~“

i

ao1-.

%1

.

i...

i

, SECURITY INFORMATION

“‘Y==NACA _—-. ———-————.—

—

RESEARCHMEMORANDUM-

EXPERIMENTAL INVESTIGATION OF AIR-COOLED TURBINE BLADES

IN TURBOJET ENGINE

XIII- ENDURANCE EVALUATION OF SEVERAL PROTECTIVE

COATINGS APPLIED TO TURBINE BLADES OF

NONSTRATEGIC STEELS

By EdwardR. BartooandJohnL. Clure

LewisFlightPropulsionLaboratoryCleveland,Ohio

cMs3rFLm EwJMEtil””

NATIONALADVISORY COMMITTEEFOR AERONAUTICS

WASHINGTONJuly 16,1953

3\9,9a/13

MICARM

8

F53E18

NATIONALADVISORYCOMMITTEEFOR

R3S?IARCEMEMORANDUM

EXPERIMENTALINSTIGATIONOFAIR-COOLED

KERONAUTICS

TURB13!31BI&lESIN

mom

XIII- ENDURANCEEVAI.UA!l?IONOF

APPLIEDTO TURBINEBLADES

By EdwardR. Bartoo

ENGINE

SEVERALPROTECTIVECOATINGS

OFNONSTRATEGICSTEELS

andJohnL.Clure

Durabilitiesgasturbinerotor

SUMMARY

of severalprotectivecoatingsa~pliedto air-cooledbladesofnonstrategicsteels(W 4130andTimken

17-22A(S))wereinvestigatedinmodifiedturbojetengines.Fourtypesof coatings,ceramic,nickel,Nicrobraz,andaluminized(diffusedalu-minum),wereappliedtoa totalof 20blades.Coatingswereendurance-testedforextendedperiodsat themaximunratedspeedandturbineinletitemperatureoftheengineused. Continuousoperationattheseconditionsislimitedtohalf-hourperiodsin service.

.Ceramic,aluminized,Nicrobraz,anda combinationofnickeland

Nicrobrazcoatingseachprovidedsatisfactorycorrosionanderosionprotectiontoatleastonebladefor100hourswithratiosofcooling-airtoconibustion-gasflowbetween0.030and0.048.Aluminizinggaveexcellentprotection,whileoneceramiccoatingprovidedexcellentpro-tectionanddemonstratedtheabilitytopreventcorrosioneventhoughthecoatingwasseverelychipped.Chemicallydepositednickelgaveadequateprotectionin thecoolermidchordregionsof theblade.Nickelovera Nicrobrazundercoatinggaveexcellentprotectionoverthemostdifficlil.tofallregions,thatis, the leating edge. Nicrobraz pro-videdexcellentprotectionovertheentireblade.

INTRODW!TION

TurbinecoolingresearchbeingconductedbytheNACALewislabora-toryincludeswnrkdirectedtowardthedevelopmentof air-cooledturbinerotorbladesofnonstrategicmetalsthatcanbe operatedin turbojetenginesatpresent-dayor slightlyhighergas-temperaturelevels.The*

2 NACARME53E18

coolingeffectiveness,ofeachofa varietyofair-cooledturbinebladeshasbeeninvestigatedM turbojetenginesthatweremodifiedto acco~o-dateeithertwoorfourair-cooledblades(refs.1 to 7). Theseinves-tigationsIndicatedthatbladesofnonstrategicmetalscanbe cooledsufficientlyatratiosofcoolantflowto-combustion-gasflowof.0.02to 0.05tomaketheirusefeasibleatpresent-dayoperatingconditions.Endurancerunningof12bladesofSAE4130orThken 17-22A(S)steels(roughly97and96percentiron,respectively)(ref.8) showedthesebladesto be capableofextendedoperation-atcurrentgastemperatureleve1s. However,corrosionofthebladeShellbec~ evidentwithin5 hoursendlimitedexperimentalbladelifetoroughly50hoursatmax-imumenginespeedandgastemperature,thusemphasizingtheneedfortheinhibitionofcorrosion.

‘—

-=—

i

.-

.

— .—

Preliminaryenduranceinvestigations.ina turbojetengineofnic_kel.. . .__=andceramiccoatingsonbladesofnonstrategicmetals(ref.9) indicatedthatnickelprovidedadequatecorrosionpr.otectionoverthemajorlor~ions =of suchbladesforabout25hoursatratedmaximumturbinespeedandinletgastemperaturebutwouldnotprotecttheleadingedgeformorethan

—-.-.10hours.Thetwoceramiccoatingsinvestigateddidnotprovideprotec- ~ ~~~tionforeven5 hours. .——-—-——

Thisreport.concernsitselfwiththeendurancetestingof fourprom- ● .:isingtypesof corrosion-resistantcoatings.Ceramic,nickel,aluminized~_-”““_andNicrobrazcoatingswereappliedtobladeshellsof eitherW 4130or .. ._Timken17-22A(S)steelanda totalof 20bladeswasrunat themaximum aratedenginespeedof11,500rpm(1300ft/sectipspeed)titha constant._ _ ‘=turbineiril.ettemperatureofapproximately_1670°F. Theratioof coolant.____flowto-combustion-gasflowperblade(hereinaftercalledcoolant-flowratio)wasmaintainedconstant,usuallyat 0.048,although,ina few -——cases,flowratiosof0,038and0.030wereused.

Thegoaloftheendurancetestswasarbitr~ilysetat 100hoursat..—

maximumratedconditionsinviewofthelackofanystandardtestfor.—

coatedbladelife. Innormalflightservice,theenginemaynotbe su.-jectedcontinuouslytomaximumratedspeed(anditsattendantgastem-

—

perature)formorethan30minutesata t@e.

Forconvenience,theresultsofthepreviouspreliminaryinvesti-—

gationofninecoatednonstrategicblades(ref.9)aresummar—

izedin .._thisreport. --

——-

—

*-

M.kxia

NACARM E53E18 3

m

.

Acknowledgementismadeof thecooperationreceivedfromtheCali-forniaMetalEnamelingCompany,TheValorizingCompany,theFerroCor-poration,theResearchLaboratoriesDivisionofGeneralMotorsCorpora-tion,andtheSolarAircraftCompanyin theapplicationof varioustypesof coatingsto theair-cooledturbineblades.

COATINGS

GeneralRequirements

A protectivecoatingforapplicationtoair-coolednonstrategicturbinebladesmustbe abletowithstandthecorrosiveanderosiveactionof thehigh-temperature,high-velocityconibustion-gasstreamtowhichthebladesaresubjectedina turbojetengine.Anyprotectivecoatingappliedto sucha blademustadherewelltothebladetienoperatingathightemperaturesundertheinfluenceofhighcentrifugalforces.Thecoatingmustbe abletowithstandthethermalshockstowhichtheturbinebladeis subjectedduringthestartingandstoppingof theengineandtherapidchangesinbladetemperaturethatmayoccurduringothertransientconditionsthatareincidenttonormalengineoperation.Thecoefficientof thermalexpansionof thecoatingmustbesufficientlyclosetothatof theblademetalovera rangeof tempera-turestopreventspal.lingandflakingof thecoating.Thecoatingmusthavesufficientductilitytowithstandthevibrationandtheelongationof theturbinebladethatoccurduringengineo~eration.Thecoatingshouldbe capableofwithstandingnormalhandling.Itspresenceon thebladesurfaceshouldnotseriouslyaffectthetensileorfatiguestrengthof thebladeshell.Itsapplicationshouldnotadverselyaffectorrestricttheheat-treatmentproceduresthatarerequiredto developthenecessaryphysicalpropertiesof theblademetal.N isalsodesirablethatthecoatingprotectthewallsof thecoolantpassagesaswellastheoutersurfaceof theblade.

Althoughthemetalshellofan air-cooledturbinebladeiscon-siderablycoolerthantheccmibustiongas,thetemperaturelevelisstillrelativelyhigh. Experimentalchordwisetemperatured3.stributionsintheshe3J.sof air-cooledturbinebl.adesofprofilesAandB (seefig.1)forcoolant-flowratiosof 0.05and0.03areshowninfigure2. Althoughthesetemperaturedistributionswereobtainedforspecificair-cooledbladesoperatingina particularturbojetengine,theyareindicativeof thetemperaturelevelsatwhichmetalsandcoatingsforforced-convectionair-cooledturbinebladesmightoperateinpresent-dayturbo-jetengines.Itmaybe seenthatleadingandtrailingedgesoperateattemperaturesashighas1200°F, whilethemidchordregionsare200°to300°F cooler.

—

4 NACAW E53E18 —.-

—

Thebladetemperaturesshowninfigure.%forprofileB aresomewh%”.

higherthanthosereportedinreference9 forthesameprofile.The .- .—valuesoffigure2--arebasedonmorecomprehensiveexperimentaldata.

— .--.—

TypesofCoating6. .—.——

In thisinvestigation,nonstrategicair-cooledturbinebladeswere _-—coatedwithceramics,nickel,orNicrobrazwhileotherswerealuminized.

%A briefsummaryof thereasonsforselectingthesecoatingsforapplicz:– .. mtiontocooledturbinebladesandof thegeneralproceduresinvolvediu_ _.-“N “:applyingthecoatingsisgiveninthefollowingparagraphs.

Ceramiccoatings.- Thesuccessfuluseof ceramiccoatingsto -. _ ..==inhibitcorrosioninnumeroushigh-temperatureal???licationsinorderto ~ --<prolonglifeand/ortoreducethestrategicmetalcontentledto theconsiderationof suchcoatingsforair-coolednonstrategicturbineblades. ““-~

Inadditiontothegeneralrequirements.outlinsdWeviouslY~a .

ceramiccoatingshouldbe as thinaspracticalinordertokeeptheshear~“-.. ,-_stressesat thebondingsurfacetoa minimm.andtoobtaini~roved .-

resistancetothermalshock.Thincoatingsalsoshowlesssusceptibil-..itytochippingthzcmghmishandling(ref.10).

