*---

T,.

FORAERONAUTICS

TECHNICAL NOTE 4130

NACA65-SERIESCOMPRESSOR

J3N3YULUS-AREARATIO,

ROTORPERFORMANCEWITHVARYING

SOLIDITY,BLADEANGLE, AND

REYNOLDSNUMBERANDCOMPARISON

WITHCASCADERESULTS

ByWallaceM. Schulze,JohnR. Erwin,andGeorgeC. Ashby,Jr.

LangleyAeronauticalLaboratoryLangleyField, Va.

Washington

October1957

TECHLIBRARYKAFB,NM

h

4

NATIONALADVISORYCOMMJTI’EEFORAERONAUTIC Iilllllll[llllllrlllllfll:l[llllllll00ibl&2

TECHNICALNOTEk130

mcA65-SERIESCOmpreSSOrROTORmmoRt4mcEwIm vmYmG

AN_NLJLUS-AREARATIO,SOLIDITY,BLADEANGLE,AND

REYNOLDSNUMBERANDCOMPARISON

W23HCASCADERESUId

ByWallaceM.Schulze,JohnR. Erwinj=d GeorgeC.Ashby,Jr.

SUMMARY

A typicalaxial-flowcompressorrotorusingNACA65-seriescompres-sorbladeswastestedat lowspeedsanditsperformagcewasmeasuredovera rangeofquantityflowratesat severalvaluesofannul-us-arearatiojblade-settingangle,solidity,andReynoldsnunberto comparewithporous-wti cascaderesults.Thedataobtainedwiththeannulusareavariedwerecorrectedtothetwo-dimensional-flowconditionby twomethods.l?romtheresultsofthisstudy,theconclusionwasreachedthattwo-dimensional-flowporous-wallcascaderesultscanbe usedto estimaterotorperformancewithgoodaccuracyovera widerangeofconditions.Themean-sxial- ,velocitymethodofconvertingtherotordatato two-dimensional-flowcon-ditionsgavegoodagreementwithcascadedataforaxial-velocitychangesacrosstherotoras largeas15percent.TherotorperformancechangedonlyslightlyastheReynoldsnumberwasdecreasedfrom500,000to 2~,000.As theReynoldsnumberwasdecreasedbelow250,000,decreasesinrotorefficiency,pressure-risecoefficient,andturninganglewereobserved.

INTRODUCTION

Theperformanceofaxial-flowcompressorbladescanbe quicklyandaccurate~measuredindetailby usingstationarymodelsin two-dimensional-flowcascadewindtunnels.Thecascadetunnelcanthusbe a veryusefulinstrumentforprovidinginformationneededinthedesignofaxial-flowcompressors.Questionsoftenariseas towhethertwo-dimensional-flowcascadedatacanbe applieddirectlyto compressorsandwhatcorrections,if any,mustbe made. Intheinvestigationreportedinreference1,

l-SupersedesdeclassifiedNACAResearchMemorandumL52H7byWallaceM. Schulze,JohnR. R’Win,andGeorgeC.Ashby,Jr.

2 NACATN4130

rotor-bladesurfacepressuredistributionsandair-turning-anglevalueswerefoundtobe similsrtothosemeasuredinporous-wallcascadetestsatdesignangleof attack.Thepresentinvestigationwasdevisedtoprovideinformationconcerningtheeffectonrotorefficiency,static-pressureandtotal-pressurerise,andturningangleof changesinbladesingle,solidity,flowrate,Reynoldsnumber,andannulusareathroughtherotor.Theperformmceoftherotorasestimatedfromcascadedatawascalculatedandispresentedforcomparison.

A

An axial-flowcompressorrotorhavingbladesofcsmiber,solidity,andhub-tipradiusratiotypicalofa centrallylocatedrotorina multi-stagecompressorwasinvestigatedatlowspeedina 28-inchtestcompres-sorwithoutguidevsaesor stators.Surveysoftheflowmadeimmediatelyupstreamsmddownstreamoftherotorwereusedincalculatingtheperfor-manceforcomparisonwithvaluesestimatedf’mmporous-wall-cascadetest .: _results. —

SYMBOLS

A

cd

cl

L/D

D

E

I

M

n

Ps

PT

Q

~

annulussrea,sq ft

sectiondragcoefficient

sectionliftcoefficient

lift-dragratio

dismeter,ft

energyaddedtoairastotalpressure,

workdoneon airby rotor,ft-lb/sec

massflow,slugs/see

rotorspeed,rps

staticpressure,lb/ft2

totalpressure,lb/ft2

quantityflowofair,ft3/sec

dynsmicpressure,lb/ft2 ‘

*.-

k’

—

-—

—

ft-lb/sec

—

d

—

●

❉

NACATN4130 3

R Reynoldsnumberbasedonbladechordlength,enteringvelocityandstsrderdstagnationdensi@andviscosity

r radius,ft

u rotorbladevelocity,ft/secexpressedas a fractionof

(exceptInfig.1 where U isthebladetipvelocity)

w

CL

P

5

v airspeedrespectiveto stationarycasing,ft/sec(exceptinfig.lwhere V is expressedas a fractionofthebladetipvelocity)

airspeedrespectivetorotor,ft/sec(exceptinfig.1 whereW ise~ressedas a fractionofthebladetipvelocity)

angleofattackrelativetobladechord,deg

inletandoutletairanglerelativetohl-ades,degfromaxis

ratiooftangentialvelocitychsagethroughtherotortoenteringaxialvelocity

ratioof specificheats

blade-anglesettingrespectivetorotoraxis,deg

adiabaticefficiency,percent

airturningangle,deg

airdensity,slugs/ft3

solidity,bladechorddividedbybladegap

airsinglein stationarycoordinates,degfromaxis

Qquantitycoefficient,—nDt3

%2 - PSIstatic-pressure-risecoefficient,

1U22P t

‘T2- P~ltotal-pressure-risecoefficient,

*putp

4

Subscripts:

NACATN 4130

1 upstreamofbladerow

2 downstreamofbladerow

a axialdirection

d designcondition

e valuebasedonvectordiagramcorrectedto

f value.basedonvectordiagramcorrectedtovelocity

meanaxialvelocity

enteringaxial

h at

P at

t at

hubsection,D/Dt= 0.784

pitchsection,D/Dt= 0.892

tipsection,D/Dt= 1.000“

ten tangentialcomponent

ch settlingchamber

APPARATUSANDTESTS

Apparatus.-A schematicdiagramofthetestcompressorispresentedinfigure2. Theflowentersfromtheatmospherethroughthreescreensintothesettlingchsmber.An entrsnceconehavinga contractionratioof 13:1isusedto acceleratetheflowintothetestsection.Therotordischargesthroughanamulerdiffuserequivalentto 6° conicalexpansion.At theendoftheannulardiffuser,theflowisturnedoutwardthrougharadialdiffuserwhichcanbe adjustedto decreaseorincreasetheexitareaandthusregulatetheflowrate. Thedriveisa 7~-horsepowerdirect-currentmotoroperablefromO to 3,600rpm.

