Jet-installation noise reduction with flow-permeable materials

Delft University of Technology

Jet-installation noise reduction with flow-permeable materials

Rego, Leandro; Ragni, Daniele; Avallone, Francesco; Casalino, Damiano; Zamponi, Riccardo; Schram,

Christophe

DOI

10.1016/j.jsv.2021.115959

Publication date

2021

Document Version

Final published version

Published in

Journal of Sound and Vibration

Citation (APA)

Rego, L., Ragni, D., Avallone, F., Casalino, D., Zamponi, R., & Schram, C. (2021). Jet-installation noise

reduction with flow-permeable materials. Journal of Sound and Vibration, 498, [115959].

https://doi.org/10.1016/j.jsv.2021.115959

Important note

To cite this publication, please use the final published version (if applicable).

Please check the document version above.

Copyright

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons. Takedown policy

Please contact us and provide details if you believe this document breaches copyrights. We will remove access to the work immediately and investigate your claim.

This work is downloaded from Delft University of Technology.

ContentslistsavailableatScienceDirect

Journal

of

Sound

and

Vibration

journalhomepage:www.elsevier.com/locate/jsv

Jet-installation

noise

reduction

with

flow-permeable

materials

Leandro

Rego

a ,∗

_,

_Daniele

_Ragni

a

_,

_Francesco

_Avallone

a

_,

_Damiano

_Casalino

a

_,

Riccardo

Zamponi

b

_,

_Christophe

_Schram

b

a Delft University of Technology, Department of Aerodynamics, Wind Energy and Propulsion, Kluyverweg 1, 2629 HS, Delft, the

Netherlands

b von Kármán Institute for Fluid Dynamics, Environmental & Applied Fluid Dynamics Department, Waterloosesteenweg 72,

Sint-Genesius-Rode 1640, Belgium

a

r

t

i

c

l

e

i

n

f

o

Article history: Received 16 October 2019 Revised 31 August 2020 Accepted 16 January 2021 Available online 22 January 2021

Keywords:

Aeroacoustics Jet-installation noise Porous materials

a

b

s

t

r

a

c

t

Thispaperinvestigatestheapplication offlow-permeablematerialsas asolution for re-ducingjet-installationnoise.Experimentsarecarriedoutwith aflat plateplacedinthe nearfieldofasingle-streamsubsonicjet.Theflatplateismodularandthesolidsurface near the trailingedgecan bereplaced withdifferent flow-permeableinserts, suchas a metalfoamandaperforatedplatestructure.Thetime-averagedjetflowfieldis character-izedthroughplanarPIVmeasurementsatthreedifferentvelocities(Ma=0.3,Ma=0.5and

Ma=0.8,whereMaistheacousticMachnumber),whereastheacousticfar-fieldis

mea-suredwith amicrophonearc-array.Acousticmeasurements confirmthatinstallation ef-fectscausesignificantnoiseincrease,upto17dBforthelowestjetvelocity,particularlyat lowandmidfrequencies(i.e.St<0.7,withtheStrouhalnumberbasedonthejetdiameter andvelocity),andmostlyintheupstreamdirectionofthejet.Byreplacingthesolid trail-ingedgewiththemetalfoam,noiseabatementofupto9dBisachievedatthespectral peakforMa=0.3and apolarangleθ=40◦,withanoverallreductionintheentire

fre-quencyrangewherejet-installationnoiseisdominant.Theperforatedplateprovideslower noisereductionthanthemetalfoam(7dBatthespectralpeakforMa=0.3andθ=40◦),

anditislesseffectiveatlowfrequencies.Thisisrelatedtothevaluesofpermeabilityand form coefficientofthe materials,whicharethe major parameterscontrolling the pres-surebalanceacrossthetrailingedgeand,consequently,thenoisegeneratedbytheplate. However,despitehavingahighpermeability,theplatewiththemetal-foamtrailingedge stillhasadistinctnoiseproductionatmidfrequencies(St≈ 0.43forMa=0.3).Basedon

theanalysesofdifferenttreatedsurfacelengths,itisconjecturedthatthesolid-permeable junctionintheplateactsasanewscatteringregion,andthusitspositionalsoaffectsthe far-field noise,whichisinlinewith analyticalpredictionsintheliterature.Nonetheless, bothtypesofinsertsprovidesignificantnoisereductionandarepotentialsolutionsforthe problemofjet-installationnoise.

© 2021TheAuthors.PublishedbyElsevierLtd. ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/)

∗_{Corresponding author.}

E-mail address: l.rego@tudelft.nl (L. Rego).

https://doi.org/10.1016/j.jsv.2021.115959

0022-460X/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ )

1. Introduction

Jet-installation noise (JIN)arisesfromthe interactionbetweenthe exhaustflow ofan aircraftjetengine anda nearby airframe surface[1,2] .Ithas beenshownthatthe presenceofsolid boundariesinthevicinity ofa jetflow highlyaffects theacousticcharacteristicsofthesystem.Forthecaseofaninstalledsubsonicjet,asidefromtheturbulence-mixingnoise component, thereis an additionalsource atthe surfacetrailing edge[3] .This region actsasa singularity forconvecting hydrodynamicpressurewaves,whicharegeneratedbythejetmixing-layer,resultingintheirscatteringasnoise.This phe-nomenonisresponsiblefornoiseincrease atlow andmidfrequencies,particularlyinthedirectionsnormalandupstream ofthejetaxis,equivalenttoadistributionofacousticdipolesatthetrailingedge[4–6] .Installationeffectsarealso respon-sibleforreflectionofacousticwavesgeneratedbyquadrupolesourcesinthejetplume[7] .Recentcomputationalresultsof aircraftacousticfootprinthaveshownthatinstallationeffectsareresponsibleforpenaltiesofapproximately4EPNdBatfull aircraftlevel[8] .Inthenearfuture,withthedevelopmentofultra-highbypassratioengines,thissourcewilllikelybecome moredominantduetoincreasedproximitybetweenengineandairframe[9] .Therefore,thedevelopmentofnoisereduction solutionsforthisparticularsourceisofinterest.

Attempts toreduceJINwere performedbyMengle etal.[10] throughtheuseofchevronnozzles.Windtunneltestsof ajet-flapconfigurationwithdifferentnozzlegeometrieswerecarriedout.Theresults,obtainedfrombeamformingapplied tophased-arraymicrophonemeasurements,showedthatthechevronsreducednoiseupto2.6dBatthespectralpeak[10] . However,althoughthechevronsprovidedsomebenefits,theywerestillnotsufficienttobringthenoiseoftheconfiguration backtoisolatedjetlevels,andthusthetrailing-edgesourcewasstillpresent.

A possiblesolution forJIN reductionis theapplicationof porousmaterials on thescatteringsurface. Similar concepts have beeninvestigated,butfordifferentapplications.Revelletal.[11] reportedflapside-edge noisemitigation whenthe surfacewastreatedwithporousmaterials. Itwasconjectured thatthereductionsinsurfacepressurefluctuationsand far-field soundwereattributedtodissipationofacousticwaves,flowcommunicationbetweenupperandlowerside(reducing thepressure imbalancebetweenthem), andlower impedanceofthesurfaceinteractingwiththevortex[11] .Experiments performedbySarradj andGeyer[12] showedthata fullyporousSD7003airfoilreducesturbulentboundary-layer trailing-edge(TBL-TE)noiseupto10dBcomparedtothesolidcase.Thepropertiesofthematerial,suchasporosityandresistivity, werefoundtoaffectthefrequencyrangeandamplitudeofnoisereduction[12] .However,suchreductionwascoupledwith a significant lossinperformance (liftdecrease anddragincrease).In ordertomitigatethesenegativeeffects,the authors restrictedtheporoussectionoftheairfoiltothetrailing-edgeregion[13] .Theshortestporouslength(last5%oftheairfoil chord)provided8dBnoisereductionwithminoreffectsontheaerodynamiccharacteristics.

