GW

w
x ME BACK TO BASICS
From Materials Evaluation, Vol. 75, No. 10, pp: 1206-1213.
Copyright © 2017 The American Society for Nondestructive Testing, Inc.

TEST
An Introduction to Long-Range
Screening Using Guided Waves
by David Alleyne, Robin Jones, and Thomas Vogt

G
uided wave testing (GW) is now a well-established method for
pipeline inspection (Cawley et al., 2003). The method exploits
mechanical stress waves that are guided along the wall of the
pipe and can travel long distances, so it rapidly provides close to
100% coverage. A transducer injects a wave signal at a chosen location on
the line and then receives echoes returning from any features or discontinu-
ities. The arrival time of the echoes indicates the distance of these reflectors
from the transducer. A single test can cover a range of 50 m or more.
Early long-range guided wave inspection applications focused on the
detection of corrosion in straight aboveground lines. Such lines could be
inspected by conventional means, but the attraction of using guided waves
was much-improved inspection speed and the assurance of 100% coverage.
As experience was gained, the method was developed to be practical for
more complex applications, including road crossings, buried lines, subsea
lines, coated lines, and complex geometries (Sabet-Sharghi et al., 2010; Rose
et al., 1994). In some of these cases there is no alternative inspection
method, so the use of guided waves avoids expensive invasive investigations
such as excavation.

. . . the use of guided waves

avoids expensive invasive
investigations.
OCTOBER 2017 • MATERIALS EVALUATION 1207
ME BACK TO BASICS w
x guided wave testing

Guided Wave Testing as a Screening Tool In pipes, the geometry of the waveguide is deter-
It is vital to recognize that the conventional role of mined by the pipes’ outer diameter and the wall
guided wave testing today is screening. The echoes thickness. Figure 1 shows group velocity dispersion
returning from discontinuities and received by the curves for a pipe with nominal 152 mm (6 in.) outer
instrumentation indicate the severity of discontinu- diameter and schedule 40 steel pipe (Pavlakovic et al.,
ities, but they do not give detailed quantitative infor- 1997). The group velocity is the speed of a wave
mation about the morphology of the discontinuities. packet—for example, a toneburst or pulse—and is the
Therefore, the inspection procedure is to use the important measure of speed in GW.
guided wave tests to indicate any locations of concern, For all practical purposes, the successful exploita-
and then to follow up with conventional detailed tion of guided waves for NDT requires
inspection at these locations. l pure mode transmission and reception by selecting
This is a powerful combination: for a given specific guided wave modes,
reporting threshold, guided waves can provide full l the modes generated to be nondispersive,
coverage of the pipe within the test range, identifying l the full cross section of the pipe wall to be insoni-
all locations where a subsequent, thorough, detailed fied, and
inspection is required. In other words, GW provides a l signals coming from each direction to be
high probability of detection (Vogt and Evans, 2009), distinguished.
and local follow-up inspection provides a high proba- Among the manufacturers of guided wave
bility of correctly assessing the severity. equipment, the piezoelectric transduction technique is
most commonly used (Alleyne et al., 2001; Mudge,
Basics of Guided Wave Testing 2001); however, there are alternatives, such as
The shape of the pipe is used to guide the wave’s magnetostrictive transduction (Kwun et al., 2003) and
energy, reducing dissipation and maximizing inspec- electromagnetic acoustic transduction (EMAT). The
tion range. piezoelectric technique is usually favored due to the
strength and quality of signal that can be achieved.
Guided Waves and Transduction Furthermore, due to the relatively low frequency of
Guided waves are waves whose propagation is guided operation, and through appropriate transducer design,
by a structural shape or boundary. The structure these piezoelectric transducers can be simply dry-
guides the direction of the wave, while the material coupled to the pipe. The coupling achieved by this
properties and shape determine its speed, stresses, technique is consistent and gives repeatably good
and displacements. The fact that the guided waves are results without the need for any coupling agent and
contained within the waveguide structure means that with minimal surface preparation (just the removal of
they can travel long distances without dissipating; this loose paint flakes, paint beads, or loose corrosion
is a major attraction for nondestructive testing (NDT) scale).
applications.
Detection of Discontinuities
Guided wave testing typically uses a pulse-echo
configuration (Lowe et al., 1998), which is displayed in
Figure 2. The wave mode generated by the transducer
6 ring propagates along the pipe and is partially
reflected from any location where there is a change in
Group velocity (m/ms)

