Download - AET-2006 Reading 3A
-
8/15/2019 AET-2006 Reading 3A
1/469
on Zhang/ Charlie Chong
My Reading on Acoustic Emission Testing2016-03 For my ASNT Level III Examination on
coming 2016 August.27 th June 2016
P r e
-
8/15/2019 AET-2006 Reading 3A
2/469
on Zhang/ Charlie Chong
Acoustic Emission Testing
-
8/15/2019 AET-2006 Reading 3A
3/469
harlie Chong/ Fion Zhang
Acoustic Emission Testing
-
8/15/2019 AET-2006 Reading 3A
4/469
harlie Chong/ Fion Zhang
Acoustic Emission Testing
-
8/15/2019 AET-2006 Reading 3A
5/469
harlie Chong/ Fion Zhang
Acoustic Emission Testing
-
8/15/2019 AET-2006 Reading 3A
6/469
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
7/469
Fion Zhang
22nd June 20
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
8/469
harlie Chong/ Fion Zhang
SME- Subject Matthttp://cn.bing.com/videos/search
https://www.youtube.com/channe
-
8/15/2019 AET-2006 Reading 3A
9/469
http://www.yumpu.com/zh/browse/user/charliechong
http://issuu.com/charlieccchong
http://independent.academia.edu/CharlieChong1
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
10/469
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
11/469
harlie Chong/ Fion Zhang
The Magical Book of Tank Inspection ICP
-
8/15/2019 AET-2006 Reading 3A
12/469
ASNT Certification Guide
NDT Level III / PdM Level IIIAE - Acoustic Emission TestingLength: 4 hours Questions: 135
1 Principles and Theory
• Characteristics of acoustic emission testing
• Materials and deformation
• Sources of acoustic emission• Wave propagation
• Attenuation
• Kaiser and Felicity effects, and Felicity ratio
• Terminology (refer to acoustic emission glossary, ASTM
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
13/469
• Signal condition
• Signal detection
• Signal processin
• Source location• Advanced signa
• Acoustic emissio
• Accessory mate
• Factors affecting
selection
2 Equipment and Materials
• Transducing processes
• Sensors
• Sensor attachments• Sensor utilization
• Simulated acoustic emission sources
• Cables
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
14/469
4 Interpretation a
• Data interpretat
• Data evaluation
• Reports
5 Procedures
6 Safety and Hea
7 Applications
• Laboratory stud
characterization
• Structural applic
3 Techniques
• Equipment calibration and set up for
test
• Establishing loading procedures• Precautions against noise
• Special test procedures
• Data displays
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
15/469
References & Catalog Numbers
NDT Handbook, Second Edition: Volume 5, Acoustic E
Catalog Number 130
Acoustic Emission: Techniques and Applications Catal
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
16/469
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
17/469
-
8/15/2019 AET-2006 Reading 3A
18/469
Reference Standards: ASTM:E 569 – Acoustic Emission Monitoring of Structures Durin
StimulationE 650 – Guide for Mounting Piezoelectric Acoustic Emiss
E 750 – Practice for Characterizing Acoustic Emission Ins
ASTM E 749-96 is a standard practice of AE monitoring
of continuous welding.
ASTM E 1932 for the AE examination of small parts
ASTM E1419-00 for the method of examining seamless, gvessels.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
19/469
API:RP 575 – Guidelines and Methods for Inspection of Existi
Low- Pressure Storage Tank.
ST 307 – An Engineering Assessment of Acoustic Method
in Aboveground Storage tanks.
ST 322 – An Engineering Evaluation of Acoustic Methods
in Aboveground storage Tank.
ST 325 – An evaluation of a Methodology for the detection
Aboveground Storage Tank.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
20/469
Reading#1
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
21/469
Acoustic Emission Non-DestrucTesting of Structures using Sou
Location Techniques Alan G. Beattie (2013)
Prepared by Sandia National Laboratories Albuquerque, New Mexico 87185 and Livermore, California 94550
Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subs
the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
22/469
AbstractThe technology of acoustic emission (AE) testing has bee
used at Sandia for the past 40 years. AE has been used o
including pressure vessels, fire bottles, wind turbines, gas
weapons, and solar collectors.
This monograph begins with background topics in acousti
instrumentation and then focuses on current acoustic emi
covers the overall design and system setups for a test, wi
blade as the object. Test analysis is discussed with an em
location. Three test examples are presented, two on expeturbine blades and one on aircraft fire extinguisher bottles
for a FORTRAN source location program is given as an e
analysis program. Throughout the document, the stress is
real structures, not on laboratory experiments.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
23/469
Nuclear Silo
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
24/469
Talented Youngster
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
25/469
Foreword Acoustic emission techniques in both Non-Destructive Te
Research and Development were first investigated by Kai
ago. The ability to triangulate to the source of an emission
same sound wave at several different sensors was recogn
In 1966, Green reported successful use of triangulation in
on a hydrotest of a Saturn S-II propellant tank. Since then
emission source location has rapidly expanded, pushed in
technological advances in the field of personal computers
simple analog systems to sophisticated digital systems. Aplaced not only on acquiring data, but also on using unive
included in the commercial systems, to analyze the data.
programs are quite versatile and significantly advanced ov
programs included in the early source location systems, th
of the possible types of analysis in the field of acoustic em
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
26/469
Saturn S-II propellant tank
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
27/469
Saturn S-II propellant tank
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
28/469
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
29/469
For the last forty years, the author has been using acoust
techniques at Sandia National Laboratories in widely vary
many of which required the development of specialized te
analysis techniques. Obviously, as in any new field, not a
techniques worked, but a general approach emerged. Thidetail the testing and analysis techniques developed in tw
Energy and Small Pressure Vessels. These applications u
to perform acoustic emission source location and involve
digital data. Because early commercial programs were no
for analysis, custom programs in FORTRAN were written.evolved, many of the custom techniques have been incorp
software furnished with the commercial systems. Howeve
commercial software are hidden in executable files. A few
developed FORTRAN programs are included here to illus
and assumptions used in acoustic emission programs. (FO
the easier languages in which to follow the workings of a cdespite the rise, decline, and death of many other comput
These working programs can be used with necessary min
an acoustic emission source location test.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
30/469
This treatise was written to describe some acoustic emiss
procedures which have worked well for the author. Neithe
rigorous scientific paper, this publication assumes that the
some familiarity with acoustic emission (AE). References
because upon my retirement 15 years ago, I gave my collpapers to a university library. For readers who want either
particular AE applications, or just a wide survey of the field
second edition of the Nondestructive Testing Handbook, v
Emission Testing, edited by Ronnie Miller. For a source o
papers on acoustic emission, I recommend the Journal ofpublished and edited by professor Kanji Ono. The Journa
internet and the entire 28 years of publication can be purc
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
31/469
Following a general discussion of acoustic emission and a
sources, this monograph then offers a general discussion
elastic waves in materials. To understand what is occurrin
either Ultrasonics or Acoustic Emission, one needs this ba
is included both to aid readers who are new to the field of and to serve as reference material for those with some pri
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
32/469
Briefly covered are elastic waves in solids including differe
reflection and refraction of waves, attenuation, superposit
of waves, resonance effects, and other phenomena which
encountered in an acoustic emission test. Also discussed
couplants, preamplifiers and other topics involved in perfoemission test.