Herr et al.[14] investigated TBL-TEnoise reduction on a DLRF16 airfoilby applyingporous materials on the last 10% of the chord. Results showed significant noise reduction atlow frequencies with respect to the solid case. Another im-portant conclusionwasthat communicationbetweenthepressureandsuction sidesoftheairfoilwasnecessarytoobtain noisereduction[14] ,indicatingthatanimprovementofthepressurebalanceisessential.Morerecently,Rubio-Carpioetal. [15,16] studiedthereductionofTBL-TEnoisebyreplacingsectionsofthetrailingedge(last20%ofthechord)withporous insertsmanufacturedoutofmetallicfoams.Acousticbeamformingresultsshowednoisereductionatlowandmid frequen-cies,ontheorderof10dBwithrespecttothesolidairfoil,andanincreaseathighfrequencies,whichwasattributedtothe surfaceroughnessoftheporousmaterial.Possibleexplanationsforthenoisereductionwererelatedtochangesinthe turbu-lenceintensityandReynoldsstressesintheboundary-layer,aswellasadifferentradiationefficiencyattheedge(reduction intheacousticimpedancediscontinuityatthetrailingedge)[16] .

DespitethepositiveresultsachievedforairfoilTBL-TEandflapside-edgenoise,therehasbeennoattemptyettoapply thistechnologyforJINreduction.Itisconjecturedthatthepressureimbalanceonasurfaceinthevicinityofajet,caused by impinging waveson only oneside, can be alleviatedthroughthe applicationofflow-permeable materials.This effect, coupledwithalessabruptjumpinimpedanceatthetrailing-edgeregion,islikelytoresultinnoiseabatement.Therefore, thiswork aimstoinvestigatethecapabilityofflow-permeablematerials toprovidenoisereduction foraninstalled jetby mitigating thetrailing-edge source,andachievingsoundlevels closeto thoseof anisolated jet. Forthispurpose, experi-ments arecarriedoutwithaflatplateplacedinthevicinity ofasingle-stream subsonicjet.Thetrailingedgeoftheplate canbereplacedwithametalfoaminsert,similarlyasperformedbyRubio-Carpioetal.[15] ,oraperforatedflow-permeable structurewithstraightholesorthogonaltothejetaxis.ParticleImageVelocimetry(PIV)isusedtocharacterizethejetflow fieldintermsoftime-averagedandroot-mean-square(r.m.s.)velocity.Acousticmeasurementsarealsoperformedwith mi-crophonesmountedinanarc-array.Far-fieldspectraofisolatedandinstalledjets(solidplate)arecomparedtotheinstalled cases withflow-permeabletrailing edges. The influenceof thejet velocity and thegeometryof the configurationis also addressed.

Thispaperisorganizedasfollows.InSection 2 ,theexperimentalset-upisdescribed,includingthecharacteristicsofthe flow-permeablematerials. InSection 3 ,theresultsof theexperimentsare discussed.Thejet flow fieldcharacterization is reported throughPIV measurements. Acoustic resultsof theisolated andinstalled configurations arealso included,along withtheeffectsofflow-permeablematerials oninstallationnoise.Finally,theconclusionsofthiswork aresummarizedin Section 4 .

Fig. 1. (a) Sketch of the FAST facility layout with a nozzle mounted in the anechoic chamber and the air supply system at the basement below. Adapted from [17] . (b) Nozzle mounted with the flat plate inside the facility.

Fig. 2. Cut view of the nozzle (dimensions as function of the exit diameter D j = 50 mm).

2. Experimentalset-up 2.1. Facilityandmodels

TheexperimentsareperformedintheFreejetAeroacouSTicfacility(FAST)atthevonKármánInstituteforFluidDynamics (VKI).Thisfacilityconsistsofacircularjetrigplacedinasemi-anechoicroom,asshowninFig. 1 a,withacut-off frequency of 350 Hz [17] . For the tests performedin thiswork, the air is supplied by a 7 bar pressure line located beneath the test chamber. The lineis alsobypassed to a seeding generator forPIV measurements. The seededflow merges with the pressurizedairinabuffertanktoensurecorrectmixing.Thejetblowsverticallyintoanextractorequippedwithamuffler [17] .Intheanechoicchamber,alasersourceandcamerasaremountedforPIVmeasurements,whereasa microphone arc-array is presentfortheacoustic ones. The picturein Fig. 1 bshowsthe jet nozzleinstalled withtheflat plateinside the facility.

Acircularconvergentnozzlewithan exitdiameterD_j =50mmandcontractionratioof36:1isdesignedbasedonthe geometryoftheSMC000nozzle,whichhasbeenusedforseveralinvestigationsofisolatedandinstalledsubsonicturbulent jets[2,18,19] .Thisnozzle,manufacturedinaluminium,isattachedtoastraightpipewith300mmofdiameter.Acutview ofthenozzleisshowninFig. 2 alongwithitsmaindimensions.Theoriginofthecoordinatessystemusedintheanalyses ispositionedatthecenterofthenozzleexitplane.

Fig. 3. Modular flat plate (dimensions as function of the exit diameter D j ). Pieces shown in blue can be replaced by the flow-permeable materials. (For

interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 4. Installed jet configuration. Geometric cases are investigated by changing the flat plate length L and radial position h .

Table 1

Jet flow conditions in terms of acoustic Mach number ( M a ), nozzle pressure ratio

(NPR), temperature ratio ( T R ) and Reynolds number ( Re ).

Condition Ma [-] NPR [-] TR [-] Re [ 10 5 ]

1 0.3 1.06 0.988 3.58

2 0.5 1.18 0.959 6.16

3 0.8 1.54 0.892 10.6

Fortheinstalledconfiguration,astainless steelflatplateismountedinthevicinityofthenozzle.Theplateisrealized with amodular structure, whichallows fordifferent surfacelengthsto be easily investigated.The length ofeach part is showninFig. 3 .The surfacehasa totaldimensionof500 mm_{× 1140}mm_{× 10}mm.The largewidthischosen toavoid side-edge scattering.Theaftpiececonsistsofasharptrailingedgewithachamferangleof40◦.Thismodulardesignalso allows foran easy replacement ofthesolid structure by theflow-permeablematerials. Twopieces atthe middlesection (shown inblueinFig. 3 ) canbe replacedby theflow-permeableones, allowing fortheinvestigationofdifferentporosity lengths(Lp=1DjandLp=3Dj).

Different geometriccasesaretestedbychangingthe lengthLandheight hoftheplate. AsshowninFig. 4 ,thelength is definedasthedistancebetweenthetrailingedge andthe nozzleexitplane, andtheheightastheradial positionwith respecttothejetcenterline.AbaselineinstalledcaseisdefinedwithL₌6D_j andh₌1_.5D_j .Theleadingedgeoftheplate ismountedupstreamofthenozzleexitplanetoavoidscatteringatthatregion.Adifferentplatelengthisalsoinvestigated (L=8D_j )forafixedh=1.5D_j ,aswellasadifferentradialposition(h=2D_j )forafixedL=6D_j .Duetoset-upconstraints, itisnotpossibletomounttheplateataradialpositionh_<1_.5D_j .Therefore,theeffectoftheplateheightisaddressedby moving itaway fromthejet. Moreover,withalengthshorterthan L=6D_j atthat position,itispossiblethatthe relative noiseincreaseduetoinstallationeffectswouldbemuchlower,particularlyformidandhighjetMachnumbers,whichcould compromisetheparametricanalysisoftheflow-permeabletreatment.

