Mechanical Systems and Signal Processing 52-53 (2015) 217–227

Contents lists available at ScienceDirect

Mechanical Systems and Signal Processing

journal homepage: www.elsevier.com/locate/ymssp

Waveform-based selection of acoustic emission events

generated by damage in composite materials
Emmanuel Maillet n, Gregory N. Morscher
Mechanical Engineering Department, The University of Akron, Akron, OH 44325-3903, United States

a r t i c l e i n f o abstract

Article history: Acoustic emission (AE) has been shown to be a promising health monitoring technique for
Received 14 February 2014 composite materials as it allows real-time location and identification of damage. When
Received in revised form attempting to relate the recorded acoustic emission to a material's mechanical behavior,
12 June 2014
the relevance of results relies on an accurate selection of AE originating from material
Accepted 1 August 2014
damage. Indeed, during mechanical tests most of the recorded AE is generated outside of
Available online 23 August 2014
the volume of interest and without proper filtering these AE signals can significantly affect
Keywords: the analysis. To date, there exists no common procedure for the selection of AE signals and
Composite materials therefore results can hardly be compared or transposed to different composite systems.
Health monitoring
In the present paper, a new waveform-based procedure is proposed for the selection of AE
Acoustic emission
events generated by damage. The procedure includes accurate localization and selection
assessment based on signal energy.
& 2014 Elsevier Ltd. All rights reserved.

1. Introduction

A wide range of composite materials are used or anticipated in aerospace applications, including ceramic matrix
composites (CMCs) that are being considered for combustor liners and rotating components in the next generation jet
engines [1]. Their successful implementation requires the acute understanding of the damage modes that are involved and
their impact on life expectancy. Among the numerous existing health monitoring techniques, acoustic emission (AE) is
unique in that it is based on the recording and analysis of elastic waves that are generated by material damage. Using few
remote sensors, AE allows real-time monitoring and localization of damage. The cumulative energy of acoustic emission
signals has been shown to be directly related to transverse matrix crack density in ceramic matrix composites [2,3] and in
polymer matrix composites (PMCs) [4]. In CMCs subjected to static loading at intermediate temperatures, the energy of AE
events has also been used as an indicator for lifetime prediction [5,6]. Finally, in both PMCs and CMCs, one of the main
current challenges is the use of AE for damage identification, based on the assumption that AE signals carry specific features
that are related to their source damage [7–9].
The literature presents promising results showing the ability of acoustic emission to monitor and identify damage in
composites. However, results can hardly be compared and the developed approaches cannot easily be transposed to other
composite systems because most works (1) suffer from the absence of a robust procedure for the selection and analysis of AE
signals originating from damage and (2) fail to account for effects of propagation on the recorded AE signals. The effects of
propagation have recently been addressed as they relate to the recorded AE energy [5,6]. The present paper will focus on the

n
Corresponding author. Tel.: þ1 330 972 7251; fax: þ1 330 972 6027.
E-mail address: [email protected] (E. Maillet).

http://dx.doi.org/10.1016/j.ymssp.2014.08.001
0888-3270/& 2014 Elsevier Ltd. All rights reserved.
218 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227

Fig. 1. Typical 3-sensor configuration for AE monitoring during tensile testing of a composite specimen. The broken lines represent the gage length. The
gray areas correspond to the surfaces in contact with the machine grips.

