ArticlePDF Available

Development of Acoustic Emission Technology for Condition Monitoring and Diagnosis of Rotating Machines: Bearings, Pumps, Gearboxes, Engines, and Rotating Structures



One of the earliest documented applications of acoustic emission technology (AET) to rotating machinery monitoring was in the late 1960s. Since then, there has been an explosion in research- and application-based studies covering bearings, pumps, gearboxes, engines, and rotating structures. In this paper we present a comprehensive and critical review to date on the application of AET to condition monitoring and diagnostics of rotating machinery.
Development of Acoustic Emission Technology for
Condition Monitoring and Diagnosis of Rotating
Machines: Bearings, Pumps, Gearboxes, Engines,
and Rotating Structures
D. Mba and Raj B. K. N. Rao
ABSTRACT—One of the earliest documented applications
of acoustic emission technology (AET) to rotating machinery
monitoring was in the late 1960s. Since then, there has been
an explosion in research- and application-based studies cov-
ering bearings, pumps, gearboxes, engines, and rotating struc-
tures. In this paper we present a comprehensive and critical
review to date on the application of AET to condition moni-
toring and diagnostics of rotating machinery.
KEYWORDS: acoustic emission, condition monitoring,
machine diagnosis, rotating machines
1. Introduction
Acoustic emissions (AEs) are defined as transient elastic
waves generated from a rapid release of strain energy caused
by a deformation or damage within or on the surface of a
material (Pao et al., 1979; Mathews, 1983; Pollock, 1989).
In the application to rotating machinery monitoring, AEs are
defined as transient elastic waves generated by the interac-
tion of two media in relative motion. Sources of AE in rotat-
ing machinery include impacting, cyclic fatigue, friction,
turbulence, material loss, cavitation, leakage, etc. For instance,
the interaction of surface asperities and impingement of the
bearing rollers over a defect on an outer race will result in
the generation of acoustic emission. These emissions propa-
gate on the surface of the material as Rayleigh waves and the
displacement of these waves is measured with an AE sensor.
Rayleigh waves are a combination of longitudinal and trans-
verse waves (Viktorov, 1967). It should be noted that sur-
face defects such as cracks and scratches attenuate Rayleigh
waves; in addition, the surface finish of metals can also
influence attenuation (Viktorov, 1967).
Judicious application of well-tried and tested acoustic
emission technology (AET) can provide powerful diagnos-
tic capabilities, which are safe, efficient and cost-effective. In
this paper we review the research and development activities
that are being pursued in the following subject areas: bear-
ings (roller and hydrodynamic), gearboxes, pumps, machin-
ery, and mechanical seals.
AE was originally developed for non-destructive testing
of static structures; however, over the last 35 years its
application has been extended to health monitoring of rotat-
ing machines, including bearings, gearboxes, pumps, etc. It
offers the advantage of earlier defect/failure detection in
comparison to vibration analysis due to the increased sensi-
tivity offered by AE. However, limitations in the successful
application of the AE technique for monitoring the perform-
ance of a wide range of rotating machinery have been partly
due to the difficulty in processing, interpreting, and classify-
ing the intelligent information from the acquired data. The
main drawback with the application of the AE technique is
the attenuation of the signal, and as such the AE sensor
has to be close to its source. However, it is often practical
to place the AE sensor on the non-rotating member of the
machine, such as the bearing or gear casing. Therefore, the
AE signal originating from the defective component will
suffer severe attenuation, and reflections, before reaching
the sensor.
AE covers a wide frequency range (100 kHz to 1MHz),
and time domain waveforms associated with AE are of two
types: burst and continuous. A continuous-type AE refers
to a waveform where transient bursts are not discernible
(Miller and McIntire, 1987). Both waveform types are asso-
ciated with rotating machinery; for instance, a continuous-
type emission may be a result of turbulent fluid flow within
a peep while a burst type could be associated with the tran-
sient rolling action of meshing bears. On rotating machinery,
typical background operational noise is of a continuous type.
Traditionally, the most commonly measured AE parameters
for diagnosis are amplitude, rms, energy, kurtosis, crest fac-
tor, counts and events (Mathews, 1983). Observations of the
frequency spectrum, whilst informative for traditional non-
destructive evaluation, have only found limited success in
machinery monitoring. This is primarily due to the broad
frequencies associated with the sources of generation of AE
in rotating machinery. For example, the transient impulse
associated with the breakage of contacting surface asperities
D. Mba (, School of Engineering, Cranfield
University, UK. Raj B. K. N. Rao, COMADEM International, Birmingham,
The Shock and Vibration Digest, Vol. 37, No. 5, September 2005 000–000
©2005 Sage Publications
DOI: 10.1177/0583102405059054
2The Shock and Vibration Digest / September 2005
experiencing relative motion will excite a broad frequency
2. Acoustic Emission and Bearing Defect
From the moment bearings leave the factory, they encoun-
ter many harsh environmental hazards, which in turn induce
a number of failure modes. It is well known how these fail-
ure modes reduce the life expectancy of the bearings. Some
of the events responsible for the bearing failures include
incorrect applications, poor maintenance, poor lubrication,
overload, over-speed, misalignment, imbalance, harsh environ-
mental conditions (temperature/humidity/dust/dirt/altitude),
etc. Bearing failure modes include friction/wear processes
producing flaking, brinelling, fluting, spalling, pitting, sei-
zure, etc. All these modes are known sources of AE. How-
ever, the most widely employed technique for condition
monitoring and diagnostics of bearings is vibration monitor-
ing. This method has been successful where the energy from
other components (shaft, gears, etc.) does not overwhelm the
lower energy content from the defect bearing. In addition, by
the time a significant change in vibration has been observed,
the remaining operational or useful life of the bearing is very
short. This is where AET offers a significant advantage. The
formation of subsurface cracks due to the Hertzian contact
stress induced by the rolling action of the bearing elements
in contact with the inner and outer races and the rubbing
between damaged mating surfaces within the bearing will
generate acoustic emission activity. Other reasons for the
generation of AE include the breakdown of the oil film, for-
eign matter in the lubricating medium, and excessive tem-
peratures. It must be noted that the propagation of the AE is
affected by material microstructure, non-homogeneities,
geometrical arrangement of free surfaces, loading condi-
tions, and the number of component interfaces. Almost all
research on the application of AE to bearing defect analysis
has been undertaken on experimental test-rigs specifically
designed to reduce AE background noise.
Catlin (1983) reported that AE activity from bearing defects
was attributed to four main factors including numerous tran-
sient and random AE signals associated with bearing defects.
Furthermore, it was stated that the signals detected in the AE
frequency range represented bearing defects rather than
other defects such as imbalance, misalignment, looseness,
shaft bending as well as the other major structural component
resonances. In addition, Catlin noted that high-frequency AE
signatures attenuate rapidly; therefore, if the transducer was
placed close to the bearing, it was possible to detect the high-
frequency content induced mainly by the bearing fault since
signatures originating from other machine components are
highly attenuated upon reaching the sensor. Balerston (1969)
published the first document that applied AET to the iden-
tification of artificially seeded defects in rolling element
bearings. Interestingly, this is probably one of the earliest
applications of AE to monitoring bearings. Defects simu-
lated included outer and inner race defects, ball defects, and
lack of lubrication. Balerston compared vibrations in the audi-
ble range, resonant range and AE, commenting on the advan-
tages that monitoring of the resonant frequency range offered
over the audible vibration range. The resonant technique
involved measurement of bearing component natural fre-
quencies initiated by shock excitation associated with minor
structural irregularities. These resonant frequencies are a func-
tion of the mass configuration and type of material involved.
The frequencies and amplitudes at resonance are much higher
than bearing element rotations, so they are ideal under con-
ditions of high background noise. Moreover, the resonant
frequencies are independent of rotational speed; however,
their amplitudes will vary directly with rotational speed, as
will the impact energy. Resonant frequencies can be as high
as 300 kHz for ball rollers, and up to 140 kHz for the inner
race, depending on the mode of vibration (Rogers, 1979). Bal-
erston suggested that the “free” resonant frequencies of the
individual components were not changed significantly after
assembly, although the assembly created a damping effect.
Furthermore, it was suggested that because of the interaction
between the components of a bearing, a defect in any com-
ponent would cause resonant frequency ringing in all com-
ponents, making interpretation difficult. Moreover, at low
rotational speeds the impact energy generated will be very
low, and this might explain why there have been limited appli-
cations of this technique to low-speed bearings. The princi-
ple of the shock pulse meter (SPM) is similar to the resonant
technique as both respond to minute transient pressure waves
generated from fault impacts in regions of contact; however,
the SPM resonates itself.
Balerston noted that two types of AE signatures were
observed during experimental testing: burst-type emissions,
associated with the seeded defects on the inner, outer race
and ball element, and continuous-type AE signatures, noted
when the bearing was run dry (starved of lubrication). In one
particular bearing defect simulations (dry run) AE counts
were noted to increase prior to bearing failure. In summary,
Balerston stated that the resonant frequency technique was
very successful and it offered a direct correlation between
defect severity and increase in amplitude level of the reso-
nant frequencies, although it was concluded that the AE
technique would become important with the development of
sensors. This was the earliest assessment on the application
of AE to bearing monitoring.
About 10 years after Balerston, Rogers (1979) utilized
the AE technique for monitoring slow rotating antifriction
slew bearings on cranes employed for gas production, and
obtained some encouraging results compared to vibration
monitoring techniques. Rubbing of the crack faces, grinding
of the metal fragments in the bearing, and impacts between
the rolling elements and the damaged parts in the loaded
zone were identified as sources of detectable AE signatures.
Roger stated that “because of the slow rotational speed of
the crane, application of conventional vibration analysis (0–
20 kHz) was of limited value for on-line condition monitor-
ing.” AE resonant transducers between 100 and 300 kHz
were found to be informative for on-line monitoring of
bearings using kurtosis at different frequency bands.
Yoshioka and Fujiwara (?, 1984) have shown that AE
parameters identified bearing defects before they appeared
in the vibration acceleration range. In addition, sources of
AE generation were identified during fatigue life tests on
thrust loaded ball bearings. Hawman and Galinaitis (1988)
reinforced Yoshioka’s observation that AE provided earlier
detection of bearing faults than vibration analysis and noted
that diagnosis of defect bearings was accomplished due to
modulation of high-frequency AE bursts at the outer race
defect frequency. Hawman and Galinaitis placed the AE
receiving sensor directly onto the bearing outer race. The
modulation of AE signatures at bearing defect frequencies
has also been observed by other researchers (Holroyd and
Randall, 1993a; Holroyd, 2001). In addition, Bagnoli et al.
(1988) investigated the demodulation of AE signatures at
the defect rotational frequency (outer race) of a bearing. It
was noted that when the defect was absent, the periodicity
of the passage of the balls beneath the load could be readily
identified by observing the frequency spectrum of demodu-
lated AE signatures; however, it was reported that the AE
intensity was less without the defect present. There was no
mention of trigger levels employed, load applied on the test
bearing, method of attaching the transducers to the rig, or
any information on background noise.
Tandon and Nakra (1990) investigated AE counts and
peak amplitudes for an outer race defect using a resonant-
type transducer. It was concluded that AE counts increased
with increasing load and rotational speed. However, it was
observed that AE counts could only be used for defect detec-
tion when the defect was less than 250 µm in diameter,
although AE peak amplitude provided an indication of defects
irrespective of the defect size. Loads applied ranged from
8% to 50% of the bearing static load rating. Choudhary and
Tandon (2000) employed AE for bearing defect identifica-
tion on various sized bearings and rotational speeds ranging
from 500 to 1500 rpm. It was observed that AE counts were
low for undamaged bearings. In addition, it was observed
that AE counts increased with increasing speed for damaged
and undamaged bearings whilst an increase in load did not
result in any significant changes in AE counts for both dam-
aged and undamaged bearings.
Tan (1990) used a variation of the standard AE count
parameter for diagnosis of different sized ball bearings. In
addition to the difficulty of selecting the most appropriate
threshold level for standard AE counts, Tan cited a couple of
other drawbacks with the conventional AE count technique.
This included dependence of the count value on the signal
frequency. Secondly, it was commented that the count rate
was indirectly dependent upon the amplitude of the AE
pulses. Tan’s variation to the standard AE counts technique
involved computing the accumulated area under the ampli-
tude–time curve of the AE waveform over a specified time
period. This was accomplished by setting four trigger levels
with amplitude multiples of 1, 2, 4, and 8, and calculating
the area under the amplitude–time AE waveform. The final
count assigned was weighted by the multiple of the ampli-
tude ratio between these levels. It was concluded that the
“new” count rates increased exponentially with increasing
defect sizes and increasing rotational speed. The dependence
of AE counts on threshold levels was also noted by Huguet
et al. (2002) during investigations on the use of AE for
identifying damage modes in specific materials; in this
instance, a trigger level of 10% of the maximum amplitude
was employed.
