMM has,andwhichprobingtechnologyisused.SomelimitationsofISO
10360withrespecttomicro-CMMswillnowbebrieflyhighlighted, referencedspecificallytosectionswithintheISO10360series
Amajorpartoftheacceptanceandreverificationtestis selec-tionandpositioningofsuitablecalibratedtestlengthswithinthe measurementvolumeoftheCMM.ISO10360-2requiresthe mea-surementoffivedifferentcalibratedtestlengthsplacedalongseven orientationswithinthemeasurementvolumeoftheCMM,fourof whichmustbethespacediagonals.Althoughtheremainingthree orientationsarelefttothediscretionoftheuser,itisusualthat
Trang 4thelongesttestlengthshouldbeatleast66% ofthemaximum
traveloftheCMMalongalinethroughthecalibratedtestlength
Thesuitabilityofthecalibratedtestlengthtoperformadequately
dependsonseveralparametersotherthanitslength.Firstly,the
uncertaintyofthecalibratedlengthsshouldbeatleasttentimes
smallerthan theexpectedEL,MPE oftheCMM.Thesuitabilityof
thecoefficientofthermal expansionofthebulkmaterialofthe
calibratedtestlengthisalsoimportant.Finally,themeasurement
featuresthatdefinethecalibratedlengthsshouldbeaccessibleto
theCMMbeingtested.Thisaccessibilityrequirementcanbe
diffi-cultformicro-CMMs,especiallywhenmeasuringacrossthespace
diagonalsandpositionsparalleltothez-axis
Ashortnoteregarding‘verysmallCMMs’isfoundinsection
6.5.2.2of ISO 10360-2, where it is stated that it is acceptable
totranslatethecalibratedtestlength,whilekeepingitoriented
alongthediagonalbeingtested.Thisallowanceisdesignedto
pro-videsufficientclearanceformeasurementandinsomecasesmay
allowcertainbodydiagonalmeasurementstobemade;aslongas
theminimumlengthrequirementsforthelongesttestlengthare
adheredto
AfurtherlimitationcomesaboutinISO10360-5:2010,
Geomet-ricProductSpecification(GPS)–Acceptanceandreverificationtests
forcoordinatemeasuringmachines(CMMs)–Part5:CMMsusing
singleandmultiplestyluscontactprobingsystems[26].Insection
6.2.3,thematerialstandardofsizefortheprobingacceptanceand
reverificationtest(single-stylusprobingerrortest)isdefinedas
aspherewithdiameterbetween10mmand50mm.Given that
theworkingstyluslengthofmicro-CMMprobesisoftenlessthan
10mm,evenaslowas2mm[27],therequirementsonthediameter
ofthetest spherecan beproblematic Alsosection 6.2.4.1
lim-itsthestyluslengthduringtheprobeacceptancetestto20mm,
30mm,50mmor100mm.Consideringcommonstylilengthsfor
micro-CMMs,thisisanotherproblematicconstraint
Smallormicro-CMMsareinfrequentlymentionedwithinthe
ISO10360series,butacommonthemewithinthedefinitionsof
scopeforeach partof theseries isthe allowancefor a mutual
agreementbetweenthemanufacturerandtheuserforthe
veri-ficationofinstrumentswithunusualdesignsorprobingsystems.It
isconceivablethatsimilaragreementscanbe,andindeedare,made
betweenmanufacturersandusersofmicro-CMMssothatthe
ver-ificationandcalibrationoftheseinstrumentscanbecloselylinked
tointernationalstandards
2.2 ASMEB89.4
Anotherstandardavailablefordefiningperformanceevaluation
methods for CMMs is ASME B89.4.1b-2001 – Methods for
per-formanceevaluationofco-ordinate measuringmachines[28].This
standarddoesnotdiffersignificantlyfromISO10360-2incontent;
however,certainchangesandadditionsaremadeinareasthatmay
bepertinenttothecalibrationofmicro-CMMs
Agreat dealofthestandardisdedicatedtotheinvestigation
oftheenvironmentalconditionsinwhichtheCMMisoperated
Micro-CMMsareabletotakemeasurementstoanuncertaintyoften
comparabletotheresolutionofmacro-scaleCMMs.Hence,
envi-ronmentalfactorscouldhaveasignificanteffectontheiroperation
SimilartoISO10360-2,thepositionsofacalibratedtestlength
aredefinedforboththedeterminationofthelineardisplacement
accuracy and the volumetric performance Any measurements
takenpurelyinthez-axisofamicro-CMMare,again,limitedby
theorientationoftheprobe.Severalofthelocationssuggestedfor
measurementofthefacediagonalsandthespacediagonalswillalso
bedifficulttomeasureusingsomemicro-CMMsduetotheaccess
limitationsaffordedbytheshortstyluslength
CertaindifferencesexistbetweenASMEB89.4andtheISO10360 seriesthatarevery relevanttothecomplianceof micro-CMMs Forexample,ASMEB89.4.1bsection6.1.1describesthetestsphere withadiameterof6mm,whichismoresuitableforusewith micro-CMMsthanthe10mmdiametertestspheredefinedinISO10360-5 [26].However,theprobinganalysisprocedureincludesthe require-mentforaprobingpointtobetaken10◦belowtheequatorofthe testsphere.Thisprobingmaynotbepossiblewithmany micro-CMMsbecause,duetothestylusgeometry,theshaft,ratherthan thetip,couldcontactthetestsphere(calledshanking).Itis sug-gestedthat,inthecaseofISO10360-5,therequiredsizeofthetest sphereisnotideal,andinthecaseofASMEB89.4,thesizeofthe testsphereismoresuitable,buttherequiredprobingstrategyis notideal
