QuasiActive Control of Axial Fan Blade Tones

1
16th Annual CAV Workshop May 8-9, 2007
Quasi-Active Control of Axial Fan Blade
Tones Using Optimally Tuned Quarter Wavelength
Resonators
Gary H. Koopmann, Dean E. Capone, Lee J.
Gorny Center for Acoustics and Vibration, The
Pennsylvania State University United States of
America Wolfgang Neise Institut für
Antriebstechnik Abteilung Triebwerksakustik Deutsc
hes Zentrum für Luft-und Raumfahrt (DLR)
Berlin, Germany Olaf Lemke Technische
Universität Berlin, Sonderforschungsbereich
557 Berlin, Germany
2

Research Objective and Relevance

• Axial Flow Fans Are Common Noise Sources in
  Turbomachines (Ventilation Units, Compressors,
  Jet Engines, etc.)
• - Sound Radiated Has Both Broadband and
  Narrowband Components
• - Tonal Noise Occurs at the Blade Passing
  Frequency (BPF) and its Harmonics
• Use of Flow-Excited Resonators Serves as a
  Non-Intrusive, Quasi-Active Means of Attenuating
  Blade Tones of Axial Flow Fans

3
Conceptual Overview - Acoustic Cancellation
Mechanism

• Blade Tone Noise Radiates as a Dipole Source

• Flow-excited Resonator Acts
• as a Secondary Monopole Source

• Resonator is Tuned To Provide Anti-phase Sound
• Near Blade Tips Resulting in Acoustic
  Cancellation

Resonator Source 180 deg out of Phase
with Downstream Noise
Flow Direction
Downstream Fan Noise (Negative Dipole Lobe)
Reduced Noise Level
Upstream Fan Noise (Positive Dipole Lobe)
Center Plane of Fan Blade
4
Flow Excitation of Resonators By Passing Rotor
Blades
Unsteady Surface Pressure
5
Modeling of Resonator as a Second Order System
Resonator Can be Modeled Damped, Mass/Spring
System
Uniform Incident Flow
Perforated Resonator Mouth
Mass mc Stiffness kc Damping
cc
m r (density of air)A (A Opening Area) k
rAcot(wL/A) (L Tube Length) c
p n a (nnumber of
holes)
(ahole diameter)
Oscillating Surface Pressure
Radiated Tonal Acoustic Pressure
6
Summary of Prior Work

• Initial Testing Performed on 10 Bladed 260mm
  Diameter Radiator Fan at the Pennsylvania State
  University
• Limited to Plane Wave Sound Propagations
• Used Perforate Mouth Quarter Wave Tubes to
  Generate Secondary Cancelling Sound Field

• Resonator Acoustic Response Driven Directly by
  Fluctuating Pressure of Passing Fan Blade Tips
• Analytical Model Developed Based on
  Transmission Line Theory to Predict Resonator
  Response
• Unidirectional Propagations of Blade Tone Noise
  Reduced by 20 dB with Negligible Efficiency Loss

7
Differences for Research Conducted At The
Deutsches Zentrum Fuer Luft-und Raumfahrt
Extend the Blade Tone Attenuation Technology
Developed for the PSU Low Speed/Low Pressure Fan
to DLRs Higher Speed/High Pressure Axial Flow
Fan

• Technical Challenges
• -Primary Noise Source Rotor-Stator Interaction
• -Onset of Higher Order Duct Modes
• -Design a More Robust Resonator with Motorized
  Control
• - Increase Number of Resonators to 16
• - Control Multiple Resonators Simultaneously to
• Achieve Optimal Reductions in Blade Tone SPLs

8
Fan Testing Facility DLR - Berlin Germany
9
Fan Testing Facility - Fan Configuration

• Ducted Low-Speed, High Pressure Axial Fan
• Max Operational Speed 4000 RPM
• 16 or 18 Bladed Impellers Mountable
• Diameter of 357.4 mm
• Shared NACA 5-63-(10) Blade Profile
• Tip Clearance of .3 mm
• 16 Stator Vanes
• Rotor Stator Spacing of 39.5 mm (From Trailing
  Blade Tip)

10
Experimental Apparatus Resonator Design Overview
Piezo Electric Controlled Motor
O Ring Seals
O Ring Seals
Tube Clamp
Coupled Hexagonal Drive Shaft
Quarter Wave Tube
Resonator Mouth Opening Pattern
Mouth Contour Flush With Fan Shroud ID

• Overall Quarter Wave Tube Length 108 to 70.5mm
• Average Cross Sectional Area 381 mm2
• Tube Wall Thickness - .74 mm