ThetestsareperformedatthreejetflowvelocitieswithdifferentacousticMachnumbersMa ,wherethejetvelocityU_j isdividedbytheambientspeedofsoundc0.Theflowcharacteristicssuchasthenozzlepressureratio(NPR)andthestatic

Fig. 5. Flow-permeable trailing-edge inserts applied to the flat plate. (a) Metal foam - d c = 800 μm. Adapted from [16] . (b) Perforated - d h = 800 μm.

Dimensions in μm.

measurementsareconductedatstaticconditions,i.e.noflowexternaltojet,ataverageambientconditionsofpamb=100.6

kPaandTamb=294K. 2.2. Flow-permeablematerials

Twotypesofnoisereductionsolutionsbasedonflow-permeablematerialsareinvestigatedinthiswork.Thefirstoneis an open-cellNiCrAlfoammanufacturedbythecompanyAlantum.Themetal foamismanufacturedthrough electrodeposi-tionofpureNionapolyurethanefoam,whichissubsequentlycoatedwithhigh-alloyedpowder[20] .Thistypeofmaterial consistsofahomogeneous microstructurewithathree-dimensionalrepetition ofa dodecahedron-shapedcell[16] . Rubio-Carpioetal.[15] haveinvestigatedtheapplicationofthismaterialwithdifferentcelldiametersdc forairfoilTBL-TEnoise reduction.Astructurewithnominaldc =800μmischosenforthisworksinceitsporosityandpermeabilitycharacteristics are available, andsignificant TBL-TEnoisereduction wasobtainedwiththisstructure [15] .Twoinsertsare manufactured fortheplate,asshowninFig. 5 ainordertoassesstheeffectofporositylengthonthenoisereduction.Thesecondtypeof flow-permeablematerialconsistsofa3D-printedperforatedinsertwithstraightholesconnectingtheupperandlowerside oftheplate,asshowninFig. 5 b.ThisinsertismanufacturedinR5,whichisaliquidphotopolymerthatoffersgoodsurface finishingandstrengthproperties[21] .Theholeshaveadiameterd_{h =}800μmandaspacingofl_{h =}2mm.

The flow-permeable materials are characterized by properties such asporosity

σ

andpermeability K.The porosity is definedastheratiobetweenthevolumetricdensitiesoftheflow-permeablematerial

ρ

p andofthe solidstructure

ρ

s ,as showninEq. (1) :

σ

=1−

ρ

p

ρ

s . (1)

The permeability isobtainedthrough the Hazen-Dupuit-Darcy equation (Eq. (2) ), whichprescribes the staticpressure loss



p acrossahomogeneoussamplewiththicknesst [22] :



p t =

μ

K

v

d +

ρ

C

v

2 d , (2)

where

μ

istheflow dynamicviscosity,

ρ

isthe flowdensity,

v

_d istheDarcian velocity(definedastheratiobetweenthe volumetricflowrateandthecross-sectionareaofthesample[22] ),andK andC arethepermeabilityandformcoefficients, whichaccountforpressurelossesduetoviscousandinertialeffects,respectively.

Forthemetalfoam,theporosityandpermeabilityparameterswereobtainedbyRubio-Carpioetal.[15] .Theformerwas obtained by measuringthe densityof smallsamples,whereas the latterwasobtained fromcharacterizationexperiments performedwithapermeabilityrig[15] .TheresultsarereportedinTable 2 .Asimilarprocedurehasbeencarriedoutforthe 3D-printedperforatedmaterial.TheresistivityR(R=

μ

/K)isalsoincludedinthetableforcomparison.

Table 2

Properties of the flow-permeable materials in terms of cell diameter ( d c ), porosity

( σ), permeability ( K), and resistivity ( R ) and form coefficient ( C).

Material dc [μm] σ[%] K [ ×10 −9 m 2 ] R [Ns/m 4 ] C [m −1 ]

Metal foam 0.8 91.7 2.71 6728 2612

Perforated 0.8 12.6 1.58 11540 7283

Fig. 6. (a) Concept of the PIV set-up with 2 cameras and 2 fields-of-view. (b) Fields-of-view from each camera (blue and red planes) and final FOV (black), with dimensions. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

2.3. Instrumentationandmeasurementtechniques 2.3.1. Flowfieldmeasurements

Atwo-dimensionaljetvelocityfieldisobtainedthroughPIVmeasurementsonthexy-plane(normaltothenozzleexit). Thismethodallowsforthemeasurementsoftime-averagedvelocitycomponentsuand

v

(intheaxialandradialdirections, respectively), andther.m.s.oftheirfluctuationsurms and

v

rms.ThePIVmeasurements areperformedonlyfortheisolated

jet configuration, since the investigatedconfigurations (length andheight) are chosen such to avoidgrazing flow onthe surface. Ithas beenshowninaprevious investigationthat thisdoesnotaffectthe noisegeneratedby turbulencemixing [6] .

SeedingparticlesareproducedbyaPIVTECPivpart45generator,comprisedby45LaskinnozzlesandusingShellOndina 919 oil,withaverage sizeof1 μm.Theseparticles havea relaxationtimeof 1μs [23] ,which issuitable duetothe flow acceleration in the nozzle. The illumination is provided by laser pulses generated witha double-cavity Quantel CFR200 Nd:YAG system.Thisequipment providesa laserwavelengthof532 nm,withamaximumenergyof200 mJ/pulse,anda pulsedurationof8ns. TwoLaVisionImagerSX4Mcameras(resolution:2360× 1776pixel;frame rate:31Hz;pixelsize: 5.5× 5.5μm;minimumtimeinterval:250ns;digitaloutput:12bit),positioned0.5mdistantofthejetaxis,areusedfor imagerecording.ThecamerasareequippedwithtwoNikkorf/1.8lensesof50mmfocallength.Thisconfigurationallowsfor measurementsoftwofields-of-view(FOV),inordertocapturealargerportionofthejetdevelopment,asshowninFig. 6 a. TheFOVs ofeachcameraareshowninFig. 6 bwithan overlapof1.25D_j betweenthem.ThefinalFOVhasadimensionof 12D_j × 4D_j (0.6m× 0.2m),anditisshownbytheblacklines.TheresolutioninthefinalFOVisapproximately6pixel/mm. Withthisset-up,1000pairsofparticleimagesareacquiredwithasamplingrateof15Hz.Theilluminationandimage acquisitionaretriggeredsynchronouslybytheLaVisionDaVis8.4software,whichisalsousedforthepost-processingofthe images.Theseparationtimebetweenpairedimagesistunedwithrespecttothejetvelocityinordertoobtainamaximum of25pixelsdisplacementatthejetcore.Thisvalueischosentoensureadisplacementofatleast3pixelsatregionsoflower velocity. A multi-pass cross-correlation algorithm [24] with window deformation [25] is applied. The final interrogation windowsizeis24× 24pixel2_{withanoverlapfactorof75%,whichprovidesafinalspatialresolutionof4mmandavector}

Table 3

PIV set-up and acquisition parameters.

Parameter Value

Cameras 2 x LaVision Imager SX4M

Acquisition frequency [Hz] 15 Max. pixel displacement [px] 25 Field-of-view 1-2 [mm] 400 × 310 Final field-of-view [mm] 600 × 200 Spatial scaling factor [px/mm] 6 Interrogation window [px 2 ] _{24 × 24}

Overlap factor [%] 75

Vectors per velocity field 600 × 200 Vector spacing [mm 2 ] _{1 × 1}

Fig. 7. Sketch of the microphone arc-array for far-field measurements. 12 microphones are placed from 40 ◦ _{to 150} ◦_{( θ= 0} ◦_{is upstream of the jet axis),} spaced of 10 ◦_{. The array is mounted at the reflected side of the plate for the installed configuration.}

detectorandarereplacedbyinterpolationbasedonadjacentdata[26] .ThemainparametersofthePIVset-uparereported inTable 3 .

