Designation: E 2192 – 02

Standard Guide for

Planar Flaw Height Sizing by Ultrasonics1
This standard is issued under the fixed designation E 2192; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (e) indicates an editorial change since the last revision or reapproval.

1. Scope 3.2.1 bi-modal—ultrasonic examination method that uti-

1.1 This guide provides tutorial information and a descrip- lizes both the longitudinal (L-wave) and shear (S-wave) modes
tion of the principles and ultrasonic examination techniques for of propagation in order to estimate or measure flaw height.
measuring the height of planar flaws which are open to the 3.2.2 corner reflector—the reflected ultrasonic energy re-
surface. The practices and technology described in this stan- sulting from the interaction of ultrasound with the intersection
dard guide are intended as a reference to be used when of a flaw and the component surface at essentially 90 degrees.
selecting a specific ultrasonic flaw sizing technique as well as 3.2.3 creeping wave—a compression wave that travels in a
establishing a means for instrument standardization.2 solid immediately adjacent to a free boundary and generates a
1.2 This standard guide does not provide or suggest accu- shear mode “headwave” (q.v.) traveling away from the bound-
racy or tolerances of the techniques described. Parameters such ary at the critical angle. (Some users reserve the term lateral
as search units, examination surface conditions, material com- wave for the creeping wave following a flat parallel surface and
position, etc. can all have a bearing on the accuracy of results. the creeping wave is used for those waves following curved
It is recommended that users assess accuracy and tolerances surfaces).
applicable for each application. 3.2.4 diffraction—a wave front whose direction has been
1.3 This document does not purport to provide instruction to changed by an obstacle or other non-homogeneity in a me-
measure flaw length. dium, other than by reflection or refraction.
1.4 This standard guide does not provide, suggest, or 3.2.5 doublet—two ultrasonic signals that appear on the
specify acceptance standards. After flaw-sizing evaluation has screen simultaneously and move in unison as search unit is
been made, the results should be applied to an appropriate code manipulated toward and away from the flaw. During tip-
or standard that specifies acceptance criteria. diffraction flaw sizing, the flaw tip signal and flaw base signal
1.5 This standard does not purport to address all of the (corner reflector) will appear as a doublet.
safety concerns, if any, associated with its use. It is the 3.2.6 echo-dynamic—the amplitude versus time of arrival
responsibility of the user of this standard to establish appro- (or distance through the component) curve created by the
priate safety and health practices and determine the applica- ultrasonic signal as the search unit is moved perpendicular to
bility of regulatory requirements prior to use. the reflector toward and away from the flaw.
3.2.7 far-surface—the surface of the examination piece
2. Referenced Documents opposite the surface on which the search unit is placed. (For
2.1 ASTM Standards: example, when examining pipe from the outside surface the
E 1316 Terminology for Nondestructive Examinations3 far-surface would be the inside pipe surface).
3.2.8 focus—the term as used in this document applies to
3. Terminology dual crossed-beam search units that have been manufactured so
3.1 Definitions—Related terminology is defined in Termi- that they have a maximum sensitivity at a predetermined depth
nology E 1316. or sound path in the component. Focusing effect may be
3.2 Definitions of Terms Specific to This Standard: obtained with the use of dual-element search units having both
refracted and roof angles applied to each element.
3.2.9 headwave—a shear wave that is generated by mode
1
This guide is under the jurisdiction of ASTM Committee E07 on Nondestruc- conversion when a compression wave travels at a grazing angle
tive Testing and is the direct responsibility of Subcommittee E07.06 on Ultrasonic
Method.
on a free solid surface.
Current edition approved April 10, 2002. Published June 2002. 3.2.10 insonify—the interrogation of an area in the exami-
2
This Standard Guide is adapted from material supplied to ASTM Subcommittee nation piece with ultrasonic energy.
E07.06 by the Electric Power Research Institute (EPRI).
3
Annual Book of ASTM Standards, Vol 03.03.

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.

1
 E 2192
3.2.11 near-surface—the surface of the examination piece examined using this guide with appropriate standardization
on which the search unit is placed. (For example, when reference blocks. The practices described are applicable to both
examining pipe from the outside surface the near-surface manual and automated examinations.
would be the outside pipe surface). 5.3 The techniques recommended in this standard guide use
3.2.12 sizing—measurement of the through-wall height or Time of Flight (TOF) or Delta Time of Flight (· TOF) methods
depth dimension of a discontinuity or flaw. to accurately measure the flaw size. This guide does not include
3.2.13 time of flight—the sound path measurement of time the use of signal amplitude methods to determine flaw size.
for the reflected or diffracted energy from a flaw. 5.4 Generally, with these sizing methods the volume of
3.2.14 30-70-70—term that is applied to the technique (and material (or component thickness) to be sized is divided into
sometimes the search unit) using an incident angle that thirds; the inner 1⁄3, the middle 1⁄3 and the outer 1⁄3. Using the
produces a nominal 70° L wave in the examination piece. far-surface Creeping Wave Method the user can qualitatively
Provided that a parallel far-surface exists, the 30° shear wave, segregate the flaw into the approximate 1⁄3 zone.
produced simultaneously at the refracting interface, reflects as
5.5 The sizing methods are used in 1⁄3 zones to quantita-
a 30° shear wave and generates a nominal 70° L wave as a
tively size the crack, that is, Tip-diffraction for the inner 1⁄3,
result of mode conversion off the far-surface. The 70° L wave
Bi-Modal method for the middle 1⁄3, and the Focused Longi-
reflects off a planar flaw and is received by the search unit as
tudinal Wave or Focused Shear Wave Methods for the outer 1⁄3.
a 70° L wave.
These 1⁄3 zones are generally applicable to most sizing appli-
4. Summary of Guide cations, however, the various sizing methods have applications
outside these 1⁄3 zones provided a proper reference block and
4.1 This guide describes methods for the following flaw
technique is demonstrated.
sizing techniques.
4.1.1 Far-surface creeping wave or mode conversion
6. Ultrasonic Flaw Sizing Methods
method,
4.1.2 Flaw-tip-diffraction method, 6.1 30-70-70 Mode Conversion or Far-surface Creeping
4.1.3 Dual element bi-modal method, and Wave Method—The far-surface Creeping Wave or 30-70-70
4.1.4 Dual element, (focused) longitudinal wave or dual Mode Conversion method (as illustrated in Fig. 1) provides
element, (focused) shear wave methods (see 3.2.9). qualitative additional depth sizing information. This method
4.2 In this guide, ultrasonic sound paths are generally has considerable potential for use when approximating flaw
shown diagrammatically by single lines in one plane that size, or, determining that the flaw is far-surface connected.
represent the center of the ultrasonic energy. 6.1.1 Excitation of Creeping Waves—The excitation of
4.3 Additional information on flaw sizing techniques may refracted longitudinal waves is always accompanied by re-
be found in the references listed in the Bibliography section. fracted shear waves. In the vicinity of the excitation, the
separation between these two wave modes is not significantly
5. Significance and Use distinct. At the surface, a longitudinal wave cannot exist
5.1 The practices referenced in this document are applicable independently of a shear wave because neither mode can
to measuring the height of planar flaws open to the surface that comply with the boundary conditions for the homogeneous
originate on the far-surface or near-surface of the component. wave equation at the free surface alone; consequently, the
These practices are applicable to through-wall sizing of me- so-called headwave is formed. The headwave is always gen-
chanical or thermal fatigue flaws, stress corrosion flaws, or any erated if a wave mode with higher velocity (the longitudinal
other surface-connected planar flaws. wave) is coupled to a wave mode with lower velocity (the
5.2 The techniques outlined describe proven ultrasonic flaw direct shear wave) at an interface. The existence of the indirect
sizing practices and their associated limitations, using refracted shear wave, headwave, has also been proven experimentally by
longitudinal wave and shear wave techniques as applied to Schlieren optics. The longitudinal wave continuously energizes
ferritic or austenitic components. Other materials may be the shear wave. It can be concluded that the longitudinal wave,

FIG. 1 Wave Generation for the Far-surface Creeping Wave/30-70-70 Mode-Conversion Search Unit

