Czichos H., Saito T., Smith L.E. (Eds.) Handbook of Metrology and Testing

Подождите немного. Документ загружается.

888 Part D Materials Performance Testing

16.7 Characterization of Reliability................ 949

16.7.1 Statistical Treatment of Reliability .. 951

16.7.2 Weibull Analysis........................... 952

16.7.3 Reliability Test Strategies............... 956

16.7.4 Accelerated Lifetime Testing .......... 959

16.7.5 System Reliability......................... 962

16.7.6 System Reliability Estimation

in Practice................................... 964

16.A Appendix ............................................. 967

References .................................................. 968

16.1 Nondestructive Evaluation

Nondestructive evaluation is an important method for

performance control and condition monitoring. In en-

gineering systems, ﬂaws and especially cracks in the

Light source Eye

Inspection tool

lens, endo-

scope

Fig. 16.1 Direct visual inspection

Light source

Inspection

tool

endoscope

Evaluation

Camera

Fig. 16.2 Indirect or remote visual inspection

materials of structural systems’ components can be cru-

cially detrimental to functional performance. For this

reason the detection of defects is an essential part of

quality control of engineering systems and their safe

successful use. Control techniques are often listed un-

der a variety of headings such as nondestructive testing

(NDT), nondestructive evaluation (NDE)orsometimes

nondestructive inspection. Applications of NDT tech-

niques, however, are much deeper and broader in scope

than just the detection of defects. The determination of

various material properties, such as elastic constants of

solids or the microstructure and texture of solids are also

covered under the NDT title.

According to the wide scope of this ﬁeld, a plethora

of physical methods are employed. Established methods

include radiography, ultrasound, eddy current, mag-

netic particle, liquid penetration, thermography and

visual inspection techniques. Applications of NDE in

industry are as wide ranging as the techniques them-

selves and include mechanical engineering, aerospace,

civil engineering, oil industry, electric power industry

etc. A large number of components are also in the focus

of interest because engineered structures are an inte-

gral part of the technological base necessary for our

lives and the public infrastructure. The operation of

NDT techniques in several industries is standard prac-

tice, for example to support condition monitoring for

the proper functioning of the daily use of electricity, gas

or liquids in which pressure vessels or pipes are em-

ployed and where the correct operation of components

under applied stress plays a large role for safety and

reliability.

16.1.1 Visual Inspection

Visual examination means, in principle, inspection with

eyesight [16.1,2]. (For the fundamentals of optical sens-

ing see Sect. 11.7.) In the case of NDT, visual inspection

has a much broader meaning. There are differences

in the aim of the visual inspection if: (i) only sur-

face characteristics, such as scratches, wear or corrosion

Part D 16.1

Performance Control 16.1 Nondestructive Evaluation 889

phenomena are of interest, or (ii) detection of cracks or

deformation is included.

The ﬁrst point (i) can be described as integral vi-

sual inspection and the second point (ii) as speciﬁc or

selective visual inspection. If between the eye of the

inspector and the surface of the test object the opti-

cal path is not broken then a direct visual inspection is

performed (Fig. 16.1). Such inspection is either integral

or selective. Direct visual inspection means examina-

tion using eyesight and sometimes simple tools such as

artiﬁcial illumination. Selective visual inspection uses

additional equipment including hand lenses, mirrors,

optical microscopes and telescopes and, for the docu-

mentation of the inspection results, storage media such

as photography or cameras in combination with a mon-

itor. If inspection is carried out by the evaluation of

photographs, video ﬁlms or by robotics an indirect or re-

mote visual inspection is performed (Fig. 16.2). Visual

inspection can be carried out in almost all areas where

preventive maintenance is required. Visual inspection is

the oldest and most common form of corrosion inspec-

tion, applied to surface corrosion, intergranular stress

corrosion, and some kinds of pitting. Several corrosion

phenomena are detectable using direct visual inspection

techniques. An example showing a corroded surface

is presented in Fig. 16.3. Clearly visible is the scarred

surface in the elbow region of the pipe. This exam-

ple illustrates that the visual technique is quick and in

most cases economical. Various types of failure are de-

tectable but the reliability of the inspection is highly

dependent on the skill and training of the examiner.

Fig. 16.3 Cor-

roded pipe

16.1.2 Ultrasonic Examination:

Physical Background

Sound travels in solids, liquids and gases with a velocity

depending on the mechanical properties of the material.