Two brief sections cover acoustic emission source mecha
parameters of acoustic emission waveforms which are us
commercial systems. All of this material was covered in th
review article and makes no pretense of being new or diffethat the most useful parts of this monograph will be the dis
design, conduction, and analysis of several types of acous
No two tests are ever identical but the general methodolo
applicable.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
33/469
1.0 CHAPTER 11.1. Introduction to Acoustic Emission
Acoustic Emission (AE) is the study and practical use of e
generated by a material subjected to an external stress. T
was recognized by early miners who exited a mine when t
supporting timbers started groaning. Tin cry, the sound pr
bar is bent, was known soon after the production of metal
C.S. Barrett mapped a low temperature phase transition in
Magnesium alloys by sticking a phonograph needle into th
recording the output as the temperature was changed. J. the signals produced by samples undergoing tensile testin
the Kaiser effect, i.e. that no signals were generated by a
second loading until the previous maximum load was exce
Kaiser’s thesis was published in 1950, several groups inv
phenomena for possible use in testing structures. In the e
Green and a group at Aerojet Corporation started using ASaturn Rocket propellant tanks. They used a form of trian
the arrival times of the acoustic pulse at several acoustic
This was the direct precursor of the work that is described
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
34/469
Despite its longstanding use, the definition of “acoustic em
source of controversy for the last fifty years. Strictly, “acou
pressure waves detected by one’s ear. However, elastic w
not limited to pressure waves, and all types of vibrational
generated by acoustic emission sources. Even so, the teremission” has become almost universally used for the phe
waves generated by an internal event in a media.
In this monograph, “acoustic” will refer to any elastic wave
acoustic emission source. Acoustic emission, then, is the
elastic wave by the rapid change in the stress state of sommaterial. This change is usually caused by the application
stimulus to the material. The material can be a solid, liquid
and the external stress (?) can be applied mechanically, t
magnetically, etc.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
35/469
The stress change must be rapid enough to transmit some
surrounding material and dissipates as an elastic wave.
On a macroscopic scale this definition includes earthquak
while on a microscopic scale it includes the fracture of cryMartensitic phase transformations. The occurrence of the
completely determined by the local conditions, the local st
physical state of the region. As a result, neither the exact
energy burst occurs nor the exact details of the generated
determined beforehand.
In general, the event generating the emission is irreversib
ruptured geologic fault nor a fractured crystallite in a meta
spontaneously.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
36/469
Once the elastic wave is generated, it then travels through
and can be detected at considerable distances from its po
Traveling from its source to the point of detection, the wav
the characteristics and variations of its acoustic path. Its e
by (1) geometric spreading and (2) scattering by both micmacroscopic variations in the material’s structure. Other k
may also be present.
The wave’s frequency content is generated by the source
travels the acoustic path.
The primary information carried by the wave is the (1) time
(2) the elastic energy detected at each sensor on the stru
of a sensor indicates that something happened in the spe
time, while the amplitude indicates the level of the disturba
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
37/469
The apparent location of the source and all other characte
detected signal are modified by the characteristics of the w
sensor as well as the characteristics the sensor. As a resu
controlled laboratory experiment, a reasonable estimate c
characteristics of the emission source; however, in a test structure, we are limited to what are basically statistical es
multiple emissions to tell us what is happening in the spec
analysis is not generally used in the study of acoustic emi
of the specimen and the location coordinates of the sourc
averages of calculated values from multiple emissions.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
38/469
Elastic energy is ubiquitous. It travels through all materials
interaction between atoms. Only a hard vacuum does not
energy. The distance from its source over which an acous
detected depends on its initial amplitude and the acoustic
the material through which it is traveling. For many structuacoustic emission signals can be detected from almost an
structure. This allows an AE test to cover an entire structu
small region. However, many acoustic signals which have
the test can be present and detected. Acoustic isolation o
highly desirable. Most acoustic emission testing filters outfrequencies below 20 KHz, thus ignoring background nois
communication in the test area.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
39/469
An acoustic emission test occurs in real time. The test doe
preexisting defects, but detects flaw growth as it occurs. If
application simulates the conditions that the test structure
experience in actual operation, serious acoustic emission
pathological flaws that could lead to failure of part or all ofamount of emission detected and the locations of the emi
depend both on the design of the structure and the materi
fabrication. For example, compare the bending of a strip o
of FRP (Fiber Reinforced Plastic) with identical dimension
will likely give a single high-amplitude burst of emission juthe FRP will show a period of low level emissions followed
level emissions and then one or more high amplitude emi
starts to fail and then tears apart or snaps. Interpreting de
the test engineer’s job. The value of the information obtain
determined first by the design of the test and second, by t
detected emission. For complex structures, both the test dinterpretation are seldom simple. It is the purpose of this m
the test engineer in both areas.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
40/469
1.2. Sources of Acoustic Emission
Acoustic emissions are acoustic waves generated by a ra
stress state of a region in a material.
Acoustic waves are one of two (?, UT? VA?) non-electromtransmitting energy through a material; the other method i
A difference between the two is that thermal diffusion invo
transfer between individual atoms, while an acoustic wave
by a cooperative motion of many atoms. An acoustic wave
as a pressure pulse in a gas or liquid, or as complicated a
of transmission in a bounded solid. The generation of an ainvariably involves a large region of atoms.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
41/469
The most common acoustic waves are sound waves in ai
are produced by vibration of a surface such as the vocal a
animal, a leaf in the wind, or the diaphragm of a speaker.
quasi-continuous, involving a modulated vibration of an in
generating region and the transmitting medium. Another tywave is generated by a sudden disturbance in or on the m
transient event that produces a transient acoustic wave. S
wave is what we define as acoustic emission. It may have
damped wave with complex frequency content or it may a
small transient events which sum into quasi-continuous nocharacteristics are that it is generated in the medium whic
is transient in nature. Any sudden movement of a group o
near the sound velocity in a material can produce a transi
The apparent quasi-continuous signals which are often se
generated by the overlap of many transient events instead
vibration of a surface.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
42/469
Figure 1-1 shows waveforms both from a transient event a
superposition of many transient events. The long decay o
is produced by reflections of the original wave in the comp
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
43/469
Figure 1-1.
Examples of burst
emission and
continuous
emission from ahigh strength
aluminum alloy
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
44/469
The size of the region generating an emission can vary fro
row or plate of atoms moving simultaneously into a new c
during a Martensitic phase transition to a fracture in a sub
between two continental plates. The energy released in a
emission event will be roughly proportional to the volumeEnergy from these waves ranges from smaller than an ele
energy contained in a thunder clap or that of a magnitude
Wave frequencies are generally related inversely to the vo
generating region, ranging from thousandths of a Hertz fo
several MHz in fine grain metals. An important characterisemission in solids is that the fracture of the region occurs
stress vector exceeds the strength of the region to withsta
other words, the exact time when the emission occurs stri
local conditions. In a metal, for example, the precise envir
crystallite differs, and the fracture of one crystallite or of th
between two crystallites will make small changes in the lothe other crystallites.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
45/469
This complete dependence on local conditions means tha
predict the exact time of any one emission or the time inte
two emissions. Acoustic emissions occur unpredictably in
response not to the applied external stress but to the loca
produced at each position throughout the material. The padetected emission depend not only on the characteristics
source, but also on the characteristics and geometry of th
the source and sensor, and on the characteristics of the s
couplant between sensor and medium.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
46/469
The signal characteristics detected by individual identical
various positions on a specimen will often vary greatly for
The more complex the medium in which the emission is g
transmitted, the more likely there will be large differences
signals from the same emission at different sensors. The of origination and the wide variety of waveforms in separa
fundamental characteristics of acoustic emission. As such
a profound effect on the type of analysis used on the dete
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
47/469
In structural metals, both inter-granular and trans-granula
crystallites can generate low level acoustic emission. Emi
generated at inclusions both by fracture of the inclusion a
the bonds between the inclusion and metal. Crack growth
several crystallites are involved can generate emission of amplitudes. The presence of corrosion on a metal surface
the fracture of brittle corrosion byproducts while active cor
emission from bubble formation. In high stress environme
of crystalline distortion) can occur in some metals and this
emission. Room temperature creep in metals may involvedislocations in the metal. Laboratory experiments have cla
detect very low amplitude emission from creep, but the au
its use in structural flaw detection. The creep rate in struct
normal usage is usually far too slow to generate acoustic
for flaw detection.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
48/469
Source mechanisms in FRP include matrix cracking, debo
matrix and fibers, fiber fracture, and crack propagation. Th
mechanisms can all be activated by the application of an
Based on the idea that the emission amplitude is related t
source, matrix cracking should produce the lowest amplitufollowed by matrix-fiber debonding. Fiber breakage would
partly due to the amount of energy released by the fractur
fiber. Finally, crack propagation, which includes all three o
mechanisms, would produce the highest amplitude emiss
generally to be the case, but trying to quantify it as a rule very well, probably because FRP is usually laid up by han
close to the structural uniformity of a well annealed metal.
an FRP structure will generate acoustic emission at loads
strength of the structure.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
49/469
The most likely cause of this emission is the relief, by min
FRP during the first loading, of high residual stresses whic
local regions during the curing process. The fractures hav
strength of the material, and the emission usually stops be
loads are reached. This emission is seldom seen upon thesubsequent loadings (Kaiser’s effect?) .