2
 E 2192
which in fact “creeps” along the surface, is completely attenu- 4. The presence of the mode-converted echo is a strong
ated a short distance from the location of the excitation. (See indication of a flaw with a height greater than 10 to 20 % of the
Fig. 2 for generation of the near-side creeping wave). With the wall thickness. In the case of smooth or at least open flaws,
propagation of the near-surface creeping wave and its continu- amplitude versus height function curves can give a coarse
ous conversion process at each point it reaches, the energy estimate of flaw height.
converted to shear is directed into the material as shown in Fig. 6.1.3.3 Far-Surface Creeping Wave Signal—If a far-surface
3. Thus, the wave front of the headwave includes the head of connected reflector is within the range of sensitivity (as
the creeping wave, direct and indirect shear waves. described above), the far-surface creeping wave will be re-
6.1.2 Far-Surface Creeping Wave Generation—When the flected and mode converted into the headwave or shear wave
headwave arrives at the far-surface of the component, the same directed to the search unit (Fig. 5). Since the far-surface
wave modes will be generated which were responsible for creeping wave is not a surface wave, it will not interact with
generating the shear wave energy, due to the physical law of weld root convexity and will not produce an indication from
reciprocity. Thus, the indirect shear wave and part of the direct the root as shown by position 1 in Fig. 6. However, if the
shear wave will convert into a far-surface creeping wave and a search unit is moved too far toward the weld centerline, the
70-degree longitudinal wave. The far-surface creeping wave direct shear wave beam could result in a root signal, but there
will be extremely sensitive to small surface-breaking reflectors is at least 5 mm (0.2 in.) difference in positioning as shown in
and the longitudinal wave will be engulfed in a bulk longitu- Fig. 6. The far-surface creeping wave signal is a clear, sharp
dinal beam created by beam spread. Additionally, these reflec- signal with a larger amplitude than the mode converted signal.
tion mechanisms are responsible for a beam offset so that there It does not have as smooth an echo-dynamic behavior as does
is a maximum far-surface creeping wave sensitivity at about 5 the mode converted signal, and it cannot be observed over as
to 6 mm (0.20 to 0.24 in.) from the ideal conversion point on long a distance as shown in Fig. 7.
the far surface. The sensitivity range of the far-surface creeping 6.2 Tip-Diffraction Method—Ultrasonic diffraction is a phe-
wave extends from approximately 2 to 13 mm (0.080 to 0.52 nomenon where ultrasound tends to bend around sharp corners
in.) in front of the index point. The far-surface creeping wave, or ends of an object placed in its path, as illustrated in Fig. 8.
as reflected from the base of a far-surface notch or flaw, will While the flaw tends to cast a shadow, diffraction occurs at the
convert its energy into a headwave since the same principles flaw tips and ultrasonic energy is bent to fill part of the shadow
apply as established earlier for the near-surface creeping wave. region. Sharp edges are diffraction centers tending to radiate
The shear wave will continue to convert at multiple V-paths if spherical or cylindrical wave fronts as though they were
the material has low attenuation and noise levels. actually ultrasonic point or line sources. If the screen signals
6.1.3 Typical Echoes of the Far-Surface Creeping Wave/30- correlating to these diffraction centers are identified, it is
70-70 Mode Conversion Technique—When the search unit possible to determine their positions relative to the thickness of
approaches a far-surface connected reflector, three different the component. The tip-diffraction method relies on this
signals will occur in sequence: (1) 70-degree longitudinal wave principle. Although the tip-diffraction concept sounds simple,
direct reflection; (2) 30-70-70 mode-converted signal; and (3) there are many other signals that may complicate screen
A far-surface creeping wave signal, as a result of mode interpretation. This is due to the fact that the ultrasound/planar
conversion of the indirect shear wave. flaw interaction is very complex. When ultrasound strikes a
6.1.3.1 Direct Longitudinal Wave Signal—If the flaw ex- flaw, specular reflection from the main plane of the flaw and
tends to within approximately 0.375 to 0.625 in. (9.5 to 15.9 texture reflections from flaw surface facets occur in addition to
mm) of the scanning surface (near surface), the direct longitu- diffraction and mode conversions. There are two standardiza-
dinal wave will reflect from the upper extremity of the flaw tion and measuring techniques for tip-diffraction sizing: (1)
face, which is very similar to the high-angle longitudinal wave The Time of Flight (TOF) technique that measures the arrival
sizing method discussed later. time of the tip-diffracted signal from the top of the flaw and
6.1.3.2 Mode Converted Signal—If the flaw exceeds a locates the top of the flaw with respect to the near surface; and
height of 10 to 20 % of the wall thickness, an indication from (2) The Delta Time of Flight (· TOF) technique that measures
the mode converted signal will occur at a typical wall the difference in arrival time of the tip-diffracted signal and the
thickness-related position. This mode converted signal results corner reflector signal at the far surface.
from the headwave or direct shear wave, which mode converts 6.2.1 Time of Flight (TOF) Sizing Technique—The TOF
the 70-degree longitudinal wave that impinges on the reflector sizing technique is a tip-diffraction technique that takes advan-
at its highest part; it is reflected as a 70-degree longitudinal tage of uniquely locating the flaw tip. The signal from the flaw
wave back to the search unit as depicted by position 1 in Fig. tip is peaked (maximized), and its arrival time or sound path is

FIG. 2 Near-Surface Creeping Wave Occurs for a Short Distance in Association with the Incident Longitudinal Wave

3
 E 2192

FIG. 3 Generation of S-Waves (Headwaves) by an L-Wave with Grazing Incidence

1—Mode-Converted Signal
2—Far-Surface Creeping-Wave Signal
FIG. 4 Search Unit Index Point Position

FIG. 5 Generation of Far-Surface Creeping Wave Signal

measured without regard to the arrival time of other signals. technique is illustrated in Fig. 9. Note that here the second
This time of flight or sound path is then a direct measurement half-V path is possible also. When the search unit is moved
of the remaining ligament (material) above the flaw, or the away from the flaw, the tip echo may again be obtained after
distance from the flaw tip to the examination surface. This the tip-diffracted signal reflects off the opposite surface of the

4
 E 2192

1—Flaw indication maximized for mode-converted wave signal

2—Flaw indication maximized for creeping-wave signal
FIG. 6 Far-Surface Creeping Wave Search Unit Position Related to Index Point

component. With the second half-V path technique, the tip height, it is necessary to note the difference in the time of
signal will occur later in time than the signal from the flaw arrival between the two signals, then apply the following
corner reflector. formula:
NOTE 1—It is very important that the user be extremely conscious of v~dt!
h 5 2cosu
the weld geometry when using the second half-V path since, for example,
the counterbore can exaggerate flaw height.
NOTE 2—Longitudinal waves should not be applied when practicing the where:
second half-V path technique as this can cause mode conversions that may h = flaw height,
interfere with the ability to interpret the instrument display. v = ultrasonic velocity in the material,
6.2.2 Delta Time of Flight (· TOF) Technique—The · TOF dt = difference in arrival time, and
Technique is applied by observing the arrival time difference u = refracted beam angle.
between the flaw corner reflector signal and the diffracted Alternately, the ultrasonic instrument may be standardized to
signal from the flaw tip while both are simultaneously present read directly in flaw height. This standardization method will
on the ultrasonic instrument display. While using this tech- be addressed in the standardization section. Separation be-
nique, the ultrasonic beam diameter must be greater than the tween the doublets should remain constant as the signals move
projected height of the flaw (actual height multiplied by the across the screen. The echo dynamic of the doublet is asyn-
sine of the refracted beam angle) and the flaw must be chronous; however, since it is the fixed interval between the
essentially perpendicular to the examination surface. In this doublet arrival times that is measured, it is not necessary to
situation, the tip-diffracted signal will occur earlier in time due maximize the response from either signal. This technique
to its shorter sound path. The tip signal amplitude is very small allows measurement when the weld crown is wide, preventing
in comparison to the flaw corner reflector signal; and the flaw maximization of the tip signal. It may also be possible to note
tip and corner signals are out of phase due to one signal being a tip signal after reflection from the back surface (second
diffracted and the other reflected twice. To measure flaw half-V path). The principles are the same as for the first half-V

5
 E 2192

FIG. 7 Echo-Dynamic Behavior of Mode-Converted Echo Signal and Far-Surface Creeping-Wave Signal

FIG. 8 Corners or Ends of Reflectors are Diffraction Centers and Tend to Radiate Spherical or Cylindrical Waves

path except that the tip signal will appear later in time than the signal need not originate singly from diffraction, since reflec-
corner reflector signal. Whether using the first or second half-V tion can also occur very near the flaw tip. In fact, reflection is
path, accuracy of the height measurement depends on the flaw the mechanism that will primarily be observed when using
orientation. If the flaw is vertical, then the measurement is notched reference blocks. It is reasonable to expect some
accurate. If the flaw is oriented toward the search unit, the first reflection to occur at an actual flaw tip. The associated rough
half-V path measurement will overestimate the height and the texture will often act as a good scattering center. It should be
second half-V path measurement will underestimate the height. noted, however, that this may not be true in every case and the
The opposite occurs for flaws oriented away from the search amplitudes of the signals received may be 20-30 dB below the
unit. flaw corner-reflector signal amplitude. Each component and
6.2.3 Application Considerations—For all of the physics material type examined should be considered as a separate
involved in tip diffraction, the method relies on the user’s examination problem. The flawed area should be adequately
ability to uniquely identify the location of the flaw tip. The scanned so that all signals, which occur in the region, can be

