Analysis on Aerodynamic and Acoustic Characteristics for

The 2nd Asian/Australian Rotorcraft Forum and The 4th International Basic Research Conference on Rotorcraft Technology

Tianjin, China, September 08-11, 2013

105

Analysis on Aerodynamic and Acoustic Characteristics for Scissors

Tail-Rotors in Forward Flight

Fan Feng Shi Yong-Jie Doctoral Candidate Associate Professor National Key Laboratory National Key Laboratory

of Science and Technology on of Science and Technology on Rotorcraft Aeromechanics, PO box 331 Rotorcraft Aeromechanics, PO box 331

Nanjing University of Aeronautics and Astronautics Nanjing University of Aeronautics and Astronautics College of Aerospace Engineering, Nanjing, China College of Aerospace Engineering, Nanjing, China

(Tel. (+86) 139-5185-2324) (Tel. (+86) 180-6144-6188) (e-mail: [email protected]) (e-mail: [email protected])

Xu Guo-Hua Ye Zhou

Professor Graduate National Key Laboratory National Key Laboratory

of Science and Technology on of Science and Technology on Rotorcraft Aeromechanics, PO box 331 Rotorcraft Aeromechanics, PO box 331

Nanjing University of Aeronautics and Astronautics Nanjing University of Aeronautics and Astronautics College of Aerospace Engineering, Nanjing, China College of Aerospace Engineering, Nanjing, China

(Tel. (+86) 139-1395-2863) (Tel. (+86) 151-6147-1778) (e-mail: [email protected]) (e-mail: [email protected])

ABSTRACT

Numerical calculations were performed on the aerodynamic and aeroacoustic characteristics of a scissored tail-rotor in forward flight. A newly established Computational Fluid Dynamics(CFD) model was adopted to simulate the unsteady flowfield and the aerodynamic characteristics of a scissored tail-rotor. Based on the computed flowfield, the aeroacoustic characteristics of a scissored tail-rotor were analyzed using the Farassat Formulation 1A, which is derived from the FW-H equation. In addition, two different scissored tail-rotor configurations, i.e., the L-configuration and the U-configuration, were studied in details. The simulated aerodynamic and aeroacoustic characteristics of a scissored tail-rotor were compared with those for a conventional one. The influence of the scissor angle was also investigated. The simulation results demonstrated that a scissored tail-rotor is significantly different from the convention one in terms of the flowfield, the aerodynamic force and the aeroacoustic characteristics. Also, the scissor angle has an important effect on the aerodynamics and aeroacoustics of a scissored tail-rotor.

INTRODUCTION

As scissored tail-rotor is an unconventional tail-rotor configuration, which is mainly installed on armed helicopters such as the AH-64 Apache and the Mi-28 Havoc. A major feature of scissored tail-rotors is the uneven blade azimuthal spacing for both the upper and the lower pair of blades. In the

literature, some previous researches have demonstrated that scissored tail-rotors have certain advantages in the noise reduction when compared with the conventional ones, which indicates its application potentials. However, most of the previous researches on helicopter tail-rotors were focused on the conventional configurations which apply an equal azimuthal spacing and only a limited amount of work has been carried out on the



106

scissored configurations. Therefore, it is essential and beneficial to have a comprehensive study of the scissored tail rotor configurations for a better understanding of its aerodynamic and aeroacoutistic characteristics for practical tail rotor design applications. So far, most of the researches on the scissored tail-rotors have been mainly focused on hover condition, and only a limited amount of studies were carried out at forward flight condition. Unfortunately, some conclusions led by different researchers were contradictory. For example, Sonneborn [1] conducted an experimental study on the scissors rotor in 1974, and his results indicated that the scissored rotor had no advantages in aerodynamics and acoustics when compared with the conventional configurations. Rozhdestvensky et al [2] carried out a new experimental research on the hover performance of scissors tail-rotor in 1996, and compared the noise level of Mi-28 helicopter equipped with both a conventional tail-rotor and a scissored one. However, their results were quite different from the Sonneborn’s. In contrast, they concluded that the scissored configuration has a superior aerodynamics and a lower level of acoustic noise when compared with the conventional one. The discrepancies between the studies of Sonneborn and Rozhdestvensky were thought to be due to the higher rotational speed of Rozhdestvensky’s rotor model. In 2007, an experiment was carried out to measure the acoustics and induced velocity characteristics of a scissored rotor at the Nanjing University of Aeronautics and Astronautics [3], including a numerical simulation of the hover performance for the test rotor using a free-wake method [4]. The experimental and numerical research demonstrated that the configuration parameters had an important effect on both the aerodynamics and the aeroacoustics of scissored rotors.