Yoshioka et al. (1999) undertook an investigation of
vibration and AE on naturally fatigued deep groove ball
bearings (bore diameter 20 mm). By removing the groove on
the inner race, Yoshioka claimed the stresses in the area of
contact were increased and this accelerated fatigue failure.
Vibration rms levels were recorded continuously through the
fatigue tests which lasted approximately 130 h. The presence
of spalls on the inner race resulted in a rapid increase in
vibration rms levels. However, AE (counts per minute) showed
a steadily increasing value at least 5 h before the observed
rapid increase in vibration. A total of 16 fatigue tests were
undertaken and the authors commented that they could pre-
dict the appearance of a spall by observing the AE response.
Whilst AE counts may highlight changes in machine state,
they will not be able to identify the origins of defect, e.g.
outer race. The successful use of AE counts for bearing diag-
nosis is dependent on the particular investigation, and the
method of determining the trigger level is at the discretion of
the investigator. Moreover, it has been shown that AE counts
are sensitive to the level and grade of lubricant within the
bearing, adding to the complexity of this measure. Morhain
and Mba (2003) undertook an investigation to ascertain the
most appropriate threshold level for AE count diagnosis in
rolling element bearings. The results showed that values of
AE maximum amplitude did correlate with increasing speed
but not with load and defect size. In addition, it has been
shown that the relationship between bearing mechanical
integrity and AE counts is independent of the chosen thresh-
old level, although a threshold of at least 30% of the maxi-
mum amplitude for the lowest speed and load operating
condition was advised. Furthermore, Morhain and Mba com-
mented that unlike the results reported by Tandon and Nakra
(1990) it was observed that AE counts could be used for
defect size detection for lengths of up to 15 mm and widths
of 1 mm. In addition, Morhain validated the observations of
Choudhary and Tandon (2000).
Kakishima et al. (2000) undertook a comparative experi-
mental study on the assessment of AE and vibration for
monitoring/detecting seeded defect simulations on the inner
race of a roller and ball bearing. Defects were seeded with an
electron discharge machine (EDM). Analysis of the AE was
based on the spectrum of the enveloped AE signals. It was
concluded that the threshold at which the AE technique was
able to identify the defect was similar to that for vibration
monitoring. Furthermore, for both AE and vibration, it was
noted that an increase in defect size resulted in an increase of
both AE and vibration levels on the envelope spectrum.
Kaewkongka and Au (2001) applied the AE technique on
a rotor dynamic system onto which multiple defects were
seeded, including a seeded defect on one of the bearings. It
was shown the AE technique offered high sensitivity, thereby
allowing for discrimination of the multiple defect condi-
tions. Success was based on minimum distance classifier.
Schoess (2000) presented the results of an assessment of six
different but relevant technologies for onboard monitoring
of a railcar bearing. It was concluded that the AE technique
offered the highest potential payoff. Schoess successfully
evaluated the AE technique on an artificially damaged bear-
ing on a railcar, concluding that the AE technique offered the
potential for condition-based maintenance in the railroad
industry. Price et al. (2001) assessed the vibration and AE
techniques for monitoring rolling element bearing failures.
Their experimental study focused on a four-ball machine
from which AE activity, vibration, temperature, friction,
etc., were monitored as a function of time. It was noted that
AE could detect distress within the test balls before the fric-
tion in the contact area increased noticeably. It was stated
that increasing damaging results in increasing friction at the
contact area.
4The Shock and Vibration Digest / September 2005
Shiroishi et al. (1997) compared vibration and AE on seeded
defective bearings operating at 1200 rpm. Interestingly, Shi-
roishi et al. defined the industry bearing failure criteria as
being reached when a defect size reached 6.45 mm2; this
value was cited from Hoeprich (1992). Defects of varying
sizes were seeded on the outer and inner races. Shiroishi et
al. noted that the vibration offered better detection than the
AE technique, and that the AE sensor was insensitive to
inner race defects. In addition, on the parameters extracted
from vibration and AE measurements, Shiroishi et al. (1997)
noted that the peak ratio was the most reliable indicator of
the presence of a localized defect with the rms, kurtosis and
crest factor showing decreasing reliability. The most signif-
icant observation from the investigation of Shiroishi et al.
was the correlation between acceleration peak value and defect
width. This correlation was first noted by Balerston (1969)
employing a monitoring system based on observations of
bearing resonant frequencies. The most recent correlation
between defect size and measuring parameter (AE) was noted
by Al-Ghamdi et al. (2004) and Al-Ghamdi and Mba (2005).
A direct correlation between defect length (circumferential,
along direction of rolling) and AE burst duration was
observed under varying simulated defect cases. In addition, a
correlation between the amplitude of the burst-type AE sig-
nature (associated with the bearing defect) to the underlying
continuous-type emission was noted to increase with increas-
ing defect width (perpendicular to rolling direction).
Li et al. (1998) undertook bearing fatigue failure tests at
1600 rpm and 167% of the rated radial load. To accelerate
failure, an initial defect was seeded with an electric dis-
charge machine. Li et al. commented that vibration and AE
rms increased with increasing defect severity. An adaptive
scheme was proposed to predict conditions of defective bear-
ings based on vibration and AE techniques. Bansal et al.
(1990) applied AE as a quality control tool on reconditioned
bearings. Bearings were tested at 3% of the load rating. It
was noted that as the load increased there was little increase
in the peak-to-peak amplitude level for standard (operational)
and reconditioned bearings; however, the peak values of the
reconditioned bearing were in some instances five times
that of a new bearing.
Li and Li (1985) presented a pattern recognition technique
for early detection of bearing faults using AE. Faults were
seeded on an outer race, a roller and multiple outer race
defects. It was noted that the occurrence of AE events at a
rate equivalent to a bearing characteristic defect frequency
was evidence of the presence of a localized defect. Li et al.
presented such a case with the seeded outer race defect but
no results on the roller defect were presented. This was
rather disappointing as Li and Li are the only investigators to
attempt to diagnose roller defects with AE.
Sundt (1979) detailed two cases where high-frequency
AE was applied to bearing defect detection. For the first case
study, high-frequency signals associated with a hairline
crack in the outer race the defect frequency were detectable
above 100 kHz. This defect condition was not observed with
vibration analysis. It was stated that the defect was at an
early stage of development and the bearing clearances had
not deviated from the normal operating condition, explain-
ing why vibration monitoring was unsuccessful in this par-
ticular study. The second case study showed the ability of
AET to detect the presence of foreign matter (sand) in the
bearings of a pump unit. Sundt commented on the use of AE
to defect defective bearings utilizing race resonance for ampli-
fication, noting that this could enhance detection sensitivity.
However, it was stated that the mechanical “Q” (dynamic
magnification factor) of the race was an unpredictable function
of the bearing type, housing constraint, etc. Furthermore, it
was noted that race resonances could be excited by normal
background noise. Also, similar readings could be obtained
from a good bearing with a high “Q” and a bad bearing with
a low “Q”. The difficulty with monitoring bearings at the
element resonating range (20–100 kHz) was also discussed
by Barclay and Bannach (1992). It was noted that wave-
lengths of vibration at these frequencies are often compara-
ble with the dimensions of parts in the bearing or bearing
housing which may create standing waves with nodes and
antinodes. The consequence of this makes sensor position
critical. Barclay and Bannach (1992) presented the spectral
emitted energy (SEE) method, which combined the high-fre-
quency AE detection within the 250–350 kHz range with the
enveloping technique. The source of AE activity was attrib-
uted to the metal-to-metal contact as a result of lubricating
film breakdown. It was concluded that the SEE method was
a viable technique for detecting rolling element bearing defects
and compliments the present-day low-frequency vibration.
Badi et al. (1990) investigated the condition of automotive
gearbox bearings using stress waves (also known as AE).
These sensors were used on a bearing test rig with simulated
faults. All the artificially seeded faults were identified by
employing the stress wave sensor method. The sensors were
easy to install and needed simple signal processing to evalu-
ate bearing faults. The only drawback was that the sensors
were bulky. Sturm et al. (1992) employed AE to investigate
damage processes (pitting and mixed friction) of sliding and
rolling element bearings under laboratory and field condi-
tions. Analysis revealed that the amplitude behavior observed
from the envelope analysis of the AE signals yielded essen-
tial information about the damage processes. Javed and Lit-
tlefair (1993) presented some general aspects of the application
of AE for detecting the early development of failures in roll-
ing element bearings. Some results of the experimental
investigation of the basic relationship between ball bearing
failures and the resulting change in AE signal were pre-
sented. Neill et al. (1998a) described the relative sensitivities
of accelerometer and AE sensors to a range of defects and
assessed their merits in an industrial environment, where
ambient noise and/or other faults were highly influential. It
was revealed that the AE signals preserve the impulsive nature
of defect-element interactions, yielding characteristic har-
monics of the defect frequencies in the spectrum. These har-
monics distinguished bearing defects from other periodic
faults induced by imbalance or misalignment occurring at the
same frequency. Also, Neill et al. concluded that the AE sen-
sors were more sensitive to small defects.
Salvan et al. (2001) adopted a triangulation technique by
employing two AE sensors with fuzzy neural networks on a
high-speed post office mail sorting machinery, which con-
tained a large number of bearings. The investigation was
limited to the detection of a simpler source and the authors
were unable to obtain a precise location, presumably due to
incorrect parameters in the sound velocity equation and the
use of an inefficient technique. Parikka et al. (2002) reported
their findings on the operation of paper machines, which
were equipped with a number of oil-lubricated rolling bear-
ings. They assessed information on the effects of higher or
significantly lower than intended bearing loads on its service
life, as the lubricant conditions or movement changes with
time. It was commented that the possibility of using AE for
monitoring critical operating situations of rolling bearings
was very promising. Based on this investigation, a window-
based diagnostic system (prototype) was developed. Morhain
and Mba (2002) investigated the application of standard AE
characteristic parameters on a lightly radially loaded bear-
ing. An experimental test rig was designed to allow seeded
defects on the inner/outer races. The test rig also produced
high background AE noise providing a realistic test for fault
diagnostics. It was concluded that irrespective of the high
levels of background noise and low radial load (between 2%
and 70% of the bearing rating), standard AE parameters pro-
vided adequate early indication of bearing defects. Fan et al.
(2005) presented data streaming technology for non-inter-
rupted acquisition of AE waveforms. In addition, Fan et al.
reiterated that modulation of the AE waveforms could iden-
tify the defective part (race, roller) within the bearing. Hol-
royd (2000) detailed laboratory studies on rolling element
bearings in which AE signals were processed in terms of
their dynamic envelop (i.e. rectification and low pass filtering).
Tests showed that the periodicity of the enveloped signal
corresponded to a bearing defect frequency. A proprietary
method of characterizing the AE time waveform was pro-
posed. Several successful applications of the proprietary
method were also presented.
Finley (1980) developed an incipient failure detection (IFD)
system based on high-frequency AEs generated from shock
pulses as a rolling element (ball) passes a defective race. A
couple of industrial case studies were presented. Finley noted
that AET has been proven to be more effective than conven-
tional low-frequency sound and vibration measurements.
Jamaludin et al. (2002) presented research findings on the
lubrication monitoring of low-speed rolling element bear-
ings (1 rpm). A test rig was designed to simulate the real
bearing used in real-life situations. Using a newly devel-
oped method called the pulse injection technique (PIT), the
variation of lubricant amount in the low-speed bearing was
successfully monitored. This technique was based on trans-
mitting a Dirac pulse to the test bearing in operation via an
AE sensor. The AE data were processed using a clustering
technique based on the autoregressive (AR) coefficient to
differentiate between properly and poorly lubricated bear-
ings. The AE technique has also been employed by Miettinen
and Salmenperä (2002), Miettinen and Andersson (2000),
and Holroyd (2000) to monitor the lubricant condition in
rolling element and plain bearings.
Whilst monitoring bearing degradation by AE and vibra-
tion analysis is relatively established at speeds above 600
rpm, at low-rotation speeds there are numerous difficulties
with vibration monitoring that have been detailed (Berry,
1992; Murphy, 1992; Canada and Robinson, 1995; Robinson
et al., 1996). The difficulty of monitoring at low rotational
speeds was summarized by Kuboyama (1987).