LikeISO 10360-2,ASMEB89.4.1bcoversmostaspectsofthe acceptancetestingandreverificationofCMMs,andshouldbe appli-cabletomicro-CMMs.However,thesespecificationstandardswere preparedwhenfew,ifany,micro-CMMsexisted.Therefore,there areconsiderablelimitationsontheapplicationofthesemethods
totheverificationofmicro-CMMs.Theselimitationsoccurmostly duetothelackofflexibilitywhenconsideringsmallstyli,bothin effectivelengthandtipdiameter,andthekinematicdesignofmost micro-CMMs
2.3 VDI/VDE2617Part12.1 TheAssociationofGermanEngineers(VDI)developsguidelines whichcoverawiderangeofapplications.Theseguidelinesareoften lessdetailedthanthoseproducedbyISOorASME,butarereleased moreoften,andalsotendtoadoptnoveltechniquesearlier.Hence, theVDIguidelineVDI/VDE2617Part12.1–Accuracyofcoordinate measuringmachines–characteristicsandtheirchecking–Acceptance andreverificationtestsforcontactingCMMmeasuring microgeome-tries[19],istheonlyguidelinepublishedthataddressestheneed foranapproachtotestingmicro-CMMs
Theguidelineclearlydefinesthescopeofmicro-CMMuseto geometriesofafewmicrometresuptoabout1mm.The guide-linealsodefinesthestylussystemofamicro-CMMtobeshorter than 5mm in length and to have a probe tip diameter below
300m.Itisimportanttonotethattheguidelinerefersspecifically
to‘contactingCMMsmeasuringmicro-geometries’while measur-ingusingsinglepointprobing.Thereforetheguidelinealsocovers macro-scaleCMMsequippedwithsmallstyliusedformeasuring micro-geometries
Theguidelinereferstotheissuesassociatedwithusingsmall styli with low probing forces and short over travel Several miniature probing systems are highlighted within the guide-lineincludingpiezo-resistiveprobes,opto-contactingprobesand vibratingprobes
Tocompletetheacceptanceandreverificationtestsfor micro-CMMs,theguidelinedescribesasimilarsetofmeasurementsto thosedescribedinISO 10360-2andASMEB89.4.VDI/VDE2617 Part 12.1requires a minimum of five test lengths along seven orientationswithinthemeasurementvolume.Fouroftheseven ori-entationsmustbethespacediagonalsofthemicro-CMM.Asuitable calibratedtestlengthshouldbeused,whichshouldhavea coef-ficientofthermallinearexpansionbelow13×10−6K−1.Inconel, Stellite,ferritictypestainlesssteelandmostceramicsmeetthis requirement
Asimilarsetoftestsisprescribedasacceptanceand reverifica-tiontestsoftheprobingsystem.Thisprocedureissimilartothat describedinISO10360-5,howeverallpreviouslydescribed limi-tations,suchasstyluslength,stylustipdiameterandtestsphere diameterareremoved
Trang 5of microgeometries with micro-CMMs These include contact
pressureexertedbyminiaturestylustips,materialpairing,
contam-inationeffects,formerrorsofcalibratedtestlengths,especiallyon
thereferencesphere,andfeasibilityofthecalibrationuncertainties
ofthecalibratedtestlengths.Theguidelinealsosuggestssome
use-fulcalibratedtestlengths,allofwhicharedescribedinthispaper
2.4 Discussion
Thecontentofthethreedescribeddocuments,theinternational
standardISO10360series,theAmericannationalstandardAMSE
B89.4andaGermannationalguidelineVDI/VDE2617Part12,all
covertheverificationofCMMsusingcalibratedtestlengths.Asno
internationalstandard existsspecificallydescribing thefull
ver-ificationofmicro-CMMs,aspectsof ISO10360and ASMEB89.4
mustbeappliedtocompletethetaskofverification,usuallyundera
mutualagreementbetweentheinstrumentmanufacturerandthe
user.VDI/VDE2617Part12goessomewaytoidentifythepartsof
thetwodescribedstandardswhichneedspecificattentionwhen
dealingwithmicro-CMMs,however,thisguidelineisnot
exhaus-tiveforallaspectsoftheoperationofmicro-CMMs.Onemajorissue,
highlightedintheVDI/VDEdocument,isthedifficultyofachieving
therequiredcalibrationuncertaintiesoncalibratedtestlengths.To
addressthisissueofuncertainties,theremainderofthisreviewwill
highlightexistingcalibratedtestlengthssuitableformicro-CMM
verification
All of the previously mentioned standards and guidelines
depend heavily on the use of calibrated reference artefacts to
ensuretraceabilityduringacceptance and reverificationtesting
ISO10360-2:2009referstotheuseofcalibratedtestlengths,as
