Drilling thick fabric woven CFRP laminates with double point angle drills

(1)

ContentslistsavailableatSciVerseScienceDirect

Journal

of

Materials

Processing

Technology

j o ur n a l ho me p age :w w w . e l s e v i e r . c o m / l o c a t e / j m a t p r o t e c

Drilling

thick

fabric

woven

CFRP

laminates

with

double

point

angle

drills

Yi˘git

Karpat

a,∗

_{, Burak}

_De˘ger

b

_{, Onur}

_Bahtiyar

b a_Bilkent_University,_Department_of_Industrial_Engineering,_Ankara_06800,_Turkey b_Turkish_Aerospace_Industries_Inc.,_Ankara_06980,_Turkey

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received18February2012

Receivedinrevisedform22April2012 Accepted25May2012

Available online 2 June 2012 Keywords:

Machining Drilling

Carbonﬁberreinforcedplastics Diamondcoatedcarbide

a

b

s

t

r

a

c

t

Carbonﬁberreinforcedplastics(CFRPs)havemanydesirableproperties,includinghigh

strength-to-weightratio,highstiffness-to-weightratio,highcorrosionresistance,andlowthermalexpansion.These

propertiesmakeCFRPsuitableforuseinstructuralcomponentsforaerospaceapplications.Drillingis

themostcommonmachiningprocessappliedtoCFRPlaminates,anditisdifﬁcultduetotheextremely

abrasivenatureofthecarbonﬁbersandlowthermalconductivityofCFRP.Itisachallengefor

manufac-turerstodrillCFRPmaterialswithoutcausinganydelaminationontheworkpartwhilealsoconsidering

theeconomicsoftheprocess.ThesubjectofthisstudyisthedrillingoffabricwoventypeCFRP

lami-nateswhichareknowntobemoreresistanttodelaminationthanunidirectionaltypeCFRPlaminates.

Theobjectiveofthisstudyistoinvestigatetheinﬂuenceofdoublepointangledrillgeometryondrilling

performancethroughanexperimentalapproach.Anuncoatedcarbideandtwodiamondcoatedcarbide

drillswithdifferentdrilltipanglesareemployedindrillingexperimentsofaerospacequalitythick

fab-ricwovenCFRPlaminates.Forceandtorquemeasurementsareusedtoinvestigateappropriatedrilling

conditionsbasedondrillgeometryandidealdrillingparametersaredetermined.Toollifetestsofthe

drillswereconductedandtheconditionofthediamondcoatingisexaminedasafunctionofdrilling

operationalparameters.Highfeedratedrillingexperimentsareobservedtobefavorableintermsofdrill

wear.Feedisobservedtobemoreimportantthanspeed,andtheupperlimitoffeedisdictatedbythe

drilldesignandtherigidityofthemachinedrill.Holediametervariationduetodrillwearismonitored

todeterminedrilllife.Athighfeeds,holediametertoleranceisobservedtobemorecriticalthanhole

exitdelaminationduringdrillingoffabricwovenCFRPlaminates.

1. Introduction

Lightweight,durable,andcorrosionresistant,carbonfiber rein-forcedplastics(CFRPs)havebeenincreasinglyusedintheaerospace industrytobuildmorereliableandfuelefficientaircrafts.Currently, fuelcostiscalculatedtobearound32%oftheairlines’operating cost,whereasitwas14%in2003(IATA,2012).Airlineshavereacted tothisfinancialpressurebyreplacingolderaircraftswithanew generationofaircraftsthataremorefuelefficient.Thisefficiency inpartcomesfromincreaseduseofsophisticatedmaterials.

Compositepartsaremanufacturednear-netshape;however, additionalmachiningoperationssuchasdrillingandmillingmay berequiredtomeetfinaldesignspecifications.Thesubjectofthis studyisdrilling,whichisthemostcommonmachiningprocess appliedtocompositelaminates.WhiledrillingCFRPlaminates, cut-tingdrillsrapidlywearoutduetothehighlyabrasivenatureof thecarbonfibersandthelowthermalconductivityofCFRP.The heatgeneratedduringdrillingislocalizedatthecuttingdrilledge, therebycausingrapiddrillwear.Delaminationisacrucialproblem

∗ Correspondingauthor.Tel.:+903122902263;fax:+903122664054. E-mailaddress:ykarpat@bilkent.edu.tr(Y.Karpat).

whendrillingCFRPlaminates,sinceitdecreasestheloadcarrying capabilityofthecompositesbyseparatingtheplies.Delamination iscloselyrelatedtowearandcontrolledduringmachining opera-tions.Dimensionalintegrityoftheholesandholesurfacequality arealsoimportantconsiderations.

Intheliterature,studiesonmachiningCFRParelimited com-paredtostudiesofmetals;however,duetoincreasingusageof thismaterialintheaerospaceindustrythenumberofstudieson machining CFRPlaminates hassignificantly increasedin recent years.Koplevetal.(1983)reportedthesignificanceoffiber direc-tiononthechipformationmechanismandobservedthatbrittle fractureisthemaincauseofchipformation.Caprinoetal.(1998) investigated machining forces during orthogonal machining of CFRPs.Theyobservedthat forcesaremainlydue tothecontact betweenworkmaterialandtheflankfaceofthedrill.Hochengand Dharan(1990)andZhangetal.(2001)developedanalyticalcutting modelsfortheorthogonalmachiningofCFRPlaminatestopredict cuttingforces.Theyusedthesemodelstopredictthecriticalthrust forcebeyondwhichdelaminationoccurs.

Experimentalstudieshavebeenpursuedtobetterunderstand the relationships between process inputs, such as machining parametersanddrillgeometry,andprocessoutputssuchascutting forces,torque,toolwearanddrilledholequality.Inanextensive 0924-0136/$–seefrontmatter © 2012 Elsevier B.V. All rights reserved.

