Transport Aircraft Wake Influenced by Oscillating Winglet Flaps

8/10/2019 Transport Aircraft Wake Influenced by Oscillating Winglet Flaps

1/14

Transport Aircraft Wake In uenced by Oscillating Winglet Flaps

Christian Breitsamter and Alexander AllenTechnische Universitt Mnchen, 85747 Garching, Germany

DOI: 10.2514/1.37307

The wake vortex development in the near eld and extended near eld behind a four engined large transport aircraft model tted with an active winglet is presented. A detailed wind-tunnel investigation is conducted using ahalf-model focusing on thehigh-lift case of a typical approach con guration at a Reynolds numberof 0 :5 10 6 basedon thewing mean aerodynamic chord andan angle of attackof 6.5deg. The ow eldis observed using advanced hot-wire anemometry mainly focusing on the cross ow plane at 5.6 spans downstream of the model. Based on the time-dependent velocity components the wake ow eld is analyzed by distributions of mean vorticity, turbulenceintensities, and spectral densities. Seven main vortical structures dominating the near- eld wingvortexsheet roll upand merge to form the remaining trailing vortex in the extended near eld. By the use of oscillating winglet aps thevelocity uctuations at the core region of the remaining vortex are signi cantly in uenced. Distinct narrowbandconcentrations of turbulent kinetic energycan be foundfor the farthest downstreamplane documentingthe presenceof the disturbances generated by the winglet aps which may result in an ampli cation of inherent far- eldinstabilities. The frequencies at which these narrowband energy concentrations occur are, on the one hand,dependent on the oscillation frequency of the winglet aps; on the other hand, there are also independent energyconcentrations within the frequency bands associated with wake instabilities.

NomenclatureAR = aspect ratiob = wing span, mb0 = main vortex pair distance, m~b0 = distance between two adjacent vortices of a four

vortex system, mCD = drag coef cient CL = lift coef cient, 2LAR= U 21 b2Cm = pitching moment coef cient c = wing mean aerodynamic chord, m f = frequency, Hz f

M = sampling frequency, Hz

k = reduced frequency, fb= 2U 1L = lift, NnB = number of band-averaging intervalsRe c = Reynolds number, U 1 c=r c = vortex core radius, mSN c;u 0 = cross-spectral density of u0 normalized with

kU 1 = u02 b=2SN u0 = power spectral density of u0 normalized with

kU 1 = u02 b=2s = load factor, = 4 0:7854 for an elliptic lift

distributionTu x, Tuy, = axial, lateral, vertical turbulence intensity,

u02p =U 1 , v02p =U 1

Tu z = w

02p =U 1U 1 = freestream velocity, m=s

u, v, w = axial, lateral, and vertical velocities, m=su, v, w = mean axial, lateral, and vertical velocities, m=su0, v0, w0 = uctuation part of u, v, and w, m=s

w0 = downwash velocity with respect to 0 , m=s,0 = 2 b 0

x = nondimensional distance in x direction, x=b; x0 at winglet tip trailing edge for 0 deg

y , z = nondimensional distances in y and z directions,2y=b, 2z=b

= aircraft angle of attack, deg0 = root circulation based on an elliptic circulation

distribution, m2=s CL U 1 b = 2 ARs = winglet ap de ection, deg = wavelength, m = kinematic viscosity, m2=s

= nondimensional axial vorticity, ! xb= 2U 1 = density, kg=m3

= nondimensional time, x 16CL = 4AR! x = mean axial vorticity component,

d w=dy dv=dz , 1=s

Subscripts

Crouch = Crouch instabilityCrow = Crow instabilityd = dominant max, min = maximum, minimumosc = oscillating

Introduction

M AJOR airports around the worldare encountering a long termcapacity problem dueto thegrowing dif culties in expandingairports and runways. Aircraft size is increased to compensate thisdevelopment. As the minimum separation distance between twoaircraft during approach, takeoff, and enroute is solely driven by themaximum takeoff weight and therewith by the size of the twoinvolved aircraft [1,2], the compensation may well lead to a loss inpassenger numbers per hour, if the separation distances of futurelarge transport aircraft are increased beyond their increase in seatingcapacity in comparison to the aircraft operating today. The FederalAviationAdministrationin theU.S.andtheCivil AviationAuthorityin Europe regulate the separation distances, which are estimatedconservative toaccount for thevaryingbehaviorof thevorticesunder different atmospheric conditions [3]. The aim of numerous activitiesis to reduce aircraft separation distances to meet the requirements of

Received 26 February 2008; revision received 10 October 2008; acceptedforpublication10 October 2008. Copyright 2008 by C. Breitsamter andA.Allen. Published by the American Institute of Aeronautics and Astronautics,Inc., with permission. Copies of this paper may be made for personal or internal use, on condition that the copier pay the $10.00 per-copy fee to theCopyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA01923; include the code 0021-8669/09 $10.00 in correspondence with theCCC.

Privatdozent, Dr.-Ing., Chief Scientist, Institute of Aerodynamics.Associate Fellow AIAA.Dipl.-Ing., Master of Science, Research Engineer, Institute of

Aerodynamics. Member AIAA

JOURNAL OF AIRCRAFTVol. 46, No. 1, JanuaryFebruary 2009

175
http://dx.doi.org/10.2514/1.37307http://dx.doi.org/10.2514/1.37307


2/14

growing air traf c, without reducing the level of safety alreadyobtained today and therewith relieve hub airports.

Wake vortical structures and related problems have beenintensively investigated in thepast[47].Thebehaviorof thetrailingwake of a lifting body is well known. It rolls up into a pair of strongcounter-rotating longitudinal vortices that persist for many bodydimensions downstream. The vortex strength is proportional to thebound circulation or body lift, andhence, forsteadyight conditionsthis is approximately proportional to the weight of the generating

aircraft. An aircraft encountering a vortex wake can experience arolling motion, if the following aircraft enters the wake of theprecedingaircraftand bycoincidencemeets thevicinityof thevortexaxis. Also a sudden upwash or downwash can occur, leading toincreasedstructural loads or a dangerouschangein climb rate, whichis especially critical close to theground. In anycasea rapidchangeof attitude possibly combined with heavy disturbances of trimmed

ight conditions can occur.Wake vortex physics and means of wake vortex alleviation have

been addressed in several integrated research projects [810].Various experimental and numerical methods have been used tostudy the wake vortex evolution and development [11,12]. Testshave been and still are conducted on generic models as well as ondetailed transport aircraft models in wind tunnels, towing tanks, andcatapults in order to deliver complementary data resulting in a sounddescription of the structure of the wake vortices progressingdownstream. Further, ight tests have been performed usingtriangular lidar measurementsto observe andreveal thedevelopment of a wake generated by an aircraft under real meteorologicalconditions [13]. Additionally, computational methods, such asvortex lament methods, Reynolds-averaged Navier Stokescomputations, and large eddy simulations, are widely used todescribe all stages of the wake vortex development [1416]. Theadvancednumerical methods includealsotheinuence of turbulenceand unsteady effects on the wakevortex development. Experimentalstudies have demonstrated the importance of such effects [17].

