TECHNICALNOTE3924

DISCHARGE COEFFICIENTS FOR COMBUSTOR-

LINER AIR-ENTRY HOLES

II-FLUSH RECTANGULAR HOLES,

STEP LOWERS ,AND SCOOPS

By RalphT.Dittrich

LewisFlightPropulsionLaboratoryCleveland,Ohio

Washington

April1958

TECHLIBRARYKAFB,NM

Illllllll[lll[llllllllllllllUUL713L

NATIONALADV3EOIBCOMMITTE3FORAERONAUTI@

TECHNICALNOTE3924

DISCHARGECOEFFICIENTSFORCOMBUSTOR-LINERAIR-ENTRYHOLES

11 - FLUSHRECTANGUIJRHOLES,STEPKTUVERS,ANDSCOOPS

By RalphT.Dittrich

An experimentalinvestigationwasconductedtodeterminedischargecoefficientsforvarioustypesof combustor-linerair-entryholessuchas flushrectsmgularholes,steplouvers,andscoops.Thedatapresentedhereinshowthevsriationindischargecoefficientof eachconfigurateionas a functionofa dimensionlessflowparameter.Withintherangeinvesti-gated,theeffectof sizeor shapeof flushholesondischargecoefficient

- wassmall.comparedto theeffectsof ductstreamvelocityorpressureratioacrossthehole.Whiletieadditionofa scooptoa flushholeincreasedthedischargecoefficientonlyat lowvaluesoftheflowparam-

3g eter,thesteplouverandthethumbnail-typescoopincreaseddischargecoefficientsthroughouttherangeoftheflowparameter.However,at lowvaluesoftheflowparameter,thedischargecoefficientsforscoopsandstep10UV=Swereaffectedbyboundary-layerconditionsoftheductstream.Theprotimityofmultipleflushholesorthewallinclinationofa convergentducthada negligibleeffectondischargecoefficient.

INTRODUCTION

Withthetrendtowardgreaterairloadingandhigherairvelocitiesthroughturbojetcombustors,a knowledgeof dischargecoefficientsforlinerwallopeningsis essentialforthedesignofaerodynamicallyeffic-ient cotiustors.Dischargecoefficientsforflushcircularholeswithflowparallelto theplaneoftheholearepresentedinreference1. Thepresentinvestigationextendstheworkofreferencelby presentingdis-chargecoefficientsforvariousothertypesof linerwallopeningssuchas slots,scoops,andlouvers.

Withflushcircularholes(ref.1)theeffectsofholediameterandwallthiclmessattheholeondischargecoefficientsweresmallcomparedwiththeeffectsof ext=nalparallelflowvelocityandpressureratioacrossthehole;theeffectsofductheight,pressurelevel,andboundary-I.ayerthicknesswerenegligible.Applicationofthedataofreference1

●

.

2.

NACATN 3924 .

to calculatedflowconditionsina mode?thatforflushcircularlinerwallholesvaryfromapproximately0.2to0.6.

cimbustor(ref.2)indicates ethedischargecoefficientmay

A tier designmayincludeother&pe8-ofair-entryopeningsthathavespecificapplication,suchasthumbnailscoopsor steplouvers(step-wallconstruction)forwallcooli& longitudinalslotsfordepthof jetpenetration,or scoopsoverhole’forjetdirection.

!~SO, such

factorsasthespacingofholes,bothi,thelongitudinalandtransversedirections,andtheinclinationofthe~ner wallmy affectthedis-chargecoefficient.

Accordingly,thefollowinggeometricqgdflowfactorswerestudiedinthisinvestigation:(1)longltudina~slotswithlengthtowidthratiosrangingfrom1 to 16;(2)steploJuve&sranginginheightfrom3/32

.

to5/8inch,bothwithandwtthoutwall,overlapora corrugatedspacer;(3)thumbnail-typescoops1/8and1/4~ch high;(4)scoopsoverholeswithscoopfaceareavaryfmgfYomO.6tol.”4timestheholearea;(5)circularflushholesinwall.swithinclihtionsof0°,8°,amd20°;(6)steplouversinwallswithinclinations~f~, 8°,and200;(7)multiple

#

circularholeswithlongitudinalspacing~rangingfrom1.5to5 diametersandtransversespacingsfrom2 to4 diam#t@.s;(8)externalflowvelocity LofO to420feetpersecond;(9)staticressureof externalstream

fapproximately2100poundspersquarefoo ab~olutej(lo)airstreamtemper-atureapproximately75°l?;and(11)pres$uredrOPacrossholeof 2 tO 250poundspersquarefoot.

Thedataforeachconfigurationare!pre8entedasa functionofaflowparameter.Thevarious typesof linerwallopeningsarecomparedanddiscussed.

Ad areaofductcrosssection}sqf%

Af. areaof louveror scoopface~sqm

4 effectiveareaof louveror scoop:face,sq ft

Ah areaof flushopening,sqft

b widthof opening,i%

c dischargecoefficient,ratioofthroughopening

CP dischargecoefficientcorrected

meas=edto theoreticalflow

forpressure-ratio”effect

—

.NACATN 3924 3

4~ 5* dischargecoefficientcorrectedforpressure-ratioeffectsands boundary-layerdisplacementthickness

h heightof louveror scoop,ft

z len@h ofrectangularslot,ft

‘d totalpressureof ductstream,lb/sqf%abs

~ Pf totalpressureof jetstreamatf’aceof louveror scoop,lb/sqft* abs

Pd staticpressureof streaminparallel-wallduct,lb/sqf%abs

pa,2 stiticpressureof streamatwall.openingin convergingduct,lb/sqftabs

+$ PJ staticpressureof jetstream,lb/sqftabs

s qd dynamicpressureofductstream,lb/sqftabs‘Y~ qj dynamicpressureof jetstream,lb/sqftabs

Td totalor stagnationtemperatureofductstream,‘R

‘bZ localvelocityinboundarylayer,ft/sec

Vd velocityof approachstreaminduct,ft/sec

Vj veloc~tyof jetstream,ftjsec

‘m measuredmassflowofairthroughopening,lb/see

‘th theoreticalmassflowof airthroughopening,lb/see

Y distancenormaltoductwall.,ft

a angleof inclinationof convergentductwall,deg

5 boundary-layerthickness,ft

&* boundary-layerdisplacementthickness,ft

e anglebetweendirectionofductflowandfaceofair-entryopening,deg

Pbz massdensityofboundary-layerair,slugs/cuft

Pj mass densityof setair,slugs/cuft

4 NACATN 3924 .

0“APPARATU8‘“

TestSection

Thetestsectionusedintheprese~tinvestigationis identicalto—.

thatdescribedinreference1. Details~of.thetestsectionareshowninfigurel(a)withtestplateflushwiththe.ductwallandinfigurel(b)asmodifiedforboundary-layerbleedoffiTheductheightforall-thepresenttestswas2.23inches.Roomairwasdrawnthroughthetestsectionby meansofthelaboratorylow-~ressureexhaustsystem.Airmass- *flowratesweremeasuredwitha calibrate”square-dgedorifice. 8.