5 L (0,1) L (0,2)
the cross section of the pipe. Such locations include
F (1,3)
benign features such as butt welds, but also disconti-
4
nuities such as patches of corrosion. The reflected
T (0,1) signals return to the transducer ring, where they are
3 F (1,2)
F (n,m) recorded.
2 A pipeline is axisymmetric in that it has symmetry
about the longitudinal axis of the pipe. Pipe features
1 that are uniform about the circumference (such as
welds) preserve this symmetry and are said to be
00 10 20 30 40 50 60 70 axisymmetric. Pipe features that disrupt this symmetry
Frequency (kHz) (such as discontinuities, drains, and vents) are said to
be nonaxisymmetric. Figure 2 shows that the reflected
Figure 1. Group velocity dispersion curves for 6 in. schedule 40 pipe. signal from an axisymmetric feature is also axisym-

1208 MATERIALS EVALUATION • OCTOBER 2017

 Weld

Incident T or L Reflected T or L Transmitted T or L Pipe

(a)

Transducer ring
Discontinuity

Reflected T or L Transmitted T or L
Incident T or L
Pipe
Reflected F Transmitted F
(b)

Figure 2. Reflection modes: (a) reflection and transmission of a symmetric mode from an axisymmetric feature; (b)
reflection, transmission, and mode conversion from a nonaxisymmetric feature (where T = torsional waves, L =
longitudinal waves, and F = flexural waves).

metric (meaning, consists of the incident mode only). • intrinsic material attenuation
On the other hand, the reflected signal from a nonax- • loss of sound energy from the pipe into the material
isymmetric feature contains both axisymmetric and with which it is in contact (for example, sand or clay
nonaxisymmetric components. This behavior can be if the pipe is buried)
exploited to enable the detection of discontinuities at • scattering from anything that creates local changes
welds. An axisymmetric butt weld should not reflect to the pipe wall (such as changes in geometry,
the flexural wave; however, a discontinuity at the weld small components attached to the pipe, roughness,
would introduce nonaxisymmetry, causing reflection of or others)
the flexural wave. For this reason, both the torsional
and the flexural waves are measured, as they provide Choice of Mode
complementary information. The initial work on GW in the mid-1990s used longitu-
The reflectivity of guided waves is governed by dinal waves. However, it was quickly recognized that
very different rules than for bulk waves; with guided the torsional mode is much superior to the longitu-
waves, it is possible to find discontinuities whose dinal mode in the vast majority of applications. This is
depths are much smaller than a wavelength. The because of several reasons:
strength of the reflection of guided waves from a l The torsional mode requires only two transducer
feature is directly proportional to the change in cross- rows (the longitudinal mode requires four rows).
sectional area of the pipe from one axial position to l The torsional mode exists over the entire frequency
the next. The axial length of a discontinuity also has range and is completely nondispersive.
an effect on its reflectivity, a complex behavior due to l The achieved torsional mode signal-to-noise ratio
an interference effect between the reflections from the (referred to in a later section) is generally higher
start and end of the discontinuity. The concept of than for the longitudinal mode, so the sensitivity is
frequency sweeping is used by some equipment higher.
manufacturers to prevent the possible underestima- l The torsional mode is not affected by nonviscous
tion of discontinuities. By collecting the frequencies of liquids inside the pipe.
the entire applicable frequency range, the GW l There are two longitudinal modes propagating in
inspector can change the displayed frequency of the the test range, which leads to false indications with
result quasi-continuously during data analysis. The this mode; the torsional mode does not have this
discontinuity is then classified when the amplitude is problem.
maximal. Further information on optimal mode selection for
guided wave testing is available (Alleyne et al., 2009).
Attenuation
One of the key attractions of guided waves is that they Performing a Guided Wave Test
can travel long distances without a significant loss of Guided wave equipment can display the results of the
energy. However, in practice, attenuation does occur test as an A-scan, showing distance along the pipe,
due to the following: and as a C-scan, providing a view of the unrolled pipe.