6
 E 2192

FIG. 9 Various Metal Paths (MP) from Different Search Unit Positions Used in the TOF Technique

identified. Care should be taken to define the tip signal since unit movement as it relates to flaw height by becoming familiar
some geometries or weld flaws produce signals that can be with the characteristics observed when sizing notches of
readily confused with the true tip signal. Some flaws produce known heights.
multiple tip signals that must be resolved. The ability of the 6.3 Dual-Element Bi-Modal Method—The Bi-Modal Sizing
operator to distinguish between tip and corner signals can be Method is based on the use of a dual-element search unit. This
compromised if several cracks are clustered in the same area. dual-element search unit is designed to insonify the entire wall
In areas of clustered cracks, corner reflections will dominate thickness by transmitting and receiving high-angle refracted
and mask tip signals. In cases of clustered cracks, the depth of longitudinal waves as well as low-angle shear waves. For this
the peaked signal may be the only reliable means to distinguish reason, the Bi-Modal sizing methods that feature the dual-
the tip signals from the corner signals. The tip-diffraction element search unit are applicable to far-surface connected
methods can be valid for a wide range of flaw heights. The planar flaws from 10 to 90 % through-wall. The TOF technique
prerequisites are that the tip of the flaw and the tip signal be requires that the first signal, the longitudinal wave, be maxi-
distinguishable from other signals. For very shallow flaws, the mized or peaked and the peaked first signal is measured along
tip signal may be masked by the flaw corner-reflector signal the instrument time base which is standardized in through-wall
due to poor resolution. A search unit with a shorter pulse depth. The · TOF technique is particularly useful because the
duration will improve this limitation. Broadband search units flaw height-related separation between the direct longitudinal
have been noted for their short pulse durations; however, due to wave and mode-converted signal can be measured before the
search unit is restricted by the weld crown. For the · TOF
dispersion in austenitic stainless steel weld metal, it may be
technique, both measurements are independent of signal am-
beneficial to select a narrow-band search unit with greater
plitudes. A 20 % far-surface notch and an 80 % far-surface
penetrating characteristics. This argument holds true for very
notch are sufficient to standardize the time base for components
deep flaws also. When the flaw is located in the weld region or
in the thickness range of 10 to 40 mm (0.4 to 1.6 in.). Flaw
very near the weld region, longitudinal waves may be consid- height may then be read directly on the screen in percent of
ered for the tip-diffraction method. Longitudinal waves may wall thickness. The extent of the flaw is indicated by the signals
help locate weak tip-diffracted signals in highly attenuative that are observed in the left half of the instrument screen. The
stainless steel but reflection from the component far surface further the direct longitudinal wave is peaked, or the greater the
should be avoided due to mode conversion. A very important separation of the signals from the mode-converted signal, or
factor in the sizing of planar flaws using the tip-diffraction peaks from mid-screen, the deeper the flaw. Signals originating
method is signal pattern recognition. To size with this method, from the interaction of shear waves with the base of the flaw,
the user must be able to identify two signals: (1) a signal that with or without mode conversion, are confined to the right half
is diffracted from the flaw tip and (2) a second signal that is of the instrument screen and merely indicate that the flaw is
reflected from the base of the flaw. The task of identifying the far-surface connected.
two signals is complicated by the high-amplitude noise signals 6.3.1 Wave Propagation Through the Material—It is ac-
and geometric signals from the component surface. Some knowledged that shear waves cannot interact effectively with
ultrasonic instruments allow the user the option of using the the upper extremities of tight and branched, medium to large
un-rectified or rectified display (RF display) signals. In many flaws that are located near the sound-scattering fusion lines of
cases, an RF display facilitates in distinguishing the tip signal austenitic welds. These may not produce readily recognizable
from noise signals by identifying the phase of the signals. The tip-diffracted signals for flaw sizing purposes. The Bi-Modal
signal from the tip of the flaw must always peak when the search unit is designed specifically for austenitic weld exami-
search unit is moved forward from the point where the corner nation, however, this is also applicable to carbon steel materi-
signal is maximized (for first half-V path) or backed up from als. As shown in Fig. 10, the Bi-Modal search unit transmits
the point where the corner signal is maximized (for second one longitudinal wave, and two shear waves and receives two
half-V path). This distance traveled is directly related to flaw longitudinal waves (one from the tip of the flaw and one from
height. The examiner must become accustomed to the search the base of the flaw), one mode-converted signal from the flaw

7
 E 2192

Subscript ’a’ denotes travel to the receiver and subscript ’b’ denotes travel from the transmitter. Enlarged detail note refers to Fig. 11.
FIG. 10 Bi-Modal Search Unit Longitudinal Wave and Shear Wave Signals

face, and one far-surface Creeping wave signal from the base in the amplitude of one signal is in unison with the amplitude
of the flaw when the search unit is operated in its normal dual rise of the other). When the center of the incident longitudinal
element mode. Depending upon search unit design, either wave beam is directed toward the flaw tip, the center of the
element can be used as the transmitter or the receiver. The incident shear wave beam is directed toward the flaw base, and
directivity patterns of the Bi-Modal search units are quite broad the amplitude of the longitudinal wave signal, as well as that of
due to the relatively small active element size and low the far-surface creeping wave signal, is maximized. Upon
operating frequency in the region of 3 MHz. Therefore, the moving the search unit closer to the flaw, the longitudinal wave
high-angle longitudinal waves and the low-angle shear waves signal will again recede into the background of irrelevant
insonify the entire component wall thickness. Four associated indications. To determine the arrival time of this signal, the
signals that move together on the instrument screen can be user typically moves the search unit toward the flaw until the
expected when the search unit is scanned over a far-surface amplitude drops.
connected flaw with broad back-and-forth movements, (Fig. 6.3.2 Principles of Bi-Modal TOF Technique—Weld crown
11). This follows from the premise that while the longitudinal
permitting, the search unit may be moved toward the weld far
waves interact effectively with both extremities of the flaw (the
enough to peak the longitudinal wave signal. Fig. 13 shows that
tip and the base), the shear waves interact only with the flaw
the relationship between the signal arrival time in screen
base. The first signal originates from the upper tip of the flaw.
divisions and the flaw height in percent of wall thickness is
If each element were a transmitter, the longitudinal wave
very nearly linear and independent of wall thickness.
energy from the two elements would converge to this area. The
usually weak tip-diffracted signal is enhanced while the back- 6.3.3 Principles of the · TOF Technique—The longitudinal
ground of irrelevant indications is suppressed by restricting the wave signal may be considered as a satellite of the mode-
longitudinal wave beams to the upper flaw tip area. The next converted signal since their separation, measured in screen
signal can sometimes be observed from a flaw and is usually divisions, is practically independent of the axial coordinate of
observed from a far-surface notch as a result of the longitudinal the search unit relative to that of the flaw. Figs. 14 and 15
wave from the transmitter reflecting at the flaw base and being illustrate the nearly linear relationship between normalized
received as a longitudinal wave by the receiver. The third flaw height and this signal separation. The most useful feature
signal is usually the strongest because it results from the of the Bi-Modal sizing method is that the flaw height can be
mode-converted shear wave from the face of the flaw. The measured anywhere along the length of the flaw as long as both
reflection of the incident shear wave at the flaw opening results the longitudinal wave and the mode-converted signals are seen
in the fourth signal which is analogous to the far-surface moving in unison on the screen, allowing height measurements
creeping wave signal. The echo-dynamic curve is broadest for to be made even when a wide weld crown is present. A second
the longitudinal wave signal and narrowest for the creeping · TOF measurement may sometimes be used to confirm the
wave signal as shown in Fig. 12. The mode-converted signal flaw height. This second measurement is obtained by noting
peaks shortly after the flaw is insonified. It follows from the difference in arrival time of the longitudinal wave signal
geometrical considerations that the echo-dynamic curves for and the longitudinal wave signal reflected from the flaw base.
the longitudinal wave signal and the far-surface creeping wave These two signals also move in unison and form a linear
signal are nearly synchronous for a large flaw (that is, the rise relationship when the flaw is oriented vertically.

8
 E 2192

Subscripts as in Fig. 10.

FIG. 11 Interaction of the Incident L-Wave and S-Wave from a Bi-Modal Search Unit with a far-surface Connected Flaw Resulting in
Four Associated Signals

6.4 Focused Longitudinal Wave or Dual-Elements Focused is placed on the weld reinforcement and directed at the flaw. In
Shear Wave Methods—The dual-element focused longitudinal this case instead of a diffracted wave returning to the search
or dual-element focused shear wave flaw sizing techniques are unit, the upper extreme of the flaw face reflects ultrasonic
essentially the TOF or sound-path measurement techniques energy back to the search unit. The result is a high-amplitude
with the use of focused longitudinal or shear wave search units, signal with a long or broad echo-dynamic pattern. Due to an
generally greater than 50 % from the far surface in depth. obliquely oriented flaw, a flaw height measurement obtained in
These techniques are particularly suitable for sizing flaws this manner tends to undersize the flaw when the location of the
which are mid-wall to very deep. The use of high beam angles peak reflection is used as the tip location. The user must
results in this technique being the most accurate for very deep compensate by moving the search unit toward the flaw until the
flaws. As with the tip-diffraction method, the signal from the signal drops by about 3 dB or by finding the slight rise in
flaw tip is maximized or peaked and its time of flight or sound amplitude along the leading edge of the echo-dynamic pattern,
path is recorded without regard to the arrival time of other which is due to the diffracted wave from the flaw tip.
signals. The focused longitudinal wave and focused shear wave
sizing techniques are used to measure the remaining ligament NOTE 3—A limitation of this method with a focused longitudinal wave
of good material between the flaw and the scanning surface. search unit is that associated shear waves (if not properly identified) may
cause confusion and could result in mode-converted signals that may
Actual flaw height is obtained by subtracting the remaining
produce erroneous measurements.
ligament from the local wall thickness. Occasionally, the signal
associated with the upper extreme of a flaw is due to beam Search unit frequency, refracted angle, element size, and
reflection rather than diffraction. This is most prevalent when a focal depth are factors for determining the effective range of
flaw follows the weld fusion line toward the outside surface of the technique. Common search unit frequencies are 2 MHz and
the weld and is oriented away from the weld and the search unit 4 MHz, with the lower frequency preferred for coarse grain