.— 8

Inpreparingceramicsforcoatings,theQroperproportionsof _materials(generallymetaloxidesandfluxingagents)arefusedandquenchedinwater.Theresultingsubstances,alongtithadditionsmad~tocontrolcertainphysicalproperties,aregroundinliquid(usuallywater)andappliedtothemetalsurfacesby sprayingor dipping.Thecoating,whichmaybe as thinas 0.001to0.002inch,isdriedat 200°.to 250°F andthenfired.Firingtemperaturesvarywidely;at thehighertemperaturesmetallurgicaleffectsuponthe_rnetalbeingcoatedmustbe,consideredwhenstress-rupturepropertiesareimportant.Coatingthe>.—.interiorsurfacesofrestrictedregionswithceramicpresentsdifficul-tieswhichmustbe consideredcarefully,particularlyincasessuchasthebladesofthepresentinvestigationwhereblockageof thecooling-airpassagesisintolerable.

Nickelcoatings.- Formanyyearsferrousmetalshavebeensuccess-fullyprotectedfromcorrosionby platingwithvariouscorrosion-resistantmetalssuchasnickel,chromium,silver,andcadmium.Theplatingmaterialuseddependsgreatlyupontheenvironmentandservice...towhichtheplatedpartwillbe exposed.Of themorecorrosion-resistantmetals,nickelisoneofthemostcomnonappliedto steel“andappearstobe wellsuitedforuseon turbineblades.Forapplyingnickelcoatingstotheexperimentalblades,chemicaldepositionwasselected@preferencetoelectroplatingfortworeasons_:first,electroplatiuwillnotcoattheinnerheat-transfersurfacesof theblades,and

—

.F —

—

-.:.-=.-

=—.

—+. —..—. —— —--=+

---*..—

~..—.

-.-—-—.

— -.——— .- ..- —-.— .

-%–” —.— .-—

Nal

~

NAM RME53E18 --”-’ 5

second,settingup equipmentto coatsmallnumbersofbladesis simplerforthechemicalprocess.Thehardnessof chemicallydepositednickelisgreaterthanthatof electrodepositednickel.As applied,thechemicallydepositednickelisbrittlebutuponheatingbecomesductileand,at thesametime,increasesinhardness(ref.il.).

Thechemicaldepositionmethodinvolvesdippingthebladeinanacidsolutioncontainingnickelchlorideornickelsulfateandotherchemicals.TableI sumarizesthecompositionof thetwoacid-nickelbathsusedinplatingbladesforthisinvestigation.Thesolutionisgenerallymaintainedinthetemperaturerangeof 150°to 200°F. Therelativelylowtemperatureof thesolutioninno wayinfluencestheheattreatmentthatmayhavebeengiventhebladepriorto theplatingpro-cess. Thetimerequiredtoforma coating0.001to0.0015inchthickisoftheorderof1 to4 hours.ReferenceU givesa detailedaccountofthemethodsandproceduresinvolvedinapplyingchemicallydepositednickeltosteel.

Nicrobrazcoatings.- Nicrobrazisthetrademe of a commericalbrazingcompoundcomyosedof about72.3percentnickel,15percentchromium,3.75percentboron,4.5percentsilicon,4.0percentiron,‘and0.45percentcarbon.Itproducesa hard,corrosion-resistantcoat-ingwhenfusedandcooled.Itssuccessfuluseinbrazingthebladeshell. to thebaseandtheobservationof itssubsequentbehaviorin serviceledto itstrialas a coating.

Inorderto coata steelsurface,Nicrobrazinpowderformmaybesuspendedina 10percentcalciumchloridesolution,brushedon thesurface,andallowedtodry. Whenheatedina dryhydrogenatmospheretoa temperatureof about2075°F, theNicrobrazwillfuseto thesur-facetoforma hardcontinuouslayer.

Aluminizedcoatings.- Aluminumcoatingshavebeenusedformanyyearstoprotectlowalloysteelsfromcorrosion(ref.12,pp.703-704andref.13,pp.704-705).Thealuminummaybe eitherin theformofa coatingof thepuremetalor as an alloyedaluminum-ironlayervary-ingfroma fewthousandthsofan inchtomorethan0.050inchin thick-ness.Thismethodofprotectionisknownas aluminizing.Usuallyaluminizedpartsarelimitedtooperatingtemperaturesof about1500°F,butshort-timeserviceup to 1750°F hasbeenobtained(ref.12).

Therearetwogeneralmethodsby whichsteelsarealuminized,namely,the“pack”processandthe“dip”process.Boththeseprocesses

. wereusedtoaluminizeturbinebladesforthisinvestigation.Theimpor-tantfeaturesof eachmethodaresummarizedinthefollowingparagraphs.

.Packprocess:Inthepackprocesstheparttobe aluminizedis

packedina boxcontainingpowderedaluminumanda smallamountofL

6 NACARME53E18

ammoniumchloride.Theboxisthensealedgas-tightandheatedtoa tem-peratureof 1500°to1800°F forabout6 to24.hours.Thisprocessimpregnates.thesurfacelayerofthemetalwithaluminumandalsoimpartaa heattreatmenttotheparentmetal.Thetemperature-timerelationforthealuminizingprocessdependsuponthesizeof thepartandtheamountofaluminumpenetrationdesired.Thedepthof.aluminumpenetrationcanbe variedas desired;valuesusuallyrangefrom0.005to0.040inch.Thesurfacelayerof iron-aluminumalloyusuallyco-ntainsabout25percentaluminum,whichresultsingoodresistanceto.heatandcorrosionandalso-exhibitsgoodtoughnessandductilitycharacteristics.A moredetailedaccountof thepackprocessisgiveninreference12,

Dipprocess:Inthedipprocessthepart__tobe aluminizediscleaned,dippedinmoltenaluminumfora contrd.ledtime,and,ifdesire{,suitablyheattreatedtopermitdiffusionof thealuminumintothesteel.Forpartsthataretooperateat temperatures~ceeding1000°F thetime--andtemperatureof thealuminumdipmustbe closelycontrolled(ref.14).Diptemperaturesareof theorderof1300°F anddipperiodsrangefrom -..15 secondsto6 minutes(ref.14). Theiron-aluminumalloyformedduring-theimmersionof thesteelinmoltenaluminumis extremelyhardandbrittleandcontainsabout55percentaluminumi.Inorder,todevelopasurfacelayerthatis softerandlessbrittlea“ndthatresistsspallingandcracking,a diffusionheattreatmentisgivenafterthedipping.Thediffusionheattreatmentiscarriedoutata temperatureofabout1800°Ffortimesvaryingfrom1 to 6 hours.Theexactdiffusionheat-treatment.._timerequireddependsuponthechemi,calcontentof thesteelandthedippingtimeinthemoltenaluminum.Reference’14 describesa patented~commericalaluminum-dipprocess. ..—

Boththepackpr”~cessandthedipprocess.,.~nvolveheatingof the ...aluminizedpartforextendedperiodsof timeat temperaturesof1500°to1800°F, whichmayadverselyaffectthephysicalpropertiesof the _parentmetalforturbinebladeap-plication.Theinteriorsurfacesof ..theblademaybe aluminizedby eitherprocess.: .

APPARATUSANDINSTRUMENTATION

PreparationofCoatedAir-CooledBlades

Bladeconstruction.- Allthecoatedair-cooledturbinebladesreportedhereinwereof theshell-supportedtypewheretheloadiscarriedby thebladeshell,Thespanwas4 inchesandthechord,approx-:

imatelyl;inches.Allthebladeswerenontwi.stedexceptblade4,whichwasgivena twisttoapproximatethatof theu~cooledblades.Theshellsofallbladesexceptblade1 wereformedby contourpressingseamlesstapered-walltubesintothedesiredairfoilshape;blade1 wascast. The

T- -.—

-

-.- .—--- ... . . .—,-.

..—

,.=--.-.-—

s “..-

—-—.—...=.u.——=.-

=—-.

.-—.-.

.—— —

- -.=-.._-— g —

.- —..—

k –--—

NACAF&lE53E18 7

shellmaterialusedin thesebladeswaseitherSAE4130or Timken17-22A(S)steel.Thesesteelswereselectedbecausestress-rupturedataforthemetaltemperaturerangeinwhichtheseair-cooledbladesweretooperateindicatedthatthesemetalsWerebestsuitedof thereadilyavaila-ble nonstrategicsteels.Ml thebladebasesweremadeof castW 4130steel.TableII indicatestheshellmaterialof eachblade.

Theinternalheat-transferareasof au bladeswereincreasedbyNOJ brazingmildsteeltubesto theinnersurfacesof theshells,as shown8 infigure1. Copperwasusedas a brazematerialforallbladesexcept

numbers26 through29;Nicrobrazwassubstitutedinthesebladesbecausecopperwouldbe attackedby moltenaluminumduringthealuminizingprocess.

.-

Bladefabricationproceduresarediscussedin detailinreference15.