11
Experimental Setup Resonator Design Features

• Length Manually Adjustable
• Variable from 103 to 70.5 mm
• Allows for Coarse Tuning of Resonator
• Controllable Mouth Opening
• -Mouth Opening Perforate Adjustable
• 0 to 42 Open Area
• (Corresponds 0 to 100 Open Resonator)
• -Impedance Governed by Opening Configuration
• Piezo-Electric Amplifier
• -Position Set by Voltage Input (0 to 5 V)
• -Control Signal Provided by 10 kW 1-Turn
  Potentiometer Coupled to Motor Shaft
• -The Control System is Accurate to 5

12
Fan Testing Facility Resonator Integration

• Resonators Equally Spaced Around Shroud
• Circumferential and Axial Mouth Positioning
• -Shroud is Manually Rotatable
• -Spacer Rings Between Shroud/Stators
• Traverse Mouths Axially Across Blades
• Mouth Surface Flush Mounted
• -Communicate with Blade Tip Pressure
• Field Without Significant Impact on Flow

• Initial Positioning Centered 30mm
• upstream of Stator Leading Edge
• (9.5mm DS of Trailing Blade Edge)
• 16 or 18 Resonator Configurations

13
Fan Testing Facility Data Acquisition
Blade Triggered Tachometer Used as Reference
Phase
Ring of 16 Equally Spaced Wall Flush-Mounted ¼
BK Microphones
¼ Microphones With 20 mm long 1.4mm ID probe
Tube
14
Flow Free White Noise Excitation of
Resonator Variation Mouth Opening Percentage
Loudspeaker

• Blocked Driving Pressure Measured Through Covered
  Resonator Opening
• Response Curves Obtained Approximately
  Characterize Flow Driven Resonators
• Quality Factor and Resonance Frequency Shift With
  Perforate

15
Measurement of Unsteady Resonator Driving
Pressure Field Incident on Fan Shroud Due to
Passing Blades

• 64 ¼ Pressure Probe Microphones Measure Incident
  Pressure on Fan Shroud Surface
• Measured Results Provide Driving Pressure for
  Flow Excited Resonator
• Measurements Taken
• Every 1.25mm Axially in
• the Blade Passing Region
• with 2,4 and 8 mm Spacing
• Outside of This Area

• Circumferential Position is Varied by 1 Degree
  For All Positions
• Between Two Adjacent Stator Vanes
• (0 Deg. Orientation Corresponds to Trailing Edge
  of Stator Vane)
• Phase Measured Relatively to Blade Triggered
  Tachometer

16
Incident BPF Pressure Magnitude and Phase on Fan
Shroud Measured Axially Along Blade (4000 RPM
Case)
Circumferential Positions in Legend Correspond to
Angular Distance of Measurement Location From the
Trailing Edge of a Stator Vane
17
Methodology of Defining Resonator
Positions Relative to Rotor and Stator Blade
Positions
4.25 deg
Direction of Positive Rotations
22.5 Degrees Stator Spacing Stator TE
Corresponds To 0 Degrees Orientation
Stator Leading Edge at 9 Deg
Optimally Cancelling Resonator at 4.25 Degrees
-12 deg
22.5 deg
No Spacer Configuration 30mm
Blade Position Corresponds to Reference
Orientation of Blades
10.5 deg
18
Quarter Wavelength Resonators Used to
Attenuate Upstream Fan Noise (All Resonators
Active)

• The Following Slides Demonstrate the
  Effectiveness of Resonators in Attenuating Plane
  Wave Propagations of Blade Tone Noise Using the
  16 Blade Rotor

Positioned 4.25 Degrees From Stator TE 100 open
Resonators (No Axial Spacer)
19
Quarter Wavelength Resonators Attenuating
Downstream Fan Noise
13.25 Degrees 100 open All Following Plane Wave
Experiments Conducted with Resonator Length
103mm
20
Comparison of Upstream and Downstream Fan Noise
Attenuation Resonator Positions

• Optimal Upstream and Downstream Optimal Positions
  Vary by 9 Degrees Orientation (144 Degrees Phase
  Difference)
• Expected Result would be 180 Degrees Phase as
  Dipole Like Fan Noise
• Since The Resonator Locations is Upstream of the
  Stator by a Distance of 30 mm The Difference in
  Source Location Must Be Considered