Another source of emission in FRP structures is often see
loading. This emission appears in the middle ranges of th
the rate of change in applied stress is highest, instead of a
appears to be caused by friction between small regions inare not bonded. This type of emission can be present dur
load test; however, it does not correlate with structural da
object. Structural damage is associated with emission tha
peak loads, especially during the rising load.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
50/469
As discussed above, most acoustic emission mechanisms
permanent change in the micro structure of the material. O
fracture occurs, it will not happen again unless there is so
mechanism. Therefore, acoustic emission appears irrever
effect, where the re-stressing of a specimen will not generemission until the previous load level has been exceeded
irreversibility.
The Kaiser effect holds very well for the immediate re-stre
specimen, but less well for composites. The problem is th
of an external stress does not necessarily take the specimmicro-stress path. One often sees emission on subsequen
lower than previously reached. The ratio of the load value
starts on subsequent loadings, to the maximum load value
previous loading, known as the Felicity ratio, indicates po
induced by the previous loading. Many NDT tests of FRP
test load in a series of steps, returning to zero between eaappearance of Felicity ratios much less than 1.0 is a good
significant damage occurred in previous loadings.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
51/469
2.0 CHAPTER 22.1. Acoustic Waves in Solids
Acoustic emissions are acoustic waves generated by a m
response to a change in stress. Once generated, these w
under the same rules as any other acoustic wave. They a
motion in a collection of atoms. Acoustic wave motion is a
movement of the atoms in a material extending over a hug
This collective motion implies that the wavelength is long
distance between the atoms. Wavelength is inversely prop
frequency of a wave; therefore, an acoustic wave is usualrelatively low frequency. For example, a frequency as high
Aluminum would still imply a wave extending over 1.5 x 1
simplest type of acoustic wave is a pressure wave which o
of material around a region is suddenly compressed by a
region. This compression can be either positive or negativ
exploding or imploding.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
52/469
The material in this shell experiences a change in its dens
change is then passed to the next shell by coupling betwe
density in the first shell then rebounds, usually going past
value to a smaller density change in the opposite direction
the density is transmitted to the next shell, and so on, throThe strength of the coupling between the material’s atoms
the density of the material determines the speed with whic
propagates . The resulting wave is known as a compressi
occurs in all materials, solids, liquids, gases, and plasmas
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
53/469
In a compressional wave, the average motion of the atom
wave’s direction of travel. As materials become more rigid
forces between atoms become far more complex and dep
geometrical positions of the atoms. More modes of wave
and the averaged atomic motions are no longer restrictedpropagation direction . The introduction of boundaries in a
introduce further complications As a wave passes through
another material, the difference in physical propertied betw
will produce reflections and refractions as well as changes
propagation. A wave traveling along a surface will have far different pa
one propagating in the bulk.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
54/469
2.2. Characteristics of Acoustic Waves
Acoustic emission signals generate complexity as the wav
the medium. Understanding acoustic emission signals req
the wave characteristics, starting with the properties of the
which the wave travels. All materials are collections of atoattractive forces while simultaneously prevented, by short
forces, from approaching each other too closely. The supe
forces results in an equilibrium position for the atom at its
material’s most stable configuration. In crystals, for examp
forces between the atoms result in defined locations for thpositions. The result is a crystal structure. In a liquid the s
amorphous, but an approximate distance between atoms
density. In a liquid, only the density is defined. There are n
the atoms as there are in a crystal lattice.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
55/469
The ability of the atoms to change position is measured by
liquid, which can range from very low to extremely high. A
are a gas and a solid. A gas is a liquid with very low visco
liquid with very extremely high viscosity. The density of a
by its total mass and the volume of its container. The acoumaterial depend on its density and the long-range couplin
between its atoms.
The long-range correlations in an acoustic wave result in
small region being displaced in the same direction from thpositions. This displacement is a local dynamic strain in th
strain’s direction and magnitude are constantly changing a
When the atomic motion is pseudo-oscillatory, so is the st
wave is an oscillating strain moving through a material. Be
strain are always directly related in a material, there is als
stress field. Therefore, an acoustic wave can be describeddynamic stress or strain field in a material.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
56/469
2.3. Wave Motion
The most familiar depiction of a wave is a sinusoidal curve
Figure 2-1a. The amplitude oscillates between positive an
a fixed rate, known as the frequency, and the curve exten
curve can be plotted equally well as a function of space oa wave has both a spatial and a time component. An equa
curve is:
Where:
A is the amplitude;
ω is 2π times the frequency, υ; andk, the wave number, is 2π over the wave length, λ.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
57/469
The frequency, wave length, and wave velocity, v are rela
2.1b. The wave described in equation 2.1a propagates in In three dimensions, the wave front, which is a surface of
the wave, is a plane perpendicular to the X axis. Such a w
plane wave. Most waves originating at a point in an exten
have a spherical wave front. However, at some distance f
origin, the spherical surface will approximate a plane oversimplicity, we will assume plane waves for the rest of this
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
58/469
Figure 2-1. (a) simple sine wave, (b) sum of two sine wav
wave, (d) spectrum of transient wave
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
59/469
If two waves exist in a medium simultaneously, their ampl
algebraically. Fig.2-2 shows the sum of two waves:
where only the time component is plotted for clarity. Thus
waves can represent a complex wave form. It has been lo
arbitrary transient function which does not contain a disco
represented by an infinite sum of sinusoidal curves knownOne form of such a series can be written as
harlie Chong/ Fion Zhang
(2.2)
(2.3)
-
8/15/2019 AET-2006 Reading 3A
60/469
where Ao and An are the amplitudes of the sine curves, the
frequencies, and the αn are the phases.
The Fourier series does not dictate that the curve or wave
A transient wave such as that shown in Fig. 2-1(c) can beFourier series. A useful method of analyzing a wave is to
components. A plot of the square of the amplitudes of the
components, An in equation 2.3, against the frequency, υfrequency spectrum of the wave. Fig. 2-1(d) shows the sp
shown in Fig. 2-1(c).
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
61/469
2.4. Acoustic Media
An acoustic wave exists only in a material, whether a solid
plasma. The material’s characteristics determine the wave
stronger the force between neighboring atoms, the more c
be their motion. On the other hand, the larger the mass ofmore force must be applied for the same acceleration. Be
synchronized movement of a large number of atoms, it is
of the material, ρ, rather than the mass of the individual atwave motion. Thus the wave velocity should be directly pr
atomic restoring force between the atoms or molecules, aproportional to the density. The actual relationship is:
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
62/469
where Vi is the velocity for the particular type of wave and
elastic constant for that type of wave. The elastic constan
strength of the coupling between atoms for that particular
Different relative motions of the atoms will have different v
constant. Another property of the material is the characterdefined by equation 2.5, as follows:
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
63/469
The reflection and transmission of acoustic waves at an in
two materials depend on the characteristic acoustic imped
materials. This dependence is given in equation 2.10. Aco
acoustic impedances, and densities for some materials of
emission tests are given in Table 1.