selection of AE events. Morscher et al. [10] showed that during mechanical testing of a typical 150 mm long composite
dogbone specimen 70–80% of the recorded AE originates from outside the gage section. If not properly filtered out, these AE
signals can greatly affect the analysis. Therefore, the objective of the present work is to develop a robust procedure for the
selection of AE events originating from damage.
When attempting to relate acoustic emission to a material's mechanical behavior, the analysis must be limited to AE
events generated by the material within the gage section (i.e. for a tensile test, typically a 25.4 mm long zone in the center
of the testing specimen where strain is measured). These AE events are here referred to as gage events. Fig. 1 shows a typical
3-sensor configuration for AE monitoring. In this configuration, the selection of gage events is straightforward as those
events trigger the middle sensor first. Top and bottom sensors are referred to as guard sensors. Events triggering one of
the guard sensors first might be associated to damage occurring between the top and bottom sensors, but could also
originate from noise-related signals generated in the grips. The present approach is similar to the one known as event-
definition-time (EDT) filtering where an AE event is kept only if it triggered the various sensors of an array of sensors within
a specified time interval after it triggered the first sensor, thus ensuring that the AE event originated from the volume of
interest. In the present approach, the distance between sensors is set so that the volume of interest around the middle
sensor matches the extensometer's gage section. The 3-sensor configuration cannot always be applied because of testing
conditions (e.g. test at high temperature) or sample geometry (e.g. single tow minicomposites). Therefore, the selection
procedure must ensure that, in tests where only two sensors are used, the analyzed AE events were indeed generated by
damage occurring within the gage section.
The location of AE events is commonly employed as a selection parameter when only two AE sensors are used (one placed
on each side of the gage section). From the times of arrival at top (ttop) and bottom (tbottom) sensors, the location x of an event
along the specimen's longitudinal axis is x ¼ Vðt bottom  t top Þ=2, where V is the wave velocity. Composite materials are highly
dispersive media, i.e. wave velocity varies with frequency. The recorded AE signals correspond mostly to the zero-order
flexural and extensional modes, A0 and S0 respectively. In the frequency range considered in acoustic emission (up to 2 MHz),
the velocity of the high-frequency S0 mode remains constant while that of the lower frequency A0 mode can vary greatly.
Therefore, accurate location of AE events relies on the detection of the high-frequency extensional mode S0, which requires the
use of wide-band sensors [11]. The determination of time of arrival is critical in order to accurately locate AE events. In this
paper, manual determination and the so-called first-threshold-crossing technique are compared to a signal onset determina-
tion method based on the Akaike information criterion (AIC) [12] that was recently developed by Sedlak et al. [13].
In addition, the major contribution of this paper is a new method to validate the selection of AE events. Indeed, location
alone does not ensure that an AE event was generated in the gage section. Two separate noise signals could trigger top and
bottom AE sensors with a difference of arrival times similar to that of a gage event and therefore be considered as such.
The proposed method compares the amount of energy recorded by top and bottom sensors and can assess that both
recorded AE signals were generated by the same gage event. This method was recently developed in order to monitor
damage progression in ceramic matrix composites under static loading by in situ determination of energy attenuation [6].
However, the approach could not be demonstrated since the AE waveforms were not recorded and the analysis only relied
 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227 219

on discrete signal features. Thus, in the present paper, the energy-based approach will first be demonstrated and then
applied for the selection of AE events originating from damage.
The following section presents the waveform-based procedure that includes accurate location of AE events and selection
of gage events based on signal energy. The procedure is then applied to two composite systems. The first composite system,
a carbon fiber reinforced carbon matrix composite (Cf/C–SiC), serves as a model material since only few AE events are
generated by damage and no significant changes in the wave propagation properties of the material are expected due to
damage progression. The second composite system is a state-of-the-art SiC fiber reinforced, melt-infiltrated SiC matrix
composite (SiC/SiC). In this case, thousands of AE events are generated by damage, which are expected to be related to
various damage modes (matrix cracking, interfacial phenomena, fiber failures). Moreover, the matrix experiences significant
damage under tensile loading and therefore substantial changes in the wave propagation properties of the material are
expected.

2. Waveform-based procedure for AE events selection

2.1. Location of AE events

Accurate location relies on accurate determination of AE signal onset. Signal onset picking can be performed manually
when the number of AE events is limited. However, it is usually not applicable in the case of AE monitoring of composites for
which thousands of events are recorded. The most widely used technique for automatic onset detection – usually performed
in real-time on acquisition systems – is the first-threshold-crossing technique. The signal onset is considered as the first
point in the signal exceeding a fixed threshold value. However, the use of a fixed threshold value can cause late triggering,
missing low amplitude extensional waves, or undesired triggering on noise. To overcome the limitations of the first-
threshold-crossing technique, a technique based on the Akaike information criterion (AIC) [12] has been developed. It was
initially aimed at signal onset determination in seismology [14] and more recently applied to AE signals [13]. The AIC-based
technique consists in calculating for each point k of a signal x composed of N points, the similarity between the signal
portion prior and after the considered point.
AICðkÞ ¼ k log10 ½varðx½1: kÞ þ ðN  k  1Þlog10 ½varðx½k þ 1: NÞ ð1Þ
The resulting function, called AIC function, reaches a minimum value at the waveform onset when the similarity is a
minimum between noise prior to onset and highly-correlated signal after the onset. The AIC-based technique showed
significant improvement when compared to the first-threshold-crossing technique [13]. For AE signals recorded during a
tensile test on a thin metal plate, 80% of arrival times were determined with an absolute deviation lower than 0.4 ms
compared to manual picks, against 20% only when using the first-threshold-crossing technique. In terms of location, 72% of
events were located with an absolute deviation lower than 1 mm, compared to 10% only using the first-threshold-crossing
technique. In the present paper, the two-step AIC-based signal onset determination developed by Sedlak et al. [13] is applied
to AE data recorded during tests on composites. The efficiency of the technique is assessed by comparing its results to those
obtained by manual determination or using the so-called first-threshold-crossing technique.
The AIC-based technique developed by Sedlak et al. is briefly summarized here. Further details are available in [13].
To calculate the AIC function, the two-step procedure uses a characteristic function, which is calculated from the AE
waveform in order to enhance amplitude and frequency changes within the signal and thus improve onset determination.
The characteristic function is expressed as follows:





