当前位置:首页 >> 兵器/核科学 >>

Experimental aeroacoustics of a two-strut, two-wheel_图文

AIAA 2014-0020 AIAA SciTech 13-17 January 2014, National Harbor, Maryland 52nd Aerospace Sciences Meeting

Experimental aeroacoustics of a two-strut, two-wheel landing gear in a propeller wake
Ra?k Chekiri? and Philippe Lavoie?
University of Toronto, 4925 Du?erin St., North York, Ontario, M3H 5T6, Canada
Downloaded by BEIHANG UNIVERSITY (CNPIEC Xi'an Branch) on March 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/6.2014-0020

Werner Richarz?
Aercoustics Engineering Ltd., 50 Ronson Drive, Suite 165, Toronto, Ontario, M9W 1B3, Canada
The e?ects of an upstream propeller on the radiated noise of a two-wheel landing gear con?guration were studied. Tests were conducted at a scale of 1:10.8 with a propeller advance ratio of 1, at Mach numbers less than 0.1 in the University of Toronto Acoustic Wind Tunnel Facility. Microphone measurements show that the propeller dominates the far-?eld noise signature for all tested con?gurations. A broadband noise increase of 10dBSPL was observed between 1 and 5 times the blade passing frequency (BPF) for M ≥ 0.045 when the drag strut is added. Spectra of the ?uctuating surface pressure measurements displayed peaks at the BPF and were strongest for the con?guration omitting the drag strut. The pressure ?uctuations at the BPF on the wheel microphones were at least 30dB less than on the strut. Cross-spectra between one far-?eld microphone against all surface microphones are also presented. These preliminary measurements present a unique dataset for the validation of models used in the design of regional turboprop aircraft.

Nomenclature
BPF Dp J M n rpm SPL V blade passing frequency propeller diameter advance ratio Mach number rotational speed (revolutions per second) rotations per minute Sound Pressure Level ?ow velocity

I.

Introduction

n recent years, landing gear have been recognized as a signi?cant contributor to airframe noise. During I the approach phase of landing, when engines operate at reduced power, the airframe noise is comparable to the engine noise. Since there is no single dominant source of airframe noise, appreciably decreasing the
1

total level requires the reduction of noise from all contributing airframe components. Empirical approaches to the problems of predicting and modeling landing gear noise remain dominant in industry. Past experimental studies, such as those by Guo2 and Dobrzynski et al.,3 have demonstrated that a landing gear noise signature depends on local ?ow conditions that can be in?uenced by the landing gear’s location on an aircraft (wing vs. fuselage) or upstream airframe airframe components. Recent studies have investigated generic landing gear models such as the four-wheel rudimentary landing gear model introduced by Spalart et al.4, 5 and the two-wheel LAGOON model introduced by Airbus.6 This
? MASc

Student, Institute for Aerospace Studies, University of Toronto, Student Member AIAA; chekiri@utias.utoronto.ca Professor, Institute for Aerospace Studies, University of Toronto, Senior Member AIAA; lavoie@utias.utoronto.ca ? Senior Scientist, Aercoustics Engineering Ltd.; wernerr@aercoustics.com
? Associate

1 of 8 American Institute of Aeronautics and Astronautics
Copyright ? 2014 by Rafik Chekiri, Philippe Lavoie. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.

Downloaded by BEIHANG UNIVERSITY (CNPIEC Xi'an Branch) on March 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/6.2014-0020

Figure 1. Photograph of the model with propeller, main strut, wheels, and drag strut (Con?guration A).

current study aims to identify the aeroacoustic signature speci?c to a two-strut, two-wheel landing gear when the components are in the unsteady wake of a propeller. The motivation of this work stems from community noise concerns related to regional turboprop aircraft such as the Bombardier Dash-8 Q400.