Table 1. Acoustic velocities and impedances for longi
Raleigh waves for several materials
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
64/469
2.5. Types of Acoustic Waves
The path traced out by a particle under the influence of an
generally be represented by an ellipse with one of its axes
direction of the wave’s travel. The type of wave in bulk ma
determined by the relationship between the average particdirection of travel of the wave. For materials with boundar
of the wave and particle motion will be determined by the
the physical geometry of the sample, and the frequency o
Waves traveling through an extended medium (one whosmuch larger than the acoustic wave length) are called bul
types of pure bulk waves are longitudinal (compressional)
(transverse) waves. In both these waves, the minor axis o
atomic paths collapses toward zero, resulting in an approx
oscillatory motion.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
65/469
For longitudinal waves, the average motion of the atomic
the material is parallel to the direction of the wave propag
have the average of this motion perpendicular to the direc
(The motion of atoms around their equilibrium position is v
with much higher frequency components than acoustic wawaves are averages over a very large number of atoms in
average acoustic motion will be ellipses around the lattice
is no requirement that the axes of the ellipse correspond t
directions in the lattice).
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
66/469
Particle motion in longitudinal and shear waves are illustra
and 2-2(b).
Since the relative motions of the particles in these two wa
the elastic constants and the wave velocities, therefore, athe shear velocity is slightly greater than one half of the lo
Waves often have both shear and longitudinal component
component traveling at its own velocity. In a non-attenuati
medium, a transient wave—sampled at some distance fro
origin—may appear to be two separate waves, one longitu
shear, as illustrated in Fig. 2-3.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
67/469
Figure 2-2. Particle
displacement for Bulk
acoustic waves, (a)
compressional wave,
(b) shear wave
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
68/469
Figure 2-3. Waveform with
compression and shear wave (a) at
origin, (b) some distance from
origin
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
69/469
Figure 2-3. Waveform with
compression and shear wave (a)
at origin, (b) some distance from
origin
harlie Chong/ Fion Zhang
shear
shear velocity is slightly greaterthan one half of the longitudinal
velocity. (Vshear = ½ Vcompression )
-
8/15/2019 AET-2006 Reading 3A
70/469
The anisotropy of the coupling forces between atoms at th
bounded solid will produce additional types of waves. A su
maximum amplitude at the surface of the solid; its amplitu
distance from the surface. The plane of the particle motion
either parallel (Love waves) or perpendicular (Rayleigh waHowever, because most acoustic emission sensors detec
perpendicular to the surface, the parallel component is se
velocity of Rayleigh waves is slightly lower than the shear
is bounded by two surfaces so that it is a plate, and the th
is on the order of a few acoustic wave lengths or less, plawaves) can occur. A plate wave is essentially two surface
synchronized either symmetrically or antisymmetrically. P
Rayleigh waves and plate waves are illustrated in Fig. 2-4
Bulk waves, surface waves, and plate waves are the mos
waves seen in the field of acoustic emission. However, th
types of waves found in solids. In general, bounded solids
symmetrical geometry can support unique types of waves
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
71/469
Figure 2-4. Particle displacements for
acoustic waves: (a) Rayleigh Wave, (b)
Plate wave, first symmetric mode (c)
Plate wave, first antisymmetric mode
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
72/469
R a y l e i g
h W a v e
P l a t e w
a v e
s y m m e t r i c m o d e
-
8/15/2019 AET-2006 Reading 3A
73/469
2.6. Dispersion and Group Velocity
The velocity defined in 2-4 is the phase velocity.
For unbounded media and surface waves on a single surf
velocity is independent of frequency.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
74/469
By contrast, all waves traveling in bounded media (where
dimensions are within an order of magnitude of the acous
dispersive, that is, the phase velocity is a function of frequ
Fig. 2-5 illustrates this principle, showing the frequency develocities for symmetric and anti-symmetric plate waves. D
have little effect on continuous waves. However, acoustic
packets of waves which can be thought of as a superposit
waves, as shown in equation 2.3. If each wave train, mak
travels at a different velocity, the wave packet will change
through the medium. As a result, the same acoustic emissdifferent when detected by the same sensor at different po
wave packet travels not at the phase velocity, but at the g
phase velocity can be defined by rewriting equation 2.4 as
harlie Chong/ Fion Zhang
(2
-
8/15/2019 AET-2006 Reading 3A
75/469
Figure 2-5. Phase velocities for
different plate wave modes
plotted against the ratio of plate
thickness to acoustic
wavelength. Plotted for steelwith a Poisson’s ratio of 0.28.
The longitudinal, extensional,
shear, and Raleigh wave
velocities are shown.
harlie Chong/ Fion Zhang
Keys:Lamb, S0 symmetric mode
Lamb, A0 asymmetric mode
-
8/15/2019 AET-2006 Reading 3A
76/469
while the group velocity is defined as:
In the absence of dispersion, these are the same velocity,
bounded solids the group velocity will be less than the pha
frequency dependent velocity can have real effects in acoone is attempting source location by measuring the differe
times at two or more sensors.
Keywords:
group velocity
phase velocity
harlie Chong/ Fion Zhang
(2.7)
-
8/15/2019 AET-2006 Reading 3A
77/469
2.7. Attenuation
A wave packet is generated with a well-defined energy. A
propagates away from its source, the energy content will r
the absence of any dissipative mechanisms.
However, if the wave front of the packet is expanding, the
area on the wave front must decrease to conserve the tota
wave front. The rate of this decrease will depend on the g
medium.
In three dimensions, the energy per unit area will decreas
the distance from the source, while in two dimensions the
area will decrease linearly with this distance. If the packet
dimension, as in propagation down a rod, the energy per
independent of the distance from the source.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
78/469
DiscussionSubject: “In three dimensions, the energy per unit area w
square of the distance from the source, while in two dimen
per unit area will decrease linearly with this distance. If theto one dimension, as in propagation down a rod, the ener
be independent of the distance from the source.”
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
79/469
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
80/469
harlie Chong/ Fion Zhang
If the packet is confined to o
as in propagation down a ro
per unit area will be indepen
distance from the source
in two dimensions thearea will decrease line
distanceIn three dimensions, the energy
per unit area will decrease as the
square of the distance from the
source
-
8/15/2019 AET-2006 Reading 3A
81/469
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
82/469
Normally, in the context of acoustic waves, one assumes
traveling in only one dimension. Therefore this geometrica
packets’ energy is ignored. However, in an acoustic emiss
neither the location of the source nor the geometry of the
the investigator’s control, this geometrical effect should beattempt to measure the energy of the generated wave pac
attenuation of a plane wave arises from dissipative mecha
as the wave propagates. In a homogeneous medium, thes
occur as a fixed percentage of the wave packet energy pe
travel.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
83/469
Mathematically, this is an exponential decrease in the wav
distance that can be expressed as:
where α is an attenuation constant per unit length, and βconstant per unit time. The two constants are related by th
as shown by:
harlie Chong/ Fion Zhang
(2.9)
-
8/15/2019 AET-2006 Reading 3A
84/469
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
85/469
Both forms of the attenuation constant are seen in the lite
Most of the many types of acoustic attenuation mechanism
of frequency dependence. Fortunately, in the normal acou
frequency range of 50 kHz to 1.0 MHz, both the frequency
the magnitude of many of these attenuation mechanisms structural materials. However, in composites, geological m
concrete the attenuation can be a severely limiting factor
tests, often restricting the useable frequency range to 100
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
86/469
2.8. Interfaces, Reflection, Transmission, and Mode C
If a plane wave strikes a surface between two materials w
impedances, part of the wave will be reflected and part tra
intensities of the reflected and transmitted components ar
where the Zi are the acoustic impedances of the materialsare symmetrical, i.e. it does not matter in which medium th
when it hits the interface. The differences in acoustic impe
in large differences in the acoustic intensities transmitted
For example, the transmitted intensity of longitudinal wave
steel-aluminum, 12% for a steel-water, and 0.004% for a s
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
87/469
When a plane wave strikes the interface, the angles of the
transmitted waves are governed by Snell's law
where θ1 is the angle of incidence, θ2 is the angle of reflerefraction, and the Vi (V1,2?) are the velocities in the mate
2.11, a transmitted velocity is positive and a reflected oneparticle motion anywhere on a wave front of a plane wave
wants to remain the same even when the wave passes an
at an interface, the direction of propagation will change ev
particle motion does not.
