air accident investigation unit ireland

Air AccidentInvestigation Unit

Ireland

SeriouS incident report Airbus A330-301, ei-ord

over iceland11 May 2010

Airbus A330-301 EI-ORD Over Iceland 11 May 2010

2

Air Accident Investigation Unit Report 2012-015

AAIU Report No: 2012-015

State File No: IRL00910032

Report Format: Synoptic Report

Published: 20/09/2012

In accordance with the provisions of SI 460 of 2009, the Chief Inspector of Air Accidents, on 12 May 2010, appointed Mr. Graham Liddy as the Investigator-in-Charge to carry out a Field Investigation into this Serious Incident and prepare a Report. Due to his retirement, on the 29 February 2012 the Chief Inspector appointed Mr. Paul Farrell as the Investigator-in-Charge to complete the Investigation. The sole purpose of this Investigation is the prevention of aviation Accidents and Incidents. It is not the purpose of the Investigation to apportion blame or liability.

Aircraft Type and Registration: Airbus A330-301, EI-ORD

No. and Type of Engines: 2 x General Electric CF6-80E1A2

Aircraft Serial Number: MSN 059

Year of Manufacture: 1994

Date and Time (UTC): 11 May 2010 @ 15.08 hrs (UTC1)

Location: Over Iceland; N64º 45.011’, W019º 20.001’

Type of Operation: Scheduled Public Transport

Persons on Board: Crew - 11 Passengers - 227

Injuries: Crew - 0 Passengers - 0

Nature of Damage: Right Hand (RH) inboard aileron, outside

servo-control, outer bracket failure2

Commander’s Licence: JAA ATPL

Commander’s Details: Male, aged 52 years

Commander’s Flying Experience: 19,200 hours, of which 4,931 were on type

Notification Source: Aircraft Operator

Information Source: AAIU Field Investigation

1 UTC:UniversalTimeCo-ordinated,whichisthesameasGreenwichMeanTime.AlltimingsquotedinthisreportareinUTC.

2 Eachwinghastwoailerons,designated“inboard”and“outboard”inthisReport.Eachaileronhastwoservo-controls,designated“inside”and“outside”

inthisReport.Eachservo-controlhastwoattachingbrackets,designated“inner”and“outer”inthisReport.Theseconventionsarenotadheredtowhen

quotingfromofficialpublications.

3

SYNOPSIS

Theaircraftwascruisingat33,000ft,whenitencounteredturbulence.Thisturbulenceresultedinamoderateun-commandedroll.

Whiletheautopilotwascorrectingthisroll,theouterbracketsupportingtheoutsideservo-controloftheRHinboardaileronfailed,causingtheRHinboardailerontooscillatequicklyfrom14degreesupwardsdeflectionto12degreesdownwardsdeflectionandsettlingatanupwardsdeflectionoftheorderof15degrees.Thisdidnotcauseanycontrolproblemsandtheautopilotwasabletocontinuetocontroltheaircraft.Theaircraftlandedwithoutdifficultyattheintendeddestination,ChicagoO’Hare(KORD).SubsequentlyacrackwasdetectedinacorrespondingbracketontheLeftHand(LH)wing.

1. FACTUAL INFORMATION

1.1 History of the Flight

TheaircraftwasonascheduledpublictransportflightfromDublin(EIDW)toKORD.Aroundthedateofthisflight,thevolcanicashsituationarisingfromtheeruptionatEyjafjallajökullinIcelandwasposingamajorproblemtoaircraftoperationsinEuropeandintheNorthAtlanticarea.Atthetimeofflight,alargecloudofvolcanicashwasstreamingdownfromIcelandinasouthwestdirection.Forthisreasontheflightwasroutedalongamorenortherlytrackthannormal,routinginitiallyfromDublintoIcelandandtheninawesterlydirectiontoKORD.DuetothefunnellingofaircrafttracksintheIcelandicarea,arisingfromthepresenceoftheashcloudtothesouthwest,therewasconsiderablecongestionintheairspaceoverIceland.AsaresultEI-ORDwasgivenacruisingaltitudeofFL330(FlightLevel330or33,000ft).TheFlightCrew,beingawareofthissituationpriortodeparturefromEIDWandtheincreasedfuelconsumptionresultingfromthislowerthannormalcruisingaltitude,andalsobecauseofthelongerrouterequiredtoavoidtheashcloud,hadtakenonextrafuel.WhileoverIceland(Figure No. 1)theaircraftencounteredsomemoderateturbulencewhichtheFlightCrewsuspectedmighthavebeencausedbywaketurbulence.Theturbulenceresultedinsomeaircraftrollandyaw.Theautopilot,whichremainedengagedduringtheturbulenceencounter,quicklyreturnedtheaircrafttolevelflight.ShortlyafterwardstheaircraftwasclearedtoclimbtoFL380.TheFlightCrewinputtedthisnewaltitudeintothealtitudeselectorandtheaircraftcommencedaclimbtothenewcruisingaltitude,undercontroloftheautopilot.Duringthisclimb,theFlightCrewobservedalowerthanexpectedrateofclimbandahigherthanexpectedfuelconsumptionrate.