9
 E 2192

FIG. 12 Asynchronous Echo-Dynamic Curves for a 50 % Deep far-surface Notch

FIG. 13 Correlation of Normalized Flaw Height With Time Delay, t, Obtained by the Bi-Modal Time of Flight Technique

materials, for example, austenitic. The effectiveness of sizing the penetration depth by reducing the frequency to increase
with high-angle longitudinal waves is strongly dependent on beam spread or by reducing the incident angle for a lower
the selection of a search unit that produces a beam shape central refracted angle, that is, a 60-degree or a 45-degree
appropriate to the application. When sizing a flaw in thin-wall longitudinal wave or shear wave search unit. It is essential to
material, select a beam angle that does not penetrate very
deeply into the component. For thicker-wall material, increase

10
 E 2192

FIG. 14 Relationships Among Normalized Flaw Height, h/t, Doublet Separation, s and Time Delay, t

FIG. 15 Correlation of Normalized Flaw Height With Doublet Separation, s, Obtained by the Bi-Modal · TOF Technique

11
 E 2192
measure the focal depth of the search unit using a reference 7.1.2 Instrument—A pulse-echo ultrasonic instrument ca-
block that contains a series of known reflectors at different pable of generating and receiving frequencies in the range of at
depths. least 1 to 5 MHz should be suitable for sizing with the
far-surface Creeping Wave or 30-70-70 mode conversion
7. Ultrasonic Flaw Sizing Standardization Requirements method. The instrument should exhibit adequate resolution and
7.1 Far-Surface Creeping Wave Method—The far-surface high filtering capabilities.
Creeping Wave Sizing Method depends upon pattern recogni- 7.1.3 Reference Block—Standardization for the far-surface
tion of the three potential signals that may be observed, for Creeping Wave technique requires special far-surface notch
example, 70-degree L wave, the mode converted signal or the reference blocks. The block must have a set of notches located
far-surface creeping wave signal. By observing the absence or at various depths from the far surface. The simplest design is a
presence of these three signals, the echo dynamics of the flat plate or pipe section with far-surface notches located at
signal, and the time of flight of the 70-degree L wave signal, a increments of 10 % or 20 % depths, for example, 10 %, 20 %,
user can classify a far-surface connected crack into the inner 40 %, 60 %, 80 %. The user should become familiar with the
1⁄3, middle 1⁄3, or the outer 1⁄3 zone of the material thickness to
absence or presence of the 70-degree L wave signal, the time of
be inspected. flight of the 70-degree L wave signal, the amplitudes and
7.1.1 Search Unit—The pattern of the three signals strongly echo-dynamic patterns of the mode-converted signal and the
depends on several search unit parameters. Before attempting far-surface creeping wave signal as different depth notches are
to apply this method with a new search unit, the sound wave encountered. If these notches are used to familiarize the user
patterns should be evaluated using known notch reflectors at with various signals that may be encountered, the block should
various depths, 20 %, 40 %, 60 %, and 80 %. There may be equal the thickness of the component to be examined.
significant variations between search units with identical face- 7.1.4 System Standardization—The far-surface Creeping
plate parameter values, even if they are from the same Wave or 30-70-70 mode conversion technique does not depend
manufacturer. Generally, a single-element search unit is suit- on the arrival time of the flaw tip signal, so the system does not
able for most applications. have to be standardized accurately for distance. The same
7.1.1.1 Beam Angle—The primary intent when sizing with search unit used for the refracted longitudinal wave sizing
the far-surface Creeping Wave method is to produce different method may be used for the far-surface Creeping Wave
beams at the far surface and utilize the responses from these technique. Use a notched block for standardization following
beams to categorize flaw height. This goal is met by using these steps:
70-degree longitudinal waves or refracted longitudinal waves 7.1.4.1 Adjust the delay to display the initial pulse at the left
at 55 degrees and higher. side of the screen.
7.1.1.2 Frequency—To limit beam spread and its degrada-
7.1.4.2 Place the search unit near the end of the reference
tion of sizing accuracy, higher frequencies than those com-
block. Observe the 70-degree mode converted L wave signal
monly used during examination are suggested. Narrow-band
and creeping wave signals.
search units will avoid the beam spread caused by the
low-frequency components of the spectrum. A frequency of 2 7.1.4.3 Peak the creeping wave signal, and adjust delay and
MHz seems ideal for austenitic material, while 4 MHz seems range controls to position the 70-degree mode converted L
more effective for carbon steel materials. wave and creeping wave signals at 4 and 5 screen divisions
7.1.1.3 Elements—Specially designed search units with respectively.
single- or dual-element search units may be used. It is very 7.1.4.4 Adjust the creeping wave signal amplitude to 80 to
difficult to eliminate internal wedge reflections and entry 100 % full screen height. Then increase the instrument ampli-
surface noise with a 70-degree, single-element search unit. tude by 8 dB. This reference level is now the primary scanning
These single-element problems can only be avoided with long and evaluation level.
wedge paths. This can lead to increased beam size and make 7.1.4.5 Place the search unit to peak the creeping wave
search unit movement difficult. Element size is significant. signal from the 20 % notch. Record echo dynamic movement
Very small elements will have excessive angular beam spread, of the 70-degree mode converted L wave signal as the search
and very large elements may produce beams with too much unit is scanned toward and away from the far-surface con-
penetration. The optimal size will probably be a function of nected notch.
wall thickness, with larger elements being acceptable for 7.1.4.6 Place the search unit to peak the creeping wave
greater thickness. Generally, a 10 mm or a 0.375 in. diameter signal from the 40 % notch. Record the echo dynamic move-
or square search unit will work for most applications. ment of the 70-degree mode converted L wave signal as the
7.1.1.4 Contact Area—The search unit contact area or search unit is scanned toward and away from the far-surface
“footprint” should be as small as practical. With dual search connected notch. If present, record the amplitude of the
units, small search unit width is necessary for proper coupling 70-degree L wave signal.
of both elements to the scanning surface. Short length is 7.1.4.7 Place the search unit to peak the creeping wave
desirable for adequate coupling in pipe weld conditions of signal from the 60 % notch. Then record echo dynamic
diametrical weld shrinkage, especially if the shrinkage is made movement of the 70-degree mode converted L wave signal as
more severe by grinding of the weld crown. If the search unit the search unit is scanned toward and away from the far-
is too long, there may be entry angle variations with strong surface connected notch. If present, record the amplitude of the
effects on the refracted angle. 70-degree L wave signal. Increase the gain to bring the

