N.

Kuppusamy
Ultrasonic Testing Of Austenitic Stainless Steel
Presented by Presented by
N.Kuppusamy N.Kuppusamy
Singapore Chapter
NDT HORIZON NDT HORIZON
Module 18A
N.Kuppusamy
INTRODUCTION
IN THIS PRESENTATION WE ARE GOING TO DICUSS THE
FOLLOWING:
PROBLEMS ENCONTERED DURING ULTRASONIC TESTING
HOW TO OVERCOME THE PROBLEMS
CALIBRATION AND SENSITIVITY SETTIN G
N.Kuppusamy
INTRODUCTION
The main practical implications of this are:
Welding procedure and preparation geometry have a strong
influence upon the capabilities of ultrasonic examination, so that
careful consideration of these factors at the design stage can
be very beneficial to the examination.
Many technical aspects of the examinations are strongly
influenced by the particular weld structure.
Only skilled, specially trained operators with a full knowledge
of the physical basis of the examination should be employed
Until recent years, austenitic steel welds were widely regarded
as uninspectable by ultrasonics.
Research and development have made it possible for a useful
level of examination to be carried out in many situations.
In general, though, the methods are more complicated and the
capabilities more limited than for the examination of welds in
ferritic steel.
N.Kuppusamy
Problems involved in the Examination of Austenitic Welds
The term austenitic covers a variety of materials and material
combinations, including austenitic stainless steels and nickel
chromium alloys such as "Inconel", "Incoloy", etc.
The capabilities of ultrasonics for the examination of welds in
austenitic materials are restricted compared to the ferritic case
because of the presence of large elongated anisotropic grains
(dendrites), often forming an ordered
columnar structure, which are
characterisitic of the austenitic weld
metal.
This type of grain structure can lead
to anisotropic ultrasonic behavior
contrasting with the isotropic
behavior of homogenous welds made in
carbon or low alloy steels.
SOLIDIFICATION BY THE
FORMATION OF
COLUMNAR GRAINS
N.Kuppusamy
Fine and coarse grained steel
at the same magnification
Fine grained steel Coarse grained steel
N.Kuppusamy
Problems involved in the Examination of Austenitic Welds
The size, the arrangement, and the elastic anisotropy of the
different grains result in high scattering associated with mode
conversion effects, beam distortion, and a variation of ultra-
sound velocity with direction and position in the weld.
The scattering of energy is observed as a relatively high noise
level (grass) and high attenuation.
The problems which occur in ultrasonic testing of austenitic
welds differ according to the parent material production
method (rolled, drawn, forged, or cast), the weld processes, and
the heat treatment as well as the composition of the parent and
weld metals.
L wave
Shear wave
Distortion
Mode conversions
Mode conversion1
Mode conversion
N.Kuppusamy
INTRODUCTION Contd
The capabilities for defect detection, positioning and size
assessment are more limited than for ferritic weld
examination.
So, monitoring the occurrence of small defects can rarely
be used for the quality control of welds, as is usual with
ferritic welds.
It may be necessary to use fracture mechanics to set less
rigorous defect acceptance standards for the particular
component. These acceptance standards should be
compatible with the limitations of the ultrasonic
techniques.
N.Kuppusamy
Difference
The ultrasonic methods applied to austenitic welds follow basically
the same principles as Ultrasonic Examination of Welds.
Some important differences do exist, to detect, locate, characterize,
and to estimate the size of weld defects.
The most important of these differences are the following:
Scattered energy from natural metallurgical discontinuities
generates noise indications at higher amplitude than would be
expected for the case of ferritic welds.
Shear wave
The choice of wave mode (longitudinal; shear)
and probe characteristics (sound field,
frequency, bandwidth, etc.) should be
optimized to allow a reliable separation of
weld defect indications from noise indications.
Scattering
S/N
Beamprofile
N.Kuppusamy
Difference
The ultrasonic beam has to cross different regions in the parent metal
and in the weld itself. The velocity of sound may vary along this path and
this may change the direction of the sound beam. Consequently, this may
result in inaccuracy in determining reflector positions.
Attenuation in the weld metal is generally more severe than for ferritic
welds and can be more or less pronounced depending on the angle of the
beam with respect to the preferred orientation direction of the grain
structure.
Therefore, the ultrasonic technique should seek to minimize beampath length
in the weld metal and, where possible, aim to take advantage of any directions
of reduced attenuation in the weld.