(2)

Fig.1.(a)FiveharnesssatinweaveCFRPand(b)CFRPlaminatecomposition.

experimentalstudy, Chen (1997)conducted tests toreveal the relationshipbetweenmachiningparametersanddelamination.A linearrelationshipwasobservedbetweenthrustforceand delam-inationwhen drilling unidirectionalCFRP composite laminates. MultidirectionalCFRPlaminatesare foundtobemoreresistant todelamination than unidirectionallaminates. It was observed thatdelaminationincreaseswithincreasingﬂankwear.Tsaoand Hocheng(2005)andHochengandTsao(2005)reportedthepositive inﬂuenceofusingbackupplatesondelamination.Itincreasesthe criticalthrustforcevalue,henceallowingdrillingathigherfeeds. Inanefforttoimprovedrillperformance,Piquetetal.(2000) ana-lyzedtheeffectsofdrillgeometryonholequality.Theyreported thatholequalitycanbeimprovedbyapplyingavariablefeedrate whilemachining.DharanandWon(2000)proposedanintelligent controlschemebasedonanexperimentalmodelofthethrustforce andtorqueasafunctionofcuttingparameters.Shyhaetal.(2009) investigatedtheeffectofdrillgeometryandoperatingparameters whendrillingsmalldiameterholeswitharobotmanipulator.They foundthatdrillgeometryandfeedratearethetwomainfactors affectingdrilllife.Theyfoundthatanuncoatedcarbidestepped drillwith140◦pointangleyieldedthebestdrilllifeperformance. RawatandAttia(2009)utilizedthe“machinabilitymaps”approach toselectcuttingconditionswhendrillingfabricwoven compos-ites.Intheirapproach,delaminationlevel,holediametererror,hole circularityerror,andsurfaceroughnessinsidetheholewereall con-sideredasprocessoutputsdependentonspindlespeedandfeed rate.Theyconcludedthathighcuttingspeedsreducethethrust forceduetothermalsoftening.Farazetal.(2009)studiedthewear behaviorofuncoatedandcoatedcarbidedrillswhilemachining fabricwovenCFRPcompositelaminatesunderdrydrilling condi-tions.Theyshowedthatwhenwornout,thecuttingedgeassumesa roundedshapethathindersitsabilitytocutthematerialeffectively. Theyalsoobservedalinearrelationshipbetweenedgerounding andcutting forces.Iliescuet al.(2010) proposeda wearmodel basedonthethrustforceandmachiningparametersfordrilling CFRP. They concluded that while carbide drills represent wear behavioraccordingtopowerlaw,diamondcoateddrills exhibit linearbehavioratthebeginningbutpowerlawtowardtheend ofdrilllife. Theyrecommend usingdiamond coateddrills hav-ing125–130◦pointangle,withoptimumcuttingspeed170m/min andfeed0.05mm/rev.Karpatetal.(2010)presentedexperimental dataonthewearbehaviorofdiamondcoatedcarbidedrills, show-ingthefracturingofdiamondcoatingandedgeroundingofthe carbidesubstrate.LazarandXirouchakis(2011)investigatedthe cuttingforces basedondrillgeometriesandcuttingparameters usingabackingplate;theycalculatedforcedistributionsalongthe cuttingedgesatlowfeedsandalsoconcludedthatfeedanddrill geometryarethemostimportantparameters.Literaturereviewson

machiningcompositesgivenbyTeti(2002)andAbraoetal.(2007) emphasizetheimportanceofdevelopingbetterdrillgeometries andmachiningstrategies.

In this study,uncoatedcarbide and diamondcoated carbide drills having double point angle geometries are tested during drillingoffabricwoventypeCFRPlaminatesunderpractical con-ditions.Experimentaltestswereperformedonaerospacequality thickCFRPmaterialusingastateoftheartmachinetoolspeciﬁcally designedforcompositemachining.Anexperimentalanalysishas beenconductedtoinvestigatetheinﬂuenceofcuttingconditions anddrillgeometryondrillingforcesandtoolwear.

2. Experimentation

ExperimentalstudieswereconductedonaDörriesScharmann Technologies5-axisprecisionmachiningcenter withmaximum 24,000rpmrotationalspeed.FabricwovenCFRPlaminateshave a total of 30 layers, composed of 0◦/90◦ (14 layers) and ±45◦ (16layers)ﬁberorientationsasshowninFig.1.TheCFRPplates wereultrasonicallyinspected forfaults beforeexperimentation. ThephysicalandmechanicalpropertiesoftheCFRPlaminateused inthisstudyaregiveninTable1.

TheCFRPplatehad asquareshape withedgedimensionsof 920mmandthicknessof10mm.Thedistancebetweenhole cen-tersistakenasequaltotwotimestheholediameter.Analuminum backingplatehaving8mmdiameterholeswasused.All experi-mentswereconductedunderwetdrillingconditionsinorderto preventcarbondust.Thecuttingﬂuidiswatersolubleoil(Fuchs Ecocool).Duringdrillingexperimentsthethrustforceandtorque weremeasuredbyaKistler9123rotatingdynamometerandits chargeampliﬁer.Thecuttingforcedata(torqueTandforcesinthree directionsFx,Fy,andFz)werecollected(samplingrateof25kHz) througha dataacquisitionsystemand processedona personal computer.TheexperimentalsetupisshowninFig.2.