Different from most of the previous researches which were mainly focused on hover conditions, this paper aims to conduct numerical studies on both the aerodynamic and the aeroacoustic characteristics of the scissored tail-rotors in forward flight. In addition, the effect of the configuration parameters of scissored tail-rotors is also investigated.

COMPUTATIONAL METHODOLOGY AND VALIDATION

AERODYNAMIC MODEL

A CFD computational model is adopted to simulate the unsteady flowfield of scissored tail-rotors in forward flight. The three-dimensional unsteady Reynolds-Average Navier-Stokes (RANS) equations were employed as the governing equations, which can be written as

[ ]( ) ( ) 0V S

dV dSt ∂

∂+ − =

∂ ∫∫∫ ∫∫W F W G W (1)

Where W is the vector of conserved variables, ( )F W and ( )G W are the convective (inviscid) and the viscous flux vectors, respectively. In the current CFD solver, a second-order upwind scheme(Roe scheme) [5] is used to calculate the convective fluxes on the faces of the control volume. For the time integration, a dual-time stepping method is applied to simulate the unsteady flow phenomenon, together with the LU-SGS scheme [6] at every pseudo-time step. The Spalart-Allmaras one equation model [7] is used as the turbulent model, which is uncoupled with the flow governing equations. An overset grid system is used to solve the rotation and the pitch motion of the tail-rotor blades. In the grid system, a C-O type grid is used for the body-fitted blade grid (the near grid) and a Cartesian type is applied as the background grid, respectively.



107

The topological relationship between the blade near-body grid and the background grid is established using the hole cutting and donor cell identifying process, and the flowfiled data between the two set of grids are exchanged using a high-order interpolation scheme.

ACOUSTIC MODEL

Since the current research is mainly focused on the acoustic characteristics of scissored tail-rotor at a moderate-speed flight condition, the well-known Farassat Formulation 1A [8] (which is based on the FW-H equation) is adopted to predict the noise level of the scissored tail-rotor, which can be formulated as

' 02

0

20 0 0

2 30

4 ( , )(1 )

( )(1 )

nT

rf ret

n i i r

rf ret

vp t dSr M

v rM r c M c M dSr M

ρπ

ρ

=

=

=

−

+ −+

−

∫

∫

x (2)

'2

0 0

2 20

20 0

2 30 0

14 ( , )(1 )

(1 )

( )1(1 )

i iL

rf ret

r i i

rf ret

r i i r

rf ret

l rp t dS

c r M

l l MdS

r M

l rM r c M c MdS

c r M

π=

=

=

=

−

−+

−

+ −+

−

∫

∫

∫

x

(3)

where ' ( , )Tp tx is the thickness noise due to the periodic motion of blade, ' ( , )Lp tx is the loading noise caused by the airloads on the blade surface. Based on Equations (2) and (3), a numerical acoustic code has been developed for scissors tail-rotor.

SIMULATION RESULTS AND VALIDATIONS

Due to the lack of available tail-rotor experimental data in forward flight, the benchmark case for the Lynx tail rotor in a hover condition [9] was selected to validate the current CFD code for aerodynamics predictions. The advance ratio can be set to

zero in the current CFD code to simulate the flowfiled in hover. For example, Figure 1 compared the CFD predicted thrust and torque coefficients of the tail-rotor with the corresponding experimental data [9]. The excellent correlation between the calculation and the measurements validates the capability of the CFD analysis.

-16 -12 -8 -4 0 4 8 12 16-0.015

-0.010

-0.005

0.000

0.005

0.010

0.015

0.020

CT

Collective pitch（ °）

Experiment Calculated

(a) Thrust coefficients versus collective pitch

angle

-0.08 -0.06 -0.04 -0.02 0.00 0.02 0.04 0.06 0.08 0.100.000

0.004

0.008

0.012

0.016

CQ/σ

CT/σ


(b) Torque coefficients versus thrust

coefficients

Fig.1 Comparison of the hover performance between CFD calculations and experiment [9]

for the Lynx tail-rotor To validate the current acoustic model, the noise level of AH-1/OLS rotor[10]was calculated and compared with available experimental data [10]. A BVI test condition (No.10014) was selected as the numerical benchmark, and the results were shown in Figure 2. The comparison between the calculated results and experimental data [10] demonstrated that, the present method



108

can predict the representative characteristics of BVI acoustics effectively except for some discrepancies in the detailed magnitude. As shown, the correlations between the CFD simulation and the measurements also demonstrated the capability of the current CFD solver for the complex BVI flowfield, which is important for the analysis of interferences among scissored tail-rotor blades.