Todays technologyis not atthe stage to track orpredict wakevor-tex locationsunder all weather conditions. Therefore, manyresearchactivities concentrate on alleviating the wake vortex hazard bymodi cations of wing geometry and/or wing loading. Strategies tominimize the wake vortex hazard concentrate, on the one hand, on alowvorticityvortexdesign,whichreducesthe wakevortexhazardbyenhancing the dispersion of the vorticityeld. It is aimed on the gen-eration of wake vortices with larger core size and smaller swirlvelocities at the core radius after roll up is completed. Also, anoptimum wing load distribution may minimize the induced rollingmomentfor a followingaircraft[18]. Thealteration of thecirculationdistribution of the wake generating wing can be obtained, for example, by using differential ap settings or spoiler settings. It hasbeen shown that a wing with an outboard partially deected ap andan inboard fullydeected ap produces, at least in the extendednear

eld, a smaller induced rolling moment than a wing with a standardapsetting[19].Ontheotherhand,thefocusisonaquicklydecaying

vortex [11]. An enhanced vortex decay may be achieved by promot-

ing three-dimensional instabilities by means of passive or activedevices [2022]. Because a multiple vortex system shows instabili-ties which can grow more rapidly, passive devices aim to promotethese kinds of instabilities through thedeliberateproductionof singlevortices in addition to those coming from the wing tip and the apedge [11,20,23]. The production of additional distinct vortices canalso be achieved by a differentialap setting. The ef ciencyof theseconcepts depends on the persistence of such additional vortex pairswhich is determined by congurational details of the aircraft.

Especially, active devices are considered as a possible powerfulmean to amplify wake vortexinstabilities.In themid 1970s,theoreti-cal and experimental studies dealt with the inuence of sinusoidallymoving aps on the wake vortex behavior and predicted a reductionin the wake life span by a factor of 3 [5,24]. In the last years, suchmeans were carefully investigated again. An active system wasproposed by Crouch et al. based on periodic oscillations of controlsurfaces, for example, ailerons and aperons [25,26]. The perturba-tions in uence thevortex waketo triggerinherent instabilitieswhich,

after suf cient ampli cation, may cause an earlier breakup of thetrailing vortices into vortex rings resulting in a rapid decay. Further investigationsconcentrate on towing tankexperiments usingmodelsof generic wing congurations. Theoscillating devices include inner and outer ailerons [27], trailing-edge aps [28,29], and winglet aps[28]. In addition, unsteady Reynolds-averaged Navier Stokescalculationshavebeen carried outfor a congurationwithoscillatingtrailing-edge aps [29]. Depending on the streamwise distance thewake vortex hazard is reduced up to a certain extent quantied, for

example, by the maximum induced rolling moment or vorticitydistribution. Other means, such as separation control by zero mass-ux blowing slots at ap edges [30] and active Gurneyaps [31] are

also under consideration.The results presented herein concentrate on the near-eld wake

characteristics of a large transport aircraft featuring a large winglet ttedwith two trailing-edgeaps. Thelower andupper winglet aps

can be de ected in the same direction (symmetrical case) or in theopposite direction (asymmetrical case). The inuence of symmetricand asymmetric static ap de ections on the wake ow eld isdiscussed in [32] whereas here the inuence of ap oscillations isaddressed. The ow eld development is characterized by axialvorticity distributions, turbulence intensity patterns, and spectraldensities. The investigation documents the wake vortex formationand analyzes the inuence of the oscillating winglet aps on therolled-up trailing vortex characteristics. The focus is to demonstratethat the frequency dependent uctuations generated in the near eld(instabilities at x 0) are present at the farthest downstream plane(at x 5:6). Such disturbances are aimed at amplifying thedevelopment of inherent far-eld wake instabilities.

Experimental SetupThe wind-tunnel facility C of the Institute of Aerodynamics at the

Technische Universitt Mnchen was used for this investigationemploying a detailed large transport aircraft conguration andapplying advanced hot-wire anemometry.

Wind Tunnel

The wind tunnel C is a closed-return type with a cross section of 1:8 2:7 m2 and has a test section length of 21 m, which covers awake distance of approximately 5.6spans downstream of themodel.Themaximum usablevelocity is30 m=s. The turbulence level at thenozzle exit is less than 0.5%. The ceiling is adjustable to control theaxial pressuregradient along thetest section. With themodelin placethe axial pressure gradient is set close to zero to ensure unaffectedwake development.

Model

Figure 1 illustrates the 1:32 scaled half-model of a typical largetransport aircraft, which was used for this investigation. The modelhas a wing semispan (b=2) of 1.242 m (aspect ratio AR 8:0, taper

Fig. 1 Half-model of large transport aircraft installed in test section of wind tunnel C.

176 BREITSAMTER AND ALLEN


3/14


4/14

regions the relative spacing is gradually enlarged to 0.024 laterallyand 0.036 vertically. Regarding the susceptibility of vorticalstructures to intrusive measurements, it was found that the presenceof the hot-wire probe has no markable inuence on the wake vortexformationandevolutioncomparingow elds obtainedbyprobeandparticle image velocimetry measurements [11].

Results and DiscussionA vortexwake canbe divided into four main regions regardingthe

downstream development: 1) the near eld, x=c 1, which ischaracterized by the formation of highly concentrated vortices shedat all surface discontinuities; 2) the extended near eld, x 10 ,where the wake roll-up process takes place, and the merging of dominant vortices (ap edge, wing tip) occurs, leading gradually totwo counter-rotating vortices; 3) the far eld,10 x 100 , wherethe wake is descending in the atmosphere and linear instabilitiesemerge; and 4) the decay region, x > 100 , where fully developedinstabilities cause a strong interaction between the two trailingvortices until they collapse. The results shown here are solely for thenear eld and extended near eld. The downstream stations aremarked by the nondimensional distance x and the nondimensionalcharacteristic time x 16CL = 4AR . The latter number

includes lift coef cient and aspect ratio to compare results of different tests and congurations with reference to an elliptic lift distribution. The ow elds are presented as contour plots of nondimensional axial vorticity ! xb= 2U 1 and verticalturbulence intensity Tu z. The term ! x indicates the mean axialvorticity component (d w=dy dv=dz).