CFðkÞ ¼
xðkÞ
þ R
xðkÞ  xðk  1Þ
ð2Þ

where x is the AE signal and R is a constant determined empirically.

The procedure also involves three empirical time constants: ΔtAM, ΔtFB and ΔtFA. The first step consists in calculating the
AIC function and finding signal onset using a truncated signal that starts at t¼0 and ends ΔtAM after the peak of highest
magnitude. In the second step, the AIC function is computed again, this time truncating the signal between ΔtFB before the
onset found in the first step and ΔtFA after that onset. The signal onset is then refined. For each composite system considered
in this paper, optimum values of the four aforementioned parameters were determined by comparing signal onsets obtained
for various sets of values with those obtained by manual determination. Only optimum results are reported here. Moreover,
comparison with the first-threshold-crossing technique will be developed for the SiC/SiC composite system only, since the
number of AE events recorded in Cf/C–SiC composites was very limited.

2.2. AE events selection based on recorded energies

The following expression has been proposed for the energy recorded by sensor i for an AE event n [6]:
Ei ðnÞ ¼ Es ðnÞAi e  Bdi ðnÞ ð3Þ
Es(n) is the energy released at the source n in the forms of elastic waves. In the case of AE generated by cracking, it was
shown using X-ray tomography combined with AE monitoring that the acoustic energy recorded after very short
propagation distance (therefore corresponding to a good estimate of the released energy) is proportional to the surface
220 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227

Fig. 2. Energies recorded by top ( þ30 mm) and bottom (  30 mm) sensors as a function of event location along the specimen's longitudinal axis for a
source energy ranging from 1 to 10. The dotted lines represent a 25.4 mm gage length.

of cracks [15]. In practice, the recorded energy Ei ranges over several orders of magnitude, which is attributed mostly to the
scatter on released energy Es. The recorded energy Ei(n) decreases exponentially with propagation distance di(n) from the
source to sensor i, B being the attenuation coefficient. The amount of recorded energy is also affected by sensor's frequency
response and coupling to the specimen's surface, which are characterized by Ai, a characteristic of sensor i ranging from 0 to
1 and considered constant for a given test. It is worth noting that, even though the frequency response of a sensor can be
accurately determined, coupling will vary from one test to another. Therefore, Ai is unknown. It is also important to note that
composite materials exhibit frequency-dependent attenuation [16–18]. Therefore, B is expected to vary depending on the
frequency content of the generated signal and so is Ai since it describes the sensor's frequency response. This will be
discussed later in particular when the considered AE signals recorded for a same test exhibit significant differences in
frequency content.
The evolution of recorded energy vs. event location is presented schematically in Fig. 2. The sensors are located at
730 mm from the specimen's center. The attenuation coefficient B is assumed to be 5  10  2 mm  1, which will be later
shown to correspond to an intermediate value for CMCs. Ai is set to 1 for both sensors for easier reading. Finally, the source
energy Es is assumed to range from 1 to 10. Fig. 2 highlights the significant decrease of recorded energy. The propagation
distance from the gage section to the sensor leads to a decrease of more than 50% of the recorded energy. Moreover, within
the gage section, for two sources of same energy located at each end (þ 12.7 mm and  12.7 mm) there is a factor of 4 in
recorded energy. However, because of the wide range of source energy values, it would be nearly impossible, from this plot,
to accurately characterize energy attenuation and validate that an event has originated from the gage section based on its
location and on the energies recorded by both sensors.
Thus, the energy ratio is defined as follows:

RðnÞ ¼ lnðEtop ðnÞ=Ebottom ðnÞÞ ð4Þ

where Etop and Ebottom are the recorded energies at top and bottom sensors, respectively. Using Eq. 3 and replacing di(n) by
dtop(n)¼L x(n) or dbottom(n) ¼Lþx(n) (where L is the distance between sensor and specimen center and x is the location
along the longitudinal axis with x¼0 at the center), the energy ratio can be expressed as follows:

