Acoustic Emissions are typically broadband and therefore can raise many propagating modes

in a beam. Holford (1999) found, that most of the energy is carried by the first symmetric anda
asymmetric lamb modes. The first asymmetric lamb mode, here referred as flexural wave,
shows strong dispersive behavior at low frequency thickness numbers. Therefore, an AE
initiated flexural wave pulse distorts and diverges as it travels along the beam. The presented
method for locating AE sources consists of two steps. In a first step, the lateral deflection due
to the incident flexural wave is measured at the end of the structure by means of a laser
vibrometer.

In a second step, the measured displcement data is reversed in time and set as the boundary
condition in a numerical model of the beam. In the simulation, the distored flexural wave form
recompresses and reaches maximal amplitude at the location of its origin. This allows the
detection of the AE source by simply finding the local mxima in an x-t diagram, also known as
Lagrange diagram.

After having converged at the location of the AE source, the different frequency components
of the flexural wave will diverge again leading to a dicrease in the amplitudes of the flexural
wave. This allows the detection of multiple AE events with a single measurement. The method
described here here requires only a single, one point, unidirectional measurement, because
only the dispersin of the flexural wave mode is used. A further advantage is, that multiple AE
can be detected because the wave amplitudes decrease after having converged at the source
position.

Multiple AE sources at an identical position are temporally separated in the Lagrange diagram.
Acoustic emission parameters extracted from the simulation of several AE sources at different
positions can easily be compared because dispersive effects o the shape of the waveform are
inherenthy compensated. However, the effect of damping on the wave form is not
compensated.

ATTENUATION

As the acoustic wave travels through the structure, its amplitude decreases. This effect is
known as attenuation and is illustrated (for a Steel plate) in figure 10-7. In this figure,
amplitude is plotted on the Y-axis using the dB decibel scale. On tjs scale, each increment of 20
decibels is a tenfold increase in the signal peak voltaje. The graph shows that the peak signal
voltaje 9 feet fro the source is about one-thirtieth of the voltaje very close to the source. The
dB scale is universally used and very convenient because, bein logarithmic, it condenses the
very wide range of AE signal amplitudes. dB is a logarithmic measure of acoustic emission
signal amplitude referenced to 1uV.

Attenuation is due to several factors. In most structures, the importan tones are geometric
spreading, scattering at structural boundaris, and absorption. When the source is oly a few
inches from the sensor, as in local monitoring of Weld details, the geometric spreading effects
are the most important. These effects are therefore especially important for crack
detectability. At distances greater tan a couple of feet, energy absortion and structural
scattering are the most influential properties. The understanding of these effects is also
important for identifying extraneous noise.
Geometric spreading effects are fundamental to wave propagation. Basically, the sound wave
is trying to spread throughout the volumen of the structure near to the source; the change in
the static stress field and the wave motion are at their mximum.

In theory, for a structure that is large in all dimensions (infinite half-space), a stress wave
would simply spread and continue to attenuate. In realistic structures, the boundaries forc
the stress wave to remain in a confined space such that attenuation from geometric beam
spreading is minimal and stress waves may travel great distances. The data show in figure 10-7
illustrate this relationship (aproximatelly). The curve falls steeply at short distances and more
gradually at larger distances. The 30% figure applies to a circular wave front spreading out in a
plate-like structure such as a girder web. In a solid mdium like concrete, the wave spreads out
in all three dimensions and it loses amplitude more rapidly 50% for each doubling of
distance. In a rod, the wave is channeled and cannot spread, so attenuation is relatively low.

The second major cause of attenuation is absorption is reflection (scatter) at structural

boundaries and geometrical discontinuities also produce mode conversions. These effects are
especially important in the complex geometry of many structures, where there may be
changes in direction, connections, stiffeners, and other boundaries along the acoustic path
from source to sensor.

The third cause of attenuation is absorption. Here the elastic and kinetic energies in the wave
are absorbed and converted into heat by the material through which the wave is passing. Steel
absobs very Little at the frequencies used for AE testing. Nonmetallics, in general, and Paint, in
particular, tend to absorbd energy more than Steel. Absorption is greater at higher frequencies
due to the shorter wavelengths at these frequencies. The distance/amplitude reduction
mechanism for absorption is different from the mechanism of beam spread/amplitude
reduction. In the case of distance/amplitude, there is a constant number of decibels
absorption per foot from the source. This is the same at all distances from the source. For
example, if there is a 6 dB reduction after the energy has traveled an additional foot, and so
on. Thus, whreas energy losses due to beam spreading near the source are high, absorption
also accounts for large osses of energy as the wave fronts move further away from the source.

Attenuation measurements are easily made with a simulated AE source. The most widely used
simulated AE source is the breaking of a pencil leas pressed against a structural member, as
illustrated in figure 10.8. as the lead is pressed against the structural member, the applied
force produces a local deformation that is suddenly relieved when the lead breaks. With good
technique, the resulting stress wave is amazingly reproducible.