—

Instrumentation

DuctS@tiC ~es8We Pd @ tdd Pr@ssure Pda stationapproximately5/8inchupstre+ofthefaceor oftheleadingedgeof flushtestholesexceptforwalls. Becauseofthesteepstatic-pres~ure.~adientinclinedductwalls,ductstaticpressmkstiththese

weremeasureda% .-of scoopsor louversthoseininclined ..—resultingfrom ●

configurationspd,2 weremeasured-ata pointin-thedubtwalloppositetheopeningasi~cated inthefollowingsketch: ‘

v

B ; ‘“” o

Thepositionof theductstatictapatB~waslocatedbyextendingthelineAC toa pointO ontheoppositewa~ andmakingtheMstanceMequaltoAO. Jetstatic-pressuretapsw&relocatedonthedownstreamfaceofthetestplateas showninfigur+l(a).Thelocationofthejet -static-pressuretapswasnotcriticalintheabsenceofparallelflowonthedownstreamfaceofthetestplate. : =

Air-EntryConfig@ations

Detailsofthe32air-entryconfig~ationsinvestigatedarepresented -intableI. Forpurposesof comparison$he32air-e@ryconfigurationsaredividedintosixgroupsonthebasis:of-geometry:flushrectangular .

● NACATN 3924 5

*hole(seriesA),steplouver(seriesB),thumbnail-typescoop(seriesC),scoopovercircularhole(seriesD),holeor louverininclinedwall(seriesE),andmultiplecticularholes(seriesF). Themetalthicknessforallconfigurationswasapproximately0.040inch.

Sincethesteplouv~ design(seriesB) isoftenusedasa continuousopeningaroundthecircumferenceofa liner,itwasdesiredto evaluatetheside-walleffects oftheexperimentalsteplouver.Side-walleffectsoftheexperimentalsteplouvermaybe causedby lateralflowofairacrosstheplaneofthesidewallswheneverthesetaM ductvelocitiesarenotequal.Inan attemptto evaluatesuchside-walleffectsthesidewallsof someofthe~erimentalsteplouv~swereextended1 inchupstreamoftheplaneof theopeningby a l/32-inch-thickplate(seriesB,tableI)whichformedafence1 inchhigh. TheleadingandupperedgesofthefencesweretaperedontheoutsidesurfacesonQ. Withthesefencesthelateralflowofairacrosstheplaneofthesidewallswouldbe eliminatedunderalloperatingconditions.Thethumbnail-typescoops(seriesC)hadtheformofa segmentofa hemispherewitha radiusof

● 0.50inch.

s Twotypesof scoop-over-holeconfigurations(seriesD)wereinvesti-gated● Inthefirsttypethebaseofthedownstreamhalfofthescoopwasflushwiththeperimeterofthehole(seriesD-1toD-3),thusforminga smoothflowpassagefortheair. In%heothertypethebaseof thescoophada radiustwicethatofthehole.

PmCwuRE

Experiments

DiscWge-coefficientdatawereobtainedforeachofthe32air-entryconfigurationsat ductvelocitiesof O,50,150,a@ 420feetpersecond.At eachductvelocitycondition,theJetvelocitywasvariedupto 650feetpersecond.Theduct-airtotalpressurewasapproxlmtel.yatmos-pheric,andthetemperaturewasapproximately75°F foralltests.

Calculations

ThedischargecoefficientC wascalculatedastheratioofthemeasuredmassflowto thetheoreticalmassflowthroughtheopeningw~wti. Thetheoreticalmassflow wth wascalculatedas theproductofthejetvelocityV~,thejetdensitypj,andtheareaof theopening.

. Thefacearea + wasusedwithstep-andthumbnail-typelouversandsomeof thescoop-over-holeconfigurations,whiletheholearea ~ was

.

6 HACATN 3924 ●

usedwithallscoop-over-holeandflushholeconfigurations. bAssuming

isentropicf~ow,theJetvelocityVj ,andtheJetdensityp~ weredeterminedfromcompressible-flowrelationsutilizingtheducttotalpressurepd andtotaltemperature~d andthejetstaticpressurep~.

RESULTS

Typical D@ta ISdDischarge-coefficientdatatypical’oftwodifferenttypesof liner- G

wallopeningarepresentedinfigure2[flushhole,fig.2(a),andscoop-

over-hole,fig.2(b),configurationsA- andD-1,respectively).Thisfigureshowsthevariationindischarge’coefficientC withstatic-pressureratiopd/p~ at ductvelocit+ Vd ofO,50,150,and420feetpersecond.At-zeroductvelocity~the-dischargecoefficientforbothtypesofopeningvariesonlysligh+lywithpressureratio.At ductvelocitiesotherthanzerothedischarg~coefficientforflushholes(fig.2(a))approacheszeroas the pres$ureratiodecreasestoward1.00,

r

butwiththescoop-over-holeconfiguration(fig.2(b))thedischargecoefficientisata relativelyhighvalqeatpressureratiosinthe B.regionof 1.00andapproacheszeroat valuesof”pd/pj lessthan1.00. “-Also,withtheflushholean increaseixductvelocitydecreasesthedischargecoefficient;whereas,withthescoop-over-holeconfigurationan increaseintheductvelocityincreasesthedischargecoefficient.