The estimationof theuncertainty inthePIV measurements is performedfollowingthe methodproposed by Wieneke [27] .ThismethodprovidestheuncertaintyofaPIVdisplacementfieldbyprojectingtheparticlesfromonepointtoanother withthe obtainedvectorsandcheckingtheresultant disparity[27] .Thecalculationsresultin amaximum uncertaintyof 0.03U_j forthemeanvelocity,and0.04u_rms insidethepotentialcoreregion.Atthelipline(y₌0_.5D_j ),duetothestrongflow unsteadiness,maximumuncertaintyvaluesof0.06U_j and0.08u_rms areobtained.

2.3.2. Acousticmeasurements

The acoustic measurements are carriedout with12 Bruel & Kjaer4938 1/4” microphones (frequency range: 4 Hzto 70kHz;pressure-fieldresponse:±2 dB;max.output:172dB ref.2× 10−5 _{Pa).ThemicrophonesareintegratedtoBruel&}

Kjaer 2670-1/4” microphone preamplifiers,anda Bruel& KjaerNEXUS Type2690-Aconditioneris alsousedtoamplify the recordedsignals.Themicrophonesare mountedonan arc-arraydimensionedformeasurementsat1m radius(20D_j , centeredattheoriginofthecoordinatessystem).Thepolaranglefollowstheconventionof

θ

₌0◦intheupstreamdirection ofthe jetaxis.Therefore,themicrophoneat

θ

=90◦ is alignedwiththenozzle exit.The microphonesare mountedfrom

θ

=40◦ to

θ

=150◦, spaced of 10◦_{, as shown inFig. 7 .}

Fortheinstalledconfiguration,thearc-arrayismountedonthereflectedsideoftheplate(jetinbetweentheplateand array), inorder to assess theeffect of theflow-permeable materials on the reflectionof jet acousticwaves aswell. The measurements areperformed witha samplingfrequency of51.2kHz for20 s.For post-processing,the acoustic dataare split intoblocksof2048 samplesforeach Fouriertransform, andwindowed withaHanningweightingfunction with50% overlap.Theseparametersresultinafrequencyresolutionof25Hz.Thespectrashowninthefollowingsectionshavebeen also scaledtoan observeratadistanceof100D_j fromtheorigin, similarlyasperformedintheJINbenchmark studiesat NASAGlenn[2] .

Fig. 8. Contour plot of the time-averaged jet axial velocity u for a M a = 0 . 5 condition.

Fig. 9. Profiles of (a) time-averaged axial velocity and (b) r.m.s. of axial velocity fluctuations at the jet centerline for three jet velocities. The centerline velocity decay follows the trend defined by Witze [28] .

Fig. 10. Profiles of time-averaged axial velocity in the radial direction at three axial stations. The dashed line corresponds to the closest flat plate position, relative to the jet in the installed configurations.

3. Resultsanddiscussions 3.1. Jetflowfield

Inthissection,theflowfieldoftheisolatedjetisdiscussed.ThePIVmeasurementsareperformedforthe3investigated acoustic Machnumbersandtheresultsare displayedintermsoftime-averagedaxialvelocity uandther.m.s.ofvelocity fluctuations(u_rms).ThejetdevelopmentforMa =0.5isshowninthecontourplotinFig. 8 .Theregioncorrespondingtothe potential coreandthedownstreamvelocity decaycanbedetected, aswellasthespreadingofthejetandsymmetrywith respecttothecenterline.

The velocity profiles are extracted atthe jet centerline andplotted inFig. 9 . The quantities are non-dimensionalized by the respectivejet nominalvelocityU_j .The potential corelength Xc , definedasthedistance betweenthepoint where u=0.98U_j andthenozzleexit,isreportedinTable 4 foralljetvelocities.Thesevaluesarecomparedwithresultsobtained

Table 4

Potential core lengths obtained from experimental mea- surements ( X c ) and Witze’s equation ( X W ) for three jet

velocities.

Ma [-] Xc /D j [-] XW /D j [-]

0.3 4.4 4.6

0.5 5.0 5.1

0.8 5.4 5.8

fromWitze’sequationforpredictingthepotentialcorelengthX_W ofcompressiblefreejets[19,28] :

XW

Dj =

4.375

(ρ

_j /

ρ

∞

)

0. 28

1− 0.16Mj , (3)

where

ρ

_j and

ρ

_∞ arethejetandambientdensities,respectively.Agoodagreementisobtainedbetweentheexperimental and predictedresults. The centerline velocity decaydownstream of the potential coreis also shownto follow the trend definedbyWitzewiththeequation[28] :

u

Uj =1− e

α/ (1−x/XW), ₍₄₎

where

α

isaconstantequalto1.43[19] .

The increase inpotential corelengthwiththejet velocity isrelatedtothe changeinthe sizeofthe structuresin the mixing-layerwiththejetReynoldsnumber[28] .ForMa =0.8,thestructures arelikelytobesmallerandthus,themerge oftheshearlayeratthecenterline occursfurtherdownstream.Thisisalsoconfirmedbyther.m.s.ofvelocityfluctuations, plottedinFig. 9 b,whicharealsolowerforhigherjetvelocities.

Velocityprofiles intheradialdirectionarealsoobtainedattwoaxial stations,correspondingto thetrailing-edge posi-tions oftheinvestigatedinstalled jetconfigurations(x=6D_j andx=8D_j ).Theprofiles areplottedinFig. 10 ,alongwitha lineaty=1.5Dj,whichistheradialpositionwheretheplateisclosesttothejet,forMa=0.3.Similar resultshavebeen obtained forthe other jet velocities.It is shownthat the axial velocity iszero aty=1.5D_j for x=6D_j and, therefore,a platewitha trailingedgeatthisposition islocated outsideoftheplume. Conversely,forx=8D_j , aty=1.5D_j , thelocal axialvelocityisnon-zeroandequalto0_.05Uj.However,duetotherelativelylowvelocityatthispoint,itisnotlikelythat the surfacesignificantly changesthe characteristicsof theturbulent structures inthe mixing-layer,i.e.nochanges inthe noiseduetoturbulencemixingareexpectedevenforthelongestsurface.Theseresultsalsoallowforthecalculationofthe jetspreadingangle

δ

.Valuesof

δ

₌9◦ (M_{a =}0_.3);

δ

₌8_.9(M_{a =}0_.5)and

δ

₌8_.6(M_{a =}0_.8)areobtained.Theseresultsare consistentwiththosefromtheNASAGlenntests[19] ,andtheyconfirmthatthejetisfullyturbulent.

3.2. Far-fieldacousticresults

Inthissection,theresultsoftheacousticmeasurementsfortheinstalledjetwithflow-permeablematerialsarereported andcomparedwiththeisolatedandinstalled(solidtrailingedge)jets,initiallyforthebaselineplateconfiguration(L=6D_j andh=1.5D_j ).Twotypesofflow-permeablematerialsare investigated:ametal foamandaperforatedplatewithstraight holes;bothinsertshavealengthLp =3D_j .TheresultsaredisplayedinFig. 11 asSoundPressureLevel(SPL-ref.2× 10−5_Pa)

versusStrouhalnumberSt (St=f× Dj /Uj ),atthreepolarangles(

θ

=40◦,

θ

=90◦and

θ

=150◦)anddifferentjetvelocities. Thespectrahavebeenplottedstartingatafrequencyof350Hz(St =0.18forMa =0.3),whichistheminimumvaluefor whichtheroomhasanechoicproperties.

