DETECTION OF SCALE INSIDE OF WATER SUPPLY PIPES
USING GUIDED WAVES
Sung-Jin Song1, Young H. Kim1, Dong-Hun Lee1, Joon Soo Park1,
Hyun-Dong Lee2 and Chul-Ho Bae2
School of Mechanical Engineering, Sungkyunkwan University
300 Chonchon-dong, Jangan-gu, Suwon, Kyonggi-do, 440-746, Korea
2
Korea Institute of Construction Technology, Goyang, Korea
Daehwa-dong, Ilsan-gu, Goyang, Kyonggi-do, Korea
ABSTRACT.
Since scale in water supplying pipes is one of the major sources of water contamination
and user's complaint, detection of scale is very important for the proper maintenance of water piping. In
the present study, the potential of guided waves was explored for the detection of scale in water
supplying pipes. Using variable angle wedges, several modes of guided waves were generated and
identified. In the experiments, it were observed that the amplitude of F(M,2) (M= 1,2,3,4) modes
decreased significantly as the increase of the amount of scale. The result of the present study
recommended that the F(M,2) modes are optimal to detect scale in water supplying pipes.
INTRODUCTION
Water piping, which is often called as the "lifeline", should be maintained soundly to
provide clean water to end-users. Unfortunately, however, water piping is degraded as the
increase of its operation time. The degradation process of water supply piping is as follows; 1)
initiation of corrosion by chemical reaction, 2) appearance of scale by the accumulation of
corrosion byproduct inside of the pipe wall, 3) growth of dents from corrosion pits, and finally
4) ending with through-wall defects which cause leakage of water.
CP657, Review of Quantitative Nondestructive Evaluation Vol. 22, ed. by D. O. Thompson and D. E. Chimenti
© 2003 American Institute of Physics 0-7354-0117-9/03/$20.00
229
Up to now, various efforts have been carried out to develop suitable nondestructive
testing methods for detection of general wall loss associated with pitting and corrosion in
various pipings [1-4]. Unfortunately, however, there have been not many of studies on detection
of scale in water pipes. Nevertheless, since scale is one of the major sources of water
contamination and user's complaints on water supplying pipes, the detection of scale should be
performed for the proper maintenance of water pipe. Especially, it should be done nondestructively from the outside of piping. Furthermore, it is strongly desired to have an efficient
method that can rapidly inspect large area of water piping.
It is well recognized that the guided waves are very suitable for the inspection of long
range pipelines[3,4]. However, the most of the works performed for the long range pipelines
using guided waves have also been concentrated on the detection of wall loss.
Thus, in this study, we have experimentally explored the potential of guided waves for
the detection of scale in water piping from the outside of pipes. The water pipe which is made
of steel, had the outside diameter of 22mm and the wall thickness of 3mm. Non-axisymmetric
guided waves were excited on the outside surface of the water pipe by using a 0.5 MHz
transducer with a variable angle shoe. Phase velocity tuning was used to generate the possible
guided wave modes, and the optimum modes were selected experimentally. The initial
experimental result is also presented here.
GUIDED WAVE MODES IN WATER PIPING
Guided wave modes and their dispersive characteristics can be obtained by solving
wave equation with proper boundary conditions [5-7]. In piping, there are infinite number of
modes that are named longitudinal modes (L(0,n)), torsional modes (T(0,n)), and flexural
modes (F(M,n)), where M is the circumferential order and n is the mode number. In most cases,
longitudinal modes are used for the inspection because they are axisymmetric modes and
efficiently excited and received with angled transducers [4]. Torsional modes are also
axisymmetric modes but are not used very often with the angled transducers because they are
hard to be excited. Flexural modes are non-axisymmetric modes and often propagate together
with longitudinal modes.
In Figure 1, the phase and group velocity dispersion curves for longitudinal and
flexural modes in the water supplying pipes under investigation are shown. The horizontal axis
represents the frequency (f) times thickness (d) of the piping (fd). For the investigation of the
dispersion characteristics, we have implemented our own program that can calculate the phase
and group velocities in the elastic hollow cylinder in MATLAB [8]. It is noticed that flexural
modes have similar patterns of dispersion with the longitudinal modes and congregated
according to the mode number.
230
SPECIMENS AND EXPERIMENTAL SET UP
In the present study, we have chosen four kinds of water pipes, as shown in Figure 2:
a) the new pipe with the amount of scale of 0% (here after "the new pipe"), b) the used pipe
with the amount of scale of 3% (here after "the used pipe with scale of 3%"), c) the used pipe
with the amount of scale of 23% (here after "the used pipe with scale of 23%"), and d) the used
pipe with the amount of scale of 42% (here after "the used pipe with scale of 42%"). The
amount of scale was measured in terms of the volume fraction.