Theemnulartestsectionhasan inner-casingdiameterof21.82inchesandan outer-casingdisneterof27.82inches;thehub-tipradiusratiois0.784.Thetestsweremadeonthe 5 = 0.6 bladesoriginallyreportedinreference2. Porous-wallcascadedesigndatafromreference3

t’

A.

t

NACATN 4130 5

indicatedthatthedesignconditionsshouldbe changedslightlyfromtheoriginalvaluesofreference2, seetableI. ThedesignisfreevortexusingNACA65-seriesairfoilsectionscamberedforisolatedairfoilliftcoefficientsof1.2,1.1,and1.0atthe30-percent(inboardsection),50-percent(pitchsection),snd70-percent(outbosrdsection)annulusheightpositions,respectively.Thebladeshavea constantchordandheightof3 inches.Thesoliditywasvsriedbychangingthenumberofblades.Forthepitch-sectionsolidifiesof 1.0and0.5,thenuniberofbladesusedwas26and13,respectively.Theaveragetipclesrancewasapproximately0.015inchor1/2percentofbladeheight.A rotatinginnercasingextending5 inchesbeyondthebladeswasattachedtotherotorto supporttherotor-mountedinstrumentsandthebalsafairingsusedtovsxytheannulussrea(fig.3). Allthreeannulus-sreachangesweremadeby alteringtheinner-casingdimneterfora shortdistancealongthetestsectionasshowninfigure4.

Thestationaryflow-surveyinginstrumentsusedwereof thetypeshowninfigure5. Oneinstrumentwasplaced1* inchesupstreaofthe

L inchesdownstreamforallrunsrotorforallrunsandtheother,lZ

. exceptthoseusingrotor-mountedinstruments.Forthesetests,thedownstreaminstrumentwasplaced~ inchesdownstreamtopreventinter-

* ferencewiththerotor-mountedrakeandprobe,figures6 and7, whichweremounted3 inchesdownstreamofthebladesonoppositesidesoftherotorspindle.Thedesigndetailssndcalibrationsoftherotor-mountedprobesrepresentedonfigures32 and33ofreference4. Therotor-mountedinstrumentsweredesignedtodeflectlessthan0.005inchduetocentrifugalforces.Thethreadedsleeves,solderedto thestreamlineshafts,wereusedformounting,angularsetting,andradialadjustmentoftheprobesatthethreepositionsused.

A sealed-ball-bearingtypepressure-trsnsferdevice,reference5,wasmountedwithinthetestblowertotransferreadingsfromtherotor-mountedinstrumentsto stationaryleadsthatwerepassedthroughthesideoftheannulsrdiffuserandconnectedtothemanometerboard.Formosttests,a verticslmultitubealcoholmsnometerwasusedto indicatethepressurereadings,However,atthelowspeedsusedtoobtaintheper-formsmceatlowReynoldsnumbers,a calibratedmanometersetat amaleof84.25°fromvertical(givinga’magnificationfactoroften)wasusedtoreadtheverylowpressuredifferences.

Testingmethods.-Whenthetestcompressorwasassembledforeachconditioncarewastakentokeepinternalsurfacesevenlyfaired,clean,andfreeof surfaceroughness.Therotorwasrunup tothetestspeed,usually2400rpm,andheldwithin% rpmduringthetest. Surveysup-streamanddownstreamoftherotorweremadesimultaneously.Theinstru-mentswerelocatedatdifferentcircumferentialpositionstopreventinterference.S&teensurveypositionsspacedto indicatethecomplete

6 NACATN4130

flowpatternfrominnerto outercasingwerenormallyused. Static-pressure,total-pressure,andyawreadingsweretakenateachsurveyposition.

Therotor-mountedinstrumentsweresetatthedesignoutletangleforeachradialpositionatwhichtheywereplacedandtestsovera rangeofflowrateswererunwithoutfurtheradjustment.Althoughtheoutletanglevariedonly+3°,no significanterrorswereintroduced.Theindi-catedoutletflowanglefromthesetestswasdeterminedby usinga yaw-calibrati.oncurve.Inasmuchasyaw-calibrationtestsindicatedlessthanl/4-percentvariationastheprobewasyawed3°,thetotal-pressureandstatic-yessurereadingswereonlycorrectedforinherentprobeerrorsat0°yaw(fig.33ofref.4). —

Testprogram.-Sixcotiigurationsweretestedwithconstantannulus

area: 10 abovethforthedesignsetting(seetableI),for72 e design10belw thedesignsettingatpitch-sectionsolidifiessetting,andfor72

of1.00and0.50.Thequantity-flowcoefficients,atdesignanglesof1° bel~ design1°abovedesign,and7~attack,forthedesi~ setting,72

are0.640,0.476,&d 0.830,respectively.Theconfigurationwitha blsd,e-anglesetting7~0abovedesignat a pitchsolidityof1.00wastestedwith

2ratiosofrotor-exitannulusmea torotor-entranceannulussreaof 1.15,0.85,sndO.70.Alltestsfortheseconditionsweremadeatrotorrota-tionalspeedof’2,400rpm. TheReynoldsnumbereffecttestsweremadewith

A

—

e

theconfigurationhavinga constantannulusarea,a bladeangleat7~0abovethedesignsetting,anda solidityof1.00atvariousrotorspeedsfrom400to 2,4oOrpm. Therotor-mounted-instr~enttestswerealsomadewiththislatterconfiguration,therotor-mountedprobeandrakebeinglocatedalternatelyattheinboard,pitch,andoutboardsectionscorre-spondingto the30-,50-,snd70-percent-annulus-heightpositions,respec-tively.Thesetestsweremadeata rotorspeedof2,000rpmtoreducethestressesontherotor-mountedinslmxnentswithoutsignificantlyreducingtheReynoldsnumiber.Testsweremadeatnumerousflowratesfromthemaximumvalueobtainabletoa conditionnearstall.exceptfortheReynoldsnumbereffecttests,whenonlyvaluesneardesignwereused.

Precisionofresults.-Flowinstrumentswerecalibratedinan8-inchcalibrationtunnel.Thestatic-pressure-ctibrationfactorsusedgaveresultscorrecttowithin1/4percentofthedynamicpressure.Theyaw

A

nullpointsweredetermined;theinstrumentsandholdersweremounted;andreadingsweretakenwitha precisionbelievedtoprovidemeasuredflow ?angleswithin1/4°oftheactualvalues.Thetachometerwascheckedwitha stroboscopeatlinefrequencyandfoundtobe accuratewithin5 rpmor1/5percentfornormaltestingspeeds.

NACATN4130 7

% Themanometerwasreadtothenearest0.01inchofalcohol,whichcorrespondsto approximately0.1percentofthedynamicpressureat aver-agetestconditions.

4ForthelowReynoldsnumbertests,theinclined

msmometerboardpermittedreadingsto thenearest0.001inchofalcohol,orabout1 percentofthe@nsmicpressureat thelowestspeedandflowrateused.

Thedatawereprocessedby anautomaticcomputingmachine.Thevariousintegrationsrequiredwerealsoperformedby themachineusingarithmeticaveraging.Computing-machineintegrationscheckedcontinuousintegrationswithin1 percent.

Estimatesofrotorperformancewerebasedonporous-wallcascadedata.Cascadetestresultswerecorrectedbyinterpolationandextrapolationoftheavailabledatato theconditionsforwhichtheywerecompared.

Themeasuredmass-flowerrorsfortheconstant-areacondition,fig-ure8, showthatthemaximumerroris lessthan3 percentsmdtheaverageerrorisapproximately1 percent.Forthevarying-areaconditions,fig-ure9,themaximumerroris lessthan5 percentandtheaverageerrorislessthan3 percent.On thebasisofthischeckandtheothertestingaccuraciesnoted,it isbelievedthatthefairedcurvesindicatetruevalueswithin2 percentfortheefficiencyandpressure-risecoefficients,andwithin1/4°forthedeflectionanglesatallconditionsexceptnearstall.Thisaccuracyisfurtherverifiedby theclosechecksobtainedwhenrerunsweremadeduringtherotor-mountedinstrumenttests.