Zero-Duct-VelocityData

Dstapresentedin figure3 showthe’variationindischargecoeffi-cientwithpressureratioat zeroductvelocity(zerocrossflaw)for27of theconfigurationstested.Thesedatka&eapplicabletothefinalair-entryopeningina combustorlinerwherealltheairapproachingtheopeningflowsthroughtheopening.At a’premxreratioof 1.02,flushholeshavedischargecoefficientsinthelrai@efromO.59to0.63,steplouversandthumbnailscoopsfrom0.67to 0.79,andscoop-over-holeconfigurationsfrom0.49to0.61.

CorrelationofVelocityData

Themethodof correlationpresented”inreference1 wasetiendedtosatisfythevariousconfigurationsofthepresentinvestigation.Thecorrelationrequireseithertwoortlxree:steps,dependingonthetypeofair-entryconfiguration. _..

.

.

NACAm 3924 7

Flowparameter.- Firstthedatasre.mlottedasa functionofadimensionlessflowParameter(pd- pj)/(pd- pd)swhichistheratioofthedifferencebetweenthetotalandstaticpressuresof thedischargejetto thedifferencebetweenthetotalandstaticmressuresoftheductstream.Forincohqmessibleflowthisparameter-isequalto(V3/Vd)2andto q& Thetypicaldataof figure2 are so replottedin figureA-. Inthisfigurethedataforthevariousductvelocitiestendto forma commoncurve.However,formy givenductvelocitythedatafallabovethiscommoncurveforthehighervaluesoftheflowparameterwheretheSktiC-preSSUre 3XLti0 pd/pJ iS high. ~i.s deviationiS consideredtObe a pressure-ratioeffectsimikrto theincreaseindischargecoeffi-cientwithan increaseinpressureratioshownforthez=o-duct-velocftyconditioninfigure3.

Pressure-ratiocorrection.- Thesecondstepisthecorrectionofdischargecoefficientsforpressure-ratioeffect.To obtaina pressure-ratiocorrectionfactor,thezero-duct-velocitydata(fig.3) for25configurationswererecalculatedas theratioofthedischargecoeffi-cientat a givenpressureratioto thedischargecoefficientat a pres-sureratioof 1.00(C/Cp).Thedatawerethenplottedagainsta generalformof thepressure-ratioterm(pd+ ~ sin8)/pj(fig-5)- For~ atr-entryconfigurationhavingitsfacenormaltothe-directionof ductflow(suchas a scoopor steplouver),19 equals90°,andthegeneralformofthepressure-ratiotermreducesto (pal+qd)/p.Jor pd/pj.Fora flushholeina parallel-walledduct,8 equalsO andthetermreducestoPd/P.~“ Forflushholesin inclinedductwalls,8 equalstheangleofinc&ationof thewall,and ~ equalsthedynamicpressureoftheductstreamat a planethroughthecenterofthehole. Thepressure-ratiocorrectioncurveofreference1 is includedin figure5. Althoughthedataforthevariousconfigurationsshowconsiderablescatterfromthecurveof reference1,thiscurve,whichdoesrepresenta meanvalue,wasusedforcorrectingdischargecoefficientsforallconfigurations.Discharge-coefficientdatawerethencorrectedforpressure-=tioeffectby dividingthedischargecoefficientC by a correctionfactorc/cpdeterminedfromfigure5. (Theparticulardataof fig.4, correctedforpressure-ratioeffect,arepresentedin figs.13(a)and(d),respectively.)

Boundary-layercorrection.- Thedatapresentedinfigures6 and7weretakenspecificallyforevaluatingboundary-~ereffectsonthedischargecoefficientsof steplouversandscoops.Reference1 indicatesthatboundarylayerhasa negligibleeffectondischargecoefficientsofflushholes.In figures6 and7 bothlouverheighth andboundsJT-layerthickness5 oftheductstreamwerevariedindependently.These

8

figuresshowthata variationthicknessmayaffectnotonly

NACATN 3924 .

in either.louverheightorboundary-layer ethedischargecoefficientbutalsothe

valueoftheflowparameterat zeroair@owthroughthelouver.

In an attempt to correct thelouverdataforboundary-layereffects,thedischargecoefficientsof figures6 and7 wererecalculatedonthebasisof effectivefaceareaof louver‘A~ asdeterminedby

~=(Af -b~*). Theboundary-layerdi~pkcementthickness .

/

8 25*= 1 (PJVJ- Pb2vbZ)W a

~svs o

wasdeterminedfrompressureprofiles($’ig.8)measuredatthefaceofthe0.623-inch-highlouverat variousf@w.conditions.Figure8 showsthattheboundary-layerthickmessat thpfaceofa louvervariesnotonlywiththeinitialboundarylayerofthe$uctstreambutalsowiththeflowthroughtheopening.Figure9 showsth+v~iationin 5* withtheflowparameterforthethreeductboundary-tiy~conditionstivestigateda uPressureprofilesshowninfigure8(b)@~ounda@-layer displacement-thicknessvaluesshowninfigure9 for@uctboun@y-1.ayerthicknessof

—

0.10inchapplytoallconfigurationsofthisiuvestigatlonexceptB-11, ‘_B-12,andE-2toE-5. --- .=1 .—

Thedataof figures6 and7, corre@edforpressure-ratioeffectsandboundary-layerdisplacementthiclme$s,arereplottedinfigures10and1.1.,respectively.Thecorrecteddi@umge coefficient~,~* ataflow-parametervalueof 1.0isapproxi@tel.y0.95foralllouverheightsandboundary-layerconditionsinvestigated.At flow-parametervaluesgreatertha,uapproximately4.0,a decre~se-in10UV= heU& increases thedischargecoefficientslightly(fig.10). —

Fldw-parametervaluesat zeroairflowthroughtheopeningareofinterestbecausetheyindicatethel.ow&limitoftheflowrangeforagivenconfiguration.Variationinthis~valueoftheflowparameterforthevariouslouverandscoopconfigurat~onginvestigatedis showninfig-ure12asa functionof louveror scoop~height.AlthoughthedataShOWconsiderablescatterfromthefairedc@ve,a definitetrendis indicated.Theoretically,foranopeninghavingzeroheight(suchasa flushhole)thelowerlimitoftheflowrangewould:beata flowparameterof 1.0;conversely,iftherewereno walleffec$(alouveror scoopdetachedfromthewallandfunctioningasa Pito~tube)thelimitingvalueoftheflowparameterfora louveror scoopwouldbe zero. —

,...CorrelatedData ‘- .

