Documente Academic
Documente Profesional
Documente Cultură
TESTING
TRAINING
HANDBOOK
Abstract
This is the official training handbook of my course Ultrasonic Level
1 training presented online. It covers all the training outlines with
the maximum information that the students need to understand the
course and to be well prepared for the official UT-L1 exam. Find my
UT-L1 course for only $5 here: https://www.udemy.com/ultrasonic-
testing/?couponCode=UTReaders
Angular incidence
Reflection ....................................................................................................................................56
Refraction ..................................................................................................................................57
Discontinuity variations
Sizing methods ........................................................................................................................164
Scanning methods ......................................................................................................................164
Evaluation of small discontinuities: the DGS methods ...........................................................167
Shape ...........................................................................................................................................173
Non-destructive Testing
The field of Nondestructive Testing (NDT) is a very broad, that plays
a critical role in assuring that structural components and systems
perform their function in a reliable and cost effective fashion. NDT
technicians and engineers define and implement tests that locate and
characterize material conditions and flaws that might otherwise
cause serious accidents such as, planes to crash, reactors to fail, trains
to derail, pipelines to burst, and a variety of troubling events.
These tests are performed in a manner that does not affect the future
usefulness of the object or material. In other words, NDT allows
parts and materials to be inspected and evaluated without damaging
them. Because it allows inspection without interfering with a
product's final use, NDT provides an excellent balance between
quality control and cost-effectiveness.
Nondestructive Evaluation
Nondestructive Evaluation (NDE) is a term that is often used
interchangeably with NDT. However, technically, NDE is used to
describe measurements that are more quantitative in nature. For
example, a NDE method would not only locate a defect, but it would
also be used to measure something about that defect such as its size,
shape, and orientation. NDE may be used to determine material
1
properties such as fracture toughness, ductility, conductivity and
other physical characteristics.
Uses of NDE
Flaw Detection and Evaluation
Leak Detection, Location Determination
Dimensional Measurements
Structure and Microstructure Characterization
Estimation of Mechanical and Physical Properties
Stress (Strain) and Dynamic Response Measurements
Material Sorting and Chemical Composition Determination
2
The first use of sound waves in water was for the detection of
submarines and was developed during World War l.
As electromagnetic waves cannot penetrate water, use of acoustic
waves to penetrate seawater, found importance and Prof. Paul Langevin
produced the first underwater transducer.
3
In the recent past, surface wave devices have evoked some interest.
They are used as delay lines, complex coded waveforms and wide band
signals for missile guidance and airport flight control. Bombs in Pacific
Ocean were probably the starting point for the ultrasonic technique.
The famous Krautkramer brothers of Germany had in the meantime
perfected the technique to such an extent that reliable ultrasonic testing
method was available.
Technique they employed was based on valves, which was
subsequently changed transistors and ultimately to use of
microprocessors. Most of the present day efforts in improving the
technique are centered on employing the microprocessor. Design of
Superior probes, and application of the technique to tough ceramics.
Defect type Eddy Current Magnetic Particle Ultrasonic Radiography Resonant Acoustic
Cracks/chips/porosity/voids
Missed processes/operations
Material property
Structurally significant
Product lot variations
4
Defect Location
Surface (external)
Eddy Current Magnetic Particle Ultrasonic Radiography Resonant Acoustic
Internal
Brazing/bonding/welding
Speed/Training/Cost
Part troughput
Training requirements
Overall inspection costs
Automation Capacity
Quantitative results
Automation requirements NA
Automation cost NA
RESONANT INSPECTION
Advantages Disadvantages
Whole part test for internal and Not diagnostic - does not indicate
external flaws where flaw is, just there is one
5
Fast as 1 part per second Materials that resonate only - metal,
composites and ceramic parts
No part preparation required Large parts (> 60lbs) difficult to test
Objective pass/fail result Significant lot to lot variations can
mask defect detection
No consumables expenses
Easily automated
Permanent record capability
Easily finds first n number of natural
frequencies for NVH applications
Best for high volume quality
inspection
Designed to be on the plant floor
ULTRASONICS
Measures thickness, velocity or detects internal defects and variations,
such as cracks, lack of fusion, delaminations and lack of bond.
Applications include wrought metals, welds, brazed joints, adhesive or
bonded joints, non-metallic materials and in-service parts.
Advantages Disadvantages
Most sensitive to cracks Couplant required
Immediate results Complex, or small parts may be
difficult to check
Automation possible Reference standards required
Permanent record capability Trained operators for manual
inspections
Portable Special probes
6
High penetration capability Surface condition
RADIOGRAPHY
Measures or detects, internal defects and variations, porosity,
inclusions, cracks, lack of fusion, corrosion, geometry variation,
density changes, misassembled and misaligned parts.
Advantages Disadvantages
Permanent records Radiation hazard
Portable Expensive
Geometry variation does not affect Trained operators needed
direction of radiation beam
Linear defect may be missed
Depth of defect not indicated
Access needed to at least two sides of
the part
EDDY CURRENT
Advantages Disadvantages
High speed Conductive material only
Low cost Shallow depth of penetration
Permanent record capability
7
No couplant required
No probe contact required
LIQUID PENETRANT
Advantages Disadvantages
Low cost Defect must be open to the surface
Portable Parts must be cleaned before and after
testing
Indications may be further examined Surface films, such as coatings, scale,
and smeared metal may visually mask
defects
MAGNETIC PARTICLE
Advantages Disadvantages
Low cost Ferromagnetic materials only
Portable Alignment of magnetic field is critical
8
Subsurface defects Demagnetization required after the test
Surface coatings can mask defects
Pre and post cleaning necessary
Messy
INFRARED THERMOGRAPHY
Advantages Disadvantages
Permanent record or thermal picture Expensive
Remote sensing Reference standards required
Portable Poor resolution on thick sections
9
2. Fundamental Properties of Ultrasound
10
Figure 2: A Wave Illustration
11
the reverse piezo-electric effect. The first people to observe the
piezo-electric effect were the Curie brothers who observed it in
quartz crystals.
Figure 3
The wavelength () is the distance of one complete wave. The
amplitude of a wave is the maximum displacement of any part of the
wave from its equilibrium position. The time it takes for a wave to
travel a distance of one wavelength is called the period (T). The
frequency (f) is the number of oscillations or cycles that occur during
12
a given period of time. Frequency is usually measured in cycles per
second, or Hertz. The period (T) is usually measured in seconds. The
frequency (f) and the period (T) are related by the following
expression:
Period (T) = 1/ Frequency (f)
One of the properties of a wave is that the velocity of a wave is
related with its frequency and wavelength by the following equation.
In 1932 Debye and Sears discovered in the USA and Lucas and
Biquard discovered in France that transparent media diffract light
when an ultra- sound wave is sent through them. This effect is a
consequence of a periodical variation of the refractive index, which
in turn is a consequence of a local periodical pressure change caused
by the ultrasound wave.
13
Figure 4: Sound diffraction
A thin slit, lit up by the lamp La serves as a source of light. The lens L1
is placed in the distance of its focal length from the gap and thus
produces a broad beam of parallel light. The light then penetrates a
transparent medium (gas, liquid or solid) in which an ultrasonic wave
transducer Q, located perpendicularly to the direction of the incidence
light beam, produces elastic waves. For experiments with liquids or
gases a container with plane parallel glass walls is required. The second
lens L2 projects a real image S0 of the gap S on a screen. If the
ultrasound wave is excited, several orders of the spectrum of the lamp
La can be seen on both sides of S0. By introducing alter F into the path
of the light beam we produce monochromatic light and obtain only one
interference strip of each order near S0.
To understand this phenomenon we must assume that the local
periodical pressure changes of the elastic wave create local changes in
the refraction index of the medium. The surfaces of equal phase S(x; y;
z) of the light wave are then no longer plains (S(x; y; z) = k0x for light
14
propagation in the x-direction.) but they become a sine function with
the same period as the ultrasonic wave. This arises from the fact that in
an area with a higher refractive index n the light wave travel slower than
in an area with a lower refractive index (c = c0n ).
In solids, sound waves can propagate in four principal modes that are
based on the way the particles oscillate. Sound can propagate as
longitudinal waves, shear waves, surface waves, and in thin materials
as plate waves. Longitudinal and shear waves are the two modes of
propagation most widely used in ultrasonic testing. The particle
15
movement responsible for the propagation of longitudinal and shear
waves is illustrated in the figure.
16
Figure 6: Longitudinal wave representation
In the transverse or shear waves, particles oscillate at a right angle or
transverse to the direction of propagation. Shear waves require an
acoustically solid material for effective propagation, and therefore, are
not effectively propagated in materials such as liquids or gasses. Shear
waves are relatively weak when compared to longitudinal waves. In
fact, shear waves are usually generated in materials using some of the
energy from longitudinal waves.
17
2.5. Ultrasound uses
Ultrasonic testing (UT) has been practiced for many decades. Initial
rapid developments in instrumentation spurred by the technological
advances from the 1950's continue today. Through the 1980's and
continuing through the present, computers have provided technicians
with smaller and more rugged instruments with greater capabilities.
18
technicians have little effect on the technician's ability to view the
screen. Screens can be adjusted for brightness, contrast, and on some
instruments even the color of the screen and signal can be selected.
Transducers can be programmed with predetermined instrument
settings. The operator only has to connect the transducer and the
instrument will set variables such as frequency and probe drive.
19
descriptions of NDE performance, such as the probability of detection
(POD), have become an integral part of statistical risk assessment.
Measurement procedures initially developed for metals have been
extended to engineered materials such as composites, where anisotropy
and inhomogeneity have become important issues. The rapid advances
in digitization and computing capabilities have totally changed the faces
of many instruments and the type of algorithms that are used in
processing the resulting data. High-resolution imaging systems and
multiple measurement modalities for characterizing a flaw have
emerged. Interest is increasing not only in detecting, characterizing, and
sizing defects, but also in characterizing the materials. Goals range from
the determination of fundamental microstructural characteristics such
as grain size, porosity, and texture (preferred grain orientation), to
material properties related to such failure mechanisms as fatigue, creep,
and fracture toughness. As technology continues to advance,
applications of ultrasound also advance. The high-resolution imaging
systems in the laboratory today will be tools of the technician
tomorrow.
Looking to the future, those in the field of NDE see an exciting new set
of opportunities. The defense and nuclear power industries have played
a major role in the emergence of NDE. Increasing global competition
has led to dramatic changes in product development and business
cycles. At the same time, aging infrastructure, from roads to buildings
and aircraft, present a new set of measurement and monitoring
challenges for engineers as well as technicians.
20
Among the new applications of NDE spawned by these changes is the
increased emphasis on the use of NDE to improve the productivity of
manufacturing processes. Quantitative nondestructive evaluation
(QNDE) both increases the amount of information about failure modes
and the speed with which information can be obtained and facilitates
the development of in-line measurements for process control.
The phrase, "you cannot inspect in quality, you must build it in,"
exemplifies the industry's focus on avoiding the formation of flaws.
Nevertheless, manufacturing flaws will never be completely eliminated
and material damage will continue to occur in-service so continual
development of flaw detection and characterization techniques is
necessary.
Advanced simulation tools that are designed for inspectability and their
integration into quantitative strategies for life management will
contribute to increase the number and types of engineering applications
of NDE. With growth in engineering applications for NDE, there will
be a need to expand the knowledge base of technicians performing the
evaluations. Advanced simulation tools used in the design for
inspectability may be used to provide technical students with a greater
understanding of sound behavior in materials. UTSIM, developed at
Iowa State University, provides a glimpse into what may be used in the
technical classroom as an interactive laboratory tool.
21
NDE, this trend will drive the emphasis on standards, enhanced
educational offerings, and simulations that can be communicated
electronically. The coming years will be exciting as NDE will continue
to emerge as a full-fledged engineering discipline.
22
3. Generation of Ultrasonic Waves
3.1. Piezoelectric crystal types and characteristics
23
exhibits strain, i.e. the dimensions of the crystal change. When the
direction of the applied electric field is reversed, the direction of the
resulting strain is reversed. This is called the converse piezoelectric
effect.
24
The direction in which tension or compression develops polarization
parallel to the strain is called the piezoelectric axis.
In quartz, this axis is known as the "X-axis", and in poled ceramic
materials such as PZT the piezoelectric axis is referred to as the "Z-
axis". From different combinations of the direction of the applied field
and orientation of the crystal it is possible to produce various stresses
and strains in the crystal. For example, an electric field applied
perpendicular to the piezoelectric axis will produce elongation along
the axis as shown in Figure 9. If, however, the electric field is applied
parallel to the piezoelectric axis, a shear motion is induced. This type
motion is shown in Figure 10.