RESULTSANDDISCUSSION

TestsWithVaryingSolidityandBladeAngle

Efficiencyandpressure-risecoefficients.-Theperformanceresultsoftherotoratthepitchsollditiesof 1.00and0.s0forbladeugles at

1°belowdesign1°abovedesign}thedesign,72 and7= arepresentedjointlyto simplifycomparisons.Theadiabaticefficienciesmeasuredatthesesixconfigurationsareshowninfigure10. Theftiredcumes indicaterelativelyhighvaluesatallbutextremeflowrates.A peakvalueof98percentis indicatedatthedesignconfiguration.Whenthesoliditywasreducedto0.5,thepeakefficiencyatdesignwas95percent.Theefficiencyishighestatthelowbladeangles.Thetotal-andstatic-pressure-ri,secoefficientsalongwiththoseestimatedfromcascadeturning-singledataforthetwosolJ.ditiesatthedesignbladesinglemeshowninfigureIl. Theseestimatedcurvesassumeno losses,someasuredvalueswouldnormallybe lower.However,laterfiguresshowthattheturninganglesproducedby therotorwerea littlehigherthantheesti-matedvalues,sothepressure-risecoefficientsshouldbe skl@Klyhigherthanestimated.Verycloseagreementis observed.

8 NACATN 4130

An attemptwasmadeto estimatetheefficiencyby usingcascadeL/Dvaluesanda proceduresuggestedinreference6. Reference3 presentsthemethodofcalculatingthecascadec1 sad cd valueswed fortheseefficiencyestimations.Theefficiencycalculation,brieflydescribedintheappendix,includesonlybladeprofile.lossesandneglectscasingboundaxy-layereffects.Theestimatedefficiencycurvefortheconfigu-

1°abovedesignrationwithsolidityof1.00anda bladeangle72 is com-psredwiththemeasuredvaluesinfigureX2 andisshowntobe lower.Efficiencieswereestimatedbythisssmemethodusingthe L/D valuescalculatedfromthedatameasuredwiththerotor-mountedinstrumentsandareseentobe higher.ItisevidentthatthecascadeL/D values meconservative.Theyareconsiderablylowerthenrotormeasuredvsluesandmorethanoffsetthecasinglossesneglected.Thisconditionmayoccurbecausethecascadevalueswereobtainedata Reynoldsnumberofabout250,000,whereastheReynoldsnumberoftherotortestswasabout~0,000.At theselowerReynoldsnumberslsminarseparationincreasesthedragval-ues,and;hence,decreaaesthe L/D ratios.Infigure13is a-comparisonofbladewakeprofileswhichindicatesthatcascadedragvalueswouldbemorelikethoseoftherotorif cascadedatatakenata Reynoldsnumbernear500,000wereused.A lackof systematiccascadedatapreventedefficiencyestimationsbasedon L/D ratiosata Reynoldsnuniberof500,000.ItIsbelievedthattheestimatedcurveusingrotormeasuredprofilelosseswouldcloselyagreewiththeoverallmeasuredefficienciesifthecasinglosseswereincluded.Thepresenceofthesecasinglossesis shownby thecurveinfigure14,inwhichblade-elementefficienciescalculatedforeachsurveypointareplott”edfora typicaltestneardesignquantityflow.

&

—

. -—

.

.

.—..

Turningangle.- Figure15presentsthemeasuredflowturninganglesattheinboard,pitch,andoutboardsectionsfortwosolidifiesas com-paredwithvalues estimatedfromcascadetestsforthesesixconfigura-tions.In allinstances,therotorresultsarehigherby 1°to l~”. Sincethisindicateddifferencewassoconsistent,thepossibilityofmeasuringerrorsduetotheeffectofthewakesuponthestationaryinstrumentwasinvestigated.Theflowdownstreamofa rotatingbladerowisdiscussedinreference1. Hgwever,nomethodofcalculatingthiseffectresultedin correctionslargerthan0.2°or0.3°assumingnormalwskeprofiles,soitwasdecidedtomeasuretheoutletsinglesdirectlyfora givencon-figurationwithrotor-mountedinstruments.Figure16 showstheresultsofthisinvestigation.Thedirectlymeasuredturninganglescompareverycloselywiththevaluescomputedfromdataobtainedby thestationary A“instrumentsattheoutboardstationandvarylessthana degreeatthepitchandinboardstationsupto anangleofattackof16°. Thus,the r10~~er th~ eactualdeflectionsare1°to15 stimatedvsluesandreason-ablytruereadingswereobtainedwiththestationaryinstrument.Anexplanationofthedifferencesissuggestedby thefactthattherotor

NACATN4130 9

dragcoefficientsarelowerthanthecascadedragcoefficientsas showninfigure16. Sincetherotorwakesweresmaller,themainflowmorecloselyfollowsthetrailing-edgebladecontourandconsequentlyexperi-encesa higherturningmgle. Thewakesinthecentralportionofthebladescaneasilybesmaller,forunlikethecascade,therotor-bladeboundarylayerscanflowtowardtheinneroroutercasingalongthebladesurfacespropelledby eithercentrifugalforceon theboundsry-lsyerpsrticlesorthestatic-pressuregradientinthemainfield,which-everpredominates.Reference1 discussesthisactionin greaterdetail.Inaddition,figure13 indicatesthatifthecascsiiedatahadbeenattheReynoldsnumberof therotortests,500,000,insteadof 250,000,thedragvaluesand,hence,turning-snglevalueswouldhaveagreedmorecloselywiththerotorresults.

In figure17,thevariationofturninganglewithairinletangleat constantsngleof attackispresentedto supplementcascadedatawhereinterpolationbetweeninletsinglesisrequired.Estimatedcurvesareagainincludedforcomparisonsmditcanbe seenthattheyfollowthessmetrendsas intherotorbutat1° to l~”lowerturninganglesasbefore.

Inorderto illustratetypicaldistributionsof flowsinglesandpressure-risecoefficientsacrosstheannulusfrominnerto outercasing,figures18and19wereprepared.Bothmeasuredandestimatedvsluesatflowratesneardesignforthedesign-blade-angleconditionoperatingat so~ditiesof1.0and0.5arepresented.Measuredturninganglesthatarelargerthsmtheestimatedvaluesandtheresultingeffectsontheoutletangleandpressure-risecoefficientsme againevidentatsllpointsfreeof thecasingboundarylayers.

Exitaxial.velocities.-Theoperationof thisfree-vortexrotoratoff-designconditionsresultedina radialvsriationofexitaxialveloc-ities,exclusiveoftheboundarylayers,atthevsrioussections.Inreference7, a methodof estimatingtheaxialvelocitiesispresented.Intheuseof themethod,it isnecesssryto estimatetheoutletflowangleexpectedsothata finalresultcanbe obtained.TwoSysm ofestimatingtheseangleswereusedandthecalculationsmadeforfiveofthetestsata sol.idi~of 1.0ateachofthethreedifferentblade-anglesettings.Thefirstmethodmakesuseoftheturninganglesestimatedfromcascadetestsindeterminingoutletanglesandthesecond,Constant’srule,reference8,whichforthepresentinvestigationwasinterpretedtomeanthattheoutletangless.reconstantatthedesignvaluesregard-lessof inletconditions(de/da=1). Figure20indicatesthediffer-encein de/da valuesobtainedfromtestcompressorresults(averageof thevaluesattheinbosrd,pitch,andoutbosrdsections),cascaderesults(formedium-csmber65-seriesairfoilsat conditionssimilsrtothoseinthetestcompressor),andtheinterpretationof Constanttsrule.Althoughsomepointchecksarenotexact,thetrendsofthetest-compressor

10 NACATN4130

andcascaderesultsaresimilarandtheagreementbetweenthetwoisconsideredgood.Figure21presentscomparisonsofmeasuredandesti-matedoutletaxial-velocities.Theoperatingconditionsevidentlyarenearenoughtodesignsothattheaxialvelocitiesdonotdepsrtfarfromconstantvalues.However,thetrendsobservedsrein goodagree-mentwiththecalculationswheneitheroutlet-angle-estimationsystemis usedinthecalculations.