Fig. 11. Isolated and installed jet spectra, with solid and flow-permeable surfaces ( L p = 3 D j ), at three polar angles ( θ= 40 ◦, θ= 90 ◦ and θ= 150 ◦) and

Fig. 11. Continued

Fig. 11. Continued

Firstly,comparingthespectraforisolatedandinstalledjets(solidplate),itisshownthatinstallationeffectsare respon-sibleforastrongnoiseincreaseatlowandmidfrequencies;forMa =0.3and

θ

=40◦,thereisa17dBincreaseinSPLwith respect totheisolatedcaseattheinstalledspectral peak(St₌0_.37).Thisstrongnoise amplificationoccursup toSt₌0_.7 forthiscondition,andathigherfrequenciesthereisaconstantshiftofapproximately3dBfromtheisolatedcurve,which characterizes reflectionofacousticwavesonthesurface.Inthesidelinedirection(

θ

=90◦),theSPLincreasesforSt<0.3, whereas for0.3<St<0.6thereisa reductionwithrespecttothe upstreamdirection.Therefore,for

θ

=90◦,thespectral peakshiftstoalowerfrequency,possiblylowerthantherangewherethemeasureddataarereliable;thisimpliesthatthe effectoftheflow-permeablematerials atthespectral peakmightbe notsignificant fora full-scaleapplication, wherethe peakislikelylocatedbelowthehearingrange.Nonetheless,forafrequencyofSt=0.25,thereisalsoa17dBincreasewith respect totheisolated case.Inthe downstreamdirectionofthejet(

θ

₌150◦),thereis amaximumamplificationof7dB atSt=0.25duetothe dipolardirectivity ofthenoisegeneratedby theplate, aswell asincreasednoisefromturbulence mixingby the jet.Forhigherjet velocities,similar trends areobtained, butthe relativeamplification withrespectto the isolatednoiselevelsislowerduetoincreasedsignificanceofturbulence-mixingnoise.

Fortheplateswithflow-permeabletreatments,thespectrashow considerablenoisereductionwithrespecttothesolid installedcase,mainlyatlowandmidfrequencies,whereinstallationeffectsaredominant.ForMa =0.3and

θ

=40◦, reduc-tionsof9dBand7dBareseenatthespectralpeak(St₌0_.37)forthemetalfoamandperforatedinserts,respectively.The spectral amplitudeisalso reducedforSt>0.7, indicatingthat theflow-permeablematerials, particularlythemetal foam, reduce theeffectsofacousticwavereflectiononthesurface. For

θ

=90◦,despitethestronglow-frequencynoiseincrease, thereisstillasignificantreductionwiththepermeabletrailingedges;9dBand6dBforSt<0.25withthemetalfoamand perforated,respectively.Finally,for

θ

=150◦,5dB and3dBreductionsoccuratSt=0.25.Moreover, sincetheinstallation effectsareweakinthisdirection,thenoiselevelswiththeflow-permeabletrailingedgesaresimilartothoseoftheisolated jet.

Comparingthetwodifferenttreatments,themetalfoamprovidesmorebenefitsthantheperforatedinsertsforalltested cases. Since theformer has a higherpermeability, itis likely that the differencesin noise levels betweenthe two cases can be attributedto a better pressure balancebetween theupper andlower sidesof theplate forthe metal foamcase, thus reducing the surfacepressurefluctuationsnearthe trailingedge and, consequently, thenoise duetoscattering. The differences betweenthe two flow-permeableconfigurations is more noticeable atlow frequencies (St<0.4). This occurs because,for

θ

=40◦,whilethenoisereductionwiththeperforatedtrailingedgeisapproximatelyconstantforSt<0.5,for themetal foamthereisa changeinthespectral shape,withanewdistinctpeakatSt₌0_.45_,inthat direction.Thisisan indicationthatthereisanadditionalnoisesourceotherthanthetrailingedge.

Similar trends are obtained forhigher jet velocities. ForMa =0.5, there is a similar absolutenoise abatement at the spectralpeakasthepreviouscase(10dBreductionwiththemetalfoamand6dBwiththeperforated).Forthisvelocity,the

Fig. 12. Directivity plots of Overall Sound Pressure Level (OASPL) of isolated and installed jets with solid and flow-permeable surfaces ( L p = 3 D j ), for three

jet velocities.

Fig. 13. Scaling of Overall Sound Pressure Level with the jet acoustic Mach number for isolated and installed configurations, with solid and flow-permeable trailing edges.

Table 5

Difference in OASPL between solid and flow-permeable (metal foam and perforated) installed cases for a polar angle θ= 40 ◦ _{and three jet velocities. The overall in-} creases due to installation effects with respect to isolated levels are also included as reference.

Ma [-] OASPL SLD −ISO [dB] OASPL SLD −MF [dB] OASPL SLD −PERF [dB]

0.3 11.5 7.7 4.3

0.5 8.2 6.4 4.1

0.8 2.5 1.7 1.4

noise increase dueto installationeffectsisrelatively lowerwhen comparedtothe Ma =0.3jet. Therefore,withthesame absolute noisereduction provided by theflow-permeable materials, thespectra approachmore thelevels of theisolated configuration.ThiseffectsbecomesmorevisiblefortheMa =0.8jet,wherethecurvesofbothtreatedsurfacespractically collapsewiththeisolatedone for

θ

>90◦,indicatingthatthetrailing-edgesourcehasbeencompletelymitigatedinthese cases.

TheOverallSoundPressureLevel(OASPL)foreachcaseiscalculatedatallpolaranglesbyintegratingtheSPLspectrain therangeof350Hz<f<20kHzandtheresultsareshowninthepolarplotsinFig. 12 ,forthreejetvelocities.

The directivity plots show that the highestdifferencesbetween isolated andinstalled (solid plate)cases are found in the upstream direction, which is consistent withnoise fromscattering atthe platetrailing edge [4] . In thedownstream direction,thisdifferenceissmallerandtheinstalledcurvestendtocollapsewiththeonesoftheisolatedjet,especiallyfor the highestjet velocity.The resultsalso show that theflow-permeablematerials areeffective inreducing jet-installation noise inallassesseddirections,particularlyupstream.Thisindicatesthat thedipolesources ontheplateare mitigated.In the downstream direction, forthe metal foamcase, the levels reach thoseof theisolated jet for

θ

>120◦ andMa >0.5, indicatingthatthereisnochangeintheturbulence-mixingnoisecomponentduetothepresenceoftheplate.

The differencesbetween the OASPL forflow-permeable and solid surfaces are reported in Table 5 , for a polar angle

θ

=40◦.Theoverallincreaseduetoinstallationeffectswithrespecttotheisolatedjetisalsoincludedforreference.Itcan be seenthatthemetalfoamprovideshighernoisereductionthantheperforatedstructure,particularlyforMa =0.3.From the 11.5 dB overall increase dueto installationeffects, it ispossible toreduce 7.7dB by applyingthe metal foamat the platetrailingedge.Forhigherjetvelocities,theinstallationnoiseispracticallyeliminatedwiththisporousmaterial.Despite havingalowerpermeability,theperforatedtrailingedgestillprovidessignificantnoisereduction,ofapproximately4dBfor Ma =0.3andMa =0.5.

Fig. 14. Isolated and installed jet spectra, with solid and flow-permeable surfaces ( L p = 3 D j ), obtained for a plate with L = 6 D j and h = 2 . 0 D j (effect of

surface height) at a polar angle θ= 40 ◦_{and three velocities.}

The dependence of the OASPLwith the jet velocity for an angle

θ

=40◦ is also calculated andplotted inFig. 13 for eachcase.ReferencecurvesarealsoaddedforOASPL∝U8

j ,whichisconsistentforturbulence-mixingnoise[29] ,andOASPL ∝U5

j ,consistentwithscatteringatthesurfacetrailingedge[3] .Byapplyingthepermeabletreatment,theexponentofnoise levelswiththejetvelocityincreasesfromn=5.8ton=6.4,fortheperforatedplate,andton=7.2forthemetalfoam.The isolated jet hasn₌7_.9.These resultsarein qualitativeagreement withthosefromGeyer andSarradj [13] .Thisconfirms that, when flow-permeable treatments are applied to the surface, the scattering becomeslessdominant withrespect to othersourcessuchasturbulence-mixing.