25
coupling, damping and face plates). Construction of three types of
transducer assemblies is shown in Fig. 11.
Figure 11a shows a straight beam contact transducer. These use thin
wear plates to prevent crystal breakage and to protect the front
electrode, which provides internal grounding. Facings of ceramic,
metal, plastic and rubber have been used. Applications include tests of
rough surfaces and electrical nonconductors. Angle beam transducers
direct the ultrasonic beam away from normal incidence toward selected
areas within a test object using a wedge between the crystal and the test
object (see Fig. 11b). These wedges or shoes are usually made from
plastic materials.
26
Figure 11: Typical ultrasonic testing transducers: (a) straight beam contact
unit; (b) angle beam transducer; (c) immersion transducer
27
resolving power. The Q of a transducer is given by the following
equation:
Fr
=
F2 F1
Where F2 - F1 = bandwidth
Fr = resonant frequency of the element
F2 = frequency above Fr where amplitude = 0.707 Fr amplitude
F1 = frequency below Fr where amplitude = 0.707 Fr amplitude
Note: In most nondestructive testing applications, the Q of the search
unit will vary from 1 to 10.
Sensitivity is the ability of the search unit to detect reflections or echoes
from small defects or flaws. Search unit sensitivity is directly
proportional to the product of its efficiency as a transmitter and its
efficiency as a receiver.
The acoustic impedance of a transducer is the product of its density and
the velocity of sound within it. Its resolving power includes the ability
to separate reflections from two closely spaced flaws or reflectors. For
example, in contact testing, the transducer must have good resolution to
separate the front surface pulse, initial pulse, or main bang from the
near-surface defect reflections. The ability of a transducer to resolve or
detect near-surface defects is related to its pulse width or length. The
transducer must stop vibrating or ringing after it is shocked in order
to resolve or see small near-surface flaws. Resolving power is also
related to crystal dampening and bandwidth. Ideally, we would like to
28
reduce crystal ringing or bandwidth without adversely affecting
sensitivity.In practice, compromises must be made to achieve test goals.
For most search units, the product of bandwidth and sensitivity is a
constant.
As bandwidth increases, sensitivity decreases; as bandwidth decreases,
sensitivity increases. By knowing the sensitivity bandwidth product and
bandwidth of the transducer, its sensitivity can be calculated.
Piezoelectric elements are usually coated with gold or silver on their
front and back surfaces to form electrodes. Figure 12 shows a few of
the available electrode configurations for flat crystals and some typical
transducers.
Small wire leads are then attached to the electrode surfaces by welding,
soldering, or cementing them in place with conductive epoxy. Good
electrical insulation must be used in search unit construction because
high-voltage pulses (1002000 VDC) are applied to the electrodes for
10ms or less. The high voltage pulses are transmitted from the pulse
circuit of the electronics unit to the search unit through a coaxial cable
to an ultrahigh frequency (UHF) or baby N connector (BNC)-type
coaxial connector on the search unit.
29
Figure 12 (a, b) Crystal electrode arrangements and (c) factory assembled
transducers. Electrode diagrams courtesy of Valpey-Fisher Corp.;
transducer photo courtesy of Krautkramer Branson, Inc., manufacturer of
Ultrasonic Nondestructive Testing Equipment.
30
The backing material of the search unit serves two purposes. First, it
controls crystal dampening and therefore bandwidth, and second, it
attenuates the energy on the back side of the crystal so that unwanted
reflections will not be received from the back side of the transducer.
Transducers can be air backed or backed by fibrous or cellular plastic
material that effectively attenuates sound on the back side of the crystal
while reducing its undampened bandwidth from as much as 20,000 to a
range of 1 to 10, depending on the application.
Some manufacturers embed a hardened surface ring and coat the front
sides of their transducers with an aluminum oxide coating to reduce
transducer wear. In contact testing, plastic wedges are used to position
crystals at an angle for shear wave testing. Delay tips are also attached
to contact-type transducers for testing thin materials. While increasing
the versatility of the transducer, the Lucite wedges and shoes decrease
transducer sensitivity. Lucite shoes are contoured wedges that are
shaped to the contour of the test surface.
In high-speed production lines, the Lucite shoes may actually ride on
the test surface, become worn, and have to be periodically replaced.
Acoustic lenses can also be attached to the front surface of the
transducer to act as a lens for focusing the sound beam. As the radius
of curvature of a curved lens increases, the focal length of the lens
increases. Ideally, the acoustic impedance of the focusing lens is
between that of the transducer and material under test. Transducers can
be spherically or spot focused and cylindrically focused. Spherically
31
(spot or point) focused transducers are used when improved resolution
to small flaws is required for the test material. Cylindrically focused
transducers are typically used for pipe and tubing inspections.
Acoustic lenses with cylindrical curvatures focus sound energy into
cylindrical surfaces normally or at right angles. Focusing lenses
effectively shorten the Fresnel zone by shifting the transition distance
N toward the transducer. Backing material variations, lens
misalignment, or lens porosity can result in the propagation of
nonsymmetrical beams when using focused transducers.
Focusing can increase the echo amplitude from small flaws near the
focal point. This technique is used to obtain better near-surface
resolution without increasing transducer frequency. The disadvantage
is that the sensitivity to defects in the far field is greatly decreased.
Cylindrical focusing is used to shape the sonic wave front to conform
to part geometry. This produces a clean front surface reflection and
better resolution of near-surface flaws in pipe and tubing. This type of
focusing is also called contoured focusing. The effects of crystal
diameter, focal distance, and water path length are given in Table1.
(Test conditions a)
32
0.75 M 4.0 0.09 1.9 0.50 0.53
0.75 L 5.2 0.15 1.0 1.00 1.07
0.50 S 1.7 0.04 0.9 0.18 0.19
0.50 M 2.7 0.06 1.1 0.38 0.41
0.50 L 3.7 0.12 1.9 0.44 0.46
0.37 S 1.3 0.03 0.8 0.13 0.13
0.37 M 2.2 0.05 1.2 0.25 0.25
0.37 L 3.0 0.08 1.4 0.38 0.41
0.25 S 0.8 0.02 0.4 0.09 0.10
0.25 M 1.5 0.05 0.8 0.16 0.18
0.25 L 2.2 0.07 0.9 0.31 0.33
0.19 S 0.6 0.02 0.4 0.05 0.05
0.19 M 1.0 0.04 0.6 0.10 0.10
a
(1) Water path may vary 25% without adversely affecting test sensitivity. (2) Data applicable
to frequencies of 10MHz or above.
b
S = short, M = medium, L = long.
33
Crystal Frequency (MHz) Crystal Thickness (in.)
0.20 0.500
0.50 0.200
1.00 0.100
2.25 0.50
5.00 0.20
10.00 0.10
15.00 0.007
20.00 0.006
25.00 0.005
(*) Crystal diameters vary widely, typically in the range 0.51.5 in. for 1 MHz to 0.1250.375
in. for 25MHz.
34
mechanically deformed; their front surfaces should be protected to
prevent scratches and minimize wear. Units should be stored at normal
operating temperatures and not subjected to excessively harsh*
chemical environments. Tank-type immersion units should be
immersed or loaded before they are pulsed, and the transmitter should
be turned off before they are removed from the immersion tank.
35
Another block (described in B.S. 3923: Part 3: 1972) used for the
determination of resolution of flaw detectors using either normal beam
or angle beam probes is shown in Figure 5.26. With this block the
resolution is determined by the minimum distance apart that flaws can
be indicated clearly and separately. In use the probe is placed on the
center line of the block over the change in radius from one step to the
next. Its position is adjusted so that echoes from the two radii are of the
same height and approximately 1/2 full screen height. The steps are said
to be resolved when their echoes are clearly separated at half maximum
echo height or less.
36
3.4. Search unit types
Contact transducers are used for direct contact inspections, and are
generally hand manipulated. They have elements protected in a rugged
casing to withstand sliding contact with a variety of materials. These
transducers have an ergonomic design so that they are easy to grip and
move along a surface. They often have replaceable wear plates to
lengthen their useful life. Coupling materials of water, grease, oils, or
commercial materials are used to remove the air gap between the
transducer and the component being inspected.
37
Contact transducers are available in a variety of configurations to
improve their usefulness for a variety of applications. The flat contact
transducer shown above is used in normal beam inspections of
relatively flat surfaces, and where near surface resolution is not critical.
If the surface is curved, a shoe that matches the curvature of the part
may need to be added to the face of the transducer. If near surface
resolution is important or if an angle beam inspection is needed, one of
the special contact transducers described below might be used.
38
Figure 16: Immersion probes
39
surface defects. The two elements are angled towards each other to
create a crossed-beam sound path in the test material.
40
thickness gauging of thin materials and delamination checks in
composite materials. They are also useful in high-temperature
measurement applications since the delay line provides some insulation
to the piezoelectric element from the heat.
41
Figure 18: Angle beam probe
Paint brush transducers are used to scan wide areas. These long and
narrow transducers are made up of an array of small crystals that are
carefully matched to minimize variations in performance and maintain
uniform sensitivity over the entire area of the transducer. Paint brush
transducers make it possible to scan a larger area more rapidly for
discontinuities. Smaller and more sensitive transducers are often then
required to further define the details of a discontinuity.
42
4. Test Material Characteristics
4.1. Velocity
The speed of sound is the distance traveled per unit of time by a sound
wave propagating through an elastic medium. In dry air at 20 C (68 F),
the speed of sound is 343.2 meters per second (1,126 ft/s). This is 1,236
kilometers per hour (667 kn; 768 mph), or about a kilometer in three
seconds or a mile in about five seconds.
In fluid dynamics, the speed of sound in a fluid medium (gas or liquid)
is used as a relative measure of speed itself. The speed of an object
divided by the speed of sound in the fluid is called the Mach number.
Objects moving at speeds greater than Mach1 are traveling at
supersonic speeds.
The speed of sound in an ideal gas is independent of frequency, but does
vary slightly with frequency in a real gas. It is proportional to the square
root of the absolute temperature, but is independent
of pressure or density for a given ideal gas. Sound speed in air varies
slightly with pressure only because air is not quite an ideal gas.
Although (in the case of gases only) the speed of sound is expressed in
terms of a ratio of both density and pressure, these quantities cancel in
ideal gases at any given temperature, composition, and heat capacity.
This leads to a velocity formula for ideal gases which includes only the
latter independent variables.
In common everyday speech, speed of sound refers to the speed of
sound waves in air. However, the speed of sound varies from substance
to substance. Sound travels faster in liquids and non-porous solids than
43
it does in air. It travels about 4.3 times as fast in water (1,484 m/s), and
nearly 15 times as fast in iron (5,120 m/s), as in air at 20 degrees
Celsius. Sound waves in solids are composed of compression waves
(just as in gases and liquids), but there is also a different type of sound
wave called a shear wave, which occurs only in solids. These different
types of waves in solids usually travel at different speeds, as exhibited
in seismology. The speed of a compression sound wave in solids is
determined by the medium's compressibility, shear modulus and
density. The speed of shear waves is determined only by the solid
material's shear modulus and density.
Dependence on the propreties of the medium
44
In gases, adiabatic compressibility is directly related to pressure
through the heat capacity ratio (adiabatic index), and pressure and
density are inversely related at a given temperature and composition,
thus making only the latter independent properties (temperature,
molecular composition, and heat capacity ratio) important. In
low molecular weight gases such as helium, sound propagates faster
compared to heavier gases such as xenon (for monatomic gases the
speed of sound is about 75% of the mean speed that molecules move in
the gas). For a given ideal gas the sound speed depends only on
its temperature. At a constant temperature, the ideal gas pressure has no
effect on the speed of sound, because pressure and density (also
proportional to pressure) have equal but opposite effects on the speed
of sound, and the two contributions cancel out exactly. In a similar way,
compression waves in solids depend both on compressibility and
densityjust as in liquidsbut in gases the density contributes to the
compressibility in such a way that some part of each attribute factors
out, leaving only a dependence on temperature, molecular weight, and
heat capacity ratio (see derivations below). Thus, for a single given gas
(where molecular weight does not change) and over a small temperature
range (where heat capacity is relatively constant), the speed of sound
becomes dependent on only the temperature of the gas.