4



Figure No. 1: Overview of aircraft location at time of event

TheFlightCrewstartedtotrouble-shootthecauseofthislossofperformance.Duringthistrouble-shootingtheyobservedthattheElectronicCentralisedAircraftMonitoring(ECAM)FlightControlpageshowedthattheRHinboardaileronwas2/3offull-scaleaboveitsnormalposition.TheRHoutboardaileronwasdeflecteddownwardsby1/3offull-scalefromitsnormalposition.BothLHaileronsweredeflectedupwardsby1/3offull-scalefromtheirnormalpositions.TherewerenoECAMwarningsorcautions.TheFlightCrewforthisparticularflightconsistedofthreepilotssotheCaptainwasabletoleavetheflightdeckandgotothecabinwhereheobtainedvisualconfirmationthatthephysicalconfigurationoftheaileronscorrespondedwiththeindicationsontheECAM.TheFlightCrewnotedthattheFlightCrewOperationManual(FCOM)gavenospecificinformationonthecauseoftheobservedproblem,butitstatedthatintheeventofanun-commandedflightcontrolsurfacedeflection,theflightcouldbecontinuedasnormalifnohandlingproblemswerenoted.TheFCOMincludedcrewawarenessguidancewhichpredictedincreasedfuelconsumptionintherange6-16%.TheFlightCrewcontactedtheOperator’smaintenancecontrolcentreinDublinviatheAircraftCommunicationsAddressingandReportingSystem (ACARS).ThepossibilityofresettingPRIM(FlightControlPrimaryComputer)2wasdiscussedbutMaintenanceControladvisedagainstthis.UsingtheHeading(HDG)selectfacility,andwithoutdisconnectingtheautopilot,theFlightCrewsatisfiedthemselvesthattheycouldsuccessfullymanoeuvretheaircraft.Thecockpitandcabinweresecured(passengersandCabinCrewseatedandstrappedin)andtheFlightCrewthenturnedofftheautopilotandverifiedthattherewerenodifficultiesinmanuallycontrollingtheaircraft.Theythenre-engagedtheautopilot,whichcontinuedcontrollingtheaircraftwithoutdifficulty.

5

AtthispointtheaircraftwastothewestofIceland.TheFlightCrewconsideredtheiroptions.Basedonthefactsthat:

• theaircraftwashandlingnormally, • theywereapproximatelyhalfwaytotheirdestination, • theyhadmorethanadequatefueltoreachtheirplanneddestinationevenallowingforthe

increasedfuelconsumption,(becauseoftheextrafuelupliftedatEIDW), • theplanneddestinationofferedalongrunway(13,000ft),whichcouldproveusefuliftheywere

toexperiencehandlingdifficultiesduringtheapproach, • areturntoEIDWatthispointwouldrequiretheaircrafttomanoeuvrearoundthevolcanicash

cloudwhichwasnowbehindthem,

theFlightCrewdecidedtoproceedtoKORD,andtheremainderoftheflightpassedwithoutfurtherincident.

OnestablishingcontactwithKORD,theFlightCrewadvisedATCthattheyhadaflightcontrolsystemsproblem,thenatureofwhichtheycouldnotdetermine.TheyrequestedanapproachtothelongestrunwayatKORD(RWY28/10)incasetheyexperiencedhandlingproblemsastheaircraft’sspeedwasreducedandtheaircraftwasreconfiguredforapproachandlanding(flapsandundercarriagedeployed).ATCqueriediftheyweredeclaringanemergencyandtheFlightCrewsaidno.ATCthendecidedthattheywouldhavetodeclareanemergency,asthiswouldensurethatRWY28/10wouldbeavailabletoEI-ORD.

TheaircraftapproachedandlandedatKORDwithoutdifficultyorincident.

1.2 Injuries to Persons

Therewerenoinjuriessustainedinthisevent.

Injuries Flight Crew Passengers Other

Fatal 0 0 0

Serious 0 0 0

Minor/None 11 227 0

6



1.3 Damage to the Aircraft

1.3.1 System Description

OntheA330aircrafteachwingisfittedwithtwoailerons,designatedasinboardandoutboard.Twoelectro-hydraulicservo-controls(actuator)areattachedtoeachaileron(inboardandoutboard).Innormaloperationsonlytheoutsideservo-controlactuatestheaileroninresponsetoinputsfromthecockpitsidesticksortheautopilot.Theinsideservo-controloperatesinadampingmodeandpreventsflutterintheeventofmultipleelectricalorhydraulicsystemfailures.