Imperfections such as cracks, pores or inclusions cause

sound-wave interactions which result in reﬂection, scat-

tering and general dampening of the sound wave [16.3–

6]. This dampening of the sound waves is responsible

for the distance traveled by the sound. In liquids sound

can travel large distances; on the other hand, in coarse-

grained solids the traveling distance may be only a few

centimeters due to scattering. Sound directions and the

estimation of the distance to the source are not easily

obtainable by simple methods. More precise techniques

are necessary to overcome these problems. This requires

the use of narrow sound beams.

In nondestructive evaluation (NDE) narrow sound

beams using short wavelengths can be formed when

the ultrasonic source size is much larger than the

wavelength. Nondestructive testing is carried out using

ultrasonic waves at high frequencies above the audible

range, higher than approximately 20 kHz and up to the

range of some hundred MHz.

Historically, the development of methods for ultra-

sonic materials testing in the industry started at the end

of the Second World War. Although the orientation of

bats using sound waves was an established phenom-

ena, the discovery of the piezoelectric effect by Jacques

and Pierre Currie 1880 and 1881 [16.3] was of critical

importance for the application of ultrasound. The the-

ory of sound propagation in solids was developed by

Lord Rayleigh between 1885 and 1910, as well as the

development of early electronic devices.

Sound-ﬁeld generation and reception is performed

using special devices, so called ultrasonic transducer

or ultrasonic probes. The active sound-ﬁeld genera-

tion tool is in the most cases a special ceramic with

piezoelectric properties. A piezoelectric material has the

characteristic that, if it is deformed by an external me-

chanical pressure, electric charges are produced on its

surface (Fig. 16.4). This effect was discovered in 1880

by the brothers Curie. The reverse phenomenon, ac-

cording to which such a material, if placed between

two electrodes, changes its form when an electric po-

tential is applied, was discovered soon afterwards in

1881 (Fig. 16.5). The ﬁrst effect is referred to as the di-

rect piezoelectric effect, and the second as the inverse

piezoelectric effect. The direct effect is used for mea-

suring and the inverse effect for producing mechanical

pressures, deformations and oscillations.

Part D 16.1

890 Part D Materials Performance Testing

Fig. 16.4 Direct

piezoelectric

effect

Fig. 16.5 Inverse

piezoelectric

effect

Generation of Ultrasound

Most transducers used in ultrasonic nondestructive

inspection are based on the piezoelectric effect. Gen-

eration of ultrasound is achieved using the inverse

piezoelectric effect, and the receiving procedure is

carried out with the direct piezoelectric effect. The prin-

ciple of the generation and reception of ultrasound was

explained and demonstrated with the help of the quartz

crystal. However for the time being other materials

with piezoelectric properties can be used to advantage.

Today, piezoelectric ceramic is most commonly used,

typical examples include barium titanate (BaTiO

Table 16.1 Sound velocities of piezoelectric ceramic materials used for ultrasonic transducers

Material BaTiO

PZT (P5) PbNb

PVDF

Sound velocity (m/s) 5200 4200 3300 1500–2600

lead zirconate titanate (PbZrO

–PbTiO

solid solution

PZT), lead meta-niobate (PbNb

), barium sodium

niobate (Ba

NaNb

) and the polymer polyvinyli-

dene ﬂuoride (PVDF). The manufacturing process and

material properties can be varied within certain limits,

achieving a large range of acoustical properties, which

are normally customized to a speciﬁc application. These

materials have a much higher so-called electromechan-

ical coupling factor than quartz. The frequency of

the ultrasonic signal comes from the thickness of the

piezoelectric ceramic plate. The correlation between

thickness and frequency f

isgivenby(16.1).

, (16.1)

where c is the velocity of sound, and t is the thickness

of the piezoelectric ceramic plate.

With knowledge of the sound velocity c of the

piezoelectric ceramic, the frequency f

is adjustable by

varying the thickness t of the ceramic. In Table 16.1 the

average sound velocities of common ceramic materials

are listed.

Such ceramic material, in the form of monolithic

discs, were, and are used for ultrasound generation.

