Defect Characterization Using Time-of-Flight Diffraction

Defect Characterization using
Time-of-Flight Diffraction
SAJU T ABRAHAM
Quality Assurance Division
Indira Gandhi Centre for Atomic Research
TOFD Principle
• The interaction of the ultrasound waves with the discontinuity will
results in the following processes: Specular reflection from the
faces of the crack obeying laws of optical reflection, Traverse
along the discontinuity surface as Rayleigh waves and radiating
when reaching the crack tips, Scattering from the faces of the
discontinuities, Mode conversion, Diffraction from the crack tips
Incident Longitudinal wave
Reflected Longitudinal Wave
Reflected Shear Wave
Diffracted Longitudinal wave
Diffracted Shear wave
Diffracted Conical Wave

Crack Surface wave
Fig.1: Interaction of a plane compression wave with a weak half-plane crack

Diffraction potential
• Consider a plane wave incident at an angle β and the observer
located at an angle θ, then for incident compression waves
diffracted as compression waves, the diffracted potential is
 p 
1/ 2
  G p ( ,  )  exp( ik p R)
 R
 1   2   
G p ( ,  )    exp( i / 4) sin  
 D  2
Gp (θ, β) is the diffraction coefficient, R is the distance from the
crack edge to the point of observation, λp is the wave length of
the compression waves, kp = 2π/λ is the wave vector, β is the
angel of incidence and θ is the angel of the diffracted wave with
respect to the crack.
Diffraction amplitude
The diffracted energy from the crack tip is distributed over a
wide range of angles. Therefore the TOFD technique is less
dependent on defect orientation. In TOFD measurement, the
angular distribution of the diffracted signals must be known in
order to best position sources and detectors. Ogilvy and
Temple have calculated the diffraction pattern for compression
and shear waves incident on a semi-infinite crack at any angle.
This mathematical model is based on the following assumptions:
1. The incident ultrasonic waves are infinite plane waves
2. The crack is a semi-infinite plane
3. Crack faces are non-interacting and smooth.
Zero points in the diffracted
pattern
• It can be shown that for the transmitter and receiver pair
placed symmetrically across the crack, the amplitude falls to
zero at angle 38˚.
• This is because the diffracted amplitude contains two terms
ψ1 and ψ2 of opposite sign which cancel out at this point.
• Similar zero points are predicted for asymmetrical
arrangements.
• The finite width of ultrasonic beam, in practical case, means
that the zero point may be observed as a minimum rather
than a true zero.
• All these predictions are made on the assumption that the
ultrasound incident on the crack tip is a continuous plane
wave.
• This will not make any serious problems with the pulsed
ultrasound since it can be Fourier treated.
Phase of the diffracted waves
• The phase of the diffracted signal is independent of
frequency and can be calculated from the diffraction
coefficient G (θ, β).
• It can be seen that at the angles corresponding to the zero

points, these is an associated phase change of 180˚.
• Generally the phase of diffracted signals affected by pulse

profile, finite crack tip width, or crack front curvature.
• For partially transparent cracks, the diffraction pattern will

be complicated so that the exact phase relation can not be
satisfied.
Compression waves in TOFD
Compression waves are usually used for TOFD studies
because of the following advantages:
• Travel faster than any mode converted shear wave so that it

will reach the receiver without mixing up with the shear
component.
• Less interference and scattering effect due to the long wave

length (hence more suitable for weld examination).
• Shear waves can also be used for TOFD examination.
• Because of its lesser wave length, the accuracy will be more.
• Scattering will be high.

Geometrical Arrangement of
TOFD
1. Lateral wave
2. Diffracted echo from the top of the defect
3. Diffracted echo from the bottom of the defect
4. Back wall reflection
Calculations
• Lateral wave:
2S
C
2
• Top diffracted echo: S2  d2
C
2
• Bottom diffracted echo: S 2  (d  a) 2
C
2
• Back wall reflection: S2  H2
C
• Defect depth from the 1
surface: C 2 t1  4S 2
2
2
1
• Defect extent: C 2 t 2  4S 2  d
2
2
Advantages & Limitations
• TOFD is not affected by • TOFD cannot characterize the
the orientation of the signals from different
defect. defects/discontinuities
• Size, shape and orientation
are accurately determined. • Due to the presence of lateral
wave, subsurface defects in a
• Probability of detection thin specimen are not easy to
(POD) is much higher and identify.
false call rate (FCR) is very
less • It requires access to both the
• It is a high-speed sides of a discontinuity.
technique.
• Because of the small signals
• TOFD technique can reveal and high gains required, image
the presence of subsurface processing system and pre-
and open surface cracks. amplifier are usually required.
• Efficient for thick-wall
inspection.
Some Experimental Images
B-Scan of vertical notch
Lateral wave
Top Diffracted
Bottom
Diffracted
Back wall
Echo
Mode
converted
part
A-Scan of vertical notch
Top diffracted echo
Bottom diffracted echo
Back wall echo