Thethreebladeprofilesusedinthisinvestigationareshowninfigure1. Therootprofilesarethesameas thetip,exceptfora slightchangeintheoutsidecontourbecauseof thetaperedwallof theshell.ProfilesA andB (figs.l(a)andl(b),respectively)wereobtainedbyformingtheshe31s,andprofileC (fig.l(c))wasobtainedby casting.ProfilesA andC areessentiallythesameandarenearlyequivalenttotherootprofileof thestandarduncooledturbinebladeusedinthetest.engine.ProfileB wasan airfoilsectionwhichwasdesignedtooperateina completelyair-cooledturbinerotorwithtwistedstatorblades.TableII indicatestheprofileof eachblade.Thevariousbladesusedwereselectedbecausetheywerereadilyavailablefromotherinvestiga-tionsandtherebyreducedthetimerequiredtoprepareforthecoatinginvestigateion.

Bladecoatingsandheattreatments.- Thecoatingsandheattreat-mentsappliedto theindividualexperimentalbladesfollow.Heattreat-mentsw&e selectedon thebasisof availabledatato obtainthebeststress-rupturepropertiesofthemetalusedandweremodifiedas addi-tionaldataandoperatingexperienceweregained.

Ceramic-coatedblades:Fivee~erimentalbladeswereceramiccoatedforthisinvestigationby commercialconcerns;thecompositionsof thecoatingsandthedetailsof theirapplicationwereconsideredproprietaryinformationandwerenotrevealed.Blades1 and2 (blades7 and8 ofref.9)werecoatedwitha modificationofNationalBureauofStandards’A-19coatingandwerenotheat-treatedforfearof damagingthecoat-ings.Blade3 wascoatedwithSolaramic8042/3FEandsubsequentlyheattreatedas indicatedin tablesIIandIIIby theNACAafterconsultationwiththecoatingvendoras to thelimitationsimposedby theceramic.*

.

8 NACAW E53E18

Blades4 and5 werecoatedwithanothermodificationofNBSA-19coatingandheattreatedas indicatedintablesIIandIII. Allceramiccoatingswereappliedtoonlytheoutsidesurfacesof thebladeshell.

Nickel-coatedblades:Nickelcoatingswereappliedtoblades6through11by theNACAusingsolution1 of tableI andtoblades12through15by a commercialconcernusingsolution2 of tableI. No dif-ferenceinthecoatingswasnoted.Theinnersurfacesof thebladeswerenickelcoatedineachcase.Allbladeswereheattreatedas indi-catedintablesIIandIIIpriortocoating.

Nickel-andNicrobraz-coatedblades:A combinationnickelandNicrobrazcoatingwasemployedonblades16 through20. A Nicrobrazcoatingwasappliedtothelea~ngedgeandtheentireairfoilsectionwassubsequentlynickelcoated.Theindicatedheattreatment(tableII)wascombinedwiththeNicrobrazcoatingoperation.

Nicrobrazcoatings:Nicrobrazwasusedto coattheentireairfoil.sectionsofblades21,22,and23. Again,theindicatedheattreatment_wascombinedwiththecoatingoperation. ,

Aluminizedblades:Sixaluminizedbladeswereobtainedforthisinvestigation,twowerealuminizedby thepackprocessandfourby thedipprocess.Bothprocessesaluminizedthe.innerheat-transfersurfacesoftheblades.Thepackprocesswasusedonblades24and25;thebladeswerepackedinaluminumoxide,aluminumpowder,andan energizerfor12hoursat 1800°F.”Blades26through29werealuminizedbythedipprocess.Theywerepreheatedina saltfluxat 1320°F for5 minutes,dippedinmoltenaluminumat 1300°F for30seconds,washedinmoltensaltat 1320°F for30 seconds,aircooled,washed,andheattreatedasindicatedintablesIIandIII. Theheattreatmentpermittedthedesireddiffusionofthealuminumintothesteel.Filletswereappliedattherootsofblades26through29titeraluminizingtoreducestressconcen-trationsinthoseregions.Intheprocess,thecoatingsintheadjacentareasweredsmagedslightly.

Engines.- Severalproductionturbojetenginesweremodifiedtoallowcoolingairtobe suppliedtoeithertwoorfourexperimentaltur--binerotorblades.Themodificationswereessentiallythosedescribedinreference1. An adjustabletail-pipenozzlewasusedtoregulateturbinegastemperatures.Bladecoolingairwassuppliedfromacompressed-airsystemexternaltotheengine.Effectivegastempera-turesat theturbinebladesweremeasuredlychromel-alumelthermo- -couplesburiedintheleadingedgesof standarduncooledbladesatasection2A inchesfromthebladetip. No thermocoupleswereinstalledon cooled%lades.Detailsofthethermocoupleinstallationaregiveninreference1.

—

.

-+

—

.- . —

——

—

—

..—

————— — —

—

—

———

- -— —.=

——-.

x—=——

.

NACARM E53E18 9

PROJEDURE

Foran evaluationof theeffectivenessofthevariousprotectivecoatings,twotypesof engineoperationhavebeenemployed- constant-speedandcyclicoperation.Inconstant-speedrunning,theenginewasoperatedatmaximumratedspeed(11,500rpm)withthetail-pipenozzleadjustedtoobtainan effectivegas(oruncooledblade)temperatureof1450°F, whichcorrespondstoayproxhatelya 1670°F turbineinletgastemperature.Cooling-airtemperaturesatthebaseofthebladewereabout180°F. Theseconditionswillhereinafterbe designatedratedtestconditions.Thecoolant-flowratiowassetat thedesiredlevelonceenginespeedandgastemperaturewereestablished.Conditionswerethenmaintainedconstantforthedurationof theendurancerun.

y

g

Cyclictestsconsistedofoperationat theratedtestconditionsfor15minutesandthenat idlingspeed(4000rpm)for5 minuteswithno changein eithercooling-airflowcontrolsor tail-pipenozzleposi-tion. Theenginewasthenacceleratedtoratedspeedandthecyclerepeated.Acceleratinganddeceleratingperiodswereof theorderof15 seconds.

Blades1, 8,9, 10,andXl,whichwerefirstreportedin.refer-●

ence9,weresubjectedto cyclicengineoperation.All’otherbladesweresubjectedto steady-speedoperation.TableII indicatesthetypeandamountofrunningtimeaccumulatedon eachblade.,Cyclicoperationwasfoundtobe verysevereonenginecomponentsandexcessiveamountsof timewererequiredforenginerepairandmaintenance.Inordertoexpeditetheinvestigationof coatingsthecyclictypeof operationwasdiscontinuedinfavorof thesteady-speedrunning.Itwasbelievedthatthesteady-speedoperationsubjectedthecoatedbladestoa sufficientnumberofrapidtemperaturechangesduringthestartingandshut-downoperationsto demonstratetheabilityofa coatingtowithstandrepeatedthermalshock.

Forflightapplication,themaximumenginespeedforcontinuousoperationis11,000rpm,whileoperationat 31,500rpmislimitedtohaM?-hourperiodsfortake-offorcombat.TheNACAtestspeedwassetat 11,500rpm. Tail-pipetemperaturesin servicesrelimitedto 1292°Fexceptforstartingandaccelerating;theI?ACA,inmaintaininga constantuncooledblade(oreffectivegas)temperatureof1450°F, encounteredtail-pipetemperaturesrangingfrom1280°to 1350°F, dependingonsaibientconditionsandtheconditionoftheequipmentbeingused,with1325°Fbeingtypicalofmostoftheoperation.

10 NACARM E53E18

Duringthecourseofthecoatinginvestigation,thecoatedblades=wereoftendsmagedby failureofuncooledbladesorof otherair-cooledbladesorby someobjectwhichpassedthroughtheturbine.Therewereinstanceswherecoatedbladesfailedinrupturewhilethecoatingwasingoodcondition.Suchfailures,whichresultedfromcausesnotrelatedto thecoatings,will.hereinafterbe referredto asmechanicalfailures.Inmanycases,thedmnagetothe-experimentalbladewascon=finedtothetipregionandthemajorportionofthebladecouldbe T-salvagedandusedforfurther-testing.Whilecentrifugalstresseswere”reducedandthevibrationalcharacteristicswerechanged,itwasfeltthat,insofaras coatingdurabilitywasconcerned,theresultsfroms’uchbladeswouldnotbe alteredappreciably.

——,-——

-T

— —.-..- ti–

“: g-

——-—

—

RESULTSANDDISCUSSION

Theresultsofenduranceinvestigationson severaltypesof ..-

corrosion-resistantcoatingsappliedto air-~ooledturbinebladesofSAE4130orTimken17-22A(S)steelarereportedintheensuingsection. * -—---QandaresumarizedintableII.

●

CersmicCoatings

Fiveceramic-coatedbladeswereendurance-tested..

Twocoatingfailureswereencounteredandbladefailuresfromcausesnotconnectedwiththeircoatingsterminatedthetestsontwoothersbeforeanysignificantamountofrunningcouldbe obtained.Thefifthbladesuccessfullycompleted100hoursofoperationattheratedtestcon-ditions.

..- ———

Blade1. - Thecoatingonblade1 wasconsidereda failureafter7.3hoursofratedenginespeedoperation,andinvestigationoftheblade_. .—wasconcludedattheendofthattime.Figure3(a)showsseveralviewsofthebladeattheconclusionofthetestsanditcanbe seenthatthe

—

coatingchippedawayfromtheleading-and+jrailing-edgeregionsoftheblade.Thereisalsoan areaneartherootinthemidchordregionof

. —

NACARM E53E18 11

thebladewherethecoatingflakedoff. Thefailureof thecoatingapparentlywasa functionof themetaltemperature,as thegreatestdam-ageto thecoatingoccurredat theleading-andtrailing-edgeregionswherethemetaltemperaturesarethehighest(seefig.2). Morecoatingwasremovedfromtheleadingedgethanfromthetrailingedge,probablybecausethehigh-velocitycombustiongasesimpingedirectlyupontheleadingedgeandtheirerosiveactionisgreatestinthisarea.