The Hsu pencil (named after the developer of the technique) and the accesory Nielsen shoe
are convenient, inexpensive aids that have been enormously valuable in practical AE testing.
The breaking of the lead creates a very short-duration, localized impulse that is quite similar to
a natural acoustic emission source such as a crack. Furthermore, the amplitude of the lead
bread source is well within the range of typical crack sources. The Hsu pencil has become so
well accepted as a simulated AE source, that in some procedures for wide-area monitoring, the
mximum permisible sensor spacing is base don the ability to detect lead breaking from
anywhere in the inspection rea.

The usual procedure for developing a graph such as shown in figure 10-7 is to break lead
several times at each of several different distances from a sensor, and then to record the
amplitude for each break. The amplitudes for each distance are averaged and, subsequently,
the average amplitudes are plotted against distance.
The atteniation curve is an important aid in determining the sensor placements for the specific
aplication. In many acoustic emission applications, to goal of the inspection is to monitor the
entire structure. In this case, it is imprtant that all parts of the structure are within detection
range of at least one sensor. In global monitoring tests of this kind, the test procedure typically
specifies how the attenuation curve can be used to determine aceptable mximum sensor
spacing.

In some tests included in these guidelines, the recommended strategy is limited-area

monitoring focusing on a known crack or possibly cracked site. Here the attenuation curve has
different applications. It shows how sensitivity to the crack compares with the sensitivity to
distant noise sources and whether distant noise sources will reach the sensor or whether they
will be attenuated before reaching it. Also, it shows that sensors close to the crack must be
placed at similar distances in order to get similar responses.

Artificial sources

The artificial vertical force sources have been used in this work. They are: 1) the fracture of a
glass capillary 0.05mm in diameter, 2) the fracture of 2H pencil leads 0.3mm and 0.5mm in
diameter, and 3) the impact of a stainless Steel ball 2.38 mm in diameter dropped from a
height of approcximately 5 cm above the plate surface. The capillary fracture is a well defined
source and has been shown to be a step function of 0.5 ms rise time is ilustrated in figure 1(a).
the pencil lead fracture is similar but as a distinctive dip at the wavefront. A typical pencil
lead fracture source signal ins shown in figure 1(b) where the amplitudes A1 and A2 and the
times t1 and t2 are dependent upon the lead type, the lead extensin length, the angle the
lead makes with the plate Surface at fracture, and the force with which it is broken.

Because of its wide bandwidth, the capillary fracture signa lis used as a reference source from
which the pencil lead fracture and ball impact sources are calculated. In this work, the rise
time of the capillary fracture source is neglected and the source signa lis considered to be a
perfect step function. The calculated source functions should show good quialitative
agreement with the signas shown in figures 1(b) and 1(c).

Experimental setup

Experimental measurements were made on two different flat plates of A533B pressure vessel
Steel. The plates were mounted in a horizonta position with the receiving transducer held in
the center of the lower Surface opposite to the receiving transducer. The experimental setup
and the dimensions of both plates are shown in figure 2. Epicenter refers to the source
position directrly above the receiving transducer, and 2h and 4h refer to the positions two and
four plate thicknesses from epicenter. The large plate had dimensions such that edge
reflections were not presented in the recorded waveforms; that is, it could be considered to be
a plate of infinite extent. However, signals recorded from the small plate contained significant
edge reflections within the recording interval.

The signals were received by several different piezoelectric transducers whose elements were
1.35mm in diameter. The transducers were spring loaded against the bottom of the plate and
acoustically cupled using a commercially available oilbased couplant. Signals of 40 us duration
were recorded with a trnsient digitizer as 512 8-bit words for subsequent analysis on a
minicomputer using an interactive AE signal processing program.

Source characterization large plate results

Pencil leads were fracture don the Surface of the large plate at epicenter, 2h and 4h. for
comparison, two different receiving transducers were used at 2h. the recorded signals are
shown in figure 4. The signals from capillary fractures at epicenter 2h and 4h were used to
obtain experimental pencil lead source functions as described under Source
Characterization. For each plate position, an inverse function was found that, when convolved
with the capillary fracture signal at that position, yielded the capillary fracture source function.
This inverse function was then convolved with the respective pencil lead fracture signal to
obtain the experimental source function. The source functions obtained in this manner are
shown in figure 4. They all show good qualitative agreement with the expected pencil lead
fracture source function shown in figure 1(b). all have an initial dip followed by a rapid rise to a
constant value. The results obtained are indeendent of the particular transducer used to make
the measurement, as can be seen by comparing the two source functions at 2h. the high
frequency noise on some of the source functions is primarily due to transducer bandwidth
limmitations and can be removed by low pass filtering.