N.Kuppusamy
Difference
Beam divergence can also be directionally dependent. The beam
profile is usually different from that measured in parent plate
(whether ferritic or austenitic) so that size estimation methods
which depend on a knowledge of the beam profile, such as the so-
called dB drop methods, are not always suitable on austenitic welds.
Conventional instruments are used for examinations, but in most
cases, special probes need to be applied
N.Kuppusamy
400 1,500 1,500 5,000 1
250 1,000 750 3,000 2
25 100 100 200 5
Grey Iron
(mm)
SG Iron
(mm)
Coarse
Grained
Steel
(mm)
Fine
Grained
Steel
(mm)
Frequency
(MHz)
Typical maximum test ranges for compression mode
These are typical ranges. In practice, maximum range will depend
on the probe design, equipment, pulse strength, probe diameter and
specific material grain structure.
For shear waves, which have approximately half the wavelength,
the maximum shear wave ranges are approximately equal to a
compression wave of twice the frequency in the table above. For
example 2 MHz shear has a similar test range to 4 MHz
compression.
The improved penetration at low frequencies is obtained at the
expense of reduced sensitivity to smaller discontinuities
N.Kuppusamy
Testing of Austenitic Welds
Austenitic welds provide special challenges due to
the nature of the coarse weld metal grains
Austenitic stainless steels are the 300 series, such
as the 304, 309, 316, and 321 grades.
The 400 series are not austenitic, but are generally
welded with type 300 welding filler, so they have a
weld metal which is substantially austenitic.
N.Kuppusamy
Testing of Austenitic Welds
The word austenitic means that the material has not
undergone a transition to the ferritic form, which is
what happens with normal carbon and low alloy steels.
The consequence is that the grains of austenitic weld
metal are very large and have a different attenuation
and acoustic velocity to the parent metal.
In contrast, conventional weld metals have a much
finer grain size, and similar acoustic velocity to the
parent metal.
The large grain size of austenitic weld metal means it
has a very high attenuation.
Austenitic welds provide special challenges due to the
nature of the coarse weld metal grains
N.Kuppusamy
Testing of Austenitic Welds . . .
Complicating the issue is that these large grains are
anisotropic.
This means that they have different properties in
different directions within each grain.
This means that each grain boundary can be an interface.
The degree of bending of
the beam can be as much
as 40 for shear waves and
20 for compression waves.
N.Kuppusamy
Effects of Austenitic Structures on Ultrasound Propagation
The effect of austenitic structures on the behavior of ultrasound
depends largely on grain size.
Small grains, as found in rolled plate, have no adverse effect on
sound propagation.
On the other hand, coarse grain cast structures and those in the
welds have marked effects, leading to increased scatter and
attenuation, variations in sound velocity, and often to beam-
distortion.
Simple apparatus to demonstrate
ultrasound propagation behavior
The effects are primarily due
to the anisotropic nature of
the austenitic grains.
N.Kuppusamy
Velocity change with refracted angle
Elastic anisotropy leads to variations in the propagation velocity of
ultrasound waves. In general, the propagation velocity depends on the angle
between the wave front and the major axes of the columnar grains.
Longitudinal Wave (C
L
)
Shear wave (horizontally
polarized -C
TH
)
Shear wave
(Vertically Polarized-C
TV
)
For the specimens with
their axes machined
parallel to the surface of
the original weld sample
the amplitude of the
transmitted signal varies
systematically
N.Kuppusamy
Conditions to be stabilized
1. Personnel already discussed
2. Required information About the Welds.
To obtain an effective ultrasonic technique with optimum flaw
detection capabilities (e.g. choice of optimum probe parameters), it is
necessary to collect maximum information on the weld characteristics.
1. Weld geometry - Selection of probe parameters depends on
weld preparation and on the degree of penetration.
2. Welding process - Knowledge of the weld process and
procedure will contribute to an estimate of the likely grain
structure
3. Heat-treatment - Information on the heat treatment cycle is
useful to provide an estimate of the parent material grain
size. Using this information, the likely transparency can be
assessed.
4. Representative Weld samples (in terms of weld preparation and
process, procedure, geometry, heat treatment, heat input, etc) to
assess the possible beam deviations resulting from the angle
of the weld fusion faces
N.Kuppusamy
Conditions to be stabilized
3. Surface Preparation and Marking
Full volumetric inspection requires dressing of the weld crown.
