DOC

Effects_0

By Veronica Sullivan,2014-07-18 02:25
7 views 0
Effects_0

    Effects

JournalofThermalScienceVo1.17.No.1(2008)l6

    ;ArticleII):1003-2169(2008)01.0001-06

    ;EffectsofaKindofNon-smoothBladeontheUnsteadyFlowFieldattheExitof ;anAxialFan

    ;HongweiMAJiboGUo

    ;SchoolofJetPropulsion,BeijingUniversityofAeronauticsandAstronautics,Beijing100083,China

    ;NationalKeyLaboratoryonAeroEngines,Beijing,100083,China

    ;Email:mahw@buaa.edu.cn

    ;Anexperimentalinvestigationofeffectsofakindofstreamwisegroovedbl

    adeontheunsteadyflowfieldatan

    ;exitofanaxialflowfanwasperformed.Tlleflowfieldat25%chorddownstreamfromthetrailingedgeathub

    ;wasmeasuredusingafastresponsefive-holepressureprobeatdifferentmassflowconditions.Tlleunsteadyflow

    ;ofthegroovedbladeswascomparedwiththatofthesmoothblades.Tllemeasurementresultsindicatethat:(1)

    ;thegroovedbladesrestrainthevelocityfluctuationandthepressurefluctuationbymodulatingthebladeboundary

    ;layers,whichcontributestotheflowlossreductioninthehubregionandintherotorwakeregionatthedesign

    ;condition;(2)thestream-wisegroovesplayanimportantroleinrestrainingtheradialmigrationintheblade

    ;boundarylayerandabatingthetipflowmixing,whichcontributestotheflowlossreductioninthetipregionat

    ;thedesigncondition;(3)atthenearstagcondition,thegroovedsurfacecannotreducetheflowloss.evenin

    ;creasethelossnearbywhentheseparationhappensinthebladeboundarylaye~

    ;Keywords:axialfan;non-smoothblade;stream-wisegroove;unsteadyflow ;Introduction

    ;niswellknownthatakindofnonsmoothsurface

    ;withmicrogroovesknownasribletscanreduceviscous

    ;dra2I”.Ribletsweretestedtoreducebladeprofilelossin

    ;compressorlinearcascadeswiththebladescoveredwith

    ;ribletfilmsand6%-10%oflossreductionwasob

    ;tained.-41.However,fewtestswerereportedonrotating

    ;rotorblades.Theauthorscarriedoutanexperimental

    ;investigationofeffectsofakindofstreamwise.groove

    ;bladeontheperformanceofanaxialflowfanL.Thetest

    ;resultsindicatedthatthegroovedbladesurfacesincrease

    ;massflowofthefan,reducetherelativetotalpressure ;lossandincreasetotalpressure15seofthefanatthetest ;massflowconditionsexceptthenearstallcondition.

    reductionmechanismwasnotvery ;However,theloss

    ;Received:September24,2007

    ;HongweiMR”Professor

    ;clear.Thispaperpresentstheunsteadymeasurementre

    ;suitsandinvestigatestheinfluenceofgrooved ;blade-surfaceontheunsteadyflowfieldintheaxialfan ;tohelptounderstandthelossreductionmechanism.

    ;ExperimentalFacilityandTestTechnique

    ;Thetestswereconductedinanaxialfantestrig.Ref- ;erence5hasgivenanintroductionaboutthefaciHty ;andtesttechnique.Theouterdiameterofthefanflow ;passageis0.355meter,withthehubtotipratioof0.49.

    ;TherotorhasthreebladeswiththeheightOf85milland ;thechordof240mn-latthetip.Therotortipclearanceis ;5mil1.about2.1%chord.Thedesignflowrateofthefan ;is91mS/min,attherotationspeedof2900l/min. ;Thenonsmoothbladeswerecoveredwithstream

    ;www.springerlink.com

    ;

;2J.Therm.Sci.,Vo1.17,No.1,2008

    ;wisegroovefilm.Reference[5hasdescribedthegroove

    surfaceindetails.Thethicknessofthegroovefilm ;blade

    ;is0.2inlnandtheheightofthegrooveis0.08mm.The ;spacingfortheribletsis0.75i/lin.