Theeffectoftheconfigurationgeometryonthenoisereductionthatcanbeachievedusingflow-permeablematerials is investigatedinthefollowing.Firstly,theeffectoftheplateradialpositionisaddressedbymovingtheplateinthisdirection. ThespectrashowninFig. 14 areobtainedforaplatewithL=6D_j andh=2.0D_j atapolarangle

θ

=40◦.Thisconfiguration isshowninFig. 14 awithsolidlines;thedashedlinescorrespondtothebaselinecase(L₌6D_j andh₌1_.5D_j ).Theresults show that atthisheight andforMa =0.3,forexample,noise reductionsup to10 dBand6 dBwithrespect tothesolid case are achieved for the metal foam and perforated plate, respectively, for St=0.37. These values are similar to those obtainedforthepreviouscasewiththeplateclosertothejet(h₌1_.5D_j ).Forthemetalfoamtrailingedge,thenoiselevels collapse withthose of the isolated configuration forMa =0.5 andhigher. The perforated insertalso provides significant noise reduction;thereisstillatrailing-edgenoisecomponentatlowfrequenciesforMa =0.5(upto4dBwithrespectto theisolatedlevels),butatMa =0.8theinstallationeffectsarealsonotvisibleforthisconfiguration.

Sincelowerabsolutelevelsareobtainedinthespectraforthetreatedplatefartherfromthejet,itisinterestingtoplot theresultsintermsofnoisereductionswithrespecttoeachsolidcase.ThecurvesinFig. 15 aregivenintermsof



SPLfor therespectiveplateheight,andforeachpermeableconfiguration,forMa =0.3.Higherjetvelocitiesarenotshownsincethe turbulence-mixingnoisebecomessignificantanditisnotpossibletoproperlyassesstheeffectofthepermeablematerials. Itcanbeseenthatthecurvesaresimilar,withminorlocaldeviations,indicatingthattheabsolutenoisereductionsprovided by thepermeable materialsare independenton theplateradial position,i.e.independenton theamplitudeofimpinging pressure waves. It is likely that this property isalso the reason whythe SPL for the installed jetswith flow-permeable trailing-edgesapproachmoretheisolatedjetlevelsforhigherjetvelocities.

The effectoftheplatelengthisinvestigatedforasurfacewithL=8D_j andh=1.5D_j ,asshowninFig. 16 for

θ

=40◦. The resultsshow that,forthisgeometry,there isa significantnoise increase atlowfrequencies (St_<0_.35_,forM_{a =}0_.3). Moreover, the benefitsprovided by theflow-permeable materials arelower than in theprevious cases(6 dB decrease at St=0.35,forMa =0.3andboth typesofinserts).At midfrequencies (0.35<St<0.7,forMa =0.3),the metal foamand perforated inserts provide similar noise reduction for thisconfiguration. The main differencesbetween the two ofthem

Fig. 15. Reduction in noise levels with respect to the solid case for each plate height, obtained at θ= 40 ◦_{and for M a = 0 . 3 . (a) Metal foam. (b) Perforated.}

Fig. 16. Isolated and installed jet spectra, with solid and flow-permeable surfaces ( L p = 3 D j ), obtained for a plate with L = 8 D j and h = 1 . 5 D j (effect of

plate length) at a polar angle θ= 40 ◦_{and three velocities.}

occur in the range of noise increase due to the increment in the plate length. This is likely the result of the different permeabilityof thesurfacesatthe trailingedge,wherelarge-scale pressure wavesimpingeon theplate;the metal foam providesabetterpressurebalancebetweentheupperandlowersidesoftheplate,thusbetterreducingthesurfacepressure fluctuations at low frequencies. On the other hand, it is likely that the noise at 0.35<St<0.7 is generated by surface fluctuationsupstreamoftheflow-permeableregion,whichisthesameforbothcases.Similartrendsoccurfortheotherjet velocities.

Thiseffectcanbe verifiedbyanalysing theinfluenceoftheflow-permeableinsertlengthonthenoisereduction, fora fixed platelength L=6D_j andheighth=1.5D_j .Measurementsaretakenforinserts withlengthLp =1D_j ,andcompared totheonespreviously shown(L_{p =}3D_j ).SpectraareplottedinFig. 17 ,forapolarangle

θ

₌40◦ andthreeM_a .Theresults show that,forthemetal foam,thesmallerinsertstillprovidessignificant noiseabatement,particularlyforMa =0.3(6dB reductionatthepeak).ForMa =0.5,similarabsolutenoisereductionsareobtainedand,forMa =0.8,thecurvesaremore similarsinceturbulence-mixingnoiseissignificant.Therefore,longerflow-permeablesectionsprovidehigherbenefitssince there isashortersolid sectionofthe platesubjectedto strongsurfacepressurefluctuations.Forthe perforatedstructure, thesmallinsert(Lp =1D_j )provideslessnoisereduction,ofapproximately4dBatSt=0.37,forMa =0.3.Thedifferencein amplitudesbetweenthecurvesforthetwoinsertlengthsisalsomoresignificantatlowfrequencies(St<0.3forMa =0.3),

Fig. 17. Isolated and installed jet spectra, with solid and flow-permeable surfaces ( L p = 3 D j and L p = 1 D j - effect of porosity length), for a plate with

L = 6 D j and h = 1 . 5 D j , at a polar angle θ= 40 ◦and three velocities.

indicatingthat theadditionalsolidlength,forthecaseswithashorterinsert,generatesnoiseinthisfrequencyrange.This is asimilarbehaviour tothat ofincreasingtheoverall platelength, asshowninFig. 16 .Nonetheless, itcanbe concluded that evensmallsectionsofpermeabletreatmentaresufficientforachievingnoise reduction.Thisisimportantsincethose typesofstructuresusuallyleadtoperformancedegradation(lossofliftanddragincrease)[12,16] .

Itisshownthatthesolidextensionoftheplateaffectsthefinalspectralshapeandamplitude,alsoshiftingthefrequency ofpeakSPL.Therefore,itisalsoimportanttoanalyzetheeffectofchangingthelengthoftheporousinsert,butkeepingthe sizeofthesolidsection oftheplateconstant.Forthatpurpose,spectraoftwocasesarecompared: L₌6D_j withL_{p =}1D_j andL=8D_j withLp =3D_j .Therefore,bothcaseshaveasolidsectionof5D_j betweenthenozzleexitandtheflow-permeable section.ResultsareshowninFig. 18 ,forthetwotypesofpermeablematerialsandthreeMa .Theresultsaresimilartothose showninFig. 17 .Thecasewithanoveralllongerplatehasmorenoisegeneratedatlowerfrequencies(St_<0_.3forM_{a =}0_.3), forbothmetal foamandperforatedinserts;atSt=0.27,thereisa5dBdifferencebetweenthemetalfoamcurvesand4.4 dB forthe perforatedones. Therefore,thisis likely attributedto the difference in totalplate length so that the noise is generatedduetothe impingementofhigh-amplitudeandlow-frequencypressurewavesontheflow-permeableregion of theplate.Ontheotherhand,thenoiseatmidfrequenciesdoesnotshowsignificantchangewhencomparingthetwocases. Therefore,itisprobablethatthedominantsourceinthisrangeisthesameforbothofthem,anditislikelythatthesource isnowlocatedatthesolid-permeablejunctionintheplate.