When the weld is fully ground, the operator will not be able to visually
identify the region to be examined (or a repaired zone) unless these
areas are clearly marked.
A good practice is to mark the surface by punch marks. These marks
may be made after completion of the weld to indicate the real fused
zone of the weld.
The material surfaces must allow free movement of the probe(s) and
provide satisfactory conditions for the transmission of the ultrasonic
waves.
Therefore, the surface roughness should generally not exceed 20 m
and the waviness should not exceed 0.5 mm over any area of 50 mm x
50 mm.
This is necessary to avoid disturbance of the ultrasonic beam which
could reduce the sensitivity and result in errors in defect location.
N.Kuppusamy
Conditions to be stabilized
3. Surface Preparation and Marking
Where there is access to only one surface of the weld, any weld
reinforcement must be ground off if the whole weld volume is to be
examined and if shear waves cannot be applied. This is because mode
conversion losses on reflection reduce the effectiveness of
examination in the second traverse with angled compression waves.
Furthermore, on each side of the weld for a minimum of 5/4 skip (1
skip) distances, the surface should be free from weld spatter, loose
scale, machining and grinding particles, dirt, paint, or other foreign
matter.
4. Condition of the Parent Metal
To allow adequate penetration of ultrasonic waves into the weld, the
parent metal should be transparent to ultrasound and free from large
flaws.
As a guide, difficulties in penetrating parent material can become
severe for grain sizes larger than ASTM 3 (average grain size 0.125
mm.)
N.Kuppusamy
Conditions to be stabilized
5. Agreements before the Start of the Examination
To avoid misunderstanding and arguments about the examination to be
carried out, a number of conditions or directives need to be
established prior to the examination.
1. Extent of the Examination
2. Sensitivity Required
3. Special Conditions
4. Regular Check of Equipment
5. Reporting
6. Visual Inspection
The visual appearance of the welded joint should be recorded with
particular reference to visible defects and the shape of the weld,
e.g. surface curvature, degree of root penetration, backing ring,
different parent metal thicknesses, extent of the reinforcement,
presence of undercut, weld finish, and alignment of parts.
N.Kuppusamy
Conditions to be stabilized
7. Couplants
A couplant, usually a liquid or semi-liquid, is required between the face
of the probe and the surface being examined to permit transmission of
the acoustic energy from the transducer to the material under test.
Typical couplants include water, oil, grease, and glycerine.
The couplant used should form a film between the probe and the test
surface.
It should not be injurious to the material to be tested - or disturb
subsequent surface treatment.
This is of particular importance in examining austenitic materials where
coupling residues may cause problems such as stress corrosion cracking
in service.
Couplants containing halogens and sulpher are to be avoided.
N.Kuppusamy
Testing of Austenitic Welds . . .
The best conditions for examination of austenitic weld metal
occur when:
the compression mode is used, as its longer wavelength
results in lower attenuation, and less beam bending occurs,
and
the beam travels either parallel or at approximately 45 to
the grain orientation for minimum bending.
Effective testing needs to be a cooperative effort between
the welding engineer and ultrasonic specialist and should be
well planned.
Do not be frightened by austenitic materials just follow
basic principles.
N.Kuppusamy
Testing of Austenitic Welds . . .
Compression probes are generally used when ultrasonic
testing of austenitic welds
Compression angle probes are constructed with very short
pulse lengths, with a wedge angle less than the first critical
angle, such that there is a shear and compression wave
generated in the material. The shear wave travels at a slower
velocity than the compression mode, and can generally be
disregarded after careful interpretation.
It is also risky to skip off far walls to test a weld, as the
process of skipping off a surface will cause the compression
wave to split off yet another shear component, which will
confuse interpretation, as well as weakening the primary
compression mode.
1. Shear wave
2. Compression wave
3. Reflected shear wave
4. Mode converted compression wave
5. Mode converted shear
6. Reflected compression wave
N.Kuppusamy
Testing of Austenitic Welds . . .
Note that this does not make it impossible to test
austenitic welds by skipping, but it does raise the
complexity.
There have been some very innovative special test
procedures developed that have used the mode converted
shear, or other waves to their advantage, but we will leave
those to the experts to worry about. Many of these
procedures use separate transmit and receive probes
(tandem tests).