0 45 90 135 180-50

-25

0

25

50

Sound Pressure(Pa)

Azimuth Angle(deg)


#210014

(a) Microphone 2

0 45 90 135 180-50

-25

0

25

50

Sound Pressure(Pa)

Azimuth Angle(deg)


#310014

(b) Microphone 3

0 45 90 135 180-50

-25

0

25

50

Sound Pressure(Pa)

Azimuth Angle(deg)


#710014

(c) Microphone 7

0 45 90 135 180-50

-25

0

25

50

Soun

d Pr

essu

re(P

a)

Azimuth Angle(deg)


#910014

(d) Microphone 9

Fig.2 Comparison of the predicted and measured time histories of sound pressure

for the AH-1/OLS rotor in hover

Results and Discussion

The tail-rotor model used in Rozhdestvensky’s experiment was selected as the benchmark numerical example in the current study of the scissored tail rotors in forward flight conditions. The detailed rotor model parameters can be found in Ref. [2]. As pointed out in Ref. [2], there are two types of configurations for a scissored tail-rotor, that is, the L-configuration and the U-configuration. In the L-configuration, the leading blade is lower than the following one. In the U-configuration, the leading blade is higher than the following one. For convenience, the leading blade is further defined as blade 1, and the following blade is defined as blade 2. According to the definition, blade L-1 represents the leading blade in the L-configuration. Similarly, the other blades can be defined as shown in Figure 3.

rotational direction

“L-1” blade

“L-2” blade

(a) L-configuration rotational direction

“U-1” blade

“U-2” blade

(b)U-Configuration



109

Scissors angle δ

(c) Top View

Vertical space h∆

(d) Side View

Fig.3 Schematic illustration of different configurations for a scissored tail-rotor

In the current study, the vertical space of the scissored tail-rotor is fixed at a constant value of Δh/R=0.1, in which R represents the tail-rotor radius. The overset grid system used for the scissored tail-rotor flowfiled simulation is shown in Figure 4.

Fig.4 Schematic illustration of the grid

system for the scissored tail-rotor

Flowfield characteristics

The BVI phenomenon for tail-rotor often

occurs at a forward flight condition [11]. The smaller blade space of scissored tail-rotors may result in a more severe BVI, which significantly affects the aerodynamics of tail-rotors. For example, Fig. 5 shows the CFD predicted representative snapshots for the tip vortex trajectories of both the L-configuration and the U-configuration at different reference rotor azimuth in a forward flight condition. It can be seen that the tip-vortex shed from the conventional tail-rotor distributes more uniformly than that from the scissored tail-rotor configuration, which is due to the nonuniform azimuthal space between the neighboring blades of a scissored tail-rotor. The severe blade-vortex and vortex-vortex interference will have important impacts on the tail-rotor aerodynamics. Comparing the front side (solid circle) and the aftward side (dashed circle) of wake structures in Figure 5, it can be seen that, for the scissored tail-rotor with both the L-configuration and the U-configuration, the vortices shed from the neighboring blades form a “vortex-pair” during the process of wake evolution due to the smaller blade spacing, which makes the wake structure of a four-bladed tail-rotor similar to that of a two-bladed tail-rotor. In addition, when vortices convect further away from the tail-rotor blades, the two vortex filaments approach each other. Furthermore, it can be seen from Figure 5 that, the downwash velocity of the scissored tail-rotor is lower than the case of conventional one at the same advance ratio.

(a) Conventional tail-rotor



110

(b) Ψ=30°（L-Configuration） (c) Ψ=30°（U-Configuration）

(d) Ψ=60°（L-Configuration） (e) Ψ=60°（U-Configuration）

(f) Ψ=90°（L-Configuration） (g) Ψ=90°（U-Configuration）

(h) Ψ=180°（L-Configuration） (i) Ψ=180°（U-Configuration）

Fig.5 Comparison of the tip vortex trajectory for a scissored tail-rotor between the L-configuration and the U-configuration (δ=30°) in a forward flight condition

Aerodynamic characteristics

Figure 6 presents the azimuthal variation of the tail-rotor thrust at different scissor angles, where the subscript ORG refers to the conventional (original) configuration, and L30 represents the L-configuration with a scissor angle of δ=30°. All the other labels can be defined similarly. When compared with the conventional one, it can be seen that the thrust of the scissored tail-rotor varies significantly with the azimuth angle, and the period is 2/Rev due to the symmetric layout of the lower/upper blades in the scissored tail-rotor. A smaller blade azimuthal spacing between the neighboring blades leads to a stronger aerodynamic interference. Meanwhile, the vertical space of the upper/lower blades can drive the tip-vortex shed from the leading blade be closer to the following blade, which may result in

a stronger aerodynamic interference. In addition, the calculation results show that the aerodynamic interference reduces with an increase in the scissor angle, which leads to a reduction of the variation amplitude of the tail-rotor thrust. It should be pointed out that the increased variation of the tail-rotor thrust will have an important effect on the performance and handling qualities of the helicopter installed with a scissored tail rotor.