Thedischame-coefficientdatafor:thevariousconfigurationswerecorrectedforpressure-ratiothiclmess,whereapplicable,

effects~ forboundary-layerdisplacementandareplbttedasa functionoftheflow

. NACAm 3924 9

4parameterin figure13. Includ~inthesefiguresaredischargecoeffi.cientsat zeroductvelocitydeterminedby extrapolationoftheapplicablefairedcurvesinfigure3 to a pressureratioof 1.00.Thesedataareplotted(fig.13)atvaluesoftheflowparameterobtainedfromtheapproximaterelation

Pd - Pj ()#2Pd-pd ~

DISCUSSION

RectangularSlots

Effectof slotsize.- Fairedcurvesofthedischargecoefficientsforfiverectangularslots(configurationsA-1toA-5)arecomparedwiththatofa 0.750-inch-diameterflushcircularhole(configurationE-l)infigure14. In general,rectangularslotswiththeirma~ordimension. parallelto thetiectionof flowhavedischargecoefficientsslightlygreaterthanthoseforcircularor squareholes.As withcircularholes

3 (ref.1),holetidthhaslittleeffecton dischargecoefficientsforg widthsgreaterthan0.5inch(compareconfigurationA-2withA-4).

Effectof lengbh-to-widthratio.- Figure14 showsthatan increaseinthelength-to-widthratioofa rectangularslotincreasesthedischargecoefficientslightlythroughouttherangeoftheflowparameter.

StepLouvers

DischargecoefficientsCp,b* forstepl-ouv-s(figs.15to 17)aremaximumata flowparameterof 1.0anddecreasegraduallyat valuesgreaterandshsrplyat valueslessthan1.0. Theminimumvalueoftheflowparameterfora givenconfigurationwasshownin figure12tobe afunctionof louverheighth.

Effectof louverheight.- Fairedcurvesrepresentingdataforsteplouverheightsrangingfrom0.104to 0.623incharecomparedinfigure15.An increasein louverheightmayeitherdecreaseor increasethecorrect~dischargecoefficient,dependingonwhetherthevalueoftheflowparam-eterisgreateror less,respectively,thanapproxi=tely4.0.

Effectof louveroverlapandspacer.- Louverwalloverlaportheuseofa corrugatedspacerwithinthisoverlapis showninfigureM tohaveonlya smalleffectonthecorrecttidischargecoefficientforlouversapproxi~tely0.10and0.25inchhigh.

10 KAC.ATN 3924

Effectof louverwidth.- Steplouv@sareoftendesignedasa con-tinuousopeningaround&e circumference~ofa combustorliner.Inordertodeterminethemagnitudeoftheside-wqlleffectsofthel-inch-wideexperimentalsteplouver,datawereobtainedforlouversbothwithandwithoutside-wallextensions.Extending~thesidewallsupstreamoftheplaneofthelouveropeningwasintended~toprevent’the3ateralflowofairacrosstheplaneofthesidewallsattheopening.ComparisonofdataforconfigurationsB-2andB-3(fig;17)indicatesthatforlouversupto 0.25inchhighside-walleffectsarenegligible.Thedata,there-fore,shouldbe.appl.icableto continuous“steplouvers.

Thunibnail-Typ eScoops

Correcteddischargecoefficientsfoqthumbnail-typescoops(fig.M)aresomeWhatlowerthanthoseforsteplouversat lowvaluesoftheflowparameters.Aswithsteplouvers,thee~fectof scoopheightonthecorrecteddischargecoefficientis small.

z

ScoopsoverCircu@rEoles &

Fourdifferentscoop-over-holeconfigurationswereinvestigated.Forthreeoftheconfigurationsthedowndtr~mhalfofthescoopcoincideswiththeperimeterofthehole,thescoo~sdifferinginfacearearelativeto holearea;forthefourthconfigurationthebaseradiusistwicethatofthehole. Theholediameterwas0.750inchforthefourconfigurations.Althoughdischargecoefficientsareusual~basedonthesmallestflowareaoftheconfiguration,forpurposesof comparisonthecalculationsforconfigurationsD-1andD-2are bas~~onholearea(alargerarea)infigureM andon scoopfhcearea(thes~llestarea)in figure20.

Effectof scoopfacearea.- CorrecteddischmgecoefficientsforscoopsOV= circularholesarecomparedtith_thoseofa flushcircukhole(configurationE-l)infigure19. @se dataarecorrectedforpressure-ratioeffectbutnotforboundsmy-~erdisplacementthickness,sincethedischargecoefficientisbased’onltheholearearatherthanonthefaceareaofthescoop.Thesedata,therefore,arecomparablewiththoseofa flushcircularho&.

Figure19 showsthat increasingscoc@faceareaby increasingscmpheight(configurationsD-1,D-2,andD-3)notonlyincreasesthecorrecteddischargecoefficientthroughouttheflowr~e butextendstheflowrangeto lowervaluesoftheflowparameter.l)his..comparisonindicatesthemagnitudeoftheeffectof scoopfaceareaondischargecoefficientsbasedonflushholearea. Theaddition~f&ri0.891-inch-highsco~”toaflushholeextendstheflowrsmgefroma’flowparameterof1.0(config-urationE-1)to valueslessthan0.05(configurationD-3)becauseofr&m

—.

—--

. .NACATN 3924 —

● pressure(fig.19). At flow-parametervaluesfaceareaofapproximately1.4timestheholeisrequiredtoattaina dischsrgecoefficienthole(configurationE-l).

11

greaterthan20a scooparea(cotiigurationD-3)equaltothatofa flush

Effectofsizeof scoopbase.- Theeffectof increasingthesizeofthebaseof a scoopfromonethatcoincideswiththeperimeterof theholeto onehavinga radiustwicethatofpareconfigurationD-3withD-4),is

,3F throughouttheflowrange.Although+ matelyequalfaceareas,thereduced

junctionwithboundary-layereffectslowvaluesof theflowparameter.

thehole,alsoshowninfigure19 (com-to decreasethedischargecoefficientthesetwoconfigurationshadapproxi-heighto? configurationD-4incon-resultedina reducedflowrangeat

Comparisonwiththumbnail-typescoop.- Becauseofgeometricsimil-arities,thedischargecoefficientsofa thumbnail-typescooparecomparedwithth;seof scoop-aver-holeconfigurationsin figure20. Sincethedis-chargecoefficientsCp,~* in figure20arebasedon scoopfacearea,

+theyarecorrectedfor.bothboundary-layerdisplacementthicknessand

2 pressure-ratioeffect.yg Thecomparisonshowsthatthethuuibnail-typescoophasthehigher

dischargecoefficientthroughoutitsflowrsnge.A studyofthedataindicatesthatthedifferencein dischargecoefficientsforthethreeconfigurationsshownisa functionoftheflushholearea(inthepheof thewall)relativetothescoopfacearea ~~. Theeffectof scoopheightontheflowrangeat lowvaluesoftheflowparameterisagainapparent.

