Iranian Journal of Oil & Gas Science and Technology, Vol. 3 (2014), No. 4, pp.

16-25
http://ijogst.put.ac.ir

A Novel Method for Ultrasonic Evaluation of Horizontal Defects Using

Time-of-Flight Diffraction
Parastoo Bagheri1, Sina Sodagar1*, Gholamreza Rashed1, and Amin Yaghootian2
1
Department of Technical Inspection Engineering, Petroleum University of Technology, Abadan, Iran
2
Department of Mechanical Engineering, Shahid Chamran University, Ahwaz, Iran
Received: September 19, 2013; revised: February 27, 2014; accepted: April 05, 2014

Abstract
Time-of-flight diffraction method (ToFD) is an amplitude-independent sizing method which is based
on the measurement of time-of-flight of defect tip diffracted waves. Although ToFD can measure
through-wall length of defect accurately, this method is not capable of measuring horizontal defect
size. In this paper, a new ToFD method for evaluating horizontal planar defects is presented. The
finite element method (FEM), using the ABAQUS software package, is employed to simulate the
ultrasonic wave behavior in the test blocks and its interaction with the embedded planar defects. The
phased array technology is also used to model the ultrasonic inspection system parameters. FEM
simulation of the new ToFD method for different crack sizes shows that, compared to the
conventional ToFD method, the accuracy of results is within an acceptable range to use the novel
technique for measuring the horizontal planar defects.

Keywords: Ultrasonic Wave, Diffracted Wave, Horizontal Planar Defects, ToFD

1. Introduction
Non-destructive testing has been increasingly used to assure the quality and reliability in the oil and
gas pipeline industries. The ultrasonic pulse-echo technique uses the pulse flight time to locate the
flaw and the echo amplitude to measure the defect size. Since the amplitude of the reflected pulses
can be influenced by many parameters, such as beam spread, surface roughness, and transparency,
using amplitude is not always sufficient for accurate defect sizing (Krautkramer, 1990). The basis of
the time-of-flight diffraction (ToFD) technique was first initiated at the National NDT Centre,
Harwell, UK. ToFD was developed mainly by Silk and his co-workers at the Harwell Laboratory over
a period of about 10 years starting in the early 1970s, from a laboratory curiosity into a sophisticated
full-scale inspection method capable of detecting and sizing defects in components accurately (Silk,
1973, 1974, 1976, 1978). The ToFD technique is an amplitude-independent sizing method, based on
the measurement of time-of-flight of flaw tips diffracted waves. Golan and Sachese suggested a
method to calculate crack size from the time delay between the arrival of a surface longitudinal
reference beam and a longitudinal or shear beam diffracted from the tips of crack (Golan, 1980). Mak
(Mak, 1983) developed a trigonometric method to calculate location, height, and angle of defect by a
transducer located in different scan positions. The ToFD technique provides the highest possible

*
Corresponding Author:
Email: [email protected]
 P. Bagheri et al. / A Novel Method for Ultrasonic Evaluation of Horizontal … 17

accuracy in measuring the depth and through-wall length of defects (Charlesworth et al., 2001), (Baby
et al., 2003), (Al-Ataby, 2012).

In 1986, the finite element simulation of ultrasonic wave propagation and its interaction with defects
were conducted by Ludwig and Lord (Ludwig et al., 1986). The numerical analysis of wave
propagation for ToFD in an austenitic stainless steel specimen, considering the effects of scattering at
grain boundaries, was carried out by Lin et al. (Lin et al., 2006) and Connolly (Connolly, 2009). They
developed an efficient method for modeling the effects of coarse grains in austenitic materials. In
2007, the simulation of the ToFD technique, using finite element method, was carried out by
Baskaran et al. They used the ANSYS finite element package to model the propagation of ultrasonic
waves in a thin cracked two dimensional specimen (Baskaran et al., 2006). In 2010, Honarvar and
Khorasani used the ABAQUS software package to simulate the propagation of ultrasonic waves and
diffraction phenomena. They compared the simulated results for drilled-hole diffraction with photo
elastic snapshots (Honarvar and Khorasani et al., 2010). Although ToFD provides better accuracy in
locating and sizing defects than other ultrasonic sizing methods and has a high probability of flaw
detection (POD), only the through-wall length of the defect can be measured and the defect real size
cannot be evaluated (Charlesworth et al., 2001). Therefore, the ToFD method cannot be used for
measuring and sizing horizontal cracks (horizontal planar defects) (ASME, 2010). In this paper, a new
ToFD method is presented for evaluating and measuring horizontal planar defects. The finite element
method, using the ABAQUS software package, is employed to simulate the ultrasonic wave behavior
in the test blocks and its interaction with the embedded planar defects. The finite element results for
different crack sizes are used to study and investigate the presence and generation of different wave
modes in the test block as well as the efficiency and efficacy of the new proposed method.