0 90 180 270 3600.008

0.010

0.012

0.014

0.016

0.018

CT

Azimuth angle(deg)

ORG L30 L45 L60 L90

(a) L-configuration



111

0 90 180 270 3600.008

0.010

0.012

0.014

0.016

0.018CT

Azimuth angle(deg)

ORG U30 U45 U60 U90

(b) U-configuration

Fig.6 Effects of scissor angle on the unsteady tail-rotor thrust

Figure 7 shows the effect of the scissor angle on the time average tail-rotor thrust. The results were normalized using the mean thrust of the conventional configuration, i.e., the constant straight-line in Figure 7 represents the conventional tail-rotor. As shown, the thrusts of both the L-configuration and the U-configuration are consistently higher than that of the conventional one at all the scissor angles considered. In addition, the L-configuration shows superior aerodynamic performance when compared with the U-configuration, which may be explained by the reduced blade vortex interaction in the L-configuration (as shown in Fig.5).

Configuration L Configuration U

20 40 60 80 1000.98

1.00

1.02

1.04

1.06

1.08

CT/CTorg

Scissors angle(deg) Fig.7 Effects of scissor angle on the time

averaged tail-rotor thrust

Figure 8 shows the azimuthal (unsteady) variation of the thrust on individual blades of both the L-configuration and the U-configuration. As shown, the general

azimuthal variation trend of the scissored tail-rotor is similar to that of the conventional one. However, the blade thrusts of the scissored tail rotor in the azimuthal range of 90°-270° are significantly different from that of the conventional one, which can be explained by different patterns of blade vortex interaction. For example, in the first quadrant (0~90°) and the fourth quadrant (270°-360°) of the rotor disk, the tip-vortex shed from the preceding blade moves away from the rotor disk, and has minor influence on the following blade. In contrast, in the second and the third rotor quadrants (90°-270°), when the tip-vortex moves toward the rear of the tail-rotor disk, it encounters the successive blade, which may cause a stronger aerodynamic interference. In hover, the blade L-2 is far from the tip-vortex shed from the preceding blade, so it suffers a very little interference. However, as shown in Fig. 8, the blade L-2 suffers a very strong aerodynamic interference in a forward flight condition. It can be concluded that, within the first 90° of the wake age, the tip-vortex convects above the rotor disk, which may cause the blade L-2 being close to the tip-vortex shed from the blade L-1. Consequently, the blade L-2 suffers more intense interference in a forward flight condition when compared with the hover condition.

0 90 180 270 3600.000

0.002

0.004

0.006

0.008

Blade L-1

CT

Azimuth angle(deg)

ORG L30 L45 L60 L90

(a) Blade L-1



112

0 90 180 270 3600.000

0.002

0.004

0.006

0.008

Blade U-1

CT

Azimuth angle(deg)

ORG U30 U45 U60 U90

(b) Blade U-1

0 90 180 270 3600.000

0.002

0.004

0.006

0.008

CT

Azimuth angle(deg)

ORG L30 L45 L60 L90

Blade L-2

(c) Blade L-2

0 90 180 270 3600.000

0.002

0.004

0.006

0.008

Blade U-2

CT

Azimuth angle(deg)

ORG U30 U45 U60 U90

(d) Blade U-2

Fig.8 Effects of scissors angle on individual blade unsteady thrust

Figure 9 shows the mean thrust of individual blade versus the scissors angle. It can be clearly seen that the mean thrust of each individual blade has large difference from each other for both the L-configuration and the U-configuration. In addition, the difference decreases with an increase in the scissors angle. At all the scissor angles considered, the mean thrusts of the blade L-1 are close to that of the blade U-1. However, the mean thrust of the blade L-2 is consistently higher than that of blade U-2,

which makes the total thrust of the L-configuration higher than that of the U-configuration (as shown in Fig. 7).

20 40 60 80 1000.7

0.8

0.9

1.0

1.1

1.2

1.3

CT/CTorg

Scissors angle(deg)

L-1 L-2 U-1 U-2

Fig.9 Effects of scissors angle on the mean thrust

of individual blade Acoustic Characteristics

To study the aeroacoustic characteristics of scissored tail-rotors in forward flight, the thickness, loading and total noises at several scissors angles were calculated using the developed acoustics solver. The noise is computed on a plane of 100m×150m, 30m away below the tail-rotor disk. Fig. 10 compares the time history of the sound-pressures between the scissored and the conventional tail-rotors. As shown, the phase of the sound-pressure peaks is changed due to the uneven blade azimuthal spacing, which is called the “modulated effect” [12].