Unlike vibration monitoring there has been considerable
success in the development and application of AE to moni-
toring slow-speed bearings. McFadden and Smith (1983)
explored the use of AE transducers for the monitoring of
rolling element bearings at speeds varying from 10 to 1850
rpm. The sensors were placed on the bearing housing. A fault,
simulated by a fine scratch on the inner raceway, formed the
basis of this experiment. It was commented that the AE trans-
ducer, with a frequency response beyond 300 kHz, failed to
perform as expected at the higher end of the rotational speed
range (850 rpm) and was inferior to the conventional high-
frequency accelerometer. However, at low rotational speeds
(10 rpm) the AE transducer appeared to respond to minute
strains (local distortions) of the bearing housing caused by
the concentrated loading of each ball in the bearing. These
minute strains appeared as spurious spikes superimposed on
the ball pass frequency. It was concluded that at low speeds
with steady loads, the base bending/strain of the bearing hous-
ing could enable the AE transducer to detect signatures from
very small defects in rolling element bearings, while at higher
speeds base bending appears as low-frequency noise.
Smith (1982) was involved in the experiment mentioned
above and, in a separate paper, reiterated the findings of
McFadden and Smith (1983), although puzzled at the behav-
ior of the AE sensor used, stating “the form of response of the
AE sensor was puzzling since the transducer was responding
to once-per-ball distorting in the casing at frequencies as low
as 1 Hz. AE transducers are not supposed to respond to fre-
quencies as low as these.”
Tavakoli (1991) investigated the application of AE to nee-
dle bearings. Interestingly, the rotational speed for this inves-
tigation was 80 rpm, which some might classify as a low-
speed application. Three simulations were undertaken: defect-
free fully lubricated, defect-free unlubricated, and a condition
in which two adjacent needle elements (rollers) were missing.
The frequency domain characteristics of the AE rms voltage
were examined in relation to the simulated conditions. It was
shown that the mean spectral density function of the rms
voltage distinguished all three simulations. It was also noted
that the source of AE in bearings was attributed to friction
and impacting.
Holroyd (1993) described in this application note the results
of AE measurements on four heavily loaded roller bearings
rotating at 60 rpm. The operation of these bearings in the
slowly rotating machine was critical indeed. This case study
clearly demonstrated the ability of this innovative and prof-
itable technology to prevent secondary damage and to mini-
mize production loss due to machine failures. Miettinen and
Pataniitty (1999) described the use of the AE method in
monitoring of faults in an extremely slowly rotating rolling
bearing, whose rotational speed varied from 0.5 to 5 rpm.
This investigation revealed that the AE measurement was
very sensitive and the fault was easily identified under labo-
ratory conditions. Jamaludin et al. (2001) reported the results
of an investigation into the applicability of AE for detecting
early stages of bearing damage at a rotational speed of 1.12
rpm. A bearing test rig was used with seeded localized sur-
face defects induced by spark erosion on the inner/outer
races and on a roller element (which resembled pitting). The
paper concluded that AE parameters such as amplitude and
energy provided valuable information on the condition of a
particular low-speed rotating bearing.
Sato (1990) investigated the use of AE to monitor low-
speed bearing damage by simulating metal wipe in journal
bearings at 5.5 rpm. It was observed that acoustic bursts
were generated as a result of slight metallic contact and the
amplitude of the waveform became larger with increasing
6The Shock and Vibration Digest / September 2005
metal wear. Sturm and Uhlemann (1985) also investigated
the application of AE to plain bearings, noting the instanta-
neous response of AE to the changes in the frictional state of
hydrodynamic fluid film.
Williams et al. (2001) noted that the majority of bearing
diagnosis experiments were undertaken with seeded defects
and, as such, undertook bearing experiments without seeded
defects; in essence, fatigue tests. The test bearings, roller, and
ball were run at 6000 rpm at 67% of the dynamic rated load,
although some tests were undertaken at varying speed con-
ditions. Vibration and AE techniques were compared, and
in one particular instance Williams et al. stated that the AE
sensor showed an increase 10 min after an increase in vibra-
tion. It was also noted that the AE sensor was unresponsive
to outer race failures. This is rather surprising considering
the number of publications confirming the ability of the AE
technique to diagnose outer race defects.
The development of AE in bearing monitoring and fault
diagnosis is the most established application of AE in rotat-
ing machinery and this is reflected in the number of com-
mercially available systems on the market today. Needless
to say, more detailed investigations are still required and there
are opportunities for applying AET for prognosis.
3. Application of Acoustic Emission to
Monitoring Gearboxes
Whilst vibration analysis on gear fault diagnosis is well
established, the application of AE to this field is still in its
infancy. In addition, there are limited publications on the appli-
cation of AE to gear fault diagnosis. Irrespective of the numer-
ous publications on the application of vibration analysis to
monitoring gearboxes, it still meets with great challenges
that monitoring and diagnosis of gearboxes present. AET offers
a complementary tool in this instance.
Miyachika et al. (1995) presented a study on AE in a
bending fatigue test of spur gear teeth. Three different gears
with common module, pressure angle, and number of teeth
were used. Two of the gears were case hardened to different
case depths. These gears were made from SC415 steel with a
face width of 10 mm whilst the second gear (face width of 8
mm) was made from S45C steel without any case hardening.
An AE sensor was fixed on the gear with a clamp arrange-
ment. AE measurements, such as frequency spectra, cumula-
tive event count, event count rate, and peak amplitude, were
recorded during the fatigue process under different tooth
load conditions. In addition, crack length measurements were
made. However, the type and characteristics of the sensor,
the sampling rate employed, and the loading frequency were
not presented in this paper. During the fatigue test, it was
observed that there was a marked increase in AE cumulative
event count and event count rate just before crack initiation
for both case hardened gears. For the normalized gear, such
an observation was not noted. It was also found that as the
tooth load decreased, the number of cycles until the marked
cumulative event count occurred increased. Miyachika et al.
drew the conclusion that the prediction of crack initiation
using the AE technique was possible for case hardened gear
but difficult in the case of the normalized gear.
Miyachika et al. (2002) extended their investigations to
supercarburized gear material. The investigation was per-
formed under the same test set and procedures as detailed
above, with additional analysis techniques: AE cumulative
energy count and wavelet transforms of AE signals. From
the results, Miyachika et al. concluded the prediction of crack
initiation by means of the AE method was possible for the
various carburized gears tested.
Wheitner et al. (1993) performed a series of gear tooth bend-
ing fatigue tests to verify the effectiveness of AE and system
stiffness measurements for monitoring the crack initiation
and propagation. The tests and instrumentation employed were
to standards detailed in the Society of Automotive Engineers
(SAE) gear geometry, testing procedure and fatigue test fix-
ture. The AE senor had a resonant frequency of 300 kHz and
was attached to the gear at the root of the tooth with super-
glue. The tooth stiffness measurements were made through
an accelerometer mounted to the base of the fixture. The test
gears were of various materials, surface finishes, and surface
treatments. All the testing was performed by applying sinu-
soidal load of 10 Hz and load ratio of 0.1. A run-out life of
106 cycles was employed for all the test cases. Wheitner et al.
noticed non-zero AE counts before the initiation point of the
gear tooth root fatigue crack, which was attributed to the
background noise of the test machine. In general, AE activ-
ity increased with crack propagation and very rapidly at the
failure point. All the test gears exhibited similar trends in
stiffness measurements. At high load and low fatigue lives,
crack propagation life contributed a significant proportion of
the gear total life as compared to crack initiation life. Wheit-
ner et al. went further to conclude that both the AE and sys-
tem stiffness measurements were effective in monitoring the
cracking processes of the gear tooth. However, in most cases,
AE activity was detected before the first change in stiffness
compliance was registered.
Singh et al. (1999) explored an alternative AE technique to
the more widely used vibration and debris monitoring meth-
ods for detection of gear tooth crack growth. They employed
a single tooth bending machine with the load on the tooth
varied sinusoidally at 40 Hz frequency. An AE sensor and
accelerometer were mounted on a spur gear near to the load-
ing tooth. The test terminated when the loaded tooth broke
off. Raw AE waveforms and fatigue cycles were recorded
during the test. There was no information given on the type
of gear, sensors, the applied load, and the sampling rate used.
The test revealed that AE detected the first sign of failure
when the gear reached 90% of its final life. As the crack pro-
gressed, AE amplitude increased. During the final stage of
gear tooth fracture, a significantly high amplitude AE burst
was detected. On the other hand, the vibration level did not
change significantly in the initial stage of crack initiation
and propagation until the final stage of failure. Hence, Singh
et al. concluded that AE method offered an advantage over
vibration monitoring techniques.
In order to study the practical aspects of sensor placement
in a real-life gearbox situation, Singh et al. (1999) performed
an assessment of the transmissibility of an AE signal within
a gearbox. The tests were performed with different torque
levels using lead pencil breaks to simulate AE activity in the
gearbox. This technique is known as the Nielsen source test.
First, various individual interfaces with varying torques were
studied and quantified. Following this, Singh et al. evaluated
the total loss of strength of the AE signal across multiple inter-
faces and compared with the sum of losses obtained from
individual interfaces. Several AE transmission paths were
examined. From the results obtained, Singh et al. concluded
that the attenuation across the gearbox was an accumulation
of losses across each individual interface within the trans-
mission path, and that the optimum path of propagation will
be that with the smallest cumulative loss.
The investigations detailed earlier (Wheitner et al., 1993;
Miyachika et al., 1995, 2002; Singh et al., 1999) have indi-
cated that the AE technique was able to detect bending fatigue
failure. In addition, the AE technique was capable of detect-
ing the fault condition in advance of the vibration monitoring
technique. This conclusion is encouraging and motivating
for the AE technique to be the new condition monitoring
tool. However, to ensure that this technique is robust, the
defect detection capability on the other modes of gear failure
(surface damage and fatigue) has to be explored.
Siores and Negro (1997) explored several AE analysis
techniques to correlate possible failure modes of a gearbox
during its useful life. The gearbox employed for the failure
interrogation includes two gear sets (input and output), a DC
shunt motor, and a variable speed controller to alter the motor
speed for the tests. The AE sensor employed was mounted
on the gearbox casing and has a resonant frequency of 175
kHz. Prior to the start of the test, the gearbox was allowed to
wear-in at 1200 rpm for four 1 h intervals at full load condi-
tion. Common gear failures, such as excessive backlash, shaft
misalignment, tooth breakage, scuffing, and worn teeth, were
seeded on the test gears. All the seeded defect conditions
were tested at 300 and 600 rpm whilst AE parameters such
as rms, standard deviation, and duration of AE were meas-
ured. Siores and Negro concluded that the monitored AE
parameters exhibit identifying qualities for the respective
failure modes.
Singh et al. (1996) performed two experiments to study
the feasibility of applying AE to detect gear pitting. Both
simulated and natural pits were used to evaluate this detec-
tion technique. The first experiment employed a UH1H gen-
erator drive offset quill, which consisted of the driver, driven,
and idler gears. In this experiment, the idler gear contains the
simulated pit of width and depth of 1.25 mm. This pit was
simulated by removing a thin strip of material from the
pitch-line on one of the teeth of the idler gear by the EDM
process. A resonant type AE sensor with a resonant fre-
quency of 280 kHz and an accelerometer were mounted on
the gearbox housing near the output shaft bearing. A tachom-
eter was used as a trigger to ensure each cycle of the meas-
urements started with the same idler tooth in contact. The
test gearbox was first run with no pit on the idler gear and
then replaced by the idler gear with simulated pits. AE and
vibration data were recorded during the run. This procedure
was repeated for several combinations of load and speed.
From the test results, Singh et al. concluded that both detec-
tion techniques were able to pick up the simulated defect but
the AE technique exhibited much greater signal-to-noise
ratio. He also suggested that both detection techniques were
unable to detect the simulated pit at extremely high speeds or
unloaded conditions as the noise level increases whilst the
amplitude of the defect signal arising from contact of the pit-
ted region decreases.