opposedtomaterialstandardandmaterialstandardofsizewhich
aredefinedinISO10360-1:2001.Theterm‘calibratedtestlength’
isintendedtoincludetheuseoflaserinterferometerswhilestill
beingequivalentto‘materialstandardofsize’(definedaccording
toInternationalvocabularyofmetrology–basicandgeneralconcepts
andassociatedterms(VIM)[29])
Thecalibratedtestlengthsdescribedinthissectionwillbe
con-sideredfor theirsuitabilityforuseduringacceptance testingof
micro-CMMs.Insomecases,thecalibratedtestlengthsmaynot
besuitableforeitheracceptancetestingorreverification.Inthis
case,theirsuitabilityforinterimtestingwillbeconsidered
Itisassumedthat,asthefollowingreviewisconcentratedon
contactingsystems,acalibratedreferencesphereisusedforthe
purposesofprobetipdiameterqualification.Commontipdiameter
qualificationproceduresresultinginasingleresultfortheeffective
stylustipdiameterareofteninsufficientfordeterminingthe
partic-ularsofthestylustipofaprobeformicro-CMMs.Severalexpanded
mappingtechniqueshavebeendeveloped[30],andtheproblem
ofspheretipdiameterevaluationisthesubjectofseveralnational
andEuropeanMetrologyResearchProgrammeprojectsatthetime
ofwriting[31]
Theuseofhighprecisionopticaldistancesensors(ratherthan
simplehighprecision2Dopticalorvideo,sensors)isnotyetcovered
byanypartoftheISO10360series,althoughatthetimeofwriting,
adraftofISO10360-8.2[32]isdueforpublication.Althoughthe
useofopticaldistancesensorsonmicro-CMMsisnotspecifically
consideredinthisreview,someofthecalibratedtestlengthsinthis
reviewaresuitableforuseforopticaldistancesensors
Severalcriterianeedtobemetifanexistingartefactwereto
performwellascalibratedtestlengths,asdescribedinISO10360-2
Firstly,thematerialthatdefinesthedimensionalquantitytobe
measuredmustbedimensionallystableovertime
Secondly,itiscommonthattheartefactshouldbemadeupof geometricfeatures.Thisrequirementisessentialsothata correla-tioncanbedrawnbetweenthemeasuredpointsandtheevaluated features.Theuseofgeometricfeaturesallowsthemeasuredpoints
tobedifferenttotheevaluatedfeatures.Itisalsoessentialthatthese featuresbeaccessiblebyamicro-CMM’scontactingmicro-probe Theartefactshouldalsoexhibitgoodtemperaturestability,both whentheartefactisstoredinacontrolledenvironment,withsmall temperaturedeviationsandlowtemperaturegradients,andalso whentheartefactistransported,i.e.,whentheartefactissubjectto largetemperaturedeviationsandgradients.Afterany environmen-talexcursions,theartefactshouldreturntothesameshapeand sizeaftersoakinginacontrolledenvironmentforseveralhours Goodtemperaturestabilityisoftengainedbyusingamaterialwith
alowcoefficientofthermalexpansion.Thiscoefficientshouldbe wellknownsothatacorrectioncanbeappliedtothereference temperatureof20◦C[33]
Finally,giventhespecialiseddesignofallcommercially avail-ablemicro-CMMs,itisimportantthattheartefactbesuitablysized
totestthefullmeasurementvolumeoftheinstrument.Atypical measurementvolumeforamicro-CMMisacubewithsides100mm long.AccordingtoISO10360-2,asuitablecalibratedtestlengthisat least66%ofthelongestmeasurablelengthwithinthatmicro-CMM measurementvolume.Forthiscommoncase,a66mmcalibrated lengthalongthemachineaxes,or93mmcalibratedlengthalonga facediagonal,ora115mmcalibratedlengthalongthespace diag-onal,willbesufficient
Aliteraturereviewhasbeencarriedouttoidentifyexisting ref-erenceartefactsforuseascalibratedtestlengthformicro-CMM acceptanceandverificationtesting.Thereviewmainlyfocusseson artefactsthatfulfiltherequirementsofthethreepreviously men-tionedspecificationstandards.Itisimmediatelyapparentthatall thecalibrated test lengthsfoundin theliterature fit intothree maincategories:gaugeblocks;ball-barsandother1Dcalibrated testlengths;ball-platesandother2Dcalibratedtestlengths;and othercalibratedtestlengths(thosethatarenotnecessarilyuseful formicro-CMMverification,butdoserveapurposeinthefieldof micro-CMMtesting)
Onemainomissionfromthisreviewistheuseoflaser interfer-ometryasameansofproducingacalibratedtestlength.Although severalresearchpapersexistonthissubject[5,8,12,34,35],andalso theapplicationoflaserinterferometersasthemeasurementscales
ofmicro-CMMsisnowacommonoccurrence[17,36–38],this tech-niquewillnotbereviewedasthistechnologyiscostly,difficultto useandinterpret,andpresentsseveralfurthertechnicalchallenges
3.1 1Dartefactsandballbars