A tone burst system that can control the frequency and the amplitude of the excited
signal precisely was used to excite the specified modes of ultrasonic guided waves in the water
pipe. Ultrasonic pulser/receivers commonly used in practice were not enough to excite
ultrasonic guided waves that can propagate far away. Therefore, we have used a high power
ultrasonic pulser/receiver (RAM-10000 made by Ritec). Also, For the excitation of guided
waves, 0.5 MHz transducers (made by Panametrics) were mounted on the variable angle wedge
that were fabricated in order to control the incident angles of excited beam.
F(1,3),R2,3),F(3,3XF(4,3)
K1,4)~F(4,4)
RM,6)~FU0
K1,7)~FU7)
3.9
4.3
4.6
5.0
fd [MHz mm]
(a)
UG,2>
F(1,3),F(2,3),R3,3),F(.4,3)
b—£—L. I I
"" ''R1,2),F(2,2),F(3,2),R4,2)
R1,4),F,(2,4),R3,4),R4,4)
R1,4),F(2,4),F(3,4),F(4,4)
F(1,4),F(2,4),R3,4),R4,4)
R1,7)~FW,7)
fd [MHz mm]
FIGURE 1. Dispersion curves of (a) phase velocity, and (b) group velocity for the steel pipe with the thickness of
3 mm, outer diameter of 22 mm and inner diameter of 16 mm. The longitudinal and shear wave velocities are 5.92
km/s and 3.2 km/s, respectively.
231
(a)
(a)
(b)
(b)
(c)
(c)
(d)
(d)
FIGURE 2. The water pipes under investigation, (a) the new pipe, (b) the used pipe with scale of 3%, (c) the used
FIGURE 2. The water pipes under investigation, (a) the new pipe, (b) the used pipe with scale of 3%, (c) the used
pipe with scale of 23%, and (d) the used pipe with scale of 43%.
pipe with scale of 23%, and (d) the used pipe with scale of 43%.
For the fd value of 1.5 MHz mm (which is the nominal value representing the test
For the fd value of 1.5 MHz mm (which is the nominal value representing the test
condition of the present study, since the center frequency of the transducer is 0.5 MHz, and the
condition of the present study, since the center frequency of the transducer is 0.5 MHz, and the
thickness of the pipe is 3.0 mm) in the dispersion curves as shown in Figure 1, it was able to
thickness of the pipe is 3.0 mm) in the dispersion curves as shown in Figure 1, it was able to
choose 10 different modes as shown in Table 1. For each mode, the phase velocity was read
choose 10 different modes as shown in Table 1. For each mode, the phase velocity was read
from the dispersion curve, and the corresponding wedge angle was determined by use of the
from the dispersion curve, and the corresponding wedge angle was determined by use of the
Snell’s law with the longitudinal velocity of the wedge of 2.72 km/s.
SnelPs law with the longitudinal velocity of the wedge of 2.72 km/s.
As shown in Figure 3, a pitch-catch experimental setup was used for the generation
As shown in Figure 3, a pitch-catch experimental setup was used for the generation
of non-axismmetric guided waves. The distance between the sending transducer and the
of non-axismmetric guided waves. The distance between the sending transducer and the
receiving transducer is 0.5 m. And, in this experimental study, all of the parameters for the
receiving transducer is 0.5 m. And, in this experimental study, all of the parameters for the
generation of guided waves (including the output power and the center frequency of the tone
generation of guided waves (including the output power and the center frequency of the tone
burst signal) were fixed, but the incident angle was changed.
burst signal) were fixed, but the incident angle was changed.
TABLE 1. Possible wave modes that can be generated in the water pipe together with their phase velocities and
TABLE 1. Possible wave modes that can be generated in the water pipe together with their phase velocities and
angles of incidence.
angles of incidence.
Modes
Phase velocity (km/s)
Modes
F(4,3)
F(4,3)
F(3,3)
Phase velocity
7.233 (km/s)
7.233
6.143
6.143
5.609
5.609
5.349
5.349
5.270
5.270
3.606
3.606
3.428
3.428
3.314
3.314
3.251
3.251
2.77
2.77
F(3,3)
F(2,3)
F(2,3)
F(1,3)
F(l,3)
L(0,2)
L(0,2)
F(4,2)
F(4,2)
F(3,2)
F(3,2)
F(2,2)
F(2,2)
F(1,2)
F(l,2)
F(4,1)
F(4,l)
232
Incidence angle ( ° )
Incidence
angle ( ° )
22.17
22.17
26.39
26.39
29.13
29.13
30.69
30.69
31.20
31.20
49.21
49.21
52.79
52.79
55.46
55.46
57.11
57.11
80.25
80.25
Sending TR
TR ^__________________^Receiving
Receiving TR
TR
Sending
pipe
0.5 m
0.5m
FIGURE 3.