RðnÞ ¼ 2BxðnÞ þ lnðAtop =Abottom Þ ð5Þ

The energy ratio is a linear function of event location x, of slope 2B and ordinate at the origin ln(Atop/Abottom). Using the
example of Fig. 2, the energy ratio can be calculated for various AE events and plotted as a function of event location along
the specimen (Fig. 3). All AE events originating from damage are expected to follow the same linear trend (dictated by the
attenuation coefficient B) regardless of their source energy. The ratio of sensor characteristics Atop/Abottom, constant for one
test, only translates the linear function along the vertical axis.
In practice, for all AE events recorded during one test, one can represent their energy ratio as a function of location.
The overall trend allows identifying the attenuation coefficient B from the slope and the ratio Atop/Abottom from the ordinate
at the origin. Even though Atop and Abottom are unknown, the ratio Atop/Abottom indicates which sensor recorded the most
energy for events located midway and therefore indicates which sensor was the most sensitive. The most sensitive sensor is
the top sensor if ln(Atop/Abottom)40 and the bottom sensor otherwise. Finally, a slight scatter is expected around the linear
trend due to uncertainties in location and energy ratio, which are discussed in the next section. However, if an event is far
from the overall linear trend, it is most likely to have not originated from damage.
 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227 221

Fig. 3. Energy ratio vs. event location along the longitudinal axis for the example considered in Fig. 2 with three different values of Atop/Abottom.

Fig. 4. Characteristic stress vs. strain and cumulative AE energy vs. stress curves for a Cf/C–SiC composite (a, b) and a SiC/SiC composite (c, d).

3. Results and discussion

3.1. Materials and AE data acquisition

The AE data used in this paper were recorded during tensile tests at room temperature on two types of ceramic matrix
composites: carbon fiber reinforced carbon-SiC matrix composites (Cf/C–SiC), with a resin-derived carbon matrix and a
chemically vapor infiltrated SiC layer; and Hi-Nicalon SiC fiber reinforced melt infiltrated SiC matrix composites (SiC/SiC). AE
data corresponding to artificial AE sources generated on Cf/C–SiC specimens were also analyzed in order to validate the
energy-based approach. They are described in the following section. Three AE sensors (ref. B1025, Digital Wave Corporation)
were installed in the configuration presented in Fig. 1 and connected to a Fracture Wave Detector acquisition system (Digital
Wave Corporation, Centennial, CO). Each AE signal consisted of 1024 points digitized at a rate of 10 MHz, giving a recorded
duration of 102.4 ms. The first 25% of each signal, called the pretrigger, corresponded to points recorded before the system
triggered. The three sensors were enslaved to one another, meaning that when any of the sensors was triggered all three
sensors recorded a signal simultaneously, therefore ensuring that the three AE signals corresponded to the same event.
As described earlier, only those events that triggered the middle sensor first were kept. Characteristic stress vs. strain and
cumulative AE energy vs. stress curves are shown in Fig. 4 for both composite systems.
222 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227

Fig. 5. Energy ratio vs. event location along the longitudinal axis for Cf/C–SiC composites. (a)–(b) Artificial AE sources, (a) pencil lead breaks and (b) boron
fiber breaks. (c)–(d) AE events generated by damage during two room temperature tensile tests. Note: location was determined by manual selection of AE
signal onsets.

3.2. Demonstration of the energy-based approach

In order to demonstrate the validity of the energy-based approach, artificial AE sources were generated on the surface of
Cf/C–SiC tensile coupons. Two AE sensors were placed 60 mm away in a configuration similar to the top and bottom sensors
used for tensile testing (Fig. 1). The artificial sources were generated at various locations along the specimen's length either
by breaking 0.5 mm diameter pencil leads (Hsu-Nielsen sources) or 0.2 mm diameter boron fibers (Specialty Materials, Inc.,
Lowell, MA). For each artificial AE event, the location was calculated based on manually-determined signal onsets in order to
minimize location uncertainties and therefore focus on the validation of the energy-based approach. The energy ratio was
calculated and reported on the energy ratio vs. location plot. Results are shown in Fig. 5 for both pencil lead breaks and
boron fiber breaks. In both cases, experimental results are in excellent agreement with the expected linear trend, which
demonstrates the validity of the proposed model for recorded AE energy. The attenuation coefficient, identified from the
slope of the linear fit, is 2.5  10  2 mm  1 for pencil leads and 3  10  2 mm  1 for boron fibers. The limited scatter is mainly
attributed to location uncertainties and variations in source frequency. Although location was calculated based on manually-
determined signal onsets, uncertainties remain due to location resolution. Signals were sampled at a rate of 10 MHz,
therefore the time uncertainty at onset picking was 0.1 ms. Considering a wave velocity of 10,000 m/s, which is common in
CMCs, the time uncertainty leads to a location resolution of 71 mm. Scatter in energy ratio can be caused between two AE
events by differences in frequency content leading to differences in attenuation coefficient. The frequency content of
generated AE sources was described by the frequency centroid (weighted frequency average) of signals recorded after
limited propagation throughout the material (0–20 mm). The frequency centroid ranged from 200 to 350 kHz for pencil lead
breaks and from 700 to 800 kHz for boron fiber breaks. It appears that for similar amplitude of frequency centroid range,
the scatter in energy ratio is similar. Moreover, as expected, lower source frequency leads to lower attenuation.