Holesad huverson Inc13medSurfaces

Inmanycombustordesignsthewallsofthelinerandtheoutershellarenotparallel.Theeffectof inclinedwallsonthedischargecoeffi-cientofa flushcircularholeanda steplouveris showninfigure21.

Flushcircularhole.- Correcteddischargecoefficientsfora 0.750-inch-diameterflushholemountedoninclinesof0°,8°,and20°arecomparedinfigure21(a).Dischargecoefficientsarepracticallyunaf-fectedlywallinclinationsupto 2Q0exceptfora smallrameffectatlowflowsthattendsto decreasethevalueoftheflowparameterforthe20°angle.

steplouvers.- Similarly,thedischargecoefficientsof a 0.105-. inch-highsteplouv~ (tithno overlap)muntedoninclinesof 0°,8°,

and20°arecoqaredin figure21(b).Sincethepressureprofilesobtained(fig.8)arenotapplicableto flowina convergentduct,the

12 NACAm 3924 *

dischargecoefficientsinfigure21(b)hrenotcorrectedforboundary. b

layerdisplacementthickness.Also,theeffectsoftheside-wallextensionsforlouversmountedonan in~linewerenotinvestigated.Thedifferencesindischargecoefficientsh@n.forthethreeconfigurationsmaybe partiallyduetovariationsinboundary-layerconditions.However,an increaseininclinationangletendstoincreasetheflowrangeat lowvaluesoftheflowparameter. .—

MultipleFlu8hHoles *4-%Theeffectofproximityofmultiplef~ushholesondischargecoeffi-

cientforbothin-lineandside-by-side,arrangementsispresentedinfig-ure22. Thedatapresentedindicatetheover-alldischarge coefficientfora givenmultiple-holeconfigurationWther thanforindividualholesof thatconfiguration.

. .

Holesinline.- Fairedcurvesrepr~sentingcorrecteddischargeco-efficientsfortwo0.750-inch-diameterf}ushholesspaced3.750and1.125 #3.nchesapart,center-to-center,ina 10

?itudinaldirectionarecompared

withthatofa singleholeinfigure22().__Theresultsindicatethatareductionin longitudinalspacingof cir&darholesfrom5 to 1.5diameters b

hasno effectontheirover-alldischmgecoefficient.In fact,thedataagreewellwiththoseforthesingleholethroughouttheflowrange.

=.Holessideby side.- A reductioniqthetransversespacingof 0.125-

inch-diameterholesfrom4 to 2 diameter+(center-to-center)is showninfigure22(b)tohaveno significanteffeqto“nthecorrecteddischargeco-efficientexceptathigh(above100)valdesoftheflowpsmuneter.

SignificanceofResults

Thedatapresentedshowthevariationindisctigecoefficientwitha flowparameterforvariousconfigurate.os of linerwallopenings.An

l!importantdifferenceinflowCharacteristiciSindicatedbetweentheflushholeandthescoopor steplouver..Withtheflushhole,theflowceasesastheflowparameterdecreasestoa valueof 1.0(i.e.,asthestatic-pressuredifferenceacrosstheope~ngapproacheszero);but,withthescoopandthesteplouver,flowconti~ues(becauseoframpressure)to somelowervalueoftheflowparameter~(dependQgontheheightofthescoop).Combustordesignshavinglowover-alltotal-pressurelossanda highairvelocityinthepassageou~sid,ethelinertendtohavealowor evena negativestatic-pressuredi~ferenceacrosstheupstreamlinerwal.1openings(ref.2). Thehighd$schargecoefficientof scoopsor louversat flow~arameterslessthan1.,0ties themessentialfor .

adequateairadmissionintheupstreamregionof suchliners..

.

.

●

✎ NACATN3924 13

*ApplicationofData

Theresultsofreferences3 and4 indicatethatdischargecoefficientsforholeshavingexternalflowonlyshouldbe applicableto ccmibinedin-ternalandexternalflow,ifthejetvelocityisgreaterthm thetiternalparallelflowvelocityandthecorrectjetoutletstaticpressureisused.Dischargecoefficientsforvariouslinerwallopeningsmsybe detezmdnedfrcmthecorrecteddischargecoefficientsinfigures14to 22 inthisreportorfrcxnapplicablefiguresinreference1 by thefold.owingmethod:

(1) At a givenvalueoftheflowpsrameter(pd- p~)/(pd- Fd)a cor-recteddischargecoefficientCp (or c-p,6*)cm be red frmna curve(figs.14to 22]selectedonthebasisofgecmetricsimil.szity(bothasto shapeandsize)to thegivenlinerwallopening.

(2)!lheccmrecteddischargecoefficient~, 6*f~ stePl~ers ~scoops(figs.15to 18 and20)mustfirstbe reducedto Cp by

.where5*,the boundary-l- tisp~~t t~c~ess~ ~Y ~ est~~dfranf@e 9 forthelocalflowconditim.Thelawerl-t oftheflowrangeforsteplouvers~d scoopsmsybe estimatedfrcmthecurveoffigure12.

(3)A pressure-ratiocorrectionfactorcj~figure5 atthegivenvalueofthepressureratio

(4)Theproductof (C/~)~ thenyieldsthecient C.

SUMMARYal?RESUL’IS

cambe obtainedfrcuu(pd+ ~ sin@/pj●

desireddischargecoeffi-

Thefollow5ngresultswereobtainedfrananevaluationoftheeffectsofvariousgecmetricandflowfactorson dischargecoefficientsforflushrectangularholes~ steplouvers~ ~d scoops:

1.Foreachconfigurationthedischargecoefficients,correctedforpressure-ratioandboundary-layereffects,Whereapplicable,werecorre-latedtitha dimensionlessflm p~ ter.