However,intheeventoffailureoftheoutsideservo-control,theinsideservo-controlcantakeovertheactive/commandfunction.Eachwingisalsofittedwithsixspoilerpanels,numbered1to6,theinnermostpanelbeingNo1.Rollcontrolinflightisachievedbyacombinationofmovementsofinboardandoutboardaileronsanddeploymentofspoilers2throughto6.Athigheraircraftspeeds(inthecruise)onlytheinboardaileronsareoperatedandtheoutboardaileronsareservoedtozero.Oneendoftheservo-controlisattachedtotheaileronitactuates;theotherisattachedtothewingrearsparthroughapairofbrackets(Figure No. 2).

Figure No. 2: Engineering schematic of the assembly (LH Wing)

7

1.3.2 Damage Found on Arrival

OnarrivalatKORDtheaircraftwasinspectedbymaintenancepersonnel.Thebrackets(PartNumberF575-50986)attachingtheoutside(active)servo-control,oftheRHinboardaileron,totheRHwingrearspar,werefoundtobefractured(Photo No. 1),andthelowerwingskindirectlybeneaththebrackets’attachmentpointswasdeflected.Furtherinspectionrevealedthatthecorrespondingouterbracketontheother(LH)sideoftheaircraftwascrackedinthesameareawheretheRHouterbrackethadfailed.

Photo No. 1: The fractured brackets

1.4 Other Damage

Therewasnootherdamage.

8



1.5 Personnel Information

1.5.1 Commander Personal Details: Male,aged52years Licence: JAAATPL,IssuedbytheIAA Last Periodic Check: 4Feb2010 Medical Certificate: Class1

Flying Experience: Total all types: 19,200 hours Total all types P1: 16,909 hours Total on type: 4,931 hours Total on type P1: 4,900 hours Last 90 days: 99 hours Last 28 days: 42 hours Last 24 hours: 9.92 hours

1.5.2 First Officer Personal Details: Male,aged35years Licence: JAAATPL,issuedbyIAA Last Periodic Check: 08Oct2009 Medical Certificate: Class1

Flying Experience: Total all types: 6,386 hours Total all types P1: 3,072 hours Total on type: 2,108 hours Last 90 days: 108 hours Last 28 days: 30 hours Last 24 hours: 9.9 hours

1.5.3 Third Pilot Personal Details: Male,aged49years Licence: JAAATPL,IssuedbytheIAA Last Periodic Check: 13Jan2010 Medical Certificate: Class1

Flying Experience: Total all types: 17,700 hours Total all types P1: 10,000 hours Total on type: 2,100 hours Total on type P1: 2,000 hours Last 90 days: 89 hours Last 28 days: 23.15 hours Last 24 hours: 9.53 hours

1.6 Meteorological Information

Theaircraftencounteredmoderateturbulenceatthestartoftheevent.

1.7 Navigation

Aidstonavigationwerenotafactorinthisevent.

9

1.8 Communications

Communicationswerenotafactorinthisevent.

1.9 Aerodrome Information KORDisamajorinternationalairport.Ithasatotalofsevenrunways,whichvaryinlengthfrom

7,500ftto13,000ft.RWY10/28is13,000ftlong,andexceedsthelengthofallotherKORDrunwaysbyatleast3,000ft.

1.10 Flight Recorders

1.10.1 Cockpit Voice Recorder (CVR)

TheaircraftwasequippedwithasolidstateCVR,whichhadarecordingdurationoftwohours.

Theeventoccurredmorethantwohoursbeforetheaircraftlanded.Consequently,therecordingatthetimeoftheeventwasnotavailableandsotheCVRwasnotofanyassistancetotheInvestigation.

1.10.2 Flight Data Recorder (FDR)

DownloadingoftheFDRprovidedthefollowinginformation:

Theaircraftwasinthecruiseat33,000ftwithAP1(Autopilot#1)andFD(FlightDirector)on;ATHR(Auto-thrust)wasengaged.

At15.08:03hrstheaircraftencounteredturbulenceandverticalloadpeaksof0.73and1.45Gwererecorded.Theaircraftimmediatelystartedtobanktotheleftatthistime.ThisrollwascounteractedbyaileronsandRHspoilers.Amaximumbankangleof11degreesleftbankwasreachedsevensecondslaterat15.08:10hrs.

Theaircraftthenstartedtorolltotheright,startingat15.08:12hrs.ThisrollwascounteractedbyaileronsandLHspoilers.TheRHinboardailerondeflectedto14degreesup.threesecondslateritwasat12degreesdownandonesecondlateritwasat13degreesup.Overthesubsequent6secondsitmovedto15degreesup.Inresponsetothisaileronmovementtheaircraftcontinuedtorolltotheright.Amaximumrightbankangleof20degreeswasrecordedat15.08:15hrs.

Thebankanglethendampedtotheleftreaching6degreesleftandrightbankandstabilisedinlevelflight(zerodegrees)at15.08:50hrs.

Inthestabilisedconditionwiththewingslevel,theinboardaileronsstabilisedatRHAileron15degreesupandLHaileron7degreesupwhiletheoutboardRHaileronstabilisedat7degreesdownandtheLHoutboardaileronat6degreesup.