3.3. Selection of AE events generated by damage in Cf/C–SiC composites

All the specimens tested exhibited similar AE behavior. About 30 gage events (that triggered the middle sensor first)
were recorded and the AE energy release was significant beyond 100 MPa (Fig. 4b). Since the non-linearity on the stress–
strain curve was very limited (Fig. 4a), these events were associated to fiber failures. The energy ratio vs. location plot was
generated for each test and exhibited very consistent results. Note that for each AE event, the location was calculated based
on manually-determined signal onsets in order to minimize location uncertainty. Results are shown for two of the tests in
 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227 223

Fig. 6. Energy ratio vs. event location (determined automatically using AIC) for all AE events recorded during room temperature tensile test of a Cf/C–SiC
composite.

Fig. 5c and d. Similar to artificial AE events, damage-related AE events were in very good agreement with the expected linear
trend. The attenuation coefficient, identified from the slope of the linear fit, was 7  10  2 71  10  2 mm  1 (mean7
standard deviation). The greater attenuation coefficient value and scatter observed on the energy ratio vs. location plot
when compared to results obtained with pencil lead or boron fiber breaks are both attributed to differences in source
frequency content. For the damage-related events, the frequency centroid of signals recorded after limited propagation
throughout the material (0–20 mm) ranged from 800 to 1200 kHz. As would be expected, higher frequency waves
experience higher attenuation. Moreover, the greater scatter in energy ratio is attributed to the greater range in source
frequency (400 kHz for damage events against 100–150 kHz for artificial events).
Finally, to illustrate the use of the energy ratio vs. location plot as a visual tool to assess the selection of AE events
originating from damage, all the AE events recorded during test were considered and the energy ratio vs. location plot was
generated automatically using the AIC technique for location determination (Fig. 6). Optimum location accuracy was
achieved with the following parameters: R¼0, ΔtAM ¼20 ms, ΔtFB ¼30 ms and ΔtFA ¼30 ms. Note that only the AE signals
recorded by the top and bottom sensors are needed to generate this plot. The true gage events are highlighted here based on
the triggering of the middle sensor. Only those events clearly follow the expected linear trend of energy ratio vs. location.
Therefore, they could accurately be selected based on this plot in tests where the use of a middle sensor is not possible.
Events for which the automatic location determination would attribute a location within the gage section even though they
did not trigger the middle sensor first can easily be filtered out based on their energy ratio value (these events are outside
the selection box in Fig. 6).

3.4. Selection of AE events generated by damage in SiC/SiC composites

The tested specimens exhibited comparable AE behavior. A total of about 4000 AE events were recorded throughout each
tensile test, and around 25% of the total were gage events (triggered the middle sensor first). The three AE signals of each
gage event were manually inspected for validation and determination of arrival times on top and bottom sensors. This
section will focus on the results obtained for two tests with different sensor/material coupling conditions. In Test 1, top and
bottom sensors are assumed to have similar coupling conditions with the specimen's surface. Coupling quality is
characterized by the average energy recorded by each sensor for all the gage events. In Test 1, the average recorded energy
was 1.6 V2.ms for the top sensor and 2.9 V2.ms for the bottom sensor. In Test 2, top and bottom sensors are assumed to have
very different coupling conditions, and the bottom sensor in particular seemed to have degraded coupling quality. Indeed,
the average recorded energy in Test 2 was 5 V2.ms for the top sensor and 0.7 V2.ms for the bottom sensor. It is worth noting
that variations in coupling quality are common in acoustic emission mainly because of variations in specimen surface
conditions and amount of coupling agent used. Moreover, it is difficult in practice to accurately characterize coupling quality
before test therefore leading to the wide range of recorded energies observed here.
One characteristic feature of SiC/SiC composites is a strong damage dependence of wave velocity [19] that needsptoffiffiffiffiffiffiffiffi
be
accounted for in order to ensure accurate location of AE events throughout tests. Wave velocity is proportional to E=ρ,
where E is the material's elastic modulus and ρ its density. Transverse matrix cracking causes a decrease in elastic modulus
and thus a decrease in wave velocity. Wave velocity can be monitored throughout test using the recorded AE signals that
triggered the top or bottom sensor first and exhibited a clear separation between extensional and flexural modes at the
furthest sensor thus indicating that the wave had propagated over a long distance and that the event originated from
outside the sensors. Wave velocity was calculated from the known distance between top and bottom sensors and the
manually-determined difference in arrival times. Results are presented in Fig. 7 for Test 2. Wave velocity decreases
significantly during test, from over 10,000 m/s initially to about 6000 m/s near failure (Fig. 7a). The comparison with elastic
modulus (taken as the tangent modulus at beginning of unload for each unload/reload cycle) in Fig. 7b confirms the direct
224 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227