Itisanessentialpartofanyspecificationstandardthatfew lim-itationsareputonthenatureofthecalibratedtestlengthusedin testingaCMM.Thisrequirementis,initially,apracticality require-ment,butalsoaneconomiconepertainingtoavailabilityandcost Likewise,thenatureofthecalibrationofthecalibratedtestlength, especiallythegeometricaldefinitionofthecalibratedtestlength, canbeflexible,providedthatthedefinitionusedincalibratingthe testlengthisreplicatedintheCMMmeasurementstrategy.Often, thisflexibilityreferstothenatureofthemeasurement,either bidi-rectionalorunidirectionalmeasurements.However,asISO10360-2
is seen asa ‘blackbox test’, an overall checkof theCMM, the measurement of calibrated testlengths is required tobemade bi-directionally
Often,bothuni-andbidirectionalmeasurementsarepossibleon manyartefactsanditisthereforeessentialthattheprobing strat-egyforthecalibratedlengthiswelldefined.Asitiscommonthat thecalibratedtestlengthisdefinedusinggeometricfeatures,the
Trang 6used
3.1.1 Gaugeblocks
Gaugeblocksareoneofthemostcommonartefactsfor
dissem-inatingtheunitoflength[25].Theircommonalitymostlystems
fromthewiderangeofavailablelengthsandmaterials.Also,gauge
blocksareideallyplacedtobeusedforthecalibrationof
micro-CMMs because of the ability to define the central length to a
measurementuncertaintyofbelow40nm(k=2)forlengthsupto
100mm[39].WiththeMPE(EL)(orEL,MPE)ofmostcommercially
availablemicro-CMMsbeingbelow300nm,onlygaugeblocksoffer
calibrationuncertaintiessuitable toverify linearmeasurements
madeonamicro-CMM.However,duetotheshapeofagaugeblock,
chosentoeasethetransferofthestandardoflengthtomeso-scale
andmacro-scale instruments,therearelimitations ontheiruse
withmicro-CMMs
Theuseofgaugeblocksdirectlyaddressestheprovisionoftest
lengthssuitableforz-axisverification,althoughthisisalso,toan
extent,dependantonthelengthofthestylus.Toperformatestin
thisorientation,acalibratedgaugeblockiswrungtoareference
platenandplacedinthemeasurementvolumeofthemicro-CMM
suchthatthereferencelengthisalignedwiththez-axis.The
micro-CMMcanbeusedtomeasureboththeplanarreferencesurfaceof
theplatenandthemeasuringfaceofthegaugeblock.This
strat-egycoincidesdirectlywiththedefinitionofthelengthofagauge
block[40].Thisstrategyisonlyvalidprovidedthelengthofthe
gaugeblockis sufficientlysmalltonot hindertheprobing
sys-tem.Themaximumlengthforwhichthisprocedureissuccessful
issimilartothestyluslengthoftheprobingsystembeingused,
whichcouldbeaslittleas5mm.Longergaugeblockscanstillbe
measured,howeverthemeasurableareaofthereferenceplaten
becomesseverelylimited,resultinginmeasurementsthatarenot
directlycomparabletothecalibratedlengthofagaugeblock
Duringthemeasurementof2Dlengths,requiringlateral
prob-ing,micro-CMMsareunabletofullyprobethegaugingfacesofthe
blocks,duetothereducedeffectivelengthofthemicro-probe
sty-lusandtheedgechamferontypicalgaugeblockmeasuringfaces
Thislimitationcouldintroduceerrorsinthemeasurements,asthe
calibratedlengthofthegaugeblockisdefinedasbeingmeasuredat
thecentreofagaugingface,andalsobecausetheparallelismofthe
gaugingfacescannotbefullydetermined.Thereareseveral
solu-tionstothisproblem,includingthecalibrationofthegaugeblocks
atamoresuitableposition,i.e.,atapositionthatisaccessiblebythe
micro-probe,orthemanufactureandcalibrationofnon-standard
gaugeblocksdesignedforeasyuseonmicro-CMMs
Asetofnon-standardgaugeshavebeenmanufacturedatthe
Physikalisch-TechnischeBundesanstalt(PTB),theNational
Metrol-ogyInstituteofGermany,wherebyagaugeblockbridgemadeof
Zerodurhasbeenmanufactured[41].Thecalibratedtestlengthis
suitablefortheverificationoftestlengthsin1Dand2D,andalsofor
thedeterminationofthestraightnessdeviationsofthehorizontal
axesofamicro-CMM.AnimageoftheZerodurgaugeblockbridge
isshowninFig.3
Thedistancesbetweentheinternalandexternalbridgefaces
werecalibratedbyPTBwithanexpandeduncertainty of20nm,
usingtheprincipleoflightinterference.Thecalibrationpositionis
lessthan1.5mmfromthetopedge,allowingthedefinitionofthe
calibratedtestlengthtobereplicatedbytheprobingstrategyofa
micro-CMM.Theresultsofmeasurementsmadeonthiscalibrated
testlengthhaveindicatedthattheinvestigatedmicro-CMM
exhib-itedstraightnesserrorsofunder50nmovera43mmlength,and
alsothatreferencedistancesweremeasuredtobetterthan100nm
[42].However,itshouldbenotedthatthisarrangementonlyallows
foronemeasureddistancepersetup
Fig 3. A non-standard set of Zerodur gauge blocks used as a calibrated test length for the calibration of micro-CMMs Image courtesy of PTB, DE.