II.
A. Experimental Facility

Experimental Setup

Tests were conducted at the University of Toronto Institute for Aerospace Studies in the Acoustic Wind Tunnel Facility. This facility features an open-circuit, open-jet, suction-type wind tunnel. The nozzle diameter is 0.70 m and nozzle-to-collector distance is 2.40 m. The wind tunnel can be operated at ?ow speeds between 10 m/s and 60 m/s. The open-jet section is located within an anechoic chamber to allow for far-?eld noise measurements external to the jet ?ow. The inner walls of the chamber are lined with ?berglass wedges to eliminate acoustic re?ections above 135 Hz. B. Model Description

A photograph of the tested model geometry is shown in Figure 1. The model geometry is based on a Bombardier Dash-8 Q400, a regional turboprop aircraft, at a scale of 1:10.8. This scaling ensures that an upstream model propeller and the landing gear are entirely in the jet potential core of the wind tunnel. The model consists of an electric motor driving a Master Airscrew two-bladed 15 × 8 propeller, a tubular nacelle, a main strut, a drag strut, an axle, and two wheels. The landing gear components are constructed of aluminum and ABS plastic. Although small-scale features such as hoses, cables, and cutouts are not replicated, it is possible to predict the additional noise of such small features by introducing a ‘complexity factor’ in future prediction models.7 A schematic drawing of the landing gear geometry is shown in Figure 2. An E-?ite Power 46 Brushless Outrunner Motor was used to power the propeller along with a 60-Amp Pro Switch-Mode BEC Brushless Electronic Speed Controller from E-?ite. Throttle was controlled through USB communication with an Arduino microcontroller. An optical tachometer was constructed to count propeller revolutions and was placed within the nacelle, adjacent to the motor. An Eagle Tree Systems eLogger V4 was also used to monitor power consumption of the system, throttle levels, temperature, and rpm (through a brushless motor sensor).

2 of 8 American Institute of Aeronautics and Astronautics

MAIN STRUT SURFACE MICROPHONE INSERT
S4 S3 S2 S1 R1 45
Downloaded by BEIHANG UNIVERSITY (CNPIEC Xi'an Branch) on March 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/6.2014-0020
o

MAIN STRUT

25

o

DRAG STRUT

FLOW WHEEL
L1 L2

R2

WHEEL TREAD SURFACE MICROPHONE INSERTS

Figure 2. Schematic drawing of the landing gear model geometry.

Various con?gurations of the model geometry were tested to deduce how each component a?ects the far-?eld acoustic signature. The advance ratio, J , de?ned as J= V nDp (1)

was set as closely to one as possible, such that results at di?erent ?ow speeds could be compared. The motor could be controlled to advance ratios within 1 ± 0.2, requiring rotational rates of between 1700 and 5000 rpm depending on the free stream velocity. For tests with no freestream ?ow, the motor was set to 1700 rpm (the same rate as the lowest speed test case) . Table 1 lists each con?guration tested.
Table 1. Model Con?gurations used in wind tunnel tests.

Con?guration Propeller Main Strut & Wheels Drag Strut

A

B

C

The model includes a support that extends partially into the airstream and is bolted to the test stand. The landing gear is cantilevered from the nacelle and is normal to the tunnel ?ow direction as shown in Figure 1. C. Instrumentation and Data Acquisition

Six Bruel & Kjaer 4134 1/2” measurement microphones are placed in a linear vertical array 2.0 m from the tunnel centreline. The linear array is traversed along the length of the test section. Microphone locations at the same vertical height as the main strut are marked M1 through M6 on Figure 3. Results presented here use data collected from far-?eld microphone position M2. Knowles Electronics EK-26899-P03 microphones have been epoxied into insets in the wheel tread and main strut to provide unsteady surface pressure measurements. Microphone wires have been routed through the model and out of the airstream, so as not to in?uence the ?ow-?eld and acoustics. Each wheel has two surface microphones: one directly facing oncoming ?ow and another 45? below it on the wheel tread surface. The main strut has four surface microphones along its length in the region directly downstream of the drag strut. These microphones and their locations are identi?ed in Figure 2. Surface microphone signals were passed through a high sound pressure level circuit to increase the dynamic range of measurements and ampli?ers with nominal gains of 40 dB. All signals were passed through antialiasing ?lters of 10 kHz and were ac-coupled to the data acquisition unit to remove any dc bias. Surface
3 of 8 American Institute of Aeronautics and Astronautics

TOP VIEW
10 cm 40 cm

TEST SECTION INLET

240 cm

JET COLLECTOR

200 cm

Downloaded by BEIHANG UNIVERSITY (CNPIEC Xi'an Branch) on March 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/6.2014-0020

15 cm M1 M2 M3 M4 M5 M6 FAR-FIELD MICROPHONE LOCATIONS

SIDE VIEW

TEST SECTION INLET

JET COLLECTOR

FAR-FIELD MICROPHONE TRAVERSE

Figure 3. Schematic drawings of the experimental setup in the test section. Not to scale.

microphones were calibrated in situ against a Bruel & Kjaer 4134 1/2” measurement microphone setting the microphone diaphragms parallel to each other and performing a frequency sweep with an aligned speaker. All microphone data were sampled for 120 s at 32 kHz with a 16-bit National Instruments PCIe-6361 A/D data acquisition card.

