Using Phased Array Beamforming to Identify Broadband Noise

Using Phased Array Beamforming to Identify Broadband Noise Sources in a Turbofan Engine Pieter Sijtsma National Aerospace Laboratory NLR Department of Helicopters and Aeroacoustics

Nationaal Lucht- en Ruimtevaartlaboratorium – National Aerospace Laboratory NLR

Introduction

PROBAND rig testing at AneCom anechoic hall

AneCom-AeroTest facility (near Berlin)

RR fan rig

12th CEAS Workshop, 23-24 October 2008, Bilbao

2

Introduction

NLR contribution to rig tests picture by Assystem UK

drooped intake

bypass ring

intake ring

liner 12th CEAS Workshop, 23-24 October 2008, Bilbao

rotor

stator 3

array response of single azimuthal mode 0 -5 -10 -15 dB

Introduction

Non-uniform arrays for azimuthal mode detection

-20 -25 -30 -35 -40 -200

-160

-120

-80

-40

0

40

80

120

160

200

mode

12th CEAS Workshop, 23-24 October 2008, Bilbao

4

Introduction

Merits of azimuthal mode detection z Valuable for tonal noise z

z

Gives information about noise generating mechanisms: – Rotor alone noise – Rotor/stator interaction noise – Interaction by steady distortion Gives information about liner performance

z Less valuable for broadband noise z z z

Gives information about liner performance Information about noise sources is limited No detailed information about source location, e.g.: – rotor or stator – leading edge or trailing edge – spanwise or concentrated at the blade tips

12th CEAS Workshop, 23-24 October 2008, Bilbao

5

Introduction

Possible added value by phased array beamforming

wind tunnel measurement

fly-over measurement

field measurement

engine outlet ‘slat horn’

landing gear

12th CEAS Workshop, 23-24 October 2008, Bilbao

nose wheel

6

Introduction

Beamforming: stationary and rotation focus (1)

Stationary focus

Rotating focus

from [Sijtsma-2001] 12th CEAS Workshop, 23-24 October 2008, Bilbao

7

Introduction

Beamforming: stationary and rotation focus (2) beamform results with intake array [sijtsma-2006]

stationary focus (stator LE) 12th CEAS Workshop, 23-24 October 2008, Bilbao

rotating focus (rotor LE) 8

Contents

Contents z Introduction (7) z Tonal and broadband noise (2) z Details of beamforming technique (5) z Axial resolution (2) z Results with intake array (17) z Results with bypass array (7) z Possible improvement (2) z Conclusions (1) z References

12th CEAS Workshop, 23-24 October 2008, Bilbao

9

Tonal and broadband

Tonal and broadband noise (1) z Tonal = rotor-bound (obtained with 1/rev pulse) z “Broadband” = what remains total

10 dB

broadband tonal

typical average spectrum of bypass array at 55% shaft speed

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

Fre quency [Hz] 12th CEAS Workshop, 23-24 October 2008, Bilbao

10

Tonal and broadband

Tonal and broadband noise (2) z Broadband is significant at all shaft speeds total

10 dB

broadband tonal

50

60

12th CEAS Workshop, 23-24 October 2008, Bilbao

70

80 Shaft spe ed [%]

90

100

110

11

Beamforming

Details of beamforming technique source plane

G

acoustic source

scan points ξ j

Cross-Spectral Matrix: C = pp∗ microphone array 12th CEAS Workshop, 23-24 October 2008, Bilbao

12

Beamforming

Beamforming summary Conventional Beamforming:

G

Estimate source auto-power Aj in scan point ξ j by minimising C − Aj g j g

∗ 2 j

, with g j = steering vector

G

(microphone pressures due to theoretical point source in ξ j ) Solution: Aj =

g∗j Cg j

(g g ) ∗ j

2

.