Baseline Con guration

To quantify theoverall development of thewake vortexow eld,a detailed investigation is conducted using the large winglet without de ection or oscillation of the winglet aps. The nondimensionalaxial vorticitydistribution and turbulence intensities of thecrossowplane at x 0:37, 0:011 , are shown in Fig. 3. The main near-

eld vortices are marked by the corresponding acronyms and their

peak vorticity levels are stated. Note that the vorticity levels1:0 < < 1:0 are blanked in all those gures, to clarify the positions of dominating vortices. Turbulence intensity levels of T uz < 0:02 arealso blanked to emphasize the vortex sheet emanating from the wingmore clearly.

Seven main vortices can be identied, namely, from outboard toinboard, the winglet vortex (WLV), the wing tip vortex (WTV), theoutboard nacelle vortex (ONV), the outboardap vortex (OFV), theinboard nacelle vortex (INV), the horizontal tail plane vortex (HTV)and the merged inboard ap vortex and wing-fuselage vortex (IFV/ WFV), Fig. 3a. Negative vorticity is attributed to the HTV and theIFV/WFV; these vortices are therefore counter-rotating incomparison to the other ve vortices. The HTV exhibits negativevorticity as thehorizontal tail plane needs to be adjusted fornegativelift due to trimmed ight. The negative circulation gradient in thewing-fuselage region leads to the counter-rotating WFV. The vortexsheet emanating from the wing trailing edge can clearly be observedand so can the vortex sheet shed at the horizontal tail plane. The twostrongestvortices based on peak vorticity levelsare theONV andtheOFV with the latter clearly being the strongest of all such distinct vortices.Themaximum peak vorticity level of theOFVis associatedwith the highly loaded outboard ap. The relatively high vorticitylevel of the ONV is caused by the inclined ring prole of the nacellesubject to thecrossowcausedbythesweptwing.Thes-like shapeof the vortex sheet in that region indicates that the two vortices haverotated around each other by approximately 180 deg. Also, severalless pronounced vortices indicated by smaller vorticity levels arepresent in between the six dominating wing vortices. These smaller vortices are caused by ow separations at slat horns, ap trackfairings, and other geometric discontinuities of the wing. Localturbulence maxima are found in the regions of high velocitygradients, especially for the area of retarded axial core ow of thedominating vortices, Fig. 3b. These turbulence maxima are evoked

by thestrong velocity gradient andstreamline curvaturein thevortexcross ow areas and by the inection in the radial proles of theretarded axial core ow. The fuselage wake is visible by a largeregion of increased turbulence intensity close to the plane of symmetry(y 0, 0:4 < z < 0:3).This regionis largerthanthefuselage diameter due to the upwash of the inclined fuselage and thedownwash of the wing, which moves the fuselage turbulence wakefrom its original position resulting in a vertical stretching of thefuselage turbulence sheet.

Progressing downstream to x 5:6, 0:164 , three vorticesremain at this station of the extended near eld, Fig.4. The strongest one is formed by the merging of the OFV with the ONV and WTV,which is thencalled the mainvortex (MV), and the second one beingtheWLV. Thethird vortexis theHTV, which is also still visible, but its peak vorticity level is strongly reduced. Because of the inducedvelocity eldsoftheOFV/ONVandtheWTV/WLVontheHTV,thelatter sinks faster than the other two vortices. The IFV/WFV and theINV have fed vorticity to other vortical structures. In comparison toFig.3, theWLVhasrotated aroundthemergedOFV/ONV/WTV by

y*

z *

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

WLV18.8

OFV55.9

ONV36.1

WTV9.8

INV31.1

IFV/WFV

-14.3

HTV-55.4

y*

z *

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

IFV/WFV0.030

INV0.111

ONV0.137

OFV0.172

HTV0.030

WTV0.047

WLV0.050

a)

b)Fig. 3 Contour plots of nondimensional axial vorticity ( ) andturbulence intensity ( Tu z) distributions at x 0:37; Re c 0:5 106 ;a) ; solid lines: positive values, dashed lines: negative values;

10 50; 1; b) Tu z; 0 :02 Tu z 0:2; Tu z 0:01.

http://-/?-http://-/?-http://-/?-http://-/?-


5/14

approximately 230 deg with the peak vorticities clearly decreasing.

Again, increased turbulence levels are attributed to the positions of the main vortices, Fig. 4b. The two strongest vortices are not in uenced by the enlarged region of the fuselage wake, whereas theHTV is now fully embedded in this area.

Figure 5 illustrates the movement of the vortices for they z plane between x 0:02 , 0:0006 and x 5:6, 0:164 . The paths of four vortices are illustrated, namely, theHTV, INV, WLV, and ONV, which later becomes part of the mainvortex. As the measuring plane at x 0:02 is upstream of thehorizontal tail plane,there isnodata fortheHTVat this station.Thereis also no data point available for the INV at x 5:6 because theINV has dissipated. The WTV merges with OFV and ONV just before x 1:0, 0:029 , but for clarity they are not illustratedhere. These three vortices form the main vortex, which exhibitsthe highest peak vorticity in the ow eld downstream of x 1:0.The vortices rotate counterclockwise around the roll-upcenter (y 0:78) which is located close to the position of themain vortex.

Further results concerning the inuence of large winglet andstaticwinglet ap de ection on the wake vortex development aredocumented in [32].

Oscillating Winglet Flaps

Measurements have been performed with both symmetrically andasymmetrically oscillating winglet aps with maximum deectionangle of 20 deg for the frequency range of 140 Hzcorresponding to kosc 0:051:987 . Based on freestream velocityandwingsemispanthe reducedfrequencyk fb= 2U 1 servesasanondimensional frequency parameter. The reduced frequency is also

used as a similarity parameter to transfer values between small-scaleand full-scale conditions. The frequency range of kosc 0:052 ischosen to cover the frequencies of dominant wake instabilities,namely, those of the Crow [34] and Crouch [35] type. Relatedreduced frequencies arekCrow 0:080:1 (long wave) andkCrouch0:120:6 (long to medium wave), [36]. Details on the reducedfrequencyvalues arepresented in thesection onSpectralAnalysis.

Themaximum apde ectionangle, 20 deg ,ischosenfortworeasons. On the one hand, the magnitude of the ap inducedvelocities at a certain frequency hasto be largeenoughto amplify theinitial disturbance level in the rolled-up vortex accelerating thedevelopment of long to medium wave instabilities responsible for wake vortex decay. On the other hand, the maximum deectionanglesat full-scale frequenciesarelimitedby theperformance of realaircraft actuator systems. Here, the velocities induced by theoscillating winglet aps are on the order of 25% with respect to thefreestream velocity.They areaimedto increase theaxial, lateral, andvertical velocity uctuations at the frequencies of long waveinstabilities in therolled-up (main)vortex byat least a factorof 3 to5.Therefore, mainly the crossow plane at x 5:6 is inspected andresults are compared with those obtained without ap oscillations(baseline).