Thickness Loading Total

0 90 180 270 360-40

-30

-20

-10

0

10

20

Sound Pressure(Pa)

Azimuth angle(deg)

(a) Conventional tail-rotor



113

Thickness Loading Total

0 90 180 270 360-40

-30

-20

-10

0

10

20Sound Pressure(Pa)

Azimuth angle(deg) (b) Scissored tail-rotor

Fig.10 Comparison of the time history of the sound pressures between the conventional and

the scissored tail-rotors

Thickness noise

Fig.11 presents the effects of the scissor angle on the thickness noise of the tail-rotor in forward flight. Since the vertical separation space on the thickness noise is insignificant, the noise between the L-configuration and the U-configuration was not explicitly distinguished here. As shown in Fig. 11, the thickness noise level of the uneven spaced layout is lower than that of the conventional one. Also, the thickness noise increases with an increase in the scissor angle. In addition, Fig. 12 shows the maximum thickness SPL value on the observation plane versus the scissors angle. It can be clearly seen that the maximum thickness SPL increases monotonically with the scissor angle, which means that the so-called “modulation effect” is gradually reducing.

-75 -50 -25 0 25 50 75-50

-25

0

25

507380 727272747381 81 7475 74757683

79798384 7577

79 757880808385 83838686.585.5

86.585.5 75

85.587.587.5 757783

8687.587.5 808587.590 7790

87.5

87

89.587

8791

77

918987.590 86.5878790

8082838586.58788 85858787.58887.5

757885.586 828284868685.5858082838585 75

76777879848474767778818282

7677788383747476

8182 7678808181 7271717277717176

7778

(a) δ=30°

-75 -50 -25 0 25 50 75-50

-25

0

25

5072737780 738073

798181 72817376788083

79798085 8385858585 818283858586 778687 7687.587.58888.5 788485.587.589.58989.5 768989.5 85

89.5

90.590 8691.589.588.587.5

86

91

89.5

768290.5

84868890.5 8285.58789.589.5

787988.58786.5 788588887476778686.58787

77798485.5868685.5747679838585.58585

74767879808373747480 73778182

7271717576788081

717777787981717880

77 (b) δ=45°

-75 -50 -25 0 25 50 75-50

-25

0

25

50767980

707477798282 71817284 828485 74757885.585.5 74848485.586.58786.5 778887.5 74848888.588.5

828888 8084

86.59090 75778787.5

90.5

919088 788086.5

919291

819091.5

7787.5

8284877980818390

8989.589.57680

85.586.587.58989808085.586

85.587.587.5 757785.586.58686.586 758285.58585.585.585 74768185 7374

76798181 727273

72767783 71717277

8183707072818282

788181 71777073

78 (c) δ=60°

-75 -50 -25 0 25 50 75-50

-25

0

25

5073777881 707476818182

83 717074 7273 73737578 73848585.5857481838486.5 778386.587 758184868887.58886.5

76

76838587888988.5 808188.590 757786909091.590.5

88.5 86

86.5

91.5

89.5

89.5

91

8691

85.59292 8590.591

90.589.5

89.587.5 8185.5908486.5

8687.58888 777887.58888

7475777879818183858787

75777777858686.5747485 79

72717174757779848471718083 77

717172

727972747880

(d) δ=90° Fig.11 Effects of the scissor angle on the

thickness noise of the tail-rotor

20 30 40 50 60 70 80 90 10090

91

92

93

94

SPL(dB)

Scissors angle(°) Fig.12 Maximum SPL value of the thickness

noise versus the scissor angle

Loading noise

Fig.13 shows the contours of the loading noise SPL calculated on the observation plane, and Fig.14 presents the variation of the maximum SPL value on the observation



114

plane versus the scissor angle. As shown from Figs.13 and 14, as the scissor angle increases, the loading noise level of the tail-rotor decreases initially and then increases. At scissor angles of 45° and 60°, the tail-rotor loading noise levels of both the L-configuration and the U-configuration are lower than that of the conventional one. Also, the L-configuration is superior to the U-configuration. The uneven blade azimuthal spacing of scissors tail-rotors changed the phase difference among the sound pressure peaks, which may decrease the amplitude of

the some pressure peaks (as shown in Figure 10), and therefore decrease the noise level. In addition, the so-called “modulation effect” further modifies the discrete frequency spectrum, which spreads the acoustic energy into more extensive frequency ranges, and therefore decreases the noise level in low- and mid-frequencies. However, as pointed out in the previous section, a reduced blade azimuthal spacing will intensify the aerodynamic interference, and therefore, a larger loading noise in forward flight.