Singh et al. (1996) performed the second experiment using
a back-to-back gearbox to study the detectability of natural
pits. Similar acquisition systems to the first experiment were
employed with both the AE sensor and accelerometer mounted
on the housing of the test gearbox. The input speed to the
gearbox was 1775 rpm with an unknown torque loading. Dur-
ing the early stage of the test, there were no defects on the
mating gear teeth surfaces and the signals (both AE and vibra-
tion) showed no significant peaks above the operational
noise level. After 30 min of operation, pits started to develop
on the pinion teeth and periodically occurring peaks were
observed from the AE signals. A further 15 min run saw pit-
ting on multiple teeth and the detected AE signals revealed
more frequently occurred peaks above noise level. There
was no visible peak noted for the accelerometer signal. Dur-
ing the test, the AE sensor was also placed at the slave gear-
box housing and bearing location between the two gearboxes
to assess the detectability of the natural pits from the men-
tioned locations. Singh et al. concluded that the AE sensor
should be as close to the monitored part as possible in order
to maximize the detection capability of pits using AE tech-
Raad et al. (2003) illustrated the application of the AE
monitoring technique for gear fault detection by employing
an industrial gear rig. No information on the gear test rig,
applied torque, and speed was given in the paper. The exper-
iment was performed above the rated load of the gears for
two weeks until near breakage of two teeth. Various types of
AE sensors (resonant and wide band) and accelerometers
were mounted on the bearing. Measured signals were taken
at regular intervals and visual inspection of gears was per-
formed at the end of each day. The recorded AE and vibra-
tion data were analyzed using four different methodologies:
visual comparison, Kurtosis, spectral density, and envelope
analysis. The visual comparison revealed that AE bursts
appeared with spalling. However, these AE bursts disap-
peared after the defect was established. There was no clear
indication from vibration signatures. The Kurtosis values were
correlated to spalling defects after 3000 cycles. However, this
method was unable to localize the spalling defect to individ-
ual tooth. The first sign of spalling observed from the vibra-
tion technique was at 5000 cycles. Using the spectral density
analysis method, the increase in energy before and after the
spall detection was common to both AE and vibration sig-
nals. In the final analysis of the AE and vibration signals, the
spectrum of the squared envelope was used. The vibration
technique was able to pick up the defect by displaying peaks
at twice the shaft frequency. However, these peaks were not
visible in the AE spectrum until the logarithm of the squared
envelope was employed. The observed peaks occurred at the
same frequency for both AE and vibration techniques. Raad
et al. concluded that this first evaluation of AE as condition
monitoring tool was promising.
Sentoku (1998) presented an investigation on tooth sur-
face failure with AE measurements. A power circulating type
gear testing machine was employed. The testing machine
consisted of a pair of test and power return spur gears with
a forced lubrication system that supply oil directly to the
engaged teeth surfaces from the side of the gear pairs. It is
important to note that the oil temperature was maintained
constantly at 40 ± 2oC. This eliminated the effect of oil film
thickness on AE activity. An ultracompact AE sensor of res-
onant frequency 350 kHz was mounted on the gear wheel
using screws. The AE signature was transmitted from the
sensor to the data acquisition card via a mercury slip ring. A
strain gage was also adhered to the tooth root to correlate the
8The Shock and Vibration Digest / September 2005
extracted AE parameters with tooth root strain waves. Dur-
ing the tests, the roughness of the gear teeth surfaces and pit-
ting size were measured at regular intervals.
The first test was performed under applied stress of 960
MPa and pinion speed of 992 rpm using hardened gears.
From the results obtained, Sentoku observed no change in
AE amplitude except that the unevenness of AE wave lines
was smaller with an increasing number of cycles. At this stage
of the test, no surface damage was noted. Subsequently, Sen-
toku performed the second test using heat-treated ground
gears. During the early stage of the test, both AE amplitude
and the pitting area ratio remained unchanged. However,
when pitting on the three monitored gear teeth began, AE
wave lines started to change. Subsequently, AE amplitudes
increased with both the pitting area ratio and the numbers of
cycle. Sentoku explained that the increase in AE amplitude
was caused by friction due to increasing pitting. Similar
observations were noted for AE energy. Hence, with the
results obtained from the test, he drew conclusion that the
AE technique could detect gear teeth pitting.
Badi et al. (1996) performed an investigation on the use of
AE and vibration monitoring techniques for condition mon-
itoring of a typical drive line. A test rig comprised of a drive
and simple spur gearbox, loaded by a pneumatically oper-
ated brake disk, was employed to simulate the essential part
of this drive line. The rotating components were connected
by flexible couplings and supported by bearing blocks. The
rig was instrumented with both accelerometers and AE sen-
sors at several locations along the drive line. However, Badi
et al. only reported the results from the sensor which gave
the optimum location for fault detection. Seeded defects
such as “blip” and “shaved” gear faults were introduced on
the test gears to simulate scuffing and pitting defects on gear
tooth. There was no further information on the testing proce-
dures used in this experiment. Analysis techniques such as
the crest factor and kurtosis were employed for both AE and
vibration techniques. For the “blip” gear fault, both monitor-
ing techniques were able to identify the defect through the
analysis techniques employed. As for the “shaved” gear fault,
only the AE technique was able to detect the defect. Badi et
al. concluded that the analysis techniques used were ideally
suited for identifying faults with an impulsive nature. How-
ever, for a more comprehensive methodology, other analysis
techniques should be explored.
Tandon and Mata (1999) performed seeded defect tests
on spur gears using an IAE gear lubricant testing machine
to assess the fault detection capability of the AE technique
and to make a comparison with the more widely used vibra-
tion technique. Both hardened and ground spur gears were
employed for the tests. The test gears were lubricated by a jet
of oil. The AE sensor and accelerometer employed had res-
onant frequencies of 375 and 39 kHz, respectively. Both the
AE and vibration signals were measured closed to the bear-
ings of the test gearbox. All the tests were carried out at a
single speed (1000 rpm) and varying load conditions (0–10 kg).
AE and vibration measurements were first taken for gears
that have no seeded defect, which were treated as reference
signals. Subsequently, a simulated pit of constant depth (500
µm) and variable diameter (from 250 to 2200 µm in incre-
mental order) was introduced on a gear tooth pitch-line by
spark erosion. From the tests, Tandon and Mata made these
observations. (a) There was some increase in AE with increase
in load. (b) The AE parameters increased as the defect size
(diameter of pit) increased. (c) The AE (ring-down) counts
showed slightly better results than the other AE parameters
measured. (d) The AE technique detected the seeded defect
at a smaller size (500 µm) compared to the vibration tech-
nique (1000 µm). (e) In general, the distribution of AE events,
counts, and peak amplitude became broader due to the pres-
ence of a defect in the gear.
Finley (1980) presented an industrial case study on the appli-
cation of an AE developed system (IFD) for gearbox moni-
toring. Al-Balushi and Samanta (2002) introduced energy-
based features extracted from AE signatures for monitoring
and diagnosing gear faults. This feature, called the energy
index (EI), was defined as the square of the ratio of the rms
value for a segment of the signal to the overall rms value of
the entire signal. Various different forms of EI were derived
and compared with existing statistical methods for early
fault detection. Experiments were undertaken on a back-to-
back spur gearbox. Three miniature ultrasound transducers
were implanted onto the rolling element bearing adjacent to
the gear wheel for collection of AE data. A triggering system
was used to ensure all the acquired data have identical start-
ing locations on the gear. The tests were performed using
brand new gears and terminated at the 40th hour when the
gear failed. AE signals were acquired for one revolution of
the test gear at hourly intervals. However, information such
as the characteristics of the sensors, the applied load, and the
reason for the varying rotational speeds was undisclosed. Al-
Balushi and Samanta illustrated that the proposed EI and the
various derived forms were able to locate the broken and pit-
ting teeth more effectively than the traditional kurtosis and
crest factor methods. By employing the proposed analysis
technique, the defective tooth was picked up in a helicopter
In a separate report Al-Balushi and Samanta (2000) pre-
sented a procedure for fault diagnosis of gears through wave-
let transforms and artificial neural networks (ANNs). The
time domain AE signals of a rotating machine with normal
and defective gears were processed through wavelet trans-
form to decompose in terms of low-frequency and high-
frequency components. The extracted features from the wave-
let transform were used as inputs to an ANN-based diagnostic
approach. The procedure was illustrated through the experi-
mental AE signals of a gearbox.
Tan and Mba (2005a, 2005b) noted difficulties in identi-
fying the location of a defective tooth during an experimental
investigation. It was noted that the lubricant temperature had
an influence on the levels of AE activity/strength during the
gear mesh. This has far reaching consequences as it implies
that whilst other researchers have stipulated the effect of load/
speed on AE activity, the time of data acquisition, in effect
the temperature of the lubricant, will influence the levels of
AE obtained.
While exploring the applicability of the AE technique to
gear health diagnosis, Toutountzakis and Mba (2003) made
some interesting observations of AE activity due to misalign-
ment and natural pitting. The test was performed on a back-
to-back spur gearbox with the AE sensors placed on the pin-
ion and bearing casing of the pinion shaft. The AE sensors
used have a relative flat response in the region between 150
and 750 kHz. A silver contact air-cooled slip ring was
employed to transmit the AE signal for further processing.
AE parameters such as rms and energy values were recorded
during the tests. Prior to the test proper, AE measurements
for defect-free gears were first recorded. As the rotational
speed increased, measured AE parameters increased for both
AE sensor locations. Furthermore, Toutountzakis and Mba
observed that change in speed resulted in changing AE param-
eters. During one of the tests, Toutountzakis and Mba noted
increasing AE rms (at pinion location) for 6 h before the
gearbox was paused for inspection. The results of the inspec-
tion revealed signs of pitting and scuffing, which indicated a
misalignment in the gearbox. The gearbox was reassembled
and the test continued. An interesting observation was made:
“a reduction in AE parameters was noted initially, but these
values gradually increased to values which did not depart
from the initial gradient of the increasing trend.” Toutount-
zakis and Mba concluded that there is a potential application
of the AE technique for gear health diagnostic.
Price et al. (2005) investigated the detection of severe slid-
ing and pitting with AE. The experimental results presented
were based on a “four-ball machine” test-rig. It was observed
that scuffing and pitting were easily detectable by observing
changes in AE energy, principally due to changes in contact
friction. More interestingly, Price et al. noted changes in the
frequency patterns of measured AE signals prior to pitting
and stated that AE monitoring was capable of detecting wear
events prior to either vibration monitoring or wear debris
analysis. Building on this statement, very recently Tan et al.
(2004, 2005a, 2005b, 2005c) have presented results of an
experimental investigation in which natural pitting of spur
gears was allowed to occur. Throughout the test period, AE,
vibration and spectrometric oil samples were monitored con-
tinuously in order to correlate and compare these techniques
to the natural life degradation of the gears. It was observed
that the AE technique was more sensitive in detecting and
monitoring pitting than either the vibration or spectrometric
oil analysis (SOA) techniques. It is concluded that as AE
exhibited a direct relationship with pitting progression, it
offered the opportunity for prognosis. From the results pre-
sented it was clearly evident that the AE monitoring indica-
tor could be linearly correlated to the gearbox pitting rates
for all torque conditions, with detection of onset of pitting as
early as 8% of the pitted gear working face area. This offered
much earlier diagnosis than vibration analysis, where only
after between 20% and 40% of pitted gear working face did
this technique offer capability for defect identification. This
near linear relationship between AE and pit progression offers
great potential, and opportunities, for prognostics in rotating
Tan and Mba (2004a, 2004b, 2005a, 2005c) ascertained
the AE source mechanism through a series of experimental
programs. These experimental programs consisted of iso-
thermal tests on undamaged gears to explore the effects of
rotational speed and applied torque on AE levels. From the
isothermal test results, it was observed that variation of the
applied torque had a negligible effect on the AE rms levels,
similar to the negligible effect of load on film thickness under
elastohydrodynamic lubrication (EHL) of non-conforming
mating surfaces. It was noted that the variation in rotation
speed had a more pronounced effect on AE rms levels rela-
tive to the load. Tan and Mba concluded that the source of
AE during gear mesh was asperity contact under rolling and
sliding of the meshing gear teeth surfaces.
Although the development of AE in gear diagnosis is in its
infancy, the papers reviewed have illustrated the potential and
viability of AE becoming a useful diagnostic tool in condition
monitoring of gears. However, more detailed investigations
are required to ensure this technique is robust and applicable
for operational gearboxes. This involves understanding the
influences of operational variables on AE generation and
investigating the effects of variable load conditions to moni-
toring with AE.
4. Pumps and Acoustic Emission Technology
Pumps play a significant role in industrial plants and need
continuous monitoring to minimize loss of production. Every
pump manufacturer supplies characteristic curves for their
equipment illustrating pump performance under given condi-
tions. These curves demonstrate the inter-relationship between
discharge capacities, pump head, power, and operating effi-
ciency. The ideal operating point for a pump is known as the
best efficiency point (BEP). This is the point where pump
capacity and head pressure combine to provide the maxi-
mum efficiency of the pump. If the pump operates too far to
the left or right of the BEP, not only may its efficiency be
compromised, but it can also be subjected to increased wear,
reducing operational life. Also, the pump manufacturer will
undertake net positive suction head (NPSH) tests on sup-
plied pumps; the significance of the latter is to determine the
3% drop in head at which serious cavitations will occur.
Cavitation occurs when the absolute static pressure at some
point within the pump falls below the saturated vapor pres-
sure of the liquid. It causes a loss of pump efficiency and
degradation of the mechanical integrity of the pump. It is
generally accepted that the critical pressure for inception of
cavitation is not constant and varies with operation fluid
physical properties, the surface roughness of the hydraulic
equipment, etc. In addition, cavitation is known to begin long
before the performance of the pump is affected (McNulty and
Pearsall, 1962).