Fig. 18. Isolated and installed jet spectra, with flow-permeable inserts and same solid length ( L = 8 D j with L p = 3 D j and L = 6 D j with L p = 1 D j ), at a polar

angle θ= 40 ◦_{and three velocities.}

Itisspeculatedthat thejunctionbetweensolidandflow-permeablesurfaceshasbecomethedominantsourcelocation forthemetalfoamcase.Theeffectofthejunctionhasbeendescribedintheliteratureasanadditionalgeometricsingularity, andthus, asanewscatteringregion,asshownby KisilandAyton [30] .Scattering atthejunction isthenresponsible for noiseincreaseatmidandhighfrequencies,alsochangingthedirectivitypatternoftheoverallconfiguration[30] .Moreover, beamformingresults fromRubio-Carpio etal.[16] showedthat, forfrequencieswhereTBL-TEnoise reductionis achieved with flow-permeablematerials, the dominantsource is placed at the solid-flow-permeable junction[16] . Therefore, it is possiblethatthereisanadditionalcontributionfromthatregion,particularlyforthecaseswiththemetalfoamduetoits highpermeability.Thejunctioneffectwouldthusbethecauseofthedifferentspectralshape,aswellasoftheSPLpeakat ahigherfrequency,relativetothefullysolidandperforatedcases.Theresultspreviouslyshownforthemetalfoamcaseare inagreementwiththishypothesis;forthereducedinsertlength,thejunctionisplacedatx=5D_j (asopposedtox=3D_j in thebaseline case),andthespectralpeak shiftstowardsalowerfrequency(Fig. 17 ).Ontheotherhand,whenthejunction isplacedatthesamepositionandtheporousextentischanged,thereissimplyanincreaseinamplitude,butthespectral peak frequencyremainsunchanged(Fig. 18 ). Thiseffectislikelynotobtainedwiththe perforatedconfiguration,since the low permeabilitydoesnotresultinastrongimpedancejumpatthejunction,and,consequently, scatteringatthat region. Furtherworkisnecessarytoconfirmthesehypotheses.

4. Conclusions

Anexperimentalstudyontheeffectofflow-permeablematerialsonthenoiseproducedbyaninstalledjetisperformed. The configuration is comprised by a single-stream subsonic jet anda nearby flat plate, placed inthe jet near-field. Two types offlow-permeable structures are investigated: a metal foam anda perforated insertwith straight holes normalto the axis.The metal foam hasa higher porosity andpermeabilitythan the perforatedstructure andits channels are also interconnected.

Planar PIV measurements arecarried out tocharacterize thejet velocity field.Based on thepotential corelength and spreadingangle,itisconcludedthatthejethasaturbulentbehaviourforalltestedvelocities.Moreover,itisconfirmedthat there isnodirectgrazingofthejetonthe plate,exceptforthelongestsurfacetested.However, forthiscase, thesurface is ina region ofvery low velocitiescompared to thepotential core, andit islikely not affectingthe noise generatedby turbulencemixing.Acousticmeasurementsshowthattheinstallationeffectsareresponsibleforstronglow-frequencynoise increasewithrespecttoisolatedlevels.Thisamplificationismoresignificantatalowjetvelocity,wherethedipolesources on thesurfacearemoreacousticallyefficientthanthe quadrupolesourcesfromturbulent mixing.The spectralshape and amplitude areshown tobe dependent on thegeometryof theconfiguration; longer surfacesproduce morenoise atlow frequencies, whereas moving theplatetowards the jet in theradial direction resultsin noise increase, especiallyat mid frequencies.

Significant noise reduction is achievedwhen the solid platetrailing edge is replaced by flow-permeable inserts, par-ticularly inthe low/mid frequencyrange,where thescattering isthe dominantmechanism. Comparing thetwo types of structures, the metal foam is moreeffective inreducing JIN, likely dueto ahigher permeability,which can mitigate the pressureimbalancebetweentheupperandlowersidesoftheplate,andthusreducethenoise generatedby surface pres-surefluctuations.Forlowjetvelocities,anoisedecreaseofupto10dBisobtainedatthespectralpeakwiththemetalfoam, buttheinstallationnoiseisstillvisible.Whenthejetvelocityisincreased,theattenuationprovidedbytheflow-permeable treatmentbringsthenoiselevelsclosertotheisolatedcase,andthetrailing-edgesourceisnolongerdominantwithrespect tothejetquadrupoles.Itisworth mentioningthatthehighestnoiselevelsfortheinvestigatedinstalledconfigurations oc-curatlowfrequencies(St<0.3forMa =0.3),particularlyatthesidelinedirection(

θ

=90◦).Forafull-scaleaircraft,these frequencies maynot be of particularsignificance. However, theflow-permeable trailingedges assessed inthis work also providenoisereductionsatmidandhighfrequencies,includingreflectioneffectsonthesurface,whichwouldbesignificant inafull-scaleconfiguration.

The effectofsurfacetreatmentis alsoassessed fordifferentconfigurationgeometries.Bymoving theplateaway from the jet, flow-permeablematerials providesimilar absolutenoise reduction asthebaseline case. Conversely, by increasing the platelength, lower noise abatement is obtainedwith theflow-permeable treatments, particularlyat low frequencies (St_<0_.35forM_{a =}0_.3),withthemetalfoamstillprovidinghigherbenefits.Ontheotherhand,thenoiseatmidfrequencies (0.35<St<0.7forMa =0.3) issimilar forthe twotypesofinsert,indicating thatit isgeneratedby theimpingementof pressurewavesinthesolidregionoftheplate,upstreamoftheflow-permeabletreatments.

Forafixedplatelength,a shorterflow-permeableinsertisshowntoprovidenoise reductionswithrespecttothesolid case, butinalower degreecomparedto thelargerinsert.Themain differencesoccuratlow frequencies,whichindicates that theincreasednoiseisduetotheadditionalsolid length,comparedto thecasewiththelonger insert.The frequency of highestSPL also shifts towards low frequencies. Onthe other hand, when theplate length ischanged, butthe solid-permeablejunctioniskeptatthesameaxialposition,theflow-permeablematerialsbehavedifferently.Forthemetalfoam, thereisanincreaseinamplitude,butnosignificantchangetothespectralpeakfrequency,whereasfortheperforatedthere isalow-frequencynoiseincreasewithachangeinthespectralpeak.Itisbelievedthatthisdifferenceiscausedbythehigh permeabilityofthemetal foam,whichproducesanewsingularityandthusanewscatteringregionatthesolid-permeable junction.

Theseresultsshowthatasurfacetreatmentwithflow-permeablematerialsisapotentiallypromisingmitigationsolution forjet-installationnoise.However,themechanismsthatprovidesuchreductionsarestillunclear.Furtherworkisrequiredto investigatethephenomenahappeningatthejunctionregionandinsidetheflow-permeablestructure,particularlyfocusing onthechangeofimpedance,pressureimbalanceandtheeffectofpermeability/resistivityoftheflow-permeablestructures, sinceitispossibletoachievesubstantialnoisereductionwithaperforatedstructure,evenwithalowporosity.

DeclarationofCompetingInterest

The authors declare that they have no knowncompeting financial interests or personal relationshipsthat could have appearedtoinfluencetheworkreportedinthispaper.

CRediTauthorshipcontributionstatement

LeandroRego:Conceptualization,Methodology,Validation, Formalanalysis, Investigation,Data curation,Writing - orig-inaldraft,Writing-review&editing,Visualization.DanieleRagni:Investigation,Writing-review&editing,Visualization, Resources,Supervision,Project administration, Fundingacquisition. FrancescoAvallone: Writing- review &editing, Visu-alization,Supervision.DamianoCasalino:Writing-review&editing,Visualization,Supervision.RiccardoZamponi:

Inves-tigation, Writing-review &editing,Visualization,Resources,Supervision.ChristopheSchram: Writing-review& editing, Resources,Supervision.