In the meanwhile, also other crystal conﬁgurations

are the basis for ultrasound generation. The use of

1–3 composite materials has the advantage due to

a higher coupling coefﬁcient and a better effective-

ness for ultrasound generation in comparison with

monolithic ceramics and polymers. At the Pennsylva-

nia State University’s Materials Research Laboratory

developed a large variety of piezoelectric composite

materials that consist of a ceramic and a polymeric

phase with different connectivities [16.7, 8]. The con-

nectivity in one, two, or three dimensions in a composite

is designated as “1, 2, or 3”. Therefore, a piezocom-

posite consisting of piezoelectric ceramic rods aligned

in parallel and embedded in a polymeric resin ma-

trix is called a “1–3” composite. The ceramic rods are

connected in only one direction, namely the poled di-

rection of the material, having a connectivity of “1”.

On the other hand, the polymer phase connects in all

three dimensions, having a connectivity of “3”. Fig-

ure 16.6 shows schematically the 1–3 piezocomposite

arrangement [16.9–11]. From the picture in Fig. 16.6,

it seems to be easily to produce such ceramic ma-

Part D 16.1

Performance Control 16.1 Nondestructive Evaluation 891

terials with the knowledge of wafer technology, but

there are some difﬁculties looking to the manufacturing

processes. These difﬁculties have limited, to a de-

gree, their applicability and have increased the cost

of manufacturing piezocomposite probes. Research and

development activities in the past try to improve these

processes und to introduce piezocomposite transducers

for NDT.

Piezoelectric ceramic materials are mainly used for

the preparation of ultrasonic transducers, nevertheless

other physical properties can be utilized for generat-

ing and receiving ultrasound. Although many of these

methods produce weaker signals than are obtainable by

the piezoelectric effect, they offer a number of advan-

tages that in special cases makes their application in the

testing of materials useful. For many of these alterna-

tive techniques the energy is transmitted by electrical

or magnetic ﬁelds, which in principle make mechani-

cal contact with the metallic test piece unnecessary. The

conversion into, or from, acoustic energy takes place in

the surface of the workpiece concerned. Compared with

the piezoelectric oscillator, which requires direct cou-

pling, normally with a suitable medium, e.g. water, to

the workpiece, the surface of the workpiece forms in

this a part of the acoustic transducer.

Other possibilities for the generation of ultrasound

are the use of mechanical shock or friction effects,

the application of thermal effects by heating the sur-

face of a body suddenly (heat shock) the exploit of

electrodynamic effects based on the Lorentz force and

the magnetostrictive effect. Apart from the generation

of ultrasound using the electrodynamic effect the oth-

ers methods play at present a small role into the NDE

ﬁeld.

Reﬂection and Transmission

of Ultrasonic Waves at Boundaries

Normal Beam Incidence. The behavior of an ultra-

sonic wave in an unlimited material is only possible

on a theoretical level because in reality every mater-

ial has boundaries. At such boundaries the traveling

wave is disturbed. If the boundary is in a direct con-

tact with a vacuum, transmission of the ultrasonic wave

through the boundary is impossible; total reﬂection

occurs. Consider an incident longitudinal plane wave

traveling perpendicular to a boundary or interface be-

tween two materials, as demonstrated in the schematic

diagram in Fig. 16.7 [16.3, 4]. For symmetrical reasons

only the reﬂected P

and transmitted P

waves are pos-

sible. The part of the reﬂected wave is characterized by

the reﬂection coefﬁcient R and the part of the trans-

Resin between the

piezoelectric rods

Piezoelectric material

Fig. 16.6 Piezocomposite arrangement

mitted wave by the transmission coefﬁcient T.Inthis

special case the reﬂection and transmission coefﬁcients

only depend on the acoustic impedance W =ρc

where

ρ is the density of the material and c

the sound ve-

locity of the longitudinal wave. The equation for the

calculation of the two coefﬁcients are as follows

R =

−W

and T =

. (16.2)

The indexes 1 or 2 are used to deﬁne the two media

(Fig. 16.7). It is clearly visible from these equations that

100% of the sound is transmitted through the interface

if W

is equal W

. In reality what happens for a steel–

water interface is a combination of impedances often

given in ultrasonic examination (W

=45 × 10

Ns/m

for steel and W

= 1.55 × 10

Ns/m

for water) for

a steel–water interface. With these values a reﬂection

Medium 1 Medium 2

Boundary

Transmitted wave P

Reflected wave P

Incident wave P

= ρ

Fig. 16.7 Normal incidence of a sound wave on a boundary

Part D 16.1

892 Part D Materials Performance Testing

Sound pressure

–1

–2

–1

–2

Sound pressure

Reflected

wave

a) b)