At16.32:44hrstheFlightCrewdisconnectedtheautopilot.Theoutboardaileronsthenservoedtozerodegrees(thisisinaccordancewiththedesignspecificationwhentheaircraftisinthecleanconfigurationabove190kts).Theaircraftinitiallyrolledtotherightduetotheasymmetricpositionoftheinboardailerons.Thiswasmanuallycorrectedwithaleftsidestickinputof1/3deflection.RollcontrolcontinuedtobeavailablebymovementsoftheLHinboardaileronandthespoilers.

10



1.11 Damage Information

AfterlandingatKORDthebracketpairsupportingtheoutsideservo-controlontheRHinneraileronwasfoundtohavefailedandwasremoved.Thepaircomprisedtheoutsideservo-controlouterbracket(F575-50989)andtheoutsideservo-controlinnerbracket(F575-50987).ThebracketassemblywasdesignatedbypartnumberF575-50986.ThecorrespondingbracketassemblyontheLHwing,whichNonDestructiveTesting(NDT)foundtobecracked,wasalsoremoved.ThefailedbracketpairwasinitiallyshippedtotheOperator’sbaseatDublinandthentakenbytheAAIUtotheaircraftManufacturer’sfacilityintheUKforexamination.

TheManufacturer’sexaminationcomprised:visualexamination;NDTexamination;Fractographicanalysis;Hardnesstesting;ElectricalConductivitytesting;Mechanicalpropertiestestingofthefailedbracketsandsurfacefinishexaminationofacurrentproductionstandardbracket.

TheManufacturer’sexaminationrevealedthattheprimaryfailurewasoftheouterbracket.Thefailurewascausedbyfatiguecrackingwhichoriginatedfromapitinthebracketsurface.Thepit,whichwasapproximately50micronsindepth,wasbelievedtohavebeenpresentinthebracketsinceitwasmanufactured.Followingfailureoftheouterbracketduetofatiguecrackingtheinnerbracketfailedduetooverload.

Detailedexaminationofthewinglowerskinintheareadirectlybelowthefailedbracketrevealedanareaofdeformationbetweenribs28and29.Thedeformationstartedattheaftsparfastenerlineandcontinuedaftreachingamaximumvalueof6.9mm.(Photo No. 2)

Photo No. 2: Deformation of lower wing skin, RH wing

Therewasnodeformationofthebottomskindirectlyundertherearspar,norwasthereanydeformationoftherearsparflangefasteners,norelongationofassociatedholes.Thelowerfastenerholesintherearsparwebforattachingthebracketstotherearsparwebwerefoundtobeelongated;theholeshadaninitialdiameterof7.94mmbutwerefoundtobeelongatedto8.6mmapproximately.

Therewasnofuelleak.

1.12 Medical and Pathological Information

Therewerenomedicalorpathologicalissuesinthisevent.

1.13 Fire

Therewasnofire.

11

1.14 Survival Aspects

Therewerenosurvivalaspects/issuesinthisevent.

1.15 Test and Research

TheManufacturercarriedoutresearchasdetailedbelow.

1.16 Organisational and Management Information

Notapplicable.

1.17 Additional Information

AtthetimeofthiseventtheaircraftManufacturerinformedtheInvestigationthattherehadbeenatotalofsixeventsrelatingtofailedorcrackedbracketsonA330andA340aircraft;theA340hasasimilarwingwithidenticalbrackets.FouroftheseeventshadoccurredpriortotheEI-ORDeventandanotherhadoccurredonthesameday.Detailsofthesebracketcrack/failureeventsareshownin Appendix A.

Replacementoftheservo-controlbracketsisnotastraightforwardtask,inspiteofthefactthatthebracketswereboltedontothewingrearspar.Thisisbecausethebracketsaresuppliedwithonlypilotholesfortheattachmentboltsandtheseneedtobecustomdrilledtothefinishedsizeandlocationforthewinginquestion.

1.17.1 Manufacturer’s Service Bulletin (SB) A330-57-3109

TheManufacturerpublishedSBA330-57-3109on13February2009.ThisSBwastitled“Wings–InnerAileronRearSparOuterattachments–InspectionofOuterBrackets”.ThereasonforissuingtheSBwasthat“anoperatorreportedcracksonbothbracketsattachingtheLHActiveServocontrol8CS1(inneraileron)attherearspar.AsubsequentinspectionoftheRHwingrevealedsimilarcracksonbothouterbrackets.Theaircraftinquestionhadaccumulated14,261flightcycles”.TheSBwasrestricted(byManufacturer’sserialnumber)toaircraftwhichhadcompletedmorethan10,000flightcycles(FC)andwasclassifiedas“Recommended”whichmeansthatitshouldbecarriedoutattheearliestopportunitysubjecttotheavailabilityoffacilitiesandqualifiedpersonnel.