III.
A. Far-?eld Measurements

Results & Analysis

Primary components were incrementally added to the landing gear assembly to obtain the far-?eld sound pressure level (SPL) spectrum for the complete con?guration (A). Each listed in Table 1 was tested at wind tunnel ?ow speeds of 0 m/s, 10 m/s, 15 m/s, and 20 m/s. These ?ow velocities correspond to Reynolds numbers based on the main strut diameter of 0, 10 000, 15 000, and 20 000, or Mach numbers of 0.00, 0.03, 0.045, and 0.06, respectively. The results comparing the narrow-band SPL of Con?gurations A, B, and C, using a far-?eld microphone at position M2 are presented in Figure 4. In all cases, the addition of the main strut and wheels to the propeller and nacelle show no appreciable change in far-?eld levels. This is demonstrated by the close resemblance of spectra for the propeller only

4 of 8 American Institute of Aeronautics and Astronautics

Far?field Narrow Band SPL Spectra ? 0 m/s 120 110 100 90 dB SPL (re 20?Pa) 80 70 60 50 40 30 20 dB SPL (re 20?Pa) A B C 120 110 100 90 80 70 60 50 40 30 20 10 frequency / BPF
0

Far?field Narrow Band SPL Spectra ? 10 m/s A B C

Downloaded by BEIHANG UNIVERSITY (CNPIEC Xi'an Branch) on March 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/6.2014-0020

10 ?1 10

10

1

10 ?1 10

10 frequency / BPF

0

10

1

(a) No ?ow test, M = 0.00.
Far?field Narrow Band SPL Spectra ? 15 m/s 120 110 100 90 dB SPL (re 20?Pa) 80 70 60 50 40 30 20 10 ?1 10 10 frequency / BPF
0

(b) 10 m/s, M = 0.03.
Far?field Narrow Band SPL Spectra ? 20 m/s 120 A B C 110 100 90 dB SPL (re 20?Pa) 80 70 60 50 40 30 20 10
1

A B C

10 ?1 10

10 frequency / BPF

0

10

1

(c) 15 m/s, M = 0.045.

(d) 20 m/s, M = 0.06.

Figure 4. Far-?eld narrowband SPL spectra for all tested con?gurations and Mach numbers, J = 1 ± 0.2. Frequencies are nondimensionalized by the blade passing frequency.

con?guration (C) and the con?guration omitting only the drag strut (B). All spectra have tonal peaks at the blade passing frequency (BPF). When no in?ow is present, the spectra from the three con?gurations resemble each other closely, as is the case when the ?ow speed is increased to 10 m/s. A strong tone is evident at the BPF however its harmonics are not prevalent in the spectra. As the ?ow speed is increased further to 15 m/s and 20 ms, the subharmonic and harmonic frequencies of the BPF become prominent. The tone at 0.5BPF can be attributed to the rotational rate of the motor since the propeller is two-bladed. To match advance ratios with lower speed tests, the rpm of the motor must be increased, leading to some structural vibrations of the model support. Alternatively, noise from the motor itself may also be contributing to the 0.5BPF tone at high rotational rates. In the higher speed test cases, a broadband increase in noise levels of approximately 10 dB between 1 and 5 times the BPF can be seen between Con?gurations A and B. This broadband increase in levels is a direct result of including the drag strut in the model setup. B. Surface Microphone Measurements

The spectra of the surface microphones in Figure 5 show tonal peaks at the BPF, as was observed with the far-?eld microphone. The strut microphone signals have a broader bandwidth in Con?guration A than in Con?guration B. Peaks of the BPF, its harmonics, and its subharmonics are apparent and superimposed

5 of 8 American Institute of Aeronautics and Astronautics

Surface Pressure Spectra ? Configuration A: Strut? 10 m/s 120 S1 S2 S3 S4 Spectral Level, dB (re 20?Pa) 120

Surface Pressure Spectra ? Configuration B: Strut? 10 m/s S1 S2 S3 S4

100 Spectral Level, dB (re 20?Pa)

100

80

80

60

60

40

40

20

20

0

0

Downloaded by BEIHANG UNIVERSITY (CNPIEC Xi'an Branch) on March 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/6.2014-0020