3. Transducer
Transducer set
set up
up for
for the
the excitation
excitationand
andreception
receptionof
ofguided
guidedwaves.
waves.
FIGURE
PROPAGATION CHARACTERISTICS
CHARACTERISTICS OF
OF GUIDED
GUIDEDWAVES
WAVES
PROPAGATION
Figure 4 shows a set of the through transmission
transmission signals
signals for
for the
the F(2,2)
F(2,2) mode
mode captured
captured
by the receiving transducer
transducer from
from the
the four
four pipe
pipe specimens
specimens with
with various
various amounts
amounts of
of scale.
scale. As
As
shown in Figure 4, the
the amplitude
amplitude of
of the
the F(2,2)
F(2,2) mode
mode signal
signal decreased
decreased significantly
significantly as
as the
the
increase of the amount of the
the scale inside
inside of the
the pipe.
pipe. And,
And, eventually
eventually this
this mode
mode disappeared
disappeared
due to the presence of a large
large of scale. The
The signals
signals for
for the
the F(l,2)
F(1,2) and
and the
the F(3,2)
F(3,2) modes
modes were
were
also captured by the
the receiving
receiving transducer
transducer under
under the
the same
same experimental
experimental set
set up,
up, even
even though
though the
the
results are not shown here. What we have found
found was
was that
that the
the amplitude
amplitude of
of both
both of
of the
the F(l,2)
F(1,2)
and the F(3,2) modes also decreased
decreased significantly
significantly as
as the
the increase
increase of
of the
the amount
amount of
of scale,
scale,
similar to the case of the F(2,2) mode.
Figure 5 shows a set of the similar through
through transmission
transmission signals
signals for
for the
the F(2,3)
F(2,3) mode.
mode.
On the contrary to
to the
the case
case of
of the
the F(2,2)
F(2,2) mode,
mode, this
this particular
particular mode
mode successfully
successfully propagated
propagated
through the water pipes even
even with
with the
the presence
presence of
of considerable
considerable amount
amount of
of scale.
scale. The
The similar
similar
trend was also observed for the
F(l,3)
and
F(3,3)
modes,
even
though
the
captured
waveforms
the F(1,3) and F(3,3) modes, even though the captured waveforms
are not shown here.
Figure 6 shows the
the summary
summary of
of experimental
experimental observation
observation for
for the
the case
case of
ofthe
theF(2,2)
F(2,2)and
and
the F(2,3) modes. The x-axis represents the
the volume
volume fraction
fraction of
of scale,
scale, and
and y-axis
y-axis denotes
denotes the
the
normalized amplitude (the amplitude of the
received
signal
form
the
pipe
under
test
normalized
the received signal form the pipe under test normalized
to that of the signal captured from
from the new pipe).
As shown in Figure 6, the
the F(2,2), which
which has
has mode
mode number
number nn equal
equal to
to 2,
2, decreased
decreased very
very
rapidly as the increase of the volume
volume fraction
fraction of
of scale. As
As for
for the
the F(2,3)
F(2,3) mode,
mode, its
its amplitude
amplitude
also decreased as the increase of the
the volume
volume fraction
fraction of
of scale.
scale. However,
However, the
the extent
extent of
of decrease
decrease
is much less than that of the F(2,2) mode.
mode. The
The F(2,3)
F(2,3) mode
mode still
still propagated
propagated through
through the
the water
water
pipes in spite of the presence of scale.
In fact,
fact, what this experimental observation
observation recommend
recommend is
is that
that the
the modes
modes with
with the
the
mode number 2 (such as F(l,2),
F(2,2),
and
F(3,2))
are
very
suitable
for
the
detection
of
scale,
F(1,2),
F(3,2))
very suitable for the detection of scale,
accumulated inside of the water supplying pipes.
233
(a)
(b)
(c)
(d)
(a)
(b)
(d)
(a)
(c)
(d)
(a)
(b)
(c)
(d)
FIGURE 4. The signals of the F(2,2) mode captured from (a) the new pipe, (b) the used pipe with scale of 3%,
FIGURE
4.
The
signals
of
the
F(2,2)
mode
captured
from
(a)
the
new
pipe,
(b)
the
used
pipe
with
scale
of3%,
3%,
FIGURE 4. The signals of the F(2,2) mode captured from (a) the new pipe, (b) the
the used
usedpipe
pipewith
withscale
scaleof
of
3%,
(c) the used pipe with scale of 23%, and (d) the used pipe with scale of 43%.