In non-ideal gases, such as a van der Waals gas, the proportionality is
not exact, and there is a slight dependence of sound velocity on the gas
pressure.
45
Humidity has a small but measurable effect on sound speed (causing it
to increase by about 0.1%-0.6%),
because oxygen and nitrogen molecules of the air are replaced by lighter
molecules of water. This is a simple mixing effect.
46
The amplitude change of a decaying plane wave can be expressed as:
47
image at the top of the page. The number of decibels between two
adjacent signals is measured and this value is divided by the time
interval between them. This calculation produces an attenuation
coefficient in decibels per unit time. This value can be converted to
nepers/length by the following equation.
48
The specific acoustic impedance z is the ratio of sound pressure to
particle velocity, and z = *v, where is the density and v the speed
of sound. So for our duct with cross sectional area A, provided that the
wave is strictly one dimensional and travelling in one direction, the
acoustic volume flow is just U = Au
For this very special case, we define the characteristic acoustic
impedance Z0, where
Z0 = p/U = p/Au = z/A
So Z0 = v/A
49
5. Sound Beam Characteristics
5.1. Intensity variations
It is convenient to define the beam 'edge' as the point, across the beam,
where the intensity of sound has fallen to one half, or sometimes one
tenth of the intensity at the center of the beam. Whenever possible we
use the Far Field in ultrasonic testing, the near field usually being
accommodated within the Perspex shoe of the probe.
The Dead Zone is a zone where it is not possible to detect defects. Due
to imperfect damping of the crystals some waves will interfere with the
returning waves. This problem can be overcome by using twin crystals,
one transmitting, and one receiving. The higher the probe frequency the
shorter the Dead Zone.
50
Figure 20: Dead zone representation
51
5.2.3. Far zone
In the Far Field the beam diverges and the signal height from the same
size of defect decreases in relation to the distance in accordance with
the inverse square law.
It can be seen from the formula, (Figure 21) that by increasing the probe
diameter or increasing the frequency (shorter wavelength), the solid
angle of the beam will decrease.
52
5.3. Probe diameter and frequency effect
53
transducer that displays good range resolution. If there is a mismatch in
acoustic impedance more sound energy will be reflected forward into
the test material.
The frequency f depends on the thickness d of the transducer crystal and
the sound velocity v of the transducer material. Transducers with higher
frequency are more complicated to produce and they are therefore more
expensive.
Transducers can be used as transmitter or as sensor alternatively or as
both of them intermittently. Probes with only one transducer are also
called transceivers.
54
is based upon the idea of superposition that allows waves generated by
the active element to be in phase with the wave reverberating in the
matching layer. Not properly designed wear plates result in disruptions
in the wave front.
Piezo-composite materials usually have a structure called 1-3, where
the piezoelectric rods are embedded in a polymer matrix by a dice-and-
fill technique. The ceramic and the resin are chosen according to the
characteristics required for the composite material. The geometry of the
microstructure itself can be adapted. One of the characteristics of a 1-3
structure is that the percentage of the ceramic can be varied by
modifying the size of the rods and their spacing.
The designation simply codes the type of composite: piezoelectric
material hits the surface in one direction; the resin does so in three.
The height of the ceramic rods long compared to their lateral
dimensions favor their vibration according to the thickness mode to
the detriment of the radial mode. This results in improved electro-
acoustic efficiency that gives the sensor a high level of sensitivity and
a high signal/noise ratio. In addition, the natural damping of composite
materials allows a relative bandwidth of 60% to 90% to be obtained
while retaining a very good level of sensitivity. The optimum size of
the rods, not only their length, depends on frequency.
The mechanical properties of the polymers are used to enable the piezo-
composite materials to be shaped for focused transducers or such that
are adapted to the surface.
55
6. Angular Incidence
6.1. Reflection
Zi and Zt are the acoustic impedances for the incident and the
transmitting material, respectively. Clearly, the transmission coefficient
is always positive, the reflection coefficient, however, can be positive
or negative. A change of sign corresponds to a phase change of the
reflected wave.
56
6.2. Refraction
57
The refraction occurs when the angle of incidence at the boundary is
zero. In the case of normal incidence, part of the beam energy is
reflected directly backwards, and the remaining energy is transmitted
into the second medium without directional change. At any other angle
of incidence, the transmitted beam is deviated from the original
direction of the incident beam, either towards or away from the normal,
depending on the relative velocities of ultrasound in the two media.
58
another type, e.g. from a shear wave to a longitudinal wave, and from a
longitudinal wave to a surface wave. The angles of reflection and/or
refraction by mode conversion can be calculated from Snells law.
Figure 26 shows a simulation result for refraction and mode conversion,
calculated by a finite difference method. We can see that an incident
plane wave (longitudinal wave) of 10 in water is refracted at the
refraction angle of 43 in steel and simultaneously converted to shear
wave at refraction angle of 22.
59
6.4. Snell's Law
Snell's law (also known as the SnellDescartes law and the law of
refraction) is a formula used to describe the relationship between
the angles of incidence and refraction, when referring to light or
other waves passing through a boundary between two different isotropic
media, such as water, glass and air.
In optics, the law is used in ray tracing to compute the angles of
incidence or refraction, and in experimental optics to find the refractive
index of a material. The law is also satisfied in metamaterials, which
allow light to be bent "backward" at a negative angle of refraction with
a negative refractive index.
Although named after Dutch astronomer Willebrord Snellius (1580
1626), the law was first accurately described by the scientist Ibn Sahlat
the Baghdad court in 984. In the manuscript On Burning Mirrors and
Lenses, Sahl used the law to derive lens shapes that focus light with no
geometric aberrations.
Snell's law states that the ratio of the sines of the angles of incidence
and refraction is equivalent to the ratio of phase velocities in the two
media, or equivalent to the reciprocal of the ratio of the indices of
refraction:
With each as the angle measured from the normal of the boundary, as
the velocity of light in the respective medium (SI units are meters per
60
second, or m/s) and as the refractive index (which is unitless) of the
respective medium.
The law follows from Fermat's principle of least time, which in turn
follows from the propagation of light as waves.
Snell's law is used to determine the direction of light rays through
refractive media with varying indices of refraction. The indices of
refraction of the media, labeled , and so on, are used to represent
the factor by which a light ray's speed decreases when traveling through
a refractive medium, such as glass or water, as opposed to its velocity
in a vacuum.
As light passes the border between media, depending upon the relative
refractive indices of the two media, the light will either be refracted to
a lesser angle, or a greater one. These angles are measured with respect
to the normal line, represented perpendicular to the boundary. In the
case of light traveling from air into water, light would be refracted
towards the normal line, because the light is slowed down in water; light
traveling from water to air would refract away from the normal line.
Refraction between two surfaces is also referred to
as reversible because if all conditions were identical, the angles would
be the same for light propagating in the opposite direction.
Snell's law is generally true only for isotropic or specular media (such
as glass). In anisotropic media such as some crystals, birefringence may
split the refracted ray into two rays, the ordinary or o-ray which follows
Snell's law, and the other extraordinary or e-ray which may not be co-
planar with the incident ray.
61
When the light or other wave involved is monochromatic, that is, of a
single frequency, Snell's law can also be expressed in terms of a ratio
of wavelengths in the two media, 1and 2:
Example 1:
Suppose you wish to calculate the refracted angle within a material
when you know the incident angle (20), incident material velocity
(2330 m/s) and refracted material velocity (5960 m/s):
A1 = 20
V1 = 2330 m/s
A2 = we dont know!
V2 = 5960 m/s
Substituting these figures into the equation above gives us:
62
6.5. Critical angles
63
First Critical Angle
Before the angle of incidence reaches the first critical angle, both
longitudinal and shear waves exist in the part being inspected. The first
critical angle is said to have been reached when the longitudinal wave
no longer exists within the part, that is, when the longitudinal wave is
refracted to greater or equal than 90, leaving only a shear wave
remaining in the part.
The second critical angle occurs when the angle of incidence is at such
an angle that the remain shear wave within the part is refracted out of
the part. At this angle, when the refracted shear wave is at 90 a surface
wave is created on the part surface.
Beam angles should always be plotted using the appropriate industry
standard, however, knowing the effect of velocity and angle on
refraction will always benefit an NDT technician when working with
angle inspection or the immersion technique.
64
7. Ultrasonic Inspection Techniques
7.1. Through transmission
65
transmission. Conventional flaw detectors usually dont have a setting
for through-transmission, but the dual setting can be used. The only
difference is that when the dual setting is used on a conventional flaw
detector for through-transmission testing, the sound path distance
indicated by the instrument must be doubled Fig.29. It is often
valuable to have an additional technique for verifying the suspected
length or depth of a crack. Three examples are described here.
66
indication brought to 100% full screen height (FSH); (B) the same
sample plate, but there is a large midwall lamination present that blocks
all of the sound from being transmitted, resulting in no echo on the
screen; (C) the transducer pair is centered over the edge of the large
lamination and half of the original sound is transmitted, resulting in a
50% FSH indication.
67
FSH indication; (B) a different location where there is a deep inside
connected axial crack that blocks all of the sound; (C) the angle beam
transducers are centered at the edge of the crack and half the sound is
transmitted, resulting in a 50% FSH indication.
68
Example 1: Radial cracks in a small-diameter branch bore hole in a
heavy-walled pipe.
A heavy-walled, high-energy pipe in a fossil-fuel power plant had a
small-diameter branch connection that leaked and was repeatedly seal
welded in a series of temporary repairs. The leaks were at the outside
edges of the fillet weld joining the branch and the main pipe Fig. 31.
After the branch was cut off flush to the main pipe surface, the source
of the leak was found to be a series of radial cracks around the bore hole
that extended down to the inside of the main high-energy pipe Fig.
32.
Engineers planning repairs were especially interested in knowing the
lengths of the cracks inside the main pipe at the upstream and
downstream orientations of the bore hole the 12 oclock and 6
oclock orientations. This would be an important factor in planning for
removal of all the cracks by boring a new, larger-diameter drilled hole
centered on the old bore hole.
The lengths of the cracks inside the main pipe at the upstream and
downstream orientations were estimated with conventional shear wave
and then verified with through-transmission
UT. The probe separation distance and the search unit angles chosen
were based on the size and schedule of the main pipe. A full-size sketch
was made, and distances and angles were measured directly from the
sketch Fig. 33.
69
Figure 31 The high-energy branch connection shown here had a
recurring series of leaks. A number of temporary repairs had been made,
which included the addition of a fitting to make the seal welded fillet
weld a larger diameter in an attempt to seal off the leak. As the cracks
grew, new leaks appeared.
70
pipe. The maximum length of the radial cracks at the upstream and
downstream orientations (12 oclock and 6 oclock) was 0.85 in. long
at the outside surface. Through-transmission UT was used to verify the
cracks axial length at the inside surface.
71
channel head, the crack lengths could be measured at the inside
surfaces Fig. 35.
It was difficult to ultrasonically measure the subsurface depth of
cracking from the inside because all the surfaces were concave. It was
also difficult to ultrasonically measure the crack depth from the outside
of the heater with conventional techniques because of the curved
surfaces, the heavy-walled thickness of the components, and the fillet
weld around the outside of the outlet nozzle.
A full-size sketch was made of the cross section at the penetration.
Scan lines were established for through-transmission testing at a series
of probe separations to measure the location of the edge of the crack at
various depths. These measurements were precisely measured and
documented. During later outages, there was no scheduled access to the
inside of the heater to look for and measure crack growth. Through-
transmission UT was repeated from the outside and the measurements
did not change, indicating that the crack had not grown Fig. 37.
72
Figure 35 View of red dry powder magnetic particle indications of the
crack during access to the inside of the heater channel head. A view of
the crack looking down the outlet; B view of the crack inside the
channel head.
Figure 37 Four pairs of scan lines were set up on the outside of the
73
A large power boiler had cracks around the outside edges of most of the
large-diameter penetrations on the inside of the mud drums. Many
ultrasonic techniques were used and cracks were estimated to be
shallow, probably less than 0.25 in. deep. Some of the cracks were at
old repair welds that visually looked severe, as if they could be quite
deep. The cracks were wide open and had longitudinal and transverse
orientations Fig. 38.
It was suggested that it would be reassuring if another technique could
be used to verify the shallowness of those cracks.
Through-transmission was used as an additional technique for
verification. Full-size sketches were developed for the section at the
worst-looking cracks Fig. 39.