2.Theeffectsof sizeor shapeof flushrectangularholesondis-. chargecoefficientweresmallccmparedwiththeeffectsof ductstreamvelcwityandstatic-pressmratioacrossthehole.

14 NACATN3924 ●.

3.Theadditionofa SCOOPto a flushholeincreasedthedischarge bcoefficientandextendedtheflowrange&tIbwvaluesoftheflowparem-eterbutdecreasedthedischargecoeffic~entathighvaluesforscoopfaceareaslessthsn1.4timestheholearea.

4.Steplouvers~d scoopshadgreaerdischargecoefficientsaad1widerflowrangesatluwvaluesofthef owparameterthm theflushholes.

However,thedischargecoefficientinth~sregionwasaffectedby boundary-leyerconditionsoftheductstream,whereastheaxtentoftheflowrangeappearedtobe a functionofscoupheight.‘Thehighdischargecoefficients ~forlouverssmdscoupsatlowvaluesoftheflowparemetermakethemessentialfw adequateairadmissionintheupstresmregionofccmbust~ 8linershavinga lowstatic-pressurediff~renceacrosstheiropenings.

5.Forflushholesandsteplounrs~mountedinwallsofa convergentductwall,inclinatimsofupto 20°ha$littleeffectondischargecmffi-Cient.

6.Theproximityofthemultipleflushholesintherangefrom5 to1.5dismeters(center-to-center}h a lo&itudinaldirectionor4 to 2

z

diametersina trsmsversedirectionhaddo sl@.ficanteffectw Mschargecoefficient. *

LewisFlight=opulsicmLaboratoryNationalAdvisoryCommitteefaAermautics

Cleveland,Ohio,Deceniber6,1957

REFEREmm

1.Dittrich,RalphT.,ad Graves,CharlesC.: DischargeCoefficientsforCcmhstor-LinerAti-EntryHoles.I - CircularHoleswithPsmllelFlow.NAC!ATN3663,1956.

2.Grobman,J.S.,Dittrich,R. T,andG~am?s,C. C.: PressureDropandAirFlowDistributicmh GasTurbineCcdmstars.Tram.ASME,vol.79,110. 7,OCt. 1957,pp.1601-1607.:

3.Cdlaghaa,EdnrundE.,md Bowden,D+ T.: InvestigationafFIw CO-efficientofCircular,E@are,ad E,.lli.pticalOrificesatHigh-PressureRatios.NACATN 1947,1949,.

4.Dewey,PaulE.: A PreMuinaryInvesti’gationofAerodynamicCharacter-isticsofSmallInclinedAirOutlets’atTraIsonicMachNunibers.NACAm 3442,1955. (SupersedesNAC~RM L53C20.) .

.

.

TABLE1.- DETAT

OFAIR-ENTRYCONFIG~TIONSINVESTIGATEDAlldimensionsinInches.)

(a)FlushrectangularholesA-1toA-5

A-1

A-3

.

16-.

NACATN 3924 -

L

TASLE1.- CONTINUSD.DETAILSOFAIS-EliTRYCONFW’RATLONSINVESTIGATED(Alldimensionsih inches.i“(b)Steplcuvem d-l tdB-6 —

30Cn

B-1EL4

—8

.

. NACATN 3924

.

.

TABLE1.- coNTrNtlED.DSTAIL9OFAIS-6?FTRYCONFICWFiATIONSS,N’VESTIGATSD(Alldimensionsin inches.)(c)SteplcuversB-7toB-12

—w-lo

B-n

18 NACATN 3924

TABLE 1. - CONTINUED. DWJ?AIL9OF AIR-ENTRYCONFIGURATIONSINVESTIGATED(Alldimensionsin Inches.)

(d) Thumbnail-typelouversC-1 andC-2 at+dscoopsover clrnularholesD-1 to D-4

D:2

.

k

gi

Cn

.

.

—

19

TASL8I. - Co?wm. DSTAIL9W AIS.SXTRYCONFIGURATIONSINVSSTIGATSD(Alldlmene.iomin inches.)

(e)Holeor louverin inclinedwallI&l to E-5-.

@mlw

.

20 NACATN3924

TABLE1. -“CONCLUDED. DETAILSOF AIR-*Y c0mI13mATIoNsINVESTIGATED(Alldlrnenslons1+ Inclae.e.)

(f)Multiplecircularholes”~-1to F-4 —

.

.

.

.

—.

* [ 4705 , ,r

/“-\k:?ic tap.

4-Inah-wJlwe duutJ’-Cca-tki.atimsfl”h”.w—-.hot t+tal-presmm pmh

Q >.

% “-%‘:~~

. .mob“AOt&tiq$riiOr&teT

,.,“---.\ :,,’.:

.>>.,.:.:>,: ‘“:. .. .-% , .:.“.,..:$;:$} . :::,;.. .

Iklct“S’t.atio-wens!ueta?JB.... .,>,?. .,.. . -+..

1. 1

(a) Test plete flush with duct wall.

Figme 1. - D3tall,aof teat appatw for stu@yof disohergeooefflc lents.

Jw’Ym@EKi&,i

. .tii

,,..

..,: ,’,,; .

,.. d. ..=-+ -., ‘>+.J ,jij=’?,..

... .. .,:,, .,,,,:

., a

\\ L

,:.

{ .,., .”””

.\\w,.-...

(b) HMfiled far boumky-@er blsedoff.

Figuro 1. - Concluded. IM.911sof test a~tus fm studyof OXso?wrgecoafficlemtm.

u

* , r 4705” #

H+lllN/” ) I I I I

A A

.8

c

// -

1

K— = — — P

/ ~ - —/ -“

* ~

.2Y-

1 H--i-

1/./~.

0(.) Flwh ewe lmle (conflguratlonA-1).

/

/ .MllllllI I Ill\/--l Illlllriliil II

.2

/

o ! I.92 ,94 .96 ,98 “ ccl) 1.02 1.04 1.06 1.08 1.10 1.12 1.14 1.16 1.18

fkStiC-PM!3i3UIY rddo) pd/pj(b) SCWP over circulerhole (mnfig.u’aticmD-1).

Mgua’c 2. - Effect of static-prWm7re ratio on Mache.rgemeff’lcientat varlOuE duct velocities.

I is F 420

I I I I

A

.

.