-75 -50 -25 0 25 50 75-50

-25

0

25

508890

8182

8791.590 81

9291.5

90.5

92.5

90

83

9191.5

93.593

.593

9185847889

7880828491

.5929088

758892.59075768484

84868986757989878382 80818381858684858378 7576787579

7876838275 777979818384888479

76 798187898477 9085

75859291.59392.5

91.577 828993

918479797878 8894

92

8079

88

91.59393.585

8684 868280 849383 818490.584 8391

.593 8291.585

81

91.587 899090

908987 (a) Conventional tail-rotor

-75 -50 -25 0 25 50 75-50

-25

0

25

509192

8487889092.5

88 83848793

92

92.5

89

8789

93.5

91.5

828385889184

9192 79

8192.59292918878828890.589 8188

89909081

78798688888681817879 757583

81868481

75788079838382 788283838584857879 77808488888687868585817975858987

8584 768491.5

9090.59090.586

868584 808092.59291.591898787 84899293.594 90

948986808491.5

92

92.5

94.5 82

87

93.5

93.588 90.5 8693.593.5 8889

92.592

91.5

899292 -75 -50 -25 0 25 50 75-50

-25

0

25

50838890

83

93.5

8991

84919393.585

81

90.5

92.5

92.5 8188

91.5

898786

91 79

91.5

83848692.5

9389777981828487929290.5

8883 798891878480

83909080808183868679

78 7978788284 78798283828080 848582

758086878790908780 7985909083 77818591.591.592.5898582 81849093.591.5

91.5 7679808492

94.591.590

.579808393

92.5

91 79838388 80

92.593

93.5 82

8295.5

91.587 88 8195

92.5

90.592.5

93.593 848887

92.594 909394

9388

92

92 (b) L-configuration，δ=30° (c) U-configuration，δ=30°

-75 -50 -25 0 25 50 75-50

-25

0

25

50

87

87

8489 8686

8590.5 8686 8888 8587

8885808691

868378

788890.588888582

818884 8085898785858180 79818685818178 7577788081848080

76 79828084828078

868484848181 7581888987878380 798284889088

86838382 8990.590.590.5 7780

87

91.5 8191.5

90.5

919088

85 809090.5 8790919086

828692

91.59190 90.590.5 8591 86909090 -75 -50 -25 0 25 50 75

-50

-25

0

25

508987 8987 89

869088

84

8884

91.5

90.5

90.58983

7986909191.5

9189817981

91.591.59089858583 778690.59190.5918987

80808888

8179 77808185868687797878

7684828583777677778484838179 80798281868786868483 757778798490.5898885 788186

878890.59291.592.5908785

79788892.5

91

87878583 8088899486

92.586 89

92.59392.5 8588

90.5

90.589

91.5 8591.591.59189

889089 9088 (d) L-configuration，δ=45° (e) U-configuration，δ=45°

-75 -50 -25 0 25 50 75-50

-25

0

25

50

8382

88

90.586

8885 8283848491.588

869190.5

8491.5

8291.591.5

90.587 767990 7678829090.59190.58683

7979878689878983848379 76

7884828378 7676797780 76808384828082767678

768284858788878580797779828690888179

7576788590.583 77819388868383 8185

84

90.58585 808185 798386

90

91.59393

91.5

8887 848791.5

8187

87 87

90 83

8989 -75 -50 -25 0 25 50 75-50

-25

0

25

50

88 8589

90.590.5 9190.5

8280

91 82

90.5 83

858890 918983 8090.58887848279 7677808389878478 7678778685828283848279 767580818181

8281777678838482 76798182828680

7981808389868383 7883879091929082 76778193.59392 7883868990

8780

90.59491.58583

838394

91.5

86 839192.592.590

90.5 909192.5

9393 879191

.592

92.586 859292

90.590.589 8489

89 (f) L-configuration，δ=60° (g) U-configuration，δ=60°



115

-75 -50 -25 0 25 50 75-50

-25

0

25

50

90.587

8183

91.59293

93.593

87 91939392.5

91

94.5

92.5

94.587

93

838590.591.5

90788290

92.5877992

.59393.592.590.58884 77

849293.592.59292898583 859390.589

7684899191.588858583838587

75798182848581828183828280 7675767777757782 79848080848382 7778828878 757683859191.5928881797979 798083868784 8892

91.59492.589 798285

90.593

93.592888483

868793.5

87 79

95.5

95.59083 7786

88

90.5

9393

8784 9090.5

94.593

83

94

95 81849191.5

90.589

92.594.592 838490

.5

9486

92.5

91.5 -75 -50 -25 0 25 50 75

-50

-25

0

25

50

90.587

8183

91.59293

93.593

87 91939392.5

91

94.5

92.5

94.587

93

838590.591.5

90788290

92.5877992

.59393.592.590.58884 77

849293.592.59292898583 859390.589

7684899191.588858583838587

75798182848581828183828280 7675767777757782 79848080848382 7778828878 757683859191.5928881797979 798083868784 8892

91.59492.589 798285

90.593

93.592888483

868793.5

87 79

95.5

95.59083 7786

88

90.5

9393

8784 9090.5

94.593

83

94

95 81849191.5

90.589

92.594.592 838490

.5

9486

92.5

91.5

(h) L-configuration，δ=90° (i) U-configuration，δ=90°

Fig.13 Loading noise characteristics of the scissor tail-rotor

20 30 40 50 60 70 80 90 10091

92

93

94

95

96

97

SPL(dB)