A pitch-catch through-transmission setup was chosen using a
0-deg straight beam search unit as the sender from the inside of the mud
drums penetrating pipe, the crossover pipe. The other search unit was
a 45-deg L-wave used on the outside surface of the mud drum head as
the receiver. Scan patterns were drawn on the inside and outside of the
mud drum Fig. 40.
The location of the edge of the cracks was verified with through-
transmission testing. The crack edge was located as in the previous
examples where the amplitude of the transmitted signal was 50% of the
sound transmitted in an unflawed area.
Through-transmission testing verified that the cracks at the 3 oclock
orientation were close to 0.5 in. deep.
74
Figure 38: Cracks at the weld toe of a large penetration into a mud drum
of a power boiler.
Figure 39: Mud drum setup utilized a 0-deg straight beam probe from
inside the crossover pipe, and a 45-deg L-wave transducer from
outside the mud drum head.
75
Figure 40: (A) Scan lines drawn on the outside of the mud drum; (B)
scan lines drawn on the inside of the mud drum crossover pipe.
76
are all suitably orientated. Discontinuities that are angled to the
scanning surface may either not reflect at all or may reflect the sound
away from the transducer.
Figure 41a shows a single crystal compression wave transducer set up
for thickness measurement of a metal part, and Figure 41b shows the
corresponding ultrasonic A-scan trace in which the time base has been
calibrated for 25 mm full scale.
The initial pulse appears at zero on the left of the trace and the back
reflection signal appears three-quarters along the time base, indicating
a sample thickness of 18.75 mm.
Thickness gauging is one of the simplest examples of compression
wave testing. Notice that the initial pulse occupies almost a quarter of
the time base so that 6 mm of metal path are obscured. This obstructed
area is known as the dead zone.
Figure 42a shows a dual element transducer set up for thickness
measurement on a sample that is 4 mm thick. Figure 42b shows the trace
for this sample with the time base again calibrated for 25 mm. Notice
that the selection of dual operation of the flaw detector isolates the
transmitter from the receiver circuit, so there is no initial pulse and,
therefore, no dead zone. The first back reflection signal (also called
back wall echo or
BWE) shows at 4 mm on the time base. Notice also that multiples of
the BWE appear at 8, 12, 16, 20, and 24 mm on the time base.
77
One way in which the reading accuracy can be improved is to take a
reading from a multiple and divide the result by the number of passes
corresponding to that multiple.
Take, for example, the reading at 24 mm (which is the sixth signal).
Divide 24 by 6 and the answer is 4 mm. However, suppose the actual
thickness was 4.15 mm. It would be difficult to read that accurately on
the first back reflection signal, but the sixth signal would have been
judged at 24.9. This number divided by 6 equals 4.15 mm. In practice,
this sort of accuracy could only be expected on samples with very
smooth scanning and back wall surfaces. For thickness measurement in
the field on corroded surfaces, errors of up to 0.5 mm are typical.
Figure 42c illustrates an actual thickness gage being used to read the
thickness of a plate that is corroded.
Figure 43a shows a single element transducer set up to detect
laminations in steel plate 20 mm thick. The lamination is smaller than
the beam. Notice in Figure 43b that the position of the lamination echo
occurs on the screen at 11 mm below the scanning surface and the back
reflection echo indicates a 20 mm thickness. The back reflection echo
is reduced in amplitude because part of the beam is reflected by the
lamination. If the lamination had been bigger than the beam, there
would be no back reflection echo. If, on the other hand, the lamination
had been smaller, the signal from the lamination would have been
smaller in amplitude and the back reflection echo bigger. It might have
been necessary to increase the equipment gain to see the lamination at
all. In the extreme case, the lamination might be so small that it could
78
not be detected at the test frequency or gain used. Therefore, the
detection of suitably orientated discontinuities is dependent on the size
of the discontinuity, the test frequency, and the gain used. Higher
frequencies can detect smaller reflectors due to their shorter
wavelength.
It must be remembered that attenuation of the ultrasonic beam also has
an effect on detection.
As the energy penetrates deeper into the material, it weakens.
Eventually, the beam is too weak to allow small echoes to get back to
the receiver. The higher the test frequency, the greater the attenuation
and the less penetration that can be achieved. The material and its grain
structure also affect attenuation. The practitioner must balance the
conflicting requirements of discontinuity size to be detected, material
properties, and ultrasonic beam properties in the choice of transducer
and test frequency.
Figures 44a and b show two discontinuities unfavorably oriented to the
sound beam. The inclined discontinuity in Figure 7-65a is reflecting the
energy away from the transducer, but also obscuring the back wall. The
result would be no signals visible on the display, but there would be a
reduction in the back reflection. In Figure 44b, the vertical crack allows
the sound to pass on either side without reflecting but would give a
normal back wall echo. The possible orientation of the discontinuity
must also be considered in devising a test procedure.
Finally, a test technique for the detection of laminations in thin plate is
illustrated in
79
Figures 45a, b, and c. The technique is called the multiple echo
technique for reasons that are obvious from Figures 7-66b and c. The
time base has been calibrated for 50 mm for a sample 3 mm thick. With
the transducer in position 1 (sound material), the multiple echo pattern
stretches to 30 mm (14 signals) as shown in Figure 7-66b. With the
transducer in position 2, over the lamination, the multiple echo pattern
only stretches to 15 mm, as can be noted in Figure 7-66c. This is
because the echoes are closer together; so close, in fact, that they
interfere with each other, leaving no clear time base in between echoes.
Contact Scanning Using Angle Beam Shear Waves. If the possible
orientation of any discontinuity is considered to be unfavorable to a
beam perpendicular to the scanning surface, it will be necessary to tilt
the beam to an appropriate angle to ensure that the beam strikes the
discontinuity as near perpendicular as possible. For small angles (up to
about 10 in the test material), compression waves may be used.
However, for larger angles, mode conversion to shear wave energy
makes the use of compression waves alone impossible.
It therefore becomes necessary to increase the incident angle beyond
the first critical angle, leaving only a shear wave in the part.
The lowest practical angle for testing with a shear wave alone is about
35 refracted shear wave angle. This does not mean that testing at angles
between 10 and 35 is impossible.
However, if an angle in this range needs to be used, the practitioner
must consider carefully the geometry of the part. The next decision is
whether to use the compression wave or the simultaneous shear wave,
80
depending on what happens to the unwanted mode. Regular off the shelf
transducers are either straight beam compression wave or shear wave
angle transducers of 35 to 70.
The common or preferred angles available in ultrasonics for shear
wave testing are 45, 60, and 70, although other angles can be made
to order. The angles marked on a shear wave transducer are for steel,
unless followed by an identifying letter for other materials.
For instance, 45Al would denote a 45 shear wave angle transducer
for aluminum.
81
Figure 42a and 42b: Thickness measurement
82
Figure 43: Perpendicular reflector
83
7.3. Angle beam
84
Distance AC = Full - Skip - Distance (FSD)
Distance AD = Half - Skip - Beam - Path - Length (HSBPL)
Distance AD + DC = Full-Skip-Bearn - Path - Length (FSBPL)
The relations used to calculate HSD, FSD, HSBPL and FSBPL for a
specimen of thickness t, are given below:
HSD = t * tan
FSD = 2 * t * tan
HSBPL = t/cos
FSBPL = 2*t/cos
If the actual probe angle is exactly equal to the nominal probe angle
then these distances can be calculated by the following formula:
Distance required = F * t
85
but cut across to the outside surface again, as shown in Figure 47 and
miss the defect.
Figure 47
For a given probe angle, the maximum wall thickness of a pipe that
allows the center of the beam to reach the bore of the pipe can be
calculated from the following formula:
(1 )
=
2
Where:
T: maximum wall thickness.
D: Outer diameter (OD) of the pipe
: probe angle
The previous Equation can be rewritten to determine the best angle for
a given wall thickness as:
= Sin-1 (1 - (2t/d))
For convenience the equation above can be simplified for standard
angle probes as
86
t=d*f
Where f is the probe factor given in Table 4.
Probe angle () 35o 45o 60o 70o 80o
Probe factor (f) 0.213 0.146 0.067 0.030 0.0076
Table 5 gives values of maximum wall thickness for various pipe sizes
and probe angles.
Probe angle 60o
87
7.4. Contact testing
7.4.1. Knowledge
7.4.1.1. Normal Beam Inspection
88
Figure 48
Figure 49
89
Figure 50
90
specimens. On-line or in-process measurement of extruded plastics or
rolled metal often is possible, as is measurements of single layers or
coatings in multilayer materials. Modern handheld gages are simple to
use and very reliable.
Figure 51
91
Figure 52
Angle Beam Transducers and wedges are typically used to introduce a
refracted shear wave into the test material. The geometry of the sample
below allows the sound beam to be reflected from the back wall to
improve detectability of flaws in and around welded areas.
Figure 53
= Angle of reflection, T=Material thickness, S= Sound path,
Skip = 2(T * Tan), Leg = T/Cos, V Path = 2 * Leg
92
Figure 54
Figure 55
93
straight beam test techniques because of their orientation with respect
to the sound beam.
Perpendicular cracks do not reflect any significant amount of sound
energy from a straight beam because the beam is looking at a thin edge
that is much smaller than the wavelength, and tilted cracks may not
reflect any energy back in the direction of the transducer. This situation
can occur in many types of welds, in structural metal parts, and in many
other critical components. An angle beam assembly directs sound
energy into the test piece at a selected angle. A perpendicular crack will
reflect angled sound energy along a path that is commonly referred to
as a corner trap, as seen in the illustration below.
Figure 56
94
Figure 57
95
Flaw location and echo display
Figure 58
Figure 59
96
Figure 60
Figure 61
97
Figure 62
Figure 63
98
Figure 64
Figure 65
99
Figure 66
Figure 67
100
Figure 68
Figure 69
101
Selecting the right angle beam assembly
The parameters that affect angle beam performance include not only the
(1) Beam angle generated by the wedge, but also (2) transducer
frequency and (3) element size. The optimum beam angle will generally
be governed by the geometry of the test piece and the orientation of the
discontinuities that the test is intended to find. Transducer frequency
affects penetration and flaw resolution:
1. As frequency increases, the distance the sound wave will travel in a
given material decreases, but resolution of small discontinuities
improves.
2. As frequency decreases, the distance the sound wave will travel
increases but the minimum detectable flaw size will become larger.
3. Similarly, larger element sizes may decrease inspection time by
increasing coverage area, but the reflected echo amplitude from small
discontinuities will decrease. Smaller element sizes will increase
reflection amplitude from small discontinuities, but the inspection may
take longer because the smaller beam covers less area.
These conflicting factors must be balanced in any given application,
based on specific test requirements.
102
Figure 70
The IIW recommends the use of a contoured wedge whenever the gap
between the wedge and the test surface exceeds 0.5 mm (approximately
0.020 in.). Under this guideline, a contoured wedge should be used
whenever part radius is less than the square of a wedge dimension
(length or width) divided by four:
Figure 71
Where
R = radius of test surface
103
W = width of wedge if testing in axial orientation, length of wedge if
testing in circumferential orientation.
Of course switching to a small wedge, if possible within the parameters
of inspection requirements, will improve coupling on curved surfaces.
As a practical matter, contouring should be considered whenever signal
strength diminishes or couplant noise increases to a point where the
reliability of an inspection is impaired.
104
Figure 72
105
7.4.1.3. Automated Scanning
106
Of interest in velocity measurements are longitudinal wave, which
propagate in gases, liquids, and solids. In solids, also of interest are
transverse (shear) waves. The longitudinal velocity is independent of
sample geometry when the dimensions at right angles to the beam are
large compared to the beam area and wavelength. The transverse
velocity is affected little by the physical dimensions of the sample.
Pulse Echo Method: Sound pressure on axis (schematic) for the incident
wave (top) and the wave reflected from a reflector in form a circular
disc (bottom).
107
Figure 73
Figure 74
108
Pulse Echo Method- Schematic screen pictures obtained by the pulse-
echo method. (a) Small flaw in sound beam; (b) two small flaws in
sound beam; (c) large flaw in sound beam, smaller second flaw and
back wall masked; (d) large, obliquely orientated flaw, back wall
masked; (e) small flaw but no back wall echo because the axis of the
beam is not incident at right angles on back wall; (f) strong attenuation
of sound beam due to scattering, no echo from flaw or back wall, only
"grass"
109
Figure 75
Pulse Echo Method- Multiple echoes in a plate. (a) schematic; (b) actual
screen picture without time or thickness scale; steel plate 50 mm thick,
frequency 4 MHz
Figure 76
110
Figure 77: Amplitude loss: Inverse Square Law
111
Figure79: Influence of reflector orientation on signal
112
Figure 81: Basic pulse echo testing presentation
Advantage:
Sensitive to near surface defect
Capable of penetrating thicker material due to pitch-catch mode.