.6

.6

.51.00 1.04 1.08 1.12 1.16 1.W 1.04 1.08 1.12 1.16 1.00 1.04 1.08 1.12 1.16

StatiO.presmn.e ratio, r+y% j

(a) Flush holes. (b) Step lcuvers and thumbnail8COOPB .

Pigllre 3. - Vamiatim in msoh.arge coefficients of 27 configurationsvelocity .

(.) Socmps over holes.

with pressure ratio at zero chmt

rid? , ‘ “

v CG4 , 4705 * v

u

(a) Blush wuare hole (oonfigmmtionA-1).

“~ p~~”’e”~~(b) Somp over oir.ularhole (ootilg!matimD-1).

mmm 4. - Va@latlOnin ~~chmge ooeffioientwith flow parawt.erat vnricw duet velooitleo.N(n

.-

I I I I I I I

CmfMMtion Ckmf@ration

0 A-1 0 c-lh A-2 0 C-20 A-3 n P-1

A-4 D P2: :: D-s

: D-4P B2 o E-1

v B-3 D E-2

1.10 A B-4 Q %3

v 2-5 a B-4A %6 o E-sP 2-7 F-1

z F-2

I.(2C2 / ‘

AJ

1.06

,..-...,-- .-- .-—. ,--..

()A

b

0Ud

i.w

7

02 s1.02 A

[1 0

1.00 1.M 1.02 1.12 1.16 l.m 1.24 l.za 1.32 1pd+~sine

Pres.wre-ratio tmrm,P$

Figure5. - RwmIre-ratio con-action factor fur discharge maffid~,

56

I . .*

t CUd bam..-.

4705 t

1.

.1 .2 .4 .6 ..9 1 8 10

b :_te,, @ 6

20 40 60 80 UN

d - Pa

Figure6. - ComDeri@nnof Uactige cnefficienimfor three heights of step louvem. LMCt boundary layer, 0.10 inch thick.Discharge coefficientsnot corrected fur bnundary-layer effects.

I. . .

Nal

I I) 1/

,0/

.6 i I

I\

I I I I 1 — ....S3. \ i i 17Ill

I t

HNIT \ I?l’i’ii!!!I

Hll+

, # 4705 “-; ,

o- .1 .2 .s .4 .5 .0 .7

t1 I ! 1 1 1 1 1 1 I 1 1 , !

Imot w lm:.@, 15ofeatv= aemn d

o .1 .’2 .3 ,4 .5 .6 .1Distanoefrm duetwall,in.

o ,1 .2 .3 .4 ,6 .6 .7

(a) rut bo.mdorg law, 0.04 1.* thiOk.

Pigurs.9.- RWmm profilesmmurecl at ram of O.623-inah-hi~steplouverat varlouMf1- ocmditi~.

N(9

1I 1

Lhot valmity, 50 feet per moond

1 I I I I I I I I I I Io .1 .2 .3 .4 .5 .6 .7

14111111111111DmtValooity,150 feetpm Ll,cond

o .1 .2 .3 .4 .5 ..2 .7Distanoe I’rcm dnot w.9.U,in.

.

0 .1 .2 .3 .4 .s .8 .7

(b)DIct bcurdary Iay*r, O.10 inch thi.k.

Ur.WO ~. . ccminuea. Premura LWOfileS measured at face of O .623-lmh-h@h step 10UWF at varicm fka comtims.

l,, .,,

I

,,. l,’

* mLTf;, .,. .118,8 :)! ,,

CNo

!2

.

!

111171111111111mm’ ‘C.dty;’Mb At ‘p& P.&h

o .1 .2 .s .4 .5 .6 .7Dinttioe from duet wall, in.

(o) ~at bundary lwer, O.XI inch thlok.

4705

, I , , , I , 1 1 , 1

I I I I I I I I I I I

I i I L I L I L I1345

h-i- 1-1 L I L I L I1146 I

, ,

0 .1 .2 .3 .4 .5 .6 ,7

~ a. - Concluded.Prmoum PI.OM1OBmummed at faoeof O.623-inch-hi@steplouverat mriaua fleaoonditioon.

.,

32 .

.

.

-. 40 60 eolooFlowparameter;~

%oCn

Figure9.- Vmiattoninboundary-layerdisplacPen~~hicknessat faceof louverulthflow

parameter forthreeductboundary-layercondilone.“

.

1.

is

\II

1

.-

— .-P. - n.

hwe ~. - COmparimm of d!acharge cmfficienta forthee valuesof duct boumdmy-lmer thidmem. S@ louver, O. 62Sinchhigh. Eiwharge cuefficientacorrected for Fessw’e-mtio effects ad bomdary-lwer displacecmnt thickusa. ;“

1’b

I ., 1.

b 1 sol!? I .

I ,1.

.

* CU-5badk 4705 !

1.0 1 1 1 1 I I

configuration Configuration

hQ B-1 o E-10D %2 o B-n

.8v B-3 o B-12A B-4 o c-lw B-5 oA B-6 n

\ v B-7 DA B-8 v

} b o B-9 a.6 \

n b L,4

QDuctvelocity,

Vs.)ft/sec

\●+

.2- Open 50 \Tailup 150Eblid 420 0

c-2D-1D-2D-3D-4

o .1 .2 .3 .4 .5 .6P~ - pj

‘ow ‘mmeter’ pd - pd

Figure12.- Variationof flow-paxemetervalueaat zeroflowthroughopeningwithEcooporlouverheight.

I

36

L?

NACATN 3924 .

Figure13.-

(a)Flushrectangtil=-holes.

Varlati.onof correcteddlsch~ge“coefficientwithflow

.-

parameter.

.

—

.--

.—

,.—

,

I —.

.

NACATN392A 37

##

v v

40 I I I J=llll~ I I #l I I Illk%# .4

8

.3

.2

.1

01 2 4 6 8 10 20 Lo 60 do 100

215

Flawp= ““~(a)Concluded.Flu8hreotewlarholes.

-e 13.- Cmtlnued. Variationorcorrecteddischargecoerf’lclentwithflowparameter.

38 NACATN3924 .

133

1.

27

,4,180

‘1”‘“-t”’ *(b)Steplouveps.

I,750

.

.

Figure13.- Continued.Vartationof corrqcted.dischargecoefficientwithflowparameter. :

—

9 I4705 8 4

a1..