OCTOBER 2017 • MATERIALS EVALUATION 1209

ME BACK TO BASICS w
x guided wave testing

The setup of the GW equipment and

collection of the data can be achieved
within a few minutes.
Information Required Prior to Testing left-hand side. Typically, these instruments are battery
It is important that the service provider is given infor- powered, portable, and fully ruggedized for deploy-
mation about the testing conditions, such as pipe ment in the field. The instrument is connected to the
size, access, layout, environment, support type and transducer ring, which is mounted on the pipe by two
spacing, insulation, coatings, linings, deposits, flexible cables. The test is controlled by special
temperature, general pipe condition, target discontinu- software on a ruggedized and portable computer,
ities, and required coverage. Any information on pipe located on the pipe beyond the ring, that is connected
history—for example, discontinuities found in the to the instrument by a USB cable. The setup of the GW
past—and isometric drawings are beneficial. equipment and collection of the data can be achieved
Some preparation of the test site and the pipe within a few minutes.
may be necessary. This could include ensuring safe
access to the test position—for example, by excavation Data Quality
or scaffolding—or removal of insulation or coatings to The quality of the data acquired during the guided
allow installation of the transducer ring on the pipe. wave test is strongly affected by equipment faults and
The GW inspector should be consulted prior to poor equipment setup. Therefore, in order to ensure
conducting this preparatory work in order to advise on that the equipment is functioning correctly and has
optimum test positions. been properly set up, GW equipment typically has a
range of tools available:
Performing the Test l checks of transducer capacitance
Figure 3 shows the typical equipment setup for a l checks of coupling strength of the transducer
guided wave test; the test instrument is shown on the segments
l transducer cable checks
l transducer ring wiring checks
l checks of instrument hardware and temperature
l signal-to-noise ratio checks
The results of these checks are recorded together
with the data. Any serious faults will result in major
warning messages given by the software and will
block viewing of data and their analysis for those
personnel with Level I certification levels.

Result Presentation
Figure 4 shows a typical result from a pipe with both
axisymmetric (for example, welds) and nonaxisym-
metric features (for example, supports). Guided wave
testing results are typically displayed with two
different types of plot: the upper plot is the C-scan
type plot, otherwise referred to as the unrolled pipe
display; and the lower plot is the A-scan result trace.

A-scan Result Trace

The conventional way to display data is to show the
Figure 3. Example setup of guided wave testing equipment, showing an inflatable received amplitudes of both the torsional and the
ring installed on a pipe. flexural guided wave as a function of distance from

1210 MATERIALS EVALUATION • OCTOBER 2017

 Normalized amplitude range
(Weld DAC is 0 dB)
12

3
Clock

12
–20.0 –10.0 0.0 –24 dB
(a) Distance (m)
Pipe
1.0 schematic
W1

W2

B1
S2

S3
0.8 Weld DAC Near zone
Amp (linear)

0.6 End of
test range
Dead zone
0.4
Call DAC
0.2

0.0
–20.0 –10.0 0.0
(b) Distance (m)

Figure 4. Example results: (a) the C-scan type unrolled pipe display; (b) the A-scan type result trace. The welds, W1 and W2, are axisymmetric
features, extending around the entire circumference, whereas the support, S2, is nonaxisymmetric and localized around the six o’clock position.
B1 is a 3D bend located close to the transducer ring in the positive direction. The DAC refers to the distance-amplitude correction.

the transducer ring, as shown in the lower plot in provides a more intuitive means of displaying the
Figure 4. The amplitudes of the reflections are shown results but also allows discontinuities to be more
as a function of distance, with the axisymmetric accurately located, particularly in the circumferential
torsional mode being represented by the black curve direction. For example, a support at the bottom of the
and the nonaxisymmetric flexural mode represented pipe can be distinguished from a discontinuity at the
by the red curve. On either side of the transducer ring same axial position but at the top of the pipe, which is
is a small area called the dead zone where analysis is not possible using the information from the A-scan
not possible, indicated on the trace by the vertical type result trace alone.
green band. Surrounding the dead zone is an area
called the near zone, where amplitudes of the Calibration
received guided waves are lower than expected, The guided wave signal amplitude decreases with
indicated on the trace by the vertical gray bands. distance along the pipe. Calibration allows pipe
Special collection protocols can be used to reduce features to be accurately sized to account for this
these lengths, for example, when testing short reduction.
sections of pipe.
Amplitude Calibration
Unrolled Pipe Display As explained in an earlier section, several mecha-
Some equipment manufacturers have developed nisms cause the guided wave signal amplitude to
focused guided wave systems, which have the capa- decrease with distance traveled. The purpose of the
bility to supplement the A-scan result trace with an distance-amplitude correction (DAC) curves is to
unrolled pipe display. This C-scan type plot is a two- correct for this decrease in amplitude.
dimensional representation of the amplitudes of the Referring to Figure 4, two different DAC levels are
reflections as a function of the circumferential position shown on the graph: the weld DAC (in black), and the
and distance from the transducer ring. This not only call DAC (in blue). It can be seen that these curves