12
 E 2192
70-degree L wave signal up to at least 40 % full screen height unit characteristics must be documented adequately prior to use
(FSH). Peak the longitudinal wave signal and record the if examination repeatability is necessary.
horizontal screen division position, for example, 2.5 divisions. 7.2.1.1 Beam Angle—Refracted beam angles in the range of
7.1.4.8 Place the search unit to peak the creeping wave 40 to 60° can generally be used with the tip-diffraction method.
signal from the 80 % notch. Then record echo dynamic Due to the long sound paths involved in the second half-V path
movement of the 70-degree mode converted L wave signal as technique, it is preferable to use refracted angles in the range of
the search unit is scanned toward and away from the 80 % 40 to 52° from the perpendicular. Higher beam angles require
notch. Record the amplitude of the 70-degree L wave signal. longer metal path distances and the energy may be attenuated
Increase the gain to bring the 70-degree L wave signal up to at to an almost nonexistent level.
least 40 % FSH. Peak the L wave signal and record the 7.2.1.2 Frequency—Search units may have either single or
horizontal screen division position, for example, 1.8 divisions. dual elements in the nominal frequency range of 2 to 5 MHz.
7.1.4.9 A typical sequence of recordings is shown below. Favorable results have been obtained with nominal frequencies
These values will be relative to the search unit design and of 4 and 5 MHz when the sizing is limited to the first half-V
frequency, and material type and thickness. path and height of the flaw is minimal. To maintain penetration
Creeping Wave 70-degree Mode Converted of shear waves through large-grain base metal structures, low
Notch Depth 70° L Wave
Signal L Wave Signal frequencies such as 2.25 MHz are desirable. However, this low
20 % none 2.5 divisions yes
40 % none 3.5 divisions yes
frequency does result in a sizeable amount of beam spread,
60 % 2.5 divisions 4.5 divisions yes which reduces the accuracy of the sizing method. Occasionally,
80 % 1.8 divisions 5.5 divisions yes tip-diffracted signals are detected with lower frequencies (for
7.2 Tip-Diffraction Method—The tip-diffraction method de- example, 1.5 MHz). The tip signal should be noted if identified
pends primarily on the arrival time of the flaw tip signal and, during the flaw detection process. The flaw depth should then
in some cases, on the arrival time of the flaw base or corner be measured using higher frequencies.
signal. Consequently, it is important to standardize the screen 7.2.1.3 Propagation Mode—Either shear or longitudinal
accurately for sound path, however, it is always acceptable to waves may be used for the tip-diffraction sizing techniques
standardize the screen to read directly in depth as a function of when using the first half-V path technique. Due to the high
a percent of component thickness. This approach is generally attenuation of stainless steel, longitudinal waves may present
more convenient. higher amplitude tip-diffracted signals, but decreased resolu-
7.2.1 Search Units—When selecting a search unit for sizing tion. Longitudinal waves are impractical with the second
planar flaws using the tip-diffraction method, the following half-V path technique since the search unit also produces a
guidelines should be considered: (1) A very high signal-to- shear wave which may cause confusion. Also, the longitudinal
noise ratio is desired. This characteristic is governed by the wave tends to mode convert to shear wave as the ultrasonic
frequency, diameter, and wave mode; (2) High-resolution beam reflects at the opposite surface. The large amount of
search units (higher frequency, shorter pulse length) will aid in mode conversion reduces the signal-to-noise ratio even more.
sizing very shallow flaws because the tip and base signals are Additionally, the presence of many spurious signals from the
close together and nearly coincident in time; (3) High- various mode converted signals may present the user with more
resolution search units work well with the time of flight (TOF) opportunities to select the wrong signal and misidentify it as
technique; (4) Large beam spread may be beneficial when the flaw tip.
sizing suspected midrange flaws with the delta time of flight 7.2.2 Instrument—A pulse-echo instrument capable of gen-
technique because this technique requires viewing of the tip erating and receiving frequencies in the range of at least 1 to 5
and base signals simultaneously; (5) The characteristics of the MHz is suitable for sizing with the tip-diffraction method.
search unit selected should be thoroughly investigated with Avoid screen displays showing signals that have minimal
reference blocks before attempting any sizing techniques; (6) filtering. Horizontal linearity should be within 2 % of full
Longitudinal waves can enhance the tip signal but may produce screen width. The ability to view the full, unrectified RF
spurious mode-converted indications; (7) The distance from waveform is helpful during sizing. It is recommended that
the front of the search unit to the beam index point should be ultrasonic instruments capable of displaying the RF waveform
minimal in order to maximize the diffracted signal from the be used.
flaw tip when a wide weld crown is present; and (8) Beware of 7.2.3 Reference Blocks—The reference block (or blocks)
reflections that may occur from within the search unit wedge. should contain special reference reflectors (for example,
These can occur in the area of interest on the display and can notches having various depths) that pertain to the standardiza-
increase the difficulty of identifying the tip signal. Various tion for specific sizing technique(s). To standardize the time
search unit designs with different element sizes and arrange- base, use 1-in. thick, flat plate of material similar to the
ments can be used. It must be pointed out that the individual component to be examined. This plate should have notches
search unit design parameters greatly influence their effective- from 10 to 90 % deep in steps of 10 %. The most desirable
ness. For the Tip-diffraction method using delta time of flight reference block is the same thickness as the component
technique, a search unit that is highly damped to a maximum containing the flaw to be sized, but standardization can be
pulse length of one and one half to two cycles at the -6 dB accomplished with any known thickness.
points is desirable. This will improve resolution for sizing very 7.2.4 System Standardization—The tip-diffraction method
shallow flaws, that is, less than 10 % wall thickness. Search depends primarily on arrival time differences between the

13
 E 2192
pulses from the base of the flaw and the flaw tip. It is, therefore, Screen Divisions Depth from Far-Surface
important that the screen be standardized accurately for sound 0 100 %
1 80 %
path. Sound path standardization may be accomplished in 2 60 %
exactly the same fashion as the standardization for flaw 3 40 %
detection if it is limited to second half-V path. The horizontal 4 20 %
5 0%
sweep is standardized in inches or mm of sound path along the
beam path. Any reference block used for sweep distance 7.2.4.2 Standardization for Delta Time of Flight Technique
standardization may be used, for example, IIW, DSC, etc. (Direct Flaw Depth Measurement)—Standardize the ultrasonic
There is an alternative sweep standardization which is simple instrument as in 7.2.4.1, record the screen divisions of separa-
and results in higher accuracy. In this case, the horizontal tion for the base and tip-diffracted signals for each of the
sweep is standardized in depth. A reference block consisting of notches. They will be approximately as follows:
a series of far-surface connected notches is required (see Fig. Notch Depth Divisions of Separation
16). 20 % 0.5 Divisions
40 % 1.0 Divisions
7.2.4.1 Standardization for the Time of Flight (TOF) Tech- 60 % 1.5 Divisions
nique To standardize for direct flaw depth measurements when 80 % N/A
using the time of flight (TOF) technique: With the deeper notches, that is, 60 % and 80 %, the tip and
NOTE 4—The first 5 steps are a simple technique to establish a coarse the base signal may not be seen on the instrument screen due
standardization. to limited beam spread of the search unit.
7.3 Bi-Modal Technique:
(1) Select a suitable reference block with at least two
7.3.1 Search Unit—The Bi-Modal search unit consists of
notches of known depths, bracketing the flaw depth range of
two moderately damped search units, one mounted behind the
interest, for example, 20 % and 80 % deep notches.
other (tandem design) on a plastic wedge with two refracting
(2) As a ranging technique, place the search unit to obtain wedge angles. Both elements are bi-modal, that is, each is
the far-surface corner of the end of the reference block. capable of transmitting and receiving both a longitudinal wave
(3) Peak this half-V path signal and using the delay control and a shear wave signal. The roles of the two elements can be
adjust this signal to 5 horizontal screen divisions. interchanged.
(4) Move the search unit from the end of the block to peak 7.3.1.1 Beam Angles—In the transmit-receive mode of op-
the near-surface signal at the edge of the reference block and, eration, one element transmits a high-angle longitudinal wave
using the range control, adjust this full-V path signal to 10 beam as well as a low-angle shear wave. The other element is
screen divisions. directed to receive all sound waves emanating from the upper
(5) Alternate between the delay and range controls to set and lower extremities of a far-surface connected flaw.
the far-surface corner to 5 divisions and the near-surface corner 7.3.1.2 Mode of Propagation—The search unit transmits
to 10 divisions. and receives both longitudinal and shear wave signals.
(6) Locate the base signal or corner signal of the 80 % 7.3.2 Instrument—A pulse-echo, ultrasonic instrument ca-
far-surface notch. Move the search unit forward to peak the pable of generating and receiving frequencies in the range of at
tip-diffracted signal from the tip or edge of the notch. Using the least 1 to 5 MHz with linear and stable time base should be
delay control adjust this signal to 1 horizontal screen division. suitable for sizing with the multi-pulse observation sizing
(7) Locate the 20 % far-surface notch to obtain the base or method. An RF display mode is not required, but is desirable.
corner signal. Move the search unit forward to peak the 7.3.3 Reference Block—Standardization for this sizing
tip-diffracted signal from the tip or edge of the notch. Using the method also requires special reference blocks. As with the
fine range control adjust the peaked signal to 4 screen other sizing methods referenced in this document, the reference
divisions. block must have a set of planar notches located at various
(8) Alternate between the delay and range controls to set depths from the far surface. Using additional blocks, the user
the 80 % tip signal to 1 screen division and the 20 % tip signal should become familiar with the echo-dynamic patterns of the
to 4 divisions. various signals expected from components of known thickness.

FIG. 16 Sizing Reference Block

14
 E 2192
7.3.4 System Standardization—A reference block contain- possible to standardize the screen to read directly in depth. The
ing far-surface notches as in Fig. 16 or Fig. 17 is necessary to depth approach is preferred.
standardize the time base of the ultrasonic instrument. Each 7.4.1 Search Units—The focal depth of the longitudinal
notch in a 25 mm (1-in.) thick reference block increases by 5 wave and shear wave search units depends on several search
mm (0.2 in.) or 20 % depth increments. unit parameters. Before attempting to apply this sizing method
7.3.4.1 Standardization for the Time of Flight technique with a new search unit, it is essential to determine the focal
should be conducted as follows: depth of the search unit with known reflectors at known depths.
(1) Position the Bi-Modal search unit at the edge of the Results between search units having identical faceplate param-
reference block to display the three primary signals, that is, the eter values may vary significantly even if they are from the
refracted L wave signal, the mode converted shear wave signal, same manufacturer. Search unit selection should be based on
and the far-surface Creeping wave signal. the component thickness and depth of penetration. The objec-
(2) As a ranging technique, using the range control, tive is to insonify a layer of the component thickness under the
separate the second and third signals by approximately 2.0 near surface without penetrating to the opposite surface.
screen divisions. 7.4.1.1 Beam Angle—The primary intent when sizing with
(3) Position the Bi-Modal search unit to peak the refracted the focused longitudinal or focused shear wave techniques is to
L wave signal from the 80 % far-surface connected notch. limit the penetration of the beam into the component in order
Adjust the delay control to position this to 1 horizontal screen to avoid confusion from unidentifiable ultrasonic signals.
division. Focused search units should only be used with the first half-V
(4) Position the Bi-Modal search unit to peak the refracted path technique. Dependent upon the examination volume
L wave signal from the 20 % far-surface connected notch. (outer 1⁄3 thickness), 45, 60, 70 degree L-waves or shear waves,
Adjust the range control to position this to 4th horizontal and near-surface Creeping Waves may be used.
screen division.
7.4.1.2 Frequency—To limit beam spread and its negative
(5) Alternating between delay and range controls ensure
effect on sizing accuracy, it is desirable to use higher frequen-
the signal from the 80 % notch is at screen division 1 and the
cies than those commonly used during flaw detection. Narrow-
40 % notch signal is at the screen division 4.
band search units should be used to limit the beam spread
(6) The instrument is now standardized such that 5 screen
caused by the low-frequency components of the spectrum.
divisions equals 0 % and 0 screen divisions equals 100 %
Ideally, the transmitting element should have a tuning circuit
through-wall depth.
that will optimize the effects of possible pulser impedance
7.3.4.2 Standardization for the Delta Time of Flight tech-
variations on the frequency spectrum. If the frequency is too
nique should be conducted as follows. With the Time of Flight
high, then the penetration depth will be inadequate for locating
standardization set, record the separation in horizontal screen
any but the deepest flaws. A search unit of 4 or 5 MHz should
divisions between the refracted L wave and mode converted
permit quantitative sizing of flaws that reach to the outer 1⁄3
signals. Approximate values are shown below:
thickness of the near surface. Lower frequencies of 2 and 3
20 % Notch 2.1 Screen Divisions
40 % Notch 2.5 Screen Divisions
MHz are acceptable dependent upon the depth of penetration
60 % Notch 2.9 Screen Divisions desired.
80 % Notch 3.3 Screen Divisions
7.4.1.3 Elements—Either single or dual-element search
7.4 Dual-Element Focused Longitudinal or Focused Shear units may be used. Note that it is very difficult to eliminate
Wave Methods—As with the tip-diffraction TOF technique, internal wedge reflections and entry surface noise when using
high-angle focused longitudinal wave or focused shear wave a single-element search unit for high-angle longitudinal wave
methods also depend primarily on the sound path travel of the examination. These single-element problems can only be
flaw tip signal. It is important that the screen be standardized avoided with the use of larger wedges, however, this can lead
accurately for sound path distance; however, it is always to increased beam size and difficult search unit manipulation.