113
Disadvantage:
It measures only sound energy loss at the receiver, without giving
details information of location.
7.4.3.1. Pitch-Catch Methods- Through Transmission
Through transmission testing uses two search units; one unit is used as
a transmitter and the other unit is used as a receiver, as shown in Figure
below.
With this technique, the ultrasonic beam passes through the test piece
or is attenuated by one or more discontinuities. Total or partial
attenuation of the signal is possible depending on the severity of the
discontinuity. Both transducers must be properly coupled with a liquid
coupling agent to obtain reliable results. As with other techniques using
two search units, greater efficiency may be obtained by using a ceramic
element in the transmitting search unit and a lithium sulfate element in
the receiving unit.
114
Figure 82: Through transmission presentation
115
surface, intersect the sound beam from the front probe at the center of
the examination zone.
Extract from: EN 583-4 Non-destructive testing - Ultrasonic examination - Part
4: Examination for discontinuities perpendicular to the surface.
Figure 83 shows the relationship between the spacing of the probes (y)
and the examination depth of the cross point (tm) and the height of the
examination zone (tz). When examining objects with plane parallel
surfaces the distance between the probes can be defined using the
following equation:
y = 2 tan (d tm) or 2 tan (bottom depth)
116
Figure 84: Distance between Transmitter / Receiver Probes
Immersion testing techniques are mainly used in the laboratory and for
large installations doing automatic ultrasonic testing. It has the
advantage that uniform couplant conditions are obtained and
longitudinal and transverse waves can be generated with the same probe
simply by changing the incident beam angle.
The three basic techniques used in immersion testing are the immersion
technique, the bubbler technique and the wheel transducer technique.
In the immersion technique both the probe and the test specimen are
immersed in water. The ultrasonic beam is directed through the water
in to the test specimen, using either a normal beam technique (Figure
117
85a) for generating longitudinal waves or an angle beam technique
(Figure 85b) for generating transverse waves.
When the normal beam technique is being used the water path distance
must always be longer than the distance S in the following equation:
When the specimen is steel the water path distance must be longer than
1/4 steel thickness otherwise the 1st back wall echo overlaps the 2nd
surface echo and defects near the back wall may not be seen.
In the bubbler or squirter technique, the ultrasonic beam is directed
through a water column in to the test specimen (Figure 86). This
technique is usually used with an automated system for high speed
scanning of plate, sheet, strip, cylindrical forms and other regularly
shaped forms.
The ultrasonic beam is either directed in a perpendicular direction (i.e.
normal direction) to the test specimen to produce longitudinal waves or
is adjusted at an angle to the surface of the test specimen for the
production of transverse waves.
In the wheel transducer technique the ultrasonic beam is projected
through a water-filled tire in to the test specimen. The probe, mounted
on the wheel axle, is held in a fixed position while the wheel and tire
rotate freely.
118
The wheel may be mounted on a mobile apparatus that runs across the
specimen, or it may be mounted on a stationary fixture, where the
specimen is moved past it (Figure 87 a and b). The position and angle
of the probe mounted on the wheel axle may be constructed to project
normal beams, as shown in Figure 87 a and b or to project angled beams
as shown in Figure 88
119
Figure 87: stationary and moving wheel transducer
120
8. Ultrasonic Testing Equipment
8.1. Basic pulse echo instrument
121
Figure 89: Basic pulse-echo instrument
122
the test object, back to the transducer the receiver (amplifier), the
gate and display. In essence, the information content of the initial
electrical pulse is modified by each of these items. It is the result of
this collective signal processing that appears on the screen. The
initial pulse may range from 100 to 500 V and have a very short rise
time.
123
simple means for altering signal levels beyond the viewing range of
the screen. Their extensive use has made decibel notation a part of
the standard terminology used in describing changes in signal levels,
such as changes in receiver gain and material attenuation.
The ratio of two pulse amplitudes (A2 and A1) can be expressed in
decibel notation (Ndb)
A2
Ndb = log10
1
124
Following a propagation delay corresponding to the ultrasonic time
of flight between the transducer and an internal reflector, the back
scattered ultrasonic signals are received by the same transducer.
These signals are then directed by the receiver preamplifier.
125
video display or A-scan. In some designs, it is possible to display
radiofrequency waveforms directly.
The horizontal axis of the display device is driven by the weep signal
generated in the systems timing section. Generally, the start of each
sweep signal is delayed with respect to the transmitter pulse or by an
interface trigger. This delay is used to offset the start of the display
to some convenient interface echo.
The signal gate delay, width parameters and alarm threshold level
typically can be selected from the front panel. To ensure reliable
results, receiver gain levels and the alarm threshold level within the
126
gate interval should be adjusted before the test using an appropriate
ultrasonic reference standard and an instrument calibration
procedure.
127
common. In the twenty-first century many designs incorporate
square wave pulsers.
If the instrument uses a spike pulser, then the operator may be able
to modify the pulse amplitude by adjusting the energy of the pulse.
Pulse is adjusted by selecting the value of the energy storage
capacitor. In addition, an adjustment of the damping resistor value
may be made to minimize transducer ringing.
128
The gain adjustment and signal gate functions are important because
they can be used to control accept/reject threshold. If the amplitude
of the signal in a discontinuity gate exceeds a pre-established
threshold, then the discontinuity alarm is activated. The
discontinuity alarm is usually built-in and can be audible or visual.
129
analysis of the amplitude and phase information. The A-scan method is
the most widely used and can be displayed on most standard ultrasonic
instruments.
With the ultrasonic B-scan, the test object is scanned along one axis to
produce a presentation of its cross section. The location along the
scanning path is shown on the X axis and time-of-flight values are
130
shown along the Y axis. Because a cross section is produced, the B-scan
is not used where large volumes of material must be inspected.
The B-scan is popular for medical diagnosis where cross-sectional
views are very useful. In medical applications, the angular manipulation
of the transducer is monitored to prevent image distortion and the
display is adjusted to account for changes in the beam angle along the
cross section of the examined area.
131
The ultrasonic C-scan is applied to the test object in a raster pattern and
presents a view of the discontinuity's area as seen from above.
Discontinuity location and size data are available from changes in
amplitude as a function of position. These are displayed on a screen or
recorded on paper. Modern C-scan systems use computers to control
the transducer position and to acquire, display, document and store the
test results. The computer synchronously acquires the digitized position
of the transducer and the associated reflected signal or the value of a
specific ultrasonic parameter. The position can be obtained by various
means, including optical encoders or sonic digitizers.
Computerized C-scan systems can acquire several ultrasonic
parameters as a function of position. In most cases, the parameter is
time of flight or the amplitude of reflection or transmission amplitude
at a certain time range. The parameters are digitized with aid of an
analog-to-digital converter.
In immersion ultrasonic testing, the C-scan systems can scan at speeds
up to 500 mm.s-l (20 in..s-l) or higher. Speeds must be kept at a level
that does not induce water turbulence, which introduces noise and
degrades the reliability of the test.
132
Figure 92: Diagram of simple ultrasonic C-scan presentation
133
The limitations of this generation of field systems are: (l) high cost, (2)
small scanning area and (3) reliability for testing objects with simple
geometries.
Crawlers have been developed to perform field scanning of structures.
Current crawlers are tethered to provide power. Air coupled systems
make it possible to test parts of complex geometry.
134
requirements, component geometry and history, material
characteristics, heat treatment, operating temperature, and pressure.
The integration of the synthetic aperture focusing technique (SAFT)
analysis module provides three-dimensional views of the inspected
zones in various cross-sections. This tool helps the qualified technician
to accurately determine type, location, and size of detected material
discontinuities important for fracture and fatigue mechanics analysis;
for example, for use in the assessment of a component's lifetime.
135
9. Ultrasonic Reference Blocks
9.1. Calibration Methods
136
by different systems. Reference standards also help the inspector to
estimate the size of flaws. In a pulse-echo type setup, signal strength
depends on both the size of the flaw and the distance between the
flaw and the transducer. The inspector can use a reference standard
with an artificially induced flaw of known size and at approximately
the same distance away for the transducer to produce a signal. By
comparing the signal from the reference standard to that received
from the actual flaw, the inspector can estimate the flaw size.
This section will discuss some of the more common calibration and
reference specimen that are used in ultrasonic inspection. Some of
these specimens are shown in the figure above. Be aware that there
are other standards available and that specially designed standards
may be required for many applications. The information provided
here is intended to serve a general introduction to the standards and
not to be instruction on the proper use of the standards.
137
cases the artificially induced defects in reference standards are better
reflectors of sound energy (due to their flatter and smoother surfaces)
and produce indications that are larger than those that a similar sized
flaw would produce. Producing more "realistic" defects is cost
prohibitive in most cases and, therefore, the inspector can only make
an estimate of the flaw size. Computer programs that allow the
inspector to create computer simulated models of the part and flaw
may one day lessen this limitation.
138
9.4. Distance amplitude blocks
139
9.5. International Institute of Welding (IIW) block
140
reference block because it was patterned after the "true" IIW block
but does not conform to IIW requirements in IIS/IIW-23-59. "True"
IIW blocks are only made out of steel (to be precise, killed, open
hearth or electric furnace, low-carbon steel in the normalized
condition with a grain size of McQuaid-Ehn) where IIW "type"
blocks can be commercially obtained in a selection of materials. The
dimensions of "true" IIW blocks are in metric units while IIW "type"
blocks usually have English units. IIW "type" blocks may also
include additional calibration and references features such as
notches, circular groves, and scales that are not specified by IIW.
There are two full-sized and a mini versions of the IIW type blocks.
The Mini version is about one-half the size of the full-sized block
and weighs only about one-fourth as much. The IIW type US-1 block
was derived the basic "true" IIW block and is shown below in the
figure on the left. The IIW type US-2 block was developed for US
Air Force application and is shown below in the center. The Mini
version is shown on the right.
141
Figure 95: IIW Type US-1
142
Figure 96: IIW Type Mini
Also known as a Rompas block, this ASTM and U.S. Air Force
miniature angle beam block is a substitute for the DSC block for
distance, beam index, refracted angle and sensitivity calibration.
Contains a 1.0 radius opposite a 2.0 radius, and a 5/64 diameter x
.750 deep flat-bottom hole. In accordance with ASTM E164 and U.S.
Bureau of Public Roads, Type B specifications. Dimensions: 1.0 thick.
Metric version available.
143
Figure 97: Miniature angle beam block
The test object is tested with a high gain setting by which the smallest
detectable reflector is displayed. An echo indication is peaked, i.e. the
maximum echo indication is achieved by careful movement of the
probe and the echo peak set by adjustment of the gain to a
predetermined height, e.g. 80% CRT screen height (reference height).
144
Figure 98: Test object with a flaw: echo at 80% (reference height)
Using the same settings, the reflector from the reference block is
scanned which is approximately positioned at the same distance as the
discontinuity.
145
Figure 99: Reference block: reference echo at 30%
The quantative unit for evaluation is now the gain change of the
ultrasonic instrument which is necessary to set the reference echoto the
reference height.
146
9.7.2. Distance amplitude curve
All reflectors in the reference block are scanned before the test, their
maximum echo heights marked on the attachment scale of the display
and joined by a curve.
Figure 101: Reference block wiht side drilled holes and resulting echoes
147
2. That the heavy reference block need not be transported to the testing
location.
3. That the recording of a DAC for certain applications is only required
once because the curve is documented on a transparency or in the
memory of a modern ultrasonic test instrument. By recording the curve
using reflectors in a test object comparable to the work piece, this curve
contains all the influences in the test object (distance law, sound
attenuation, and surface losses). Corresponding corrections are
therefore not necessary. Regarding the evaluation results, we must
understand here that the effect of the discontinuity (geometry,
orientation and surface quality) is not taken a great deal into account
the same as the DGS method. Therefore, the result of a discontinuity
evaluation with the reference block method has the same uncertainty as
the
DGS method.