Uncoatedanddiamondcoatedcarbidedrillswereusedinthe experiments.Drillsemployedinthisstudyhavetwistdrill geom-etrywithdoublepointangles.Theyareobtainedfromthesame manufacturer. Table 2 shows experimental conditions used in drilling force analysis and drill tip angles. Drill tip angles are measuredwithanopticaltoolmeasurementdevice(Zoller Ventu-rion450/6).UCandDCC-IIdrillshavethesamegeometry.DCC-I and DCC-IIarediamond coatedcarbide drillswithdifferenttip anglesanddifferentprimary(OA)andsecondary(AB)cuttingedge lengths.

ThegeometryofthedoublepointangledrillisshowninFig.3(a) andacomparisonoftooltipgeometriesareshowninFig.3(b).The helixangle(),therakeangle(),andtheclearanceangle()of thedrillsare30◦,7◦,11◦,respectively.Aregiononthedrillnearits

(3)

Table1

MaterialpropertiesofCFRPlaminates.

Material Fibervolume(%,v/v) Strength(MPa) Modulus(GPa) Density(g/cm3₎ _Cured_ply_thickness_(mm)

Epoxyimpregnatedgraphitefabric/5-harnesssatina ₅₀ ₅₂₀ ₅₂ _1.485 _0.33

a_Mechanical_properties_of_the_laminate_are_valid_for_both_warp_and_ﬁll_directions_at₂₃◦_C.

Table2

RangeofcuttingconditionsusedindrillingforceanalysisonfabricwovenCFRP.

Acronym Toolcoating Diameter(mm) Drilltipangles(2˛,2ˇ) Feed(␮m/rev) Rotationalspeed(rpm)

UC Uncoatedcarbide 6.35 140–60◦ 40,60,100 5000,7500,10,000

DCC-I Diamondcoatedcarbide 6.91 130–60◦ 40,60,100 5000,7500,10,000

DCC-II Diamondcoatedcarbide 6.38 140–60◦ _40,_60,_100,₁₅₀ _5000,_7500,_10,000

chiseledge(OAasshowninFig.3(b))isgroundinordertoimprove

drillingperformanceatthetipofthedrill.Fig.3(a)alsoshowsthe differentialcutting(dFc),thrust(dFz)andtangential(dFt)forces actingonthedrill.Multiplyingthedifferentialcuttingforceand thedistancefromthecentergivesthedifferentialtorque.Adding togetheralldifferentialthrustforcesandtorquesalongthecutting edgesyieldstotaltorqueandthrustforceduringdrilling.

3. Experimentalinvestigationofthrustforcesandtorques

A fundamental step in understanding the drilling mechan-ics of CFRP material is the investigation of cutting forces. As

Fig.2. Experimentaltestsetup.(a)FabricwovenCFRPplateand(b)CFRPplateand aluminumbackingplateassembly.

Fig.3. (a)Doublepointangledrillgeometryand(b)proﬁlesofthedoublepoint angledrills.

(4)

Fig.4.Thrustforce(N)andtorque(Ncm)measurementsduringdrillingwithdoublepointdrillgeometryandcharacteristicthrustforcelocations.

mentionedinSection1,theforcesduringdrillinghavebeenthe sub-jectofmanystudiesintheliterature.Cuttingforcemeasurements havebeenusedonlineand/orofﬂinetomonitordrillcondition. Themagnitudeanddevelopmentofdrillingforcesaredependent onmechanicalpropertiesofthecompositematerial,operational drillingparameters (rotational speed and feed), drillgeometry, anddrillmaterial/coating.Fig.4showsatypicalthrustforceand torquemeasurement,wheredrillgeometrydependent character-isticthrustforceandtorquelocationsareindicated.

Duetodoublepointangle,theentranceregionhastwodifferent slopeswheretheentry(OA)valueisdenotedasFz-enterandthepeak (OB)valueisdenotedasFz-peak.Torquevaluereachesitspeakas soonasthedrillfullyengages(OB)withthematerial.Thethrust forcevaluesmeasuredintheentryregionindicatetheinﬂuence ofdrilltipanglesonthrustforces.Asuddendropinthrustforces isobservedjust afterthedrilltipleavesthematerial.Sincethe outercuttingedge,whichiscarryingthetorque,wouldstillbein contactwiththematerial,dropintorquefollowsthrustforceafter ashortdelay.Similartotheentryregion,twodifferentslopesare alsoobservedontheexitregion.Thethrustforcevalueimmediately beforethedrilltipleavesthematerialisnamedasFz-int.Thethrust forceattheendofthesecondslopeisnamedasFz-exit.Theamount ofthrustforcesFz-intandFz-exitareimportantsincetheyindicate themagnitudeoftheforcewithwhichthedrillpushesthebottom layersofthelaminate.Thecharacteristicthrustforceandtorque locationscanbeidentiﬁedusinggeometricalmeasurementsofthe drill,drillingfeedrate,andtimehistoryofthethrustforcesand torques.Itmustbenotedthatthelocationoftheforcepointsis relatedtothedrillanglesandcuttingedgelengths.Foragivendrill tipangle,alargerprimarycuttingedgemovesthelocationofthe peakforcetowardtheleftonthetimeaxisandupwardsonthe thrustforceaxis.ThedecreasebetweenFz-peakandFz-intisrelated bothtothethicknessofthelaminateandthefeedvalue.

Fig.5representsexperimentalforceandtorquemeasurements whendrillingwithuncoatedcarbidedrills asafunctionoffeed underconstantrotationalspeedof5000rpm.Alinearrelationship betweenfeedandthrustforcesandtorquesexistswithinthe exper-imentalrange.

Fig.6representstheinﬂuence ofcutting speed andfeedon thrustforcesandtorquewhendrillingwithDCC-Idrill.InFig.6(a),

thelinearrelationshipbetweenthrustforcesandfeedisshownfor threedifferentrotationalspeeds.Fig.6(b)showstorque measure-mentsasafunctionoffeedandspeed.