For a wave perpendicular to the surface (θ1 = 0°), this reschange of 180° in the relative motion of the particle to the without changing the character of the wave.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
88/469
Wave Perpendicular To The Surface (
1 = 0
)
harlie Chong/ Fion Zhang
For a wave
the surface
results in a180° in thethe particle
direction w
character o
-
8/15/2019 AET-2006 Reading 3A
89/469
For non-perpendicular angles of incidence, the reflected a
waves will have both longitudinal and shear components
motion is parallel to the interface) because of the change
between the particle motion and the propagation direction
in Fig. 2-6. The process of generating both modes of bulkreflection or refraction is known as mode conversion.
In acoustic emission, where there is no control of the wav
inevitably the wave reaching the sensor is composed of b
shear components, no matter what its original polarizationsituations, surface waves are also present. Since mode co
almost every reflection, it is an almost continuous process
propagates in a bounded medium. Because of this continu
between modes traveling at different wave velocities, the
will lengthen in time as it travels instead of dividing into se
and shear components as shown in Fig. 2-3.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
90/469
Figure 2-3. Waveform with
compression and shear wave (a) at
origin, (b) some distance from
origin
harlie Chong/ Fion Zhang
Because of this continuous
transformation between modes
traveling at different wave
velocities, the transient wave form
will lengthen in time as it travelsinstead of dividing into separate
longitudinal and shear
components as shown in Fig. 2-3.
-
8/15/2019 AET-2006 Reading 3A
91/469
Figure 2-6. Reflected and transmitted waves across an in
incident wave is a longitudinal wave with an angle of incid
double arrows show the direction of particle motion assoc
wave.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
92/469
-
8/15/2019 AET-2006 Reading 3A
93/469
-
8/15/2019 AET-2006 Reading 3A
94/469
Figure 2-7. Reflected and transmittedwaves inside a plate immersed in water.
(a) Successive reflections have been
displaced for clarity.
(b) Strain in a plate one half wavelength
thick.
(c) Amplitude as a function of frequency
in the plate for high and low Q
materials.
(d) Strain in a plate one wavelength and
one and one half wavelength thick.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
95/469
If there are a great many reflections, the peak strain can rlevel. If there are only a few reflections, the amplitude of tmuch less, and the peak will spread over a wider frequencfrequency of this wave need not be exactly that of a half w
some reinforcement; however, the greater the number of rnarrower will be the allowed frequency range at maximumlarger this maximum strain will be. This is illustrated in Fhigh amplitude peak is said to have a high Q, where Q is thenergy stored to energy dissipated. This increase of the str
a material at a half-wave thickness is known as a resonancfrequency. From Fig. 2-7b, we see that at resonance, the athroughout the plate is a maximum. Resonances can occugeometry allows acoustic waves to reflect in such a way thseveral reflections of the wave are superimposed. In piezo very high Q resonances allow precise generation of single
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
96/469
Figure 2-7. Reflected and transmitted waves inside a plateimmersed in water. (a) Successive reflections have been
displaced for clarity. (b) Strain in a plate one half wavelength
thick. (c) Amplitude as a function of frequency in the plate for
high and low Q materials. (d) Strain in a plate one wavelength
and one and one half wavelength thick.
-
8/15/2019 AET-2006 Reading 3A
97/469
Longitudinal waves- Plane pressure pulse waveLongitudinal waves, Longitudinal "l waves", are waves in which the displacement o
direction as, or the opposite direction to, the direction of travel of the wave. Mechan
also called compressional waves or compression waves, because they produce co
when traveling through a medium. The other main type of wave is the transverse w
displacements of the medium are at right angles to the direction of propagation. Somechanical, meaning that the wave needs a medium to travel through. Transverse
called "t-waves" or "shear waves".
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
98/469
Longitudinal waves- Representation of the propagation
omnidirectional pulse wave on a 2d grid (empirical shape)Longitudinal waves, Longitudinal "l waves", are waves in which the displacement o
direction as, or the opposite direction to, the direction of travel of the wave. Mechan
also called compressional waves or compression waves, because they produce co
when traveling through a medium. The other main type of wave is the transverse w
displacements of the medium are at right angles to the direction of propagation. So
mechanical, meaning that the wave needs a medium to travel through. Transverse
called "t-waves" or "shear waves".
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
99/469
Shear Wave– Plane shear waveS-waves, secondary waves, or shear waves (sometimes called an elastic S-wave)
and are one of the two main types of elastic body waves, so named because they
object, unlike surface waves. The S-wave moves as a shear or transverse wave, s
the direction of wave propagation. The wave moves through elastic media, and the
from shear effects. These waves do not diverge, and they obey the continuity equamedia.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
100/469
Shear Wave– Propagation of a spherical S-wave in a 2
model) S-waves, secondary waves, or shear waves (sometimes called an elaselastic wave, and are one of the two main types of elastic body waves, so named b
the body of an object, unlike surface waves. The S-wave moves as a shear or tran
perpendicular to the direction of wave propagation. The wave moves through elast
restoring force comes from shear effects. These waves do not diverge, and they o
for incompressible media.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
101/469
Longitudinal WavesIn a longitudinal wave the particle displacement is parallel to the direction of wave
at right shows a one-dimensional longitudinal plane wave propagating down a tube
down the tube with the wave; they simply oscillate back and forth about their individ
Pick a single particle and watch its motion. The wave is seen as the motion of the
pressure wave), which moves from left to right. The second animation at right showoscillatory motion of individual particles and the propagation of the wave through th
also identifies the regions of compression and rarefaction.
The P waves (Primary waves) in an earthquake are examples of Longitudinal wave
the fastest velocity and are the first to arrive.
harlie Chong/ Fion Zhang http://www.acs
-
8/15/2019 AET-2006 Reading 3A
102/469
-
8/15/2019 AET-2006 Reading 3A
103/469
Transverse WavesIn a transverse wave the particle displacement is perpendicular to the direction of w
animation below shows a one-dimensional transverse plane wave propagating from
do not move along with the wave; they simply oscillate up and down about their ind
as the wave passes by. Pick a single particle and watch its motion.
The S waves (Secondary waves) in an earthquake are examples of Transverse waa velocity slower than P waves, arriving several seconds later.
harlie Chong/ Fion Zhang http://www.acs
-
8/15/2019 AET-2006 Reading 3A
104/469
Water WavesWater waves are an example of waves that involve a combination of both longitudi
As a wave travels through the waver, the particles travel in clockwise circles. The r
decreases as the depth into the water increases. The movie below shows a water
right in a region where the depth of the water is greater than the wavelength of the
particles in yellow to show that each particle indeed travels in a clockwise circle as
harlie Chong/ Fion Zhang http://www.acs
-
8/15/2019 AET-2006 Reading 3A
105/469
-
8/15/2019 AET-2006 Reading 3A
106/469
More Reading on“Phase velocities for different p
modes”
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
107/469
Condition monitoring of large oil anstorage tanks using guided waves
AbstractLarge storage tanks containing hazardous liquids such as
products and food processing liquids are common through
Corrosion in the tank floor is a serious environmental and
order to monitor the condition of the tanks and prevent lea
must be inspected at regular intervals. Currently before ancarried out, the tank must be emptied and cleaned. This is
dangerous process due to weeks of lost production, trans
to temporary storage tanks and exposure of the workers to
inspection and cleaning.
harlie Chong/ Fion Zhang http://www.twi-global.com/technical-knowledge/published-papers
-
8/15/2019 AET-2006 Reading 3A
108/469
TWI is managing a European CRAFT project called TANK
overcome the drawbacks of current inspection practices. B
wave sensors outside the tank and using reconstructive to
techniques there is potential for carrying out an inspection
instantly and without the requirement of emptying and cleaoperator entry inside the tank. TWI has been working with
Lithuania, using numerical modelling to study the potentia
technique. The effects of lap joints in the tank floors and a
liquid contents of the tank have been considered using bo
finite element modelling methods.