Whenanyfacediagonalsincludingz-axistravelandanyspace diagonalsaretested,accessproblemsaffectthemeasurement pro-tocol.Currentlytherearenocommerciallyavailableangledprobes availableformicro-CMMs,andhenceasecondauxiliarygauging surfaceisrequiredtocompletethemeasurement.Aphysicalsetup
ofthisprocedure,suggestedin VDI/VDE2617Part12,isshown
inFig.4, wherethelengthLis inferredfromlengthofthe ref-erencegaugeblock.Thisrepresentsamethodfor unidirectional measurementofagaugeblock
Thisarrangementofgaugeblockscouldpotentiallyintroduce errorsduetothewringingofthetwogaugeblocksandduetothe flatnessandparallelismerrorsofthegaugingfaces.Eventhough thelengthofagaugeblockincludestheeffectofone-facewringing [40],thiswringingeffectistoareferenceplatenandnotasecond gaugeblock.Areasonableestimateoftheresultingcontributionto theexpandeduncertaintyduetotheseerrorscouldbeoftheorder
of50nm,basedonthetoleranceforthevariationinlengthofa gradeKgaugeblock(calibrationgrade)[40]
Eventhoughthisincreasederrorisacceptableintermsofthe manufacturer’sspecifiedvalueofEL,MPEofmostcommercial micro-CMMs,thelogisticalcomplicationsofusingthissetupresultsin completionofaverificationtesttakingmanyhours,orindeed sev-eraldays
EventhoughthecompletionofanISO10360-2likeacceptance test ispossible usinggaugeblocks, and a devisedprocedure, it wouldbetime consumingandnotadherenttotherequirement
ofbi-directionalmeasurement.Agooddealoftimewillbespent
insettingupthevariouslengthgaugeblocksinthemeasurement volume,andalsoonwaitingforthesystemtothermallystabilise afterinterferencefromtheoperator.Thetestwouldalsorequirea greatamountofskillandverypreciseoperationoftheCMM,and
ishenceunsuitablefornon-expertusers
Fig 4. A suggested physical setup for testing a length, L, along any face diagonal
Trang 7Fig 5. METAS miniature ball bars Image courtesy of METAS, CH.
3.1.2 METASminiatureballbars
Apreviouslystatedcriterionforthedesignofacalibratedtest
lengthisthatitshouldcompriseseveralgeometricfeatures,such
thatacorrelationcanbemadebetweenthemeasuredpointsand
theevaluatedco-ordinates.Inthecaseofgaugeblocksandauxiliary
gaugingfaces,thegeometricfeatureisaplane,andtheevaluated
featuresareperpendiculardistancesbetweenaplaneandapoint
Therefore,theuncertaintyinthelengthisdependentontheform
andtheparallelismoftheplanes.Thesegeometriesbecome
inac-cessiblewhenanytiltisintroducedonthecalibratedtestlength
toincludemeasurementinthez-axisofthemicro-CMM.An
obvi-oussolutionforthisinaccessibilityofgeometricalfeaturesatan
angleistouseafeaturewhosenominalgeometryisinvariantwith
rotation,i.e.,asphere.Spheresandspherebeamsarecommon
ref-erencestandardsinco-ordinatemetrology,andthedevelopment
ofaminiatureballbarartefactisanidealstartingpointforthe
developmentofcalibratedtestlengthssuitableformicro-CMMs
AsetofminiatureballbarshavebeenmanufacturedbyMETAS
usingZerodurbarsandsyntheticrubyspheres[43].Theballbars
rangeinlengthfrom20mmto100mm,andhavebeenusedto
verifytheglobalmeasurementprecisioninthewholevolumeof
amicro-CMMtobe15nm+L/1400m/mm.Theseminiatureball
barsareshowninFig.5
Thisdesignofballbarcanbeusedparalleltothexandy
mea-surementaxes,alongfacediagonalsandalongspacediagonals.The
probingoftheendspheresisrelativelyfast,anditissuggestedthat
duetoshortmeasurementtimeandthematerialsused,thismethod
isalmostunaffectedbythermaldrift[43].However,aswithgauge
blocks,anyproceduretofullydetermineELofthemicro-CMMwill
beverytimeconsumingduetotheneedtomanuallychangeand
rearrangeallofthedifferentlengthballbarsandwaitforthermal
stability
Toreducethisdisruption,anartefactthatincludesseveral
mea-surablelengths(atleastfive,accordingtoISO10360-2)wouldallow
eachorientationoftheartefacttobemeasuredatonce
3.1.3 A*STARmini-spherebeam
Amini-spherebeamhasbeendevelopedattheNational
Metrol-ogyCentreinSingaporethatconsistsoftenevenlyspacedGrade3
[44]rubyspheres,eachofdiameter5mmwhosecentresarespaced
at10mmintervals[45].Thespheresaremountedontoacarbon
fibrerod.Aspecifictypeofcarbonfibrewaschosenbecauseofits
lowco-efficientofthermalexpansionof−0.8×10−6/ C
Thesphereshaveaspecifieddeviationfromsphericalformof
0.08mandanarithmeticaveragesurfaceroughnessspecification
of0.10m.Animageofthemini-spherebeamisshowninFig.6
Calibrationofthemini-spherebeamwasundertakenby
Eid-genössischesInstitutfürMetrologieMETAS(theFederalInstitute
forMetrology, Switzerland),theNationalMetrologyInstituteof
Switzerland, onthe micro-CMM shown in Fig 1.An expanded
Fig 6. The mini-sphere beam, as developed by A*STAR, Singapore.