Acknowledgements

ThisworkispartoftheIPER-MANproject(InnovativePERmeableMaterialsforAirfoilNoiseReduction),projectnumber 15452,funded bythe NetherlandsOrganization forScientificResearch (NWO).The authorswouldliketo thankAlejandro Rubio-Carpio,forprovidingtheparametersfromtheflow-permeablematerialscharacterization.Theauthorswouldalsolike tothankDr.MirjamSnellenandProf.SybrandvanderZwaagfromtheDelftUniversityofTechnology,forcollaborationin theproject.

References

[1] J.L.T. Lawrence, M. Azarpeyvand, R.H. Self, Interaction between a flat plate and a circular subsonic jet, in: 17th AIAA/CEAS Aeroacoustics Conference, Portland, OR, USA, 2011, doi: 10.2514/6.2011-2745 .

[2] C. Brown , Jet-surface interaction test: far-field noise results, in: Proceedings of the ASME Turbo Expo 2012: Power for Land, Sea and Air, Copenhagen, Denmark, 2012, pp. 1–13 .

[3] J.E. Ffowcs-Williams, L.H. Hall, Aerodynamic sound generation by turbulent flow in the vicinity of a scattering half plane, J. Fluid Mech. 40 (4) (1970) 657–670, doi: 10.1017/S0 022112070 0 0 0368 .

[4] R.W. Head, M.J. Fisher, Jet/surface interaction noise: - analysis of farfield low frequency augmentations of jet noise due to the presence of a solid shield, in: 3rd AIAA Aeroacoustics Conference, Palo Alto, CA, USA, 1976, doi: 10.2514/6.1976-502 .

[5] I. Belyaev, G. Faranosov, N. Ostrikov, G. Paranin, A parametric experimental study of jet-flap interaction noise for a realistic small-scale swept wing model, in: 21st AIAA/CEAS Aeroacoustics Conference, Dallas, TX, USA, 2015, doi: 10.2514/6.2015-2690 .

[6] L. Rego, F. Avallone, D. Ragni, D. Casalino, Noise amplification effects due to jet-surface interaction, AIAA Scitech 2019 Forum, San Diego, CA, USA, 2019, doi: 10.2514/6.2019-0 0 01 .

[7] A.V.G. Cavalieri, P. Jordan, W.R. Wolf, Y. Gervais, Scattering of wavepackets by a flat plate in the vicinity of a turbulent jet, J. Sound Vibr. 333 (24) (2014) 6516–6531, doi: 10.1016/j.jsv.2014.07.029 .

[8] D. Casalino , A. Hazir , Lattice Boltzmann based aeroacoustic simulation of turbofan noise installation effects, in: 23rd International Congress on Sound & Vibration, Athens, Greece, 2014, pp. 1–8 .

[9] C. Hughes , The promise and challenges of ultra high bypass ratio engine technology and integration, AIAA Aerospace Sciences Meeting, Orlando, FL, USA, 2011 .

[10] V.G. Mengle, L. Brusniak, R. Elkoby, R.H. Thomas, Reducing propulsion airframe aeroacoustic interactions with uniquely tailored chevrons: 3. Jet-flap interaction, in: 12th AIAA/CEAS Aeroacoustics Conference, Cambridge, MA, USA, 2006, doi: 10.2514/6.2006-2435 .

[11] J.D. Revell, H.L. Kuntz, F.J. Balena, B.L. Storms, R.P. Dougherty, Trailing-edge flap noise reduction by porous acoustic treatment, in: 3rd AIAA/CEAS Aeroacoustics Conference, Atlanta, GA , USA , 1997, doi: 10.2514/6.1997-1646 .

[12] E. Sarradj, T. Geyer, Noise generation by porous airfoils, in: 13th AIAA/CEAS Aeroacoustics Conference, Rome, Italy, 2007, doi: 10.2514/6.2007-3719 . [13] T. Geyer, E. Sarradj, Trailing edge noise of partially porous airfoils, in: 20th AIAA/CEAS Aeroacoustics Conference, Atlanta, GA , USA , 2014, doi: 10.2514/

6.2014-3039 .

[14] M. Herr, K. Rossignol, J. Delfs, N. Lippitz, M. Mößner, Specification of porous materials for low-noise trailing-edge applications, in: 20th AIAA/CEAS Aeroacoustics Conference, Atlanta, GA , USA , 2014, doi: 10.2514/6.2014-3041 .

[15] A. Rubio Carpio , R. Merino Martínez , F. Avallone , D. Ragni , M. Snellen , S. Van Der Zwaag , Broadband trailing edge noise reduction using permeable metal foams, in: InterNoise Conference, Hong Kong, 2017 .

[16] A. Rubio Carpio, R. Merino Martínez, F. Avallone, D. Ragni, M. Snellen, S. Van Der Zwaag, Experimental characterization of the turbulent boundary layer over a porous trailing edge for noise abatement, J. Sound Vibr. 443 (2019) 537–558, doi: 10.1016/j.jsv.2018.12.010 .

[17] D. Guariglia, A. Rubio-Carpio, C. Schram, Design of a facility for studying shock-cell noise on single and coaxial jets, Aerospace 5 (1) (2018) 25, doi: 10.3390/aerospace5010025 .

[18] C. Brown , J. Bridges , Small Hot Jet Acoustic Rig Validation, Technical Report, NASA/TM-2001-214234, Cleveland, OH, USA, 2006 .

[19] J. Bridges, M. Wernet, Establishing consensus turbulence statistics for hot subsonic jets, in: 16th AIAA/CEAS Aeroacoustics Conference, Stockholm, Sweden, 2010, pp. 1–41, doi: 10.2514/6.2010-3751 .

[20] S. Kim, C. Lee, A review on manufacturing and application of open-cell metal foam, Procedia Mater. Sci. 4 (2014) 305–309, doi: 10.1016/j.mspro.2014. 07.562 .

[21] Tenco DDM, DLP Digital Light Processing Data Sheet, URL http://tenco- online.com/wp- content/uploads/2017/03/Datasheets- Tenco- DDM.pdf . [22] D.B. Ingham, I. Pop, Transport Phenomena in Porous Media, first ed., Pergamon, Kidlington, Oxford, UK, 1998, doi: 10.1016/B978- 0- 08- 042843- 7.

X50 0 0-4 .

[23] D. Ragni, F. Schrijer, B.W. Van Oudheusden, F. Scarano, Particle tracer response across shocks measured by PIV, Exp. Fluids 50 (1) (2011) 53–64, doi: 10.10 07/s0 0348- 010- 0892- 2 .

[24] F. Scarano, M.L. Riethmuller, Advances in iterative multigrid PIV image processing, Exp. Fluids 29 (7) (20 0 0) S051–S060, doi: 10.10 07/s0 03480 070 0 07 . [25] F. Scarano , Iterative image deformation methods in PIV, Meas. Sci. Technol. 13 (1) (2001) .

[26] J. Westerweel, F. Scarano, Universal outlier detection for PIV data, Exp. Fluids 39 (6) (2005) 1096–1100, doi: 10.1007/s00348- 005- 0016- 6 . [27] B. Wieneke, PIV uncertainty quantification from correlation statistics, Meas. Sci. Technol. 26 (7) (2015), doi: 10.1088/0957-0233/26/7/074002 . [28] P.O. Witze, Centerline velocity decay of compressible free jets, AIAA J. 12 (4) (1974) 417–418, doi: 10.2514/3.49262 .

[29] M.J. Lighthill, On sound generated aerodynamically I. General theory, Proc. R. Soc. A (1952) 564–587, doi: 10.1098/rspa.1952.0060 .

[30] A. Kisil, L.J. Ayton, Aerodynamic noise from rigid trailing edges with finite porous extensions, J. Fluid Mech. 836 (2018) 117–144, doi: 10.1017/jfm.2017. 782 .