Fig. 7. Damage dependence of wave velocity in a SiC/SiC composite. (a) Wave velocity determined from AE events that occurred outside the outermost
sensors vs. applied stress. (b) Comparison with elastic modulus (tangent modulus at beginning of unload) extracted from unload/reload cycles.

Fig. 8. Percentage of events with an absolute deviation from manually-determined signal onset lower than that indicated on the x-axis (in terms of time in
ms or location in mm) for all the gage events recorded during tensile test of two SiC/SiC composite specimens. (a) Test 1. (b) Test 2.

relationship between wave velocity and transverse matrix cracking. For the determination of location of gage events, wave
velocity was determined from Fig. 7 based on the stress at which the event was recorded.
The performance of both signal onset determination techniques was first evaluated in terms of location accuracy.
For both tests, manually-determined arrival times at top and bottom sensors for all gage events served as reference. Event
location accuracy relies on the accuracy in difference of arrival times between top and bottom sensors rather than accuracy in
arrival time itself. Therefore, the absolute deviation in difference of arrival times was calculated for each AE event between manual
determination and first-threshold or AIC techniques. The absolute deviation in location was estimated from the deviation in
difference of arrival times assuming a constant wave velocity of 10,000 m/s for simplicity. The parameters leading to optimum
results were 20 mV for the first-threshold technique and R¼1, ΔtAM ¼20 ms, ΔtFB ¼30 ms and ΔtFA ¼ 10 ms for the AIC technique.
Fig. 8 shows that the AIC technique slightly outperforms the first-threshold-crossing technique leading to 5% more events located
with a deviation of 74 mm from manual determination. Moreover, coupling quality seems to have a significant effect on location
accuracy for both techniques. In Test 1, 85% of gage events were located by AIC with a 74 mm deviation from manual
determination whereas in Test 2, only 70% of gage events were located with the same accuracy.
The performance of both signal onset determination techniques was also evaluated in terms of accuracy in gage events
selection. As mentioned earlier, when only the top and bottom sensors are used, gage events can be selected based on their
location along the longitudinal axis (the gage section corresponding to 712.7 mm from the center). The selected events were
compared to those triggering the middle sensor first (Table 1). True gage events correspond to events with location within
12.7 mm from center that indeed triggered the middle sensor first whereas false gage events have a location wrongfully
determined within 12.7 mm from center but did not trigger the middle sensor first. In Test 1, location-based selection gives very
accurate results, more than 90% of the actual gage events being selected. The AIC outperforms the first-threshold technique
leading to the selection of 3% more true gage events and 3% less false gage events. In Test 2, the difference between first-
threshold and AIC becomes more significant. Decrease in coupling quality leads to the selection of a lower fraction of true gage
events. With AIC, 75% of the gage events are selected against 67% with first-threshold. The main difference appears on the false
gage events, which are wrongfully located within the gage section. While the AIC maintains a low number of false gage events
(6% of the total number of gage events), the first-threshold technique leads to the selection of 25% of false gage events.
To conclude the comparison between first-threshold and AIC-based techniques, performance evaluation based on
location accuracy shows similar results because it uses gage events for which the waveforms are well defined and signal
 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227 225

Table 1
Selection of gage events based on location for two tensile tests on SiC/SiC composites. True gage events correspond to events located within the gage
section that indeed triggered the middle sensor first (number and percentage of total number of gage events). False gage events correspond to events
located within the gage section that did not trigger the middle sensor first (number and percentage of total number of gage events).