harlie Chong/ Fion Zhang http://www.twi-global.com/technical-knowledge/published-papers
-
8/15/2019 AET-2006 Reading 3A
109/469
1. IntroductionIn producing a system for the inspection of tank floors the
technical challenges to overcome. Guided waves, by natu
prismatic structures and therefore any changes to the cros
lap joint can cause undesirable reflections and noise, mak
signal difficult to interpret. In addition to this, the sound en
by materials surrounding the tank such as the liquid conte
sand or concrete underneath the tank. Modeling has been
study such effects in isolation and therefore gain a better
these issues so that practical solutions can be found.
harlie Chong/ Fion Zhang http://www.twi-global.com/technical-knowledge/published-papers
-
8/15/2019 AET-2006 Reading 3A
110/469
2. Global matrix model An initial global matrix model (a set of analytical formulae
waves in a multi-layer structure) was set up to consider th
varying connectivity possible at a lap joint. The bonding m
region where the steel plates overlap and these variations
the Lamb waves velocities and attenuation. The bond stre
layers has been considered as three different states: 'perf
The 'perfect' bond state means that all displacements are
the interface. In this case, the bonding layer properties ha
be the same as the steel plate properties and the bondingrelatively small. The 'poor' bond state means that displace
boundary are reduced compared with the 'perfect' bonding
interface. This was achieved by reducing the longitudinal
velocities in the bonding layer by a factor of two. The 'slip'
that shear displacements are not transferred through the i
state was reached setting shear velocity to 0m/s. A schem
shown in Fig.1.
harlie Chong/ Fion Zhang http://www.twi-global.com/technical-knowledge/published-papers
-
8/15/2019 AET-2006 Reading 3A
111/469
Fig.1. Schematic of lap joint model
harlie Chong/ Fion Zhang http://www.twi-global.com/technical-knowledge/published-papers
-
8/15/2019 AET-2006 Reading 3A
112/469
-
8/15/2019 AET-2006 Reading 3A
113/469
Fig.2. Phase velocity dispersion curves for 6mm steel pla
harlie Chong/ Fion Zhang http://www.twi-global.com/technical-knowledge/published-papers
-
8/15/2019 AET-2006 Reading 3A
114/469
Fig.3. Attenuation curves for 6mm steel plate with diesel o
harlie Chong/ Fion Zhang http://www.twi-global.com/technical-knowledge/published-papers
-
8/15/2019 AET-2006 Reading 3A
115/469
The phase velocity dispersion curves for a multi-layered z
one side are shown in Fig.4. There is little difference betw
states considered. However, in the 'slip' case, it was found
generated that propagates in the bond interface with zero
a Stoneley wave.
harlie Chong/ Fion Zhang http://www.twi-global.com/technical-knowledge/published-papers
-
8/15/2019 AET-2006 Reading 3A
116/469
Fig.4. Phase velocity dispersion curves for steel lap joint w
bonding layer conditions
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
117/469
3. Finite element model A finite element model was used to analyse the propagati
a lap joint with diesel on one side and a vacuum on the ot
was assumed to exist between the two plates. The plates
be joined by a 45° fillet weld. Figure 5 shows the propagawave at different moments in time.
The first two figures in Fig.5 shows the lamb waves in the
attenuation into the liquid before the waves reach the weld
in Fig.5 shows the interaction of the waves with the joint.
This is shown in more detail in Fig.6. A lot of attenuation i
observed, induced by the fillet weld geometry.
Figure 7 shows the energy still existing in the lower steel p
through the lap joint.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
118/469
Fig.5. Lamb wave propagating in steel lap joint at differen
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
119/469
-
8/15/2019 AET-2006 Reading 3A
120/469
Fig.5. Lamb wave propagating in steel lap joint at differen
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
121/469
-
8/15/2019 AET-2006 Reading 3A
122/469
Fig.7. Finite element modeling results after lamb wave ha
fillet weld
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
123/469
5. ConclusionsThere is great potential for the use of guided waves to ins
service. The work has shown that modelling can be us
modes to optimise the minimisation of attenuation loss
contents of the tank. The modelling has also shown tha
propagate past fillet welds and therefore there is poten
corrosion anywhere on a typical tank floor with lap join
Acknowledgement
The authors would like to thank the other partners working on this project: Spree E
ISOtest Engineering Srl, Coaxial Power Systems Ltd., Royal Vopak, ST ServicComputer Consultancy Ltd. and the European commission for funding this wo
References
Non-destructive testing Handbook, 2nd edition, Vol. 7. Ultrasonic Testing / A.S. Bir
USA: American Society for Non-destructive Testing, 1991.
Demcenko A., Mazeika L. Calculation of Lamb waves dispersion curves in multi-la
Ultragarsas. 2002. Vol. 3 (44). P. 15 - 17.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
124/469
More on Lamb WaveVelocity dispersion inherent in the characteristic equa
Lamb waves exhibit velocity dispersion; that is, their veloc
depends on the frequency (or wavelength), as well as on and density of the material. This phenomenon is central to
understanding of wave behavior in plates. Physically, the
the ratio of plate thickness d to wavelength {\displaystyle \
determines the effective stiffness of the plate and hence th
wave. In technological applications, a more practical para
derived from this is used, namely the product of thickness
harlie Chong/ Fion Zhang
since for all waves
-
8/15/2019 AET-2006 Reading 3A
125/469
The relationship between velocity and frequency (or wave
in the characteristic equations. In the case of the plate, th
not simple and their solution requires numerical methods.
intractable problem until the advent of the digital compute
Lamb's original work. The publication of computer-generacurves" by Viktorov[3] in the former Soviet Union, Fireston
Worlton in the United States, and eventually many others
theory into the realm of practical applicability. Experiment
observed in plates can be understood by interpretation wi
dispersion curves.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
126/469
Dispersions curves of free Lamb waves for two different P
x-axis shows the product of angular frequency and plate t
by the shear wave velocity. The y-axis shows the phase v
wave normalized by the shear wave velocity. For high freq
have the Rayleigh wave velocity, approximate 92 % of thevelocity.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
127/469
-
8/15/2019 AET-2006 Reading 3A
128/469
3.0 CHAPTER 33.1. Detection of Acoustic Emission
3.1.1. Sensors
A sensor is a device which generates an electrical signal w
by an acoustic wave. Acoustic emission (AE) sensors can
several physical principles. The signals can be generated
devices such as phonograph pickups, capacitive micropho
magnetostrictive devices, piezoelectric devices, and by th
interferometers to detect the surface displacement of the s
relationship between the characteristics of the wave and twill depend on both the sensor and the wave. An ideal se
a voltage-time curve identical to the amplitude-time curve
point where the sensor is located. Although no sensor app
for certain types of acoustic waves, laser interferometry w
available sensors operate quite well for specified types of
ranges of parameters.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
129/469
-
8/15/2019 AET-2006 Reading 3A
130/469
harlie Chong/ Fion Zhang
Magnetostriction (cf. electrostriction) is a property of ferromagthem to change their shape or dimensions during the process of magnetization. Th
magnetization due to the applied magnetic field changes the magnetostrictive strai
value, λ. The effect was first identified in 1842 by James Joule when observing a s
This effect causes energy loss due to frictional heating in susceptible ferromagnetiresponsible for the low-pitched humming sound that can be heard coming from tran
oscillating AC currents, which produce a changing magnetic field
-
8/15/2019 AET-2006 Reading 3A
131/469
Magnetostrictive ultrasonic transducers utilize the principle of m“ferromagnetic” materials which include iron, nickel and cobalt as well as many alloys of these th
waves in a liquid. The magnetostrictive principle was described in some detail in a previous blog
magnetostriction and ferromagnetism can be found in a “reader friendly” paper entitled Magneto
written by Geoffrey P. McKnight of the UCLA Active Materials Lab. Devices to create motion us
magnetostriction were first developed in the 1930’s and perhaps earlier although earlier work wa
1950’s, the technology expanded into the ultrasonic frequency range in response to a need for urobust than those using the (then) fragile crystal compositions that were used in piezoelectric ult
Some of the developmental work on magnetostrictive transducers in the 1950’s was done by Ale
Murdoch Laboratories, a predecessor of Blackstone Ultrasonics and Blackstone-NEY Ultrasonic
ceramic piezoelectric materials and advanced Langevin type piezoelectric ultrasonic transducers
magnetostrictive transducer was, probably, the most reliable and powerful ultrasonic transducer
ultrasonic manufacturers.