j

12th CEAS Workshop, 23-24 October 2008, Bilbao

13

Beamforming

Point source description

⎛ g1, j ⎞ G G ⎜ ⎟ Steering vector: g j = ⎜ # ⎟ , g n , j = F ( xn , ξ j ) ⎜g ⎟ ⎝ N, j ⎠ G xn are microphone locations, F is the transfer function F can be a Green's function G (monopole source assumption) G G (solution of partial differential equation LG = −δ x − ξ

(

)

+ boundary conditions) F can also be a derivative of G (dipole source assumption) 12th CEAS Workshop, 23-24 October 2008, Bilbao

14

Beamforming

Options for Green’s function z Option 1: Free-field Green’s function in uniform flow z Option 2: Green’s function in hard-walled duct and uniform flow

[Lowis-2007]

z Option 3: Green’s function [Rienstra-2005] in soft-walled duct

and uniform flow

z Option 4: Green’s function values calculated with CAA z Option 5: Measured Green’s function values

z Unknown source directivity is a critical issue for all options!

12th CEAS Workshop, 23-24 October 2008, Bilbao

15

Beamforming

Option chosen here z Free-field Green’s function is chosen z To minimize effects of reflections, a liner is required z

simulations shown in [Sijtsma-2007]

z Non-circular geometry (drooped intake) is not a problem

z Beamforming with rotating focus (ROSI) is analogous to

Conventional Beamforming(CB) [Sijtsma-2001]

12th CEAS Workshop, 23-24 October 2008, Bilbao

16

Axial resolution

Lack of axial resolution (1) z Simulation study (lined duct)

0.4 m

point source 0.4 m

cylindrical duct

array plane

12th CEAS Workshop, 23-24 October 2008, Bilbao

M=0.29

scan plane (rotor LE) 17

Axial resolution

Lack of axial resolution (2) z Simulation study

Conventional Beamforming x,y-scan

12th CEAS Workshop, 23-24 October 2008, Bilbao

18

Intake array

Intake array (1)

CB results with intake array scan grid on stator LE 55% shaft speed

z No source details visible

12th CEAS Workshop, 23-24 October 2008, Bilbao

19

Intake array

Intake array (2)

CB results with intake array scan grid on stator LE 75% shaft speed

z Details visible at higher

frequencies

z Best results with broadband

noise

12th CEAS Workshop, 23-24 October 2008, Bilbao

20

Intake array

Intake array (3)

CB results with intake array

(using 2 principal components of the CSM)

scan grid on stator LE 75% shaft speed

z Coherent structure visible z Sound sources due to vibrations

rather than to aero-acoustic mechanisms?

12th CEAS Workshop, 23-24 October 2008, Bilbao

21

Intake array

Intake array (4)

ROSI results with intake array scan grid on rotor LE 55% shaft speed

z Blades vaguely visible

12th CEAS Workshop, 23-24 October 2008, Bilbao

22

Intake array

Intake array (4)

ROSI results with intake array scan grid on rotor LE 70% shaft speed

z Blades visible z Best result with broadband noise

12th CEAS Workshop, 23-24 October 2008, Bilbao

23

Intake array

Intake array: rotor vs stator sources z Difficult to unravel rotor and stator sources: z z z

lack of axial resolution application of source integration technique not useful rotor sources are “spread out” in stator plane, and vice versa

z Looking at mode spectra might be helpful

12th CEAS Workshop, 23-24 October 2008, Bilbao

24

Intake array

Intake array: azimuthal modes

Azimuthal modes of intake array broadband noise 55% shaft speed

z Positive modes prevail

12th CEAS Workshop, 23-24 October 2008, Bilbao

25

Intake array

Why do positive modes prevail? z Possible explanations for dominance of positive modes: 1. 2. 3.