2. Review of the conventional ToFD method

The ToFD technique uses tip diffraction to identify the top, bottom, and ends of a discontinuity. Silk
chose to use an angled compression wave for the ToFD technique rather than a shear wave, for two
reasons. First, the tip diffraction signal is stronger than a shear wave diffraction signal; second, with
an angled compression wave, a lateral wave is produced that can be used to measure the horizontal
distance between the transmitter and the receiver.

The tip diffraction signal is generated at the tip of the discontinuity, effectively a “point” source.
According to Huygens (Krautkramer et al., 1990), a point source produces a spherical wave. Figure 1-
a shows a typical ToFD transducer set-up on a component with a vertical discontinuity. Figure 1-b
shows both the lateral wave and a diffraction beam from the tip of a reflector. There are four sound
paths from the transmitter to the receiver. Path “A” is the lateral wave path traveling just below the
surface. Path “B” is the tip diffraction path from the top of the discontinuity. Path “C” is the tip
diffraction path from the bottom of the discontinuity, and path “D” is the back wall echo path. Figure
2 shows a typical un-rectified received signal using ToFD. It should be noted that the phase
relationships A and C are in the opposite phase to B and D. The important difference to note is
between paths B and C; the top and bottom diffraction signals are in the opposite phase. This phase
difference allows the practitioner to identify those points. Assuming that the diffracting tip is centered
between the two transducers, the depth of crack tips below the inspection surface can be calculated
by:

(Ct 2 ) 2  4S 2
d1  (1)
2
18 Iranian Journal of Oil & Gas Science and Technology, Vol. 3 (2014), No. 4

(C t3 ) 2  4S 2
d2  (2)
2

and thereby,

a  d 2  d1 (3)

where, a is the defect through-wall size, d1 is the depth of the top edge from surface, d2 is the depth of
the bottom edge from surface, and 2S is the probe separation (Figure 1). C is the longitudinal wave
velocity inside the material, t2 and t3 are respectively the travel times of waves diffracted from the top
and the bottom of the crack (Figure 2).

a)

b)

Figure 1
Configuration of time of flight diffraction technique (Hellier, 2003).
 P. Bagheri et al. / A Novel Method for Ultrasonic Evaluation of Horizontal … 19

Figure 2
The time-of-flight diffraction signal (Charlesworth et al., 2001).

3. Finite element modeling of the ToFD method

In this section, the finite element method (FEM) is used to simulate the ultrasonic wave propagation
in the time of flight diffraction technique. The FEM modeling consists of two basic steps: defining the
mesh configuration and problem discretization, and modeling the transmitting and receiving
transducers. The ABAQUS finite element software package is employed for analysis and a two-
dimensional four-node quadrilateral plane strain element, CPE4R, is used in ABAQUS to discretize a
carbon steel test block including vertical crack. The mesh size depends on the frequency of the
propagated wave in the sample and the corresponding wavelength. The piezoelectric angle wave
transducer, transmitter, is simulated by a transient single frequency pulse wave applied on the surface
of the sample. The transient excitation is modeled using a cyclic single frequency pressure/force
function as follows (Mardani et al., 2012):

 2 f t
(1  cos( )) cos(2  f t ), 0  t  N / f
p t    N (4)
 0 otherwise

where, f is the excitation wave frequency and N represents the number of cycles.