Reflected

wave

Incident wave

Incident

wave

Steel Water SteelWater

Transmitted

wave

Transmitted

wave

Fig. 16.8 Sound pressure values by reﬂection at a steel/water inter-

face

coefﬁcient of −0.935 and a transmission coefﬁcient

of 0.0065 is calculated. That means that 93.5% of

the sound ﬁeld is reﬂected at the steel–water interface

(Fig. 16.8). The negative value of the reﬂection coefﬁ-

cient means that the phase of the reﬂected ultrasonic

wave has changed by 180

◦

compared with the inci-

dent wave. If we assume an interface steel–air then R

is −0.99998, i. e. nearly 100%. Such a material combi-

nation is given for instance if a crack is located in a steel

component. Therefore, when the sound wave strikes the

crack surface at a normal angle it can be detected with

a high probability.

Equations (16.2) are valid if the sound wave is trav-

eling perpendicular to the interface. The behavior is

Medium 2

Medium 1

Incident

longitudinal wave

Transmitted

longitudinal

wave

Transmitted

shear wave

Reflected

longitudinal

wave

Reflected

shear wave

Boundary length

greater than

sound wavelength

Fig. 16.9 Wave mode conversion at a boundary

more complex if the sound strikes the surface with an

angle > 0

◦

(Fig. 16.9). A mode conversion as illustrated

in the ﬁgure must be assumed, i. e. an incidence longitu-

dinal wave is reﬂected at the boundary as well as a shear

wave. A longitudinal and a shear wave are transmitted

into the second medium. This circumstances must be

taken into consideration by the calculation of the re-

ﬂection and transmission coefﬁcient. The situation is

similar if an incidence shear wave is assumed. For each

case presented in Fig. 16.9 two reﬂection and two trans-

mission factors are obtained.

Inclined Beam Incidence. If an ultrasonic longitudinal

plane wave strikes a liquid–solid interface at an an-

gle α as demonstrated in Fig. 16.10 one reﬂected wave

and two transmitted waves are generated. The reﬂected

angle α is equal to the incidence angle whereas the

transmitted angles are different. These two angles are

dependent on the sound velocity for the different wave

types corresponding to the two materials. The calcula-

tion of these angles can be carried out using Snell’s law

sin α

. (16.3)

Using (16.3), all angles can be calculated if the sound

velocities and one angle are known. In the follow-

ing some consequences derived from the Snell’s law

are considered. The Plexiglas–steel interface is impor-

tant for ultrasonic angle beam probe construction and

therefore we will regard this material combination in

the following. The sound velocity for the longitudinal

wave is on average 2740 m/s for Plexiglas (perspex) and

5920 m/s for steel as well as 3255 m/s for the shear

wave in steel. Consideration of Snell’s law indicates

that two critical angles exist with respect to the refrac-

tion process on the interface. The ﬁrst critical angel is

given when α

=90

◦

, i. e. the total reﬂection of the lon-

gitudinal wave. In this case only shear waves exist in

material 2. The second critical angle occurs when the

shear refracted angle is 90

◦

. If only shear waves will be

used in the practical application the Plexiglas wedge an-

gle β should be between 27.6 and 57.3

◦

. These values

are calculated using the above printed sound velocities.

Angle beam probes generating shear waves used for the

examination therefore have angles of incidence between

35 and 80

◦

Ultrasonic Transducers

In the following, the description of generating and re-

ceiving ultrasound with transducers (probes) will focus

on piezoelectric ceramic material [16.3, 5, 6] due to the

Part D 16.1

Performance Control 16.1 Nondestructive Evaluation 893

importance in practical applications. Such ultrasonic

transducers are optimized for different areas of appli-

cation [16.12–14] but the basic construction principle

nearly always remains the same. Figure 16.11 shows

schematically typical commercial ultrasonic transduc-

ers as a part of the basic equipment for ultrasonic

examinations.

