Development of an advanced ultrasonic phased array for the ...
Transcript of Development of an advanced ultrasonic phased array for the ...
Development of an advanced ultrasonic phased array
for the characterization of thick, reinforced concrete
components
Laurence Jacobs, Georgia Institute of Technology
Kim Kurtis, Jin-Yeon Kim, Prasanth Alapati, Max May (Georgia Tech)
Jianmin Qu, Xiang Gao (Tufts University)
Develop ultrasonic phased array device that uses wave mixing and
nonlinear acoustic techniques to quantitatively characterize and image
microscale (100 μm) damage in thick, reinforced concrete components.Total Project Cost: $0.87M
Length 24 mo.
Project Vision
The Concept
‣ Increase concrete durability by developing an ultrasonic phased array device to
characterize and quantify microscale damage throughout component thickness
‣ Combine ultrasonic phased array technology with nonlinear wave mixing and
concrete material modeling to enable “medical quality” imaging of reinforced
concrete components
The Team
‣ Team of four faculty co-PIs from Georgia Tech (3) and Tufts University (1)
‣ Laurence Jacobs: linear and nonlinear ultrasound, NDE of concrete
‣ Kim Kurtis: emerging and novel methods to understand cement based materials
‣ Jin-Yeon Kim: measurement and modeling methods for NDE, phased arrays
‣ Jianmin Qu: ultrasound and micromechanics of concrete
‣ Jacobs and Kim: development of the phased array device; material modeling
‣Qu: computational modeling to optimize wave mixing scheme
‣ Kurtis: material modeling and development of specimens with known damage
2
Project Objectives
‣ First three quarters: computationally demonstrate and optimize wave mixing
conditions and physically demonstrate capability to generate forward mixed wave
with phased array and array (antennae) sensors
‣ First year go/no-go demonstration of forward wave mixing to detect and
characterize microscale (on the order of 1.0 mm) damage in 3-D with a scan time
under 30 minutes for a volume of 500 cubic inches for forward propagation
(10/15/2020)
‣ Team will develop and commercialize phased array capable of nonlinear wave
mixing in the 10-100 kHz range
‣Working to identify the best first commercial applications: where is an image of
microscale damage through the thickness critical to inform and prioritize
maintenance and repair strategies?
3
Motivation: Nonlinear Rayleigh waves to track ASR damage
‣ Application of nonlinear Rayleigh surface waves to
characterize ASR damage (Previous work by team)
‣Measurements on both EPRI slabs and ORNL/UTK slabs
4
Kim et al, JNDE 2017 and Con Bldg Mat, 2018
Motivation: Wave mixing fundamentals
( , ) cos[ ( )]R R
T
yu y t A t
c
2
0 0 0( ) ( ) ( )
2 ( )
T T
T L T
y U y V y lA
c c c
Resonant
Shear Wave L TR T
L T
c c
c c
Vl
Vs_new
Thickness d
Lo
ng
itu
din
al
wav
e
tra
nsd
uce
r
Received
signal
Vs
y,vx,u
Sh
ear
wa
ve
tra
nsd
uce
r
2L L
T L T
c
c c
Resonant
Condition
y0
( , )Lu y t( , )Tu y t
( , )Ru y t
6
Out-of-plane polarization:
In-plane polarization:
2cos Δ𝜃 2− 1 = cos(2Δ𝜃)
Model for wave mixing: Polarization analysis
Model for wave mixing: Beam steering with an array
• Typical side lobes with peak side lobes
much smaller than main lobe
• Phased arrays are designed for fixed
frequency, nonlinear wave mixing
techniques with varying frequencies would
add additional design constraints
• Array for beam steering only, not detection
• Normalized pressure for beam steering at 𝜃_𝑠=30°, with parameters: 𝛿_𝑥=25𝑚𝑚, 𝑁=8,
𝑓=80𝑘𝐻𝑧, 𝑐_𝑇=2702 𝑚𝑠^(−1), 𝜆_𝑙=3.38𝑐𝑚
• Main lobe at 30°
• Grating lobe appears due to spatial aliasing
for 𝜆_𝑙/2<𝛿_𝑥 at −58°
7
Model for wave mixing: Maximum steering angle 𝜃𝑠8
• Steering angles of both arrays are limited depedent on their
input frequency
• Frequency ratio is fixed by condition at mixing points
• Lower frequencies degrade steering ability
11/6/2020
Computational results: Amplitude of mixing wave at receiver
𝐿 = 𝜉ℎ
h
1 2
Array Transducer #1 Array Transducer #2
Signal Receiver
𝜙
Δ𝜃𝑑1 𝑑2
𝑑3
21 21 2 1 2 1 1 2 2 3 32
( ) cos 1 exp( )T T
L
U UA d d d
C
Amplitude with attenuation
2
2
1 2
1 1 1 2 3 2cos( ) 1 , ,
2T T
m m
Frequency(MHz) Attenuation (dB)
0.1 73.2
0.08 58.5
0.05 36.6
0.03 22
0.01 7.3
0:00:14/9:49
Computational results: Polarization choices
Resonant mixed waves cos 1 2
1 2( ) ( ) ( )SV SV L 2
11 cos2
1 1
1 1
1 2( ) ( ) ( )SH SH L 2
11 cos2
1 1
1 1
1 2( ) ( ) ( )L SV L 1
1 cos
1
2
1k
2k
1 2
k k k
1 2
k k k
2L Tc c
SV – Transversely Polarized Shear Wave
SH – Horizontally Polarized Shear Wave
L – Longitudinal Wavec k
Measurement results: Array equipment
13
Array Antennae
(32 elements x2)
Array
Controller
Possible setup on concrete slabUltrasonic array system
Surrogate Concrete Slab
Interface
Cables
Measurement results: Samples
14Insert Presentation NameNovember 6, 2020
Samples: mini mortar and concrete blocks 15x15x40 cm^3
Concrete
Mortar
Array Antennae
Measurement results: Single-sided inspection using the reflection of the
mixed wave off the bottom surface
Is the amplitude of the nonlinear mixed signal
large enough to be detected after reflection from
the bottom surface?