TheOperator’sMaintenanceContractedServiceProvider(CSP)receivedtheSBinFebruary2009.TheCSPdeterminedthattheSBwasapplicabletothesubjectaircraftandanotheroftheoperator’saircraft.TheSBcalledforaHighFrequencyEddyCurrent(HFEC)inspectionofinnerandouterservo-controlattachmentbrackets,designated8CS1(lefthandwing)and8CS2(righthandwing).TheSBincludedapictureofthepairofbracketstobeinspected;however,thepictureincorrectlyshowedthebracketsfortheinactiveservocontrolratherthantheintendedactiveservocontrolbrackets.TheCSPraisedthenecessaryEngineeringOrderandrelatedinstructionsfortheaccomplishmentoftheSB.TheSBwasaccomplishedduringaC-CheckataMaintenance,RepairandOverhaul(MRO)facility,on22December2009.NoadversefindingswererecordedfollowingaccomplishmentoftheSB.

On9July2009,SBA330-57-3109wasrevised(Revision01,09July2009)followingthediscoveryofcrackingonthebracketsofanaircraftwhichhadaccumulated8,930flightcyclesi.e.lessthantheoriginalinspectionthresholdsetbytheManufacturer.Thenewthresholdwassetat7,500flightcycles.

12



1.17.2 HFEC Inspection

TheInvestigationinterviewedtheInspectorwhocarriedoutSBA330-57-3109attheMROfacility.TheInspectorheldcertificationissuedbytheBritishInstituteofNon-DestructiveTesting,whichwasvaliduntil6November2011.TheInvestigationwasalsoshownhisCompanyProceduresandHumanFactorstrainingcertification,whichwerealsovalidandindate.

TheInspectorinformedtheInvestigationthatonthedayoftheSBaccomplishment,itwashisonlyworktaskandthathehadnotbeensubjecttoanydistractions.UsinganotheraircraftattheMROfacility,theInspectordemonstratedhowhehadcarriedouttheInspection.Thedemonstrationshowedthatusingtheprescribedmethodofinspectionitwasdifficultforaninspectortopreventtheprobefromslippingofftheedgeofthebracketbeingsurveyed.Suchslippingcausedfalsereadings.DuringthedemonstrationtheInspectorshowedgoodunderstandingofthetechnique,andrepositionedtheprobeeachtimeitslippedtoensurethatthefinalresultoftheHFECinspectionwasvalid.

TheInvestigationreviewedthecalibrationhistoryforthetestequipment.Theunithadbeencalibratedon28January2009andwasnextduecalibrationon28January2010.Atthecalibrationon28January2010theunitwasfoundtobewithinthecalibrationtolerance.Inaddition,priortoeachuseoftheHFECequipmentAirbusNon-DestructiveTestingManualrequireslocalcalibrationusingsamplecalibrationblocks.Thesecalibrationblocksarelocallymanufacturedblockswithmachinedslotsrunningtoavarietyofdepths.TheInvestigationfoundthattherewasappropriatecertificationfortheseblocks.

DuringhisinterviewtheHFECInspectorrecalledthattheFINcodemarkingdecalwasmissingfromtheRHinboardaileronactiveservo-control.SubsequentinspectionofEI-ORDbytheOperatorconfirmedtheInspector’srecollection.TheOperatorarrangedforreplacementdecalstobefitted.

1.18 Meetings with the Aircraft Manufacturer

1.18.1 June 2010

On16June2010theInvestigationmetwiththeAircraftManufacturertodiscussthematter.TheManufacturerinformedtheInvestigationthatatthattimetherehadbeenatotalofseven(7)aileronactiveservobracketcrackingeventsrecorded.(Appendix A)Ofthese,tworesultedinfractureandseparation;thisincludedthesubjectevent.FouroftheeventsweredetectedasaresultoftheapplicationofSBA330-57-3109Version00.

Atthetimeofthemeetinglaboratorytestsonsixaffectedbracketshadrevealedsurfacepittingtodepthsof50microns;thispittingwasthesiteofthecrackinitiation.Fatiguestriationswereinevidenceinmostofthecracks;crackswithalengthuptoabout4mmexhibitedatypicalstriationspacingofapproximately2microns.

Therewaslittleevidenceofcorrosionwithinthepitsandthepresenceofprimerinthepitssuggeststhatthepittingwastheresincemanufacture.

TheManufacturerhadidentifiedadegreasingprocessknownas“pickling”usedinthemanufacturerofthebracketsasthelikelycauseofthesurfacepitting.PriortoMay2003thepicklingprocesswasamanualprocess,thereafteritwasautomated.TheManufacturerstatedthatabracketrandomlyselectedfromtheproductionlinein2008wasfoundtogenerallyhaveasurfacefinishbetterthantheproductiondrawingrequirementof2.4microns,thoughasmallnumberofisolatedpitsofupto20micronsdepthwereidentifiedandbelievedtobeduetomachining.During

13

manufacturingthebracketsarefullymachinedbeforethesurfacefinishischecked.Thebracketsarethendegreasedandcleanedpriortoanodisingandpost-anodisingthebracketsarepaintedandre-workedasnecessary.Withnosurfacefinishcheckafterthedegreasing(“pickling”)thevariabilityassociatedwiththemanualnatureoftheprocessmeantthatpitsofvaryingdepthcouldbepresent(undetected)inthefinishedbracket.