measurementuncertaintyof63nmwasquotedforthe90mm max-imum lengthof the spherebeam Thebeam wassubsequently usedtore-verifyamicro-CMMattheNationalMetrologyCentre, Singapore.Resultsfromthesemeasurementsrevealedapossible squarenesserroronthemachine[45]
Themini-spherebeamwasproposedasanartefactsuitablefor evaluatingthevolumetricmeasurement error,specificallyasan alternativetogaugeblocks.Itwasdesignedtoreducetheproblems whichoccurwhenmeasuringgaugeblocksacrossanydiagonalthat includesverticalinclination.Assuch,itcanbepositionedwithin themeasurementvolumeinmanyoftheorientationsrequiredby ISO10360-2.Theonlygeometricallimitationofthiscalibratedtest lengthisitsunsuitabilityforusetoverifythez-axisofamicro-CMM Eventhoughtheco-efficientofthermalexpansionforthis cal-ibrated test length is low, it should benoted that sometypes
ofcarbonfibrecanhavecoefficientsofthermalexpansionupto 2.0×10−6/ C.Also,thesusceptibilityofcarbonfibreandepoxyto changesinhumidity,whichhavenon-zerocoefficientsofmoisture expansion[46],andalsopossiblelowdimensionalstabilityover time,suggeststhatalowexpansionmetalstructuremaybemore suitable
A similarmini-sphere beamconcepthasbeen developedby TrapetPrecisionEngineering[47],howevernoliteraturecouldbe foundthatdescribeditsperformance
3.2 Ballplatesand2Dartefacts Whenextendingthecalibratedlestlengthsofferedbyatest arte-factintotwodimensions,thedistinctionbetweenbi-directional and unidirectionalmeasurement becomes unnecessary Due to theirdesign,all2Dartefactsareunidirectional.Theirdesign there-fore begins tolimit thecalibration methods for suchartefacts,
as length measurement using the methodof light interference becomesmoredifficultintwodimensions
3.2.1 CarlZeissminiatureballplatefortheF25
AminiatureballplatehasbeendevelopedbyCarlZeissthat consistsofaZerodurbasewithninesiliconnitridehemispheres wrungtoitssurface.Thedesignoftheplateresultsinatotalof thirty-sixindividualball-to-balldistancesrangingfrom13mmto
100mm.ThisartefactisshowninFig.7 Thedesignoftheminiatureballplatecoversseveral require-ments set by ISO 10360-2, including its ability to be used to
Trang 8Fig 7.Miniature ball plate manufactured by Carl Zeiss (left) in its inclined position (right).
completemeasurementsin1D,2Dand3D.Thedesignoftheplate
issuchthatthehemispherescanbeprobedfromaboveandfrom
below, and is shown schematicallyin Fig 8 The result of this
featureisthatareversalmethod,incorporatingclassicalrotation
reversals,can beappliedthat largely eliminatesthe systematic
deviationsoftheCMM.However,becausethereferencesurfaces
arehemispherical,measurementofafullhemispheretothe
equa-torisimpossible.Anymeasurementtakenonthehemisphereswill
onlyincludepointsto88◦fromthepole,resultingin96.5%coverage
ofthetotalavailablehemisphericalsurfacearea.Whenprobingthe
spherefrombelow,thisamountofcoverageisreducedtoabout5%
TheminiatureballplatehasbeencalibratedatPTBusingaZeiss
F25micro-CMM[48].Thetraceabilityofthemeasurementstaken
onthemicro-CMMisensuredusinggaugeblocks.Thefinal
cen-tretocentredistancesofthenineballsontheplatearequoted
withanuncertaintyof110nm(k=2).Acomparisonbetweenthe
currentresultsfromPTBandamicro-CMMcapableof
interfero-metriclengthmeasurementwascompletedattheendof2011,and
confirmedcomparablemeasurementsoftheballplatetowithin
100nm[49]
It hasbeennoted that this artefactis not suitablefor some
reversaltechniquesbecausethehemispheresarenot
symmetri-callyplacedontheplate[50,51]
Onelimitationontheuseofthishemisphereplatearisesfrom
thekinematicdesignofmanymicro-CMMs.Anumberofface
diag-onalsandspacediagonalscanbetestedusingthisartefactbytilting
it,asshowninFig.7.However,duetothehemisphericalnatureof
thereferencespheres,initsinclinedposition,thedefinitionofthe
spherecentresbecomeslessaccurate,asthemeasurementareais
furtherreducedbyupto25%
Withthecalibrationof thisballplatebeingperformedusing
amicro-CMM,thereislittlescopefordirectcomparisontoother
primarymeasurementtechniques,suchasthoseusedforgauge
Fig 8. The geometry of the hemispherical plate allows the hemisphere to be
mea-blockorstepgaugecalibration.Also,itcouldbeproblematicthat ballplateisintendedforuseasaverificationartefactfor micro-CMMswhenitisitselfcalibratedusingamicro-CMM