Blade2. - Blade2 wasconsidereda failureafterabout4.7hoursof operation.Inspectionat thistimeindicatedthatthecoatingappar-entlysoftenedwhenheatedandflowedtowardthebladetipundertheinfluenceof thehighcentrifugalforces.A photographof thisbladeisshowninfigure3(b). Theflowlineswereessentiallyparalleltothebladebaseovertherelativelycoolmidchordregiononboththepres-sureandsuctionsurfaces.Intheleading-andtrailing-edgeregions,whichoperatehotterthanthemidchordregion,theflowlineswerenearlyradialandwereverypronouncedas showninfigure3(b).Baremetalwasvisibleinthevalleysbetweenflowlinesintheleading-andtrailing-edgeareas.Inasmuchas thebladewasnotheat-treated,therewasnopossibilityof damagefromthissource,anditmustbe assumedthatthecoatinglackedthenecessaryphysicalpropertiesfortheapplicationinquestion.

.Blade3. - Blade3 wasgivenheattreatment5 of tableIII. This

heattreatmentwasspecifiedandappliedtothebladeby theNACAafter. correspondencewiththecoatingsupplierrelativeto therestrictions

@osed on theprocessby theceramiccoating.Afterthebladewasheat-treated,thecoatingwasdiscoloredinscatteredareasandmayhavebeendamaged.Neverthelessthebladewastestedtoseewhetherthedamagewouldprogress.After8 hoursof operationthebladewasinspectedandthecoatingappearedtobe inas goodconditionaswhenthetestbegan.Operationwascontinuedandat 11.3hoursdamagetothecooling-airsupplysystemresultedin theair-cooledbladeoverheatingandfrac-turingat a sectionabout1/3spanfromtheroot. Thecoatingontheremaining1/3of thebladeappearedtobe inthesameconditionaswhenthetestbeganexceptforslightevidenceof erosionat theleadingedge.Althoughno definiteconclusionscouldbemaderegardingthedura-bilityof thiscoating,itappearedpromising.

Blades4 and5. - Blades4 and5 werebothcoatedwiththesaneceramic,a modifiedNBSA-19typecoating,andweregivenheattreat-ment1 (tableIII)by thecoatingsupplier.InspectIonofthebladesafter8.3hoursof operationindicatedthatbothwereinexcellentcondition.Blade4 failedinfatigueat therootshortlyafterthisinspectionand

dthetestswerecontinuedwithblade5 only.At approximately20hoursoperationa foreignobjectpassedthroughtheturbineandstruckblade5,withtheresultthata portionof thecoatingalongtheleadingedgeof

NACARM E53E18

thebladewaschippedoffandbaremetalappearedtobe exposed.Opera-tionwascontinued,however,andafter34.3hoursthebladewasagainstruckby a foreignobjectanddamagedinaboutthesameareaaspre-viously.Thisdsmagecanbe seenon theleadingedgeof thebladein~~figure4(a).Severalsmallerchippedareaswereobservedinthecoatingon theforwardportionof thesuctionsurfacenearthebladetipandcanalsobe seeninfigure4(a). Investigationofthecoatingwascontinueduntilthebladewasoperatedfor100hoursatratedtestconditions.Figure4(b)showsthebladeuponcompletion-of100hoursofoperation.

.--- - .

-.. .. . +..

i-..-—-

..*

.._

(Thedamtigetovisibleinthemishandlingofexaminationofindicatedthatvialedcomplete

thecoatingalongthetrailingedgeof thebladewhichis ‘“ g-..-.—

viewof thepressuresurfaceinfig.4(b)wascausedby”.- Co

thebladeaftercompletionof.thetests.)MicroscopicN

a sectionthroughthedamage@_areaat theleadingedge.—

someceramicstillclungtothemetalandapparentlypr6-.<

corrosionprotectionsinceno corrosioncouldbe observed.~—

underthemicroscope.Duringtheinvestigationofblade5 no softeningorflowingofthecoatingwasobserved.No__ghipping‘orflakingother..thanthatcausedby foreignparticlespassingthroughtheturbine&sobserved.Slightlyraisedmarkingsthatbecameevidenton thesurface

——

of thecoatingafterheattreatingwere stillvisibleafter100hoursofrunning,indicatingthatthecoatinghadnoterodedtoanygreatextent.After100hoursofoperationthecoatingwasinexcellentcon-ditionexceptwhereithadbeenhitandappearedtoprovidecompleteprotectionforthebladeshelleveninthose-areas:Corrosionwwi ‘–observedon theuncoatedinnersurfacesof~he shellandon thecooling-airtubesto depthsof0.003to0.005inch.Theseobservationsweremadeintheimmediatevicinityof theleadingedge.

.

..._.—.—_

.—.——...z—--..——

.—== —...

.<—

Theinvestigationoffiveceramic-coatedturbinebladesindicatesthata ceramiccoatingiscapableofproviwngcorrosionprotectionto -an air-cooledbladeofSAE4130steelfor100hoursofoperationat

.-.

speedsandtemperaturesequaltoormoreseverethanthoseencountere&.

atmaximumratedconditionsina present-@”yturbojetengine.WhetherthiscoatingwouldalsoprovidesatisfactoryprotectionforTimken “-17-22A(S)materialisnotdefinitelyknown;thecoatingsupplierbelieves, _

____ -—

however,thatTimken17-22A(S)canbe successfullycoatedwithperhaps -.

slightmodificationof thecoatingandthe.applicationprocedure.The—

abilityof thiscoatingtoadhereandprovideprotectionaftersevere : - Z-

chippingis significantandisa verydesirablecharacteristicfora .—ceramiccoatingforturbinebladeapplication.

—=Thecoatingonblade=

appearedtobe promising,althoughtheshortendurancetestgivenhere.—.

makesfurtherworknecessarybeforedefiniteconclusionscanbe reache=. =—

Tenhavinga

NickelCoatings i.

nickel-coatedblades(blades6 through15)andfiveblades —-conibinationofnickelandNicrobrazcoatings(blades16 —-w%

—

NACARM E53E18

.through20)wereshellsandeight

.

13

investigated..Sevenof thesebladeshadSAE4130steelbladeshadshellsofTimken17-22A(S)steel.

B~des 6 through15.- Of thetenbladesthatdependedentirelyonnickelcoatingforcorrosionprotection,blades6, 8,9, 11,12,14,and15operatedforsufficientlengthsof timetoobtainsignificantcoatingresults.Blades7,10,and13 failedmechanicallyandinvesti-gationof thecoatingson thesebladeswasterminatedaftershortperiodsof time,as shownintableII. Thelastinspectionsof thecoatingsofthesethreebladespriorto theirmechanicalfailureindicatedthecoat-ingswerestill.ingoodcondition.

Blades6, 8,9, 1.1,12,14,and15wereoperatedatratedtestcon-ditionsforperiodsof timerangingfromabout11.5hourstoasmuchas25hoursbeforethecoatingswereconsideredfailures.Thefailuresofthenickelcoatingswereallsimilar,thatis,thenickelbegantoblis-terandchipawayfromtheleading-andtrailing-edgeregionsof thebladesas showninfigure5(a). As operationof thebladeswascon-tinued,corrosionof theparentmetalintheleadingandtrailingedgesoccurredandthenickelcoatingalsobegantofailin themidchordregionof theblade.Thecoatingsgenerallybegantofailfirstin thetipregionof thebladeafterabout11 to 15hoursof ratedenginespeedoperation.8 Temperatureandcentrifugalforcearebothfactorsin thebreakdownof thecoating.Failureusuallyprogressedrapidlyalongtheleadingedgeof theblade,whereerosionalsoseemedtobe a factorin

. removalof thecoating.A bladethatexhibitedextensiveerosionandcorrosionat theleadingedgeisshowninfigure5(b),whichisa photo-graphofblade6 at thecompletionof 44hoursatratedtestconditions.Thefailureof thecoatingalongthetrailingedgewasgenerallylessextensivethanthatat theleadingedge,probablybecausethescouringactionof theconibustiongasesisnotso greatin thisregion.Also,trailing-edgetemperaturesaresomewhatlowerthanthoseat theleadingedge.Thetemperaturelevelof thebladematerialinfluencesthelifeof thenickelcoating;failureof thecoatingin thecoolermidchordregionwasmuchmoreinfrequentthanat theleadingandtrailingedges(seefig.2). Frequently,no evidenceof failurewasobservedin themidchordregion.Thelowertemperatureapparentlyenablesthenickeltoadhereandprovidethenecessarycorrosionresistance.

Blades16 through20.- Becauseinvestigationofblades6 through15 indicatedthatthenickelcoatinggenerallyfailedinitiallyin theleading-edgeregionandthatthisfailureappearedtobe causedbycor-rosionanderosion,itwasthoughtthata corrosion-resistantmaterial,harderthannickel,shouldbeappliedtotheleadingedgeof theblade.Thebalanceof thebladesurface(exceptperhapsthetrailingedge)mightdthenbe adequatelyprotectedbya nickelcoating.Fivesuchblades(blades16 through20)wereprepared.A Nicrobrazcoatingwasfirst

—

. appliedtotheleading-edgeregionof thebladesas describedin the .-

14 NACARM E53E18

PFWEDUREsectionof thisreport.me blad& werethennickel-coated,andinthismannertheNicrobrazformedanundercoatforthenickel .alongtheleadingedge.