Using linear delay law for phased array transducers, the compressional excitations can be applied to
the sequential elements so that ultrasonic wave propagates at a specific angle, θS. The delay time
between adjacent elements, or nodes, is calculated using Hyphen’s principle (Olympus NDT, 2007):

d sin  s
t  (5)
C

where, d is the distance between two adjacent elements, θs stands for the steering angle of
propagation, C is the longitudinal wave speed in the media, and Δt is the time delay between two
adjacent elements.
20 Iranian Journal of Oil & Gas Science and Technology, Vol. 3 (2014), No. 4

To investigate the convergence of the results and the appropriate element size for a 2 MHz frequency
ultrasonic wave, the signal-to-noise ratio (SNR) is obtained for different element sizes. As it can be
seen in Figure 3, at f = 2 MHz, the maximum SNR and SNR convergence occurs for the element sizes
smaller than 60 μm.

60
SNR

30

0
0 50 100 150 200

Element size (μm)

Figure 3
The effect of element size on SNR convergence.

4. The proposed method

As it was mentioned before, using the conventional ToFD method in Equations 1 to 3, the difference
between time-of-flight diffractions of upper and lower crack tips gives the defect through-wall size,
whereas the actual defect size cannot be measured. This means that the conventional ToFD technique
leads to large errors for oblique defects and cannot be used for horizontal defects. In the proposed
method, a novel configuration and the corresponding formula are used so that the ToFD method can
be employed to evaluate horizontal planar defects. Figure 4 shows the proposed ToFD configuration
on the specimen with a horizontal planar discontinuity. As it can be seen, in this configuration, two
transducers including a transmitter/receiver, No. 1, and a receiver, No. 2, are located at the both sides
of the defect.

Figure 4
The proposed ToFD configuration for horizontal planar defects.
 P. Bagheri et al. / A Novel Method for Ultrasonic Evaluation of Horizontal … 21

The ultrasonic wave propagation, in this configuration, is simulated using the ABAQUS software to
study the behavior of ultrasonic wave modes in the test block and their interaction with the defect. In
Figure 5, the different incident ultrasonic wave modes are shown. The transmitting transducer, T1,
emits a short pulse of ultrasonic wave, longitudinal wave, into the component and energy spreads out
as it propagates into the specimen. If the crack face is smooth, there will be a mirror-like reflection of
the wave incident on the face (see Figure 6). For any horizontal planar discontinuity, whether smooth
or rough-faced, diffraction from the edges of the defect causes some fraction of the incident energy
travel towards the receiving transducers R1 and R2 in longitudinal and shear modes at different wave
velocities. As it can be seen in Figures 6 and 7, the mode conversion behavior due to the interaction of
ultrasonic wave with the defect leads to the presence of longitudinal and shear waves from each tip of
the defect. Moreover, three different wave modes, including longitudinal lateral, shear lateral, and
Rayleigh waves travel from the transmitting transducer, T1, to the receiving transducer, R2. (Figure
(5)).

Figure 5
Ultrasonic wave excited modes in ToFD method (before striking defect).

Figure 6
Ultrasonic wave modes during striking defect.
22 Iranian Journal of Oil & Gas Science and Technology, Vol. 3 (2014), No. 4

Figure 7
Ultrasonic wave propagating modes after striking defect.

If the crack is large enough, the signals from the two ends of the defect will be sufficiently separated
in time to be recognized as coming from separate sources. Therefore, using this configuration and the
related ultrasonic wave propagation simulation, the time difference between the received longitudinal
diffracted waves from the left and right defect tips to each receiver, R1 and R2, can be employed to
measure the horizontal defect size. It should be noted that the new method can also be used for
evaluating the vertical defects. The calculation of the horizontal defect size using Pythagoras’ theorem
reads:

( L2  ( L1 / 2))2  ( L1 / 2)2
l (6)
2S
and
L1  t1 CL , L2  t2 CL (7)

where, t1 is the arrival time of the signal diffracted from the left tip of the defect by receiver 1, R1, and
t2 is the arrival time of the signal diffracted from the right tip of the defect by receiver 2, R2. CL is the
longitudinal wave velocity and 2S represents the separation between the transducers.