OCTOBER 2017 • MATERIALS EVALUATION 1211

ME BACK TO BASICS w
x guided wave testing

incorporate a combination of exponential decay and Follow-Up Inspection

instantaneous drops at features. The welds are used The extent of secondary follow-up inspection of indica-
as reference features to correctly position these DAC tions is determined by the reporting requirements.
curves, and thereby calibrate the amplitude. Technically, this follow-up is not actually part of the
There are advanced amplitude calibration tech- GW inspection—note, for example, that GW and the
niques that some manufacturers offer, such as using follow-up inspection are governed by separate inspec-
the absolute calibration technique, calibrating from tion procedures, the latter being either that of the
measured weld dimensions, or calibrating from a self- contractor or end client. Nevertheless, GW as a
measurement of the transducer ring output. screening method is of course most effectively used in
conjunction with follow-up inspection.
Distance Calibration
The software assumes a certain value for the torsional Application Examples
wave speed in steel (equal to the shear bulk wave Guided wave testing is applicable to a wide range of
speed), which is accurate enough that it is not usually different scenarios, offering efficiency and coverage
necessary to calibrate the reported distance. However, improvements over alternative NDT methods.
if the pipe is at an elevated temperature, the torsional
wave speed will decrease, leading to an error in the Rapid Long-Range Screening
calculated positions of features on the pipe. To correct One of the main strengths of GWT is the rate of
for this, the distance to a known feature can be inspection coverage of pipelines that can be achieved.
measured using a tape measure and the distance This is due to the ability to screen many meters of
scale adjusted so the calculated and actual positions pipe from a single test point. During the inspection of
match. a 508 mm (20 in.) gas pipe, from a single installation
of the transducer ring on the pipe, it was possible to
Signal-to-Noise Ratio inspect 230 m of pipe: 100 m in the negative
Signal-to-noise ratio (SNR), which is the ratio of signal direction and 130 m in the positive direction. The
amplitude relative to the background noise level, is inspection results indicated that there was no
used to quantify the quality of a nondestructive corrosion present in the test range, but they did
testing signal. For a given reporting threshold (call highlight misalignment in some girth welds.
level), the higher the SNR, the higher the probability of
detection and the lower the occurrence of false indica- Corrosion Under Insulation
tions. For a typical guided wave application that is The widespread problem of corrosion under insulation
concerned with finding corrosion of about 6% cross- was one of the driving applications for the initial
sectional loss or greater, a minimum SNR of approxi- development of GW. A short section of insulation
mately 32 dB should be achieved. (approximately 1 m) is required to be removed from
In order to achieve the right balance between the pipe in order to install the transducer ring. From
detection sensitivity and minimizing false calls, a this test position, many meters of pipe in either
detection threshold is defined at 6 dB above the back- direction can be assessed for the presence of
ground noise (meaning, at twice the amplitude of the corrosion under insulation. This represents a signifi-
general noise level). This gives a lower limit, below cant improvement in productivity and integrity
which indications should not be called. assurance over conventional techniques of inspecting
for this type of corrosion (for example, radiographic
Interpretation of the Results techniques), which can miss degraded areas because
The interpretation of the processed data is the most they do not inspect the full cross section of the pipe.
difficult part of the guided wave test. As with any
nondestructive method, successful interpretation High Temperature Pipe
depends heavily on proper training and the user’s Some equipment manufacturers produce equipment
amount of experience in using the guided wave that has been specifically designed for the inspection
equipment. of high temperature pipework, up to temperatures of
Once all weld, flange, and support locations are 350 °C. In one example, a high temperature inflatable
identified, all other indications above the “call DAC” ring was used in the inspection of a 203 mm (8 in.)
level must be investigated. All indications above the diesel pipe that was at a temperature of 240 °C. An
detection threshold should also be investigated, even area of corrosion was identified and followed up with
if they are below the “call DAC.” conventional inspection techniques.