Blades17 and18 failedinfatigueat thebladerootafter11.4and23.7hoursofoperation,respectively..~e coatingswereingoodconditionat theinspectionspriorto thetimeofthefatiguefailures.Blade19wasdamagedbeyondrepairby fragmentsfroma failedbladeafter25.6hoursof operation.Thecoatingon thisblade.hadshownev~~denceof slightflakingin themidchordregionnearthetip,butthecoatingon therest-ofthebladewasingoodconditionpriorto thetfi-ethebladewasdamaged.Blades16and20weti=operatedsuccessfullyfor_101.2and.100hours,respectively.Thecoatingonblade16wasingoodconditionexceptforscalingof thenickelo_ntherearthirdof thesuc-tionsurface.No corrosionof theblademetalwasapparentinthisarea.At theendof100hoursof operationonblade20thecoatingwasinexcellentconditionexceptfora smallareaonthepressuresurfacenearthetrailingedgewherethecoatinghadbeguntoflakeoff(fig.6). Nocorrosionof theparentmetalwasobserved.In theleading-edgeregionnearthetipon thesuctionsurfaceofblade20thereappearedtobe. __severalsmallareaswherethenickelhadflakedaway. Thisdsmagewas..causedby foreignparticlespassingthrough.$heenginewhichstruckthe-bladeintheleading-edgeregionwheretherewasa Nicrobrazundercoat,and,althoughthenickelchippedaway,theNicrobrazprotectedtheparentmetal.Theadherenceof thenickelto theNicrobrazappearedtoelim-inateflakingofthenickelandenabledit tQ protecttheblade.The _.underlying”Nicrobraz,whilecapableofprovi~ngprotection,wasnotcalleduponto dosoexceptwherethenickel_.waschip~e,da~y by foreignparticlesstrikingtheblade. —

Duringtheinvestigationofthefivenickel-coatedbladeshavingNicrobrazundercoatat theleadingedges,noneof thetestsweretermi-natedbecauseof coatingfailure.Adherenceof thenickelovertheentirebladeappearedtobe betterinthisgroupofbladesthanintho~epreviouslyemployed.Blades16 through20had shellsofTimken17-22A(S)steel;whilethepreviousgroup,withtheexceptionofblades7 and8,.

——

,. _—

-—.. —

—

.-, .- —.— —

a-,:.—

—————

~ _ —--.—.——

———

●“ ~..—.—— .-—

—-+

w—— .— —

.—.<

.——

—---...-

——_

hadshellsof SAE4130steel(seetableII). Basedsolelyon observa-tionof theseblades,itappearsthatthea~erenceofnickel,whenappliedin themannerpreviouslydescribed,isgreatertoTimken17-22A(S)steelthantoSAE4130steel. .-

NicrobrazCoatings

Blades21,22,and23.- Becauseof thesuccessobtainedinusingNicrobrazasanundercoatfornickel,itwasbelievedthata completely

——

-

~ -=-.

.

NACARM E53E18 15

mla~

.

Nicrobraz-coatedblademightoperatesuccessfully.Furthermore,theeasewithwhicha Nicrobrazcoatingcanbe appliedandthefactthatitsapplicationcanbe combinedwithotherbrazingoperationsduringbladefabricationmadetheNfcrobrazcoatingappeardesirable.Con-sequently,blades21,22,and23wereNicrobraz-coated.

Blade21 successfullycompleted100hoursofoperationat rated.testconditionswitha coolant-flowratioof 0.048butfailedmechan-icallyafteran additional1.2hoursof operationat a flowratioof0.03.TheNicrobrazcoatingwasinexcellentconditionafter100hoursandshowedno evidenceof impendingfailure.Blades22and23wereoperatedfor31.5and5.9hours,respectively.Bothbladesfailedinfatigueat thebladeroot. Priorto failurethecoatingsonbothbladeswereinexcellentcondition.

ThisinvestigationindicatesthatNicrobrazcoatingsprovidegoodcorrosionanderosionprotectionforTimken17-22A(S)steel.It iS

believedthatNicrobrazwouldoffersimilarprotectionto SAX4130steel.Nicrobrazpossessesstrongalloyingcharacteristicsandmaypenetratetheparentmetalto somedegree;itseffecton thefatiguestrengthofthebladeshellremainstobe determined.

.AluminizedCoatings

. Blades24and25.- Blades24and25werealuminizedby thepackprocess.Blade24 successfullycompleted100hoursof operationandthealuminizedcoatingwasinexcellentconditionat thecompletionofrun-ning. Slighterosionof theleading-edgesurfacewasnoted,butitwasnotextensiveandno evidenceof corrosionwasobserved.Themidchordandtrailing-edgeregionsof thebladeshowedno evidencesof erosionor corrosion.Blade25failedstructurallyafter8 hoursof operation;thebladesurfacewasin excellentconditionuntilthattime.

Thesuctionsurfaceofblade24wasstruckby fragmentsof a failedbladeafter50hoursof operationanda numberof shallowscratchesresulted(seefig.7(a)).Thealuminizedlayerwassufficientlytoughtowithstandtheimpactandno damagewassustainedotherthanthescratchesthemselves.Uponfurtherrunningthescratchesdisappearedandwerepresumablyfilledinby iron-aluminumcompoundsfromtheadja-centsurfaces.Thesurfaceof thebladeafteran additional50hoursofoperationshowedno tracesof thedamage(seefig.7(b)).

Thesurfacesofblades24and25wereconsiderablyaluminizingthantheywerepriortobeingcoated.This.notchangewithoperatingtime. Thissurfaceroughnessbladesaluminizedbythepackprocess.

rougherafterroughnessdidistypicalof

16 NACARM E53E18

Beforeoperationthecolorofthebladesurfacewasdeepgray.Afteroperation,blade24exhibiteda reddishcoloralongtheleadingedge,overmostof thesuctionsurface,andoverabout20percentofthepressuresurface.Thischangeincolorisapparentlytypicaland ..istheresultof theformationof complexiron-aluminumoxideswhichprovidea protective,adherentrefractorycoatingon thesurfaceofaluminizedsteels(ref.14).

Blades26 through29.- Blades26through29werealuminizedbythedipprocess;a= thebladeswereoperatedsuccessfullyfora minimumof100hoursatratedtestconditionsas shownintableII. Blades26and27wereoperatedfor124.4hours.Thecoatingso?allthebladeswereinexcellentconditionat theconclusionof thetests.Allthe .bladesof thisgroupweredamagedduringthejcourseof theinvestigationby severalmechanicalfailures;inordertocontinueinvestigationofthecoatingsthetipsof thebladesweregroundoff. Thebladeswere,,,thereforeshorterthannormal,particularlyblades26,27,and28,asshowninfigure8. Theshorteningofthebladesresultedinlowerstressesinthebladesshells,butthiswouldnotbe expectedtoaffectthelifeof thecoatingsappreciably.Anypossibleweakeningof theparentmetalas a resultof thealuminizingtreatmentwould,of courselnotshowup soreadily.

Thesurfacesof thebladesthatwerealuminizedbythedipprocesswereonlyslightlyrougherafteraluminizirv..thanbeforebeingcoated~_..andthesurfacesof thisgroupofbladeswereconsiderablysmootherthanthosealuminizedbythepackprocess.Afteraluminizingthecolorofthebladesurfaceswasa deepgrey.Duringoperation.thebladesurfacesdevelopeda reddish-orangecolorovermostof theblade.Thiswassim-ilartothecolorobservedonblades24and25.

Figure8(a)showsthesuctionsurfacesofblades26,27,and28earlyin theinvestigationandofblade29priortooperation.Fig-ure8(b)showsboththepressureandsuctionsurfacesof thebladesatthecompletionof theinvestigation.Comparisonof thesuctionsurfacesshowninfigures8(a)and8(b)indicatesthattherewaslittlechangeintheconditionofthecoatingon theblades,leventhoughtheblades..of__figure8(b)haveabout100hoursmoreoperatingtimethaninfigure8(a).Withoneexception,theentiresurfaceof eachbladeincludingtheleading-andtrailing-edgeregions,whichwe~ethefirsttoexhibitfailureson theceramicandnickelcoatings,wasinexcellentconditioti”at thecompletionof thetests.Surfacescratchesandabrasionsshowedno tendencytoprogressandno corrosiondevelopedinthemas theinvesti- _ ~gationcontinued.Thetipregionsofblades26md 27weredamagedwhenstruckby fragmentsofanotherblade,butthealuminizedcoatingscon-

—2=

tinuedtoprotectthesurroundingareas.No..corrasionwasevidentexcelt.___– _onblade27,wherethemetalwasactuallytornandthebaresteelwasexposedto theactionof thegases.

=e..-

NACARM E53E18 17

b

“

%.~

.