Scissors angle(°)

L Configuration U Configuration

Fig.14 Maximum SPL value of loading noise versus the scissor angle

Total noise

Fig.15 presents the contours of the total noise SPL on the observation plane in forward flight. Fig. 16 shows the effects of the scissor angle on the maximum SPL of the total noise. The constant straight-line in the plot represents the conventional tail-

rotor. As shown, there exists an optimal scissor angle for both the L-configuration and the U-configuration in the range of 30°-60° which can make a lowest tail-rotor noise level. For the current tail rotor configuration, δ=45° is the optimal scissors angle.

-75 -50 -25 0 25 50 75-50

-25

0

25

50

82

84858990.5

9089 889090.5 9292.592.592.59090.590.5

88

84

92.59091

8689

93.589 7885

9292.59191

8890.5

9390

83848586878990.578828889

89909293

92

787991 7677818293.5

93798090

90

92

92.594.5 76767683859189 77789293.5

95.5

94.5

93.5 7878788391.5 8384

95

94.592.5

90.5

87

93.5

94.590 80838593

.593918983

90.5

92.5939187 839392

90

9090.5

91.591

8584 8984 (a) Conventional tail-rotor

-75 -50 -25 0 25 50 75-50

-25

0

25

5092 92

86 8593

838586

87

91.591 82

9191.592

91.5

91.5

80909192.59391

.58990.59192

92 8489 788492.592.5

788285

84 7880909092.5 8385

8792.5

92.590 90.5

92.5

90 81839091.5

87 8083838687909591 83838487

949388

90.590 83 82

85

90.5

92.589

84

92.5

92.592

91

88

90.5

929189

91 8593

91

9291.5 -75 -50 -25 0 25 50 75-50

-25

0

25

50

92.5 9190 8485

92.5

90

8991

91

8390.5

93.592

91.5

91.5929392.5

8291 8591.59191859191.590.591

7984899279848585899389

88

829091.591.593 7790.5

90.589 859292.5

9188 808292.593 779194

93

90

87 88

93.5

90 8386

9395.5 79

90.590.5

95.5 7984859191.5

93.594

94.5 81

95.5

94.5

93.5

90.5 8793

92.591 8990 828894.5

91 9094.5 839094.594

9294.59189

91.59490

(b) L-Configuration，δ=30° (c) U-Configuration，δ=30°



116

-75 -50 -25 0 25 50 75-50

-25

0

25

50

90

88

90.5 83

828990.58283859081

8487867980

85 788185

8991 78798083899191 77808588

8892

89

80 757790.593

75869092929389 76798489 85888993 788494

828991

91.59383

889188 8591929292.5 8592.590.5 84

91

92

90.5 8690.5

91.5

89

90.5

88

88 -75 -50 -25 0 25 50 75-50

-25

0

25

50888788 878890.58889

90.5

90.58989

92

879092

9291.5

9090838490

86909083

868791.590.5

82828690919189 788291

7579

81839179

8086

85

90 7990.590.592

92.5 777988919192

79799092.5

93.5

9288 818382868793

93.5 828485 81

91.591

8793.592 84919488 8487

91.593

92.592

91 87939387 85908886 899090

9191.5929090 90 908887 85898887 (d) L-Configuration，δ=45° (e) U-Configuration，δ=45°