Manual AIC First threshold

Gage events True gage events False gage events True gage events False gage events

Test 1 1050 981 (93%) 37 (3%) 949 (90%) 63 (6%)

Test 2 1270 950 (75%) 79 (6%) 848 (67%) 317 (25%)

Fig. 9. Energy ratio vs. event location along the longitudinal axis for a SiC/SiC composite. (a) All test. (b) First 250 events only and energy ratio calculated
for highest frequency components. (c) Last 250 events only and energy ratio calculated for highest frequency components. Note: location was determined
by manual selection of AE signal onsets.

onset is clear and therefore it can be determined rather accurately even with the first-threshold technique. However, in
terms of performance in gage events selection, AIC clearly outperforms the first-threshold technique especially when sensor
coupling quality is not optimum, reducing the loss of information as well as the addition in the analysis of events originating
from outside the gage section.
Finally, the contribution of the energy-based approach is discussed for Test 1. First, the energy ratio vs. location plot was
generated with manually-determined locations. When all gage events are considered (Fig. 9a), a clear linear trend can be
observed. However, the scatter is significantly greater than that observed on Cf/C–SiC composites. The reasons for this are
twofold. First, the frequency centroid of signals recorded after limited propagation throughout the material (0–20 mm)
ranged from 600 to 1200 kHz, which is almost twice the frequency range observed on Cf/C–SiC composites, most likely due
to the generation of AE events from various damage modes. Second, the matrix experiences significant damage during
tensile test in the form of transverse matrix cracks, which are expected to increase energy attenuation due to increased
wave scattering. In order to reduce the observed scatter due to scatter in source frequency, the energy ratio was calculated
based on the energy carried only by the highest frequency components recorded by the sensors (from 1500 kHz to
2000 kHz). Moreover, to account for increase in attenuation due to damage, only 250 consecutive events were considered at
a time (Fig. 9b and c). The scatter in energy ratio is greatly reduced and as expected the attenuation coefficient increases
significantly throughout test (from 2.5  10  2 mm  1 for the first 250 events to 3.2  10  2 mm  1 for the last 250 events).
At last, the energy ratio vs. location plot was generated automatically using the AIC technique for location determination
(Fig. 10). As already mentioned, the AIC technique provides a very accurate selection of gage events. In Test 1, only about 40
226 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227

Fig. 10. Energy ratio (highest frequency components) vs. event location (determined automatically using AIC) for all AE events recorded during room
temperature tensile test of a SiC/SiC composite.

events are wrongfully located within the gage section. An automated selection of gage events could be set using for instance
a confidence interval around the linear regression. Only those events within the confidence interval would be considered as
gage events. The events outside of the admissible interval could either be discarded or manually inspected. In the present
case, manual inspection is acceptable since only about 20 events appear clearly away from the linear trend within the gage
section. These events may correspond to noise signals or damage related events for which automatic onset determination
failed to find the actual signal onset. Their selection or location could then be manually adjusted.

4. Future applications of the energy-based approach

The energy ratio vs. location plot offers a visual tool for assessing the selection of AE events generated by damage. Using
only the events recorded during test, it also gives a measure of energy attenuation. In composites where damage
progression is expected to affect wave propagation properties, such as the SiC/SiC composites presented in this paper, the
energy-based approach could now be used to monitor damage throughout test.
Also, by determining the energy attenuation for different frequency ranges, one could characterize the frequency-
dependent attenuation in a given composite system and use this information to evaluate the frequency content of an AE
event at its source based on the frequency content of the signal recorded at a known distance from the source.

5. Conclusion

The energy-based approach presented in this paper allows validating that AE events originated from damage therefore
providing robust data for subsequent analysis such as damage mode identification. The approach only requires two remote
sensors and is thus applicable when specimen geometry or testing conditions prevent from placing a sensor in the middle of
the gage section. It was demonstrated using artificial sources and then used on two different composite systems. The
energy-based approach also offers a potential for damage monitoring and in-situ determination of wave propagation
properties that could be used to improve the description of AE sources.
In addition, the AIC-based AE event location determination was shown to outperform the so-called first-threshold-
crossing technique, being less dependent on the quality of material/sensor coupling. Up to 85% of events were located with a
deviation from manual determination not exceeding 4 mm. Moreover, using the AIC-based technique to select events
originating from the gage section ensured accurate selection, maximizing the number of selected true gage events and
limiting the addition of false gage events to the analysis.

Acknowledgments

The authors gratefully acknowledge Dr. Mrityunjay Singh from the Ohio Aerospace Institute for providing the Cf/C–SiC
specimens, Rolls-Royce Corporation (Indianapolis, IN) for providing the SiC/SiC specimens and Christopher Baker for helping
with mechanical testing.