3.2.2 METASballplate
A ball plate for measurement on various contacting micro-CMMshasbeendevelopedbyMETAS[43].Itconsistsoftwenty-five precisionrubyspheres,nominally3mmindiameterthatpush-fit intoanInvarbaseplate,85mm×85mminsize.Animageofthis ballplateisshowninFig.9
Theballplateiscalibratedonaprecisionmicro-CMMatMETAS
byusinganerrorseparation technique.Thistechniquecalls for measurementstobetakenfrombothsidesoftheplate,sothedesign
oftheballplateallowsittobereversible.Thismethodofcalibration alsoallowsextrainformationaboutthemicro-CMMcompleting themeasurementtobecomputed,suchastheorthogonalityand straightnessoftheaxesandanyotherangulardistortions TheplateismanufacturedfromInvar,alowexpansion mate-rial.Toremoveresidualthermaldrift,allthespheresontheplate aremeasuredtwice,withthesecondsequencebeingmeasuredin reverseorder
Thedesignissimilartoclassicalmacro-scaleballplatesavailable commercially.However,thepushfitdesign,coupledwithstyliof shortlength,resultsinlittleornotiltanglebeingpossibleduring measurement
ThisartefactisoneofseveralusedduringtherecentEuramet project1088,‘Towardstruly3Dmetrologyforadvancedmicro-parts’
Fig 9. A ball plate manufactured from Invar with 25 precision ruby spheres Image courtesy of METAS, CH.
Trang 9writing
3.2.3 Krugerplates–column,ballandhole
Asetof2Dplateartefactshavebeendesignedandmanufactured
attheNationalMetrologyInstituteofSouthAfrica(NMISA,ZA)[51]
Thespecificaimofthedevelopmentwastobeabletoundertake
acompleteperformanceevaluationofseveralCMMs,both
meso-scaleandminiature,usingcontactingandopticalprobes
Aclassicalballplate,similartothatproducedcommercially[53]
wasmanufacturedonasmallscale.Thisballplatecanbeusedto
verifymicro-CMMsusingareversalmethod,becausetheprecision
spherescanbeaccessedfrombothsidesandarealsoarrangedina
symmetricalpattern
TwosimilarKrugerplatesweredevelopedtoaddresstheneed
forcrosscalibrationofcontactingandvisionCMMs.Thepreviously
mentionedsmallscale ballplateisunsuitablefor measurement
usingavisionCMMbecausetheopticalsensorisunabletoperform
automaticedgedetectionoftheprecisionspheres.Toovercome
thisproblem,twofurtherplatesweredevelopedthatincorporated
columnsandholes.Thesharpedgesofthesegeometricalfeatures
areeasilydetectedbyvisionCMMs,andtheplatescanbecalibrated
onamicro-CMMusingacontactingmicro-probe.Imagesofallthree
platesareshowninFig.10
During calibrationof the holeplate and ballplate, theyare
rotatedandinvertedtofacilitatereversalcalculationtechniques
Thethicknessoftheseplatesisapproximately4mm,allowing
con-tactingmicro-probes to access themeasurement features from
bothsides
3.2.4 PTBmicro-ballplates–smoothandroughened
AballplateconcepthasbeendevelopedbyPTBresultingintwo
separateartefactsthatactastransferstandardsbetweenCMMs
withcontactingandopticalsensors
Thefirstartefactconsistsof asetofthirty-sixstainlesssteel
spheres,2mmindiameter,arrangedina6×6array.Thefullarray
coversanareaofonly20mm×20mm.Noindicationof
measure-mentuncertaintyhasbeenfoundintheliterature,however,itis
estimatedthatthismeasurementstandardcouldbecalibratedto
anexpandeduncertaintyof500nm
Asecondmicro-ballplatehasbeenmanufacturedusing
thirty-six roughened balls, 500m in diameter, covering an area of
6.5mm×6.5mm[54].Thisartefactis a goodexample ofa ball
platesuitableforcross-checkingofmicro-CMMsusingcontacting
probingsystemsandopticaldistancesensors,especiallycontrast
detectionsystemssuchasvariablefocustechniques[55].Animage
ofthetwoartefactsisshowninFig.11
Themagnitudeofthemeasurementuncertaintyforbothplates
isunsuitableformicro-CMMs.Thishighuncertaintyismostlydue
tothehighformdeviationsofthespheres,especiallytheroughened
spheres,whichmaybeseveralhundrednanometres.Theneedfor
roughenedsphereswithlowroundnessdeviationshallbeessential
astheuseofopticalmicro-CMMsbecomesprevalent.Thisneedis
beingaddressedbyKefersteinetal.[56]throughthedevelopment
ofprecisionreferencespheresdesigned formulti-sensor
micro-CMMs
3.2.5 Siliconmicro-machineddimensionalcalibrationartefact
SandiaNationalLaboratories,USA,havedevelopedameso-scale
dimensionalartefactmanufacturedusing siliconbulk
microma-chining[57,58].Theartefactwasdesignedtobeusedtoevaluate