A magnetostrictive ultrasonic transducer consists, essentially, of series of laminations of a magn
active material attached directly to a vessel or tank which holds the liquid to be ultrasonically acteither around or in proximity to the laminations of magnetostrictive material to provide the oscilla
cause cause them to, in turn, vibrate. The basic construction is shown below.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
132/469
-
8/15/2019 AET-2006 Reading 3A
133/469
-
8/15/2019 AET-2006 Reading 3A
134/469
Initially, all piezoelectric devices were made from single c
useful of these were quartz, Rochelle salt, and ammonium
phosphate. Later, a class of materials known as ferroelect
materials which have a polarization even in the absence o
investigated and found useful. Ceramics made of ferroelea uniform direction of polarization similar to that found in a
crystal. It became possible to produce ferroelectric ceram
properties superior to piezoelectric single crystals with the
all acoustic emission sensors today are made from a varie
ceramics.
harlie Chong/ Fion Zhang http://www.acs
-
8/15/2019 AET-2006 Reading 3A
135/469
3.1.3. Size Effects
An infinitesimal piece of piezoelectric material with many d
electrodes totally embedded in a sample would come clos
sensor. However, when we scale up the piezoelectric to a
placing one set of electrodes on the outside surface of thedepart from that ideal. The physical size of the sensor res
effects, resonance and strain averaging. Both can becom
the physical dimensions of the sensor approach or exceed
the acoustic wave. Since the output of a piezoelectric crys
the strain (and proportional to the average strain for a crysdimensions), the maximum output of a sensor occurs at it
frequencies.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
136/469
The fundamental resonant frequency of a plate
occurs when the plate thickness is one-half
wave length as shown in Fig.2-7b. The reflected
wave is in phase with the incident wave at the
surface and the strains add. If the frequency isincreased until there is one full wavelength in
the crystal, there will again be strain re-
enforcement due to the reflected waves.
However, we can see in Fig. 2-7d that while the
strain level may be very great at this frequency,2υo, the average strain over the crystal exactlycancels so that the output of the sensor is zero.
Increasing the frequency to 3υo, we see 1.5wavelengths in the crystal and re-enforcement
again occurring. The average strain over two-
thirds of the crystal now cancels out, but theaverage strain over the last third reaches a
maximum.
harlie Chong/ Fion Zhang http://www.acs
Figure 2-7. Rimmersed in water.
clarity. (b) Strain in a
a function of frequen
Strain in a plate onethick.
-
8/15/2019 AET-2006 Reading 3A
137/469
The result is that a piezoelectric sensor will have a maxim
whenever the thickness, d, is:
d = (2n - l)λ/2 (3.1)
and no output when
d = nλ (3.2)
harlie Chong/ Fion Zhang
Thus a sensor can be operated either
at its fundamental frequency,υo (1/2,1/4λ?), or its harmonic frequencies
nυo where n is odd.
Comments: Why odd?
-
8/15/2019 AET-2006 Reading 3A
138/469
-
8/15/2019 AET-2006 Reading 3A
139/469
Standing Wave FormationThe animation below depicts two waves moving through a medium in opposite directions. The blue wave is moving to the right and the green w
situation in which two waves meet while moving along the same medium, interference occurs. The blue wave and the green wave interfere to fo
The resultant in the animation below is shown in black. The resultant is merely the result of the two individual waves - the blue wave and the gr
principle of superposition. The result of the interference of the two waves above is a new wave pattern known as a standing wave pattern. Stan
identical frequency interfere with one another while traveling opposite directions along the same medium. Standing wave patterns are characte
which undergo no displacement. These points of no displacement are called nodes (nodes can be remembered as points of no desplacement)
animation above. The nodes are always located at the same location along the medium, giving the entire pattern an appearance of standing st
inspection of the above animation will reveal that the nodes are the result of the destructive interference of the two interfering waves. At all timegreen wave interfere to completely destroy each other, thus producing a node.
harlie Chong/ Fion Zhang http://ww
-
8/15/2019 AET-2006 Reading 3A
140/469
First Harmonic
Standing Waves on a String A standing wave pattern is a pattern which results from the interference of two or more waves along the same medium. All stan
positions along the medium which are standing still. Such positions are referred to as nodal positions or nodes. Nodes occur a
one wave is displaced upward the same amount that a second wave is displaced downward. This form of interference is know
point of "no displacement." A node is a point of no displacement. Standing wave patterns are also characterized by antinodal p
vibrate back and forth between a maximum upward displacement to a maximum downward displacement. Antinodes are locatetwo interfering waves are always undergoing constructive interference. Standing wave patterns are always characterized by an
There are a variety of patterns which could be produced by vibrations within a string, slinky, or rope. Each pattern corresponds
frequency and is known as a harmonic. The lowest possible frequency at which a string could vibrate to form a standing wave
or the first harmonic. An animation of a string vibrating with the first harmonic is shown below.The frequency associated with e
which waves move through the medium and the wavelength of the medium. The speed at which waves move through a mediu
medium (tension of the string, thickness of the string, material composition of the string, etc.). The wavelength of the harmonic
the harmonic number (first, second, third, etc.). Variations in either the properties of the medium or the length of the medium wthe string will vibrate.
harlie Chong/ Fion Zhang http://www
-
8/15/2019 AET-2006 Reading 3A
141/469
-
8/15/2019 AET-2006 Reading 3A
142/469
There are a variety of other low energy vibrational pattern
established in the string. For guitar strings, each pattern i
some basic traits:
There is an alternating pattern of nodes and antinodes
There are either a half-number or a whole number of w
pattern established on the string.
Nodal positions (points of no displacement) are establ
the string where the string is clamped down in a fixed
One pattern is related to the next pattern by the additio
one or more nodes (and antinodes).
harlie Chong/ Fion Zhang http://wwwhttp://www.physicsclassroom.co
-
8/15/2019 AET-2006 Reading 3A
143/469
Harmonic:
harlie Chong/ Fion Zhang http://www.physicsclassroom.co
-
8/15/2019 AET-2006 Reading 3A
144/469
Fundamental and HarmonicsThe lowest resonant frequency of a vibrating object is called its fundamental freque
have more than one resonant frequency and those used in musical instruments typ
the fundamental. A harmonic is defined as an integer (whole number) multiple of th
Vibrating strings, open cylindrical air columns, and conical air columns will vibrate
fundamental. Cylinders with one end closed will vibrate with only odd harmonics ofmembranes typically produce vibrations at harmonics, but also have some resonan
harmonics. It is for this class of vibrators that the term overtone becomes useful - t
non-harmonic overtones.