A longitudinal wave is generated using a straight

beam probe, presented in a schematic sketch in

Fig. 16.11a. The piezoelectric ceramic plate is behind

a protective coating (wear shoe/plate) and the back-

ing material (damping element) is largely responsible

for the pulse form. Together with an electrical adapta-

tion the ultrasonic pulse can be formed and optimized

to a particular application. As explained the thickness

of the ceramic delivers the middle frequency of the

pulse. This kind of probe has such an construction

that the waves have a perpendicular (straight) radia-

tion in relation to the test object’s surface. The size

of the piezoelectric ceramic together with the middle

frequency will determine important ultrasound probe

parameters, including the near-ﬁeld length, divergence

angle and beam diameter, which are required to perform

reliable inspection. All these parameters of an ultrasonic

probe can be measured or calculated using simple equa-

tions. Figure 16.11b shows the principle drawing of an

angle beam probe for shear-wave generation. The shear

waves are generated due to refraction of the incident

longitudinal wave on the coupling surface. A common

ultrasonic angle beam probe can be employed to gener-

ate shear waves only. This is achieved if the wedge has

an angle between the ﬁrst and the second critical an-

gle. So far, for the material combination perspex/carbon

steel, conventional probes are available generating shear

a) b)

Wire leads

Metal casing

Socket

Electrical

matching

Acoustic

insulation

Damping element

Electrical matching

Damping element

Perspex wedge

Object to be examined

Transducer

Wear plate

Fig. 16.11a,b Normal or straight beam probe (a) and angle beam probe (b)

Medium 2

Medium 1

(liquid)

Incident

longitudinal wave

Transmitted

longitudinal

wave

Transmitted

shear wave

Reflected

longitudinal wave

Boundary

Fig. 16.10 Inclined incidence at a liquid/solid boundary

waves with incidence angles of 45, 60 and 70

◦

. In recent

years with the development of electronic beam form-

ing so-called phased array probes [16.3] with variable

incidence angles have become available. The princi-

ple of the phased array probe can be explained with

the help of Fig. 16.12. The crystal is divided into sev-

eral elements. In the given example, eight elements

are illustrated. Each element is connected with the

transmitter/receiver part of special ultrasonic control-

ling equipment (phased array equipment). The time for

transmitting the main bang (transmitter pulse) is con-

trolled by the equipment in such a way that different

transmitter times can be achieved between the indi-

vidual elements. When the delay time Δt between the

elements is constant decreasing or increasing a varia-

tion of the incidence angle will occur. If Δt follows

Part D 16.1

894 Part D Materials Performance Testing

Transmitter pulses

8 crystal elements

Sound field direction

Variation of the incidence angle Variation of the focal distance

Fig. 16.12 Principle of a phased array probe

Initial

pulse

Echo

rear wall

Echo defect

Amplitude varies

with defect

Fig. 16.13a,b Pulse-echo technique (a) and through-transmission

technique (b)

a predetermined curvature (like a lens) the sound ﬁled

can be focused at a certain distance. Further a combina-

tion of both is also possible. With such a phased array

probe all incidence angle can be realized for longitudi-

nal as well as for shear waves. For the second case, if

Δt is speciﬁed by the curvature of a lens, a variation of

the focal distance is given by changing the curvature.

With advanced phased array equipment both changing

the angle and focal distance can be carried out in one

stage.

In the following some of sound-ﬁeld distributions of

different kinds of probes are presented.

Sound Propagation in Solids

Sound propagation in materials is always connected

with an interaction of the ultrasonic pulse and the struc-

ture of the material. Especially for solid materials such

as steel the interaction with the grain size (scattering

and reﬂection) is sometimes not negligible mainly if

the wavelength and the grain size are within the same

dimensional range. Due to this interaction the sound

losses energy i. e. an attenuation of the sound will hap-

pen along the propagation direction. Another effect is

the frictional motion of the particles of the solid which

also has an attenuation effect but is based on the ab-

sorption of the ultrasonic intensity. Therefore the total

attenuation of the ultrasonic intensity is given as the

sum of the scattering and absorption coefﬁcient (16.4)

μ =μ

+μ

. (16.4)

This attenuation must be taken into consideration during

the examination.

Principle of the Pulse-Echo

and Through Transmission Methods

Defects in materials caused scattering and reﬂection

of the ultrasonic wave and the detection of the re-

ﬂected or transmitted waves allows the location of the

defect. A general illustration of the pulse-echo and

through transmission methods is shown in Fig. 16.13.

The through transmission normally requires access to

both sides of a component and is applied if the back-

reﬂection method cannot provide sufﬁcient information

about a defect. For example, when small defects are

detectable which do not give adequate reﬂection signals.