    ;Theflowfieldat25%chorddownstreamfromthe ;trailingedgeathubwasmeasuredusingafast-response ;fiveholepressureprobe.Thepressureprobehasaconi

    ;calheadwithadiameterof6inln.Itsmeasurementan

    ;certaintiesofanglesarelessthan1degreewhilethean

    ;certaintyoftotalpressureislessthan0.5%. ;Themeasurementsweretakenatfivemass..flowcon.. ;ditions,fromdesignconditiontonear-stallcondition.The ;flowfieldsatfourteenpositionsweremeasuredfromthe ;hubtothetip.Inordertoobtainaccurateresults,theflow ;fieldateachstationwasmeasuredfor400periods.500 ;pointsatthesamplespeedOf2000oHz.Sometestswere ;repeatedtoensurethetestresultswererepeatable.In ;ordertoinvestigatetheinfluenceofnonsmoothblades

    ;ontheflowfield,thesamemeasurementsweretaken ;usingthesame5holepressureprobeatthesarlleaxial ;planeandatthesameboundaryconditionsforboth

;smoothbladesandnonsmoothblades.

    ;“restresultsatthedesigncondition

    ;Fig.1showsthedistributionsofrelativetotalpressure ;losscoecient(?Pwt,O.5pUt2)attheexitsofbothsmooth

    smoothrotorpassagesatthedesigncondition. ;andnon

    ;TheflowlossesmainlyOCCurintheendwallandwake ;regions.Thelossinthehubregionresultsfromflow ;mixingamongthehubboundarylayer,thepassagevor- ;texandthebladeboundarylayers.Thewakelosscomes ;fromthemixingbetweentheboundarylayersonthego

    ;torbladesurface.Duetotheradialmigrationinthe ;boundarylayeronthesuctionsurface.moreandmore ;lowenergyfluidsaretransportedtothecomerofthe ;casingwallandthesuctionsurface.Thewakebecomes ;widerandthelossincreaseswiththespan.Theflow ;mixingbetweentheboundarylayersonthesuctionsag- ;faceandonthecasingwallcausessubstantialflowloss ;whilethetipleakageflowenhancestheinteractioninthe ;tipregionleadingtothehighestlossnearby. ;(a)smoothfan

    ;(1))nonsmoothfan

    ;Fig.1Distributionsoflosscoefficient(?Pwt,O.5pUt2)atthe

;rotorexitinthedesigncondition

    ;Comparedwiththemeasurementresultsinthesmooth ;fan,Fig.1(b)indicatesthatthenon-smoothbladesag- ;facesreducetherelativetotalpressureloss.Thehubre

    ;gionofhighlossbecomesthinwhilethehighestlevelof ;lossgoesdown.Thebladewakesbecomenarrowwhile ;thelossesarereduced.Itismorenotablenearthesuction ;side.Theimprovementshouldresultfromthemodulation ;ofthenon-smoothbladesurfaceonthebladeboundary ;layers.Themeasurementresultsnearthesuctionsurface ;inReference[4indicatedthatsuchanonsmoothsurface

    ;canreducethethicknessoftheboundarylayerwhilethe ;streamwisegroovescouldrestraintheradialmigration ;intheboundarylayer.Bothactionscanreducetheflow ;撼撼三兰

    ;

    ;

    ;4

    ;wherearerespectivelycorrespondingwimtheregionsof ;highloss.

    ;Inthemeasurementtipregion,theaxialcomponentis ;foundtobethelargestofthethreeturbulenceintensities,

    ;andtheradialcomponentsecond,whilethetangential ;componentisthelowest.Thevariationsconfirmthatthe ;tipleakageflowplaysaleadingroleincausingflow ;fluctuationandenergydissipationinthetipregion. ;Inthehubregion,theaxialandtangentialturbulenee ;intensifieshavehighlevelwhiletheradialcomponentis ;thelowest.Thevariationsindicatethattheflowfluctua. ;tionandlossmainlyresultfrominteractionofboundary ;layersonthehubandontheblade.Thetangentialmigra. ;tionoflow.energyfluidsintheboundarylayeronthe ;hubmayplayallimportantroleincausingflowfluctua. ;tionandenergydissipationinthehubregion. ;Inthewakeregion.thelevelofaxialturbulenceinten. ;sitvisthehighestwhilethelevelofradialcomponentis ;thelowest.