. TestsWithVsryingAnnulusArea

4

In ordertodeterminetheeffectsofvqing theaxialvelocitythroughcompressorbladerows,oneblade-settingconditionwastestedwithfairingsattachedto therotorhubto increaseordecreasethesxialvelocitythroughtheblades.Theconditionofa sokhiityof1.0anda blade

1°ab&e designsingle7Z wasselectedforstudy.Someoftheresultspre-viouslydiscussedforthisconditionwithconstantannulusareasreincludedinseveralofthefollowingfigureswhichshowtherotorperfor-mancewithnonconstantannulusmea forconvenienceinmakingcomparisons.Figures22and23presenttheefficienciesandpressure-risecoefficientsmeasuredatratiosof rotor-exitannulusareato rotor-entrsmceannulusareaof1.1.5,0.85,“and0..70as comparedwiththeconstant-arearesults.For Az/Al= 1.0,thepeakefficiencyis about97.5percentas comparedwith93,96.5,and97.5percentfor A2/Al=,.1.15,0.85,and0.70,respectively.Withinthelimitsofmeasuringaccuracy,theefficienciesincreaseasthestatic-pressure-risedecreases.Thetotal-pressure-risecoefficientsfortheseveralsrearatios,whenplottedagainstflowcoef-ficient,figure23,areseentodifferconsiderably,particularlyforA2/A~=0.70.At a qyantitycoefficientof0.52,thetotal-pressure-risecoefficientsat A2/AI= 1.15,I.00,0.85,and0.70 are0.60,0.665,0.575,and0.355,respectively.Becauseofthelargedifferenceinmeanvelocityforthedifferentarearatios,thequantitycoefficientis probablynotthebestbasisonwhichto comparetotal-pressurerise.A betterbasismightbe theeffectiveangleofattack~e whichisbasedonthemeanaxialvelocity.Infigure2k,thetotal-pressure-risecoefficients~ areplottedagainsttheeffectiveangleofattack.Nesrthedesignangleofattackthedifferencebetweenthe VT valuesfor A2/Al= o.85.@ 1.0 isverysmalL(approximately2 percentof*T at A2/Al= l.O)jwhereas,thedifferencebetweenthe *T valuesfor A2/A~= 1.15ahd1.00 isnotas small(approximately7 percentofVT at ~/Al = 1.0)● Theincreaseddiffusionwith +/Al = 1.15decreasestheefficiencyand,hence,decreases‘$T.me chage int~-gentidVelOCityAVtU, whichis_prOpOrtiOIIdto VT fOrCOIIStUtefficiency,isaboutthesamenearthedesignangleof attackfora,, .-—..

.

0

.4

NACATN4130 Il.

15-percentdecreaseinannulussxeaas forconstantannulusareathroughtherotor.Evidentlytheeffectiveturninganglesneardesignsrenearlythessmeforthesameeffectivesingleofattackeventhoughtheareachangeconsiderablyalterstheinletandoutletairangles.Thestatic-pressure-risecoefficientisgreatlyaffected,however,andislarge-dependentonthearearatioas shouldbe expected.

Thecurvesof a againste atthesefourannulus-srearatiosfortheinbosrd,pitch,andoutboardsectionsaregiveninfigure25. Themeasuredturnihganglesvsry~eatlyfromthevaluesestimatedfromcas-cadetestsfor A2/Al= 1.00.Thevariationissystematic,butdiffer-encesinturningasmuchas4°,arounddesignangleofattack,existwhentheannulus-srearatiois15percentaboveorbelow1.00. Thedif-ficultyofestimatingdirectlytheturninganglesofbladerowshavingannulus-srearatiosotherthan1.0fromconstant-sreacascaderesultshasbeenWown forSOEEtime.However,methodshavebeensuggestedtocofiatthisdifficulty.

Onemethodof convertingthedatatotwo-d~nsional-flowconditionsisto assumethattheeffectiveoutletangleisfoundiftheoutletaxialvelocityistakentobe thesameas theinletvalue,theoutlettangen-tialvelocitynotbeingchsmged,seevectordiagraminfigure26(a).Thisassumesa constantcirculation.Thiscorrectionsystemwasappliedto thesetestsandtheresultmtcurvesof a against19 areshowninfigure27. Cascadecurvesalsocorrectedinthissamemannerareincludedforcomparison.Thesystemis seentoresultina laxgeover-correctionineveryinstancewiththediscrepanciesnearlyas lsrgeas thoseoftheinitialuncorrectedvalues.Consequentlythissystemdoesnotappeartobe valid.

A secondmethodof correctionistoretainconstantcirculationbutto correctbothinletmd outletsnglestothemeanaxialvelocity.Thisof courseintroducesa differentvalueforboth a and 19,as indicatedinthediagramoffigure26(b).Theresultsusingthissystemareshowninfigure28withestimatedcurvescorrectedinthesanemanner.Theagreementismuchbetterbutcsrefulobservationrevealsthatthisisalsoan over-correctionsystem.

Inan idealsystem,allthecurves,regardlessofaxial-velocitychsmge,wouldfallupontheconstant-annaus-arealine. Althoughcor-rectionto an axialvelocityofmagnitudebetweenthemeanandoutletvaluewouldneWly producethisidealresultforthesetests,itIsnotbelievedthatthissamecorrectionwouldbe optimumat otherrsmgesofinletairangleorforothermethodsofproducinga changeinaxialvelocitythroughbladesections.l%erefore,a morefundamentalsystemmustyetbe devisedifhighaccuracyat allconditionsisdesired.Atpresent,themean-axial-velocitysystemappearstoyieldresultsof

—

12 NACATN4130

sufficientaccuracyformostapplicationsinwhichsyial-velocitychangesup to about15percentoccur.

Theman-axial-velocitycorrectionsystemwasusedto estimatefromcascadeteststhepressure-risecoefficientsandflowanglesacrosstheannulusfora testneardesigninletairangleat eachof thefourarearatios.Theseestimationsarecompsredwithmeasuredvaluesinfigures29and30. Forthearearatiosof1.00and0.85,thecomparisonsarenearlyexact.Thedifferencesinflowanglesandpressure-risecoefficientsevidentfor A2/A1= 1.15 probablyresultfromthelocaleffectsoffairingsandincreasedboundary-layerthicknessdueto theseverestatic-pressureriseacrosstherotor.Theconfigurationwith A2/A1= 0.70hasa verylowestimatedstatic-pressure-risecoefficient,0.21comparedto 0.56fortheconfigurationforwhich A2/Al= 1.0,sothelowvaluesof total-pressure-risecoefficientcannotlogicallybe attributedtothickboundarylayers.Morelikely,theassumptionof constantcircu-lationforthiscaseofincreasingsxialvelocitythroughtherotorisunsound.

Sincetheoutletsxialvelocitiesforthesearea-changetestswerequitedifferenthorntheprevioustests,a comparisonwasmadebetweenmeasuredvaluesand-thoseestimatedusingthesystemofreference7.Thiscomparisonwasmadeforonetestnearthedesigninletairangleforeachareachangecondition,seefigure31. Themeasuredtrendisagainestimatedquitewellby thesystemusingeithercascadedataortiieinterpretati&of Const&’sruieforestimating

TestsofVaryingReynoldsNumber

Therotorspeed,qmtity-flowcoefficient,and

outletairangles.

resultantmeansec-

A

—

—

e.