Sources are rotor bound, and prevalence is due to source rotation Sources are stator bound, and prevalence is due to source directivity Sources are stator bound, and prevalence is due to rotor shielding

12th CEAS Workshop, 23-24 October 2008, Bilbao

26

Intake array

Option 1: Rotor source rotation (1) Expression for rotating monopole:

1 p (ω ) = 2π

∞

∑ σ (ω − mΩ)e

− im (θ −φ )

m =−∞

source spectrum

∞

e ∑ μ

Dm (ε mμ r ) Dm (ε mμ ρ )

iα mμ ( x −ξ )

iQmμ Dm (ε mμ ) 2

=1

fan rotation speed

schematic source spectrum (from ROSI)

10 dB

increasing m → decreasing ω-mΩ → increasing levels 0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

Fre quency [Hz]

12th CEAS Workshop, 23-24 October 2008, Bilbao

27

Intake array

Option 1: Rotor source rotation (2)

Simulation

(including intake liner)

z Not like actual results

12th CEAS Workshop, 23-24 October 2008, Bilbao

28

Intake array

Option 2: Stator source directivity (1) axial flow

stator dipoles swirl

rotor rotation axial flow 12th CEAS Workshop, 23-24 October 2008, Bilbao

29

Intake array

Option 2: Stator source directivity (2) Expression for dipole:

1 p (ω ) = 2π

∞

∑ σ (ω )e

m =−∞

− im (θ −φ )

m ⎛ ⎞ iα mμ ( x −ξ ) Dm (ε mμ r ) Dm (ε mμ ρ ) ∑ ⎜ α mμ sin(ψ ) − cos(ψ ) ⎟ e r Qmμ Dm (ε mμ ) 2 ⎠ μ =1 ⎝ ∞

asymmetric with m

ψ

12th CEAS Workshop, 23-24 October 2008, Bilbao

30

Intake array

Option 2: Stator source directivity (3)

Simulation, ψ=30°

(including intake liner)

z More like actual results

12th CEAS Workshop, 23-24 October 2008, Bilbao

31

Intake array

Option 3: Rotor shielding (1) stator

rotor sound propagation

transmission 12th CEAS Workshop, 23-24 October 2008, Bilbao

shielding 32

Intake array

Option 3: Rotor shielding (2) Expression for “shielded fraction” [Schulten-1992]:

⎧⎪ B Ω m S mμ ( ρ ) = min ⎨1, ( xTE − xLE ) + 2 π α ρ M 2 Re( ) m μ ⎩⎪

⎫⎪ ⎬ ⎭⎪

monopole: 1 p (ω ) = 2π

∞

∑ σ (ω )e

− im (θ −φ )

∞

∑ (1 − Smμ ( ρ ) ) e μ =1

m =−∞

iα mμ ( x −ξ )

Dm (ε mμ r ) Dm (ε mμ ρ ) iQmμ Dm (ε mμ ) 2

dipole 1 p (ω ) = 2π

∞

∑ σ (ω )e

m =−∞

− im (θ −φ )

m ⎛ ⎞ iα mμ ( x −ξ ) Dm (ε mμ r ) Dm (ε mμ ρ ) − − S ρ α ψ ψ 1 ( ) sin( ) c os( ) ( mμ ) ⎜ mμ ∑ ⎟e r Qmμ Dm (ε mμ ) 2 ⎝ ⎠ μ =1 ∞

12th CEAS Workshop, 23-24 October 2008, Bilbao

33

Intake array

Option 3: Rotor shielding (3)

Simulation monopole

(including intake liner)

z Quite like actual results

12th CEAS Workshop, 23-24 October 2008, Bilbao

34

Intake array

Option 3: Rotor shielding (4)

Simulation dipole, ψ=30°

(including intake liner)

z Quite like actual results (maybe even better)

12th CEAS Workshop, 23-24 October 2008, Bilbao

35

Bypass array

Bypass array (1)

CB results with bypass array scan grid on stator TE 55% shaft speed

z Stator vanes clearly visible

12th CEAS Workshop, 23-24 October 2008, Bilbao

36

Bypass array

Bypass array (2)