-75 -50 -25 0 25 50 75-50

-25

0

25

50848990

86

90.591.591 86

909090.5 91

798085

91

858690.5917681

91

91.591

818591.5

818491 74788290.59292.5 848585

8993 91.592

92.591 767987

90.5

91.5 787883

92.5

9188 7981

94.5

93

818283

92.594.59390

89878081

94

94.5

939086 838795

93.5

89 808194.5 818593.592.5

90

90 90

90.5 91.5 83

85

8889909090 -75 -50 -25 0 25 50 75-50

-25

0

25

5089 898989 9090.5899090 8690

83

9090 91.590.5

90 91

84

90.591 81839191

.590.59084858891

90.5 90.59190.591.5 80

838890 78899091.5927883848992.5 818389919193.5

90.5 75909093

9189 91.5929590.587 7980848592

95.5

94

92.5 8081828393.5

92.5

89 9395.5

94

92.5898686 82859595

8885 82

91.5

93 828594 85929493 93.592

90.58585

8892.5

898684 92.590.5 8691 899189 87

(f) L-configuration，δ=60° (g) U-configuration，δ=60°

-75 -50 -25 0 25 50 75-50

-25

0

25

50

92 92 90.5

929390.590.5

8081

889092

80

909291 94.5

9493.5

8490.5 94929090.5

8892.590

777891.5

949393.590.59091.587

7791

92.593.59091.5 86868789 8190.591.591.589869090 91909190 84909192.5

92.5 77899493

7983899490 767677

8691.59189 7883868788 8081858788

9288 83

91.5

94.5

89 7980

88

92.5

95.594.591

95.5

96

9388 8085878896.5 83

95

969290 858790

95.5

93.587 8588929587

8894.59084 94.5

939289 84889094 8592.593.5 -75 -50 -25 0 25 50 75

-50

-25

0

25

50

92 92 90.5

929390.590.5

8081

889092

80

909291 94.5

9493.5

8490.5 94929090.5

8892.590

777891.5

949393.590.59091.587

7791

92.593.59091.5 86868789 8190.591.591.589869090 91909190 84909192.5

92.5 77899493

7983899490 767677

8691.59189 7883868788 8081858788

9288 83

91.5

94.5

89 7980

88

92.5

95.594.591

95.5

96

9388 8085878896.5 83

95

969290 858790

95.5

93.587 8588929587

8894.59084 94.5

939289 84889094 8592.593.5

(h) L-configuration，δ=90° (i) U-configuration，δ=90°

Fig.15 Total noise characteristics of the scissored tail-rotor

20 30 40 50 60 70 80 90 10093

94

95

96

97

98

SPL(

dB)

Scissors angle(°)

L Configuration U Configuration

Fig.16 Maximum SPL value of the total noise versus the scissors angle

CONCLUSION

In this paper, an unsteady N-S solver and an acoustic solver based on the FW-H equation were coupled to study the aerodynamic and aeroacoustic characteristics of a scissored tail-rotor. Based on the simulation results, the following conclusions can be obtained.

A smaller blade azimuthal spacing of the scissored tail-rotor makes the wake characteristics of a four-bladed tail-rotor similar to that of a two-bladed tail-rotor.

The scissored tail-rotor suffers a more severe aerodynamic interference when compared with the conventional one



117

due to the reduced blade azimuthal spacing.

The aeroacoustic characteristics of the scissored tail-rotor are significantly different from that of the conventional one. The scissor angle has an important effect on the tail-rotor aeroacoustics. There exists an optimal scissors angle that can achieve a lowest tail-rotor noise level.

The L-configuration is superior to the U-configuration in both aerodynamic and aeroacoustic characteristics.

REFERENCES

[1] Sonneborn W. G. O., Drees J. M. “The

Scissor Rotor”. Presented at the American Helicopter Society 30th Annual Forum, Washington D. C., 1974: 18-27.

[2] Rozhdestensky M. G. “Scissors Rotor Concept: New Results Obtained”. Presented at the American Helicopter Society 52nd Annual Forum, Washington D. C., 1996: 1231-1241.

[3] Xu G. H., Zhao Q. J., Peng Y. H. “Study on the Induced Velocity and Noise Characteristics of a Scissors Rotor”. Journal of Aircraft, 2007, 44(3): 806-811.

[4] Xu G. H., Wang S. C., Zhao J. G. “Experimental and Analytical Investigation

on Aerodynamic Characteristics of Helicopter Scissors tail rotor”. Chinese Journal of Aeronautics, 2001, 14(4): 193-199.

[5] Roe P. L. “Approximate Riemann Solvers, Parameter Vectors, and Difference Schemes”. Journal of Computational Physics, 1981, 43(2):357-372.

[6] Luo H., Baum J. D. “A Fast, Matrix-Free Implicit Method for Computing Low Mach Number Flows on Unstructured Grids”. AIAA 99-3315, 1999.

[7] Spalart P. R., Allmaras S. R. “An One-Equation Turbulence Model for Aerodynamic Flows”. AIAA 92-0439, 1992.

[8] Brentner K. S. “Prediction of Helicopter Rotor Discrete Frequency Noise”. NASA Technical Memorandum 87721, 1986.

[9] David B. S., Gloria K. Y., Charles A. S., Martin J. H. “Performance and Loads Data from an Outdoor Hover Test of a Lynx Tail Rotor”. NASA TM-101057, 1989.

[10] Schmitz F. H., Boxwell D. A., Splettstoesser W. R., and Schultz K. J. “Model-Rotor High-Speed Impulsive Noise: Full-Scale Comparisons and Parametric Variations”. Vertica, 1987, 8(4): 395-422.

[11] Newman S., “The Foundations of Helicopter Flight”. Edward Arnold, 1994.

[12] Brentner K S, Edwards B D, Riley R, et al. “Predicted Noise for a Main Rotor with Modulated Blade Spacing”. Journal of the American Helicopter Society, 2005, 50(1):18-25.

Documents

Analysis on Aerodynamic and Acoustic Characteristics for