—

—tionRe~oldsnfier-o~thetestsmadeto investigatetheeffectsofReynoldsnumbersreshowninfigure32. Thedesignflowcoefficient,0.476,isalsoindicatedinthefigure.~e”Reynoldsnumberatdesignflowforgivenspeedswasusedinthepreparationof theothergraphs.Theadiabaticefficiencyisshowninfigure33. Thescatterinresultsisbelievedduetotestinginaccuraciesillustratedby theerrorsinmeasuredmassflowsasalsoshowninthisfigure.Thetrendisquitedefinite,however,indicatingan appreciabledecreaseinefficiencybelowR= 250,000.A comparisoncurve,estimatedusingcascaderesultsandtheequationofreference6, showscloseagreementintrendandabsoluteval-

—

ues. BecauseitiscomputedusingqnlydataforamI?ACA”65-(12)1Osec- A

tionanditdoesnotincludecasinglosses,theestimatedcurveisexpectedtobehigherandatbestanapproximation.Thetotal-and vstatic-pressure-risecoefficients,figure34)alSOshowa decreasewithreducedReynoldsnuuiber.Thereductionis10“to15percentforbothcurves,butnodefiniteReynoldsnumberbelow400,000“canbe described

.—

NACATN 4130 13

. asthe“drop-offpoint.” Thecurvesof turningangleagainstReynoldsnumber,fairedat a’designquantity-flowcoefficientof 0.476fortheinboard,pitch,andoutboardsections,figure35,showunexplained-4trends.A gradualreductionisobservedas theReynoldsnumberisdecreasedfrom400,000to 150,000.At R = 150,000,thetrendreversesandatR= 80,000,a returntovaluesmeasuredathigherReynoldsnum-bersoccurs.Theestimatedcurvefor theinbosrdsectionfollowsthistrendtothepointofreversal.CascadedataforlowerReynoldsnuniberssrenotavailable.Thedisplacementof theestimatedcurvefromthelomeasuredcurveby 1°to lZ isinagreementwiththecomparisonsmadeinfigure15. Similarturning-angleresultsat theselowestReynoldsnum-bershavebeenobservedinotherunpublishedinvestigations.

CONCLUSIONS

An investigationofa typicalaxial-flowcompressorrotoroverarangeof quantityflowrates,bladeangles,annul.us-arearatios,solid-

. ities,sndReynoldsnumberswasmadeandtheperformancecomparedtovaluesestimatedusingporous-wall-cascadedata. Af3a resultofthisstudy,itisconcludedthat:

s1.Low-speedcascaderesultscanbeusedto estinaterotorturning

sqles,static-andtotal-pressure-risecoefficientsandefficienciesaccuratelyfora tidevarietyof’conditions.

2.Themean-axial-velocitymthod of convertingthedatato two-dimensional-flowconditionscanbeusedwithgoodresultsinestimatingrotorperformancefromcascadedataforaxial-velocitychangesacrosstherotoras largeas 15percent.

3. Thecalculatedoutletsxialvelocities,excludingtheboundarylayer,werefoundtobe ingoodagreementwithmeasuredvaluesforallcomparisonsmade.

4.Theflowturninganglesproducedbythetestrotorwereconsis-0 tol~”forallconditionstested.tentlyhigherthancascadevaluesby 1

14 NACATN4130

5.Theeffectof’decreasedReynoldsnuiiberwasfoundtobe verysmallintherangebetween250,000and500,000.As theReynoldsnumberwasdecreasedbelow250,000,decreasesinrotorefficiency,pressure-rise-coefficient,andturningsnglewereobserved.Goodagreementbetweencascadeandrotortrendswasobserved.

LangleyAeronauticalLaboratory,NationalAdvisoryCommitteeforAeronautics,

LangleyField,Vs.,December16,19~2.

NACATN4130 15

APPENDIX

REDUCTIONOFDATA

Therelevsmtrelationshipsandthemethodsusedin calculatingtheperformancefroIuthetestdatatillbe presented.Allperformsnceqy.sm-titiesarebasedonenteringconditionsof2,116poundspersquarefootand519°F absolute.

Therotorinletdensityforthesetestswasdeterminedfromthefollowingexpansionof theisentropicpressure-densi~relationship:

.

P

[

pch1 --(PS-P‘ch s

-P‘ch

YP‘ch

2y_psch

Y -1 )]-1

Theflowwasassumedtobe incompressibleandthepowerinputiscalculatedfrommomentumconsideration:

e

J(rtL12-I~=p)‘%v~ - ‘+%2x% d(r2~

rh2

Totalpressuresweredeterminedfromradialsurveypoweroutputbasedon chaniberpressuremaybe veryforlowvaluesof PT2/PT1by

measurements.Thecloselyapproximated

‘2=‘~”‘a2~T2-‘T.h)d(r2)Themassflowat eachpointis

Jrt2M=fi ()pVad rz

‘h2

Foreqyalmeasuredwss flows,rotorefficiencyis

‘2 - ‘1nr =

12 - ‘1

Sincemeasured mass flow varied by as much as 5 percent (figs.8 and 9)a mm-flow correction

was applied to the efficiency. Ifthemassflowat position 2 is assumed correct and if the massflow errors are assumed to be the result of errors in static-pressuremeasurement,rotm effi-ciency becoms

Actually the correctionshad ody a minor effect because El and 11 were always close to zero.

In reference 7 theoretical equationsare presented for the calculationof axial velocitydistributionsupstreamor downstream of blaie rows. If the general equation is mdified for thecondition of rotor testE with no guide vane and with constsmt inlet axial velocity (neglectingwall-boundarylayers), the rotor exit axial-velocitydisl@bution becom?s:

JW% SIA2—(, d D/Dt’

$.(co6 %):%@ ‘/Dt

1

z.1’

‘1 9. i I111 , I ,1

31?NACATN 4130

Determination (1‘a2ofthevalueof necessaryto satisfyUt CosP2 ,=

~ut

continuityreqtiresa trial-and-errorsolution.As a firstapproxima-tionan estimationofthepitchsectionaxialvelocityismadeby usingthe

The

for

the

followingequation:

generalequationC= thenbe writtenfor(va2~t)%,Dt ~ sOlve~

()Va2

()

vWiththisfirstapproximationof a2

‘tc”s ‘2~Dt0 Utcos i32wDt’

generale~ticm canbe usedtodeterminetherotorexitaxialveloc-itiesforsev&alradialstationsacrosstheannulus.To satisfycon-tinuity,theintegratedareaunderthecurvesof V@Jt ‘d ‘a2/”t

(/ )2

plottedagainstD Dt mustcoincide.Ad~ustmentofthevalueof/ \

()‘a2Ut CosB2ismadeto satisfycontinuity.Generally,continuity

%@canbe satisfiedwithin1 percentforno

()

Vamentsofthevalueof 2 .

‘t Cos‘2Dh/Dtitydistributionwascalculatedforfive

moret- twoor threeadjust-

Inthispapertheaxialveloc-

radialstationsfromrootto tip.

A generalmethodforpredictingefficienciesforbladerowsforwhich L/D valuesareknownormaybe estimatedisgiveninreference6.Thebasiceqmtion

Powerlosses~ =1-

Powerinput

-—_

18 NACATN 4130

whenappliedinthis.investigationbecomes.-.

where

datathis

Estimatedvaluesof L/D wereobtainedfromtheavailablecascadeforthevaluesof j3,a,and G expectedatvariousradii.In —paper.thiswasdoneattheinboard~pitch,andoutboardradii.‘I’@

sectio=lefficiencyateachpointwascalculatedby usingtheaboveequation.Thecalculatedefficiencieswereplottedagainstradiusandfairedtothecasingstoprovideanefficiencydistributionacrosstheannulus.Theesthatedrotorefficiencieswereobtainedby mechanically .-

integratingtheseefficiencydistributionsandcomputingaveragevalues. A

NACATN4130 19

REFERENCES

1.Westphal,WillardR.,endGcdwin,WilliamR.: ComparisonofNACA65-SeriesCompressor-BladeWessureDistributionsandPerformanceina Rotorad inCascade.NACATN38Q6,1957. (SupersedesNACARM L5U120.)