5. Results and discussions

To investigate the efficiency and efficacy of the proposed method, using finite element modeling, the
novel method is applied to eight carbon steel blocks with different size embedded horizontal cracks.
The test blocks have 100 mm lengths and 20×20 mm2 cross sections and are modeled with the
ABAQUS finite element software package using CPE4R plane strain elements. The acoustic and
elastic properties of carbon steel are given in Table 1. Each block contains a horizontal planar defect.
The defects have lengths of 2, 4, 6, 8, 10, 12, 14, and 16 mm and a depth of 12 mm depth, as can be
seen in Figure 4. The transmitter is modeled as an 8-element 2 MHz phased array transducer. Each
element of the phased array transducer has 0.5 mm length and the gap space between two adjacent
elements is 0.1 mm. The first receiving transducer is located on the position of the transmitter and the
second is located at 35 mm distance from the transmitter on the inspection surface, 2S = 35 mm. The
ABAQUS finite element software package is used to simulate the new ultrasonic ToFD method. The
received signals at the first and the second receivers are shown in Figures 8 and 9. In Figure 8, the
 P. Bagheri et al. / A Novel Method for Ultrasonic Evaluation of Horizontal … 23

first echoes are related to the transient pulse waves given by Equation 4. These echoes are generated
by the eight elements of the phased array transducer with a specific delay time, Equation 5, and are
received by the first receiver. In this signal, the second echo is related to wave diffraction from the left
tip of the defect, which is detected by the first receiver, R1. The back-wall reflection from back
surface of the block is shown as the third echo in this Figure. Figure 9 shows the signal received by
the second transducer, R2. In this signal, the first echo is due to the longitudinal mode of the lateral
wave, which travels from transmitter to the receiver 2, R2, and the second echo is diffracted wave
from the right tip of the horizontal defect. Figures 5, 6, and 7 show the corresponding waves
propagated in the test block. Using the signals detected by the receiving transducers, the
corresponding times due to diffracted echoes from the defect tips (left and right) are determined, and
then the horizontal defect size is measured using Equations 6 and 7.

Table 1
The acoustic and elastic properties of carbon steel.

ρ (g/cm3) E (GPa) ν CL (m/s) CS (m/s)

7.85 207 0.3 5985 3210

0.12 Back wall wave

Longitudinal
diffracted wave
Amplitude

0.02

-0.08

Initial pulse
-0.18
0 2 4 6 8
Time (µsec)

Figure 8
Time-of-flight diffraction signal detected by receiver 1, R1.

0.015 Longitudinal diffracted wave

Lateral longitudinal
wave
0.005
Amplitude

-0.005

-0.015

-0.025
0 2 4 6 8
Time (µsec)

Figure 9
Time-of-flight diffraction signal detected by receiver 2, R2.
24 Iranian Journal of Oil & Gas Science and Technology, Vol. 3 (2014), No. 4

The new method is carried out on eight carbon steel blocks with different size embedded horizontal
cracks. The measured crack size resulting from the FEM simulation of each block is shown in Table
2. The comparison of the simulated and the measured crack size results shows that the maximum error
which occurs at 2 mm crack size is 19.7%. As it can be seen in Table 2, the measured crack size error
is minimized within the crack size range of 8-14 mm, and it slightly increases for larger defects. This
agrees with the conventional ToFD results, which show higher measurement errors at smaller crack
sizes (Charlesworth et al., 2001). The review of the simulated results of different crack sizes shown in
Table 2 indicates that the accuracy of the proposed method for horizontal cracks, comparing to the
conventional method for vertical cracks, is within an acceptable range.

Table 2
Comparison of the measured defect size with the actual defect size.
Actual crack Measured crack Relative Incident angle
t 1 (μs) t 2 (μs)
size (mm) size (mm) error % (degree)
2 3.3759 3.6409 2.39 19.73 50
4 3.1529 3.6409 4.41 10.25 50
6 4.0003 4.8450 6.57 9.50 60
8 3.7532 4.8506 8.54 6.75 60
10 3.5196 4.8556 10.41 4.18 60
12 3.3138 4.8657 12.12 1.07 60
14 3.1315 4.8655 13.55 3.21 60
16 2.9418 4.8602 14.97 6.40 60

6. Conclusions
In this paper, a new time of flight diffraction (ToFD) method to evaluate horizontal planar defects was
presented. The finite element method was employed to simulate the ultrasonic wave behavior in the
test blocks and its interaction with the embedded planar defects such as cracks. The phased array
technology was also used to model the ultrasonic inspection system parameters. The simulation of the
new ToFD method for different crack sizes, using the ABAQUS finite element package, showed that,
compared to the conventional ToFD method, the result accuracies are within an acceptable range to
use the novel technique for measuring the horizontal planar defects. The application of the new
method for eight carbon steel blocks with different size horizontal cracks (2–16 mm) indicated that
the maximum error occurs at a 2 mm crack length. Finally, it was observed that the measured crack
size error was minimized within the range of 8-14 mm, and that it slightly increased for larger defects.