Theoneexceptionto thegeneralexcellentconditionof thebladesurfaceswasalongtheleadingedgeofbh”de28,whereseveralshort,finecrackswerevisibletotheunaidedeye. Thesurfacesfromthelead-ingedgeas farbackas thefirstcooling-airtubewerea differentcolorfromtherestof theblade,beinga muddygraywithoutredor orangetingesnotedelsewhereon thesamebladeandon theotherbladesof thisgroup,includingtheleadingedges.Examinationshowedthatthecoolingpassagealongtheleadingedgehadbeenalmostcompletelyblockedbyloosescale,presumablyfromtheexternalcooling-airsystem,whichhadwedgedintoa restrictedregionat thebladeroot. Completeblockagewouldhavegivena leading-edgetemperatureof theorderof 1400°F.Microscopicexaminationof sectionsalongtheleadingedgeandat90°totheleadingedgeabout1.8inchesfromthebladebasebroughtoutthefollowingfacts:TheTimken17-22A(S)steelintheleading-edgeregionof thebladeshellhadsoftenedbecauseof itshighertemperature;thehardnessneartheleadingedgewaslessthanRockwellC-10as contrastedwithRockwellC-28intheregionoppositethefirstcoolingtube. Thecoatingat theleadingedgeshoweda RockwellC-19hardnessas comparedwitha C-28hardnessoppositethefirstcooling-airtube.Thealuminizedlayercontaineda seriesof finecracksovertheentiresection.Inthecoolerregionnearthecooling-airtubes,thesewerehairlinecrackswhichdidnotpenetratethecoatings;neartheleadingedgetheyoccurredwithaboutthesamefrequencybutwereheavierandsometimespenetratedthealuminizedlayer.Whenthecoatingwaspenetrated,a corrodedregionmushroomedoutfromthecrackintotheunderlyingsteel.At thesectionexaminedthecoatingwasthickerontheinsidethanontheexteriorofthebladeshell;possiblythegreatererosiveactionof theexhaustgasesaccountsforthiseffect.TheSAE1020steelcooling-airtubesofblade28werealuminizedalmostcompletelythrough;thealuminumpene-tratedallexcepta 0.001-to0.0015-inch-thickregioninthecenterofthetubewall.

me coatingsintherootregionsof thisgroupofbladeswereroughandtheblades.themselvesappeartobe damaged.Thisconditionarosewhenthefilletswereappliedatthebladerootssubsequenttothealumi-nizingoperation.Theaffectedareasdidnotincreaseinsizeas theenduranceoperationprogressedandthebladeshellswereprotectedade-quately.Withinthefilletedareaitself,however,the”shellofblade28wasdamaged,presumablyduringthefilletingprocess,andcracksdevelopedtiter118hoursofoperation;reducingthebladelengthby cuttingoffdamagedportionsneerthetipundoubtedlyforestalledanearlybladefail-ureinthiscase.

TheworkwithaluminizedbladesindicatesthataluminizingisasatisfactorymethodofprotectingTimken17-22A(S)turbinebladesinpresent-dayturbojetenginesatmaximumratedconditionswithbladetem-peraturesup to1200°F. Undertheconditionsofoperation,thecoatingapparentlybreaksdownatmetaltemperatures14-00°F.

-

betweeu1200°&d about -

——

18 9m9mBirGeneralComments

Ithasbeendemonstratedthatair-cooled

NACARM E53E18

turbinebladesofnonstra-tegicsteelscanbe successfullyoperatedina turbo~etengineforextendedperiodsoftimewithreasonablecooling-airflowrateswhenpro-tectedbyceramic,nickel-Nicrobraz,Nicrobraz,oraluminizedcoatings.Itisbelieved,however,thatmoreresearchisrequiredbeforeanyoftheseprotectivecoatingscanbe s??ecificallY.reco~endedforservice.useina turbojetengine.

Theendurancelifeofa nonstrategicturbinebladedependsnotonlyon thedurabilityof thecoatingbutalsoon__thefatiguestre~th~creeyJandstress-rupturepropertiesof theparent_~terialat elevatedtempera-tures.Theeffectsof thevariouscoatingsonthesepropertiesare.no{”entirelyknownat thepresenttime. m.-

Thedepthofpenetrationof ceramiccoatingsintotheparentmetalisbarelymeasurableand,it isbelievedat thistime,canbe neglected”insofaras anyeffecton thestreuthpropertiesof thebladeiscon--cerned.Thelimitationsthata givenceramiccoatingmayimposeupontheheattreatmentof a steelarean importantconsideration,however.Themechanicalfailureofblades1 and10can”beattributedto theabsenceofheattreatment(seeremarksoftableII). Fromtheltiitede~erienceattainedwiththeceramic-coatedbladesofthisinvestigation,itappearstobe desirabletousea coati~.@osefiring!ernperatureis_...aboutthesameasorlowerthanthatofthedesirednormalizing”tempera:ture,sothattheheat-treatandfiringoperationscanbe combined.T&i.sisnotonlyconvenient,butmaybe necessarytoavoiddetrimentaleffectstotheceramiccoatingthatmayresultwhentheheattreatmentisa.sub-sequentoperationor tothemetalproperties-whentheceramicisfired_afterthebladeisheattreated.

Thenickelcoatingonbladeshavingnickelornickel-Nicrobrazcoatingsdoesnotpenetratetheparentmetalsignificantlyandwouldnotbe expectedtoinfluencethestrengthor,creeppropertiesof thebladeshell.

Nicrobrazalloysreadilywithsteelsandundoubtedlyaffectstheirstress-ruptureandfatigueproperties.However,suchdataarenotava+l-able. It isbelievedthatthehighpercenta&e_of.fatiguefailureson .blades16 through23wasduetofabricationtechniquesratherthandetr~-mentaleffectsoftheNicrobrazonthemetalof thebladeshells.

Whenbladesarealuminizedtheymustbe heatedto1500°F orhigherforseveralhoursinorderto diffusethealuminumproperly.InthecaseofSAE4130andTimken17-22A(S)steels,extendedheatingat tem-peraturesaboveabout1750°F mayadversely-affecttheirstrengths.Aluminizedsteel,ofcourse,hasa layerof iron-aluminumalloYat thesurface.Whateffect,ifany,thisalloylayermayhaveuponthestrengthofthesteel,particularlyinthinshellsof turbineblades,isnotknown.

wallssuchas usedin the .,

_.

NACARM E53E18 19

d

It isobviousthatbeforeanycoatingcanbe recommendedforserviceuseitseffectson thephysicalpropertiesof theblademetalmustbe

. known.

Theimportanceofprovidingprotectionovertheinteriorsurfacesof thebladeswasbroughtoutina numberof casesduringthisinvesti-gation.Corrosionwasnotedonthemildsteelcooling-airtubesofbladesafterthreeor fourdaysexposureto theatmosphere,evenwhenthosebladeshadnotbeenoperated.inan engine.After100hoursof

~ operation,corrosiontoa depthabout0.005inchwasnotedintheleading-“ edgeregionofblade5 on theuncoatedinnersurface;inthinshellsofair-cooledblades,corrosionto thisdepthwillhaveanappreciableeffecton thestrengthof theblade.Inmoredestructiveatmospheres,suchas encounteredincarrier-basedoperations,corrosionwillbe moresevereandtheneedforprotectioncorrespondinglygreater.

SUMMARYOFRESULT’S

Theresultsof an experimentalinvestigationto determinethedura-bilityof severalprotectivecoatingsappliedtononstrategicair-cooledturbinebladeswhichwereoperatedina twbojetengineareasfollows:

. 1. Satisfactoryprotectionof thebladeswasprovidedbyceramic,nickel-Nicrobraz,Nicrobraz,andaluminizedcoatings.Eachof thesecoatingsindicatedthatitwouldgivesatisfactorycorrosionanderosion.protectionto theturbinebladesfor100hoursofmaximumratedenginespeedoperationwithcoolant-flowratiosin therangeof 0.030to0.048.However,theadherenceofthenickelwaserraticandunpredictable.

2.Bladesaluminizedby thepackandby thedip-and-diffusionpro-cessesgaveexcellentprotectionagainstoxidattonandcorrosion.Allthealuminizedblades(exceptonethatfailedmechanically)operatedforatleast100hours.Twobladeswererunfor124.4hoursandwereinexcellentconditionwhenthetestswereterminated.

3. Oneceramiccoating(amodifiedNBSA-19ceramic)providedexceXlentservice.Thiscoating,althoughseeminglychippedcompletelythroughincertainregionsby foreignpsrticlesstrikingtheblade,wassufficientlyafierentto leavea thinfilmwhichgavesatisfactorypro-tectionto theunderlyingmetaloveranextendedperiodofoperation.

4.Chemicallydepositednickelcoatingswerefoundtorequireanundercoatingof sometypeattheleadingedgeofthebladeinorderto

20~

NACARME53E18 “- ‘--Z

preventflakingandfailureofthenickelinthisregionof theblade.A Nicrobrazundercoatprovideda satisfactorybaseforthenickelat theleadingedge.Bladeshavinga Nicrobrazundercoatat theleadingedgeanda chemicallyd@positednick”elcoatingovertheentirebladesurface-gaveadequateprotectionfor100hoursofratedenginespeedoperation.However,thenickelcoatingsinthisinvestigationwereerraticinsofarasadherencewasconcernedandoftenfailedinmuchshorterperiods.

5.Nicrobrazcoatingsappliedto theleading edge or to theentirebladesurfaceexhibitedverysatisfactorycoricosion-resistantproperties.

6.AluminumandNicrobrazalloywithanduntkmbtedlyaffecttosomeextentthefatiguestrengthandstress-to-rupture‘pro~ertiesof the .,..metalstowhichtheyareapplied.Applicatig_nof thealuminumorNicro-brazcoatingsmayresultinoverheatingtheparentmetalpriortoheattreatment.Theeffectsof thesevariousfactorson thestrengthof thebladeshell.arenotknownat thepresenttime,buttheyshouldbe evalu-atedbeforeanyprotectivec“oatingisconsideredforserviceuse.

7.In-viewof-theinsignificantyenetra~ion.of._ceramiccoatingsintothemetalshell,itis.believedthatthemetalpropertiesarenot__affectedappreciably.However,thefiringt~eratureof.theceramic..andtheheattreatmentofthemetalmustbe carefullycorrelatedtoavoiddamageto eithertheceramicor themetal.

LewisFlightPropulsionLaboratoryNationalAdvisoryCommitteeforAeronautics

Cleveland,Ohio,April14,1953 .

1.Ellerbrock,HermanInvestigationofI - RotorBladesRME50104,1950.

2.Hickel,RobertO.,Investigationof

REF-CES

H.,Jr.,andStepka,FTancisAir-CooledTurbineBladesinwith10TubesinCooling-Air

.,

.

._-=

_.● ✍✍

�

.—

——-—

-—=

. .-—

..T.._

.- .:.