Fig.7(a)and(b)showsthesameobervationsforDCC-IIdrill.It canbeconcludedthattheinﬂuenceofcuttingspeedisnot signif-icantonthethrustforcesandtorquesfordiamondcoatedcarbide drills.HigherthrustforcesandtorquesaremeasuredwithDCC-II whichhasalargertipangle.

(a) (b) 0 10 20 30 40 50 60 70 80 0 0.02 0.04 0.06 0.08 0.1 0.12 Thr u st For ce , Fz (N) Feed (mm/rev) F z_peak F z_exit F z_int 0 0.05 0.1 0.15 0.2 0.25 0.3 0.12 0.1 0.08 0.06 0.04 0.02 0 Tor q ue, T (Nm ) Feed (mm/rev)

Fig.5.Comparisonofthrustforceandtorquemeasurementsatthreedifferentfeed levelsforuncoatedcarbidedrill.

(5)

(a) (b) 0 20 40 60 80 100 120 140 160 0.12 0.1 0.08 0.06 0.04 0.02 0 Thr u st Force, Fz (N) Feed (mm/rev) Fz_peak (5000 rpm) Fz_int (5000 rpm) Fz_exit (5000 rpm) Fz_peak (7500 rpm) Fz_int (7500 rpm) Fz_exit (7500 rpm) Fz_peak (10000 rpm) Fz_int (10000 rpm) Fz_exit (10000 rpm) 0 0.05 0.1 0.15 0.2 0.25 0.3 0.12 0.1 0.08 0.06 0.04 0.02 0 Tor que, T (Nm) Feed (mm/rev) Torque (Nm) -5000 rpm Torque (Nm) -7500 rpm Torque (Nm) -10000 rpm

Fig.6.Thrustforce(a)andtorque(b)measurementsasafunctionoffeedandspeed forDCC-I.

Fig.8showsthecomparisonofthrustforcesforeachdrilling drillexaminedinthisstudy.UncoatedcarbideandDCC-IIhavethe samegeometry,butthedifferencesinthrustforcesarequitelarge duetothickdiamondcoatinganditsroundingeffectonthecutting edges.DCC-I,withasmallertippointangle,yieldedlowerthrust forcesthanDCC-IIeventhoughithasan8%largerdrilldiameter. ThelowestexitthrustforcesarerecordedwithDCC-Iwhichhas

(a) (b) 0 50 100 150 200 250 0.2 0.15 0.1 0.05 0 Thr u st For ce, Fz ( N ) Feed (mm/rev) Fz_peak (5000 rpm) Fz_int (5000 rpm) Fz_exit (5000 rpm) Fz_peak (7500 rpm) Fz_int (7500 rpm) Fz_exit (7500 rpm) Fz_peak (10000 rpm) Fz_int (10000 rpm) Fz_exit (10000 rpm) 0 0.050.1 0.150.2 0.250.3 0.350.4 0.2 0.15 0.1 0.05 0 Torque, T (N m ) Feed (mm/rev) Torque (Nm) -5000 rpm Torque (Nm) -7500 rpm Torque (Nm) -10000 rpm

Fig.7.Thrustforce(a)andtorque(b)measurementsasafunctionoffeedandspeed forDCC-II.

Fig.8. ComparisonofthrustforcesforeachdrillatN=5000rpm,f=40␮m/rev.

ashortersecondarycuttingedge.Theforceandtorque measure-mentsobtainedfordiamondcoatedcarbidedrillsareinvestigated indetailinthenextsection.

3.1. Inﬂuenceofdrilltipanglesondrillingforces

Fig.9showsthethrustforcesactingonadoublepointangledrill andunchipthicknessvariationwithrespecttocuttingedges.Thrust forcemeasurementscanbedividedbytwotocalculateforcesper edge.Thetotalthrustforceactingonthedrillisthesumofthrust forcesactingonthreedifferentsections.Thethrustforceinthe primaryregion(OA)isthesumofchiseledgeforce(Fch)andprimary edgeforce(Fpr).Thethrustforceinsecondaryregion(Fsc)canbe foundbyconsideringpeakandentrythrustforces.Thrustforcesin eachsectionarerepresentedinEq.(1).Thedrilltipangles˛andˇ andthelengthsoftheprimaryandsecondaryedgesaredrilldesign variables.Ifthediameterofthedrillandtheheightofthedrilltip areknown,thelengthsoftheprimaryandsecondaryregionscan becalculated.

Fz-OA= Fz-enter₂ =Fch+Fpr

Fz-AB= Fz-peak−₂Fz-enter =Fsc

(1)

Theuncutchipthicknessesatprimaryandsecondarycutting edgescanbecalculatedasshowninEq.(2),wherefisthefeedper revolution.Foragivenfeed,increasingtipangleincreasesuncut chipthicknessintherespectivearea.Therefore,increasingtipangle

F

r

α

β

O

A

B

pr ch sc OB OA

f/2

t

₂

A

1

B

Fig.9.Thrustforcesactingonthedrill(righthandside)anduncutchipthickness variation(lefthandside).

(6)

(a) (b) 0 200 400 600 800 1000 1200 1400 1600 0.04 0.06 0.1 0.15 Speciﬁc C u ng Force (Kf) (N/mm^2) Feed (mm/rev) DCC-I (OA) DCC-I (AB) DCC-II (OA) DCC-II (AB) 0 200 400 600 800 1000 1200 0.04 0.06 0.1 0.15 Speciﬁc Cung Force (Kt) (Nmm/mm^2) Feed (mm/rev) DCC-I (OA) DCC-I (AB) DCC-II (OA) DCC-II (AB)

Fig.10.Speciﬁccuttingforcescalculatedfor(a)thrustforceand(b)torqueasa functionoffeedperrevolution.

isusuallyaccompaniedbyadecreaseintheedgelengthinorderto keepthetotalareasameunderthecuttingedge.

t1=

_f 2

sin(˛), t2=

_f 2

sin(ˇ) (2)

Theareaoftheuncutchipunderneathprimaryandsecondary edgescanbewrittenasinEq.(3).