TheInvestigationraisedanumberofissueswiththeAircraftManufacturer,namely,

• ThelackofclarityinthetitleofSBA330-57-3109.“WINGS–INNERAILERONSREARSPAROUTERATTACHMENTS–INSPECTIONOFOUTERBRACKETS”couldmistakenlybeconstruedtomeanthatonlytheouterbracketsontheoutsideservo-controloftheinboardaileronrequiredinspection.

• ThediagramintheSBshowedthebracketpairfortheinsideservo-controlinsteadofthebracketpairfortheoutsideservo-control.(Figure No. 3)

• WhywastheSBone-offratherthanrepetitiveinnature? • WhydidtheSBnotprescribeacomplianceperiod? • Whyweresomeaircraftexcludedfromtheapplicabilitytable? • Wasthereariskofafuelleakduetotheassociateddeformationofthebottomskinofthe

wing? • NeedtoreviewtheaircrafthandlingimplicationsofEI-ORD’sbracketfailure. • NeedtoconductacrackagestudyonEI-ORD’sfailedbracketstoassessifitwaslikelythat

therewerecrackspresentwhenthebracketswereHFECinspectedinDecember2009.

Figure No. 3: Original SB Diagram showing incorrect bracket pair

14



1.18.1 October 2010

On27October2010theInvestigationmetwiththeAircraftManufactureragaintoreviewprogressonthemattersincethepreviousmeeting.

InrelationtotheSB,theAircraftManufactureradvisedthatitwasre-issued(Version02,08September2010)withanewtitle“WINGS-INNERAILERONSREARSPARATTACHMENTS-INSPECTIONOFOUTBOARDACTUATORBRACKETS”.ServiceBulletinA330-57-3109Version02increasedtherecommendednumberofoperatorsrequiredtoaccomplishtheinspectionfromonetotwo.ThereasonfortheSBwaschangedfrom“AnoperatorreportedcracksonbothbracketsattachingtheL/HActiveServoControl8CS1(inneraileronsattherearspar…”to“AnoperatorreportedaruptureofboththepairofattachmentbracketsfortheoutboardL/HServoControl*8CS1)oftheinneraileron...”ServiceBulletinA330-57-3109Version02alsopicturesthecorrectbracketpair(Figure No. 4),includesnotesonnon-destructivetesting,graphicsdepictingthetypicalcracklocationanddetailsforalocallymanufacturedplasticguidetoaddresstheproblemswiththeprobeslippingoffthebracketedge.OnthequestionofrepetitiveintervalstheManufacturersubsequentlyinformedtheInvestigationthatthefollowingintervalswerepublishedintheServiceBulletinA330-57-3109Revision03on31May2011:

• FortheA330,ISB57-3109R3setstherepetitiveintervalasThreshold=6000FC/19400FH3,Interval=1800FC/5700FH.

• FortheA340,ISB57-4119R3setstherepetitiveintervalasThreshold=3600FC/20600FH,Interval=1700FC/9600FH

Figure No. 4: Revised SB diagram showing correct bracket pair

3 FH:FlightHours

15

Onthequestionofpossiblefuelleak,theAircraftManufacturerstatedthatinbotheventsinvolvingbracketfracturetherewasnodeformationofthebottomskinundertherearsparnorwasthereanyholeelongationorrecordeddamagetothesparflangefasteners.TheManufacturerconcludedthattherewasthusnoriskoffuelleakfromthesparflangebolting(Primaryleakpath).InthecaseofEI-ORDtherewassomelocaliseddeformationofthetwolowerfastenersthroughtheaileronattachmentbracketandRearSparweblocations.However,theManufacturerconcludedthattherewasalowriskoffuelleakfromthesefastenersandthatinanycasetheriskwasmitigatedbythesealingandovercoatingofthefastenersduringinstallationandthelowmagnitudeoftheholedeformation.

AsaresultofthisanalysisbytheManufactureritwasconcludedthattheriskofasignificantfuelleakduetothefractureofanaileronbracketwasclassifiedasverylow.

TheAircraftManufactureralsoconsideredthepossibleimplicationsforaircrafthandlingofabracketfailure.TheManufacturerconductedanengineeringassessmentofthehandlingqualities/behaviouroftheaircraftwith1or2inboardaileronsatzerohingemomentposition,asaresultofrupturedactiveservocontrolbrackets.TheManufacturerconcludedthatthereisalwaysenoughlateralcontrol(usually,lessthan5degreesspoilerdeflectionwithover35degreesavailable)tocounterthefailure(s)withverysimilardynamics.TheManufactureradvisedthattheriskofflutterhadbeenreviewedandpresentedtoEASAatthe75thAirworthinessReviewMeeting(ARM)on22/23September2010).TheminutesoftheARMmeetingrecord:“Fluttercanonlybecausedbythecompleteseparationofbothactiveanddampingservocontrols.Thiscannotbecausedbythepresentissueasonlytheactiveservocontrolbracketsoftheinneraileronsisimpacted.Thereisnoriskofflutter”.TheManufacturerstatedthattheclassificationof“MINORimpactonAirworthiness”wasagreedbyEASA(atthe75thARM).