The preference regarding which method to use is subjective. The
corresponding national test specifications normally state the test
method to be used so that the operator is not able to make his own
decision. If no data is available, the test situation should be analyzed in
order to decide which method be best used:
Firstly, it must be established whether a reference block exists which
corresponds to the test object. If yes, then the test can be carried out
simply and reproducibly with the reference block method. If no
reference block is available then the DGS method can be used, or a
reference block must be subsequently produced comparable to the test
148
object. However, in many cases the DGS method can be used without
difficulty, namely when the test object is made of low alloy steel, has a
simple geometry, a low sound attenuation and an even surface quality.
The test should be carried out with a narrow band standard probe with
a frequency between 1 MHz and 6 MHz for which there is a DGS
diagram or a DGS scale. The new computer controlled instruments
normally support the program controlled recording of DACs. With the
USD 10 the recorded DAC is automatically converted to a horizontal
line. This is known as time corrected gain (TCG).
Figure 102: DAC of the reference echoes (left) and with time corrected gain
(right)
149
10. Inspected Part Variations
The fact is that roughness is the natural state of surfaces, and left to
its own devices, nature will make sure they are rough. The roughness
of a surface is a measure of its lack of order. Disorder is entropy
under another name, and if a solid surface is considered as a closed
system then the Second Law of Thermodynamics predicts that its
entropy will tend to a maximum. To reduce its roughness, its entropy
must be reduced, and the Second Law tells that it can only be done
this by doing work. Thus if the axes of the well-known figure are
150
transposed which relates machining time to roughness, it can easily
seen that, it is nothing but an entropy diagram.
Figure 103: Relationship of surface texture to production time (b) the same
figure replotted as work reducing entropy
Types of Surfaces
151
Real Surface: A real surface is the actual boundary of an object. It
deviates from the nominal surface as a result of the process that created
the surface. The deviation also depends on the properties, composition,
and structure of the material the object is made of.
152
Figure 104: An Exaggerated Surface Shape
153
Sometimes these surfaces are said to have a non-directional,
particulate, or protuberant lay. Several different types of lay are
possible depending on the manufacturing and machining processes.
How the surface roughness affects the results of the ultrasonic test
Figure 105: Poor coupling results due to rough surface and thin couplant
154
In addition to reduced coupling, which will reduce signal
amplitudes, the rough surface increases the rate of wear on the probe.
On an otherwise smooth surface isolated sticky regions such as weld
spatter can hinder or stop probe motion or in the case of mechanized
systems there may be sufficient force to move the probe past the
obstruction but this could result in damaging the probe by either
tearing it from its mounting or severely scoring the plastic wedge.
When the dirt on the test piece is very fine (similar to a flour texture)
coupling can be prevented due to surface tension preventing the
liquid couplant penetrating to the metal. Unless a transfer value has
been established between test piece and calibration piece, this could
go undetected.
155
resulting from scatter on rough surfaces will normally have short sound
paths. They can be eliminated as true flaws by failing to locate any trace
of indication from the full skip or from the opposite side.
156
be compared to the uncoated reference block by a simple transfer value.
Even a slight loss due to the coating may be preferable to removing the
coating and trying to inspect on the rough surface it hides. Tightly
adhering coatings on the surface generally allow good transfer of the
energy. Loose or flaking coatings are undesirable and should be
removed prior to conducting the examination. When calibrating the
equipment for reference sensitivity on critical applications, it is
essential to evaluate the component for any energy losses due to surface
condition and apparent attenuation variations. The procedure for this is
fairly simple and is performed by using two transducers in a pitch
catch technique.
157
To accurately predict the effects of curved surfaces on pulse/echo
immersion inspection, it is necessary to have a set of models for the
transducer radiation field, the flaw response signal, and the
microstructure induced noise. Over the years, many fast and reasonably
accurate models have been developed in a number of laboratories. For
the transducer radiation field, the Gauss-Hermite beam model produces
good results for transducer radiation into liquids and for transmission
of those fields through flat or curved liquid-solid interfaces. By taking
advantage of the paraxial approximation, the model runs much faster,
by many orders-of-magnitude, than other numerical methods. This
"beam" model enables the transducer wave fields on the solids to be
rapidly computed for a series of transducers and focal lengths as a part
of the inspection design process. For the flaw response signal, a "flaw
signal" model has been developed that uses a modified born
approximation and requires a numerical integration of the incident
pressure field over the flaw volume. This model can predict time
domain flaw signals that can be compared, on an absolute level, to
actual measurements. As for the beam models, use of an appropriately
selected approximation allows the code to run much faster than purely
numerical methods such as the finite element method. For the
microstructure-induced noise, a backscattered noise model has been
developed that uses a single-scattering approximation and assumes that
the observed RF noise is an incoherent summation of echoes from all
insonified grains. The model can predict the backscattered grain noise
158
level for a given inspection scenario in a computationally efficient
manner.
In this work we describe the use of these physics-based models of an
ultrasonic pulse/echo inspection process to guide technique
optimization for detecting inclusions through cylindrical surfaces in
steel bar stock. Models are used for three purposes. The beam model is
used to select transducer focal properties that will be minimally affected
by the cylindrical geometry of the billet surface, which acts to defocus
the beam in a plane perpendicular to the billet axis. The flaw response
model is used to predict the strength of echoes from undesirable
inclusions that can lead to premature failure. The microstructural
response model is used to predict backscattered grain noise levels for
microstructures of a given type and grain size. The flaw response and
noise predictions combine to yield estimates of the minimum detectable
inclusion size.
As an example of the integrated use of these tools, we discuss two
competing immersion designs for a system to inspect 3-inch diameter
cylindrical billet used for the manufacture of gears. One design uses a
standard, spherically focused transducer; and the other uses a
bicylindrically focused transducer, designed to better inspect the zone
where gear teeth will later be machined. We will assume that the
inclusions are hypothetical "stringers" whose elongation direction is
aligned with the billet axis. Our main goal is to estimate stringer
detectability for both the standard and optimized transducers.
159
10.4. Grain size
160
Ahmed and Thompson, and Panetta have extended the theory to treat
polycrystalline materials with texture and elongated grains.
161
Han and Thompson have extended Rose's theory to treat duplex
titanium alloys. Rose's theory and related extensions will be used in
this work to analyze backscattering of longitudinal waves. It is also
desirable to extend the theory to calculate backscattering for
transverse waves since it can provide additional valuable
information about the microstructure.
162
measuring longitudinal wave backscattering from three orthogonal
directions. The third one was based on measuring longitudinal and
transverse wave backscattering from a single surface.
163
11. Discontinuity Variations
11.1. Sizing methods
164
Figure 106: A large reflector in the sound beam
165
Figure 107a: Straight beam probe on the reflector boundry
This means that the acoustic axis is exactly on the boundary of the
discontinuity. The probe position is marked and the operator
determines further boundry points until a contour of the
discontinuity is formed by joining the marked points together, figure
107b.
166
11.1.2. Evaluation of small discontinuities: The DGS method
Under optimal conditions, e.g. drill holes with flat bottoms and at equal
depths, this law can be confirmed:
167
The echo heights are proportional to their area or The echo heights are
proportional to the square of their diameter
168
With accurate tests using flat-bottom holes at different depths a simple
law can be found, at least in the far field of the applied sound beam:
If such curves are put on transparent scales having the CRT format then
we immediately have the possibility to comparatively evaluate echoes
from unknown reflectors and those from natural reflectors, i.e. the echo
height of the discontinuity is compared to that of a circular disk.
The discontinuity in figure 111 reflects the sound waves the same as a
circular disk having a diameter of 4 mm.
169
Figure 111: Evaluation of a discontinuity (F) using evaluation curves
Due to the fact that we can only assess the sound reflected from the
discontinuities we must of course not equate the diameter of 4 mm with
the "true size" of the discontinuity. We therefore refer to them as an
equivalent disk-shaped reflector or as equivalent reflector size (ERS).
The equivalent reflector size only correspondsto the true reflector size
of a discontinuity in an ideal case which is when it is circular and
exactly hit vertical to the acoustic axis. In practise this almost never
occurs which means that the true size of a discontinuity is normally
larger than the equivalent reflector size. A law for this cannot be derived
because the echo height is strongly dependent on the characteristics of
the discontinuity, this means its geometry, orientation to the sound
beam and the surface quality. For example, a pore (spherically shaped
gas inclusion) with a diameter of 2 mm has an equivalent reflector size
of 1 mm; an angled flat reflector 5 mm long gives, according to
orientation, a result of ERS 0 (not detectable) to perhaps ERS 2.
This uncertainty in the evaluation of the discontinuity is however
neutralized when other possibilities and techniques in ultrasonic testing
170
are used to inspect detected discontinuities closer. An experienced
ultrasonic operator can, without additional expense, accurately give
information about the discontinuity which he has detected.
Scanning the discontinuity from different directions, assessing the echo
shape and the behavior of the display when moving the probe (echo
dynamics) are just a few techniques which can be successfully applied.
Despite the remaining uncertainty with evaluation of natural
discontinuities theabove method of discontinuity evaluation is applied
in many countries due to the fact that the method is based on well
proven laws in the sound field. It is therefore reproducible, i.e. the
evaluation results are independent of testing device and operator.
The socalled DGS scales or discontinuity evaluation can be obtained
from the probe manufacturer for many probes and various calibration
ranges. DGS means that the scale is allocated an echo at the Distance,
with correctly set Gain and (equivalent reflector) Size. However, the
modern version of the DGS scale would need some explanation because
it was developed to fulfill the requirements of the most common
specifications in practical testing: If, on a certain test object whose
purpose and therefore stress values are known, an ultrasonic test is to
be carried out then firstly, if necessary with destructive testing, it should
be established how large the permitted material flaw should be. Of
course, the position of such a flaw in the material and its rate of
occurance play a part.
If a permitted flaw size has been determined then this size is multiplied
with the safety factor which, amongst others, also takes the evaluation
171
uncertainty of the ultrasonic test into account. The corresponding echo
amplitude curve for this size is now of importance for the ultrasonic
test. The ultrasonic operator scans the test object with the probe and
only needs to record the indications which exceed this recording curve,
figure 112.
172
the test instructions provide the following measures: rejection, repair
or further tests for exact assessment of the discontinuity (diagnosis).
Figure 113 shows testing of a forged part. The recording curve
corresponds to Equivalent Reflector Size 3. The detected discontinuity,
at a depth of 110 mm, exceeds the curve, i.e. all reflector data must now
be recorded into a predetermined form.
11.2. Shape
173
Crack is tight linear separations of metal that can be very short to
very long indications. Cracks are grouped as hot or cold cracks. Hot
cracks usually occur as the metal solidifies at elevated temperatures.
Cold cracks occur after the metal has cooled to ambient temperatures
(delayed cracks).
Lack of fusion (Cold Lap) is a condition where the weld filler metal
does not properly fuse with the base metal or the previous weld pass
material (inter pass cold lap). The arc does not melt the base metal
sufficiently and causes the slightly molten puddle to flow into base
material without bonding.
174
Porosity is the result of gas entrapment in the solidifying metal.
Sometimes porosity is elongated and may have the appearance of
having a tail This is the result of gas attempting to escape while the
metal is still in a liquid state and is called wormhole porosity. All
porosity is a void in the material.
The moisture turns into gases when heated and becomes trapped in
the weld during the welding process. Cluster porosity appear just like
regular porosity but the indications will be grouped close together
(Please verify the applicable norme of the project to know more
about indications groupement).
175
Slag inclusions are nonmetallic solid material entrapped in weld
metal or between weld and base metal.
176
Figure 120: Internal concavity
177
Offset or mismatch are terms associated with a condition where two
pieces being welded together are not properly aligned. The
difference in density is caused by the difference in material
thickness. The dark, straight line is caused by failure of the weld
metal to fuse with the land area.
178
Figure 125: Excess weld reinforcement
The flaw position is used for both region of interest (ROI) selection
the depth normalisation (which is required for the feature extraction
performed later). It is therefore important that the position estimation
is accurate and robust. The current method to find the flaw position
is based on fitting an hyperbolic function to the flaw response in B-
scan data. The algorithm operation is illustrated at Fig 126. Consider
the A-scan 48 mm from the centre of the weld, marked with a vertical
line in Figure 126b (also included in the box in the same figure). The
maximum response yielded by the algorithm is approx. Rmax= 45mm.
179
Figure 126: Illustration of defect position estimation.
Depth Normalisation
180
suitable angular range given the depth of the flaw. This implies
interpolating features from flaws located close to the probe, and
down sampling features for flaws that lie further away from the
probe.
Fig 127: Illustration of the effect of the cone beam geometry for two defects
at different depths. Defect f1 will be seen in fewer A-scans than defect f2.