AOA= rOA

sin(˛)t1, AAB= rOB−rOA

sin(ˇ) t2 (3)

ThenEq.(4)canbeusedtocalculatetheaveragespeciﬁccutting forces(relatedtothrustforceandtorque)separatelyforthe pri-maryandsecondarycuttingedges.Torquemeasurementsarealso dividedbytwotocalculateperedgetorquemeasurements.

Kf−p= F_Az-OA OA, Kf−s= Fz-AB AAB Kt−p=₍_r TOA OA/2)AOA, Kt−s= TAB ((rOA+rOB)/2)AAB (4)

Fig.10showsaveragevaluesofspeciﬁccuttingforcescalculated forDCC-IandDCC-IIdrillsat5000rpm.Adecreasingtrendin cut-tingforcesisobservedwithincrerasingfeed.Thisphenenomenon isknowninmachiningassizeeffect.Cuttingedgeradiusandwork materialpropertiesarebelievedtobereasonsforsizeeffect.

DrillingperformancesofDCC-Iand DCC-IIgeometriescanbe comparedbasedonthespeciﬁccuttingforces(KfandKt) calcu-latedat the primary and secondary cuttingedges. Under ideal drillingconditions for a given drill geometry, thrust force and torquemustbedistributedevenlybetweenprimaryandsecondary drillingedges.Ifthesecondarycuttingedgeofthedrillisdesigned tobesmall,thenexitthurstforceswillbesmall.However,athigh

0 20 40 60 80 100 120 0 0.02 0.04 0.06 0.08 0.1 0.12 Thr u st Force, Fz ( N ) Feed(mm/rev) Fz peak Fz exit

Fig.11. Thrustforcesmeasurementsduringdrillingwithapilothole(N=5000rpm).

feedsthetorquecarryingcapabilitywilldecrease.InFig.10(a),itis seenthattheprimaryedgeofDCC-Icarriesmorethrustloadthan thesecondaryedgeduetoitslargersize.Thetorqueloadis car-riedequallyatlowfeedlevel.Asfeedincreases,torqueismainly carriedonthesecondaryedgeduetoitsshorterlengthasshownin Fig.10(b).AsforDCC-II,around100␮m/rev,itdistributesthethrust forceandtorquealmostevenlybetweenprimaryandsecondary edges(Fig.10(a)and(b)).Ingeneral,cuttingcoefﬁcientsaresmaller forDCC-Iduetosmallertipangle,thereforeDCC-Imaybeabetter optionfordrillingthinCFRPlaminates,wherelowfeedsareusually usedtokeepthrustforceslow.DCC-IIgeometrycanbeemployedin drillingthicklaminatesathigherfeeds.Themechanicalproperties oftheCFRPlaminatesdependonmanyfactors,includingcarbon

Fig.12.SEM images ofDCC-II drills (a)N=10,000rpm,f=40␮m/rev and(b) N=10,000rpm,f=100␮m/rev(afterdrilling1728holes).

(7)

Fig.13.SEMimagesoftheﬂankfaceofDCC-Icuttingdrills:(a)N=15,000rpm,f=75␮m/rev,(b)N=15,000rpm,f=225␮m/rev,(c)N=5000rpm,f=225␮m/rev,(d) N=5000rpm,f=75␮m/revand(e)holeexitscorrespondtoeachexperimentalcondition.

ﬁberpercentage,carbonﬁberdiameter,typeofresinusedinthe matrix,curingconditions,etc.Developinganoptimizeddrill geom-etrywhichwillperformwellatallcompositematerialsmaynot bepossible,asexplainedabove,butdeterminingfavorabledrilling conditionsforagivendrillgeometryorselectingdrillgeometry basedonlaminatethicknessispossible.

Inordertoinvestigatetheinﬂuenceofthetipofthedrillon thethrustforces,drillingexperimentswerealsoconductedon pre-drilledholesof2.5mm.Thrustforcemeasurementsareshownin Fig.11.Thediameterofthepilotholeisselectedsothatitmatches

thesizeoftheregionOA(seeFig.3).Peakforcesbecomestationary afterthefeedvalueof40␮m/revwhileexitthrustforcecontinues increasingatalowerratethanitdoeswhendrillingwithouta pre-drilledhole.ThecontributionoftheregionOAonthethrustforce measurementscanbecalculatedas50%basedonmeasurements giveninFigs.7and11.

ItmustbenotedthatthechipformationprocessinCFRP lam-inatesconsistsofshearingandbrittlefracturingoftheﬁbers.As feedincreases,duetolargerloads,ﬁbersarecrushedmore eas-ilyanddominantchipformationmechanismbecomesthebrittle

(8)

Fig.14.SEMimagesofthefractureddiamondcoatingzone.

fracturingoftheﬁbers.Thismayexplainplateauincuttingforces withincreasingfeedasshowninFig.11.