Force measurements are conducted with a maximum static apde ection of 20 deg . A slightdecrease in lift can beobservedfor both symmetrical (1:7%) and asymmetrical (0:6%) winglet apde ections. The increase in the drag coef cient is about 2.1% for symmetrical winglet ap de ections and approximately 1.2% for asymmetrical deections. Any change of the overall forces andmoments is an important issue as the horizontal tail plane cannot bemoved at such high frequencies as the winglet aps, in order to keepthe aircraft in a steady ight condition. The pitching moment coef cient stays almost constant at asymmetric deections of lower and upper aps, that is, aps are de ected opposite at the same

y*

z *

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

Main Vortex(OFV/ONV/WTV)

30.1

WLV1.3

HTV-6.2

y*

z *

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

Main Vortex(OFV/ONV/WTV)

0.146

WLV0.028

HTV0.084

a)

b)Fig. 4 Contour plots of nondimensional axial vorticity ( ) andturbulence intensity ( Tu z) distributions at x 5:6; Re c 0:5 10 6 ;a) ; solid lines: positive values, dashed lines: negative values;

10 50; 1; b) Tu z; 0 :02 Tu z 0:2; Tu z 0:01.

y*

z *

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4INVWLVHT VONV-MV

HTV x*=5.6

HTV x*=0.37

INV x*=0.02

INV x*=4.0

WLV x*=5.6

WLV x*=0.02

ONV x*=0.02MV x*=5.6

Fig. 5 Positions of the vortices between x 0:02 and x 5:6; Re c 0:5 106 .

BREITSAMTER AND ALLEN 179
http://-/?-http://-/?-


6/14

nominal de ection angle, indicating a stable, straight,and levelight condition.

Wake Flow eld

The axial vorticity eld at station x 5:6 is evaluated for four cases, namely, twosymmetrical andtwo asymmetricaloscillations at two frequencies, f osc ;1 4 Hz , kosc ;1 0:199 and f osc ;2 12 Hz ,kosc;2 0:596 , Fig. 6. The results for these two frequency values are

selected because they are related to the Crow-type instability for the(rolled-up) main vortex system (kCrow 0:080:1, kosc ;1 2kCrow )and the Crouch-type instability for the vortex system consisting of the neighboring corotating winglet vortex and main vortex(kCrouch 0:120:6).

The vorticity elds indicate that the peak vorticity level attributedto themainvortex (mergedOFV/ONV/WTV) is increased comparedto the baseline case, compare Figs. 4 and 6. This raise in peakvorticity is between 5 and32%. TheWLV, whichwasstillvisible for the baseline conguration, has disappeared in the case of the lower oscillation frequency (kosc 0:199 , Figs. 6a and 6c. Here, the WLV

has nearly completely merged with the main vortex leading to anincreased peak vorticity level. In contrast, the WLV can still bedetected at the higher oscillation frequency (kosc 0:596 , Figs. 6band 6d). The peak vorticity level is of the same magnitude as for thebaseline case (Fig. 4), but the WLV has progressed further in itscounterclockwise rotation with respect to the baseline. The vorticitysheet connecting the WLV and the main vortex indicates theadvancedmergingof thetwo vortices.Theshapeandsizeof themainvortex does not differ markedly from the baseline conguration

neither in vorticity nor in turbulence intensity distribution, Fig. 7.The presence of the WLV in case of the higher oscillation frequencycan also be seen in the turbulence intensity pattern by an ellipticregion of moderate turbulence intensity. Compared to the baselinecase, increased turbulence levels between 17 and 20% are found for the main vortex core, except for the asymmetrical case at kosc 0:596 .

Figure 8 summarizes the downstream development of nondimen-sional axial peak vorticity max , vortex core radius rc =b, axialvelocity de cit U 1 umin =U 1 , and peak turbulence intensitiesTu x;y;zmax for the main vortex as a function of nondimensional

y*

z *

0.6 0.7 0.8 0.9 1 1.1-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

OFV/ONV/WTV31.7

y*

z *

0.6 0.7 0.8 0.9 1 1.1-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

OFV/ONV/WTV32.5

WLV1.6

y*

z *

0.6 0.7 0.8 0. 9 1 1 .1-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

OFV/ONV/WTV39.8

y*

z *

0.6 0.7 0.8 0.9 1 1.1-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

OFV/ONV/WTV33.4

WLV1.3

a) b)

c) d)

Fig. 6 Contour plots of nondimensional axial vorticity ( ) distributions at x 5:6; Re c 0:5 10 6 ; 1 50; 1; a) symmetricaloscillation at k 0:199; b) symmetrical oscillation at k 0:596; c) asymmetrical oscillation at k 0:199; d) asymmetrical oscillation at k 0:596.

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


7/14

distancex and time . Data corresponding to staticapde ectionsare given in [32]. All results are obtained on measuring grids withidentical spatial resolution. Figure 8a illustrates the decay in peakvorticity for the main vortex and the HTV due to turbulent diffusion(d = d jMV 167 ). The decay rate of the HTV is clearly higher, asthe decay process is mainly inuenced by the interaction with thehighly turbulent fuselage wake. The peak vorticity levels for allvorticesexcept themainvortexdecreasedownstream.Vorticityfromother vortices,mainly theONF andWTV, is fed duringmerging intothe main vortex keeping its peak vorticity on a high level. The coreradius is de ned as the distance from the vortex center with zerocross ow velocity to the radial point of maximum crossowvelocity. The vortex core size shows a signicant increase upstreamof x 3:0 followed by a drop progressing downstream. The strongincrease in vortex core size is associated with the turbulent mergingof dominant vortices (ONF and WTV) [37]. After roll up of thevorticity sheets the vortex core radius may decrease to some extent.At x 3:0,themergingoftheWLVwiththemainvortexstarts.Theaxial velocity decit shown in Fig. 8c illustrates a decreasing decit up to x 4:0, and thereafter the decit increases to approximately

11%. This axial velocity decit is a characteristic feature of thetrailing vortices reecting the retarded axial velocity due to the wingboundary layers and local trailing-edge ow separation. Figure 8dpresents the peak turbulence intensities of all three velocitycomponents. At x 5:6 such peak turbulence intensities reachlevels between 5 and 15%.