References

[1] K. Wood, Ceramic–matrix composites heat up, High-perform. Compos. (2013) 〈http://www.compositesworld.com/articles/ceramic-matrix-composites-
heat-up〉.
[2] G.N. Morscher, Modal acoustic emission of damage accumulation in a woven SiC/SiC composite, Compos. Sci. Technol. 59 (1999) 687–697.
[3] G.N. Morscher, Stress-dependent matrix cracking in 2D woven SiC-fiber reinforced melt-infiltrated SiC matrix composites, Compos. Sci. Technol. 64
(2004) 1311–1319.
 E. Maillet, G.N. Morscher / Mechanical Systems and Signal Processing 52-53 (2015) 217–227 227

[4] C. Baker, G.N. Morscher, V.V. Pujar, J. Lemanski, Transverse cracking in carbon fiber reinforced polymer composites: modal acoustic emission and peak
frequency analysis, Compos. Sci. Technol. (2014). submitted for publication.
[5] E. Maillet, N. Godin, M. R'Mili, P. Reynaud, J. Lamon, G. Fantozzi, Analysis of acoustic emission energy release during static fatigue tests at intermediate
temperatures on ceramic matrix composites: towards rupture time prediction, Compos. Sci. Technol. 72 (2012) 1001–1007.
[6] E. Maillet, N. Godin, M. R'Mili, P. Reynaud, G. Fantozzi, J. Lamon, Real-time evaluation of energy attenuation: a novel approach to acoustic emission
analysis for damage monitoring of ceramic matrix composites, J. Eur. Ceram. Soc. 34 (2014) 1673–1679.
[7] R. Gutkin, C.J. Green, S. Vangrattanachai, S.T. Pinho, P. Robinson, P.T. Curtis, On acoustic emission for failure investigation in CFRP: Pattern recognition
and peak frequency analyses, Mech. Syst. Signal Process. 25 (2011) 1393–1407.
[8] M.G.R. Sause, A. Gribov, A.R. Unwin, S. Horn, Pattern recognition approach to identify natural clusters of acoustic emission signals, Pattern Recogn. Lett.
33 (2012) 17–23.
[9] E. Maillet, N. Godin, M. R'Mili, P. Reynaud, G. Fantozzi, J. Lamon, Damage monitoring and identification in SiC/SiC minicomposites using combined
acousto-ultrasonics and acoustic emission, Compos. Part A 57 (2014) 8–15.
[10] G.N. Morscher, Modal acoustic emission source determination in silicon carbide matrix composites, in: D.O. Thompson, D.E. Chimenti (Eds.), Review of
Progress in Quantitative Nondestructive Evaluation, vol. 19, American Institute of Physics, NY, 2000, pp. 383–390.
[11] M.R. Gorman, Plate wave acoustic emission in composites, in: A.K. Mal, Y. Rajapakse (Eds.), Impact Response and Elastodynamics of Composites, ASME,
New York, NY, 1990, pp. 107–134.
[12] H. Akaike, Markovian representation of stochastic processes and its application to the analysis of autoregressive moving average processes, Ann. Inst.
Stat. Math. 26 (1974) 363–387.
[13] P. Sedlak, Y. Hirose, M. Enoki, Acoustic emission localization in thin multi-layer plates using first-arrival determination, Mech. Syst. Signal Process. 36
(2013) 636–649.
[14] R. Sleeman, T. van Eck, Robust automatic P-phase picking: an on-line implementation in the analysis of broadband seismogram recordings, Phys. Earth
Planet. Inter. 113 (1999) 265–275.
[15] E. Maire, V. Carmona, J. Courbon, W. Ludwig, Fast X-ray tomography and acoustic emission study of damage in metals during continuous tensile tests,
Acta Mater. 55 (2007).
[16] T.D. Lhermitte, S.M. Handley, M.R. Holland, J.G. Miller, Anisotropy of the frequency-dependent ultrasonic attenuation in unidirectional graphite/epoxy
composite material, Ultrasonics Symp. (1991) 819–823.
[17] W. Huang, S.I. Rokhlin, Y.J. Wang, Effect of fibre–matrix interphase on wave propagation along, and scattering from, multilayered fibers in composites.
Transfer matrix approach, Ultrasonics 33 (1995) 365–375.
[18] S. Biwa, Y. Watanabe, N. Ohno, Analysis of wave attenuation in unidirectional viscoelastic composites by a differential scheme, Compos. Sci. Technol. 63
(2003) 237–247.
[19] G.N. Morscher, A.L. Gyekenyesi, The velocity and attenuation of acoustic emission waves in SiC/SiC composites loaded in tension, Compos. Sci. Technol.
62 (2002) 1171–1180.