In general, defect sizing and characterization cannot be

reliable for indications from within austenitic weld

metal. However, the assessment routine detailed below is
helpful since indications which plot out as having
significant through wall extent are unlikely to be
associated with insignificant defects. (It must be
remembered that the reverse is not necessarily true. That
is, it is possible for a large defect to be represented by
a small indication.) For each probe used, indications
detected at the specified testing sensitivity shall be
examined by probe movement in two mutually perpendicular
directions. Angle beam scans shall be made directions
towards and away from the reflector, and at right angles
to these directions. The response of each defect shall be
noted, and the defect shall be classified in terms of
Patterns 1 to 3, as detailed below.

Pattern 1
A single sharp signal rising smoothly in height to a
maximum then falling smoothly to zero (see Figure A).

Pattern 2
A single sharp signal rising smoothly in height to a
maximum which is held with or without minor amplitude
fluctuations with probe movement and falling smoothly to
zero (see Figure B).

Pattern 3
A signal, or group of signals, which may be fully or
partially resolved, fluctuating in height as it/they rise
to and fall from one or more principal maxima (see Figure
C).
For Pattern 1 behavior, it is not possible to estimate the
defect size because the width Of the beam in austenitic
weld metal Cannot be accurately known. Such indications
are often associated with volumetric defects, but it is
recommended that indications of this type be assessed by
other angles of probe to confirm a volumetric character.

Pattern 2 and Pattern 3 behavior are likely toindicate the

presence of a planar defect. A rough estimate of defect
height can be obtained by plotting the positions of the
last maxima in the echo response (for Pattern 2 this
involves plotting the positions corresponding to the
extremes of the “plateau” in the echo response).

These techniques should only be applied when the operator

has a clear understanding of the principles involved.

In some circumstances, specialized techniques (Example:

Modal Technique) may be of value,
Fig. A

Fig. B

Fig. C
Bi - Modal Technical

1.0 Description

1.1 The Bi-Modal Technical uses a dual element search

unit, either side-by-side or tandem orientation, producing
a pulse train of longitudinal and shear waves. Sizing
estimates are obtained by reflecting a refracted
longitudinal wave from the crack tip, mode converting a
direct shear wave from the crack face, and mode converting
an indirect shear wave to produce an ID Creeping Wave,
which reflects from the crack base.

1.2 The two basic crack sizing techniques for the Bi-Modal
Tecnical are:
1. Time of Flight (TOF), Tau or M-AATT, which is the
peaked Pulse 1 signal, set to specific screen ranges.
2. Delta Time of Flight (Δ TOF), Sigma or M-RATT, which
is the screen division of separation between the Pulse
1 and Pulse 2 signals.

1.3 The Bi-Modal techniques are very effective for sizing

ID connected planar cracks in the material thickness range
of 0.3" to 1.5 inches. Due to the pseudo-focusing effect
of the incident angles of the transmit and receive
crystals, the Bi-Modal technique is most effective for
sizing ID connected mid-wall cracks, generally 30 to 70 %
through wall.
1.4 Search units designs are as specified by the
manufacturer. Examples of acceptable search units include:
ADEPT 60, SLIC 40, and Sigma Bi-Modal. Other search units
for the Bi-Modal Method may be used as demonstrated by an
acceptable Bi-Modal calibration on the sizing calibration
block.

1.5 Generally, the frequency of the Bi-Modal Search is

approximately 3 to 3.5 MHz. Other frequencies may be used.

1.6 The Bi-modal Method is defined as follows. Three main

signals are noted:
 Pulse 1 is the reflected longitudinal wave from the
crack tip.
 Pulse 2 is the direct shear wave which mode converts
to a longitudinal wave and then reflects a
longitudinal wave from the face of the crack.
 Pulse 3 is an indirect shear wave which mode converts
to an ID Creeping Wave at the ID surface of the
component.
 A fourth signal is noted as a satellite pulse signal
of Pulse 1. Sometimes called 1 star. Generally, this
signal is not used for calibration.