4. Investigatingtherelationshipsbetweendrillwearand drillingoperationalparameters

Diamondcoatedcarbideanduncoatedcarbidedrillsareusedin thisstudy.Diamondcoatingisgrownonthesurfaceofthecutting drillthroughchemicalvapordeposition,wherehydrogenandagas containingcarbonaremixedinsideachamberatveryhigh tem-peratures.Thethicknessofthecoatingandthesizeofthediamond particlesareknowntobeimportantintermsofdrillperformance. Diamondcoatingcanbeappliedtoawiderangeofdrill geome-tries;however,thecoating’sthicknessdecreasesthesharpnessof thecuttingdrilledge,whichisnotdesirablewhenmachiningCFRP laminates.Thissectioninvestigatestheconditionofthediamond coatingonthedrillasafunctionofdrillingparameters.Fig.12(a) and12(b)showsthephotosoftheDCC-IIdrillafterdrilling1728 holesat40and100_␮m/revat10,000rpm,respectively.Diamond coatingonthedrillisobservedtobedamagedinbothdrillingcases. Lowfeeddrillingcase(Fig.12(a))resultedinmoreseveredamage

Fig.15.(a)ForcevariationinDCC-IIdrillwithrespecttonumberofholesand(b) worndrillthrustforceswithrespecttodrillingparameters.

tothedrill.Thediamondcoatingismostlyfracturedfromtheﬂank faceofthedrill,andthefracturedareaislargerinthesecondary region.Fig.12(b)showsthatinhighfeeddrillingcase,fractureon thecoatinginitiatesfromthesecondaryregionandgrowstoward theprimaryregion.Atlow feeddrillingcase, thetotaldistance thedrilltravelswithintheholeis2.5timeslargerthanthehigh feedcase. Whenthediamondcoatingisfractured,thrustforces andtorquesincrease.Asaresult,thebottompliesofthelaminate arepushedoutbylargerthrustforces,andeventuallydelamination occurs.

In order to investigatethe inﬂuence of largerfeed rates on drillingperformance,testsareperformedwithDCC-Idrillsunder alargerspeedandfeedexperimentalrange(5000-15000rpmand 75–225␮m/rev).Fig.13revealsthewearzoneattheﬂankfaceof thedrillsafterdrilling1728holes.

Similarly,fracture ofthediamondcoatingismore severeon theflankfacethantherakefaceofthedrill.Thesmallestfracture areaonthediamondcoatingontheflankfacewasobservedinthe experimentalcaseofN=15,000rpm,f=225␮m/rev(ExperimentB) wherealargefeedwastakenathighcuttingspeed(Fig.13(b)). Dia-mondcoatingonthedrillwassignificantlyfracturedattheouter edgeofthecuttingedge,possiblyduetohightorque,fortherewas nosignificantdelaminationattheexitoftheholesasshownin Fig.13(e).Coatingontheprimarycuttingedgeand chiseledge ofthedrillisintact.Theexperimentalcasewiththelowestfeed rate(ExperimentD)yieldedtheworstexperimentalperformancein termsofdiamondcoatingfracturezoneareaandresultedin prob-lemsattheexitoftheholeduetoexcessiveedgeroundingand ineffectivecuttingperformanceasshowninFig.13(e).Highfeed andlowspeeddrillingcase(Fig.13(c))yieldedasmallerfractured areaofthediamondcoatingcomparedtolowfeedandhighspeed drillingcase(Fig.13(a)).Fig.14showsdetailedSEMimagesofthe fracturedcoatingzonethatcorrespondtoexperimentalcase(D) showninFig.13.Thethicknessofthediamondcoatingismeasured tobe9–10␮m.

(9)

0 0.5 1 1.5 2 2.5 3 3.5 4 0 20 40 60 80 100 120 140 160 Time (s) Thrust Force (N) 1st hole 2nd hole 4th hole 2nd Hole 1st Hole 4th Hole

(b)

(a)

0 50 100 150 200 250 1 3 5 7 9 1113151719212325272931 Thrust Force (N) Number of Holes Fz_peak (5000 rpm) Fz_int (5000 rpm) Fz_exit (5000 rpm) Fz_peak (10000 rpm) Fz_exit (10000 rpm) Fz_int (10000 rpm)

Fig.16.(a)Timehistoryofthrustforcemeasurementsasafunctionofnumberofholesand(b)inﬂuenceofrotationalspeedonthrustforces(atf=40␮m/rev).

Fig.15(a)showsthevariationinthrustforceswithrespecttothe numberofdrilledholesforDCC-IIafterdrilling1728holesattwo differentfeedswithconstantrotationalspeedof10,000rpm.Forces increaseaccordingtopowerlawandcontinueincreasinglinearly. Forthesamenumberofholes,highfeeddrillingcaseresultedin lowerthrustforcesthaninthelowfeeddrillingcase.Consequently, drillingatlowfeedinitiallyresultsinlowerexitforces,butdue tofasterdrillwear,thrustforcesbecomeevenlargerafterdrilling thesamenumberofholes.Fig.15(b)representsthecomparison ofthrustforcesmeasuredafterdrilling1728holesunderdifferent drillingconditions.Thepositiveinﬂuenceofhighspeedandfeed canbeseen.Thedifferencesbetweenhighandlowspeedwhen drillingatahighfeedarenotlarge.At5000rpmand75␮m/revfeed drillingconditionwithDCC-Idrill,peakthrustforcewasmeasured as1553Nafterdrilling1728holes.

Fig.16(a)representsexperimentalforcemeasurementswhen drillingwithuncoated carbide cutting drills.After drillingonly four holes, its thrust force value surpasses that of the DCC-II withthesamedrillgeometry.Therateofincreaseinpeakthrust forcesisquitehigh,indicating rapiddrillwearespeciallyatthe primaryregion ofthe drill(Fz-enter and Fz-peak).Theincrease in exitthrustforceisrelativelylowcomparedtoothercharacteristic thrustforcepoints,indicatingalowerwearrateatthesecondary cuttingedge.Fig.16(b)showstheinﬂuenceofspeedondrilling forces,wherethevariationofcharacteristicthrustforcepointsis shownwithrespecttonumber holesatconstant feed.Apower

lawequationcanbeusedtorepresenttherelationshipbetween thrustforceand numberofholesfortheuncoatedcarbidedrill. Basedonholeexitqualityforuncoatedcarbidedrillsat5000rpm and 40␮m/rev theacceptable drill life was observed tobe 20 holes, aroundwhich theexitthrustforcewas50N (Fig.16(b)). Thisthrustforcevalue(Fz-int)isnamedasthecriticalthrustforce value.