—.

s Experimental -T-—.:TurbojetEngine.

——-Passages.NACA —. ,.—..-

andEllerbrock,Herman-H.,Jr.: Experimental —

Air-CooledTurbineBladesinTurbojetEngine.—

.,-=.—..——II - RotorB~des tith15FinsinCooling-AirPassages.NACA _. ._ .=RME50114,1950. .- —— —

3.Hickel,RobertO.,andSmith,GordonT.: ExperimentalInvestigationofAir-CooledTurbineBladesinTurbojetEngine.III- RotorBladeswith34SteelTubesinCooling-AirPassages.NACARM E50J06j1950,..

------~..—.—

.

NJm8

NACARM E53E18 21

.4.Ellerbrock,HermanH.,Jr.,Zalabak,CharlesF.,andSmith,GordonT.:

ExperimentalInvestigationofAir-CooledTurbineBladesinTurbojet. Engine.IV - Effectsof SpecialLeading-andTrailing-EdgeModifi-

cationonBladeTemperature.NAcARME51A19,1951.

5.Smith,GordonT.,andHickel,RobertO.: ExperimentalInvestigationofAir-CooledTurbineBladesinTurbojetEngine.V - RotorBladestithSplitTrailingEdges.NACARM E51A22,1951.

6.Arne,VernonL.,andEsgar,JackB.: ExperimentalInvestigationofAir-CooledTurbineBladesinTurbojetEngine.VI - ConductionandFilmCoolingofLeadingandTrailingEdgesofRotorBlades.NACARME51C29,1951.

7.Smith,GordonT.,andHickel,RobertO.: ExperimentalInvestigationofAir-CooledTurbineBladesinTurbojetEngine.VIII- RotorBladeswithCappedLeadingEdges.NACARM E51.H14,1951.

8.Stepka,FrancisS.,andHickel,RobertO.: E@erimentalInvestiga-tionofAir-CooledTurbineBladesinTurbojetEngine.IX - Evalua-tionof theDurabilityofNoncriticalRotorBladesinEngineOpera-tion.NACARM E51%L0,1951.

* 9.Esgar,JackB.,andClure,JohnL.: ExperimentalInvestigationofAir-CooledTurbineBladesin TurbojetEngine.X - EnduranceEvalua-tionofSeveralTube-FilledRotorBlades.NACARM E52B13,1952..

10.HarrisonjWillismN.,Moore,DwightG.,andRichmond,JosephC.:CeramicCoatingsforHigh-TemperatureProtectionof Steel.Res.PaperRP1773,U.S.Dept.Commerce,Jour.Res.Nat.Bur.Standards,vol.38,Mar.1947,pp.293-307.

11.Brenner,Abner,andRiddelJ_jGrace:DepositionofNickelandCobaltby ChemicalReduction.Res.PaperRP1835,U.S.Dept.Comuerce,Jour.Res.Nat.Bur.Standards,vol.39,Nov.1947,pp.385-395.

12.Sayles,B. J.: AluminumImpregnation.MetalsHandbook,Am.Sot.Metals,1948.

13.Finkbone,B. P.: HotDippedAluminumCoatingson Steel.MetalsHandbook,Am.Sot.Metals,1948.

.

22 NACARM E53E18 .—

14.Hanink,D.K.,.

andBoegehold,A. L.: CoatingSteelby theAltiP ___._ _ _ <Process.Preprintofpaperpresented..a%SAEAnnualMeeting .—

(Detroit),Jan.12-16,1953..,-_. =—-#-

15.Long,RogerA.,andEsgar,JackB.: ExperimentalInvestigationof ., ~Air-CooledTurbineBladesinTurbojetEngine.VII- Rotor-Blade. =- –

-.--—

FabricationProcedures.NACARME51-E23,1951. —

●

✎

�

.

—

.

*

.

NACARM E53E18

TABLEI. - COMPOSITIONOFACID-NICKELSOLUTIONSUSED

FORNICKELCOATINGTURBINEBLADESa

Solution

I II

Nickelchloride,NiC12”6H20,g/liter 30 --

Nickelsulfate,NiS04S7H20,g/liter -- 30

Sodiumhypophosphite,NaH2P02-H20,g/liter 10 10

SoMum hydroxyacetate,NaC2H303,g/liter 10 --

SOdiUmacetate,NaC2H302”3H20,g/liter -- 10

PH 4t06 4t06

/Rateof deposition,in.hr o●0005 0.001

aInformationfromtable3 ofreference11. v

23

.

.

.1—8

7

b8

%

‘lb’

’11

12

Is

14

15

——

ml. of mating

;mmmic, mOdi-%s.3 m A-19

:wamio, modi-lied 9E3 A-19

:-10, 301ara-ti 80+3F2

Hlokel

Iuokel

Imkel

: l:!diokel

NiOkBl

Ninkel

N&el

hmcatod forFl hr. thenrdckcl SQmw

mom

,,i,:dJ.<,.

SAE 4130 COmt

SA6 4uKl A

BAB 4130 A

2AB 4130 A

IYakEm A17-22A(S)

lwke?l A17-=(s)

SA2 4134 A\ !+, ,.,,

SA6 413# :A

31B 4130 A

sA2u3a B

M2 4& B

263 ii301 B’

-L~iB“17-22A(S]

TA23X 11 - 6THARY OF FA2R1CAT1CHAHD OF=U~ D3TAES

Batreat-mlttable11)—

lule

im

5“

1

1

1,

6

6

2

1-+

2

.2

:2

‘2

‘7

T.

Cyoles

m

107

59

“’ 2

46

or pstIitlmm

he-mat-cbr

2.3

4.7

11.2

X5.9

100.0

44.3.

8.0

!1

24,0

2::;

S.tl1.0

21.O’5.0W.S

23.91

—,01-mlmBCio

—.046

.W6

.046

.048

.c48

.04s

.046

.03

j

.0s,,

.044

.038

.030

.(24B

.030

Sx&

.WQ

,.

.ik

—

Cmditim or ooatim namll?a

.temlve chipping on leadingendtrailingedges and mmtlon .urfaoem

ltirecc.stingnmti tam’d tip;I@.Otalexpom.1 et leadhg and trail-ing edgeszoellentq 6 hr tipotim

zoellmt at lant”ine++otim (6.2 br)

IBlade W.kd at 1/3 awn m rmult

of la!. of cOolill&alrsupply

T&:;lade, ftiled in fatig,m

.ellmt th-cughcut; miginnl cark- 3T088ure surfacebowlm cut andlngn m ooatlng still evident,no @lins 10M. frm tibee at BaI-mrrmim et lemding edge where Clunial of testdmnmed bY debrin

mting Oh.lpmd off 1ead43 edge emd Premire mrfaoe b.nfe.lmt MCIWm of tralng edge, gti oca-ditica .lmMere

Pilled away frm tube,

004 =ed by fa.iiureof anotbm blade

w.tiw erded awaym lending end CutOr 1/S of premum surfaoeia-ailbgedges and .mmmim set failed becml,eor qu.rmlal, pcaInj ommmlm etidmt at leading ttmd, am VimlmedKe nfixr 76 oynlea

‘M~&~ti~eyw *m leading and Bltie Still servioeabl.

mt41t3chimed Orr lc.wllngC&0 St El* w= thmz100w16d fm -25 hr and a made f-d, blade me ~_vr: ~tiwtutlm tantareccated;009.tinsompletely gonsfrm leadiw edge mm other lmal— at ommletim of tent

,,005

j,measure sbrfaoe mlhd away. firm ~Lame.

mcatad bwe ccrrbdadbadly in ;

I

Portich of blade broken may frm= hr, mukwcnnt!ocatlwblis- pmmvre inn-faceat tiptered In 10 *j at 51.2 br Ccmtlilgbad maled Off iilS- _UOJ 1=6d-iu e6sa man b- ma had LIorrodedbadly /

,,. “q, . ~, I ; ..! .,’, .’? :im.,.w

.18

17

18

19

20

a

22

2s

24

Z5

2s

27

28

e9

Uiok,l MitJlUiorobraxonleading *.3

Nlokel witiNiorobr= MIleading edga

Hiobl WithIIiorob= onlaad.blgedge

liiokmlwithUicmbrsdm1- edw

Ilicrobraz

NiorobraE

Alummixbd(pank WIMem )

N.umlD.ised(Pank worms)

tlumltind[dipPrcloem)

illminixad{alp ImOesrl)

N.-IA(dipprcuem)

Numinieti(dip prmosa)

B

B

B

B

B

B

B

B

A

A

B

B

B

B

7

7

7

7

7

7

7

7

1

1

b

&

L

k

—

100.0l.e

11.4

23.7

24.11.5

1L13.O

lW. O1.2

?0.01.5

5.9

100.0

8.0

124.4

124,4

1.lfl.7

lW.6

(X-4, .

.048 00cd

.04S clodexmpt fctrtiokel moaling offleadlng 0.3.s0mm? FOOt

.042 tiding cdze very gcmdi Blight

.03 flaMJIs at tidahord mm. tlp mbotb Bllrrmea

.046 v- s-l alkzht flakingwhareleading * waa bit end mtralliw *e

.C4# KxOellent●t aJJ Btagan 0? opratim

.03

.04a WRelleut

.0s

,042 ~ellent

.048 Very WOO e.sneptwham hit by dmbris

.04.9 Verysoml

.03

[

dlant OxrMpt formgs nearrti~~r;~tltingfrm qplimtian

.03 2x0e11ent .xaapt far dnuaz. MAVrmb mnvltiilgrrm mlicaticuof fillet .mxtfm tip damffm

.0$ Exmllont BXOapt rw! We ml..~~~tlm frm Bpplimtdon

,9s m.11.nt exwpt whmu tit by debrimmad where wmnd away at tip

2880 ,,

~ W failureofblade 21

FatiW failureat blade mot

Fatigue failure at blm+e ,oot

~ bY failure of blalm 22

y: #mludOa Ulxrli-tubing

Fatimefailure aV blade root

Fatir.m ?ailum at bleda Fcmt

Fatigue rai.lureat bmde root

ml% &lOludOd upon maoblng

Bhell pulled cut of base

ltipdamsMI by ntmdard bladefailu@n at 5.1 IWJblmle notbaak ti 3.92 in.