References
Al-Ataby, A. A., Automatic Detection, Sizing and Characterization of Weld Defects Using Ultrasonic
Time-of-Flight Diffraction, PhD Dissertation, Liverpool University, No. 1, p. 95-96, 2012.
American Society of Mechanical Engineers (ASME), Boiler and Pressure Vessel Code, Section V,
Non Destructive Examination. Appendix N - Time of Flight Diffraction (TOFD) Interpretation,
Article 4, p. 93-112, 2010.
Baby, S., Balasubramanian, T., Pardikar, R.J., Palaniappan, M., and Subbaratnam, R., Time-of-Flight
Diffraction (TOFD) Technique for Accurate Sizing of Surface-Breaking Cracks, Insight, June,
Vol. 45, No. 6, p.426-430, 2003.
 P. Bagheri et al. / A Novel Method for Ultrasonic Evaluation of Horizontal … 25

Baskaran, G., Balasubramaniam, K., and Lakshmana Rao, C., Shear Wave Time-of-flight Diffraction
(S-ToFD) Technique, NDT&E International, Vol. 39, p. 458-467, 2006.
Charlesworth, J. P. and Temple, J. A. G., Engineering Applications of Ultrasonic Time of Flight
Diffraction, England, RSP Press, No. 3, p. 20-28, 2001.
Connolly, G.D., Modeling of the Propagation of Ultrasound through Austenitic Steel Welds,
Ph.D.Dissertation, UK Research Centre in NDE (RCNDE) Department of Mechanical
Engineering, Imperial College London, No. 4, 2009.
Golan, S., Sizing of Cracks with Scattered Ultrasonic Waves, Proceedings of the First International
Symposium on Ultrasonic Characterization, p. 29-36, 1980.
Hellier, C. J., Handbook of Nondestructive Evaluation, McGraw Hill, 2003.
Honarvar, F., and Khorasani, S., Simulation of Time of Flight Diffraction (ToFD) Technique by
Finite Element Method, Online Workshop in www.ndt.net, No. 5, September, 2010.
Krautkramer, J., and Krautkramer, H., Ultrasonic Testing of Materials, Berlin, Springer-Verlag, P104
& 108-221,4th Edition, No. 6, 1990.
Lin, S., Fukutomi, H. and Ogata, T., Analysis of Wave Propagation for the ToFD Method by Finite
Element Method: Optimization of Test Configuration and Proposal of a New ToFD Method,
Review Progress in Quantitative Nondestructive Evalauation, Maine, USA, 2005.
Ludwig and, R. and Lord, W., Developments in the Finite Element Modeling of Ultrasonic NDT
Phenomena, Review of Progress in Quantitative Nondestructive Evaluation, 5A, American
Institute of Physics, p. 73-81, 1986.
Mak, D. K., Ultrasonic Method for Measuring Crack Location, Crack Height and Crack Angle,
Ultrasonics, p. 259-270, 1983.
Mardani, M., Sodagar, S., and Rashed, G. R., Modeling of Ultrasonic Phased Array Method Using
Finite Element Method, No. 7, ISME, Shiraz, Iran, 2012.
Olympus NDT, Advances in Phased Array Ultrasonic Technology Applications, Waltham, No. 8, p.
31-42, 2007.
Silk, M.G., Defect Detection and Sizing in Metal Using Ultrasound, International Materials Reviews,
Vol. 27, p. 28-50, 1973.
Silk, M.G., Accurate Technique for Defect Sizing in Pressurized Components, London, Institution of
Mechanical Engineers, Vol. 3, p. 155-162, 1974.
Silk, M.G., Defect sizing Using Ultrasonic Diffraction, British Journal of Nondestructive Test, Vol.
21, p. 12-15, 1976.
Silk, M.G., The Use of Diffraction-Based Time-of-Flight Measurement to Locate and Size Defects,
British Journal of Nondestructive Test, Vol. 26, p. 208-213, 1978.