The results for the cases with winglet ap oscillations taken at station x 5:6 are included in Fig. 8 and listed in Table 1. Thevalues for static ap de ections are also presented in Table 1.Winglet ap oscillations lead to an increase in peakvorticity relativeto the baseline case. Higher peak vorticity levels result from static

ap de ections because additional stationary single vortices arecreated enhancing the merging process. Compared to the baselinecase, the vortex core radius decreases stronger for oscillating apsthan forstatic apde ections, re ecting again theaccelerated roll-upprocess. A signicant difference is shown for the axial velocityde cit for which static ap de ections exhibit higher values withrespect to the baselinecase whereas reduced values are found for thecase of oscillating aps, the latter due to enhanced turbulent mixing.The peak axial turbulence intensity of the vortex core area stays

y*

z *

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

OFV/ONV/WTV0.199

y*

z *

0.6 0.7 0.8 0.9 1.11 0.6 0.7 0.8 0.9 1.11

0.6 0.7 0.8 0.9 1.110.6 0.7 0.8 0.9 1.11

-0.3

-0.2

-0.1

0

0.1

0.2

OFV/ONV/WTV0.174

WLV0.032

y*

z *

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

OFV/ONV/WTV0.175

y*

z *

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

OFV/ONV/WTV0.144

WLV0.026

a) b)

c) d)

Fig. 7 Contour plotsof turbulence intensity( Tu z) distributions at x 5:6; Re c 0:5 10 6 ; 0 :02 Tu z 0:2; Tu z 0:01; a) symmetricaloscillationat k 0:199; b) symmetrical oscillation at k 0:596; c) asymmetrical oscillation at k 0:199; d) asymmetrical oscillation at k 0:596.

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


8/14

almost constant, whereas the lateral and vertical turbulenceintensities increase in their levels markedly. An increase in lateraland vertical turbulence intensity is also present for static apde ections,butthermsvaluesaregenerallynotashighasforthecase

of oscillating aps. Thedata substantiate thefact that theoscillationsof the winglet aps can enhance vortex merging and lead to higher rms levels in velocity perturbations at the core of the remainingrolled-up vortex.

x*

m a x

0 1 2 3 4 5 6-80

-60

-40

-20

0

20

40

60

Baseline, Main VortexBaseline, HTVSymmetrical k=0.199Symmetrical k=0.596Asymmetrical k=0.199Asymmetrical k=0.596

x*

r c / b

0 1 2 3 4 5 60

0.01

0.02

0.03

0.04

0.05

Main VortexSymmetrical k =0.199Symmetrical k =0.596Asymmetrical k= 0.199Asymmetrical k= 0.596

x*

( U

- u m i n

) / U

0 1 2 3 4 5 60

0.05

0.1

0.15

0.2

Main VortexSymmetrical k=0.199Symmetrical k=0.596Asymmetrical k=0.199Asymmetrical k=0.596

x*

T u x , T

u y ,

T u

z

0 1 2 3 4 5 60

0.1

0.2

0.3

0.4

Main Vortex; Tu xMain Vortex; Tu yMain Vortex; Tu zSymmetrical k=0.199Symmetrical k=0.596Asymmetrical k=0.199Asymmetrical k=0.596

a) b)

c) d)

Fig. 8 Development of ow eldvariables between x 1:0and x 5:6; Re c 0:5 106 ; a) max forthe main vortexand theHTV;b) r c= b forthe mainvortex; c) U 1 u min =U 1 for the main vortex; d) Tu x;max , Tu y;max , and Tu z;max for the main vortex.

Table 1 Flow variables for core region of main vortex at x 5:6

Case name max rc =b U 1 umin =U 1 Tu x Tu y Tu zBaseline 30.1 0.034 0.111 0.050 0.153 0.146Static ap de ections [32]Symmetrical inboard 47.3 0.024 0.114 0.044 0.179 0.153Symmetrical outboard 42.4 0.028 0.126 0.044 0.165 0.150Asymmetrical inboard 46.4 0.025 0.126 0.044 0.135 0.150Asymmetrical outboard 44.0 0.025 0.157 0.042 0.144 0.147Oscillating apsSymmetrical k 0:199 31.7 0.031 0.043 0.043 0.169 0.199

Symmetrical k 0:596 32.5 0.028 0.029 0.048 0.193 0.174Asymmetrical k 0:199 39.8 0.031 0.036 0.047 0.184 0.175Asymmetrical k 0:596 33.4 0.029 0.037 0.046 0.170 0.144



9/14

Spectral Analysis

Power spectral densities are analyzed to detect peaks associatedwith the velocity perturbations introduced into the vortex wake bythe oscillating winglet aps. They are aimed to enhance thedevelopment of inherent wake instability mechanisms. SpectraldensitiesSN u0 of theaxial velocityuctuationsu0 areevaluatedfor thecore region of the WLV at x 0:02 and x 0:37 and also for thecore region of the main vortex at x 5:6. The calculation uses fast Fourier transformation and linear band averaging with n B 1024

frequency bands. The spectral values are normalized with thevariance of the velocityuctuationsu02 andthefrequencyinterval of band averaging f f M = 2nB kU 1 = b=2 , therefore multi-plied by kU 1 = u02 b=2 . These nondimensional spectraldensities are plotted as a function of reduced frequency k andpresented for both symmetrical and asymmetrical oscillations of thewinglet aps.

Considering the WLV core at the most upstream station(x 0:02), distinct spectral peaks can be detected at reducedfrequencies corresponding to the winglet ap oscillation frequenciesand at reduced frequencies associated with the higher harmonics,

Fig. 9. The symmetrically, that is, in-phase, deected aps producelarger spectral peaks than the asymmetrically deected aps with180deg phaseshift.In that case, thelower andupper winglet apsarede ected in oppositedirections creatingadditional sideedgevorticesso that turbulent mixing diminishes the spectral values related to theoscillation frequencies. Moving downstream to the WLV core at station x 0:37, Fig. 10 reveals even increased values of the well-de nedspectralpeaks causedbyap oscillations. Thespectralpeaksof the symmetrical case show values up to 18.8, which are again

higher compared to the asymmetrical case with peak values of 5.4.Concentratingonthemainvortex core at stationx 5:6,thespectraof the symmetrical case exhibit signicant energy overshoots at thefrequencies of the winglet ap oscillations for all ve consideredreduced frequencies, Fig. 11. For the asymmetrical case, smaller spectral peaks are observed increasing in amplitude with increasingoscillation frequency. Overall, the velocity uctuations ingested at x 0:0 by the oscillating winglet aps into the vortices at the wingtip are still present in the rolled-up vortex at x 5:6. The relativelevelsof thesefrequency dependentvelocitiesare in therange of10%with respect to the freestream velocity.

k

S N u

10 -1 10 0 10 110 -3

10-2

10 -1

10 0

10 1

10 2

k=0.099k=0.199k=0.497k=0.994k=1.987

a)

b)

k

S N u

10 -1 10 0 10 110 -3

10 -2

10 -1

10 0

10 1

10 2

k=0.099k=0.199k=0.497k=0.994k=1.987

Fig. 9 Normalized power spectral densities of the axial velocityuctuations at x 0:02; Re c 0:5 10 6 ; a) symmetrical oscillations;

b) asymmetrical oscillations.

k

S N u

10-1

100 10

110-3

10-2

10 -1

10 0

101

10 2

k=0.099k=0.199k=0.497k=0.994k=1.987

k

S N u

10 -1 10 0 10 110

-3

10 -2

10 -1

100

101

10 2

k=0.099k=0.199k=0.497k=0.994k=1.987

a)

b)Fig. 10 Normalized power spectral densities of the axial velocity

uctuations at x 0:37; Re c 0:5 10 6 ; a) symmetrical oscillations;b) asymmetrical oscillations.