Onthequestionofcrackageestimation,theManufactureradvisedthatthegrowthofasmallflaw(defect)tothepointofrupturecouldnotbedeterminedexactly,butisestimatedtobeseveralthousandflightcycles.EI-ORDwasinspectedinaccordancewiththeSBat15,987FC(55,303FH).Thebracketfracturedat16,183FC(56,180FH).Thus196FC(877FH)hadelapsedbetweentheSBaccomplishmentandtheruptureofthebrackets.Examinationrevealedthatthefinalcrackhadgrownto48mmpriortobracketfracture.Striationspacingwasrelativelyuniformforthefirst10mmofgrowthat1micron.AlthoughitwasdifficulttorelateFCdirectlytoloadapplicationsManufacturer’sengineeringdataandanalysisindicatedthatanassumptionofonemajorfatiguestriationperflightcyclewouldberepresentative.Basedonthisassumptionthelaboratoryanalysissuggeststhatthecrackwouldhavebeenoftheorderof10mminlengthatthetimeoftheSBaccomplishment(HFECinspection).Inconclusion,theManufacturer’slaboratorytestingandanalysissuggestedthatitwashighlyprobablethatthecrackswerepresentduringtheoriginalSBinspectioninDecember2009.

1.19 Useful or Effective Investigation Techniques

Notapplicable.

16



2. ANALYSIS

Thiseventwascausedbythefailureoftheinboardaileronoutsideservo-controlouterbracket(F575-50989)ontheRHwingduetofatiguecracking.Theoutsideservo-controlinnerbracket(F575-50987)failedinoverloadasaconsequenceofthefailureoftheouterbracket.Thecorrespondingbracketpaironthelefthandwingwasinspectedandfoundtobecracked.ThetriggeringeventforthisfailurewasanaircraftencounterwithturbulenceoverIceland.

TheFlightCrewdetectedtheeffectsofthefailurealthoughtherewasnospecificindicationoftheeventontheECAM.TheFlightCrewfollowedastructuredapproachtotroubleshootingtheproblemandconsideredallpertinentfactorsbeforemakingthebalanceddecisiontocontinuetotheirdestination.

TheManufacturerhadissuedarecommendedSBrequiringinspectionoftheaffectedbracketsusingHFEC.TheOperator’sMaintenanceManagementSystemcorrectlyscheduledtheSBforaccomplishment.On22December2009,196FCpriortofailure,theSBwascarriedoutbyasuitablyqualifiedinspectorusingproperlycalibratedequipment,freefromanydistractions.TheInspectorwhocarriedouttheinspectionhadclearrecalloftheaircraftandevenrecalledamissingdecalwhichtheOperatorsubsequentlyconfirmed(andreplaced).However,theinspectiondidnotdetectanycracks.Followingtheevent,theManufacturerindicatedfromstudiesofthefeaturesofthefatiguecrackinthefailedbracketthatitwaslikelythatcrackswerepresentwhentheSBwascarriedout.ThereisnoclearexplanationastowhytheinitialapplicationoftheSBdidnotdetectthepresenceofcracks.However,theInvestigationdididentifythattheprocedureprescribedintheSBwaspronetoprobeslippagewhichcausedfalsepositivereadings.InadditiontheoriginalSBcontainedadiagramofthewrongbracketpairandcouldhavecausedconfusion,althoughthereisnoevidencethatsuchconfusionwasafactorinthisparticularapplicationoftheSB.TheManufacturerre-issuedtheSBwith,interalia,adiagramofthecorrectbracketpairanddetailsofalocallymanufacturedguidetoamelioratetheprobeslippageproblem.

Thefailedbracket’sfatiguecrackswerefoundtohaveoriginatedfroma50micronpitwhichwasprobablypresentsincemanufacture.Thepitwaslikelyduetoamanual“pickling”process,whichhasbeenautomatedsince2003.Testsonaproductionstockbracketindicatethattheautomatedprocedurehasaddressedthepittingproblem.

TheeventwasclassifiedbytheManufacturer(andacceptedbyEASA)ashavingaminorimpactonairworthinessandtheriskofanyassociatedfuelleakwasdeemedtobeverylow.

17

3. CONCLUSIONS

(a) Findings

1. TheaircraftencounteredturbulenceoverIcelandandduringautopilotcorrectionstoleveltheaircrafttheouterbracketattachingtheRHinboardaileron’sactive(outside)servotothewingrearsparfractured.

2. TheinnerbracketattachingtheRHinboardaileron’sactive(outside)servotothewingrearsparfailedasaresultoftheouterbracketfracturing.

