Upload
phamnhu
View
229
Download
1
Embed Size (px)
Citation preview
Penn State Center for Acoustics and Vibration (CAV)
Structural Vibration and Acoustics GroupPresented as part of the 2014 CAV Spring workshop
Stephen Hambric, Group Leader
April 2014
Andrew Barnard Kevin KoudelaRobert Campbell Tim McDevittJames Chatterley Scott MillerStephen Conlon Dan RussellTyler Dare Micah ShepherdJohn Fahnline Alok Sinha Sabih Hayek Marty TretheweyTony Jun Huang
2/18April 2014
CAVToday’s topics
• Quiet rotorcraft roof panels• Dr. Steve Hambric
• Vibration and high cycle fatigue of turbine blades• Alok Sinha
• Flow-excited ribbed panel optimization• Micah Shepherd, ARL and PhD student, Acoustics
• Winner of INCE-USA Beranek Student Medal, 2014
• 2015 Noise and Vibration Conference
• Acoustic black holes• Dr. Stephen Conlon• Phil Feurtado, PhD student, Acoustics
3/18April 2014
CAV2015 Conference
4/18April 2014
CAV
Quiet Rotorcraft Roof Panels
Principal Investigators: Dr. S.A. Hambric, Dr. K.L. Koudela, M.R. Shepherd (PhD, Acoustics), and D.B. Wess
Sponsor:
Collaborators:
5/18April 2014
CAVRotorcraft Cabin Noise
• Strong transmission gear meshing tones excite roof panel
6/18April 2014
CAVBaseline Panel
• Manufactured at Bell Helicopter (Textron)
7/18April 2014
CAV
• Stiff and lightweight– Carbon fiber face sheets– Nomex core
Honeycomb core sandwich panel
8/18April 2014
CAVBaseline Panel – FE/BE modeling
• All solid quadratic elements, smeared face sheet properties
• BE model of surrounding air
9/18April 2014
CAVTransmission Loss
0
5
10
15
20
25
30
35
40
100 1000 10000
Transm
ission Loss (d
B)
Frequency (Hz)
Measured
Simulated ‐ FE/BE
Simulations within 3 dB of NASA SALT measurements
10/18April 2014
CAVDamping of face sheets
• 3M VHB 9469– Very thin (0.005”) adhesive
with high material losses
Honeycomb core (0.50 thick)
Visco(0.005 thick)
Carbon bag ply(0.0079 thick)
Carbon tool ply (0.0079 thick)
Carbon tool ply + adhesive(0.0079 thick)
Carbon bag ply + adhesive(0.0079 thick)
Visco(0.005 thick)
11/18April 2014
CAVDamping of face sheets –coupon tests and FE simulations
FE within 4% of measurements
12/18April 2014
CAVDamping of face sheets –coupon tests and FE simulations
13/18April 2014
CAVDamping of face sheets –Projected performance
0
5
10
15
20
25
30
35
40
100 1000 10000
Transm
ission Loss (d
B)
Frequency (Hz)
Simulated ‐ lf=0.05
Simulated ‐ lf=0.01
14/18April 2014
CAV
Vibration and High Cycle Fatigue of Turbine Blades
Principal Investigator: Dr. Alok Sinha
Sponsors: GUIde 4 ConsortiumAir Force Research Lab
15/18April 2014
CAVTurbine Blade Geometric Mistuning
V. Vishwakarma, A. Sinha, Y. Bhartiya and J. Brown, “Modified Modal Domain Analysis of Coordinate Measurement Machine Data on Geometric Mistuning,” GT2013 - 94393, Proceedings of ASME Turbo Expo, San Antonio, Texas, June 2013
MMDA (Modified Modal Domain Analysis): High Fidelity Reduced Order Model– Geometric Mistuning: variations in blade geometry due to manufacturing
tolerances– Proper Orthogonal Decomposition (POD) of Coordinate Measurement
Machine (CMM) data on blades’ geometries– Validated on an industrial rotor
16/18April 2014
CAV
Flow-excited ribbed panel optimization1
Principal Investigator: Micah Shepherd, PhD student, Acoustics
Dr. S.A. Hambric, Advisor
Sponsor:
1Shepherd, M., Structural-acoustic optimization of structures excited by turbulent boundary layer flow, PhD dissertation in Acoustics, May 2014.