10/14

A spectral investigation for a complete frequency band of winglet ap oscillations is conducted for the rolled-up (main) vortex at

station x 5:6. The relative difference between the spectra for the

cases of oscillating winglet aps and baseline conguration isanalyzed for axial, lateral, and vertical velocity uctuations.Considering, for example, the axial velocity uctuations thisdifference is given by SN u0osc k S

N u0baseline

k =SN u0baseline k . Therelationship corresponds to discrete Fourier transformation of ergodic time series with same sampling frequency and samplinginterval[38].ResultsareplottedinFigs.12and13asafunctionofthereduced frequency k and the reduced oscillation frequency kosc .Figure12refersto thecaseof symmetricalaposcillations,Fig.13tothe asymmetrical case. As proven in Fig. 11, the spectral peaksassociated with the oscillation frequencykosc are mostlymuch larger than the corresponding values of the baseline conguration(oscillation off). Consequently, the relativedifference between thesevalues results in local maxima along a diagonal line, indicated byvalues beyond 2. This diagonal trace reects a synchronous forcingof desired disturbancesby introducing frequency dependentvelocity

uctuations. In addition, vertical traces of peaks in the spectraldifferences are detected, for example, at reduced frequencies of

k

S N u

10-1

100 10

110-3

10-2

10 -1

10 0

101

102

k=0.099k=0.199k=0.497k=0.994k=1.987

k

S N u

10 -1 10 0 10110

-3

10 -2

10-1

100

10 1

102

k=0.099k=0.199

k=0.497k=0.994k=1.987

a)

b)Fig. 11 Normalized power spectral densities of the axial velocity

uctuations at x 5:6; Re c 0:5 10 6 ; a) symmetrical oscillations;b) asymmetrical oscillations.

k

k o s c

0 0.5 1 1.5 2 2.5 30

0.5

1

1.5

2

k

k o s c

0 0.5 1 1.5 2 2.5 30

0.5

1

1.5

2

k

k o s c

0 0.5 1 1.5 2 2.5 30

0.5

1

1.5

2

a)

b)

c)

Fig. 12 Relative difference of the power spectral densities for different frequenciesof symmetrical ap oscillationswith referenceto thebaselinecon guration at x 5:6; Re c 0:5 10 6 ; isolinesplotted for1, 2,3, 4,5,10, 15, 20; a) Su 0 ; b) Sv0 ; c) Sw0 .



11/14


12/14

correlated with the second probe at station x2 5:6. The probes arepositioned in the core region of the WLV (x1 0:02) and the mainvortex (x2 5:6) at the same locations as taken for evaluation of thepower spectral densities shown in Figs. 9 and 11.

Normalized cross spectral densities SN c;u 0 of the axial velocityuctuations u0 are plotted against the reduced frequency k, Fig. 14.

The calculation of the cross-spectral densities is performed usingthesame parameters as for the power spectral densities. Results areshown again for symmetrical and asymmetrical winglet aposcillations. Distinct spectral peaks are clearly visible at reducedfrequencies corresponding to the oscillation frequencies and their higher harmonics. The amplitudes of these peaks are comparablyhigh for both types of oscillations indicating the dominance of theforced oscillations beingconvected fromnear-eldto extendednear-

eldpositions.Figure15containsthephase angle distributionsof theaxial velocity uctuations as a function of the reduced frequency for the asymmetrical oscillations. Phase angles at the values of thereduced oscillation frequencies are within a band of 180 30 deg .

ConclusionsThe wake vortex development of a large transport aircraft in

approach con guration tted with large winglet and oscillatingwinglet aps is investigated by experimental means. A detailed four engined half-model of 1:32 scale is studied at an angle of attack of 6.5 deg and a Reynolds number based on the mean aerodynamicchord of 0:5 106. Vorticity and turbulence intensity as well asspectral density and cross-spectral density distributions areanalyzedin the near and extended near eld based on advanced hot-wireanemometry. The winglet ap oscillations are aimed to introducedesired perturbations in the wake ow eld to promote and amplify

the development of inherent far-eld wake instabilities. It is shownthat the generated near-eld disturbances (at x 0) are still present at the farthest downstream plane (at x 5:6). The oscillationfrequencies are chosen to match the frequency range of the wakeinstabilities. The main results of this study can be summarized asfollows:

1) The wake near eld shows seven particular vortices, namely,from outboard to inboard, the WLV, the WTV, the OFV, the ONVand INV, the IFV/WFV, and the HTV. These vortices are connectedthrough thevortexsheet shed at thewing trailing edge, except fortheHTV. Positive axial vorticity is attributed to WLV, WTV, OFV,ONV, and INV, whereas IFV/WFV and HTV show negative axialvorticity. The latter is caused by the negative circulation gradient at the wing-fuselage junction and the negative lift of the horizontal tailplane needed for trimmed ight. The highest level of axial peakvorticity is given by the OFV because of the highly loaded outboardap. It is located close to the free circulation centroid which is thewake vortex roll-up center.

2) In the extended near eld the position and strength of theparticular vortices change progressively because of the wake roll-upprocess and vortex merging. This process forms the remainingrolled-upvortex to whichmost of thevorticity is fedup to 0:16.The core regions of the particular wake vortices are characterized byaxial vorticity peaks and considerable axial velocity decits (811%). The vortex core structure exhibits maximum turbulenceintensities on the order of 520%.

3) Spectral analysis of the velocity uctuations shows a specicnarrowband concentration of turbulent kinetic energy in the core of the rolled-up vortex, mainly at the frequencies of the winglet aposcillations. The velocity spectra show peak values at thesefrequencies raising the perturbation level by a factor of 5 to 20compared to the case of static ap de ection or oscillation off. Thissynchronous forcing can amplify the long wave Crow-type

k

S c , u

10-1

100 10

110-3

10-2

10-1

10 0

101

102

k=0.099k=0.199k=0.497k=0.994k=1.987

k

S c , u

10-1 10 0 10 1

10-3

10 -2

10 -1

100

10 1

102

k=0.099k=0.199k=0.497k=0.994k=1.987

a)

b)Fig. 14 Normalized cross-spectral densities of the amplitude of theaxial velocity uctuations for stations x1 0:02 and x2 5:6; Re c 0:5 10 6 ; a) symmetrical oscillations; b) asymmetrical oscil-lations.

k

P h a s e a n g l e [ ]

10-1

100

1010

45

90

135

180

225

270

315

360

k=0.099k=0.199k=0.497k=0.994k=1.987

Fig. 15 Phase spectrum of cross-spectral densities of the axial velocityuctuations for asymmetrical winglet ap oscillations for stations x1

0:02 and x2 5:6; Re c 0:5 106 .