3. Pittingofupto50micronsdepthwasfoundonthesurfaceofthefailedbracket. 4. Suchpittingprovidedthesitefortheinitiationoffatiguecracking. 5. Theproductiondrawingcalledforthebracketsurfacefinishleavingnopittinggreaterthan

2.4micronsindepth.Qualitycontrolchecksonproductionstockbracketsin2008foundthatgenerallythesurfacefinishwasbetterthantheproductiondrawingrequirementof2.4microns,thoughasmallnumberofisolatedpitsofupto20micronsdepthwereidentifiedandbelievedtobeduetomachining.

6. ThecauseofthepittingwastracedbytheManufacturertoamanualpicklingprocessusedpriortoMay2003,andwhichhadbeenautomatedsincethen.

7. InFebruary2009,theManufacturerpublishedarecommendedSBcallingforHFECinspectionoftheaffectedbracketstodetermineifcrackswerepresent.

8. ThisbulletinwasaccomplishedonEI-ORDinDecember2009andnocracksweredetected. 9. TheInspectorwhocarriedouttheHFECcheckwasproperlyqualifiedandtheequipment

usedwasproperlycalibrated. 10.DuringtheexaminationprescribedintheoriginalSBitwasfoundthetestingprobewas

pronetoslippingofftheedgeofthebracketcausingfalsereadings. 11.TheoriginalSBhadapotentiallymisleadingtitleandcontainedadiagramshowingthe

wrongsetofbrackets. 12.TheManufacturerre-issuedtheSBthreetimesreducingtheinspectionthresholdto

7,500flightcycles(Version01,9July2009),correctingthetitleandbracketdiagram,recommendingtwoInspectorscarryouttheworkanddetailingaplasticguidefortheHFECinspection(Version02,8September2010)andmandatingrepetitiveinspections(Version03,31May2011).

13.TheManufacturer’slaboratorytestingandanalysissuggestedthatitwashighlyprobablethatcrackswerepresentduringtheoriginalSBinspectioninDecember2009.

14. Intotal,acrosstheA330/A340fleet,therewerefivepartialcrackingeventsandtwofullruptureevents,thelastofwhich(partialcrack)wasin2010.

15.Atthe75thAirworthinessReviewMeetingbetweenEASAandtheManufacturertheeventwasclassifiedashaving“MINORimpactonAirworthiness”.

16.Theriskoffuelleakfollowingthefailureofthebracketswasclassifiedasverylow.

(b) Probable Cause

1. FractureoftheouterbracketattachingtheRHinboardaileron’sactive(outside)servo,duetofatigue.

(c) Contributory Cause(s)

1. Manual“pickling”processwhichcausedsurfacepittinggreaterthanthatallowedorprescribedintheManufacturer’sdrawings

2. FailuretodetectcracksduringHFECinspectionprescribedinaManufacturer’sSB.

18



4. SAFETY RECOMMENDATIONS

AstheManufacturerhaschangedthebracketmanufacturingprocesssince2003andhasrevisedtheSBtotakeaccountofissuesidentifiedduringthisInvestigation,thisInvestigationdoesnotsustainanySafetyRecommendations.

APPENDIx A

Year Aircraft Bracket Finding LH/RH FC FH P/N2007 A330-300 Fullcrackonbothlugsof

theInboardAileronouterbracket(8cS1).StaticfailureonIBbracket

LH 14216 45346 F575-50989(OBbracket)F575-50987(IBbracket)

2007 A330-300 Partialcracksonbothbrackets;fatiguegrowthevident

RH 14853 48116 F575-50989(OBbracket)F575-50987(IBbracket)

2009 A340-300 Partialcrack LH 8930 71688 F575-50989(OBbracket)

2009 A340-200 Partialcrack RH 10198 71989 F575-50989(OBbracket)

2010 A330-300 Fullcrack(subjectaircraft) LH/RH 16183 56180 F575-50989(OBbracket)F575-50987(IBbracket)

2010 A330-300 Partialcracks LH/RH 17457 59106 F575-50989(OBbracket)F575-50987(IBbracket)

2010 A330-300 PartialcrackMultiplebrackets

LH&RH 16828 58853

- END -

Produced by the Air Accident Investigation Unit

AAIU Reports are available on the internet www.aaiu.ie

In accordance with Annex 13 to the International Civil Aviation Organisation Convention, Regulation (EU) No 996/2010, and Statutory Instrument No. 460 of 2009, Air Navigation (Notification

and Investigation of Accidents, Serious Incidents and Incidents) Regulation, 2009, the sole purpose of these investigations is to prevent aviation accidents and serious incidents. It is not the purpose of any such accident investigation and the associated

investigation report to apportion blame or liability.

A safety recommendation shall in no case create a presumption of blame or liability for an occurrence.

A.A.I.U., Department of Transport Tourism and Sport, 2nd Floor, Leeson Lane, Dublin 2, Ireland.Tel (24x7): +353 1 604 1293 or +353 1 241 1777 Fax: +353 1 604 1514 Email: [email protected] Web: www.aaiu.ie

air accident investigation unit ireland

Documents