In this paper we review the application of AE for condi-
tion monitoring of pumps. Prior to detailing some recent
attempts at applying AE to pump health diagnosis, the inves-
tigations of McNulty and Pearsall (1962) and McNulty and
Deeprose (1978) are worth mentioning. They undertook high-
frequency measurements (up to 160 kHz) taken at the suc-
tion and discharge sides of the pump and detected incipient
cavitation. However, it was noted that the success was depend-
ent on the operational background noise levels. These results
relating NPSH to varying noise levels are of great interest,
although undertaken at 40 kHz. This clearly relates the audi-
ble intensity and high-frequency energy to the varying cavi-
tation stages experienced by a pump as the head drops to the
3% level. It was noted (McNulty and Pearsall, 1962) that dur-
ing cavitation the high-frequency noise increased. In sepa-
rate paper, McNulty (1981) showed that the minimum noise
intensity levels of a pump were obtained at the BEP. Sources
of noise were noted as turbulence, impeller and volute inter-
actions and hydraulic interactions.
Derakhshan et al. (1989) investigated the cavitation bub-
ble collapse as a source of AE and commented that the high
amplitude pressure pulse associated with bubble collapse
generated AE. When the AE sensor was placed on the actual
specimen experiencing cavitation, Derakhshan et al. observed
10 The Shock and Vibration Digest / September 2005
increasing AE rms levels with increased pressure of flow
and cavitation. However, with the AE sensor mounted on the
tank wall the reverse was observed: decreasing AE rms lev-
els with increasing pressure and cavitation. This was attrib-
uted to a visible bubble cloud that increased with pressure. It
was commented that this cloud attenuated the AE signature
prior to reaching the transducer on the wall casing. In addi-
tion to the high amplitude pressure pulse associated with
cavitation, pressure pulses associated with centrifugal pumps
have been detailed (Guelich and Bolleter, 1992); these include
wake flow from the impeller blade trailing edge, vortices
generated by flow separation and recirculation. The influ-
ence of the latter on pump performance has been presented
(Fraser, 1981).
Neill et al. (1996, 1997) assessed AET for detecting early
cavitation. It was also noted that the collapse of cavitation
bubbles was an impulsive event of the type that could gener-
ate AE. These transients cause very high local transient pres-
sure that can damage the internal parts of pumps. It was
observed that when the pump was under cavitation, the AE
operational background levels dropped in comparison to non-
cavitating conditions. To ensure a more direct transmission
path between the fluid and the sensors, metal wave guides
were put into the venture tube wall at different locations. It is
worth stating that prior to, and during cavitation, vibration
measurements showed no significant change. In conclusion,
Neill et al. stated that loss in NPSH before the 3% drop-off
criterion was detectable with AE and evidence of incipient
cavitation was detectable in the higher frequency band (0.5–
1 MHz). It is interesting to note that Neill et al. (1998b) also
successfully applied AET to detect the recirculation in a cen-
trifugal pump. Recirculation is defined as a flow reversal at
either the inlet or the discharge tips of the impeller vanes. It
occurs in axial, centrifugal shrouded and unshrouded pumps.
It is important to detect this phenomenon at the earliest stage
and distinguish it from other undesirable phenomena, such
as cavitation.
Hutton (1969) investigated the feasibility of detecting AE
in the presence of hydraulic noise. It was noted that artificially
seeded AE bursts were detected above background opera-
tional noise for turbulent flow, with and without cavitations.
Furthermore, Hutton noted that the presence of cavitations in
the system increased the operational AE noise levels by a
factor of 50. In addition, cavitation was found to generate a
significant increase in noise level below 500 kHz. Hutton
placed AE sensors on the pipe. Darling and Johnston (1991)
found that AE from a high pressure hydraulic pump during
cavitation was wide band noise, up to 1 MHz. Darling and
Johnston noted that during cavitation there was little change
in the vibration signature from normal operation, which was
not the case with AE observations. It was also commented
that the position of the AE sensor was insensitive to mount-
ing position whilst the reverse was observed with the vibra-
tion senor.
Al-Maskari (1984) attempted to detect incipient cavitation
with AE but concluded that whilst the inception of cavitation
was not detectable with AE, fully developed cavitation was
detectable. Another interesting observation by Al-Maskari
was the variation in AE activity at flow rates below the BEP
and it was suggested that investigations on applying AE to
cavitation detection should be concentrated at the BEP. Al-
Maskari placed the AE sensor on the pump casing. Sundt
(1979) detailed a case study on the application of AE for
detecting pump cavitation. It was shown that during cavita-
tion AE levels increased whilst vibration levels dropped.
Also, Finley (1980) presented an industrial case highlighting
the successful application of AE to cavitation detection.
Whilst promoting the use of audible acoustics (less than
20kHz) for monitoring pumps, Cudina (2003) cites some appli-
cations of AE for detecting broad-band noise associated with
cavitation. Al-Sulti et al. (2005) noted that the use of the
power spectrum density of AE acquired over a range of flow
rates was not effective in detecting cavitation. However, it
was noted that the use of higher-order spectral analysis (bi-
coherence) showed improved sensitivity of AE over vibra-
tion for early detection of cavitation. The results are in contrast
to nearly all published work on AE for monitoring cavitation
where a clear increase in AE levels was noted without the
need for advanced signal processing.
The papers reviewed above have clearly associated AE
with the collapse of cavitation bubbles. The presence of cav-
itation has been shown to increase operational AE noise levels.
Recently, Alfayez et al. (2005) and Alfayez and Mba (2005)
undertook experimental tests on a range of pumps in an
attempt to correlate incipient cavitation with AE activity.
The results showed a clear relationship between AE activity
measured from the pump casing, suction and discharge
pipes, and incipient cavitation. At a high NPSH value, when
incipient cavitation is known to occur, a significant increase
in AE was observed. Experiments were conducted for several
flow rates on different sized pumps to validate this assump-
tion. Further reduction in the NPSH resulted in a decrease in
measured AE levels due to the presence of bubble clouds.
Observations of the frequency content of captured AE time
waveforms showed a shift in frequency range for incipient
and developed cavitation. The results of this study also
showed that the measurement of AE rms levels could be
employed for determining the BEP of pumps, which offers
enormous opportunities within the industry. Sikorska and
Hodkiewicz (2005) reiterated the observations of Alfayez et
al. (2005) and Alfayez and Mba (2005), noting that AE was
able to detect off duty conditions in double suction pumps.
Furthermore, Sikorska and Hodkiewicz noted that AE could
be used to detect cavitation and recirculation and postulated
that low-flow AE activity was initiated by recirculation
whilst high-flow AE activity was due to incipient cavitation.
Cavitation is known to occur more easily at higher flow rates
(Cudina, 2003).
5. Monitoring Engines and Rotating Structures
with Acoustic Emission
Industries all over the world use various machines and
structures to manufacture and distribute various goods and
services to global customers. These include rotating and recip-
rocating machines and mechanical structures of all sizes,
shapes, and complexities. Damage assessment of these assets
(both old and new) is very crucial as it determines the quality,
reliability, availability, maintainability, and the life expect-
ancy. The reliability and health monitoring of both old and
new machineries and structures form the subject of exten-
sive research in many academic institutions, government
laboratories, defense research establishments, and industrial
organizations worldwide. AET is now becoming a widely
accepted practice in the field of engine and rotating struc-
tural monitoring.
Holroyd et al. (1996) illustrated the background to the AE
approach and its technological developments, which enabled
it to be used as a means of dynamically probing the opera-
tion of machineries and mechanisms, and attempted to clar-
ify the opinions held on the similarities and differences of
the AE and vibration monitoring techniques when applied to
machinery condition monitoring (Holroyd and Brashaw,
1999). Holroyd and Randall (1993b) illustrated with exam-
ples some real benefits of using AE techniques as a highly
sensitive, simple to use, and cost-effective maintenance tool.
Gill et al. (1998) described how AE techniques could be
implemented as a condition-based maintenance strategy to
monitor the inlet and outlet valves of reciprocating compres-
sors. The investigation was based on an eight-cylinder, hor-
izontally opposed, single acting, two-stage compressor used
to compress ethylene at a large plastics plant. Gill et al. high-
lighted the possibility of detecting fluid movement with AET.
The sensor required very little space and was non-intrusive,
which was a major benefit in the hostile conditions. The
results revealed the practical deployment of AE sensors for
condition monitoring applications.
Fog et al. (1998) conducted an experimental investigation
into detecting exhaust valve burn-through on a four-cylin-
der, 500 mm bore, two-stroke marine diesel engine with an
output of approx. 10,000 BHP. The investigation comprised
monitoring three different valve conditions (normal, leak,
and large leak). Vibration and structure-borne stress waves
(AE) were monitored. The results showed that the AE sig-
nals contained more information for identifying valve and
injector related mechanical events during the combustion
process than time series recorded from other sensors. Fea-
tures of the AE signals were extracted using principal com-
ponent analysis (PCA). A feedforward neural classifier was
used to discriminate between the three valve conditions. Friis-
Hausen and Fog (2001) identified efficient classifiers for
the detection of two different failure modes in marine diesel
engines: exhaust valve leaks and defective injection (misfire).
The purpose of the exhaust valve is to seal the combustion
chamber from the surroundings during compression, thus
securing maximum pressure in the cylinder during the com-
bustion event. This ensures maximum engine performance
in terms of output power. This study identified an efficient
classifier, which could discriminate completely between
leak sizes in the exhaust valve, based on the recorded rms
AE signals. An efficient classifier for detection of misfire
was also developed. El-Ghamry et al. (1998) illustrated the
potential of AE sensing to determine the strength of the air–
fuel mixture in 30.56 liter Perkins four-stroke, eight-cylinder
turbocharged gas engine. AE, acceleration, inside cylinder
pressure, and timing signals were monitored during the tests.
The results revealed that the AE signal showed additional
features, which could be used to identify the strength of the
gaseous fuel mixture.
Results that showed that indirect measurements of cylin-
der pressure from diesel engines with AET were presented
by El-Ghamry et al. (2005). The AE rms was correlated
to the pressure in the time and frequency domains. Fur-
thermore, the complex cepstrum analysis was used to model
the pressure readings from the complete combustion phase
of the engine. El-Ghamry et al. noted the advantage of employ-
ing the cepstral analysis for the model, stating that it used
the frequency content of the AE rms signal rather than the
energy content, which gave the advantage over signals with
low-energy content. The application of AET to modeling
the pressure originated from previous studies by El-Ghamry
et al. (2003). In the latter investigation El-Ghamry et al.
attempted to develop generic techniques for diagnosing
faults in reciprocating machines. The generic pattern recog-
nition technique developed was based on the time-domain
AE rms signals, statistical feature extraction from the time-
domain signal, and correlation of the AE response to specific
events in the engines. Steel and Reuben (2005) recently
reviewed developments in monitoring engines with AE. It
was noted that AE signals could be associated with the
actual operational and degrading processes in the engine.
Furthermore, this could be accomplished non-intrusively. It
was also stated that analysis of AE data could be enhanced
with a detailed knowledge of the operating conditions of
the engine, such as injector timing, running speed, and valve
Mba (2002) presented a case study on the application of
high-frequency AE as a means of detecting the early stages
of loss of mechanical integrity in low-speed rotating machin-
ery. Investigations were centered on the rotating biological
contactor (RBC), which is used for sewage treatment in small
communities and rotates at approximately 1 rpm. The results
presented were obtained from an operational unit that suf-
fered a fractured stub shaft retaining bolt head. Evidence to
support the inadequacies of vibration analysis and the appli-
cability of AE to detecting this fault condition were detailed.
The results of the case study pointed to the potential of AE
for diagnosing serious mechanical defects where vibration
analysis would be ineffective. The investigation showed that
AE activity could be related not only to the fractured bolt but
also to loose bolts. The mechanism for generating AE signa-
tures was the rubbing in the threaded bolt within its recess
and the rubbing and/or crushing of the fractured bolt shank
with wear and rust particles within the clearance hole of the
stub shaft. A typical AE parameter such as amplitude can
provide valuable information on the clamped condition of a
component on a low-speed rotating machine. These observa-
tions confirmed the finds of Hanel and Thelen (1995, 1996a,
1996b) where a relationship between AE activity and the
tensile stress on a bolt was established. A direct correlation
between increased AE activity and plastic deformation of
the bolt was presented. Furthermore, the investigators pro-
posed that the low-level AE activity in the elastic range of
the bolt corresponded to the friction process in the thread.
Smulders and Loob (1994) employed the use of envelop-
ing and high-frequency AET for monitoring: bearing fault
detection, very slow speed bearing, rail car turntables, and
lubrication condition in paper mill machinery. Mba et al.