13/14

instability in which the time is reduced when the process of vortexlinking leading to rapid wake vortex decay starts.

4) Quasiperiodic uctuations occur alsoforasynchronous forcing.Fluctuations are amplied at a reduced frequency of k 0:1corresponding to the frequency of the Crow-type instability. Inaddition, increased amplitudes are present at k 0:3 which is in therange of reduced frequencies for the Crouch-type instability. Thisinstability is related to the mutual induction of two vortex pairs withcorotating neighbored vortices which is here themain trailing vortex

interactingwith theweakwinglet vortex. This interaction leads to theampli cation of the involved instabilities also contributing to wakevortex collapse.

As these investigations are limited to the range of the extendednear eld, further experiments are planned to prove the expectedenhanced wake vortexdecay in thefar eld.The focus isona towingtanktest usingan existingdetailedlarge transportaircraft model, thusincreasing the wake distance up to about 80 wing spans. In addition,the measured disturbance velocities are used as start conditions for stability analysis simulations.

Acknowledgment Thesupport of this investigation within theframeworkof IHK

Innovative High-Lift Congurations by the German Government,

Federal Ministry of Economics and Technology, under contract number 20A0301J is gratefully acknowledged.

References[1] Hnecke, K., Structure of a Transport Aircraft-Type Near Field

Wake, The Characterisation & Modi cation of Wakes from LiftingVehicles in Fluids , CP-584, AGARD, May 1996, pp. 5-15-9.

[2] Brenner, F.,AirTraf c Control Procedures forthe Avoidanceof WakeVortex Encounters: Today and Future Developments by DeutscheFlugsicherung GmbH, The Characterisation & Modi cationof Wakes from Lifting Vehicles in Fluids , CP-584, AGARD, May 1996, pp. 4-14-14.

[3] Gerz, T., Wake Vortex Prediction and Observation: Towards anOperational System, Proceedings of ODAS 2001, S1-3 , ONERA-DLRAerospace Symposium, ONERA, Chatillon, France, June 2001,ISBN 2-7257-006-X.

[4] Betz, A., Das Verhalten von Wirbelsystemen, Zeitschrift fr Angewandte Mathematik und Mechanik , Vol.12, No. 3, 1932, pp. 164174.doi:10.1002/zamm.19320120307

[5] Donaldson, C. P., and Bilanin, A. J., Vortex Wakes of ConventionalAircraft, AGARDograph, No. 204, May 1975.

[6] Devenport, W. J., Rife, M. C., Liapis, S. I., and Follin, G. J., TheStructure and Development of a Wing-Tip Vortex, Journal of Fluid Mechanics , Vol. 312, 1996, pp. 67106.doi:10.1017/S0022112096001929

[7] Spalart, P. R., Airplane Trailing Vortices, Annual Review of Fluid Mechanics , Vol. 30, 1998, pp. 107138.doi:10.1146/annurev. uid.30.1.107

[8] Rossow, V. J., Lift-Generated Vortex Wakes of Subsonic Transport Aircraft, Progress in Aerospace Sciences , Vol. 35, No. 6, 1999,pp. 507560.

doi:10.1016/S0376-0421(99)00006-8[9] Hnecke, K., From Formation to Decay Extended-Time WakeVortex Characteristicsof Transport-Type Aircraft, AIAA Paper 2002-3265, June 2002.

[10] Henke, R., Validation of Wing Technologies on an Airbus A340Flying Testbed: First Flight Test Results from the European ProgramAWIATOR, International Council of the Aeronautical Sciences(ICAS) Paper 2004-4.4.1, Sept. 2004.

[11] Coustols, E., Stumpf, E., Jacquin, L., Moens, F., Vollmers, H., andGerz, T., Minimised Wake: A Collaborative Research Programme onAircraft Wake Vortices, AIAA Paper 2003-0938, Jan. 2003.

[12] Gerz, T., Holzpfel, F., and Darracq, D., Commercial Aircraft WakeVortices, Progress in Aerospace Sciences , Vol. 38, No. 3, 2002,pp. 181208.doi:10.1016/S0376-0421(02)00004-0

[13] Kpp, F., Rahm, S., Smalikho, I., Dol, A., Cariou, J.-P., Harris, M.,

and Young, R. I., Comparison of Wake-Vortex Parameters Measuredby Pulsed and Continuous-Wave Lidars, Journal of Aircraft , Vol. 42,No. 4, 2005, pp. 916923.doi:10.2514/1.8177

[14] Winckelmans, G. S., Vortex Methods, The Encyclopedia of Computational Mechanics , edited by E. Stein, R. de Borst, and T. J. R.Hughes, Wiley, Vol. 3, 2004.

[15] Stumpf, E., Study of Four-Vortex Aircraft Wakes and Layout of Corresponding Aircraft Congurations, Journal of Aircraft , Vol. 42,No. 3, 2005, pp. 722730.doi:10.2514/1.7806

[16] Holzpfel, F., Hofbauer, T., Darracq, D., Moet, H., Garnier, F., andGago, C. F., Analysis of Wake Vortex Decay Mechanism in theAtmosphere, Aerospace Science and Technology ,Vol.7,No.4,2003,

pp. 263275.doi:10.1016/S1270-9638(03)00026-9[17] Breitsamter,C., Bellastrada,C., andLaschka,B.,Investigationson the

Turbulent Wake Vortex Flow of Large Transport Aircraft, ICAS Proceedings, 23rd International Congress of the Aeronautical Sciences , ICAS, Sept. 2002, pp. 382.1382.13.

[18] Elsenaar, A., The Optimum Wing Load Distribution to MinimizeWake Hazard, Conference on Capacity and Wake Vortices , ImperialCollege, London, Sept. 2001.

[19] Breitsamter, C., Beein ussung des Nachlaufwirbelsystems vonGro ugzeugen, Proceedings DGLR-Symposium: The Airport as a Neighbour Challenges and Solutions for future Air Transport Systems , DGLR Rept. 2004-06, Oct. 2004.