The pulse-echo technique requires access to only

one side of the component. This is one advantage of this

method compared with the through-transmission tech-

nique. Another advantage of the pulse-echo method is

given due to the time-of-ﬂight measurement because

the defect location (distance from the coupling surface)

can be estimated from the time required for the pulse

to travel from the probe to the discontinuity and back

and can be displayed on an oscilloscope. If the hori-

zontal sweep of the oscilloscope is calibrated with the

sound velocity into the material to be inspected, the re-

quired travel time from the probe to the reﬂector and

back corresponds to the distance of the reﬂector from

the coupling surface. This method is used extensively

in practice.

16.1.3 Application Areas

of Ultrasonic Examination

The application range of ultrasonic examination is very

broad. It covers, for example, wall thickness measure-

ment of structural components and the detection of wall

Part D 16.1

Performance Control 16.1 Nondestructive Evaluation 895

Probe

Echo

height

Object to be examined

Back wall

Sound beam

Initial pulse

Transit time of US pulse

897654321010

Transit time

Back wall echo

Fig. 16.14 Principle of wall thickness measurements

thickness losses due to corrosion, erosion, wear etc.

Other application areas include the measurement of ma-

terial properties such as sound velocity, elastic constants

and the detection of ﬂaws, e.g. cracks, slag inclusions,

lack of fusions, pores. The following examples illustrate

the application of ultrasonic examination for perfor-

mance control of components and engineering systems

in various areas of technology and industrial sectors.

Wall Thickness Measurement of Components

During the lifetime of a component their surface are

subjected to environmental stresses. Depending on the

environment corrosion phenomena must be taken into

consideration. A reduction of wall thickness can oc-

cur and as a consequence the safety factor that was

the basis for the construction may no longer be valid.

The principle of thickness measurement is illustrated

in Fig. 16.14 [16.15]. A straight-beam longitudinal

probe is coupled on the surface and the pulse echo from

the parallel ﬂat opposite surface (back wall) is received

and displayed on the screen. The transit time of the

pulse corresponds with the wall thickness if the equip-

ment was calibrated with the correct sound velocity.

This method is sometimes known as a simple length

measurement.

897654321010

8.4 0 °, 50

Fig. 16.15 Wall thickness measurement using multiple

echoes

Another method is to analyze multiple echoes which

are observed, i. e. multiple reﬂections of an acoustic

pulse propagating between the coupling surface and

the back wall (Fig. 16.15). Since the constant distances

between the echoes corresponds to the specimen thick-

ness d, then the whole transit time of the multiple

echo indications on the screen is also a multiple of the

wall thickness. An example of a corroded aluminum

plate from the aircraft industry is plotted in Fig. 16.16.

>1.8 mm

<1.7 mm

<1.6 mm

<1.5 mm

<1.3 mm

<1.2 mm

<1.4 mm

<1.1 mm

1.8 mm

Fig. 16.16 Wall thickness measurement

Part D 16.1

896 Part D Materials Performance Testing

For wall thickness estimation the measurement was

carried out using a mechanized ultrasonic immersion

technique. The different gray scales correspond with

different wall thicknesses.

All thickness measurement using ultrasound have in

common that the coupling conditions of the transducer

must be performed in the same manner, however also

care must be taken to use the correct sound velocity in

the material.

Weld Inspection

Ultrasonic weld inspection is done extensively around

the world, and the pertinent literature is published

in a multitude of journals, proceedings and technical

reports [16.14]. Usually, a shear-wave angle beam trans-

ducer is used to examine the weld and the adjacent

region in a halfskip and skip distance (Fig. 16.17). In

the ﬁgure the method of examining the root, the interior

and the top layer of the weld is presented. The skip dis-

tance can be calculated (16.5) where d is the thickness

of the specimen and α the beam angle

=2d tan α. (16.5)

