Trame de larticle : Contrle de la zone raccordement dun piquage par un traducteur conformable instru

1
Inspection of complex geometries using smart
flexible phased array
O. Casula, G. Toullelan  CEA/LIST, France Ph.
Dumas Imasonic, O. Roy M2M
2
Context
Monolithic wedge transducer adapted to control of
smooth surfaces
nozzle (3D)
Butt-weld (2D)
No acoustical coupling
Complex components The wedge is not adapted to
changing geometries
3
Simulation of the transmitted field through the
buttweld (1/4) By using simulation software (CIVA)
CAD file definition Simplification of 2D
butt-weld profile
Irregular thickness of the coupling layer
Monolithic wedge probe 10 x 10mm² 2MHz LW45
5mm
10mm
Field Strongly Distorded
optimized characteristics
0dB
? dB
field distortion is due to the irregular
thickness of the coupling layer
4
Optimization of the transmitted field through the
butt weld (2/4)
Geometry compensation by using ? wedge phased
array and adapted delay law
Adapted delay law for focusing LW45
Array of 20 elements 25.8 x 20mm² 2MHz
Irregular thickness of the coupling layer
Double refraction
5mm
optimized characteristics
Distorded Field
-9dB
0dB
the adapted delay law is not sufficient to
compensate the distortion ? single refraction ?
computations made with the CIVA software
5
Optimization of the transmitted field through the
buttweld (3/4)
Geometry compensation by reducing the number of
refraction ? immersion phased array and
adapted delay law
Array of 20 elements 25.8 x 20mm² 2MHz
Adapted delay law For focusing LW45
single refraction
20mm
Water coupling
5mm
-7dB
Split beam Distorded Field
optimized characteristics
-11dB
0dB
the transmitted field is not improved ? No
refraction
computations made with the CIVA software
6
Optimization of the transmitted field through the
buttweld (4/4)
Geometry compensation by suppressing refraction
? flexible phased array and adapted delay law
Adapted delay law For focusing LW45
Array of 20 elements 25.8 x 20mm² 2MHz
Optimised coupling
no refraction
5mm
focus beam Low losses
optimized characteristics
0dB
-2dB
interest of the Flexible Phased Array
computations made with the CIVA software
7
Principle of smart flexible phased array
Optimised coupling
Mechanical device
Embedded Profilometer
Linear distribution of piezoelectric elements
Delay law Computation
8
Set-up of the UT Acquisition System

• UT Acquisition System MULTIX (M2M)
• full parallel multi-channel system (up to 128)
• 2 embedded processors real time process
• driving the profilometer,
• computing the delay laws adapted to the
  distortion
• simulation and imaging tools

profilometer
2D Flexible Transducer
Mechanical system
data from profilometer
US channels
32 linear piezoelectric elements max mechanical
travel 15 mm

• global error lt 2
• time loop 1ms

9
Inspection with smart flexible phased array on
realistic 2D surface
LW45 - depth 40mm
scanning
reconstructed Profile
focal point

• Validation of the process

10
Experimental detection of SDH on realistic 2D
surface
2 sets of 4 SDH (2mm-diameter) located under -a
flat surface (ref.), -an irregular profile.
Beam steering configuration 32 focus points from
0 up to 55 at 40mm depth
11
Experimental detection of SDH on realistic 2D
surface
Results at 2 positions
Distortion of 5mm
scanning
-2.0dB
G7 Irrégular Surface
-4.0dB
G3 planar Surface
G3
0dB (ref)
G7
-3.5dB
Accurate detection (SNR, Bandwidth) Low
sensitivity to surface irregularities (-3.5dB /
5mm)
12
Contact Inspection of 2D and 3D complex geometries
2D geometries
Cladding
Weld crown
3D geometries
nozzle
13
3D geometry Extension of the 2D-concept

• The UT acquisition system (M2M)
• driving the 3D-profilometer,
• computing the adapted delay laws

Matrix distribution of 12x7 (84) elements Molded
in a soft resin
US channels
data from profilometer
Bending radius gt 20mm repetition rate gt 150
Hz overall error lt 5
Manufacturer IMASONIC
14
Inspection of a 3D component. example nozzle
Detection of notches and Flat Bottomed Holes from
the external geometry
Experimental set-up
15
Experimental result of notch detection from the
cone
height 7mm
Focusing parameters skew 30 LW 45 D66 mm
Nozzle front view

• Detection, positioning and sizing
• corner echo (SNR 20 dB) and the tip diffraction
  echo (SNR 7 dB)
• Accurate depth sizing of the notch (hmeasured7
  mm)

16
Experiment from the cylinder/cone junction
focusing parameters LW0 40mm-depth
3D-Reconstruction
Angular scanning
LW0
LW 0
FBH
Backwall echo
Accurate detection (SNR)
17
Conclusion Technological solution for the control
of 2D-3D geometries

• Flexible Array
• - evaluation of the interest by comparison with
  wedge and immersion PA
• - probes manufactured by Imasonic (licence
  agreement)
• mastery of focus beam characteristics (with the
  instrumentation)
• matrix arrays allows 3D-focalisation with good
  spatial resolution
• MultiX electronics and software (M2M)
• real time process rate compatible with the
  scan velocity (rate 2D gt 800Hz and 3D gt 150Hz)
• - Functionality available with only up-grading
  software of the MultiX acquisition system (M2M)
• CIVA software
• simulation tool allowing
• the probe design,
• the predictions of control capabilities on
  complex geometries
• 3D-reconstruction of experimental results (for
  interpretation of flaw detections and
  characterisations)

18
Real time reconstruction of external and internal
complex 2D geometries
Reconstruction with the embedded profilometer
scanning
Backwall echo LW0
CIVAs reconstruction tool
LW0
LW0
LW0
35mm
Backwall echo Segmentation

• Real Time Reconstruction of
• external surface thanks to the embedded
  profilometer (errorlt0.5mm)
• internal surface
• Very close to the theory (average error0.4mm)

Superposition of theory and exp profiles
-0.7mm
0.4mm
CAD
Measurement
19
Validation of the behaviour of the probe with the
CIVA software

• For each position, the focusing parameters are
  adapted to radial notch detection
• - Longitudinal waves
• - incidence angle in the notch reference 45
  (top view)
• Beam perpendicular to the plan defined by the
  radial notch (front view)

Front view
Top view
45
Radial notch
? 3D focalisation Validation of the design of
the probe