theperformanceofmicro-CMMsusingeithercontactingoroptical
sensors.Thishybridfunctionalitywasachievedbyusingafeature
geometry,flankedwallsandedges,whichcanbeprobedbyboth
contactingandopticalprobes.Imagesoftesteddesignsforthese
artefactsareshowninFig.12
Calibrationof theseartefacts, toanexpanded uncertaintyof
400nm(k=2),isundertakenusingamacro-scaleCMM,a micro-CMM, a high accuracy optical CMM and also a profile stylus instrument[59,60]
The seemingly high uncertainty associated with these arte-factsarisesfromtheiruseas calibratedtestlengths specifically forhighaccuracyoptical(orvideo)CMMs.Futuredevelopmentof theconceptisaimedatcalibrationoftheartefacttoanexpanded uncertaintyof100nm.Anyimprovementisdirectlydependanton improvedsiliconbulkmachiningtechniques
A1Dartefactisalsoavailableinthisproductfamily,actingasa lowcostoptiontothe2Dartefact.Anexamplestandardisshown
inFig.13 3.3 Otherartefacts Several miniaturegeometrical standards exist that although unsuitableascalibratedtestlengthsforacceptancetests,doserve
apurposeinthefieldofmicro-CMMcalibration,eitherforinterim checksorastaskspecificreferenceartefacts.Severalofthese stan-dardswillbedescribed,butthisisnotanexhaustivelist,asmany moreareknowntoexist,andarereviewedelsewhere[61] 3.3.1 Aztecstandard
A 3D dimensional pyramidal artefact has been designed at PTB, which is manufactured in silicon using micromachining fabricationtechniques[62].Thispyramidalartefact,orAztec stan-dard, hasseveralmeasurement pointsin thexy-plane and also
at differentz-heights.The overall dimensionofa 2×2 arrayis
13mm×13mm×1.4mm The Aztec standard consists of three structuredwafersandonenon-structuredsiliconwaferresulting
infourreferencez-planesandvariousothertiltedplanes(angledat 54.7◦).Withacontactingprobingsystemitispossibletodetermine thepositionsof theinclinedplanes,and subsequently calculate theirintersectionpoints,resultingin3Dreferencepoints.ASEM imageandaphotographoftheAztecstandardareshowninFig.14 3.3.2 Micro-holestandard
ThePTBmicro-holestandardisataskspecificartefactfocusing
onmicro-holes,suchasinjectionnozzlemeasurement[61].The calibrationofthismicro-holeisdifficultbecauseitshighaspect ratioisbeyondthecapabilityofmostmicro-probes.Themaintype
ofprobethisartefactisdesignedforisopto-mechanicalprobes,
orfibreprobes,which,althoughprovidehighaspectratioprobing, aresubjecttomanygeometricaleffects[63].Afurtherissueisthat the measurement uncertainty of these opto-mechanical probes increasesastheydescendintohighaspectratioholes
Tosolvetheissueofcalibration,themicro-holestandardis man-ufacturedfromseveralsheets,eachcontainingawell-positioned hole Theseindividual holes are then calibrated separately and assembled to create a continuous and well calibrated micro-hole Thealignment of the individualsheets is defined by two vee-grooves,oneforrotationalpositioningandoneforheight posi-tioning Aschemaof the conceptand an imageof therealised standardareshowninFig.15
3.3.3 Polytecstepheightstandard Testingthez-axisofamicro-CMMisdifficulttoperformdue
tothelackofprobe/styluscombinationswithorientationsother thanvertical,andbecausetestsusinggaugeblockscanbeverytime consuming.Asuitablesolutionforthisproblemcouldbeastep artefactcoveringa certainrangeofthez-travelof themachine
AstepartefacthasbeendevelopedbyPolytecGmbH[64],andis showninFig.16
Thisfinestepheightartefactcoversarangeof2.25mmin nine-teenstepsofheight0.125mm.Althoughinitiallydesignedforthe
Trang 10Fig 10.The three Kruger plates: ball plate (left), cylinder plate (middle) and hole plate with ceramic inserts (right) Image courtesy of the NMISA, ZA.
Fig 11.Two ball plate artefacts suitable for measurement on a contacting micro-CMM and various optical systems Image courtesy of PTB, DE.
Fig 12.Two tested designs of the silicon micro-machined dimensional calibration artefact, left [59] and right [60] Image courtesy of Sandia National Laboratories, USA.
Fig 13.A 1D artefact for cross-calibration of optical and contacting micro-CMMs Image courtesy of Sandia National Laboratories, USA.
Fig 14. The Aztec standard, fabricated through the micromachining of silicon Image courtesy of PTB, DE.