Where possible, it is preferable to keep it as simple as
possible by working within the half skip section of the
beam path and ignore the echoes that follow!
N.Kuppusamy
Calibration
Calibration is different to conventional angle shear wave probes
Calibration is best carried out with a V2 (AS2083 Block 5)
block made of austenitic steel.
Range
It is possible to calibrate the range with a zero compression
probe, as the acoustic velocity is the same for the angle and
zero compression beams. Calibrate the range as if using a zero
compression probe.
Zero
Set the zero as you would for a conventional angle probe using
the 50 mm radius of the V2 Block. Be careful that you calibrate
against the shortest reflection, as the longer one may be a
shear wave.
Sensitivity and DAC
This is best done using a block similar to the IOW Block
(AS2083 Block 2). Be careful when setting sensitivity, because
the shear wave will also be striking holes and reflecting back to
the probe. Make sure by calculation and measurement that you
are striking the correct hole with the right mode!
N.Kuppusamy
Calibration Block for examination of Austenitic Welds
This block is one design that may be useful for setting sensitivity and
has a series of inclined surfaces and reference holes for verifying
equipment performance.
Reference blocks welded with similar materials and welding procedures
are invaluable aids to effective testing and final validation.
Validation should also be
made on a reference block
containing a weld of similar
geometry and welding
procedure, with reference
reflectors located at key
point in the cross section to
demonstrate that the
combination of weld and test
procedure are compatible.
This will demonstrate if
there are any major
anomalies due to beam
bending or variations in
attenuation.
N.Kuppusamy
Validation block for Austenitic steels
This block allows the performance to be verified when scanning through
both weld and parent metal.
Austenitic plate does not have the same limitations as welds or castings
due to the grain size reduction in rolling.
Fortunately, austenitic steel plate has very similar ultrasonic properties
to carbon and low alloy steel plate, so can be tested similarly, with an
adjustment to calibration for the slight variation in acoustic velocity.
In certain situations, you can use conventional zero and compression
probes to undertake a useful examination of an austenitic weld by
approaching the fusion faces through the parent metal.
This is of course a limited
inspection, but may be useful in
some instances if you do not
need a volumetric scan of the
weld metal and have limited
probes.
SKETCH TO
CKECK
N.Kuppusamy
Summary
Wave Type
The use of conventional shear wave probes should always be
considered first. If the signal-to-noise ratio is insufficient for an
effective inspection,
Special probes like refracted longitudinal angle beam probes,
including the surface wave type, need to be used.
Selection of Probes
The major variables to be considered when selecting a probe are
(1) wave type
(2) angle
(3) frequency
(4) type of probe
(5) size and geometry of probe and component
Signal-to-Nolse Ratio
For a reasonable ultrasonic examination, the noise level should be
at least 6 dB below the recording level for the whole sound path
N.Kuppusamy
Summary
Probe Angle
In general, the probe angle should be selected to be suitable
for the defect orientation expected.
This means that if possible, the incident angle should be chosen
to strike a defect perpendicularly for maximum echo amplitude.
For weld examination,it istherefore necessary to know the weld
geometry priorto and after welding;
Particularly the details of the weld root if this is not machined
or ground flush.
Recording level
In the absence of any experience and of any prescribed criteria, 50% of the DAC for
reference reflectors indicated through the weld might be a good first approach.
In certain cases, it may be necessary to go down to 25%.
However, the recording level in the appropriate depth zone should be at least 6 dB
higher than the maximum noise level in that depth zone.
N.Kuppusamy
Probe Angle
The entire volume of weld and fusion zone should be
covered with at least two angles of incidence is shown
in the Figure.
1. Surface and near surface area can be
tested with 70 (2MHz) probe.
2. Rest of the weld volume and Root can
be tested with 45 probe.
If the weld penetration is not ground
flush, geometrical or false echoes may be
generated with a 45 probe, and
for that reason, a probe witha larger
angle (60) should be used to evaluate
the root along with 45 probe.
N.Kuppusamy
Probe Angle
The choice of probe angle largely depends on the
weld preparation and weld surface condition.
It should also be remembered that examination at
more than halfskip distance is not very effective
due to mode conversion effects at the backwall.
To reach the lower part of the weld with large angle
probe requires a long sound path and this can also
reduce the effectiveness of the examination
N.Kuppusamy
Special Techniques
Surface wave technique
A derivative of the angled longitudinal wave probe is the "surface
wave" probe.