17/18April 2014
CAVAutomated design optimization
Use stochastic, global optimization procedure
AnalysisUpdate Design Variables
Model
Objective Function
Finite Element
Boundary Element
Shape Properties
18/18April 2014
CAVFlow-excited structure
Simply‐supported edges Aluminum Pressurization (55 kPa) Radiation from plate only TBL flow at 216 m/s
19/18April 2014
CAVUnconstrained Optimization
20/18April 2014
CAVConstrained Optimization(Limited center displacement)
ARL Penn State
1
Acoustic Black Holes for
Noise and Vibration Control 29 April 2014
Center for Acoustics & Vibration
Annual Spring Workshop
Stephen C. Conlon Head, Noise Control & Hydroacoustics Division
Applied Research Laboratory
814-863-9894 [email protected]
ARL Penn State
2
ABH Research Team: Theory & Applications
Students:
Phil Feurtado, Angelina Conti
Research Faculty:
Steve Conlon, John Fahnline, Micah Shepherd
Recent Publications:
Conlon, S.C. and Semperlotti, F., “Passive control of vibration and sound transmission for vehicle structures via embedded Acoustic Black Holes,” Noise-Con 2013, August 26-28, Denver Colorado (2013).
Liuxian Zhao, Stephen C. Conlon, Fabio Semperlotti, “Broadband Energy Harvesting using Acoustic Black Hole Structural Tailoring,” (accepted for publication) Smart Materials and Structures.
Philip A. Feurtado, Stephen C. Conlon and Fabio Semperlotti, “A normalized wavenumber variation parameter for acoustic black hole design,” (submitted for publication) Journal of the Acoustical Society of America.
Stephen C. Conlon, John B. Fahnline and Fabio Semperlotti, “Acoustic Black Holes for Noise and Vibration Control,” (submitted for publication) Smart Materials and Structures.
Zhao L., Semperlotti F., “Vibration Control and Energy Harvesting Based on Embedded Acoustic Black Hole,” Noise-Con 2014, Fort Lauderdale Florida.
Students:
Liuxian Zhao, Hongfei Zhu
Research Faculty:
Fabio Semperlotti
ARL Penn State
3
Motivation
• Increasing needs for lightweight high loss structures • Vibration attenuation
• Transmission loss
• Vehicle applications • Interior acoustics
• Radiated sound power
• New design opportunities • Embedded “designed-in” high loss treatments – new systems
• Advanced composites
• Multifunctional structures
• “Add-on” treatments - problematic existing systems
ARL Penn State
4
Motivation: Vehicle Applications
Efficient Passive Noise &
Vibration Control
ARL Penn State
5
Background: Black Hole History
Concept of an object from which light could not escape -black hole
• Originally proposed by Pierre Simon Laplace in 1795
• Using Newton's Theory of Gravity, Laplace calculated if an object were compressed into a small enough radius - the escape velocity of that object would be faster than the speed of light
• Singularity point has zero volume and infinite density
ARL Penn State
6
Black Holes Background: Acoustic Wave Propagation
Pekeris, JASA 18 (1946), absence of reflected waves - layered inhomogeneous media w/ sound velocity profile that decays linearly to zero w/ increasing depth
Mironov, Sov. Phys. Acoust. 34 (1988), no reflection of waves in plate w/ thickness decreases smoothly to zero over finite interval
“Acoustic Black Holes”
Krylov, (multiple publications 2001- 2013), power law one-dimensional wedges, two-dimensional ABH feature in plates
More recently many other authors examining fundamental 1D and 2D concepts theoretical and preliminary measurements of single features
Next Step – need to examine how to apply ABH’s to practical noise control applications
ARL Penn State
7
ABH Theory • Bending Wave Speed
– Aluminum – 6.8 mm (0.268 inch) – Power taper = 2.2
( )( )
( )
[ ] ( )
12
2
12 2
22 3 1
, 0 2, 2
constant
0
b L
f
m
m
b taper
b
c c
h Ef
Lh x h x
LL x x L
c f L xc as x L
ωκ
πρ ν
ε
−
=
= −
= ≤ ≤
= − < ≤
= −
→ →
0L/2
L100
1k 10k100k1e6
10-2
100
102
104
Frequency, (Hz)plate span
wav
e sp
eed
(m/s
)
0 50 100 150 200 2500
10
20
30
40
plate span (mm)
plat
e th
ickn
es (m
m)
Plate span increments= 100, Taper exponent= 2.2
power profileuniform plate thicknessplate with end power taper
ARL Penn State
8
• Many authors cite no reflections (ABH effect) with power taper – Due to the vanishing wave speed as tip is approached – Also cite need for added damping layer to dissipate resulting
reflections due to “practical manufacturing limitations” (tip thickness does not reach zero)
• Singularity at tip of power taper structure – Wave speed vanishes (infinite time to reach tip, no reflections) – Velocity becomes infinite (the second effect of the singularity)
• The velocity result tells us for practical structures we NEVER want the thickness profile to reach zero even if we could manufacture it!