Inordertodeterminedrilllife,holeexitsareusuallyinspected fordelamination.Somevisualtechniqueshavebeenproposedin theliteratureinordertodecidewhetherimperfectionsatthehole exithavereachedanunacceptablelevel.Thesetechniquesmeasure thediameterofthedelaminatedareausingacameraandcalculate anindexrelativetonominalholediameter(Schulzeetal.,2010). Inspectionofdelaminationrequiresadetailedexaminationofthe holesandalsoremovaloftheCFRPlaminatefromthefixturein ordertocheckholeexits.Ontheotherhand,holediameterscan easilybecontrolledduringproductionbyusinggaugesor microm-eters.Inordertoinvestigatetheinfluenceofhighfeedsonhole quality,holediametersareautomaticallymeasuredusinga coor-dinatemeasurement machine(DEA-DeltaSlant) every96holes. Fig.17showsthediametermeasurementvariationsfordifferent drillingconditions.Thedecreaseindiameterisacceptableaslong asitstayswithinspecifiedtolerancesdefinedbythemanufacturer. Adecreasingtrendinholediameterswithrespecttothenumber ofdrilledholesisclearlyobserved.Thelowerlimitofacceptable holediameterisassumedtobe6.35mmwhereDCC-IIdrillwith

(10)

(a)

(b)

6.2 6.25 6.3 6.35 6.4 6.45 1 168 360 457 648 840 ₁₀₃₂ ₁₁₂₈ ₁₃₂₀ ₁₅₁₂ ₁₆₀₉ ₁₈₀₀ ₁₉₉₂ ₂₁₈₄ ₂₂₈₀ ₂₄₇₂ ₂₆₆₄ ₂₇₆₁ ₂₉₅₂ ₃₁₄₄ ₃₃₃₆ ₃₄₃₂ ₃₆₂₄ ₃₈₁₆ ₃₉₁₃ ₃₉₃₃ ₄₀₀₈ Hole Diameter (mm) Hole Number # 6.2 6.25 6.3 6.35 6.4 6.45 Hole Diameter (mm)

Hole Number #

(c)

6.2 6.25 6.3 6.35 6.4 6.45 1 192 384 576 672 864 ₁₀₅₆ ₁₁₅₃ ₁₃₄₄ ₁₅₃₆ ₁₇₂₈ ₁₈₂₄ ₂₀₁₆ ₂₂₀₈ ₂₃₀₅ ₂₄₉₆ ₂₆₈₈ ₂₈₈₀ ₂₉₇₆ ₃₁₆₈ ₃₃₆₀ ₃₄₅₇ ₃₆₄₈ ₃₈₄₀ ₄₀₃₂ Hole D iameter (mm) Hole Number#

Fig.17.Variationofdiameterwithrespecttoholenumber:(a)N=15,000rpm, f=225␮m/rev, (b) N=5000rpm, f=225␮m/rev and (c) N=15,000rpm, f=300␮m/rev(DCC-IIdrillwith6.38mmdiameter).

6.38mmnominaldiametersareusedinexperiments.Hole diame-tervariationmeasurementsrevealthathighfeedandspeeddrilling conditions(N=15,000rpm,f=225␮m/rev)reachthelower toler-ancelevelof6.35mmaround4000holesasshowninFig.17(a).No delaminationproblemwasobservedattheholeexits.Whenspeed isreducedto5000rpm,holediametertoleranceisalsoreached around1700holes(Fig.17(b)).Theinﬂuenceoffasterdrillwearat lowfeedisobservedonholediameters.Whenfeedisincreasedto 300␮m/revat15,000rpm,alargescatteronmeasureddiameters wasobserved.Tolerancelevelwasreachedwithin1700holesas showninFig.17(c).

Basedontheﬁndings givenabove,increasingfeedand rota-tionalspeedhelpsprotectdiamondcoatingonthedrill,whichin turnresultsinbetterdrillperformanceintermsofholequality. Theresultspresentedinthisstudydonotagreewithmajorityof theresultsreportedin theliterature,where low feedandhigh speeddrillingarerecommendedintermsofdelaminationbased holequalitycriteria.ItmustbenotedthatthepropertiesofCFRP material,drillinggeometry,and rigidityof themachinetool all

playanimportantroleintheseobservations.Therigidityofthe machinetoolisknowntoinﬂuencedrillingstability,drilllife,and holequality.Increasingthrustforcewithincreasingfeedmayresult invibrationsduringdrilling,whichadverselyaffectprocess out-puts.WhileFig.2maygiveanideaabouttherigidityofthemachine toolusedinthisstudy,animpacthammertestwasperformedto measureitsdynamicproperties.Theresultswereobtainedthrough CutProTM_software_and_the_spindle-tool_holder-drill_system_is mea-suredtohaveanaturalfrequencyof3500Hz,stiffnessof1e9N/m, andadampingratioof1.846%(equalinxandydirections).

Intheliterature, thepositiveimpactof usingbackingplates whiledrillingCFRPlaminateshasbeenreportedespeciallywhen drillingunidirectionaltypeCFRPlaminates.Inthisstudy,testsare repeatedwithoutusingabackingplate,andnosigniﬁcant differ-enceswereobservedintermsofmeasureddrillingforcesorhole quality.Fabricwovenlaminatesaremoreresistanttodelamination thanunidirectionalCFRPlaminates.Therefore,duringproduction ofunidirectionalCFRPlaminates,fabricwovenCFRPsareusually usedatthetopandbottomofthelaminatestodecreasethe like-lihoodofdelaminationduringdrillingoperations.Arecenttrend inaerospacemanufacturingrequiresCFRPlaminatestobedrilled togetherwithtitanium/aluminumalloyplatesasstacksinorderto makeassemblyoperationseasier.Drillgeometryselectionand/or designbecomesmorechallengingduetodifferentmaterial prop-erties.