The surface wave is generated at the first critical angle of incidence,
as shown in the Figure, and propagates along the surface as a
compression wave.
It is also referred to by other names, e.g. head wave, lateral wave,
fast surface wave at the first critical angle. Unlike Rayleigh waves,
the surface waveis not damped by couplants on the component
surface, nor does the beam follow undulations in the surface.
A surface wave probe generates
compression waves at large angles between 70 and 90
shear waves according to Snell's Law
Probe
33 S wave
Primary Surface waves
Secondary Surface waves
N.Kuppusamy
Special Techniques
Surface wave technique
The surface wave sound velocity is identical to that of compression waves.
Although it generates a beam with "complicatedcharacteristics, it is a very
useful probe for detecting surface defects.
A surface wave probe can also be considered for inspection of the weld
root. The 33 shear wave component of the surface wave probe converts to
a secondary surface wave at the backwall of the component, as shown in
the Figure.
In this case, the weld penetration echoes can largely be eliminated
(because it does not follow contour).
The advantage of a 70 probe with a strong surface wave component is
useful for thin materials and near surface evaluation of thick materials.
Probe
33 S wave
Primary Surface waves
Secondary Surface waves
N.Kuppusamy
Special Techniques
Tandem Techniques
Special tandem techniques as shown in the figure can be used as
directed by experts.
Longitudinal Longitudinal waves tandem technique (or)
Longitudinal Shear wave tandem technique can be used.
R
T
T R
45L
70L
70L
70L
31S
31S
N.Kuppusamy
Rate of probe movement
The rate of probe movement shall not exceed 50 mm/sec,
unless the examination capability has been verified at the
higher scanning speed.
This reduced speed is because of the low signal-to-noise
ratio compared to ferritic welds.
N.Kuppusamy
Frequency
Frequencies from 1 mHz to 5 mHz can be used, depends up
on thickness and graininess of the material.
For thickness over 25mm 2 & 1 mHz probes are
recommended for course grained materials.
For thickness up to 25mm 4 & 2 mHz are
predominently used.
N.Kuppusamy
Type of Probe
Varities of probes are available for selection. Following probes can be
selected for testing. The selection is often orbitrary.
Single Crystal (Advantage of regularly decreasing DAC)
0 Longitudinal wave probe for on the weld scanning.
45, 60, 70 Longitudinal wave probes for weld scanning
(limited to half skip)
45, 60, 70 shear wave probes for weld scanning (can be
used up to full skip & useful for low thickness materials).
Surface wave probes for top & bottom near surface testing.
Twin Crystal (eliminates dead zone)
45, 60, 70 Longitudinal wave probes for weld scanning
(limited to half skip)
45, 60, 70 shear wave probes for weld scanning (can be
used up to full skip & useful for low thickness materials).
N.Kuppusamy
size and geometry of probe and component
The probe should be of a size which allows manual examination
with good acoustic coupling.
Coupling involves an interaction between probe size, roughness,
and the geometry of the component.
If the probe dimension W fails to meet the requirement:
R > W
2
/4, (dimensions in mm) or
the coupling gap exceeds 0.5 mm.
The probe shoe should be adapted to the geometry of the
component.
R = Radius of the curved surface under examination and
W = Probe dimension (length or width).
N.Kuppusamy
Multi TRL Probe Concept
In the case of thick welds, more
probes are necessary.
The advantage in more probes
is that the probe angle can be
an optimized selection for flaw
detectability in each depth
zone.
All compression wave TR
probes have a very low
sensitivity in the first part of the
range behind the acoustic zero
point (mechanical zero).
This causes a very small dead
zone.
The surface wave probe fully
eliminates this effect as shown
in the Figure.
100MM
40MM
20MM
5
0
I
II
III
IV
TRCreep 70TRL 60TRL 45TRL
f10 f25 f50 f75
PROBE 1 2 3 4
( f =Transit Distance
at Focus)
f
f
f
DEPTH(THICKNESS)
f
f
f
f f f
1 2 3 PROBE
0dB
-6dB
80%
60%
40%
20%
0
I II III IV ZONE
0 5 20 40 100mm
DEPTH(THICKNESS)
4
A
M
P
L
IT
U
D
E
N.Kuppusamy
End of Session
(End of Tech-Talk)