(1996) and Mba and Hall (2001) presented the results of a
study into the use of stress wave analysis as a means of detect-
ing early stages of loss of mechanical integrity in low-speed
rotating machinery. The source of AE was attributed to the
breakage, and entrapment, of surface asperities as a result of
relative movement of clamped components that had lost pre-
defined tightening torques (loose clamped components). AR
coefficients associated with each AE provided an efficient
parameter for classification and diagnostics. Holroyd (2002)
reviewed some of his development work in applying AET to
12 The Shock and Vibration Digest / September 2005
machinery condition monitoring over the last decade, and
also introduced new developments in the field of condition
monitoring of structures.
6. Conclusion
AET is a continuously evolving multidiscipline and is now
the focus of intense research- and application-based studies.
The wealth of knowledge discovered, generated, and dis-
seminated in this evolving discipline is itself proof of its
diverse applicability. The interest in developing new tech-
nologies to overcome the many hitherto unsolved problems
in condition monitoring and diagnostics of complex indus-
trial machinery applications offers immense opportunities
for AET to grow unabated. This is also reflected by the sig-
nificant growth in global demand for AE sensors. With the
accelerating speed in the growth of intelligent information,
sensor and data acquisition technologies, combined with the
rapid advances in intelligent signal processing techniques, a
healthy growth in the application of AE in many engineer-
ing, manufacturing, processing, and medical sectors is to be
expected. The application of AE in prognosis has yet to be
fully explored and exploited. We are still a long way away
from interpreting and fully understanding the wonderful
“sounds of AE” from rotating machines.
Al-Balushi, K. R. and Samanta, B., 2000, “Gear Fault Diagnostics Using
Wavelets and Artificial Neural Networks,” in Proceedings of
COMADEM 2000, Houston, TX, 1001–1010.
Al-Balushi, K. R. and Samanta, B., 2002, “Gear Faults Diagnosis Using
Energy-based Features of Acoustic Emission,” Journal of Systems
and Control Engineering, Vol. 216, 249–263.
Alfayez, L. and Mba, D., 2005, “Detection of Incipient Cavitation and
Determination of the Best Efficiency Point for Centrifugal Pumps
Using Acoustic Emission,” Journal of Mechanical Process Engi-
neering, Part E, IMechE, in press.
Alfayez, L., Mba, D., and Dyson, G., 2005, “The Application of Acoustic
Emission for Detecting Incipient Cavitation and the Best Efficiency
Point of a 60KW Centrifugal Pump: A Case Study,” NDT and E
International, Vol. 38, No. 5, 354–358.
Al-Ghamdi, A. M. and Mba, D., 2005, “A Comparative Experimental
Study on the Use of Acoustic Emission and Vibration Analysis for
Bearing Defect Identification and Estimation of Defect Size,”
Mechanical Systems and Signal Processing, MSSP04-98R2.
Al-Ghamdi, A. M., Cole, P., Such, R., and Mba, D., 2004, “Estimation of
Bearing Defect Size with Acoustic Emission,” INSIGHT, Vol. 46,
No. 12, 758–761.
Al-Maskari, A., 1984, “Detecting of Cavitation in Centrifugal Pump Using
Acoustic Emission,” M.Sc. Thesis, Cranfield University, UK.
Al-Sulti, S., Samanta, B., Al-Balushi, K. R., Al-Zedjali, M., Al-Araimi, S.
A., and Siddiqi, R. A., 2005, “Comparison of Signal Processing
Techniques for Detecting Pump Cavitation Inception,” in Proceed-
ings of the 18th International Congress and Exhibition on Condition
Monitoring and Diagnostic Engineering Management (COMADEM
2005), Cranfield, UK, 31 August–2 September.
Badi, M. N. M., Johnson, D. E., and Trmal, G. J., 1990, “The Use of Stress
Wave Sensors for the Diagnosis of Bearing Faults,” in Proceedings
of COMADEM 1990, Brunel University, UK, 20–25.
Badi, M. N. M., Engin, S. N., and Schonfeld, D., 1996, “Fault Classifica-
tion of a Model Drive Line Using Time Domain Data,” in Proceed-
ings of COMADEM 1996, Sheffield, UK, 43–50.
Bagnoli, S., Capitani, R., and Citti, P., 1988, “Comparison of Accelerome-
ter and Acoustic Emission Signals as Diagnostic Tools in Assessing
Bearing,” in Proceedings of the 2nd International Conference on
Condition Monitoring, London, UK, May, 117–125.
Balerston, H. L., 1969, “The Detection of Incipient Failure in Bearings,”
Materials Evaluation, Vol. 27, 121–128.
Bansal, V., Gupta, B. C., Prakash, A., and Eshwar, V. A., 1990, “Quality
Inspection of Rolling Element Bearing Using Acoustic Emission
Technique,” Journal of Acoustic Emission, Vol. 9, No. 2, 142–146.
Barclay, J. and Bannach, R., 1992, “Multiparameter Condition Monitor-
ing,” Noise and Vibration Worldwide, March, 17–20.
Berry, J. E., 1992, “Required Vibration Analysis Techniques and Instru-
mentation on Low Speed Machines (Particularly 30 to 300 RPM
Machinery),” Technical Associates of Charlotte, Inc., Advanced
Vibration Diagnostic and Reduction Techniques.
Canada, R. G. and Robinson, J. C., 1995, “Vibration Measurements on
Slow Speed Machinery,” in Predictive Maintenance Technology
National Conference (P/PM Technology), Indianapolis, IN, Vol. 8,
No. 6, 33–37.
Catlin, J. B. Jr., 1983, “The Use of Ultrasonic Diagnostic Technique to
Detect Rolling Element Bearing Defects,” in Proceedings of the
Machinery and Vibration Monitoring and Analysis Meeting, Vibra-
tion Institute, IL, April, 123–130.
Choudhary, A. and Tandon, N., 2000, “Application of Acoustic Emission
Technique for the Detection of Defects in Rolling Element Bear-
ings,” Tribology International, Vol. 33, 39–45.
Cudina, M., 2003, “Detection of Cavitation Phenomenon in a Centrifugal
Pump Using Audible Sound,” Mechanical Systems and Signal
Processing, Vol. 17, No. 6, 1335–1347.
Darling, J. and Johnston, D. N., 1991, “The Use of Acoustic Emission for
Condition Monitoring in High-pressure Hydraulic Pumps,” in
EUROTECH Direct 91, IMechE, Birmingham, UK, 2–4 July, Paper
C414/011, 137–143.
Derakhshan, O., Houghton, J. R., and Jones, R. K., 1989, “Cavitation Mon-
itoring of Hydroturbines with RMS Acoustic Emission Measure-
ments,” in World Meeting on Acoustic Emission, Charlotte, NC,
March, 305–315.
El-Ghamry, M. H., Brown, E., Ferguson, I. G., Gill, J. D., Reuben, R. L.,
Steel, J. A., Scaife, M., and Middleton, S., 1998, “Gaseous Air–fuel
Quality Identification for a Spark Ignition Gas Engine Using Acous-
tic Emission Analysis,” in Proceedings of COMADEM 1998,
Monash University, Australia, 235–244.
El-Ghamry, M. H., Reuben, R. L., and Steel, J. A., 2003, “The Develop-
ment of Automated Pattern Recognition and Statistical Feature Isola-
tion Techniques for the Diagnosis of Reciprocating Machinery
Faults Using Acoustic Emission,” Mechanical Systems and Signal
Processing, Vol. 17, No. 4, 805–823.
El-Ghamry, M., Steel, J. A., Reuben, R. L., and Fog, T. L., 2005, “Indirect
Measurement of Cylinder Pressure from Diesel Engines Using
Acoustic Emission,” Mechanical Systems and Signal Processing,
Vol. 19, No. 4, 751–765.
Fan, Y. E., Gu, F., and Ball, A., 2005, “Condition Monitoring of Rolling
Element Bearings Using Advanced Acoustic Emission Signal Anal-
ysis Technique,” in Proceedings of the 18th International Congress
and Exhibition on Condition Monitoring and Diagnostic Engineering
Management (COMADEM 2005), Cranfield, UK, 31 August–2 Sep-
Finley, R. W., 1980, “Computerization Preventive Maintenance Systems
Using Modified Acoustic Emission Techniques,” Materials Evalua-
tion, Vol. 38, No. 8, 15–20.
Fog, T. L., Brown, E. R., Hanson, H. S., Madsen, L. B., Sørensen, P., Han-
sel, E. R., Steel, J. A., Reuben, R. L., and Pedersen, P. S., 1998, in
Proceedings of COMADEM 1998, Monash University, Australia,
Fraser, W. H., 1981, “Flow Recirculation in Centrifugal Pumps,” in Pro-
ceedings of the 10th Annual Turbomachinery Symposium,” Tur-
bomachinery Laboratory, TX, 95–100.
Friis-Hausen, A. and Fog, T. L., 2001, “Monitoring Exhaust Valve Leaks
and Misfire in Marine Diesel Engines,” in Proceedings of
COMADEM 2001, University of Manchester, UK, 4–6 September,
Gill, J. D., Brown, E. R., Twite, M., Horner, G., Reuben, R. L., and Steel, J. A.,
1998, “Monitoring of a Large Reciprocating Compressor,” in Proceed-
ings of COMADEM 1998, Monash University, Australia, 317–326.
Guelich, J. F. and Bolleter, U., 1992, “Pressure Pulsations in Centrifugal
Pumps,” Journal of Vibration and Acoustics, Transactions of the
ASME, Vol. 114, 272–279.
Hanel, V. and Thelen, W., 1995, “Determination of Load Condition of
Standard Screws During Tensile Tests Using Acoustic Emission
Analysis,” in Proceedings of the Instrumentation and Measurement
Technology Conference (IMTC/95), Integrating Intelligent Instru-
mentation and Control, IEEE, 284.
Hanel, V. and Thelen, W., 1996a, “Monitoring Screws Under Tensile Load
Using Acoustic Emission Analysis,” IEEE Transactions on Instru-
mentation and Measurement, Vol. 45, No. 2, 547–550.
Hanel, V. and Thelen, W., 1996b, “Acoustic Emission Testing of Bolted
Connections Under Tensile Stress,” in Proceedings of the 22nd
European Conference on Acoustic Emission Testing,” Robert Gor-
don University, Aberdeen, 29–31 May, 185–188.
Hawman, M. W. and Galinaitis, W. S., 1988, “Acoustic Emission Monitor-
ing of Rolling Element Bearings,” in Proceedings of the IEEE Ultra-
sonics Symposium, 885–889.
Hoeprich, M. R., 1992, “Rolling Element Bearing Fatigue Damage Propa-
gation,” Transactions of ASME, Journal of Tribology, Vol. 114,
Holroyd, T. J., 1993, “Acoustic Emission Monitoring of Slow Moving
Bearings,” in Proceedings of COMADEM 1993, University of West
of England, Bristol, 518.
Holroyd, T. J., 2000, “Acoustic Emission as a Basis for the Condition
Monitoring of Industrial Machinery,” in Proceedings of the 18th
Machinery Vibration Seminar, Canadian Machinery Vibration Asso-
ciation, Halifax, Nova Scotia, Canada, 27–29 April, 125–131.
Holroyd, T., 2001, “Condition Monitoring of Very Slowly Rotating
Machinery Using AE Techniques,” in Proceedings of the 14th Inter-
national Congress on Condition Monitoring and Diagnostic Engi-
neering Management (COMADEM’2001), Manchester, UK, 4–6
September, 29.
Holroyd, T. J., 2002, “Simplifying the Condition Monitoring of Machines
and Structures Using Acoustic Emission,” in Proceedings of
COMADEM 2002, COMADEM International, Birmingham, UK,
Holroyd, T. J. and Brashaw, C., 1999, “A Critical Appraisal of Monitoring
Elastic Waves as a Means of Detecting and Diagnosing Machine
Faults,” in Proceedings of COMADEM 1999, Sunderland, UK, 491–
Holroyd, T. J. and Randall, N., 1993a, “Use of Acoustic Emission for
Machine Condition Monitoring,” British Journal of Non-Destructive
Testing, Vol. 35, No. 2, 75–78.
Holroyd, T. J. and Randall, N., 1993b, “Field Application of Acoustic
Emission to Machinery Condition Monitoring,” in Proceedings of
COMADEM 1993, University of West of England, Bristol, 217–
Holroyd, T. J., Randall, N., and Lin, D., 1996, “Acoustic Emission as a
Means of Dynamically Probing the Operation of Machines and
Mechanisms,” in Proceedings of COMADEM 1996, Sheffield, UK,
Huguet, S., Godin, N., Gaertner, R., Salmon, L., and Villard, D., 2002,
“Use of Acoustic Emission to Identify Damage Modes in Glass Fiber
Reinforced Polyester,” Composites Science and Technology, Vol.
62, 1433–1444.
Hutton, P. H., 1969, “Detecting Acoustic Emission in the Presence of
Hydraulic Noise,” Non-Destructive Testing.