5. Conclusions

ThisstudyexaminesdrillingoffabricwovenCFRPlaminates andobservestheperformanceofdrillgeometryasafunctionof drillingoperationalparameters.Theexperimentalstudywas con-ductedunderindustrialconditionsusingspecializedequipmentfor CFRPmachining.

• The interplay among the exit thrust force, thetorque carry-ingcapabilityofthesecondaryedge,andtheloaddistribution betweentheprimaryandsecondaryedgesispresentedasa func-tionofdrillanglesandedgelengths.

• DelaminationinfabricwovenCFRPlaminatesisobservedtobe closelyassociated withtheconditionof thediamondcoating. Peakthrustforcesareobservedtoincreasesigniﬁcantlyasaresult offracturingofthediamondcoating.

• DCC-IIdrillgeometryisfoundtobemoresuitabletohighfeed drillingthanDCC-Igeometry.

• Hole diameter toleranceis observed to bemore critical than delaminationduringthedrillingexperimentsconductedinthis study.

Acknowledgements

TheauthorswouldliketothankTheScientiﬁc andTechnical ResearchCouncilofTurkey(TUBITAK-Teydeb)andODAGEMA.S. fortheirﬁnancialsupportofthisstudy.

References

Abrao,A.M.,Faria,P.E.,CamposRubio,J.C.,Reis,P.,Davim,P.,2007.Drillingofﬁber reinforcedplastics.Areview.JournalofMaterialsProcessingTechnology186, 1–7.

Caprino,G.,Santo,L.,Nele,L.,1998.Interpretationofsizeeffectinorthogonal machiningofcompositematerials.PartI:unidirectionalglassﬁbrereinforced plastics.CompositesPartA29A,887–892.

Chen,W.C.,1997.Someexperimentalinvestigationsinthedrillingofcarbonﬁber reinforcedplastic(CFRP)compositelaminates.InternationalMachineTools& Manufacture37,1097–1108.

Dharan,C.H.K.,Won,M.S.,2000.Machiningparametersforanintelligentmachining systemforcompositelaminates.InternationalMachineTools&Manufacture40, 415–426.

(11)

Faraz,A.,Biermann,D.,Weinert,K.,2009.Cuttingedgerounding:aninnovativedrill wearcriterionindrillingCFRPcompositelaminates.InternationalMachineTools &Manufacture49,1185–1196.

Hocheng,H.,Dharan,C.H.K.,1990.Delaminationduringdrillingincomposite lami-nates.JournalofEngineeringforIndustry112,236–239.

Hocheng, H., Tsao, C.C., 2005. The path towards delamination-free drilling of composite materials. Journal of Materials Processing Technology 167, 251–264.

Iliescu,D.,Gehin,D.,Gutierrez,M.E.,Girot,F.,2010.Modelinganddrillwearin drillingofCFRP.InternationalMachineTools&Manufacture50,204–213. International Air Transport Association (IATA) Fact Sheet:Fuel, 2012.

http://www.iata.org/pressroom/facts ﬁgures/fact sheets/pages/fuel.aspx. Karpat,Y.,Camuscu,N.,Kılıc¸,A.,Sonat,F.,De˘ger,B.,Bahtiyar,O.,2010.Drillingcarbon

ﬁberreinforcedplasticswithdiamondcoatedcarbidecuttingdrills.In: Pro-ceedingsofthe36thInternationalMatadorConference,Manchester,UK,pp. 205–208.

Koplev,A.,Lystrup,A.,Vorm,T.,1983.Thecuttingprocess,chipsandcuttingforces inmachiningCFRP.Composites14,371–376.

Lazar,M.B.,Xirouchakis,P.,2011. Experimentalanalysisofdrillingﬁber rein-forced composites. International Machine Tools & Manufacture 51 (12), 937–946.

Piquet,R.,Ferret,B.,Lachaud,F.,Swider,P.,2000.Experimentalanalysisofdrilling damageinthincarbon/epoxyplateusingspecialdrills.CompositesPartA: AppliedScienceandManufacturing31,1107–1115.

Rawat,S.,Attia,H.,2009.Characterizationofthedryhighspeeddrillingprocessof wovencompositesusingmachinabilitymapsapproach.CIRPAnnals– Manufac-turingTechnology58,105–108.

Schulze,V.,Becke,C.,Weidenmann,K.,Dietrich,S.,2010.Machiningstrategiesfor holemakingincompositeswithminimalworkpiecedamagebydirectingthe processforcesinwards.JournalofMaterialsProcessingTechnology211(3), 329–338.

Shyha,I.S.,Aspinwall,D.K.,Soo,S.L.,Bradley,S.,2009.Drillgeometryandoperating effectswhencuttingsmalldiameterholesinCFRP.InternationalMachineTools &Manufacture49,1008–1014.

Teti,R.,2002.Machiningofcompositematerials.CIRPAnnals– Manufacturing Tech-nology51,611–634.

Tsao,C.C.,Hocheng,H.,2005.Effectsofexitback-upondelaminationindrilling compositematerialsusingasawdrillandacoredrill.InternationalJournalof MachineDrills&Manufacture45,1261–1270.

Zhang,L.B.,Wang,L.J.,Liu,X.Y.,2001.Amechanicalmodelforpredictingcritical thrustforcesindrillingcompositelaminates.ProceedingsoftheInstitutionof MechanicalEngineers215,398–405.