harlie Chong/ Fion Zhang http://hyperp
-
8/15/2019 AET-2006 Reading 3A
145/469
Fundamental and Harmonics with one EnCylinders with one end closed will vibrate with only odd harmoni
Vibrating membranes typically produce vibrations at harmonics,
harlie Chong/ Fion Zhang http://hyperp
-
8/15/2019 AET-2006 Reading 3A
146/469
harlie Chong/ Fion Zhang http://hyperp
Fundamental and Harmonics with one EnCylinders with one end closed will vibrate with only odd harmoni
Vibrating membranes typically produce vibrations at harmonics,
-
8/15/2019 AET-2006 Reading 3A
147/469
Open End Standing Wave Patterns
harlie Chong/ Fion Zhang http://www.physics.louisville.ed
-
8/15/2019 AET-2006 Reading 3A
148/469
Non-Harmonic
harlie Chong/ Fion Zhang http://www.physics.louisville.ed
-
8/15/2019 AET-2006 Reading 3A
149/469
Harmonic: If both ends of the string are fixed the pulses t
left will interfere, left below. If, instead of pulses, we imag
of the same amplitude, frequency and wavelength, but tra
directions we obtain the standing wave pattern, below righ
harlie Chong/ Fion Zhang http://www.physics.louisville.ed
-
8/15/2019 AET-2006 Reading 3A
150/469
The Q of the transducer depends only on the number of re
sensor; therefore, the Q is independent of the harmonic a
operating, as long as the material of the sensor does not s
dependent attenuation. Also, the sensor will always have
frequencies below the fundamental frequency, υo. At frequabout 3υo/4, the resonance will have no effect and the outessentially independent of frequency. In reality, a materia
in one dimension without producing strains in other directi
Fig. 3-1a. Many acoustic emission sensors use this to get
response, using a piezoelectric element in the shape of a wave with a vertical displacement on the cylinder face wil
resonance. This resonance will then give a large output s
sensor is a very sensitive detector of acoustic emission. H
not be used to measure the frequency spectrum nor the a
the acoustic wave since this cross coupling of vibration m
distorted representation of the wave
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
151/469
Figure 3-1. (a) Deformation of a material
showing multiaxial strain resulting from
uniaxial force. (b) spectral response of an
acoustic emission sensor
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
152/469
-
8/15/2019 AET-2006 Reading 3A
153/469
In addition to resonance effects, there is another importan
averaging by a sensor. Fig.3-2a shows a block with a sen
the sensor is excited with a compressional wave moving p
surface, the entire sensor face will move in phase. Exclud
average strain in the sensor will be independent of frequea Rayleigh wave which is traveling parallel to the sensor f
particle motion perpendicular to the sensor face. In this ca
distribution in the transducer will vary as a function of dista
wave. Fig.3-2b shows the strain variation where the diame
less thanλ
/2. Here the output is still proportional to the awave. In Fig. 3-2c, the diameter of the sensor is larger tha
In this case, for every complete wave-length under the se
averages to zero. Only the extra fraction of the wavelengt
contributes to its output.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
154/469
Figure 3-2. (a) Sample block with a sensor mounted on o
compressional wave is shown traveling perpendicular to t
a Rayleigh wave travelling parallel to the face. (b) Instanta
block’s surface from Rayleigh wave with wavelength much
sensor diameter. (c) Strain on surface from Rayleigh waveshorter than sensor diameter. (d) Sensor output as a func
Rayleigh waves with equal amplitudes.
harlie Chong/ Fion Zhang
λ
-
8/15/2019 AET-2006 Reading 3A
155/469
This averaging essentially reduces the effective area of th
higher the frequency, the greater the reduction. Additiona
frequencies, depending on the shape of the sensor and th
of the sample, the total strain averages to zero. These effe
Fig. 3-2d where the response for this type of surface wavesensor with a flat frequency response to compressional w
to its face. The high frequency response of such a sensor
drastically on the angle of incidence with which the wave s
This averaging effect depends on the acoustic wave-lengt
Therefore, the sensor response not only is going to vary wangle of incidence, but also it is going to vary when used w
materials. The best answer to this problem of averaging s
of sound over the surface of the sensor is to make the sen
For steel, a 3mm diameter sensor should work reasonably
kHz. The inevitable tradeoff is that the smaller sensor has
capacitance and thus, as will be discussed later, a reducesensitivity.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
156/469
3.1.4. Couplants
We have assumed to this point that the sensor has simply
surface of the material containing the acoustic wave. Whe
found that the sensor produces a very weak signal. If a th
placed between the sensor and the surface, a much largeThis fluid acts a couplant that ensures good contact betwe
a microscopic level. The use of some type of couplant is a
the detection of low level acoustic signals. Physically, this
by looking at the acoustic wave as a pressure wave trans
surfaces in contact. On a microscopic scale, the surfaces the material are quite rough; they actually touch in only a
are in contact. Stress is force per unit area and the actual
force is very small. If the microscopic gaps are filled with a
will be uniformly transferred between the surfaces. For a s
variable strain component parallel to the surfaces, very litt
transferred between the surfaces because of the few poinIn this case, filling the gaps with a low viscosity liquid will
such a liquid will not support a shear stress. However, a h
or a solid will help transmit the parallel strain between sur
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
157/469
Although the terms bond and couplant have been used in
many AE reports (including the author’s), their meanings s
Strictly, a couplant is any material which aids the transmit
waves between two surfaces, while a bond is a couplant w
holds the sensor to the surface. For example, water is a cepoxy resin is a bond. Many problems have come about f
an inapplicable way. If a rigid bond is used to attach a sen
which elastically deforms during the test, the normal resul
and poor or no sensitivity to the acoustic wave.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
158/469
Similarly, in an experiment where the temperature is chan
the use of a rigid bonding material can lead to broken bon
differential thermal expansion between the sensor and the
agents, then, must be chosen carefully, considering the co
materials under the test conditions. Usually, if the bond won, it will be an adequate couplant. For a compressional w
act as a couplant. A highly viscous fluid will transfer some
the boundary which may or may not be an advantage. In o
author tested a large number of couplants with compressi
all couplants showed an increase in the signal strength ov30+2 dB. The variation was little more than the uncertaint
measurement. Practically, a couplant can be a thin layer o
that wets both surfaces. The sensor should be held again
some pressure furnished by magnets, springs, tape, rubbe
secret is to use as thin a layer as possible. If a rigid bond
be minimal differential expansion between the two surface
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
159/469
A flexible bond can also be used. Over the years, the auth
excellent results with GE Silicone II sealant. It is available
household glue which will set up within about 12 hours wh
one inch diameter sensor and a metal or plastic surface. I
glue, a perpendicular sinusoidal force of about 100 G prodfailures. It is quite flexible, allowing for sensor removal fro
knife blade or wood chisel slid between the sensor and th
2, a few commonly used couplants are listed along with th
range in which they can be used.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
160/469
Table 2. Some Common Acoustic Emission Couplants
Approximate Tempera-ture Ranges
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
161/469
3.1.5. Temperature Effects
The temperature dependence of piezoelectricity is compli
discussion of its effect on sensors is beyond the scope of
However, certain effects can lead to problems when a sen
different temperatures. First, ferroelectric materials, such ceramics, have a Curie temperature above which the mat
another, and usually non-ferroelectric, phase. Taking a fe
through the Curie temperature will remove the polarization
piezoelectric properties of the sensor, and may shatter the
Ferroelectric sensors will usually work well up to temperatthe Curie temperature, if the other materials in the sensor
temperature. The Curie temperatures of PZT ceramics lie
and 400°C.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
162/469
Ferroelectric ceramics are poly-crystalline. Each crystallite
or more ferroelectric domains, i.e. regions where the spon
polarization is all in one direction. This polarization exists
directions in the crystal structure. When the ceramic is po
are aligned, as closely as the crystal orientation allows, topolarization. Because of the random orientation of the cry
number will have several possible orientations approxima
polarization in the ceramic. Small strains may be enough
to change orientation. Such a flip of a domain may cause
in the polarization of the sensor. However, this change is magnitude as the change caused by a small acoustic wav
impossible to distinguish an electric signal caused by a fli
one caused by acoustic emission.
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
163/469
-
8/15/2019 AET-2006 Reading 3A
164/469
3.1.6. Sensor Sensitivity - Effects of Cables
The sensitivity of a sensor is governed by the intrinsic sen
piezoelectric material, the dimensions of the piezoelectric
design and materials used in its case. Practically, one rec
manufacturer a measured response curve to a standard scapacitance of the sensor. This curve is often presented a
independent of the measurement technique. However, the
sensor will always depend, in part, on the equipment with
harlie Chong/ Fion Zhang
-
8/15/2019 AET-2006 Reading 3A
165/469
The open circuit voltage produced by a sensor is a proper
piezoelectric element and is:
Vo(S) = Q(S)/Co (3.3)
where Q is the charge produced by a strain S, and Co is tthe sensor. When connected to a preamplifier, the actual
input resistor of the preamplifier (given large input resistan
V(S) = Q(S)/(Co + Cc + CI)