Jamaludin, N., Mba, D., and Bannister, R. H., 2001, “Condition Monitor-
ing of Slow-speed Rolling Element Bearings Using Stress Waves,”
Journal of Process Mechanical Engineering, Vol. 215E, No. E4,
Jamaludin, N., Jailani, M., and Nor, M., 2002, “Lubrication Monitoring in
a Low Speed Bearing Using Acoustic Emission Technique,” in Pro-
ceedings of COMADEM 2002, COMADEM International, Birming-
ham, UK, 521–527.
Javed, M. and Littlefair, G., 1993, “Neural Networks Based System for
Rotating Machinery,” in Proceedings of COMADEM 1993, Univer-
sity of West of England, Bristol, 418–423.
Kaewkongka, T. and Au, Y. H. J., 2001, “Application of Acoustic Emis-
sion to Condition Monitoring of Rolling Element Bearings,” Meas-
urement and Control, Vol. 34, No. 8, 254–247.
Kakishima, H., Nagatomo, T., Ikeda, H., Yoshioka, T., and Korenaga, A.,
2000, “Measurement of Acoustic Emission and Vibration of Rolling
Bearings with an Artificial Defect,” Quarterly Review of Railway
Technical Research Institute, Vol. 41, No. 3,127–130.
Kuboyama, K., 1987, “Development of Low Speed Bearing Diagnosis
Technique,” in Proceedings of the 41st Meeting of the Society for
Machinery Failure Prevention Technology, NKK Fukuyama Works,
Fukuyama City, Hiroshima, Japan.
Li, C. J. and Li, S. Y., 1985, “Acoustic Emission Analysis for Bearing
Condition Monitoring,” Wear, Vol. 185, 67–74.
Li, Y., Billington, S., Zhang, C., Kurfess, T., and Danyluk, S., 1998,
“Dynamic Prognostic Prediction of Defect Propagation on Rolling
Element Bearings,” in ASME/STLE Tribology Conference, Toronto,
Ontario, Canada, 26–28 October, 385–392.
McFadden, P. D. and Smith, J. D., 1983, “Acoustic Emission Tranducers
for the Vibration Monitoring of Bearings at Low Speeds,” Report no.
McNulty, P. J., 1981, “Measurement Techniques and Analysis of Fluid-
borne Noise in Pumps,” National Engineering Laboratory, NEL
Report No. 674.
McNulty, P. J. and Deeprose, W. M., 1978, “Scaling of Non-cavitating and
Cavitating Pump Noise Acoustic for Pumps,” Risley, UK, Paper 3,
McNulty, P. J. and Pearsall, I. S., 1962, “Cavitation Inception in Pumps,”
Journal of Fluids Engineering, Vol. 104, 99–104.
Mathews, J. R., 1983, Acoustic Emission, Gordon and Breach, New York.
Mba, D., 2002, “Applicability of Acoustic Emissions to Monitoring the
Mechanical Integrity of Bolted Structures in Low Speed Rotating
Machinery: Case Study,” NDT and E International, Vol. 35, No. 5,
Mba, D. and Hall, L., 2001, “Monitoring the Integrity of Low Speed Rotat-
ing Machines,” in Proceedings of COMADEM 2001, University of
Manchester, UK, 4–6 September, 697–707.
Mba, D. U., Bannister, R. H., Leslie, J. T., and Findlay, G. E., 1996, “Con-
dition Monitoring of Low Speed Rotating Machinery,” in Proceed-
ings of COMADEM 1996, Sheffield, UK, 73–84.
Miettinen, J. and Andersson, P., 2000, “Acoustic Emission of Rolling
Bearings Lubricated With Contaminated Grease,” Tribology Interna-
tional, Vol. 33, No. 11, 743–802.
Miettinen, J. and Pataniitty, P., 1999, “Acoustic Emission in Monitoring
Extremely Slowly Rotating Rolling Bearing,” in Proceedings of
COMADEM 1999, Sunderland, UK, 289–297.
Miettinen, J. and Salmenperä, P., 2000, “Acoustic Emission Monitoring of
Grease Lubricated Rolling Bearings,” in Proceedings of
COMADEM 2000, Houston, TX, 21–30.
Miyachika, K., Oda, S., and Koide, T., 1995, “Acoustic Emission of Bend-
ing Fatigue Process of Spur Gear Teeth,” Journal of Acoustic Emis-
sion, Vol. 13, No. 1/2, S47–S53.
Miyachika, K., Zheng, Y., Tsubokura, K., Oda, S., Kanayama, Y., Koide,
T., Namba, C., and Hayashi, T., 2002, “Acoustic Emission of Bend-
ing Fatigue Process of Supercarburized Spur Gear Teeth,” Progress
in Acoustic Emission XI, The Japanese Society for NDI, 304–310.
Miller, R. K. and McIntire, P., editors, 1987, Non-Destructive Testing Hand-
book: Acoustic Emission Testing, Vol. 5, ASM, Materials Park, OH.
Morhain, A. and Mba, D., 2002, “Defect Diagnosis of Lightly Loaded
Bearings with Acoustic Emissions,” in Proceedings of COMADEM
2002, COMADEM International, Birmingham, UK, 492–502.
Morhain, A. and Mba, D., 2003, “Bearing Defect Diagnosis and Acoustic
Emission,” Journal of Engineering Tribology, I Mech E, Vol. 217,
No. 4, Part J, 257–272.
Murphy, T. J., 1992, “The Development of a Data Collector for Low-speed
Machinery,” in Proceedings of the 4th International Conference on
Profitable Condition Monitoring, bHr Group Ltd, Stratford-upon-
Avon, UK, 8–10 December, 251–258.
Neill, G. D., Reuben, R. L., Sandford, P. M., Brown, E. R., and Steel, J. A.,
1996a, “Detection of Incipient Cavitation in Pumps Using Acoustic
Emission,” in Proceedings of COMADEM 1996, Sheffield Univer-
sity, UK, 391–401.
Neill, G. D., Reuben, R. L., and Sandford, P. M., 1997, “Detection of
Incipient Cavitation in Pumps Using Acoustic Emission,” Journal of
Process Mechanical Engineering, Vol. 211, No. E4, 267.
Neill, G. D., Benzie, S., Gill, J. D., Sandford, P. M., Brown, E. R., Steel, J.
A., and Reuben, R. L., 1998a, “The Relative Merits of Acoustic
Emission and Acceleration Monitoring for the Detection of Bearing
Defects,” in Proceedings of COMADEM 1998, Monash University,
Australia, 651–659.
Neill, G. D., Brown, E. R., Reuben, R. L., Sandford, P. M., and Steel, J. A.,
1998b, “Detection of Recirculation in Pumps Using Acoustic Emis-
sion,” in Proceedings of COMADEM 1998, Monash University,
Australia, 643–649.
Pao, Y-H., Gajewski, R. R., and Ceranoglu, A. N., 1979, “Acoustic Emis-
sion and Transient Waves in an Elastic Plate,” Journal of the Acous-
tical Society of America, Vol. 65, No. 1, 96–102.
Parikka, R., Vidqvist, V., Vaajoensuu, E., and Tervo, J., 2002, “Test
Equipment and Diagnostic System for Determining Lubrication
Condition and Running Clearance of Rolling Bearings,” in Proceed-
ings of COMADEM 2002, COMADEM International, Birmingham,
UK, 153–162.
14 The Shock and Vibration Digest / September 2005
Pollock, A. A., 1989, “Acoustic Emission Inspection,” Physical Acoustics
Corporation, Technical Report, TR-103-96-12/89.
Price, E. D., Lees, A. W., and Friswell, M. I., 2001, “Application of High
Frequency Monitoring for Classification of Rolling Element Bearing
Failures,” Key Engineering Materials, Vol. 204–205, 173–182.
Price, E. D., Lees, A. W., and Friswell, M. I., 2005, “Detection of Severe
Sliding and Pitting Fatigue Wear Regimes Through the Use of
Broad-band Acoustic Emission,” Proceedings of the I MECH E,
Part J, Journal of Engineering Tribology, Vol. 219, No. 2, 85–98.
Raad, A., Zhang, F., Randall, B., and Sidahmed, M., 2003, “On the Com-
parison of the Use of AE and Vibration Analysis for Early Gear
Fault Detection,” in Proceedings of the 8th Western Pacific Acous-
tics Conference, Melbourne, Australia, 7–9 April.
Robinson, J. C., Canada, R. G., and Piety, R. G., 1996, “Vibration Monitor-
ing on Slow Speed Machinery: New Methodologies Covering
Machinery from 0.5 to 600 rpm,” in Proceedings of the 5th Interna-
tional Conference on Profitable Condition Monitoring – Fluids and
Machinery Performance Monitoring, bHr Group Ltd, Publication 22,
Harrogate, UK, 169–182.
Rogers, L. M., 1979, “The Application of Vibration Analysis and Acoustic
Emission Source Location to On-line Condition Monitoring of Anti-
friction Bearings,” Tribology International, Vol. 12, No. 2, 51–59.
Salvan, S. M. E., Parkin, R. M., Coy, J., and Li, W., 2001, “Intelligent Con-
dition Monitoring of Bearings in Mail Processing Machines using
Acoustic Emission,” in Proceedings of COMADEM 2001, Univer-
sity of Manchester, UK, 4–6 September, 67–74.
Sato, I., 1990, “Rotating Machinery Diagnosis With Acoustic Emission
Techniques,” Electrical Engineering in Japan, Vol. 10, No. 2,115–
Schoess, J. N., 2000, “Development and Application of Stress-wave
Acoustic Diagnostics for Roller Bearings,” Proceedings of the SPIE,
Vol. 3986, 58–70.
Sentoku, H., 1998, “AE in Tooth Surface Failure Process of Spur Gears,”
Progress in Acoustic Emission IX, AEWG & AE Group, S19–S24.
Shiroishi, J., Li, Y., Liang, L. S., Kurfess, T., and Danyluk, S., 1997,
“Bearing Condition via Vibration and Acoustics Emission Measure-
ments,” Mechanical Systems and Signal Processing, Vol. 11, No. 5,
Sikorska, J. Z. and Hodkiewicz, M., 2005, “Comparison of Acoustic Emis-
sion, Vibration and Dynamic Pressure Measurements for Detecting
Change in Flow Conditions on a Centrifugal Pump,” Proceedings of
the 18th International Congress and Exhibition on Condition Moni-
toring and Diagnostic Engineering Management (COMADEM
2005), Cranfield, UK, 31 August–2 September.
Singh, A., Houser, D. R., and Vijayakar, S., 1996, “Early Detection of Gear
Pitting,” in Power Transmission and Gearing Conference, San
Diego, CA, ASME, Vol. 88, 673–678.
Singh, A., Houser, D. R., and Vijayakar, S., 1999, “Detecting Gear Tooth
Breakage Using Acoustic Emission: A Feasibility and Sensor Place-
ment Study,” Journal of Mechanical Design, Vol. 121, 587–593.
Siores, E. and Negro, A. A., 1997, “Condition Monitoring of a Gear Box
Using Acoustic Emission Testing,” Material Evaluation, Vol. 55,
No. 2, 183–187.
Smith, J. D., 1982, “Vibration Monitoring of Bearings at Low Speeds,”
Tribiology International, Vol. 15, No. 3, 139–144.
Smulders, A. and Loob, C., 1994, “Machine Condition Monitoring Using
Multiparameter Measurements,” in Proceedings of COMADEM
1994, Delhi, India, 147–152.
Steel, J. A. and Reuben, R. L., 2005, “Recent Developments in Monitoring
of Engines Using Acoustic Emission,” Journal of Strain Analysis for
Engineering Design, Vol. 40, No. 1, 45–57.
Sturm, A. and Uhlemann, S., 1985, “Diagnosis of Plain Bearings by
Acoustic Emission Analysis,” Measurements, Vol. 3, No. 4, 185–
Sturm, A., Ellmer, G., and Billhardt, S., 1992, “Analysis of High-frequency
Acoustic Signals of Machines with Rolling Element Bearings and
Sliding Bearings,” 184–190.
Sundt, P. C., 1979, “Monitoring Acoustic Emission to Detect Mechanical
Defects,” InTech, December, 43–44.
Tan, C. C., 1990, “Application of Acoustic Emission to the Detection of
Bearing Failures,” in the Institution of Engineers Australia, Tribol-
ogy Conference, Brisbane, Australia, 3–5 December, 110–114.
Tan, C. K. and Mba, D., 2004a, “Acoustic Emission Source at the Gear
Mesh,” in Proceedings of the 17th International Congress on Condi-
tion Monitoring and Diagnostic Engineering Management
(COMADEM), Cambridge, UK, August, 145–151.