600

1.

00

‘1~ p@J’meter)~(b)Conaluded,Step10UVCrE.

Figura13.- Continued.Variationof oorremteddioohargecmffIolentwithflewparameter. (.Nm

I

40 NACATN 3924

1.

28,380

.

1‘

5780

.

“w ‘=meter’ ~-”

(c)Thumbnail-type’scoops.

Figure13. - Continue~. Variationof correcteddischargecoefficientwith flowparameter.

—

$jUI

.

.

.

I

mm -Oter, ~

(d) doocw over airmlar holcm.

Ww’e 1s.- Cmtlnmd. ‘fariatim of o-tad dimhnrge ocmffloinn% wibhClm v*,

42 NAC!ATN 3924 ●

.,1s0

....

$jCn

,llW

1103

n- Pumtal-,“*-- .(e)Flushholeor steplouvedon tnclinedwall.

Figure13, - Continued.Variationof cotirecteddischargecoefficientwithflowparameter.

P

-..

—

—.

.

,.

NACATN3924

Ino5 1.

,260

.

1.0

..9,000 ,

.6

.4

.2

0.2 .4 .6 .a1 20 40 60 601WJ

‘f” ‘“”’~’” &g

(e)Oonaluded.Flushholeor steplouveron inclinedwall.

.

Figure13.- Continued.Variationof correcteddi8chargeooefficientuithflowparameter.

.

I

.6

,6

.4

.5

.2

M“”rt/mo

.1ml dmu o

J_c.’mml up

r solid1%420

0b

.1

.6

1 24 681OZO Wmmlm m m 1 04 Ll alo -xl 4e .mmux 5c0 ue

w,- 9mawta-,~

(f) Multlple circular holes.:1

Figure 13. - Concluded, Verlatlon of oorrected discharge coefficient with flm parametxzr.

I , < 1 4705 I t

Rlgore 14. - Effect of fluuh ractanaularimle dlmsn.sioneon correcteddiwhzwa cnofficient. (@rve for oircul.mflnehhole includedfor ccmparlmn.)

I&.rn

l——

4=--—

6

4

-. .,..

2 /

/

0 /.1 .2

1.0

Flowwr.ter, ~

Figure1.5.- EPfect of step louver k%ht on corrected dimharge coefficlmt.

I

1 ,

II 1

01

s.

#

E-—1 IL 1

I1 I I I I

I

.6

.4

.Z

0EE.4 .6

* , 4705 a

.8 1 2 4 6 -8 10 20 4060 80 100—

Flow parameter, ~

(a) 6tep louver height, 0.10inch.

Figure 16. - Effect of Btep louver overlap and corrugated spacer within the overlap on correcteddischarge coefficient.

—..- ..

1

f

.

.4 .6 .8 1 2 4 6

1 #

1 1 I t 1 I I

Conflguratlon

—— E-3—-— B-5

E-.

10 20

—“

—

40 60 80 103

L&.

P~ - PjFlawWameter, ~d- ~d

(b)Step lower height, 0.25 inch.

Figure 16. - Concluded. Effect of step louver overlap and corrugated 8pacer wlthln the overlap w

on corrected discharge coefficient. i

:, I

. . ● rSO,l.v

1 CU-7 ‘ 4705 ‘

1.0

.8

.6

.4

.2

0.4 .6 .8 1 2 4 6 8 10 20 40 60 80 100

pa - Pj

‘lOw ‘met=’ Pd - Pd

Figure 17. - Effect of side-well exkmiona for Blq louvers on corrected discharge coefficient.Step louver height, 0.25 Inch. IF.

m

I I

.

1.0

.8

.6

.4

. .

:,

. .2

0.

—

/7

.6 1 2 4 6 8 10 20 40 En 80 100g

Figure 18. - Effect of height of thumbnsil-type scoops on corrected discherge coefficient.

SO.L7!’1

r ‘%,’ ,“

I I

n

.

CU*7lack 4705 “

J4 1.0

..9

.6

.4

.2

0.01 .02 .04 ,06 .08 .1 .2 .4 .6 .81 _ 2 46 8 10 20 40 m 50100

Flow parawt.,;=

Figure 19. - Effectof Bizeof scoopnwunttioverflushcircularbole on mrrected clisc=ge coefficient.Mscharge coaf-flcientserebesedon flushW area.

WP

1

rnRi

1

Fi@re 20. - CmpariOOn of dischergecoefficientsfor =aOp-over-tiletiththmb~l-m Bcoop. DiecharsecOefficietibasedon sceapface area.

‘1 I

x w S0,L7 . ,I

“

.

NACATN 3924 53

.6Configuration

—--- — E-2—-— E-3

.5

A’

.4

t

.3

.2

KI

.1

i’ttibt-t m-i--H-o~.8 1 2

<2’‘1

Qx

K/.,5 MA..

- 11

4 6 8 10 20 40 w 80 100

Flow parameter,Pa - pJpd - Pa,2

(a)Flush circularhole tith 0.75-inchdhneter.

Figure 21. - Effectof well inclinationon correcteddischargecoefficient.

+--u2kal

.

.2 .4 .6 .8 1 2 4

!!!”,,~

6 8 10 20 40E41 W 100

% - Pj

?FI.OVpa==ter, Pd . ~d,~

s(b) kWp louver 0.10 inch high.

Q

Figure 21..- Concluded. Effect of wdl inclination on corrected discharge coefficient. CN

$

P , SOL+ ‘ ‘,.,

, , ,4705

1 2 4 6 8 10 20 4060 so 100 zoo 4(X

324‘ow ~metir’ Pfl- p~

(a)Holes In line. (Coufigoration E-1 includedfor ccawarieon.)

Figure22. - ~ect of proximityof multiplelmles on correcteddiechargecoefficient.

.7

.E

.5

.4

.3

.1

.1

c

=TrrrtT

/

m.G tm+mt---1----- - -- A. -------- ----- ..

I 1 1 1 1 1 1 I

I I I I I 1 I I

2 4 6 8 10 20 40 EJ2 m lm m 4

m.w Wmm=ter, ~

(b)Eole66i.5eby aide.

-22. - CmcMd. Effectof rn-=iti~ or ~ti~e -m on mrrmted M3cher.secoefficient.

, J

I 1’, :, .

II 1,

0

SOL?

mm

. ,