1212 MATERIALS EVALUATION • OCTOBER 2017

Summary
Guided wave testing has become a well-established
method for inspecting pipelines. It is important to
remember that it is a screening technique that should
be used to detect, locate, and classify discontinuities.
Follow-up inspection with a quantitative nondestruc-
tive testing technique should be performed to
precisely measure discontinuity dimensions. w x

AUTHORS
David Alleyne: Ph.D., Guided Ultrasonics Ltd., Wavemaker
House, The Ham, Brentford, TW8 8HQ United Kingdom;
email [email protected]
Robin Jones: Eng.D., Guided Ultrasonics Ltd., Wavemaker
House, The Ham, Brentford, TW8 8HQ United Kingdom;
email [email protected]
Thomas Vogt: Ph.D., Guided Ultrasonics Ltd., Wavemaker
House, The Ham, Brentford, TW8 8HQ United Kingdom;
email [email protected]

ACKNOWLEDGMENTS
Figure data and photos provided courtesy of Guided
Ultrasonics Ltd.

REFERENCES
Alleyne, D., B. Pavlakovic, M. Lowe, and P. Cawley, “Rapid
Long Range Inspection of Chemical Plant Pipework Using
Guided Waves,” Insight, Vol. 43, No. 2, 2001, pp. 93–96.
Alleyne, D.N., T. Vogt, and P. Cawley, “The Choice of
Torsional or Longitudinal Excitation in Guided Wave Pipe
Inspection,” Insight: Non-Destructive Testing & Condition
Monitoring, Vol. 51, No. 7, 2009, pp. 373–377.
Cawley, P., M.J.S. Lowe, D.H. Alleyne, B. Pavlakovic, and P.
Wilcox, “Practical Long Range Guided Wave Testing: Appli-
cations to Pipes and Rails,” Materials Evaluation, Vol. 61,
No. 1, 2003, pp. 66–74.
Kwun, Hegeon, Sang Y. Kim, and Glenn M. Light, “The
Magnetostrictive Sensor Technology for Long Range Guided
Wave Testing and Monitoring of Structures,” Materials Eval-
uation, Vol. 61, No. 1, 2003, pp. 80–84.
Lowe, M.J.S., D.N. Alleyne, and P. Cawley, “Defect Detection
in Pipes Using Guided Waves,” Ultrasonics, Vol. 36, Nos. 1–
5, 1998, pp. 147–154.
Mudge, P.J., “Field Application of the Teletest Long-Range
Ultrasonic Testing Technique,” Insight, Vol. 43, No. 2, 2001,
pp. 74–77.
Pavlakovic, Brian, Mike Lowe, David Alleyne, and Peter
Cawley, “Disperse: A General Purpose Program for Creating
Dispersion Curves,” Review of Progress in Quantitative
Nondestructive Evaluation, Vol. 16, Plenum Press, New York,
New York, 1997, pp. 185–192.
Rose, J.L., J.J. Ditri, A. Pilarski, K. Rajana, and F. Carr, “A
Guided Wave Inspection Technique for Nuclear Steam
Generator Tubing,” NDT & E International, Vol. 27, No. 6,
1994, pp. 307–310.
Sabet-Sharghi, Reyaz, Kianoush Samani, Erik Burr, Travis
Stelly, Randy Denardi, Jim McNew, and Mark Susich, “Ultra-
sonic Guided Wave Testing of Concrete Anchor Penetra-
tions,” Materials Evaluation, Vol. 68, No. 10, 2010, pp.
1085–1090.
Vogt, T., and M. Evans, “Reliability of Guided Wave Testing,”
4th European-American Workshop on Reliability of NDE,
Berlin, 2009.

OCTOBER 2017 • MATERIALS EVALUATION 1213