CB results with bypass array scan grid on stator TE 70% shaft speed

z Stator vanes clearly visible

12th CEAS Workshop, 23-24 October 2008, Bilbao

37

Bypass array

Bypass array (3)

CB results with bypass array scan grid on stator TE 85% shaft speed

z Stator vanes clearly visible

12th CEAS Workshop, 23-24 October 2008, Bilbao

38

Bypass array

Bypass array (4)

CB results with bypass array scan grid on stator TE 100% shaft speed

z Stator vanes clearly visible

12th CEAS Workshop, 23-24 October 2008, Bilbao

39

Bypass array

Bypass array beamforming summary z Results are remarkably good z z

Duct is annular No liner between stator and array

– Uniform flow assumption in simulations maybe not adequate

z Spanwise sources rather than tip sources

12th CEAS Workshop, 23-24 October 2008, Bilbao

40

Bypass array

Bypass array: azimuthal modes (1)

Azimuthal modes of bypass array broadband noise 55% shaft speed

z Symmetry in modes

12th CEAS Workshop, 23-24 October 2008, Bilbao

41

Bypass array

Bypass array: azimuthal modes (2) z Azimuthal modes are symmetric z Sound probably generated at trailing edges

dipoles stator dipoles

12th CEAS Workshop, 23-24 October 2008, Bilbao

42

Improvements

Possible improvement: cage array (1) 0.4 m

point source 0.4 m

M=0.29

“Cage” array

array (5 rings)

scan plane (rotor LE)

from [Sijtsma-2007] 12th CEAS Workshop, 23-24 October 2008, Bilbao

43

Improvements

Possible improvement: cage array (2)

z Much better axial

resolution

12th CEAS Workshop, 23-24 October 2008, Bilbao

44

Conclusions

Conclusions z Beamforming was done using the free-field Green’s function z

tonal sound was filtered out

z Beamforming with the intake array on the stator did not show much detail of

broadband noise source locations. z Only at 70% and 75% shaft speed, the OGV stator vanes seemed to be visible. z In those cases, there appeared to be large coherent source structures, probably due to an instability.

z By beamforming with the intake array on the rotor, the blades were vaguely

recognized as sound sources.

z The mode detection results seem to indicate that most noise in the intake is

generated at the stator.

z By beamforming with the bypass array, sources on the stator vanes could be

made clearly visible.

z The noise sources on the stator vanes seem to be distributed along the span. z

Hence, tip noise sources seem to be of lower importance.

z The bypass mode detection results seem to indicate that the noise sources are

located at the trailing edges of the stator vanes.

z Improvements in beamforming are possible with other array configurations

12th CEAS Workshop, 23-24 October 2008, Bilbao

45

References

References 1.

[Lowis-2006] C.R. Lowis and P. Joseph, A focused beamformer technique for separating rotor and stator-based broadband sources, AIAA 2006-2710

2.

[Rienstra-2005] S.W. Rienstra and B.J. Tester, An analytic Green’s function for a lined circular duct, AIAA 2005-3020

3.

[Schulten-1992] J.B.H.M. Schulten, Transmission of sound through a rotor, DGLR/AIAA 14th Aeroacoustics Conference, 1992

4.

[Sijtsma-2001] P. Sijtsma, S. Oerlemans and H. Holthusen, Location of rotating sources by phased array measurements, AIAA 2001-2167

5.

[Sijtsma-2006] P. Sijtsma, Using phased array beamforming to locate broadband noise sources inside a turbofan engine, AARC Engine Noise Phased Array Workshop, Cambridge, MA, 2006

6.

[Sijtsma-2007] P. Sijtsma, Feasibility of in-duct beamforming, AIAA 2007-3696

12th CEAS Workshop, 23-24 October 2008, Bilbao

46

12th CEAS Workshop, 23-24 October 2008, Bilbao

47