ROI Selection
181
However, for realistic defects the echo-dynamics curves may be
skewed, have more than one peak, etc. The approach used previously
was to smooth the echo-dynamics, with low-pass filtering, which
partially solves the problem. This method was suitable for the simulated
and artificial defects. Experiments have been performed using entre-of-
mass calculations in order to find a robust estimate for the echo-
dynamics centre point. This approach was, however, too sensitive to
long tails with high amplitude (energy) in the echo-dynamics.
Therefore, the previously used algorithm was adopted here as well. The
algorithm includes: low-pass filtering and finding the maximum of the
echo-dynamics in the first step, and selecting a number of A-scans
centred around the A-scan corresponding to max amplitude in the
second step.
Classical Features
182
problems, but at the expense of some loss of information. Figure 128
shows three examples of envelopes corresponding to different types of
defects.
In spite of the very different shape of the waveforms, the rise time,
pulse duration, and fall time are rather similar for all of the signals in
Figure 128. It is evident that more powerful features are needed if the
classification should be feasible for this type of signals.
Conclusions
183
ultrasonic signals within one type (class) of defects is much larger for
real than for artificial defects. Our goal was to separate soft (or
volumetric) defects from the sharper ones (crack-like defects), but if
one studies the echo-dynamics and the pulse shapes (i.e. the envelope)
it becomes apparent that some sharp and soft defect types are very hard
to separate. This implies that overlapping feature regions are
encountered, especially when using classical features (fall/raise times,
pulse duration and echo-dynamics). To avoid overlapping class
boundaries, more powerful feature extraction algorithms are needed to
achieve a good classification performance. High variation of the
ultrasonic signals also has two further consequences: flaw position
estimation (needed for feature extraction) may be poor and the amount
of data needed to construct reliable classifiers large. The following is
recommended during design of a self-learning classifier:
184
The first recommendation may be not fulfilled using single B-scan
measurements only. A common practice is to combine measurements
from several transducers (with different angles, centre frequencies, etc.)
and TOFD measurements. The second condition is clearly not fulfilled
if only classical features are used and therefore more powerful methods
are needed.
185
Appendix 1: Locating a discontinuity
186
1st Echo = t,
2nd Echo = 2t,
3rd Echo = 3t, etc.
187
(display scale) is 10 mm per scale
graduation. We couple the straight-
beam probe to the V1 block (laying
flatwise), Fig. 39. The backwall echo
sequence now comes from the 25 mm Fig. 40 Calibration range: 0-
steel path. The allocation of sound 10mm
paths si to the corresponding scale
positions Ti is carried out using the calibration table:
1 25 10 2.5
2 50 10 5.0
3 75 10 7.5
4 100 10 10.0
188
Fig. 41 USK 7: Calibration in the 100 Fig. 42 USK 7 D: Consideration of
mm range the probe delay
189
Echo-No Sound path si Scale factor k Skalen-position Ti
i [mm] [mm/scale grad.] [scale grad.]
1 80 25 3.2
2 160 25 6.4
3 240 25 9.6
190
a 250 mm range with an 80 mm Fig. 44 USK 7 D: Sound path
aluminum path measurement.
191
b) Now we couple the probe to the 6 mm step and
bring the echo to the 6th scale graduation with the
range control.
c) Steps a) and b) are alternately repeated until both
echo flanks are exactly on the 3rd and 6th scale
graduations, Fig. 45a+b.
192
echo at the 6th graduation
(bottom)
193
Fig. 49a Sound path in the V1
block without angle reflection
The advantage with echoes from the circle segment of the calibration
block is that the same sound path is always given independent of the
probe angle, Fig. 48. When the angle-beam probe is exactly coupled in
the center of the circle segment, a first echo is exactly received from
100 mm out of V1 block. According to the reflection law, the sound
waves coming out of the arc are reflected away from the coupling
surface to the back, this means away from the arc, Fig. 49a. A second
echo out of the arc, needed for the calibration sequence, cannot
therefore be produced. For this, there are two saw cuts made in the
center of the quarter circle: in the edges, which these saw cuts form with
the surfaces, the sound waves are reflected back within themselves due
194
to double reflection (angle reflection effect) so that they go back to the
arc, Fig. 49b.
For the miniature angle-beam probe one uses the considerably smaller
and lighter Standard Calibration Block 2 (V2 block). This has, as
opposed to the V1 block, two circle segments with a common center
point, however it does not have saw cuts. The required echo sequence
is produced here by the alternating reflection of the sound waves, Fig.
51a+b.
195
Fig. 51b Path of a sound wave in a Fig. 52 Range: 100 mm calibrated
V2 block, radius 25 mm on V2, radius 25 mm.
The echo of a discontinuity on the instrument display does not now give
us any direct information about its position in the material. The only
available information for determination of the reflector position is the
scale position and therefore the sound path s, this means the distance of
the discontinuity from the index point(sound exit point) of the probe,
Fig. 53.
196
The mathematics of the right-
angled triangle helps us to evaluate
the Surface Distance and
the Depth of a reflector which are
both important for the ultrasonic
test, Fig. 54a. We therefore now
have the possibility to instantly Fig. 53 Scanning a reflector using
mark a detected flaw's position on an angle beam probe
the surface of the test object by
measurement of the surface distance from the sound exit point and to
give the depth. For practical reasons, the reduced surface distance is
used because this is measured from the front edge of the probe. The
difference between the surface distance and the reduced surface
distance corresponds to the x-value of the probe, this is the distance of
the sound exit point to the front edge of the probe, Fig. 54b.
197
and immediately displayed so that the operator does
not need to make any more time-consuming
calculations, Fig. 55. This is of great help with weld
testing because with the calculation of the flaw depth
Fig. 55 USN 50:
an additional factor must be taken into account,
A hole being
namely: whether the sound pulses were reflected
scanned with the
from the opposing wall. If this is the case then
probe MWB 60-
an apparent depth of the reflector is produced by
4E
using the depth formula which is greater than the
thickness T of the test object. The ultrasonic operator must acertain
whether a reflection comes from the opposite wall and then proceed
with calculating the reflector depth, Fig. 56b.
198
the discontinuity from theindex point (sound exit point) of the probe,
Fig. 53. The mathematics of the right-angled triangle helps us to
evaluate the Surface Distance and the Depth of a reflector which are
both important for the ultrasonic test, Fig. 54a. We therefore now have
the possibility to instantly mark a detected flaw's position on the surface
of the test object by measurement of the surface distance from the sound
exit point and to give the depth. For depth of the reflector is produced
by using the depth formula which is greater than the thickness T of the
test object. The ultrasonic operator must acertain whether a reflection
comes from the opposite wall and then proceed with calculating the
reflector depth, Fig. 56b.
199
A-scanA method of data presentation on an ultrasonic display
utilizing a horizontal baseline, that indicates distance, and a vertical
deflection from the baseline, that indicates amplitude.
A-Scan presentationA method of data presentation utilizing a
horizontal baseline to indicate distance, or time, and a vertical
deflection from the baseline to indicate amplitude.
AmplitudeThe vertical height of a signal, usually base to peak, when
indicated by an A-scan presentation.
Angle beamA term used to describe an angle of incidence or
refraction other than normal to the surface of the test object, as in angle
beam examination, angle beam search unit, angle beam longitudinal
waves, and angle beam shear waves.
Area amplitude response curveA curve showing the relationship
between different areas of reflection in an material and their respective
amplitudes of ultrasonic response.
AttenuationA factor that describes the decrease in ultrasound
intensity or pressure with distance. Normally expressed in decibels per
unit length.
B-scan presentationA means of ultrasonic data presentation that
displays a cross section of the specimen, indicating the approximate
length (as detected per scan) of reflectors and their relative positions.
Back reflectionAn indication, observed on the display screen of a
UT instrument, that represents the reflection from the back surface of a
reference block or test specimen.
Back echoSee back reflection.
Back surfaceThe surface of a reference block or specimen that is
opposite the entrant surface.
Beam spreadA divergence of the ultrasonic beam as it travels
through a medium.
BubblerA device using a liquid stream to couple a transducer to the
test piece.
C-scanAn ultrasonic data presentation that provides a plan view of
the test object and discontinuities.
200
CollimatorA device for controlling the size and direction of the
ultrasonic beam.
Contact testingA technique in which the transducer contacts directly
with the test part through a thin layer of couplant.
CouplantA substance, usually a liquid, used between the transducer
unit and test surface to permit or improve transmission of ultrasonic
energy.
Critical angleThe incident angle of the ultrasonic beam beyond
which a specific refracted wave no longer exists.
DACDistance amplitude correction. Electronic change of
amplification to provide equal amplitude from equal reflectors at
different depths. Also known as swept gain, time corrected gain, time
variable gain, etc.
DAC curveA curve (usually drawn on the screen) derived from equal
reflectors at different depths.
Damping, search unitLmiting the duration of a signal from a search
unit subject to a pulsed input by electrically or mechanically decreasing
the amplitude of successive cycles.
dB controlA control that adjusts the amplitude of the display signal
in decibel (dB) units.
Dead zoneThe distance in the material from the surface of the test
specimen to the depth at which a reflector can first be resolved under
specified conditions. It is determined by the characteristics of the search
unit, the ultrasonic instrumentation, and the test object.
Decibel (dB)Logarithmic expression of a ratio of two amplitudes or
intensities. (UT) dB = 20 log10 (amplitude ratio).
Delay lineA column of material such as Plexiglas that is attached to
the front of a transducer. It behaves similarly to a water path and allows
the initial pulse to be shifted off the scree. This often improves near
surface resolution.
Delay sweepAn A-scan or B-scan presentation in which an initial
part of the time scale is not displayed.
201
DiscontinuityA lack of continuity or cohesion; an intentional or
unintentional interruption in the physical structure or configuration of a
material or component.
Distance amplitude, compensation (electronic)The compensation
or change in receiver amplification necessary to provide equal
amplitude on the display of an ultrasonic instrument for reflectors of
equal area that are located at different depths in the material.
Distance amplitude, response curveSee DAC. A curve showing the
relationship between the different distances and the amplitudes of an
ultrasonic response from targets of equal size in an ultrasonic
transmitting medium.
Distance linearity rangeThe range of horizontal deflection in which
a constant relationship exists between the incremental horizontal
displacement of vertical indications on the A-scan presentation and the
incremental time required for reflected sound to pass through a known
length in a uniform transmission medium.
Doppler effectThe change in frequency of a sound wave due to
movement of the reflector. Movement toward or away from the sound
will result in a change in frequency (e.g., the tone of a train whistle
changing as the train passes).
Dual search unitA search unit containing two elements, one a
transmitter, the other a receiver.
Dynamic rangeThe ratio of maximum to minimum reflective areas
that can be distinguished on the display at a constant gain setting.
Entrant surfaceThe surface of the material through which the
ultrasonic waves are initially transmitted.
Far fieldThe zone of the beam (beginning at the Y0 point) where
equal reflectors give exponentially decreasing amplitudes with
increasing distance.
FlawA discontinuity in a material or component that is unintentional.
Flaw characterizationThe process of quantifying the size, shape,
orientation, location, growth, or other properties of a flaw based on
NDT response.
202
Frequency (examination)The number of cycles per second (Hz).
Frequency, pulse repetitionThe number of times per second that a
search unit is excited by the pulser to produce a pulse for ultrasonic
imaging. This is also called pulse repetition rate or pulse repetition
frequency (PRF).
GateAn electronic means of selecting a segment of the time range
for monitoring, triggering an alarm, or further processing.
Immersion testingAn ultrasonic examination technique in which the
search unit and the test part are submerged (at least locally) in a fluid,
usually water.
Impedance, acousticA mathematical quantity used in computation
of reflection characteristics at boundaries. It is the product of wave
velocity and material density
IndicationA response or evidence of a response disclosed through
an NDT that requires further evaluation to determine its full and true
significance.
Initial pulseThe response of the ultrasonic system display to the
transmitter pulse (sometimes called main bang).
Lamb waveA specific mode of propagation in which the two parallel
boundary surfaces of the material under examination (such as a thin
plate or wall of a tube) establish the mode of propagation. The Lamb
wave can be generated only at particular values of frequency, angle of
incidence, and material thickness. The velocity of the wave is
dependent on the mode of propagation and the product of the material
thickness and the examination frequency.
Linearity, amplitudeA measure of the proportionality of the
amplitude of the signal input to the receiver and the amplitude of the
signal appearing on the display of the ultrasonic instrument or on an
auxiliary display.