2.Bogdonoff,SeymourM.,andHerrig,L.Joseph:PerformsmeofAxial-FlowF= ud CompressorBLadesDesignedforHighLmiiings.NACATN1201,1947.

3. Herrig,L.Joseph,Einery,JamesC.,andErwLn,JohnR.: systematicTwo-DimensionalCascsdeTestsofNACA6~-SeriesCompressorBladesat LowSpeeds.NACATN3916,1957. (SupersedesNACARML51G31.)

4. Schulze,WallaceM.,Ashby,GeorgeC.,Jr.,andI&win,JohnR.:SeveralCombinationProbesforSurveyingStaticandTotalPressuresndFlowDirection.NACA~ 2830,lg52.

5. Davey,RichardS.: MultiplePressure-TransferDevice.Instrumentsjvol.23,no.4,Apr.1950,p. 350.

6. Sinnette,JohnT.,Jr.: AnalysisofEffectofBasicDesignVariableson SubsonicAxial-Flow-CoqressorPerformance.NACARep.901,1948.

7. Bowen,JohnT.,Sabersky,RolfH.,andRannie,W. Duncan:TheoreticalandExperimentalInvestigationsofAxialFlowCompressors.SummaryReport,ContractN6-oR1-1o2TaskOrderIV,OfficeofNavalRes.,Mech.E@. Lab.,C.I.T.,Jan.1949.

8. Howell,A.R.: ThePresentBasisofAxialFlowCompressorDesign.PartI. CascsiieTheoryandPerformance.R.& M. No.2095JBritishA.R.C.,1942.

.

v

20 NACATN 4130

TABLEI.-DESIGNCONDITIONS

[ 1Descriptionofbladesectionsisgiveninreference3

Section e, P1) 132, ~) ~NACAbladeprofiled~~ deg deg deg deg D/Dt

Originaldesigndetailsfromreference2

Root

1

65-13.5)10 16.7 24.1 48.8 24.7 32.1 1.135 .784Pitch 65-11)10 13.1 17.4 52.4 35.0 39.3 1.000 .892Tip 65-8.5)10 10.0 12.9 55.5 42.6 45.5 .892 1.000

I Design conditionsforthisinvestigation IRoot 65-13.5)10

[15.6 24.2 48.9 24.7 33.3 1.135 .784

Inboard 65-12)10 13.3 19.9 51.1 31.2 37.8 1.051 .849Pitch 65-11)10

[12.0 17.5 52.5 35.0 40.5 1.000 .892

Outboard 65-10)10 10.8 15.5 53.8 38.3 43.0 .954 .935ITip 65-(8.5)10 8.9 13.0 55.6 42.6 46.7 .892 1.000

v

.

#

NACATN4130 21

pl= 52.5—

I a = 12.0

I

u = o.892 UarQ= 0.416

Designcondition

Ie=uo3 61=0(

/

I CC-12.1

f=4005–

,9‘1

I

Figurel.-Velocitydiagramsatpitchsectionfordesiguconditionandmeasuredvaluesneardesignflowrateexpressedas a fractionof Ut.

— —-—

lum

‘rAdjustableendplata

*ttlm ‘=’3 - SoreensRadialdiffuser

Armulardiffuser2121

75 m Isotar

l::

t 1 I,? ‘------

!mansferdevice~~. p ~ driveshaftN- @-----—-. = ~ / Rotor= I# 1. -=--T I

12%n--nt. Rota’ blades

Figure 2.- Schematic diagram of test compressorshowing rotor andinstrumentationpositions.

NACATN4130 23

.

*

.

,. .-.—~“-—”- “’9-Z

.4

3/.--..;:-,.—._

Y- .7--- *..

-. , . ._=. — . . — ..—

_?:.- ~.--

? .,

.— .. ,--

1.I

t

.

/7 3

,-...—-----

● ✎ ✎✍✍

-———

.-

-. -—J.—J—

— ?—— ,’—L= ;-+d,—..— .—- :—.+ .

&.r.- —-—— .—— .—. .- . - - -. —..,-. ,---- —.. .--- . ----- - ~ I:aw:..--—-.—..a. :,..,,.,-T-— , .... ,-.J

“““+iiik—.+-.—.—_— ----1-..-. .L. :— :. ..- —.=”

~Y .=----- ,. .-.,

-=’.-”*–.+- “~ g

Figure3.- Partialviewoftestsectionshowingrotorconfigurationasslteredtoprovideexit-to-enteringannulus-arearatioof1.15.Solidity,1.0;7~0abovedesignbladeangle. 1

NACATN4130

.

k ---L--- _----_-J.__.

‘---l----------t--’ +

‘---+----------+ --’

(a)‘Z/AI.1.15,

A ) -t——-—__—-—-—-__&B

f 1-——_—__—_—__—-B+

c -—-t_______ ---+___

(d k/k~.,3..9,5

A-AB-Bc-a

Figure4.-Cross-sectionalview‘areachangesmadetangentoftheblade

H-A—---—-——--—-–A

B—--———---__——--B

o-- -——-----——— .-— c

(b)‘Z/AI. low,

—..

A-— r t-—————__--A-

B—-

f t

---- -— —___ --- B

–cm---”--i--c

‘d)‘~Az.0070>CntbxrdsectionPitabmotionInb3ard motion

throughrotorillustratingtheannulus-testedinthisinvestigation.Thebslsafairingwastotheaxial.directionattheleadingandtrailingedgesandarbitrarilyfairedinbetween.

.

.

4

.

.=

—

NACATN4130

.

●

.

.-

-——.

.

.

25

\

NACA.,

Figure5.-Surveyinstrumentwithmeasuringheadinstalledshowingarrangementofyaw,total-pressure,andstatic-pressmetubes.

b

... .,* Jlu..>!,

Figure 6.-Instrument

*.

.+. . . ,,. ,,,. LL.. -AJLLuLU

used on rotor to meamrce blade wake profiles.“

*

mm

r #

,,

Figwre 7.-Inslmumentusedonrotortomeasureandoutletangle.

a-lJ-”[mik).1.,! .,, ,,, .ti A 4LA.U!L

static and total pressure

.

h

2

0

-2

4

h

2

0

4

4A s .6 .7 d .9 1.0

*ma maw - ●t,kti.m2, Slug8/.ck

F@Ore 8.-Difference in measured mass flows at statlona 1 and 2 for thelo

constant-areatests with design, 7Z aove des~, @ 7~0bel~ des~

blade angles at solidifies of 1.0 and O.5.

NCD

a . L,

I 1 ,1● ✎

.

h

2

0

:

i

-+

L

;

{,

a.

! e

5

I

0

-2

-43.5 .bO .50 .&J .70 .00 .5U

— - n. *t.taum2,Sluul/...

Figure 9.-Differenceunmeasuredmassflowsatstatfone1 and2 for7~0,5

abuve desi~ blade angle and solidityof 1.0 with anmihm-area ratiosof 1.15, 1.(M, 0.85,and0.70.

.c

m)

%

90

M

80

100

55

90

E$

w.2 Jl .6 .8 1.0 1.2

Qwtii?owffidmt, @

Figure 10.- Vsriation of adiabatic effici.encywith quantity coefficient

10 abovefor constant-areaconditionwith blade angles at design, 7Zdesign, and 73°below design settinge at solidifies of 1.0 and 0.5.