B-Scan of tilted notch
Lateral wave
Top Diffracted
Bottom
Diffracted
Back wall
Echo
Mode
converted
part
A scan views of tilted notch
Here we
can
calculate
the
orientation
B-Scan views of SDH
1. 4mm SDH
2. 3mm SDH
3. 2mm SDH
4. 1mm SDH
Mode
converted
part
A-scan view of 4mm SDH
• It can clearly
distinguish the top
and bottom
diffracted
echoes.
Lateral wave
Diffracted
echoes.
Back wall
reflection
A-scan view of 1mm SDH
• The signal
appearing is the
sum of the top
and bottom
diffracted waves
• It is not possible
to differentiate
these two.
• Use signal
processing
algorithms.
D-Scan of Lack of Penetration
Specimen: SS
Thk: 25 mm
Lateral wave Weld: Double V
D-Scan of Slag
Specimen: SS
Lateral wave Thk: 25 mm
Weld: Double V
Aim of the study
• Train an Artificial Neural Network to

classify the defects.
Reason
• It is not possible to differentiate signals
from different defects.
• The signals from different defects (LP, LF,
PO, Slag, Cracks) have its own
characteristics.
• If the defect dimension is less or it is a thin
specimen, it is not possible to distinguish the
signals (one solution is immersion technique).
• Use the advanced signal processing algorithms
to extract the features.
Work so far
• Literature survey on ultrasonics, TOFD,
Phased Array and other branches of NDT
(RT, LPT, MPT, ECT).
• Basic experiments on UT, TOFD and RT.
• Specimens: Pipes, weld pads and parent
materials (steel); both field and in-lab.
Reference _ Basics
1. J.A. Ogilvy and J.A.G. Temple, Diffraction of elastic waves by
cracks: application to time of flight inspection, Ultrasonics, Nov.
1983, PP 259 – 268.
2. F.A. Ravenscraft, K. Neuton and C.B. Scruby, Diffraction of
ultrasound by Cracks: comparison of experiment with theory,
Ultrasonics Jan. 1991 vol.28, PP 29-37.
3. J.A. G. Temple, “Time-of-flight inspection: theory”, Nucl. Energy,
1983, 22, No. 5, Oct., 335 – 348.
4. J. B. Keller, Geometrical Theory of Diffraction, JOSA, Vol 52, No
2, 1962, pp. 116 – 130.
5. J. B. Keller, Diffraction by an Aperture, J. Appl. Phy., Vol 28, No 4,
1957, pp. 426 – 444.
6. J. M. Coffey and R K Chapman, Application of elastic scattering
theory for smooth flat cracks to the qualitative prediction of
ultrasonic defect detection and sizing, Nucl. Energy, 1983, 22, No
5, Oct., pp. 319 – 333.
Reference _ Signal Processing
• E. P. de Moura, M. H. S. Siqueira, R R da Silva, J. M. A. Rebello
and L. P. Caloba, Welding discontinuity pattern recognition in
TOFD signals, Part 1. Linear classifiers, Insight Vol 47 No
12 December 2005, 777-782
• E. P. de Moura, M. H. S. Siqueira, R. R. da Silva, J. M. A.

Rebello, Welding discontinuity pattern recognition in TOFD
signals, Part 2. Non-linear classifiers, Insight Vol 47 No 12
December 2005, 783-787.
Reference _ Signal Processing
• G. Baskaran, Krishnan Balasubramaniam, C V Krishnamurthy
and C Lakshmana Rao, Ultrasonic TOFD flaw sizing and
imaging in thin plates using embedded signal identification
technique (ESIT), Insight Vol 46, No. 9, Sep. 2004.
• W AI-Nuaimy and O Zahra, Time-of-flight diffraction -

from semi-automatic inspection to semi-automatic
interpretation, Insight Vol. 47, No. 10, Oct. 2006, pp. 639-
644.
• J.L.B.C. Veiga, A.A. de Carvalho, I.C. de Silva and J.M.A.

Rebello, The use of artificial neural network in the
classification of pulsed-echo and TOFD ultrasonic signals,
J. of the Braz. Soc. of Mech. Sci. & Eng., Oct-Dec. 2005, Vol
27, No 4, pp. 394-398.
Thank you

Defect Characterization Using Time-of-Flight Diffraction

Uploaded by

Copyright:

Available Formats

Defect Characterization Using Time-of-Flight Diffraction

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Defect Characterization Using Time-of-Flight Diffraction

Uploaded by

Copyright:

Available Formats

Defect Characterization using

Incident Longitudinal wave

Reflected Longitudinal Wave

Reflected Shear Wave

Diffracted Longitudinal wave

Diffracted Shear wave

Diffracted Conical Wave

Fig.1: Interaction of a plane compression wave with a weak half-plane crack

• It can be seen that at the angles corresponding to the zero

• Generally the phase of diffracted signals affected by pulse

• For partially transparent cracks, the diffraction pattern will

• Travel faster than any mode converted shear wave so that it

• Less interference and scattering effect due to the long wave

• Shear waves can also be used for TOFD examination.

• Because of its lesser wave length, the accuracy will be more.

• Scattering will be high.

Top diffracted echo

Bottom diffracted echo

Back wall echo

• Train an Artificial Neural Network to

• E. P. de Moura, M. H. S. Siqueira, R. R. da Silva, J. M. A.

• W AI-Nuaimy and O Zahra, Time-of-flight diffraction -

• J.L.B.C. Veiga, A.A. de Carvalho, I.C. de Silva and J.M.A.

You might also like