The examination of welds requires that the skip dis-

tance, the beam angle, the beam spread and the probe

index point are well known. Further information, such

as the surface conditions, coupling media to be used,

system of marking welds, system of indicating defects

found, access conditions for examination of welds and

use of other, possible NDT methods should be known

before examination. The number of beam angles de-

pends strongly on the wall thickness of the component

to be examined. After the checking sensitivity and set-

ting up of the whole ultrasonic system (transducer and

equipment) the inspection procedure can be carried out

for the detection of longitudinal defects. The shear-

wave probe should be placed on the coupling surface

and kept during the movement perpendicular to the weld

centerline while swiveling the probe head lightly. The

ultrasonic probes used for weld inspection must take

Half skip Full skip

Fig. 16.17 Weld inspection

Diameter: 1932 mm

Length: 4173 mm

Fig. 16.18 Turbine shaft

into consideration the geometry of the component and

the weld. Different weld joints include square groove,

bevel groove, v-groove, u-groove and weld types in-

cluding butt, lap, T, edge, and corner; although this list

is not complete, different ultrasonic probes with special

sound-ﬁeld characteristics must be used to overcome

this examination problem. Examination procedure and

problems are described in several standards, codes and

guidelines. For example the detection of defects direct

or in the neighborhood of the coupling surface require

so-called transmitter receiver probes. The main mea-

suring principle that justiﬁes the use of such probes

is the separation of the transmitter and receiver func-

tion which is accomplished by using two crystals in

the probe. However the choice of the optimal inspec-

tion techniques for reliable examination results requires,

in addition to the knowledge of the standards, a skilled

examiner.

Examination of Forged

and Rolled Machinery Parts

With regard to high machining cost, the examination

of large forgings should start in the earliest stages of

production. In recent times the inspection of compo-

nents such as turbine shafts, railway wheels etc. are

carried out using mechanized testing. Figure 16.18

shows an example of the examination of a turbine shaft

with a diameter of about 2 m. The examination was

performed using phased array probes [16.16]. During

the shaft rotation the incidence angles for longitudi-

nal waves were changed in a range between −30

◦

and

◦

with steps of 3

◦

. For every full circle a echotomo-

gram was reconstructed. After the whole examination

of the shaft the defect situation in the shaft, in the ra-

Part D 16.1

Performance Control 16.1 Nondestructive Evaluation 897

Phased array probe Phased array probe

Disc breakWheel

Fig. 16.19 Railway axel inspection with phased array

probes

dial and axial directions, was visualized with the help

a computer program called glassy shaft.

Another example where mechanized testing is used

is the inspection of railroad axels with phased array

probes (Fig. 16.19). Two coupled phased array probes on

the axel surface are presented in the ﬁgure (two probes

for each side). Thus for this special application case

four phased array probes are necessary for the examina-

tion of the area beneath the wheels and the disc brakes.

Phased array probes with the possibility of steering the

incidence angle have advantages because of the limited

accessibility between the wheel and the disc brake.

Inspection

of Semi-ﬁnished Industrial Products

AwideﬁeldforNDT is the inspection of semi-ﬁnished

products such as rods, pipes plates, casting products

etc. Defects in this group of products can either be on

or near the surface and internal (embedded) defects.

In the case of steel rods, manual ultrasonic inspec-

tion using the pulse-echo technique is implemented in

practice as well as mechanized inspection. Straight and

angle beam probes are for the detection of the different

defects such as shrinking cracks, inclusions, cavities,

pores etc. There are several different techniques for

identifying defects, which can be achieved by mecha-

nized inspection. One main difference is the handling

of the rod during the inspection, e.g. whether only lon-

gitudinal movement or rotation of product is performed.

The probe arrangement will depend on the type of me-

chanical rod movement that takes place. A more or less

simple movement of the rod can be carried out if phased

array probes or real-time scanners are used. The reason

Fig. 16.20 Array transducers for the examination of longi-

tudinal welds in pipes (GE Inspection Technologies)

is, that a complicated mechanical movement is substi-

tuted by the sound movement. One example of such

a transducer arrangement is presented in Fig. 16.20.

The geometry of the array transducer is adapted to

the geometry of the pipe to be inspected. In this case

the longitudinal weld of pipes will be examined. In

addition with the help of some of ultrasonic transduc-

ers a simple mechanized movement of a rod is also

possible. Three probe holders in a circumferential po-

sition of 120

◦

in each and the combination of angle

and straight-beam transducers sufﬁciently guarantee the

detection of embedded and near-surface longitudinal

defects.

16.1.4 Magnetic Particle Inspection

The physical principles of magnetic particle inspection

have been well known for a long time. At the begin-

ning of the 20th century it was discovered that, when

steel parts are machined while held in a magnetic chuck,

the iron grindings formed patterns around cracks on the

surface. The disturbance of magnetic ﬁelds by surface

defects in magnetized ferromagnetic specimens is the

Magnetic field Iron powder Crack

Fig. 16.21 Principle of magnetic particle inspection

Part D 16.1