FIG. 17 A Bi-Modal Search Unit Positioned on the Reference Block for Maximum L-wave Signal Amplitude from the Fourth Step

15
 E 2192
Element size is an important consideration. Very small ele- to observe the response from either a near-surface notch or the
ments will have excessive angular beam spread, and very large end of the block. These signals appear at the same horizontal
elements may produce beams with too much penetration. The position as a 100 % through-wall flaw. When a planar flaw
optimal size will probably be a function of depth of penetration signal is obtained and maximized, its distance below the near
and wall thickness, with larger elements being acceptable for surface will be indicated by the time-of-flight (sweep position),
greater thicknesses. or depth from the near surface. This dimension is subtracted
7.4.1.4 Contact Area—The search unit contact area or “foot from the local thickness of the component to determine the
print” should be small. With dual-element search units, small flaw height. If notches are used, the block should be of the
search unit width is necessary for proper coupling of both same thickness as the component. If the reference block is used
elements to the surface. Short length is desirable for adequate only for establishing the screen distance standardization, its
coupling in conditions of diametrical weld shrinkage, espe- thickness is not important provided that it is a known thickness
cially if the shrinkage is made more severe by grinding of the and it is thick enough to fully characterize the ultrasonic beam.
weld crown. 7.4.4 System Standardization—The focused longitudinal
7.4.1.5 Focal Depth or Focal Sound Path—Dual-element wave or focused shear wave techniques depend primarily on
search units focus at a point directly beneath or only slightly the time of flight or sound path. The instrument must be
ahead of the housing to maximize the signal. This is the point standardized accurately for distance or depth. Standardization
at which the sound beams would cross or Beam Crossover may be performed using a notched block.
Point. This may result in insufficient access to deep flaws if the 7.4.4.1 Adjust the delay to display the initial pulse at the left
weld crown is present. The exit point-to-front of search unit side of the screen.
distance should be as small as practical and the crossover point 7.4.4.2 Obtain a signal from the end of the reference block
distance should be as long as practical. The focal sound path is with the index point placed very near the block end. This will
generally noted for each type of focused dual-element search not be a corner reflection from the bottom of the block; it will
unit. The more important measurement is the focal depth of the be a reflection from the top part of the end of the block. Verify
search unit. This can be calculated by the following formula: that the correct signal is obtained by finger-damping near the
FD 5 FS ~Cos of the refracted angle of the search unit! top of the end face.
7.4.4.3 Adjust the delay and range controls to place the
where: end-of-block signal at 0 horizontal divisions.
FD = focal depth, and 7.4.4.4 Place the search unit index point directly above the
FS = focal sound path.
reference notch or hole located at 0.10 in. (or metric equiva-
7.4.2 Instrument—A pulse-echo ultrasonic instrument ca-
lent) from the near surface. Move the search unit backward,
pable of generating and receiving frequencies in the range of at
slowly. With focused L waves, one or more signals from the
least 1 to 5 MHz should be suitable for sizing using the
associated shear wave component may appear. Continue to
high-angle longitudinal wave method. Horizontal linearity
move back until a separate signal appears. This will be the
should be within 2 % of full screen width. The system should
focused longitudinal wave or focused shear wave signal. It can
exhibit adequate resolution and high filtering capabilities.
be recognized by the fact that it usually travels more along the
7.4.3 Reference Block—Standardization for the focused lon-
screen baseline than the mode converted signals.
gitudinal wave or focused shear wave sizing techniques re-
7.4.4.5 When the focused L wave or focused shear wave
quires special reference blocks. The block must have a set of
signal is maximized, visually verify that the search unit index
planar or rounded reflectors located at various depths from the
point is in the proper position to detect the reflector with a high
examination surface. An example is a flat plate or pipe section
beam angle. Adjust the delay control to place this signal at 1
with far-surface notches located at increments of 0.10 in. in
division.
depth below the near surface, that is, 0.10 in., 0.20 in., 0.30 in.,
0.40 in., etc. or if metric, 2 mm, 4 mm, 6 mm, etc. Alterna- 7.4.4.6 Obtain and maximize the high-angle longitudinal
tively, a reference block with side-drilled holes located at wave or shear wave signal from the notch or hole located at the
increments of 0.10 in. (or metric equivalent) in depth below the desired depth below the near surface, for example, 0.500 in. or
near surface can be used. The signals noted from the holes are typically 12.5 mm if using metric reference blocks. Adjust the
due to reflection rather than diffraction, and the central part of range control to place this signal at 5 screen divisions.
the ultrasonic beam is apparently bent as it propagates into the 7.4.4.7 Continue with the other notch or hole signals until
component, resulting in the beam interacting with the holes at they are too far from the near surface to be detected with a
various depths. The beam may reflect off of the side of the hole signal-to-noise ratio of at least 3. Increase instrument gain as
at the 0.1-in. depth and near the top of the hole at the 0.4-in. necessary.
depth. This is dependent upon the refracted angle of the search 7.4.5 The screen setup should now be as follows:
unit and the depth of the hole. Starting with the notch or hole Reflector Signal Horizontal Divisions
End of block 0
located 0.10 in. (or metric equivalent) in depth below the near 0.1 in. deep 1
surface, set the ultrasonic response from the notch tip at 1 0.2 in. deep 2
horizontal screen division. Standardize the instrument screen 0.3 in. deep 3
0.4 in. deep 4
so that each horizontal division represents 0.10 in. (or metric 0.5 in. deep 5 (may not be detectable:
equivalent) of depth into the component. The beam entry point dependent on search unit)
is placed at the extreme left side of the screen. It is also useful (For metric standardization, select accordingly). See Fig. 18.

16
 E 2192
The standardized region of the screen is from 1 division to signals observed on an actual far-surface connected flaw. If
the position of 0.500 in. below the surface. As a general rule, only the far-surface creeping wave signal is observed, the flaw
do not attempt to size cracks outside this standardized range. is approximated to less than 10 to 15 % depth. If the far-surface
Instrument gain will likely need to be increased to obtain the creeping wave signal and the mode converted signal, are
last one or two signals. Also, the near-surface notch signal observed, the flaw is approximated to greater than 15 to 20 %
would be well over 100 % full screen height. Note that the area depth. If the far-surface creeping wave signal, the mode
of the screen where the mode converted shear wave signals converted signal, and a 70-degree L wave are observed, the
appear from the far surface is to the right half of the screen flaw is approximated to greater than 40 to 50 % depth.
outside the standardization region. NOTE 5—These estimates are a typical evaluation from a far-surface-
notch reference block. These estimates will vary with type of search unit,
8. Sizing Applications frequency, size, material thickness and type. Another key point is to
8.1 Far-Surface Creeping Wave or Mode Conversion observe the echo dynamics of the mode converted signal. A broad echo
Technique—Once the flaw’s existence has been verified with dynamic movement of the mode converted signal would confirm the
the detection and characterization techniques, the next decision presence of a 70-degree L wave signal to verify a deep flaw greater than
40 to 50 % through wall.
is to estimate the vertical extent of the flaw as measured from
the far surface. The far-surface Creeping Wave or Mode 8.2 Tip-Diffraction Technique—Tip-diffraction sizing may
Conversion Method will place the flaw into 1⁄3-depth zones. be performed using the first or second half-V path techniques,
The far-surface Creeping Wave method will provide a qualita- and the user may choose either the Time of Flight (TOF) or the
tive estimate of the flaw depth. The technique also confirms Delta Time of Flight (· TOF) techniques. The results should be
that a flaw is surface connected by the presence of the Creeping the same if the flaw is vertical and the actual material thickness
Wave signal. The component thickness is divided into 1⁄3 is known. If more than one of these combinations of methods
thickness zones; inner 1⁄3, middle 1⁄3 and the outer 1⁄3 volume or is used and different results are obtained, the user must
zone. Using the far-surface creeping wave, the user can determine whether to accept the results of one method or
approximate the flaw depth into one of the 1⁄3 zones. Once the disregard all results and attempt sizing with one of the other
flaw is qualitatively sized, for example, 20 to 30 % deep, then two methods. If the flaw is not vertical, the · TOF technique
the other sizing methods and techniques are used to provide a will provide inaccurate results; therefore if the flaw is sus-
finite flaw depth estimate, for example, 23 % through-wall. pected to be oriented other than essentially vertical, the user
Generally, the following steps are used for evaluation with the should choose the TOF technique. Experience has shown that
far-surface creeping wave to approximate the flaw depth. occasionally the second half-V path, TOF technique may
However, the standardization performed on the reference exaggerate the flaw height as the ultrasonic beam may be
blocks addressed in Section 7 is used for comparison with redirected slightly at the reflection. This leads to the conclusion