2. Calibration

2.1 Time of Flight (TOF), Tau / M-AATT Calibration

2.1.1 Determine the thickness of the component to

be sized and obtain a calibration block of the
 similar material and nominal thickness with notches
in a range of at least 20% to 80% through-wall
depth.
2.1.2 Place the search unit on the edge of the
sizing calibration block. Identify the three basic
signals, or "Pulse Train", Pulse 1, Pulse 2, and
Pulse 3. Also, note the 1 star signal.
2.1.3 As a ranging technique, adjust the range
control to separate Pulse 2 and Pulse 3 to
approximately 2 major divisions.
2.1.4 Place the Bi-modal search unit to obtain the
Pulse 1, 2, and 3 signals from the 80% notch. Peak
the Pulse 1 signal. Using the delay or zero offset
control; adjust this Pulse 1 signal to 1 horizontal
screen divisions.
2.1.5 Place the Bi-Modal search unit to obtain the
Pulse 1 signal from the 20% notch. Using the range
or sweep control, adjust the Pulse 1 signal from
the 20 % notch to 4 screen divisions.
2.1.6 Repeat steps 2.1.4 and 2.1.5 until a linear
screen range is obtained such that the 80% Pulse 1
signal is at 1 screen division and the 20 % Pulse 1
signal is at 4 screen divisions. The 60% and the 40
% Pulse 1 signals should be at 2 and 3 divisions
respectively.

2.2 Delta Time of Flight (Δ TOF), Sigma / M-RATT

2.2.1 With the (TOF), Tau / M-AATT technique

calibration complete, record the horizontal screen
divisions of separation between the Pulse 1 and
Pulse 2 signals for each of the notches, e.g., 80%,
60%, 40%, and 20%.

2.3 With the TOF/Tau and Δ TOF/Sigma calibrations

complete, record the screen divisions of separation
for the Pulse 2 and Pulse 3 signals.

2.4 If during the crack sizing evaluation, the

separation of the Pulse 2 and Pulse 3 signals
change from that observed during calibration, then
the Delta Time of Flight or Sigma depth sizing
estimate maybe incorrect due to a crack that is
oriented other than perpendicular to the ID
surface, or there has been a change in thickness.

2.5 As such, the Pulse 1 signal sizing estimate is the

most accurate depth estimate.

3.0 Scanning/Evaluation

3.1 Scan the area of interest. First observed the Pulse 3

signal to verify the crack is ID connected. Then observe
the Pulse 1 signal. Peak the Pulse 1 signal to obtain a
sizing TOF/Tau depth estimate.

3.2 Once the TOF/Tau sizing depth estimate is obtained,

then measure the screen divisions of separation between
the Pulse 1 and Pulse 2 signals. This is ΔTOF/Sigma depth
estimate. Compare the separation of Pulse 1 and Pulse 2 of
the crack to the separation estimates from the notches to
determine the crack height estimate.

3.3 Adjust the instrument gain such that the average noise
level is about 10% to 15 % full screen height (FSH).

3.4 If possible, repeat the sizing examination from the

weld side to verify the signals obtained from the base
material side of the weld.

3.5 Scan along the length of the crack to determine the

deepest sizing estimate.

4.0 Limitations

4.1 The ultrasonic instrument should have a RF displays to

aid the sizing examiner to resolve the crack tip signal.
The crack tip signal may have a low signal-to- noise
ratio. The RF display may help the sizing examiner to
detect the crack tip signal.

4.2 The Bi-Modal techniques are somewhat difficult. The

sizing examiner must be thoroughly trained in the
applications of the TOF/Tau and ΔTOF/Sigma techniques.

4.3 The Bi-Modal Method may detect multiple crack tip

signals. The sizing examiner must be aware that the
highest amplitude signal does not always indicate the
deepest point of the crack.

4.4 If Pulses 2 and 3 are difficult to interpret, then the

Delta Time of Flight or Sigma measurement may produce
false sizing estimates.

4.5 The crack depth must be estimated on the basis of the

Time of Flight or Tau measurements when both Pulse 2 and
Pulse 3 are absent.

4.6 The Pulse 1 signal for TOF/Tau may be difficult to

detect since oriented cracks may not produce Pulses 2 and
3.