(c)
used pipe
pipe with
(c) the
the used
used pipe
pipe with
with scale
scale of 23%, and (d) the used
with scale
scale of
of 43%.
43%.
(a)
(b)
(c)
(d)
(a)
(b)
(c)
(d)
(a)
(c)
(d)
(a)
(b)
(d)
FIGURE
5.
The
signals
of
the
F(2,3)
mode
captured
from
(a)
the
new
pipe,
(b)
the
used
pipe
with
scaleof
of3%,
3%,
the
F(2,3)
mode
captured
from
(a)
the
new
pipe,
(b)
the
used
pipe
with
FIGURE
5.
The
signals
of
captured
from
(a)
the
new
pipe,
(b)
the
used
pipe
with
scale
FIGURE 5. The signals of
the used pipe withscale
scale of
of 3%,
3%,
(c)
the used
used pipe with
with scale
scale of
of43%.
43%.
of 23%,
23%, and
and (d)
(d) the
(c)the
the used
used pipe
pipe with
with scale
scale of
of
43%.
(c)
the
used
pipe
with
scale
of
23%,
and
(d)
the used pipe
pipe with scale
Normalizedamplitude
amplitude
Normalized
Normalized
amplitude
1
11
•8
3 0.8
0.8
0.8
0.8
F(2,3)
F(2,3)
0.6
0.6
0.6
0.4
0.4
0.4
F(2,2)
F(2,2)
F(2,2)
0.2
|0,
0.2
0.2
00
0
000
0
10
10
10
10
20
20
20
30
30
30
40
40
40
Volume
of
(%)
Volume fraction
fraction of
scale
Volume
fraction
of scale
scale (%)
(%)
50
50
50
FIGURE
amplitude
of
the
received signal
signalaccording
accordingto
tothe
theamount
amountof
ofthe
thescale.
scale.
FIGURE6.
6. Variation
Variation of
of the
the normalized
normalized
normalized amplitude
amplitude of
of the
the
received
FIGURE
6.
Variation
of
the
normalized
amplitude
of
the received
received signal
signal according
according to
to the
the amount
amount of
of the
the scale.
scale.
234
CONCLUSION
Ultrasonic guided waves were explored for the detection of scale inside of water
supplying pipes. Firstly, we have theoretically calculated dispersion curves for identifying the
possible modes that can propage through the water piping under investigation. Then, the nonaxisymmetric guided wave modes were excited to reveal the possibility of the detection of
scale. From this analysis, it has been found that the guided wave modes with the mode number
2 were optimal for the detection of scale, since their amplitude decreased significantly due to
the increase of the amount of scale. Currently, the further theoretical investigation for this
phenomenon is undertaken. And the more detailed discussion will be given in a separate paper
shortly.
REFERENCES
1. Brook. M, Ngoc, T. D. K., and Eder, J., "Ultrasonic inspection of steam generator tubing by
cylindrical guided waves," Review of Progress in Quantitative Nondestructive Evaluation,
Vol. 9, pp. 243-249, (1990)
2. H. J. Shin, Yi, R., and Rose, J.L., "Defect Detection and Characterization in Power Plant
Tubing Using Ultrasonic Guided Waves," 14th WCNDT, New Delhi, India, December 8-13,
(1996)
3. Alleyne D. N., Lowe M., Cawely P., "The inspection chemical plant pipe work using lamb
waves: defect sensitivity and field experience", Review of Progress in Quantitative
Nondestructive Evaluation, D. O Thompson and D. E. Chimenti eds, Plenum Press, New
York, Vol 15, P. 1589, (1996)
4. H. J. Shin, S-J. Song, J. S. Park, "Long range inspection of polyethylene coated steel pipes
by using guided waves", Review of Progress in Quantitative Nondestructive Evaluation, D.
O Thompson and D. E. Chimenti eds, Plenum Press, New York, Vol 19, P. 1194, (2001)
5. D. C. Gazis, "Three-Dimensional investigation of the propagation of waves in hollow
circular cylinders. I. Analytical foundation," J. Acoust. Soc. Am., Vol. 31, No. 5, pp. 568573, May, (1959)
6. K. F. Graff, Wave Motion in Elastic Solids, Dover Publication INC., New York, (1975)
7. J. L. Rose, Ultrasonic Waves in Solid Media, Cambridge University Press, (1999)
8. D. H. Lee, "A study on scale detection in water piping using guided waves", MS. Thesis,
Sungkyunkwan University, (2002)
235