Linearity, time or distanceA measure of the proportionality of the
signals appearing on the time or distance axis of the display and the
input signals to the receiver from a calibrated time generator or from
multiple echoes from a plate or material of known thickness.
203
Longitudinal waveA wave in which the particle motion of the
material is essentially in the same direction as the wave propagation.
(also called compressional wave).
Metal pathSee Sound path
ModeThe type of ultrasonic wave propagating in the material as
characterized by the particle motion (e.g., longitudinal, transverse, etc.)
Mode conversionPhenomenon by which an ultrasonic wave that is
propagating in one mode refracts at an interface to form ultrasonic
wave(s) of other modes.
Multiple back reflectionsSuccessive signals from the back surface
of the material under examination.
Near fieldThe region of the ultrasonic beam adjacent to the
transducer having complex beam profiles and intensity variations. Also
known as the Fresnel zone.
NoiseAny undesired signal (electrical or acoustic) that tends to
interferes with the interpretation or processing of the desired signals.
Normal incidence (also see Straight beam)A condition in which the
axis of the ultrasonic beam is perpendicular to the entrant surface of the
part being examined.
Penetration depthThe maximum depth in a material from which
usable ultrasonic information can be obtained and measured.
ProbeSee Search unit.
Pulse-echo techniqueAn examination method in which the presence
and position of a reflector are indicated by the echo amplitude and time.
Pulse lengthA measure of the duration of a signal as expressed in
time or number of cycles.
RangeThe maximum distance that is presented on a display.
Rayleigh waveAn ultrasonic surface wave in which the particle
motion is elliptical and the effective penetration is approximately one
wavelength.
Reference blockA block of material that includes reflectors. It is
used both as a measurementscale and as a means of providing an
ultrasonic reflection of known characteristics.
204
ReflectorAn interface at which an ultrasonic beam encounters a
change in acoustic impedance and at which at least part of the sound is
reflected.
Reject, suppressionA control for minimizing or eliminating low-
amplitude signals (electrical or material noise) so that true signals are
emphasized.
Relevant indicationAn indication caused by a discontinuity that
requires evaluation.
ScanningThe movement of a transducer relative to the test part in
order to examine a volume of the material.
Search unitAn electroacoustic device used to transmit and/or receive
ultrasonic energy.
The device generally comprises a piezoelectric element, backing,
wearface and/or wedge. Sometimes known as a probe or
transducer.
SensitivityA measure of the smallest reflector that produces a
discernible signal on the display of an ultrasonic system.
Shear wavewave motion in which the particle motion is
perpendicular to the direction of propagation (transverse wave).
Sound pathThe path of the sound energy from the time that it leaves
the transducer and reflects back to the transducer.
Skip distanceIn angle beam testing, the distance along the test
surface from sound entrant point to the point at which the sound returns
to the same surface. It can be considered the top surface distance of a
complete vee path of sound in the test material.
TransducerA piezoelectric element used to produce ultrasonic
vibrations.
Through-transmission techniqueA technique in which ultrasonic
waves are transmitted by one search unit and received by another at the
opposite surface of the material being examined.
Vee pathThe angle beam path in materials starting at the search-unit
examination surface, through the material to the reflecting surface,
continuing to the examination surface in front of the search unit, and
205
reflecting back along the same path to the search unit. The path is
usually shaped like the letter V.
Water pathThe distance from the transducer to the test surface in
immersion or water column testing.
WedgeIn angle beam examination by the contact method, a device
used to direct ultrasonic energy into the material at an angle.
Wheel search unitAn ultrasonic device incorporating one or more
transducers mounted inside a liquid-filled flexible tire. The beam is
coupled to the test surface through the rolling contact area of the tire.
Also known as a wheel probe or roller search unit.
Density Acoustic
Longitudinal Shear Surface
Metals Velocity Velocity Velocity
Impedance
g/cm3
206
g/cm2-sec
cm/s in/s cm/s in/s cm/s in/s
x105
AL 1100-0
0.635 0.25 0.310 0.122 0.290 0.114 2.71 17.20
(2SO)
AL 2014
0.632 0.2488 0.307 0.1209 N/A N/A 2.80 17.80
(14S)
AL 2024 T4
0.637 0.2508 0.316 0.1244 0.295 0.116 2.77 17.60
(24ST)
AL 2117 T4
0.650 0.2559 0.312 0.1228 N/A N/A 2.80 18.20
(17ST)
Babbitt 7.4 -
0.230 0.0906 N/A N/A N/A N/A 23.20
Bearing 11.0
Brass, Half
0.383 0.1508 0.205 0.0807 N/A N/A 8.10 31.02
Hard
Brass, Naval 0.443 0.1744 0.212 0.0835 0.195 0.0770 8.42 37.3
Bronze,
0.353 0.139 0.223 0.0878 0.201 0.0790 8.86 31.28
Phospho
Cesium
0.0967 0.0381 N/A N/A N/A N/A 1.88 1.82
(28.5oC)
207
Hafnium 0.384 0.1512 N/A N/A N/A N/A N/A N/A
Indium
0.222 0.0874 N/A N/A N/A N/A 7.30 16.21
(156oC)
Iron, Cast 0.480 0.189 0.240 0.0945 N/A N/A 7.80 37.44
Lead 5%
0.217 0.0854 0.081 0.0319 0.0740 0.0291 1.9 23.65
Antinomy
Magnesium
0.579 0.228 0.310 0.122 0.287 0.113 1.74 10.07
(AM-35)
Magnesium
0.547 0.2154 0.303 0.1193 N/A N/A 1.69 9.24
(FS-1)
Magnesium
0.567 0.2232 0.301 0.1185 N/A N/A 1.70 9.64
(J-1)
Magnesium
0.576 0.2268 0.309 0.1217 N/A N/A 1.75 10.08
(M)
Magnesium
0.580 0.2283 0.304 0.1197 N/A N/A 1.82 10.56
(O-1)
Magnesium
0.571 0.2248 0.305 0.1201 N/A N/A 1.83 10.45
(ZK-60A-TS)
0
Platinum 0.396 0.1559 N/A N/A N/A 21.4 84.74
.167
208
Plutonium
(1% 0.182 0.0717 N/A N/A N/A N/A N/A 28.6
Gallium)
Potassium
0.182 0.0717 N/A N/A N/A N/A 0.83 1.51
(100oC)
Silver, Nickel 0.462 0.1819 0.232 0.0913 0.169 0.0665 8.75 40.43
Silver,
0.476 0.1874 N/A N/A N/A N/A 8.70 41.41
German
Steel, 302
0.566 0.2228 0.312 0.1228 0.312 0.123 8.03 45.45
Cres
Steel, 347
0.574 0.226 0.309 0.1217 N/A N/A 7.91 45.4
Cres
Steel, 410
0.539 0.212 0.299 0.118 0.216 0.085 7.67 56.68
Cres
Steel, 1020 0.589 0.2319 0.324 0.1276 N/A N/A 7.71 45.41
Steel, 1095 0.590 0.2323 0.319 0.1256 N/A N/A 7.80 46.02
Steel, 4150,
0.586 0.2307 0.279 0.1098 N/A N/A 7.84 45.94
Rc14
Steel, 4150,
0.589 0.2319 0.318 0.1252 N/A N/A 7.82 46.06
Rc18
Steel, 4150,
0.587 0.2311 0.320 0.126 N/A N/A 7.81 45.84
Rc43
Steel, 4150,
0.582 0.2291 0.277 0.1091 N/A N/A 7.80 45.4
Rc64
Steel, 4340 0.585 0.2303 0.319 0.1259 N/A N/A 7.80 45.63
Thallium
0.162 0.0638 N/A N/A N/A N/A 11.9 19.28
(302oC)
209
Thorium 0.240 0.0945 0.156 0.0614 N/A N/A 11.3 27.12
Titanium
0.827 0.3256 0.516 0.2031 N/A N/A 5.15 42.59
Carbide
Uranium
0.518 0.2039 N/A N/A N/A N/A 6.03 31.24
Dioxide
References:
210
1. C.G. Windsor, Can we train a Computer to be a Skilled Inspector, Insight,
37, January 1995, p. 36.
2. A. MacNab and I. Dunlop, Review of Artificial Intelligence Applied to
Ultrasonic Defect Evaluation, Insight, 37, January, 1995, p 11.
3. K. Fukanaga, Introduction to Statistical Pattern Recognition, Academic
Press, 1990.
4. B. Eriksson and T. Stepinski, Ultrasonic Characterization of Defects, Part 2.
Theoretical Studies, SKI Report 95:21, Swedish Nuclear Power Inspectorate,
1995.
5. B. Eriksson, T. Stepinski and B. Vagnhammar, Ultrasonic Characterization
of Defects, Part 3.Experimental Verification, SKI Report 96:75, Swedish
Nuclear Power Inspectorate, 1996
6. F. Lingvall, Efficient Data Representations for Eddy Current and Ultrasonic
Applications, Lic. Thesis, Uppsala University, Signals and Systems, May
2000.
7. Friedrich Frster: On the way from the "know - how" to the "know - why" in
the magnetic leakage field method of nondestructive testing (Part
Two). Materials Evaluation. October 1985. 43 (10): 1399 - 1404
8. Kai - Hua Zhao, Xi - Mo Chen: Electromagnetism. Higher Education
Publisher, Beijing. Apr. 1978
9. Wei - Chang Zhong: Theoretical fundamentals of magnetic dipole for
longitudinal magnetization of a square steel component. (Chinese Journal of
Non - Destructive Testing. 1997.19 (4): 95 - 98; <The 9th Asia - Pacific
Conference on Nondestructive Testing in conjunction with ASNT's 1998
Spring Conference and 7th Annual Research Symposium> Anaheim,
California, USA, March 1998: 67 - 70; Materials Evaluation. 1999.57 (9):
937 - 939
10. Wei - Chang Zhong: Linear magnetic charge density along all edges of a
cuboid steel magnetized longitudinally. <ChSNDT 7th Conference on NDT
and International Research Symposium> Santou, China. Oct. 1999: 641
11. Xi - Chun Huang ed. Electromagnetic energy and electromagnetic force.
People's Education Publisher, Beijing. 1983
12. Ci - Zhang Feng ed. Static electromagnetic field. Xi - an Jiao - Tong
University Publisher, Xi - an, China. 1985
13. De - Fu Wan, Shi - Hua Luo ed. Magnetic physics. Electronic Industry
Publisher, Beijing. 1987
14. ULTRASONIC TESTING OF STEEL CASTINGS by J. D. Lavender
Research Manager. Quality Assurance Group Steel Castings Research &
Trade Association Sheffield. England
211
15. INTRODUCTION TO NONDESTRUCTIVE TESTING A Training Guide
Second Edition by PAUL E. MIX
16. NDT training program : Ultrasonic method-ASNT continuing education in
nondestructive testing
17. Nondestructive testing handbook 2nd edition, volume 10 : Nondestructive
testing overview- The American Society of Nondestructive Testing.
18. Handbook of Nondestructive Evaluation - Charles J. Hellier
19. Angle Beam Ultrasonic Testing Models and Their Application to
Identification and Sizing of Surface Breaking Vertical Cracks. Journal the
Korean Society for Nondestrwtioe Testing Vol. 72, No. 6 (2002. 12) Sung-
Jin Song' Hak-Joon Kim* * Hee Jun Jung* and Young H. Kim
20. Through-Transmission UT: An Alternative for Crack Inspection BY JIMMY
ELLIS AND STEFAN PETRUC
21. Application of the through transmission ultrasonic technique for estimation
of the phase velocity dispersion in plastic materials-R. Raiutis, A. Voleiis,
R. Kays-Ultrasound Institute, Kaunas University of Technology, Studentu
50, LT-51368 Kaunas, Lithuania
22. THE SCIENCE AND ENGINEERING OF MATERIALS - SIXTH
EDITION DONALD R. ASKELAND PRADEEP P. FIJLAY WENDELIN
J. WRIGHT
23. Non-Destructive Evaluation Methods - Thomas Luthi
24. NONDESTRUCTIVE TESTING METHODS AND NEW APPLICATIONS
Edited by Mohammed Omar
25. INTRODUCTION TO ULTRASONIC TESTING
Albert S. Birks, AKZO Nobel Chemicals, Axis, Alabama
Matthew J Colis, Columbus, Ohio
Robert E. Green, Jr., Johns Hopkins University, Baltimore, Maryland
212