TIP .iMM@d b? mtmdm?d bladefailurn lb 24 Ill.;L!lde nutbaok to S.S6 b,

lm dmasedbY utandardbladefailure ab 17.2 hrj blade cutbaok to 2.66 in., oraakn atbane Mt+? 118.7 hr

Blu7e mtr.&ah6dud robbed-t bail .-art.mlhrlout bnokta s.2m.

.%01. ncmisted of 15 mimten at ratad tsat omditima follmmd by S m.tmtaa at idling omditimo.%evimmly reported in rera.mce 9.

i,

26 NACARM E53E18

.

TABLEIII. - ~T mTMENT8 APPLIEDTOA~-COOIXDTUR8~ BLADES– — “ –

[Allnormalizingwasdoneinan inertatmosphere.]

——

.—.—

2

7

~eat-treatment Heat-treatmentprocedurenumber

SAE4130steelblades

1 Normalizeat1600°F for30-rninutes;aircooltoroomtemperature.Drawat100Q”F for15minutes;aircooltoroomtemperature..1 ..=.—

Heatin1800°F saltbathfor15minutesfouc)wedby ““

isothermalquenchin1000°F saltbath;holdfor15minutes.Aircooltoroomtemperature.

3 He%tin2000°F saltbathfw”15minutesfollowedby“-isothermalquenchin1000?_._Fsaltbath;holdfor15minutes.Aircooltoroomtemperature.

Timken17-22A(S)steelblades4 Normalizeat1725°F for30%iinutes;aircooltoroom .

temperature.Drawat1225°F for6 hours;aircooltoroomtemperature. — —

5 Normalizeat 1750°F for30‘tinutes,cooltoroomtemperatureinnitrogenblast.Drawat1225°F for4 hours;coolinnitrogenblast.

6 Heatin1800°F saltbathfor15minutesfol.lowedbyisothermalquenchin1200°T saltbath;holdfor15minutes.Aircooltor~omtemperature.

Normalizeat 2075°F for15%inutesfollowedbycooli~-inhydrogenatmos~hereatrateequivalenttoaircool-.Drawat1225°F for4 hoursfollowedby aircoolto ‘-roomtemperature.

.

—.

.-

. ..— ._—

.

.

NACARM E53E18 ---mr 27

.

.

0coaN

(a)Formed-shellbladetithprofilel!..

.

(b)Fonued-shellbladewith~ofileBi

.. ....- .--— -—

-—.-— . . —. — ~———------ ..-. . . . . . =!5=.

.

(c)Cast-shellbladewithprofileC.

Figure1.-Crosssect~ona& experimentalblades.

C-29180

28

1600

1400

8 1200U“

s~

g 1000

800

600

NACARM E53E18

1

Profile

—.-.

Effeotivegastemperature

\1,<A\ \ /\ 4 /

tmlPressuresurface

Tttll=d=Suctionsurface

.-

.-

.

-,.

N-(nco0

.

.

— -—

0 20 40 60 80 100 80 60 40 20 0Leading Trailing Leadingedge- edge edge

Bladechord,percent

(a)Coolant-flowratio,0.03.

Figure2. -Peripheraltemperaturedistributionatsection2%inohesfromtipforbladesofprofilesA andB.

—.-.

●

NACARM E53E18

.

29

1601

140(

.

80(

60~

I 1 IYrofile

——. — AB

I I

Effectivegastemperature

#

- \/

\~ \ I{!\

\-~

\ \ /

Pressuresurface Suctionsurfaoe

o 20Leadingedge

Figure2.

seotion

40 60 80 100 80 60 40 20 0!&ailing Letiingedge edge

Bladechord,percent

(b)Coolant-flowratio,0.05.

- concluded.%%ripheraltemperaturedistributionat

2% inohesfromtipforbladesofprofilesA andB.

(8) Blade lafter7.5houre (20.7GlespluE Z.3 houreat rated teat oo~t,l~) with cools.nt-fm

ratio of 0.048. ccmpl.de b~ of czmt~ along leadlrg and treilllng edgee evident.

F@re 3. - T&ical cerardo-cimt~ ~fallweB.

i! t.

● 0!382 ● x,.11,1 ,,!,. ,, ,11 ,’tll,l’ .’ 1,,, I 1:1 11, il :1, IJI II.

NACARM E53E18

.

.

“

31

(b) Blade2 after 4.8hours at rated teet conditions with coolant-flowratioof0.048.CoatingsoftenedEndflowedtowardtip.

Figure3.- Concluded..Typioaloeramic-ooat~failures.

,,I

-dOd sreaB ZIU3to fordm ~Lides

l--from ,.9*,UN of =.otLer bled? -,

~:li,!

\l:l

i.

1,,

,,

c-32339

(.) Bla& 5 after 34.3 hours at rated teet comiiti.ms with ooelfmt-flw ratio of 0.046. CMt* h exaellcmt LxuMlticm [

except where &m+@ by foreign ob &cts atiild.ng blade.‘+

Figure 4. - C.gi.emit-otxited bliitle5.w

!,I

● I. 0882 ● L

,

‘1‘*

, ,

,.. ,

Yr

(b) Blade 5 d’ter 100 houre at rated test w

~ti *TO hit by fOrCJl@Iobjeota.

Figure 4

‘ s (2X-5 088?. * ,

c-szis40

Ilaltlons Ultb CaOlent-flw mtio C& 0.048. Coating In exoellent ocmdltion

. - Concluded. Ceramic-coated blade 5.

NACARM E53E18

.

●

&%?&. .—..-’:1—-----

)Corrodedarea~ ...

~,. ,,

7“-----”””:”“-”--’-“’

.

=!Y=-C-32341

(a) Blade12after ”55.3 hours at rated test oondi.tlonstith coolant-fluw ratio of .

0.03to 0.048.Coatingcompletel? ercdedawayet-Iead~edgeandblisteredandflakedh soatteredareaeoverentireblade.‘-

.=-. ——.=

Figure5.-Typloalnlokel-ooatlngfailures. .—

~CA RM E53E18

&

&

.

.

—--===”*,a. .

Corrodedarea

—. --.–-——

35

=?3=’C-32342

(b) Blade6after44hoursatratedtea%condltlcmswithcoolant-flowratioof0.048.Coat-chippedanderodedonleadingedgeanierodedalongtrailingedge.Lead*edgeoorrodeaneartIp.

Figure5.-Conoluded.Typicalnickel-oati~failures.

WOY

,\

\,.m~—mixohrizd

!.;. ,.,, !’

—. .,. . ,,, ,

!

I Ptiure6. - Nlokal-caated blade 20 with Nicrobraz along leadbg edge.dth Oooknt-flw ratio of 0.048.

Blade after 102 hours at rated teat ocmd.ltiansleadlng edge end midohcad raglcm h excellent omxl,ition.

.

I 1 1’,1 *

0813Z w *,,,l. .,, ..lb I 1,, ,,. 1,,,, ,. ,1,

, ,

Suetlonsurface

,!

I

,

2B80 * ,

!2

------Pressuresurface C-32344

(a) Blade fir 50 houm at rnted teat oond.itionswith coolant-flow mtio IX’0.048. BlaAe 8UJ?fa08B h ~l~llt

c.cudltlm except where hit by fomigu Furtlclm?.

I’igure 7. - Aleminked blmle 24.tn-4

(2

aElE-

(b) Blade Wter 100 hours at rated test adltiorm with Uc&lant-fl.w mtio cK 0.048. SwSaoes h ExcOumlt 00Iditial.~Vi0U81Y &amM8d amaB ehmm cu fi8w13 7(a) Idistiwuishablm fmm adJo~ S@a=.

F@&e ‘7.- Conclded. Alknlzed blade 24.

i,t,/ * ,! , .

i: 1,4, II I, ,i,, illi, d,!, lll ,, ,,, 1, 11,;’

own ‘ ‘II ,,

Cnal

*

2s80 , ,

,,

~Coatlng damagedduring applioatlonoffil.lets~---

,,, :,

,, ,-,.I ., “,”lRllum :’=’,., .——.

1, 1 Suction surface ., . . .,.,

(a) Bbde8durlw ear~stages ofemlumnoe rmmlngviti .oolamt-flcmmtloae 0.03. Timsonbladm 261Mu3 .27,23.8 bourB;on blmie 28, 16.1 homa; on blda 29, m time. Surfaoe.9h emellent oOz@lticm.

l?l@re 8. - Alumlnizd bladm 26, 27, 26, and 29.

Suctionsurfacs

c-32347~r Preamre surfaoe= “(i) Blxdes at Cmclusicm 02 enawanc e ~ vith a ODOLmt-fkw mtio d 0.02. Tim m blmkm 26 and 27, 124.4 hOW4;

ii m blade 28, 118.7 hours; ~ b~e 29, KIO.6 ho~. S@aoes ~ e~ell~t c~~tlon.I

Flgwe 8. - CoMl@ea. Alumhized blades 26, 27, 28, 81xI29.

r L d .

0882 ‘I II i!, ;1 illl” i II ill, Ill II II!