?20 ?1 10

10 frequency / BPF

0

10

1

?20 ?1 10

10 frequency / BPF

0

10

1

(a) Strut Microphones, Con?guration A.
Surface Pressure Spectra ? Configuration A: Wheels ? 10 m/s 60 L1 L2 R1 R2 Spectral Level, dB (re 20?Pa) 60

(b) Strut Microphones, Con?guration B.
Surface Pressure Spectra ? Configuration B: Wheels ? 10 m/s L1 L2 R1 R2

40 Spectral Level, dB (re 20?Pa)

40

20

20

0

0

?20

?20

?40

?40

?60 ?1 10

10 frequency / BPF

0

10

1

?60 ?1 10

10 frequency / BPF

0

10

1

(c) Wheel Microphones, Con?guration A.

(d) Wheel Microphones, Con?guration B.

Figure 5. Surface narrowband SPL spectra for Con?gurations A and B, M = 0.03 and J = 1 ± 0.2.. Frequencies are nondimensionalized by the blade passing frequency.

on a large broad energy peak. In Con?guration B, many harmonics of the BPF are clearly distinguishible above the broad peak and in the case of microphone S1, dominate the acoustic signature. The inclusion of the drag strut attenuates the harmonic peaks evident in Con?guration B though the broadband levels increase. Figure 5 (b) shows that propeller tip vortex-strut interaction generates the largest surface pressure ?uctuations. Apart from peaks at 0.5BPF and the BPF, the levels of the wheel surface pressure ?uctuations are signi?cantly lower compared to the ?uctuations on the main strut surface. This may be an indication that tip vortices shed by the propeller do not impinge on the upstream facing portion of the wheels as illustrated in Figure 6. C. Cross-spectral Analysis

Figures 7 and 8 show the cross-spectra of each surface microphone with the Bruel & Kjaer 4134 1/2” measurement microphone at position M2 for a ?ow velocity of 10 m/s and J = 1. Examining the cross-spectra of the strut microphones with the far-?eld microphone, it is observed that the strongest cross-spectral magnitude at the BPF is given by microphone S1. For Con?guration A, microphone S2 has the lowest peak level, while the peaks for S3 and S4 are nearly equal. This is in contrast with Con?guration B where the strength of the cross-spectra generally decreases moving toward the nacelle on the strut. This may be due to the presence of the drag strut reducing the strength of impinging vortices on S2. In the case of Con?guration B, cross-

6 of 8 American Institute of Aeronautics and Astronautics

Downloaded by BEIHANG UNIVERSITY (CNPIEC Xi'an Branch) on March 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/6.2014-0020

SLIPSTREAM
Figure 6. Illustration of the propeller slipstream with respect to the model. Not to scale.

spectral magnitudes are 2 to 3 times larger than in Con?guration A. This can be explained by the addition of the drag strut creating an obstruction between the propeller tip vortices and the main strut surface. The wheels are not a signi?cant source of far-?eld noise as is evident from the two order of magnitude lower cross-spectral levels at the BPF compared to the main strut (see Figure 7 and 8).

IV.

Conclusions

Noise measurements taken in the far-?eld of a two-strut, two-wheeled landing gear with an actuated upstream propeller have been compared and discussed for various landing gear con?gurations at Mach numbers of 0.00, 0.03, 0.045, and 0.06 and an advance ratio of 1. The propeller and model support were shown to dominate the far-?eld acoustic signature. A broadband increase in levels in the region of 1 to 5 times the BPF was observed upon inclusion of the drag strut. Subharmonic and harmonic frequencies of the BPF were dominant in the far-?eld SPL spectra only for wind tunnel ?ow speeds greater than or equal to 15 m/s (M = 0.045). Unsteady surface pressure measurements on the landing gear main strut and wheels were recorded and used to compute surface and far-?eld cross-spectra. The pressure ?uctuations on the main strut were observed to be the strongest sources of far-?eld noise at the BPF. When the drag strut was removed, these pressure ?uctuations were shown to have greater cross-spectral magnitudes. At the BPF peak, the phase of the cross-spectra are shown to cross zero and swing 180? . The cross-spectral magnitudes provide support to the claim that the ?ow structures of the propeller slipstream do not impinge on the wheels. The recorded sound pressure level spectra can be used as an indication of what to expect during higher velocity testing thereby enabling the development of more accurate noise prediction tools. The results of this study also provide a new benchmark case for computational aeroacoustic codes that are used in design and analysis of aircraft landing gear.