    ;Comparedwiththemeasurementresultsinthesmooth ;fan,thetestdatainthenon-smoothfanshowninFig.2 ;indicatesthatthelevelsofturbuleneeintensitiesinthe ;mainflowpassage,especiallythetangentialandradial ;components,alenotablylowerthanthoseinthe ;smooth.bladepassage.Theregionsoflowfluctuarion ;enlargeinthemainflowpassage.Inthemeantime,the

    ;hubregionsofl1ighfluctuationbecomeslim.Thecorn. ;prehensivevariationsareshowninFig.3.Theturbulence ;Ca)smoothfan

    ;(b111011-smoothfan

    ;Fig.3Distributionsofturbulencekineticenergyattherotor ;exitatthedesigncondition

    ;J.Therm.Sci.,Vo1.17.No.1,2008

    ;kineticenergyisnormalizedby0.5oUtz.Fig.3showsthat ;theturbulencekineticenergyinthetipregionofthe ;non.smoothfanpassageisobviouslylowerthanmatin ;thesmoothpassage.Thesevariationsconfirmthatthe ;stream..wisegroovesurfacecanrestraintheflowfluctua.. ;tioninthefanbymodulatingtheboundarylayeronthe ;bladesurfacesatthedesigncondition.

    ;Fig.4showsthedistributionsofthefluctuationstatic ;pressurecoefficient(APs/0.5pUtz)attheexitofthefan ;rotorpassage.TIlefluctuationofpressurefAPs)were ;obtainedbythefollowingformulas:

    ;?Ps=(8)

    ;Ps=?f(9)N

    ;whereN--400

    ;Fig.4indicatesthatthepressurefluctuationinthe

    ;non-smoothbladepassageislowerthanthatinthe ;smoothone.Thismeansthatthenon.smoothbladesur. ;facecanreducetheaerodynamicnoiseinthefan. ;(a)smoothfan

    ;Co)non-smoothfan

    ;Fig.4Distributionsofsmilepressurefluctuationattherotorexit ;atthedesigncondition

    ;Testresultsatthenearstallcondition

    ;Fig.5showsthedistributionsofrelativet0tajpressure ;詈薹88吾售

    ;0000000000000000

    ;0?拿已抖兹伯博?,?????????

    ;

    ;HongweiMAeta1.EffectsofaKindofNon-smoothBladeontheUnsteadyFlo

    wFieldattheExitofanAxialFan

    ;l0ssattheexitsofbothsmoothandnonsmoothblade

    ;passagesatthenearstallcondition,whichareobviously ;difierentfromthoseatthedesigncondition.Withthe ;massflowdecreasing,theflowlossesinthefanincrease. ;Theregionofhiglllossenlarges,occupyingnearly ;half-spanofthepassage.Theresultsindicatethatthe ;flowlossinthehubregionofthenonsmoothrotoris

    ;lighflylowerthanthatofthesmoothrotor.Ontheother ;hand,theflowlossesinthemainflowpassageandthetip

    smoothrotorarcnotablyhigherthan ;regionofthenon

    ;thoseofthesmoothrotor.

    ;(a)smoothfan

    ;(b)nonsmoothfan

    ;Fig.$Distributionsofrelativetotalpressureloss(AP~/0.5 ;pUt2)attherotorexitatthenear-stallcondition ;Atthenearstallcondition,theflowseparationhap. ;pensintheboundarylayeronthesuctionsurface.Fig.6 ;showsthedistributionofaxialvelocityattherotorexitat ;thenearstallcondition.Thesubstantialflowblockage ;happensinthetipregion,whichshouldresultfromthe ;interactionamongtheflowseparation,thetipleakage ;flowandboundarylayersonbothbladesurfacesandon ;thecasingwal1.

    ;Measurementresultsinthecascadeindicatedthat ;thegroovedsurfacecannotreducetheflowloss,even ;(a)smoothtan

    ;(b)non-smoothfan

    ;5

    ;Fig.6Distributionsofaxialvelocitycoefficient(W#U0atthe

Report this document

For any questions or suggestions please email
cust-service@docsford.com