[20] Jacquin, L., Fabre, D., Sipp, D., Theolis, V., and Vollmers, H.,Instabilities and Unsteadiness of Aircraft Wake Vortices, AerospaceScience and Technology , Vol. 7, No. 8, 2003, pp. 577593.doi:10.1016/j.ast.2003.06.001

[21] Rennich, S. C., and Lele, S. K., Method for Accelerating theDestruction of Aircraft Wake Vortices, Journal of Aircraft , Vol. 36,No. 2, 1999, pp. 398404.doi:10.2514/2.2444

[22] Durston, D. A., Walker, S. M., Driver, D. M., Smith, S. C., and Savas,., Wake Vortex Alleviation Flow FieldStudies, Journal of Aircraft ,Vol. 42, No. 4, 2005, pp. 894907.doi:10.2514/1.7904

[23] Fabre, D., and Jacquin, L., Stability of a Four-Vortex Aircraft WakeModel, Physics of Fluids , Vol. 12, No. 10, 2000, pp. 24382443.doi:10.1063/1.1289397

[24] Crow, S. C., and Bate, E. R., Jr., Lifespan of Trailing Vortices in aTurbulent Atmosphere, AIAA Journal , Vol. 13, No. 7, 1976, pp. 476482.

[25] Crouch, J. D., Miller, G. D., and Spalart, P. R., Airplane TrailingVortices, Boeing Aero Magazine , No. 14, 2001, pp. 18.

[26] Crouch, J. D.,Miller, G. D.,and Spalart, P. R.,Active-Control Systemfor Breakup of Airplane Trailing Vortices, AIAA Journal , Vol. 39,No. 12, 2001, pp. 23742381.doi:10.2514/2.1244

[27] Haverkamp, S., Neuwerth, G., and Jacob, D., Active and PassiveVortex Wake Mitigation Using Control Surfaces, Aerospace Scienceand Technology , Vol. 9, No. 1, 2005, pp. 518.doi:10.1016/j.ast.2004.08.005

[28] Kauertz, S., and Neuwerth, G., Excitation of Instabilities in the Wakeof an Airfoil with Winglets, AIAA Journal , Vol. 45, No. 3, 2007,pp. 577598.doi:10.2514/1.26462

[29] Vo, G., Stumpf, E., Konrath, E., v. Carmer,C., Krckeberg, C. P., andMeyer, H., Wake Vortex Alleviation by Differential and OscillatingFlap Settings:A Numerical and ExperimentalStudy, TheDLR Project Wake Vortex Detecting, Characterizing, Controlling, Attenuatingand Predicting Aircraft Wake Vortices , Deutsches Zentrum fr Luft-und Raumfahrt e.V., DLR Research Report 2008-15, 2008, pp. 126(ISSN 1434-8454).

[30] Greenblatt, D.,ManagingFlapVorticesviaSeparationControl, AIAA Journal , Vol. 44, No. 11, 2006, pp. 27552764.doi:10.2514/1.19664

[31] Vey, S., Paschereit, O. C., Greenblatt, D., and Meyer, R., Flap VortexManagement by Active Gurney Flaps, AIAA Paper 2008-286,Jan. 2008.

[32] Allen, A., and Breitsamter, C.,TransportAircraft Wake Inuenced bya Large Winglet andWingletFlaps, Journal of Aircraft ,Vol.45,No.2,2008, pp. 686699.doi:10.2514/1.32787

[33] Breitsamter, C., Turbulente Strmungsstrukturen an Flugzeugkon-gurationen mit Vorderkantenwirbeln, Ph.D. Thesis, DM 18432,Technische Universitt Mnchen, Herbert Utz Verlag Wissenschaft

(Aerodynamik), Germany (ISBN 3-89675-201-4), June 1997.[34] Crow, S. C., Stability Theory for a Pair of Trailing Vortices, AIAA Journal , Vol. 8, No. 12, 1970, pp. 21722179.doi:10.2514/3.6083

http://dx.doi.org/10.1002/zamm.19320120307http://dx.doi.org/10.1017/S0022112096001929http://dx.doi.org/10.1146/annurev.fluid.30.1.107http://dx.doi.org/10.1146/annurev.fluid.30.1.107http://dx.doi.org/10.1146/annurev.fluid.30.1.107http://dx.doi.org/10.1016/S0376-0421(99)00006-8http://dx.doi.org/10.1016/S0376-0421(02)00004-0http://dx.doi.org/10.2514/1.8177http://dx.doi.org/10.2514/1.7806http://dx.doi.org/10.1016/S1270-9638(03)00026-9http://dx.doi.org/10.1016/j.ast.2003.06.001http://dx.doi.org/10.2514/2.2444http://dx.doi.org/10.2514/1.7904http://dx.doi.org/10.1063/1.1289397http://dx.doi.org/10.2514/2.1244http://dx.doi.org/10.1016/j.ast.2004.08.005http://dx.doi.org/10.2514/1.26462http://dx.doi.org/10.2514/1.19664http://dx.doi.org/10.2514/1.32787http://dx.doi.org/10.2514/3.6083http://dx.doi.org/10.2514/3.6083http://dx.doi.org/10.2514/1.32787http://dx.doi.org/10.2514/1.19664http://dx.doi.org/10.2514/1.26462http://dx.doi.org/10.1016/j.ast.2004.08.005http://dx.doi.org/10.2514/2.1244http://dx.doi.org/10.1063/1.1289397http://dx.doi.org/10.2514/1.7904http://dx.doi.org/10.2514/2.2444http://dx.doi.org/10.1016/j.ast.2003.06.001http://dx.doi.org/10.1016/S1270-9638(03)00026-9http://dx.doi.org/10.2514/1.7806http://dx.doi.org/10.2514/1.8177http://dx.doi.org/10.1016/S0376-0421(02)00004-0http://dx.doi.org/10.1016/S0376-0421(99)00006-8http://dx.doi.org/10.1146/annurev.fluid.30.1.107http://dx.doi.org/10.1017/S0022112096001929http://dx.doi.org/10.1002/zamm.19320120307


14/14

[35] Crouch, J. D., Instability and Transient Growth for Two Trailing-Vortex Pairs, Journal of Fluid Mechanics , Vol. 350, 1997, pp. 311330.doi:10.1017/S0022112097007040

[36] Breitsamter, C., Nachlaufwirbelsysteme groer Transport ugzeugeExperimentelle Charakterisierung und Beeinussung. Habilitations-schrift, Technische Universitt Mnchen, Herbert Utz Verlag(Aerodynamik), Germany (ISBN 3-8316-0713-3), Feb. 2007.

[37] Meunier, P., and Leweke, T., Three-Dimensional Instability DuringVortex Merging, Physics of Fluids , Vol. 13, No. 10, 2001, pp. 27472750.doi:10.1063/1.1399033

[38] Marple, S. L., Jr., Digital Spectral Analysis with Applications ,PrenticeHall, Englewood Cliffs, NJ, 1987, pp. 2556, 111125, 386394.

http://dx.doi.org/10.1017/S0022112097007040http://dx.doi.org/10.1063/1.1399033http://dx.doi.org/10.1063/1.1399033http://dx.doi.org/10.1017/S0022112097007040

Documents

Transport Aircraft Wake Influenced by Oscillating Winglet Flaps