Primary wave
Secondary wave
Phased Array 1 Phased Array 2
Non-contact receiver
Primary wave
Concrete block
A
B
Inspection volume
Detect reflection of mixed wave off of
the bottom surface
Bottom Surface
1
23
4
Changing the
separation
distance
between
transducers.
Measurement results: Reference mixing in the mini concrete block
@ spot #3 (10 cm from the top surface)
Multiply reflected signals
Window where the
nonlinear mixing
signal is expected,
based on time of
flight
Signal processing
(windowing, band-pass filtering)
Frequency Spectrum
Filtered
nonlinear mixing signal
Bounce mixed ultrasonic signal off bottom surface for single side operation
Original
Measurement results: Reference mixing in the mini concrete block
0
12″
Cross section at
AA1
4″
5″
4.5″
12″
Recycled
aggregate
3″4.5″
Cross section at
AA1
4″
5″
4.5″
12″
ASR
3″4.5″
Cross section at
CC1
4″
5″
4.5″
Entrained
air voids
3″4.5″
Cross section at
DD1
4″
5″
4.5″
Fire
damage
3″4.5″
12″ 12″
Mechanical
damage
Thermal
damage
5 ft
12″
Recycled
aggregate
A
A1
3″4″
4″
5″
B
B1
3″4″
4″
5″
6″
C
C1
3″4″
4″
5″
D
D1
1-3″4″
4″
5″
6″ 14″6″
ASREntrained
air voids Fire
damage
Mechanical
damage
12″Thermal
Measurement results: Sketch of the concrete block showing 4 types of embedded
microscale damage
19
Challenges
‣ One challenge is the inherent physics of the problem: high attenuation due to
scattering from the aggregate and material interfaces in the ultrasonic frequency
range
‣ A second challenge is lack of array (antennae) manufacturers in the required
frequency range: this application is too low for medical (and metallic material)
applications, and too high for near field geophysics applications
‣ A third challenge will be that many owners take the “ignorance is bliss” approach.
‣Wave mixing and nonlinear ultrasound allow for a lower frequency approach and
still enable imaging at the microscale level
‣Once the device has been developed and its accuracy demonstrated, it will be
critical to find the right field demonstration application
Potential Partnerships
‣ Team with concrete NDT suppliers. EPRI and Georgia DOT. Any capabilities to
offer other teams, or potential collaborations with other teams?
21Insert Presentation NameNovember 6, 2020
Summary Slide
‣ Combine ultrasonic phased array technology with an understanding of nonlinear
wave propagation in heterogeneous media and concrete material modeling to
enable “medical quality” imaging and characterization of thick reinforced concrete
components
‣ Nonlinear mixing techniques are not wavelength dependent, thus enabling
characterization of microscale damage (much smaller than inherent
microstructure) in heterogenous concrete material
‣ Team has expertise NDE of concrete, nonlinear wave mixing, phased arrays and
material modeling of concrete
‣ Develop ultrasonic phased array device that uses wave mixing and nonlinear
acoustic techniques to quantitatively characterize microscale damage in thick,
reinforced concrete components
22Insert Presentation NameNovember 6, 2020
0:00:14/9:49
Computational results: Choices of incident wave pairs
1k
2k
1 2
k k k
1 2
k k k
Forward Mixing
Backwards Mixing
1 1 c k
2 2 c k
1 2 c k k
1 2 c k k