FIG. 18 Focused Longitudinal Wave or Focused Shear Wave Standardization

17
 E 2192
that the first half-V path technique is more accurate and the as with austenitic stainless steel, or when the tip signal cannot
second half-V path data is used only for confirmation of flaw be resolved from the upper corner reflector signal. If both the
height in the middle to outer 1⁄3 volume of material. tip and the upper corner signal are present, which can occur
8.2.1 First Half-V Path Technique: with less attenuative material, then both a quantitative and a
8.2.1.1 Time of Flight Technique—After the specific area confirming qualitative signal are attained as illustrated in Fig.
containing the far-surface connected planar flaw is identified, 22.
position the search unit to obtain a signal from the flaw base at 8.2.2.2 Delta Time of Flight Technique—After the specific
half-V path. Move the search unit forward to obtain the area containing the planar flaw is identified, position the search
diffracted signal from the flaw tip. If the signal is observed, unit to obtain a signal from the flaw base at half-V path. Move
peak the diffracted wave signal from the flaw tip, then measure the search unit back to obtain a tip-diffracted signal from the
the sound-path distance to the signal. This distance reveals the top of the flaw. Manipulate the search unit back and forth and
amount of material above the flaw tip. The flaw height is then look for a doublet. If the doublet is observed, measure the time
determined by subtracting this dimension from the local difference between the two signals. The separation indicates
material thickness. Fig. 19 shows the display of a first half-V the height of the flaw, according to the standardization. Fig. 23
path TOF standardization and the ultrasonic beam paths related shows the instrument display of a second half-V Delta Time of
to that standardization. Flight standardization and a sketch of the ultrasonic beam paths
8.2.1.2 Delta Time of Flight Technique—After the specific related to that standardization.
area containing the planar flaw is identified, position the search 8.3 Bi-Modal Method—Flaw sizing on the basis of time of
unit to obtain a signal from the flaw base at half-V path. Move flight or the delta time of flight techniques should be performed
the search unit toward the flaw and back again and look for as follows:
evidence of a tip-diffracted signal moving in unison with the 8.3.1 Connect the search unit to the instrument operated in
flaw base signal (a doublet). If the doublet is observed, measure the transmit-receive mode.
the time difference between the two signals, (the tip signal and
8.3.2 Standardize the instrument for the appropriate wall
the flaw base signal). The separation indicates the height of the
thickness using the nomograms prepared from Fig. 13 or the
flaw, according to the standardization. Fig. 20 shows the
table prepared from the nomograms. Check the standardization
instrument display of a first half-V path Delta Time of Flight
by observing the initial pulse at t = 0 from the edge of a block
Technique standardization and a sketch of the ultrasonic beam
and a tip-diffracted signal from the reference block.
paths relative to the far-surface notches in the reference block.
8.2.2 Second Half-V Path Technique: 8.3.3 Using the notches in the most appropriate reference
8.2.2.1 Time of Flight (TOF) Technique—After the specific block, review the echo-dynamic behavior of the various signals
area containing the planar flaw is identified, position the search that are expected from all flaw sizes.
unit to obtain a signal from the flaw base at half-V path. Move 8.3.4 Scan a component volume void of flaws from both
the search unit away from the flaw to obtain the diffracted wave opposing directions and observe the irrelevant indications, both
signal from the tip of the flaw after reflecting from the far statically and dynamically, to the left of mid-screen. Adjust the
surface. If the signal is observed, peak the diffracted wave gain so that the average background noise is approximately
signal from the flaw tip, then measure the distance to the signal. 10 % of full screen height.
Subtracting the local wall thickness from this measurement 8.3.5 Scan over the flawed area with broad back-and-forth
reveals the flaw height. Fig. 21 shows the display of a second movements. Look for the mode-converted and creeping wave
half-V path Time of Flight (TOF) Technique standardization signals to the right of mid-screen. If the flaw is perpendicular
and the ultrasonic beam paths related to that standardization. to the surface, only two signals separated by about 2.2 screen
This technique may be effective for wide weld crowns which divisions and peaking between divisions 5 and 7 should be
would not allow for effective use of the first half-V path observed. The longer the signal duration, the deeper the flaw.
technique. If the tip signal is not observed but there is the 8.3.6 Scan the area from the opposite direction with broad
presence of a second half-V path corner reflector signal, this movements and look for signals only to the right of mid-screen.
would indicate the presence of a very deep flaw. This signal If the signal patterns are essentially the same when the flaw is
would only provide qualitative information that the flaw was viewed from the two directions, it is likely that the flaw has
very large. This may occur in highly attenuative material, such grown perpendicular to the surface and is oriented vertically.

FIG. 19 Example of First Half-V Path Time of Flight (TOF) Standardization

18
 E 2192

FIG. 20 Example of First Half-V Path Delta · TOF Standardization

FIG. 21 Example of Second Half-V Path Time of Flight Standardization

FIG. 22 Example of the Presence of Both the Tip Signal and the Upper Corner Signal

FIG. 23 Example of Second Half-V Path Delta Time of Flight Standardization

8.3.7 Look for a tip-diffracted longitudinal wave signal to 8.3.9 If possible, repeat these measurements from the other
the left of mid-screen, from both sides of the component. If the direction and profile the flaw from both directions. If a
amplitudes of the tip-diffracted signals are comparable and longitudinal wave signal from the flaw base is received, then
smaller than that of the appropriate reference signal, then it is relative arrival time measurements between the tip-diffracted
still likely that the flaw is oriented vertically. The separation longitudinal wave signal and the longitudinal wave signal from
measured between the tip-diffracted longitudinal wave signal the flaw base provide additional confirmation of flaw height.
and the mode-converted signal along the component from each 8.4 Focused Longitudinal Wave and Shear Wave Method—
side of the flaw yields the first two estimates of flaw height Longitudinal waves are not strongly affected by weld metal.
according to the appropriate curve in Fig. 15. The more Given a suitable surface, ideally a flush-ground and blended
extensive the flaw, the larger the distance between the signals. weld crown, this method can work as well through weld metal
8.3.8 Move the search unit toward the flaw far enough to as through base metal.
maximize the tip-diffracted longitudinal wave signal (weld
8.4.1 Signal Presentation
crown permitting). The time between the peaked tip-diffracted
signal and the mid-screen mark yields a third estimate of height NOTE 6—Instrument gain for investigating the flaw should be set such
in percent of local wall thickness. that the noise level from an unflawed area is low in amplitude but visible,