(Hnesacrosscurvesindicatedesignpoints.)

1 .I ,, I I

uo

# t , .

1..0

.8

.6

&

.2

0

.6

.b

.2

0.2 .b .6 ,8 LO

malJti*0w2fird9nt**

Figure IJ..- Vaxiation of total- and static-~ essu.re-risecoefficientquanti~ coefficientfor the com3tamt-anraM.s-areacondition with

10 below design blade angles at solldities#0 above design, and 722and 0.5.

1.2

with

design,

of 1.0uP

32 NACATN4130

100

95

90

85

80

?5

70

65

60

I Estimatedfranrotorwakes +— I I I I— Estimatedfromcascadewakes-–D--

Measuredvalues o—.. —

,v

I

I 1 I I I I_ .1 I I I I 1

?.

●

●3 J! .5 .6 .7 .8

Quantitycoefficient,@.

Figure12.- Variationofmeasuredandestimatedefficiencieswithquan-titycoefficientforconstant-annulus-areaconditionwithbltieangle 8

set7*”abovedesigmatsolidity1.0. (Verticallineacrossthecurveindicatesdesi~point.)

. , # * , .

,

r— --t Y> ~

=K9=–0 2 4 6 8 10 1.2 Ill 16 18 20 22 24

Mstmce, in pwwentof blade spacing

Figure 13.-fkoparison of wakes measured h the test rotor at the pitchdiameter and in a low-speed cascadetunnel. The rotor pitch diameterhad EKINACA 65-(H)Ioblade section. The blades j-nc~cade were ofNACA 65-(u)1osection.P1. 60°, a . U’”.

34 NAC~TN4130

F

o 1 2 3Hub ‘rip

Annulusheight,inches

Figure14.- Vsriationofefficiencyacrosstheannulusfora typical

testoftheconstant-annulus-areaconditionwith7~0abovedesign

bladeangleat solidity1.0. @ approximatelyequsltodesignvalue.

.

“B

. ..-

NACATN4130 35

●

✎

.

.

-o & 8 1.2 160

Figure15.- Variationofandestimatedforthe7L0abovedesign,and2

20II 8 12 16 20

0 b. 8 12 16 20

lnglaofattaok,a ,&g

(a)Inboardsection.

turninganglewithangleof attackasmeasuredconstsmt-annulus-areaconditionwithdesi~,

7~”belowdesignbladeanglesofsolidifies2

of1.0and0.5. (Bsrsacrosscurvesindicatedesignpoints.).

.

An@ of atti, e,CM

(b) Pitch section.

I Figure I-5.- Continued.

. ,* 1 . ●

Um

, ,1 #

24

20

16

12:a.00

m“G94

g )lmmnwd Esttited Maasured

1

o—10

A fd+ 7.50

12

B

4

0 4 8 Is 16 200 4 s 12 16 &l

Angh OfattlOk,a,*g

(c) Outboard section.

Figure 15.. Concluded.

. ,

38 NACATN4130

.In

mmm An@m 21ad0W*BStatkmry Rotor-xmntadOMcw!+ Xmwred 2ntimatadinstmmeatim&mmlt EdLmatad—~— ,---.+ . ~ -?+- htboud

65-(10)10

24

20

16

12

8

.,?--- - .. ... —.04

.0s

3

4--.02;

$

3.01

80

0 4 8 lE 16 20 -. . .—.0 4 e 12”16 m

0 4 a 18 16 20

—-

—.

.-

..- .

---●

.-

.-

.—. .

..<

Angleofattmk, a , deg

Figure16.- Compsrisonofestimatedvartationofturningsnglewithangleofattackwithmeasuredveluesusingstatio~ androtor-mountedinstruments,andcomparisonofmeasuredandestimatedwakesfortheconstant-annulus-

sreaconditionwith

(Barsacrosscurves

a 7L0abovedesignbladeangleata solidityof1.0.2

indicatedesignpoint.) .

NACATNk130 39

.

2a

24 ..+--------,------ -- ------ --- ---~--,--. _ --

- - - . -:--p --%) - - - .

- ~- .

- . --n-.

3

16~ ~

— ~ = .~ .

6+(I.2)1O12

65-(ll)lo65-(10)10Eetilnateaau-I.@ a-1.CXJo-.950 A c1 8❑ & o ——. 12

24 0 n o ------- 16---n---------. -- --,---- ---

20 -k ---_ --- - S-..- -- -.

- . ------- - k4 . --’_

- -16 4 — _ - -

— _ *- —

— ~

12

8 --+ --------20

---~ -- A---

-. v- ---

e -—- _ -.-.— 7 — — — -

- —.- —

- .

16 ~- - e-

*-

— -— -n-

— _ %— _

12-

v–

a32 36 40 44 48 62 66 60 64 68

Inletairangle, ~1 , deg

(a)Solidityofpitchsectionof1.0.

Figure17.- Variationofturningaaglewithinletairangleat constsmtanglesofattackasmeasuredandestimatedattheinboard,pitch,andoutboardsectionsforconstsnt-annulus.weatestsata solidityof1.0.

40 lWK!ATN4130

16

12

a

4

0

I I I I I I I I I I I I I I.- -

--- --- Ju--- --- ------- -..------—. . --- -.—— —L . --- ---

- . --— h. ---

— — +-1

d c- - w~ -

65XI2)1O 65-(U)1O 6-(10)10Estimated a

O ● ●52 a-.50 e=.)J8o A-- n. ItEl h 0 .—— 80. D .- (’) ---------U. 2.. . .

$ 166a -- -

0---- ~-

- ----

12--

+- -- ._ -- --- - --- ---------0“ - h- —- ‘---“---- ._-{ ‘--‘a1

— -IL --—

mn — - — —zh

4

0

12--+

--- __ - -------~* —–I I I I I I 1 1 v,— 7 I [

I I I I I I

----- --- :9:’:::.--+

t-+--l-l--l t- - -- -- ‘------F.a-

. .— ?-- L —

8 — — — — — ~ -

1

-c)— . —

4 I

-=w=’–

032 36 40 44 48 62 56 60 64 68

InletairaX@e, @l , deg

(b)Solidityofpitchsectionof

Figure17.- Concltied.

0.50.

a

.

.

.

.

.

I * , 1 . ,

72

a

56

46

:i?$40

i ,2

2

24

16

8

01 2

E&ADnulua )A.@it,I.nohea

EimWaaured~ted

h—~ ---n —-—0 T--—

Anuulun Might, inohes

Figure 18.- Variation of measured and esti&ted flow angles relative to therotar aeroBs the anmdm for typical tests nem design flow rate fordesign blade angle, constant annuha area, md mlidities of 1.0 - 0.5.

+

.9

.8

.7

.6

.5

.4

.5

.LIIL : =#Ll

.1

.OO 1 Q sHa alp

Armulumheight,inohaa

1 2 3&b !l’lp

Annul.u8helsht,lnchem

Figure 19.-Variation of nmumred and estimatedtotsl- and static-presaure-rise coefficientsacross the annulusfor typical tests near design flowrate at the design blsileangle, constant-amulus-srea conditionforsolidifiesof 1.0 and 0.5.

, , , , . Y

1 1,

I tI I

.9

.6

1.0

.5

.4

.3

Inlet air angle, ~,deg

F&ure 20.- Variation of average measured end estimated M/da valueswith inlet air angle fcm the constaut-annolus-areatests at soli.ditiesOf 1.0 and 0.5.