21
Resonator Fan Tone Reduction For Downstream
Propagation Fan Speed 3000 RPM - Resonator
Located at 13.75 Degrees
For Fan BPFs Closer to Resonance, The Mouth
Impedance Is Increased by Reducing the Opening of
the Resonator Such that the Response is of the
Appropriate Magnitude
(Shroud position is adjusted slightly to optimize
resonator phasing)
22
Resonator Fan Tone Reduction For Downstream
Propagation Fan Speed 2285 RPM - Resonator
Located at 15 Degrees
Resonator Driven Below Resonance
(Resonator Mouth Opening Adjusted to 50 Opening)
23
Resonator Fan Tone Reduction Using Reduced
Number Of Resonators Located at 13.25 Degrees
The Number of Active Resonators can be Reduced
from 16 to 8, 4 and 2 Resonator Configurations
(Evenly Spaced)
Reduced Number of Resonators can be Used
Obtaining Similar Reductions to All 16 Plane
Waves Excited - Source Location Less Critical
24
m 2 BPF Tone Reduction Using 16 Resonator
Shroud Experimental Determination of Optimal
Circumferential Positions Upstream and Downstream
Attenuations (3300 RPM Fan Speed)
Optimal Upstream Position at 9.75 deg (Position
7.5) for m 2 Mode
Optimal Downstream Position at 6 deg (Position
0) for m 2 Mode
Here Optimal Bi-Directional Attenuation Occurs at
approximately Position 3 Reducing Upstream and
Downstream BPF levels to 107 dB from 112.4 and
113.7
The Following Slides Are for Higher Order Modes
18 Blade Rotor (6 mm Spacer Ring In Place)
25
m 2 BPF Tone Reduction Using 16 Resonator
Shroud Diagrams of Optimal Resonator Positions
for Upstream and Downstream Attenuations (3300
RPM Fan Speed)
Optimal Upstream
Optimal Downstream
Once Optimal Positions for Resonator Sources are
Determined Resonator Modification Length and
Opening Impedance Allows for Source Strength to
be Adjusted for Optimal Fan Noise
Cancellation Here Optimal Positions are Very
Close (lt4 degrees) Bi-directional Noise
Reduction is Possible For Higher Order Mode
Propagations
26
m 4 2XBPF Tone Reduction Using 16 Resonator
Shroud Experimental Determination of Optimal
Circumferential Positions (3300 and 4000 RPM Fan
Speeds)
Two Optimal Locations for Upstream Reduction
9.75 and 21 degrees m 4 Mode
Optimal Downstream Reductions 17.25 and 3.5
degrees m 4 Mode
27
Optimized Upstream m 2 BPF Tone Reduction
Resonator Position 9.75 degrees (3300 RPM Fan
Speed)
28
Optimized Upstream m 2 BPF Tone Reduction
Resonator Position 9.75 degrees (Varying Fan
Speed)
Reductions of Fan BPF Tone Using Mouth Tunable
Resonator at Fixed Position (9.75 deg)
Fixed Resonator Position can be used to Attenuate
a Specific Mode Generated By A Variable Speed Fan
Finer Resolution Testing Would Have Likely
Resulted in Higher Attenuations as Optimal
Resonator Response Would Have Been Used. This
would Be particularly evident in the 3100 RPM
test case where the resonator responds near to
its resonance and is effective with only 20
opening.
29
Further Analysis of Measured Data to Verify
Expected Resonator Response

• Flow Driven Resonator Response can be Determined
  Through Measurement Closed End Sound Pressure.
• This Measurement with Transmission Line Theory
  Can Be Used to Track Resonator Response
  (Potential for Adaptive Control)
• Downstream Sound Pressure is Measured to
  Determine the Minimum Noise Output

Closed End SPL Measurement
Resonator Sound Field
Resonator Microphone
Downstream Microphone
30
Further Future Work In Axial Fan Noise Control

• Dipole Resonator Configurations are Being
  Investigated to Attenuate Plane Wave Tonal Noise
  Propagations at the Pennsylvania State University
  Facility
• Further Analysis of Measured Data From DLR
  Study to Improve Understanding of Physical
  Mechanisms of Noise Attenuation
• Future Testing At DLR using Resonators
• Dipole bi-directional Plane Wave Reductions
• Quasi-Active Noise Control Possibilities
• Larger Scale Higher Order Mode Attenuations Using
  Resonators
• Further Testing To Understand Physical
  Attenuation Mechanisms
• Further Fan Noise Applications for Both
  Statored and Unstatored Fans

31
A Vision for Using Quarter Wavelength Resonators
to Control Turbofan Noise from Jet Engines
Resonators integrated into nacelle