Acknowledgments
The authors would like to thank the Green Aviation Research and Development Network (GARDN), the Natural Sciences and Engineering Research Council of Canada (NSERC), and Bombardier Aerospace for their ?nancial support. The authors would like to thank Stephen Collavincenzo and Raymond Wong from Bombardier Aerospace for their technical support and guidance. The authors would also like to thank Nicole Houser for her assistance in preparing ?gures.

7 of 8 American Institute of Aeronautics and Astronautics

10 CPSD Magnitude

4

Cross?spectra ? Configuration A: Strut ? 10 m/s CPSD Magnitude S1 S2 S3 S4

10

4

Cross?spectra ? Configuration B: Strut ? 10 m/s S1 S2 S3 S4

10

2

10

2

10 0.5 4 2 0 ?2 ?4 0.5

0

1 frequency / BPF

1.5

2

10 0.5 4 2 0 ?2 ?4 0.5

0

1 frequency / BPF

1.5

2

CPSD Phase (radians)

Downloaded by BEIHANG UNIVERSITY (CNPIEC Xi'an Branch) on March 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/6.2014-0020

1 frequency / BPF

1.5

2

CPSD Phase (radians)

1 frequency / BPF

1.5

2

(a) Strut Microphones, Con?guration A.

(b) Strut Microphones, Con?guration B.

Figure 7. Cross-spectra of strut surface microphones and a far-?eld microphone at position M2, M = 0.03 and J = 1 ± 0.2. Frequencies are nondimensionalized by the blade passing frequency.
10 CPSD Magnitude
2

Cross?spectra ? Configuration A: Wheels ? 10 m/s CPSD Magnitude L1 L2 R1 R2

10

2

Cross?spectra ? Configuration B: Wheels ? 10 m/s L1 L2 R1 R2

10

0

10

0

10

?2

0.5 4 2 0

1 frequency / BPF

1.5

2

10

?2

0.5 4 2 0

1 frequency / BPF

1.5

2

CPSD Phase (radians)

?2 ?4 0.5

CPSD Phase (radians) 1 frequency / BPF 1.5 2

?2 ?4 0.5

1 frequency / BPF

1.5

2

(a) Wheel Microphones, Con?guration A.

(b) Wheel Microphones, Con?guration B.

Figure 8. Cross-spectra of wheel surface microphones and far-?eld microphone at position M2, M = 0.03 and J = 1 ± 0.2. Frequencies are nondimensionalized by the blade passing frequency.

References
1 Lopes, L., Brentner, K. S., Morris, P., Lilley, G., and Lockard, D., “Complex landing gear noise prediction using a simple toolkit,” AIAA Paper 2005-1202 , Jan. 2005. 2 Guo, Y., “E?ects of Local Flow Variations on Landing Gear Noise Prediction and Analysis,” Journal of Aircraft , Vol. 47, No. 2, March 2010, pp. 383–391. 3 Dobrzynski, W. M., Pott-Pollenske, M., Foot, D., and Goodwin, M., “Landing Gears Aerodynamic Interaction Noise,” ECCOMAS Conference , Multi-Science, Jyv¨ askyl¨ a, 2004, pp. 115–135. 4 Spalart, P. R., Shur, M. L., Strelets, M. K., and Travin, A. K., “Towards Noise Prediction for Rudimentary Landing Gear,” Procedia IUTAM , Vol. 1, 2010, pp. 283–292. 5 Reger, R., Liu, F., and Cattafesta, L., “An Experimental Study of the Rudimentary Landing Gear,” AIAA Paper 20130464 , Jan. 2013. 6 Manoha, E., Bult? e, J., Ciobaca, V., and Caruelle, B., “LAGOON: further analysis of aerodynamic experiments and early aeroacoustics results,” AIAA Paper 2009-3277 , May 2009. 7 Guo, Y., “A statistical model for landing gear noise prediction,” Journal of Sound and Vibration , Vol. 282, No. 1-2, April 2005, pp. 61–87.

8 of 8 American Institute of Aeronautics and Astronautics


相关文章:
ZZ_5_Actran_HVAC_Aeroacoustics_图文.ppt
ZZ_5_Actran_HVAC_Aeroacoustics_机械/仪表_工程科技_专业资料。汽车空调通风口...experimental measurements A better agreement could be reached : A longer CFD...