5

44 NACILTN4130

1.0

.9

.8

.7

.6

.8

.7

.8 ~

.7

.6

(a) 7~0belowdesign,an@.eat”a &~ity of1.0. ‘

Figure21,.-Variationofmeasuredandestimatedaxislvelocitiesacrestheannulusintermsof Ut forfiveflowratesattheconstant-

+

.-.

s .

annulus-areacondition.

NACATN

.’

.6

s

.

.lt

.3

.8

.7

.6

.5

4130 45

.6J~_ -y

_ —- --

.5 - ‘h >Jf

\~

●h

\IL.3 A

F * + 3 * - +a—— ——--—. .==P 9

●7 1

.6 /

.5

●ll/ p

●9+ + )_ -~ n 1

d J— .= * _

.8 !)

/ Pitchsection,Bl Meaeurea.7 46.4

48.6 :52.6 056.6 A

.6 \/ 58.7 “ &

Estimatedfrom=sCSJ&— —— - —Eetimtedfrcmtheinter’pre~tion–-–– ---of Conetant’srule

I I I l=.5 ~ 1 2 3

Hub Tip

hmlue height,inchee

(b)Designbladeangleat a solidityof1.0.

Figure21.- Continued.

46 NACATN4130

.5

A

.3

.7

.6

.5

●h

●5

●3/

[s..2 I

.7

.6

.5

0 1 2 3Hub Tip

ANIulueheight,inches

.

.

(c] 7~0abovedesignbladeangle.

Figure21.-Concluded.

. , * , k

m

95

w

s

65

lm

95

90

85

80

~2.b .6 ,8

.2 .I1 .6 .0

Cwlwv C-=.Orriniant,●

Figure 22.- Varlation of adiabaticefficiemy with quantity coefficient

10 Aove designfor 75 blade angle, solldi~ of 1.0, and annulus-area

‘4!%*

ratios of 1.15, 1.00, 0.85, and 0.70. (Verticallines across curve

indicate design pointa.)3

Lo

.8

.6

* .h

.8

.6

H-t--t+ I I I I I1’

I I I I I

W-H---H I 1%.1 1%1 I I 1%1 I 1%1I 1 1 I n 1 1 Y 1 1 1 I 1 1 la I

. ..2

Figure 23..

quantity

.h .6 .8

.2 .b .6 ,6

Q.lult.itY~, *

Variation of total- and static-pressure-risecoefficientswith

coefficientfor 7~0 above design blade angle, solidi~ of 1.0,

Lltlilannulu8-2mea ratios OfL1.15, 1.00, 0.85, and O.~0.

1 >

4

,

s’

.9

..9

.7

.6

.5

J!

.3

.2

.1

0-2 0 2 4 6 8 10 12 u16 12 m

Figure 24.- W3riation of’tots.1-~esme-rj.se coefficient *T With effective

sngle of attack ~ at the pitch diameter fcm the several annulus-area

ratios tested. (Bern across curves indicate &esign points.)

, h

I I

‘E

8

)4

00 4 8 u! 16 mo h 0 J

II 8 lz 161.2 m

Allg10 of attack,a, *

Figure 25. - Ccmpsrison of estimatedand measured turning-anglevariationwith angle of attack at the Mboard, pitch, and outboardblade sections

for7~”above design blade angle, solidi~ of 1.0, md annulus-area

ratios of 1J5 1.00, 0.8?5,and 0.70. (Barn across curves indicatedesign points.~

, . , . , “

‘a-!=

Go

# . # , 1 *

(a) Veloci@ vector diagram cmrected (b) VelocitYvectm diagram cmectidto enteringaxial velocity. to man axial velocity.

Figore 26.- hkthodsof cmecting velocity vector d@ram6 of VaI’Yl.W

axial veloci~ to cmpare with constant-axial-velocity caacade or

compresam result~.

UP

Auglaofattwk,a, (k

Figure ~.- CcrQarisonof estimatedand measured turning-anglevariationwith angle of attack at the Mboard, pitch, and outboardblade sections

for7~0 above design blade angle, solidity of l.O,aroi annulus-area ratios

of 1.15, 1.00, 0.85, and 0.70 using the constant-circulationsystemcorrected to the ent~~ axial velocity. (Ears across curves j.33Mcate

design points. )

*t .

II, ●

, , ●

Figme 28.- Comparison of estimatedawl measured turning-emglemrlationwith angle of attack at the inboard,p,ltch,ad outboardblade sections

for ~~” above design bhde angle, solidity of l.O,and -us-area ratiofi

of 1.15, 1.00, 0.85, and 0.70 using the constant-circulationsystemcorrectedto the mean axial velocities. (B=rs across curve= indicatedesign points.) U

u

(a) Ammlus-mea ratios of 1.15 a33d1.00.

Figure 29.- Variation of measured total- and static~pressure-risecoeffi-cients across the annulua M comparedwith values esthatsd usimg the

man-axial-velocity, constant-circulation system fac 7~0 abo& design

blade angle}ani solidi~ of 1.0.

. .. ,

, , , 4

o 1 2 3tbb m

JmEoDa M.@, iul-

(b) Annulua-arearatios of 0.85 ando.70.

Figure 29.-Concluded. Uul

12

a

%

w

w

32

2b

M

8

00 1 2 3

m &Acailwhdgm,irlcina

o 1 2 3&b =P

~w,-

(a) Annulms-arearatios of 1.L5 endl.00.

Figure 30.- Wriation of meaaured flow angles relative to the rotor acrossthe annul.us8S cqed with values estimated using the mean-axial-

velocity, comtant-ctidation system for 7~0 above design blade -e,I

and Solialty of 1.0.

, .1

. .,,.

. .

72

&

56

ho

Ml

32

$%

16

8

00 1 2 3Hub %

AmIllIMILd@t, Iwlma

o 12.

3

IIQb ‘rip

Amnau9 height,Im&a

(b)AnnuUm-area ratios of 0.85 and 0.’70.

Figure 30. - Concluded.

58 NACATN41.30

i%

.8

*7 — =

.6 Measured Eet5nletea EstimatedfromthefromCaacade interpretationof

A.J~

Conetant’srule

u ——— —–--–—–— ‘J 1.M

.5 1.00: .05A .-IQ

.7

=.6

* 7%\

<

s v

.6 ,4.

s -— ----

A

•1●3 .5T

%- :s

●4

0

.3- ~I

1 2Hub T/p

Annulusheight,Iriih;s

.

.

Figure31.- Variationofmeasuredandestimatedsxialvelocitiesacross

theannul.usintermsof Ut 10 aboveatflowratesneardesignfor72

.

—*

.

designbladeangle,solidityof1.0,andannulus-arearatiosof1.15,1.00,0.85,and0.70.

* ●✎ ✜ ✌ ✌

:*

i! .3

.2

.1

00 lcu,om 2oo,um XX)*OOO hoo,ml

Sc@mlds number

1Sclta500,000

Figure 32.-Relation of Reynolds numlxx effect tefitpoints with the design

quanti~ flow coefficientat solidity of’1.0, 7~0 dove design blade

angle, and constant annulus mea.v)

-4

1, .

o 100,000 200*000 300,000 400,000 500,000-~ ~

Fi@me 33..Variation of measured mass flov error and measured and estimated

10 above designefficienciesTELthReynolds number at solidity of 1.0, 7Z

blade ax@e, E@ constant srmulusarea.

mo

Figure 34.-

Reynolda

Conetant

SOynoldanmber

Variation of static.

number at a solidi~

anmilua area.

and totel-prefisure-rifiecoefficientwith

of 1.0, 7~0 above design blade angle, and

UY

.

20 Imw! 1

Figure 35.- Vsrktion of me8sw~ =d cascde est-t~ t~ -es”with Reyno1.dEnurber at the iribd, pitdl,andoutb~ sections

, ,. .