ABH Theory: The Singularity
fh h= ABHx
( )0
0
x xh x h=
= ( )0
0 0xh=
=
( )h x
ARL Penn State
9
ABH Theory: Wedge Simulation
Tapered wedge simulations • Taper power= 2.2 • Material= Aluminum • Thickness= 6.8 mm (0.268 inch) • Material Damping: 2%
Damping layer: • Thickness= 2.2 mm • Damping : 10% • Excitation= harmonic at 5kHz • NASTRAN Sol 109 – Time Transient
•Finite thickness at wedge termination – significant reflections
•Addition of damping layer eliminates majority of reflections
With added damping layer
ARL Penn State
10
Simulations
Finite Element / Boundary Element Models & Response Simulations
• Transient excitation (FE) - wave attenuation
• Harmonic excitation – Structural acoustic coupling (FE/BE)
- accelerance / vibration reduction - radiated sound power reduction
- structural intensity
10
ARL Penn State
11
Transient Excitation Results: Vibration Reduction
• Aluminum Plate – 6.8 mm thick (0.268 inch) – 5 x 5 grid of ABH – Material damping: 2% – ABH
– Diameter: 10cm – Taper exponent: 2.2 – Damping layer thk: 2mm – Damping: 10%
– Sinusoidal excitation (5 kHz) – NASTRAN Sol 109, Δt=1.e-7
• Vibration completely attenuated ~ 4th ABH Row
ARL Penn State
12
Results: Vibration & Radiated Sound Reduction
FE / BE Models • Aluminum plate thk= 6.8 mm (0.268 in) • Panel masses, uniform = 9.09 kg , ABH = 7.58 kg • Panel length and width = 0.7 m (both panels) • Aluminum damping loss factor = 0.001 • Damping layer thickness = 2.0 mm • Damping layer loss factor = 0.1
• Air loading on one side of panels • Boundary conditions
• Mechanical – edges unsupported (free) • Acoustic – baffled
• Unit force point drive • Uniform panel critical frequency ~ 1900 Hz
25 ABH plate
Uniform plate
ARL Penn State
13
-20
0
20
40
60
80
100
120
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
dB
Frequency (Hz)
Results: Vibration & Radiated Sound Reduction
Total radiated sound power
Surface averaged accelerance
Vibration & sound power reduction (5 to ~20 dB) above 2.5 kHz ABH panel = 80% mass of uniform panel
Uniform panel
ABH panel
Uniform panel
ABH panel
ARL Penn State
14
50
55
60
65
70
75
80
85
90
95
100
2000 3000 4000 5000 6000 7000 8000
Aco
usti
c Po
wer
Out
put (
dB re
1x1
0-12
W/N
^2)
Frequency (Hz)
ABH broadband absorption
Results: Radiated Sound Power Reduction
( )6500
*1.0ABH
B plate f Hz
f Diac − ≈
=
Significant ABH panel sound power reduction
ARL Penn State
15
Results: Modal Loss Factors
0.001
0.01
0.1
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Loss
Fac
tor
Frequency (Hz)
• With air loading • Uniform plate
~ 1900critf Hz
~37x
ARL Penn State
16
0.001
0.01
0.1
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Loss
Fac
tor
Frequency (Hz)
Results: Modal Loss Factors
• No air loading • ABH & uniform plates have same damping layers / locations
~37x
ARL Penn State
17
Results: Modal Loss Factors
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Loss
Fac
tor
Frequency (Hz)
ARL Penn State
18
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Loss
Fac
tor
Frequency (Hz)
(0,1) Mode
(1,1) Modes
(2,1) Modes
Results: Modal Loss Factors
ARL Penn State
19
Results: Energy Flow 2x2 ABH
Point force drive
Point damper
Structural intensity: Local ABH mode ~ 2700 Hz
Structural intensity: Global mode ~1000 Hz
ARL Penn State
20
Results: Energy Flow 2x2 ABH
ABH side of plate
Backside of plate
Structural intensity: Local ABH mode ~ 2700 Hz
ARL Penn State
21
Results: Energy Flow 2x2 ABH
Structural intensity: Local ABH mode ~ 2700 Hz
ARL Penn State
22
Summary & Conclusions
• Results demonstrated the potential for arrays of Acoustic Black Hole features embedded in structures for vibration and noise control
• Panel with periodic ABHs showed significant vibration and radiated sound power reduction • ABH panel mass = 83% of uniform panel mass
• Low frequency energy dissipation mechanisms were characterized • Global vs local ABH modal loss factors
• Future noise control work will focus on design sensitivity studies for ABH features & optimization