19
 E 2192
that is, 10 to 20 % full screen height. The examiner should only be h 5 t 2 0.1 in. ~s 2 2!
concerned with the noise level in the part of the screen that is standard-
ized. Noise at the extreme left and right sides of the screen will have no where:
effect. h = height,
Select the area of the component where the flaw is located. t = wall thickness at the flaw location, and
Scan the area and look for signals in the standardized area of s = sweep position of the maximized recordable signal,
the screen. Signals near 2 divisions should be expected to have expressed as the number of major sweep positions.
higher amplitude and signal-to-noise ratio than those occurring NOTE 7—The 2 in the parenthesis results from placing the beam center
deeper into the wall (a strong signal at 2 divisions indicating a entry point at 2 major sweep divisions. The constant in front of the
through-wall flaw). The screen may display the longitudinal parenthesis is shown in inches. This would be converted to the applicable
wave extremity signals and the shear wave flaw opening value in mm if metric standardization was used.
signals moving in unison. As the search unit is scanned toward Compare the height measurements obtained from both sides
the flaw, the shear wave signal will appear late in time at high of the flaw. They should be almost equal. For weld inspection,
amplitude, and the high-angle longitudinal wave signal will the measurement made from the weld side may be slightly
appear earlier in time (in the standardized region), probably lower if the flaw follows the weld fusion line because the
with less amplitude. If signals were obtained in either of the received signal will be a reflected wave from the top of the flaw
previous investigations, return to the unflawed area of the weld face instead of the extreme tip. If the heights are very different
and try to reproduce them. If similar signals are obtained in the (more than two wavelengths), then one of the signals may have
unflawed area, the recorded signals were probably not flaw- been a misinterpreted noise signal. If confirmation is achieved,
related; if the signals cannot be reproduced in the unflawed the conservative approach should be taken. Compare the height
area, they were probably flaw-related. Time permitting, repeat measurements obtained from different points along the length
the investigation at several points along the flaw. If the flaw is of the flaw. A flaw profile can be constructed from these
large enough to penetrate into the beam center, the signal-to- measurements. If the profile shows large height changes with a
noise ratio should be sufficient to resolve the indications. In the small position change (small compared to the flaw length),
shallower, less intense areas of the beam, it is possible to miss suspect that some of the measurements are incorrect.
the tip signal. The distinction between the high- and low-
intensity areas can be very clear, particularly for search units 9. Use of Complementary Methods for Flaw Sizing
with a very high refracted angle and relatively low beam 9.1 General Guidelines for Complementary Methods—The
spread. When evaluating a signal in a low-intensity area of the various sizing methods have certain limitations that prohibit
beam, the user should confirm the result using another sizing their application to all flaw sizes. By understanding each
technique. When a large beam is used (large due to high method in detail with its limitations, the methods must be
angular beam spread or large elements), the tip signals may combined to complement one another so that the user is more
exhibit extensive travel across the time-base. Such signals are confident of the flaw size measurement obtained from several
difficult to maximize accurately and therefore should be used methods than of a flaw size obtained from only one method.
non-quantitatively. Use a smaller beam or a different sizing The methodology described in this document assumes that all
method to obtain a numerical value for the flaw height. of the necessary search units are available to the user to
8.4.2 Calculation of Flaw Height—If no relevant signals combine the methods in the prescribed manner. The extent to
were obtained in the standardized region, then the high-angle which complementary flaw sizing methods may be useful in a
longitudinal wave method has not found any evidence that the field environment will depend on the physical conditions of the
flaw penetrates near the near surface. If relevant signals were component, the flaw morphology, the search unit selection and
present in the standardized region but excessive signal travel the time constraints imposed. Fig. 24 shows the optimum
precluded accurate definition of their sweep positions at ranges of applicability for the three most important sizing
maximum amplitude, then a quantitative value for flaw height methods.
cannot be obtained. The only conclusion is that the flaw was 9.2 General Guidelines for Flaw Sizing Evaluations—To
deep enough to approach the near surface to within the provide optimum conditions for flaw sizing in welds, the weld
standardized depth range. If accurately maximized signals are crown should be ground flush with the weld for the entire flaw
obtained, calculate the indicated height by the formula: length. Prior to attempting to size a planar flaw, the user must

FIG. 24 Optimum Ranges of Applicability for Flaw Sizing Methods

20
21
E 2192

FIG. 25 Guide to Confirm Flaw Existence, Location, and Depth Sizing

 E 2192
first verify its location and extent. This is typically accom- 9.3 Principles for Flaw Sizing—The activities as listed
plished with the original equipment used for detection. The should be accomplished.
next step involves searching for evidence of an extremely large 9.3.1 Verify location and extent of flaw.
flaw. The initial approach is to use the creeping wave tech- 9.3.2 Approximate the flaw depth with the far-surface
nique. The two methods of determining that a planar flaw is Creeping Wave or Mode Conversion Method.
large (High-Angle Longitudinal Wave Method and obtaining a 9.3.3 Search for evidence of a very large flaw:
signal from the second half-vee path corner reflection) should 9.3.3.1 Focused Longitudinal Wave or Focused Shear Wave
both be attempted to provide greatest confidence. If neither of Method,
these show that the flaw is large, the tip-diffraction method 9.3.3.2 Second Half-V Path Corner Reflection, and
should be attempted with the first half-V path technique. The 9.3.3.3 Tip-Diffraction.
Mode Conversion and Bi-Modal Methods should add confi- 9.3.4 Search for evidence of a small flaw:
dence to the results obtained. It is recommended that whatever 9.3.4.1 Tip-Diffraction, and
results are obtained, the user must confirm them using a 9.3.4.2 Bi-Modal Method.
9.3.5 If evidence of a large flaw exists:
complementary method. Additionally, if the results show that
9.3.5.1 Confirm with complementary method, and
the flaw is small, the user should also prove that it is not large
9.3.5.2 Prove that a small flaw does not exist.
by using a High-Angle Longitudinal Wave technique. This is
9.3.6 If evidence of a small flaw exists:
important since flaws can be multifaceted and a lower facet
9.3.6.1 Confirm with complementary method,
may provide signals indicating flaws less than 50 % through- 9.3.6.2 Confirm that a large flaw does not exist, and
wall while there are other facets that may be virtually 100 % 9.3.6.3 Confirmation may also be achieved by a different
through-wall. It is very important that large flaws are not angle of approach or by the opposite direction of approach.
missed and that small flaws are not called large. Flaw height 9.3.7 If two or more methods exhibit differing results:
determinations should also be made at several locations along 9.3.7.1 Eliminate those results with lowest confidence based
the flaw length to increase confidence levels and to increase the on range of applicability or repeatability or result,
chances of finding the deepest point of the flaw. In the case 9.3.7.2 In case of doubt, take value with greater height, and
where two or more methods produce different estimations of 9.3.7.3 Be aware of possible indications from weld fabrica-
flaw height, the user should be knowledgeable enough to select tion flaws.
the correct data. This requires understanding of the theory
behind each method as well as the limitations. The diagram 10. Keywords
shown in Fig. 25 presents a flaw evaluation and sizing flow 10.1 evaluation; examination; flaw-height sizing; flaw siz-
chart with five possible sequences for estimating flaw height. ing; nondestructive testing; ultrasonic

BIBLIOGRAPHY

(1) Advanced Ultrasonic Flaw Sizing Handbook, 1998 Edition, J. Mark (9) Electric Power Research Institute Training Course, “UT Operator
Davis, published by: The Art Room Corporation, 1998. Training for Sizing Intergranular Stress Corrosion Cracking (IG-
(2) Advanced Ultrasonic Flaw Sizing Training Course, Davis NDE, SCC),” Competency Area 911.
Inc., June 1999. (10) Full Characterization of Near-surface Flaws with Multimode
(3) A Time-of-Flight Method for Crack Evaluation Using Focused Straight-Beam Transducers, by G.J. Gruber & S.R.Burger, Intn’l
Ultrasonic Probes, L. Paradis, Y. Serruys, and R. Saglio, Materials Journal of Pressure Vessels & Piping, # 22, 1986.
Evaluation, Vol. 44 No. 4, April 1986, pp 568-570. (11) IIW Handbook on the Ultrasonic Examination of Austenitic Clad
(4) Changes in Ultrasonic Defect Sizing, by M.G.Silk, NDT Interna- Steel Components, R.J. Hudgel, compiled by Commission V
tional Vol. 20 No. 1 February 1987. IIS/IIW 1080-90 ex.doc. V-939-90, published by European Com-
(5) Crack Height Measurement-An Evaluation of the Accuracy of mission, Joint Research Centre, Institute of Advanced Materials,
Ultrasonic Timing Methods, by K.Date, H.Shimada & N.Ikenaga, 1994.
NDT International December 1982. (12) Ultrasonic Defect Sizing—Japanese Tip Echo Handbook, second
(6) Creeping Wave Probes in Ultrasonic Testing—Physical Principles, version, Edited by: K. Harumi, Y. Ogura, M. Uchida, Translated by:
Application in Welded Constructions, A. Erhard, H. Wustenberg, E. M. D. C. Moles, N. Miura Published by: Tip Echo Working Group
Schulz, E. Mundry and B. Ludwig, Welding and Cutting, May of 210 and 202 Subcommittee of Japanese Society for Non-
1983. Destructive Inspection, 1996.
(7) Defect Identification and Sizing by the Ultrasonic Satellite—Pulse (13) Nondestructive Testing Handbook, Volume 7, Ultrasonic Testing,
Technique, G.J. Gruber, Journal of Nondestructive Evaluation, Vol. American Society for Nondestructive Testing, 1991.
1, No.4, December 1980, pp 263-276. (14) On the Nature of the So-Called Subsurface Longitudinal Wave
(8) Detection of Near-Surface Cracks with Creeping-Longitudinal and/or the Surface Longitudinal “Creeping” Wave, K.J.Langenberg,
Waves, A. Erhard, H. Wustenberg, E. Mundry, FASE, Warschau, P. Fellinger, R. Marklein, Research in Nondestructive Evaluation,
18-22, 1978. Springer-Verlag, 1990, pp 61-81.

22
 E 2192
(15) Sizing of Planar Defects by Use of Double-Angle Ultrasonic Beams (17) The Use of the Head-type Waves for Ultrasonic Monitoring, I.N.
and Tip-Echo Times, M.A.M. Mahmoud, G.A. Hewitt and D.J. Ermolov, N.P. Razygraev, and V.G. Shcherbinski, Soviet Journal of
Burns, Materials Evaluation, Vol. 44 No. 9, August 1986, pp NDT, S 27/33, 1978.
1125-1131.
(16) Stress Corrosion Cracking, Recommended Practices, Canadian
Energy Pipeline Association (CEPA), Inspection Modu1e, Same
Side Sizing of Stress Corrosion Cracking Manual Procedure, by R.
Ginzel, 1997.

ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@astm.org (e-mail); or through the ASTM website
(www.astm.org).

23