Tan, C. K. and Mba, D., 2004b, “The Source of Acoustic Emission During
Meshing of Spur Gears,” in Proceedings of the 26th European Con-
ference on Acoustic Emission Testing (EWGAE), Berlin, Germany,
15–17 September, 559–565.
Tan, C. K. and Mba, D., 2005a, “Identification of the Acoustic Emission
Source During a Comparative Study on Diagnosis of a Spur Gear-
box, Tribology International, Vol. 38, No. 5, 469–480.
Tan, C. K. and Mba, D., 2005b, “Limitation of Acoustic Emission for Iden-
tifying Seeded Defects in Gearboxes,” Journal of Non-Destructive
Evaluation, Vol. 24, No. 2.
Tan, C. K. and Mba, D., 2005c, “Correlation Between Acoustic Emission
Activity and Asperity Contact During Meshing of Spur Gears Under
Partial Elastohydrodynamic Lubrication,” Tribology Letters, in
Tan, C. K., Irving, P., and Mba, D., 2004, “Opportunities Offered by Acous-
tic Emission for Prognosis of Spur Gears,” in Proceedings of the 17th
International Congress on Condition Monitoring and Diagnostic Engi-
neering Management (COMADEM),” Cambridge, UK, 152–163.
Tan, C. K., Irving, P., and Mba, D., 2005a, “A Comparative Experimental
Study on the Diagnostic and Prognostic Capabilities of Acoustics
Emission, Vibration and Spectrometric Oil Analysis for Spur
Gears,” Mechanical Systems and Signal Processing, submitted.
Tan, C. K., Irving, P., and Mba, D., 2005b, “Prognostic Potential of the
Acoustic Emission Technique for Spur Gears. Essential Technolo-
gies for Successful Prognostics,” in Proceedings of the 59th Meeting
of the Society for Machinery Failure Prevention Technology
(MFPT), Virginia Beach, VA, 18–21 April, 67–76.
Tan, C. K., Irving, P., and Mba, D., 2005c, “Diagnostics and Prognostics
with Acoustic Emission, Vibration and Spectrometric Oil Analysis
for Spur Gears: A Comparative Study,” INSIGHT, Vol. 47, No. 8.
Tandon, N. and Mata, S., 1999, “Detection of Defects in Gears by Acoustic
Emission Measurements,” Journal of Acoustic Emission, Vol. 17,
No. 1–2, 23–27.
Tandon, N. and Nakra, B. C., 1990, “Defect Detection of Rolling Element
Bearings by Acoustic Emission Method,” Journal of Acoustic Emis-
sion, Vol. 9, No. 1, 25–28.
Tavakoli, M. S., 1991, “Review of Bearing Monitoring – Application of
Acoustic Emission,” in Proceedings of the 1st International Confer-
ence on Acoustic Emission in Manufacturing, S. J. Vahaviolos, edi-
tor, Boston, MA, 16–19 September.
Toutountzakis, T. and Mba, D., 2003, “Observation of Acoustic Emission
Activity During Gear Defect Diagnosis,” NDT and E International,
Vol. 36, No. 7, 471–477.
Viktorov. I. A., 1967, Rayleigh and Lamb Waves: Physical Theory and
Application, Plenum Press, New York.
Wheitner, J., Houser, D., and Blazakis, C., 1993, “Gear Tooth Bending
Fatigue Crack Detection by Acoustic Emissions and Tooth Compli-
ance,” ASME Technical Paper 93FTM9, 1–7.
Williams, T., Ribadeneira, X., Billington, S., and Kurfess, T., 2001, “Roll-
ing Element Bearing Diagnosis in Run-to-failure Lifetime Testing,”
Mechanical Systems and Signal Processing, Vol 15, No. 5, 979–993.
Yoshioka, T. and Fujiwara, T., ?, “New Acoustic Emission Source Locat-
ing System for the Study of Rolling Contact Fatigue,” Wear, Vol. 81,
No. 1, 183–186.
Yoshioka, T. and Fujiwara, T., 1984, “Application of Acoustic Emission
Technique to Detection of Rolling Bearing Failure,” American Soci-
ety of Mechanical Engineers, Production Engineering Division Pub-
lication, Vol. 14, 55–76.
Yoshioka, T., Korenaga, A., Mano, H., and Yamamoto, T., 1999, “Diagno-
sis of Rolling Bearing by Measuring Time Interval of Acoustic
Emission Generation,” Journal of Tribology, Transactions of ASME,
Vol. 121, 468–472.
... Over the past years, acoustic emission (AE) sensors have been used in bearing condition monitoring and fault diagnosis [4]. AE signals can capture bearing fault features and detect incipient faults [5,6]. Detection is performed using the frequencies of selected components as key features. ...
Full-text available
The demand for the condition monitoring of induction motors is increasing in various fields, such as industry, transportation, and daily life. Bearing faults are the most common faults, and many fault diagnosis methods have been proposed using artificial pitting as the fault factor in most cases. However, the validity of a fault diagnosis method for other kinds of faults does not seem to be evaluated. Considering onsite scenarios and other possibilities of faults, this paper introduces scratches on the outer raceways of bearings. A study was performed on the detection of several kinds of bearing scratches using a proposed method that was based on an auto-tuning convolutional neural network. The developed approach was also compared with other diagnostic methods for validation. The results showed that the proposed technique provides the possibility of diagnosing several kinds of scratches with acceptable accuracy rates.
... Preemptive measure and real time monitoring of these signals can avoid severe losses and disastrous failures [7,8] and hence has received considerable attention [9]. Different methods of condition based monitoring are being developed and used, including but not limited to oil debris, vibration, acoustic emission, electrostatic and temperature analysis [10][11][12][13]. ...
Full-text available
In condition based maintenance, different signal processing techniques are used to sense the faults through the vibration and acoustic emission signals, received from the machinery. These signal processing approaches mostly utilise time, frequency, and time-frequency domain analysis. The features obtained are later integrated with the different machine learning techniques to classify the faults into different categories. In this work, different statistical features of vibration signals in time and frequency domains are studied for the detection and localisation of faults in the roller bearings. These are later classified into healthy, outer race fault, inner race fault, and ball fault classes. The statistical features including skewness, kurtosis, average and root mean square values of time domain vibration signals are considered. These features are extracted from the second derivative of the time domain vibration signals and power spectral density of vibration signals. The vibration signal is also converted to the frequency domain and the same features are extracted. All three feature sets are concatenated, creating the time, frequency and spectral power domain feature vectors. These feature vectors are finally fed into the K- nearest neighbour, support vector machine and kernel linear discriminant analysis for the detection and classification of bearing faults. With the proposed method, the reduction percentage of more than 95% percent is achieved, which not only reduces the computational burden but also the classification time. Simulation results show that the signals are classified to achieve an average accuracy of 99.13% using KLDA and 96.64% using KNN classifiers. The results are also compared with the empirical mode decomposition (EMD) features and Fourier transform features without extracting any statistical information, which are two of the most widely used approaches in the literature. To gain a certain level of confidence in the classification results, a detailed statistical analysis is also provided.
... The first part of Table 8 reports several relevant studies of the last 20 years for CM. A comprehensive review of previous works about the AE-based structural diagnosis of bearing defects, gearbox faults, and pumps can be found in Mba & Rao [255]. ...
Full-text available
A complete surveillance strategy for wind turbines requires both the condition monitoring (CM) of their mechanical components and the structural health monitoring (SHM) of their load-bearing structural elements (foundations, tower, and blades). Therefore, it spans both the civil and mechanical engineering fields. Several traditional and advanced non-destructive techniques (NDTs) have been proposed for both areas of application throughout the last years. These include visual inspection (VI), acoustic emissions (AEs), ultrasonic testing (UT), infrared thermography (IRT), radiographic testing (RT), electromagnetic testing (ET), oil monitoring, and many other methods. These NDTs can be performed by human personnel, robots, or unmanned aerial vehicles (UAVs); they can also be applied both for isolated wind turbines or systematically for whole onshore or offshore wind farms. These non-destructive approaches have been extensively reviewed here; more than 300 scientific articles, technical reports, and other documents are included in this review, encompassing all the main aspects of these survey strategies. Particular attention was dedicated to the latest developments in the last two decades (2000–2021). Highly influential research works, which received major attention from the scientific community, are highlighted and commented upon. Furthermore, for each strategy, a selection of relevant applications is reported by way of example, including newer and less developed strategies as well.
... Since then, there has been an explosion in research-and application-based studies covering bearings, pumps, gearboxes, engines, and rotating structures. Mba and Rao (2006) presented a comprehensive and critical review on the application of AET to condition monitoring and diagnostics of rotating machinery. ...
Conference Paper
Full-text available
A brief history of rotor dynamics field has been documented in the present review paper. It reviews early development of simple rotor models starting from the Rankine to Jeffcott rotor models and physical interpretations of various kind of instabilities in rotor-bearing systems. It also reviews developments of analysis methods for the continuous and multi-degrees-of freedom systems to allow practicing engineers to apply these methods to real turbo-machineries. The paper also summaries work on conditioning monitoring and the recent trends in the area of rotor dynamics. 1. Definition: A rotor is a body suspended through a set of cylindrical hinges or bearings that allow it to rotate freely about an axis fixed in space. Engineering components concerned with the subject of rotor dynamics are rotors of machines, especially of turbines, generators, motors, compressors, blowers and the like. The parts of the machine that do not rotate are referred to with general definition of stator. Rotors of machines have, while in operation, a great deal of rotational energy, and a small amount of vibrational energy. It is very evident from the fact that a relatively small turbine propels a huge aircraft. The purpose of rotor dynamics as a subject is to keep the vibrational energy as small as possible. In operation rotors undergoes the bending, axial and torsional vibrations. 2. From Rakine to Jeffcott Rotor Models: Rotor dynamics has a remarkable history of developments, largely due to the interplay between its theory and its practice. Rotor dynamics has been driven more by its practice than by its theory. This statement is particularly relevant to the early history of rotor dynamics. Research on rotor dynamics spans at least a 135-year history. Rankine (1869) performed the first analysis of a spinning shaft. He chose a two-degrees-of-freedom model consisted of a rigid mass whirling in a circular orbit, with an elastic spring acting in the radial diection. He predicted that beyond a certain spin speed ".. . the shaft is considerably bent and whirls around in this bent form." He defined this certain speed as the whirling speed of the shaft. In fact, it can be shown that beyond this whirling speed the radial deflection of Rankine's model increases without limit. Today, this speed would be called the threshold speed for divergent instability. However, Rankine did add the term whirling to the rotor dynamics vocabulary. Whirling refers to the movement of the center of mass of the rotor in a plane perpendicular to the shaft. The frequnecy of whirl depends on the stiffness and damping of the rotor and the amplitude is a function of the excitation force's frequency and magnitude. A crtical speed occurs when the excitation frequency coincides with a natural frequency, and can lead to excessive vibration amplitudes. Rankine's neglect of Coriolis acceleration led to erroneous conclusions that confused engineers for one-half century. The turbine built by Parsons in 1884 (Parsons, 1948) operated at speeds of around 18000 rpm, which was fifty times faster than the existing reciprocating engine. In 1883 Swedish engineer de Laval developed a single-stage steam impulse turbine (named after him) for marine applications and succeeded in its operation at 42000 rpm. He aimed at the self-centering of the disc above the critical speed, a phenomenon which he intuitively recognized. He first used a rigid rotor, but latter used a flexible rotor and showed that it was possible to operate above critical speed by operating at a rotational speed about seven times the critical speed (Stodola, 1924). It thus became recognized that a shaft has several critical speeds and that under certain circumstances these were the same as natural frequencies of a non-rotating shaft. In order to calculate the critical speeds of cylindrical shafts with several discs and bearings the general theory of Reynolds (Dunkerley, 1895) was applied. The gyroscopic effect was also considered, together with its dependence on speed. The required solution of the frequency equation was at that time only possible for simple models.
... The condition monitoring (Davies, 1998;Williams, et al., 1994;Collacott, 1977) is the use of advanced technologies in order to determine equipment condition, and then potentially predict failures. It includes, but is not limited to, technologies such as (i) the vibration measurement and analysis (Rao, 2007;Mitchell, 1993), (ii) the acoustic emission technology (Mba and Rao, 2006), (iii) tribology e.g. the oil/wear debris analysis/ferrography (Newell, 1999), (iv) the motor-current analysis (Nandi et al. 2005;Acosta, et al., 2006), (v) Infra-red thermography, (vi) Ultrasonics, etc. Of the techniques available, vibration monitoring is the most widely used technique in industry today. ...
Conference Paper
Full-text available
The omnipresence of gears as critical elements in complex rotating machinery has made the study of vibration a more interesting subject. One of the major repercussions of gear faults is the excessive vibration, which eventually transmits to foundations through bearings. Through gear vibration analyses, a lot of features are acquired, and the next step is the classification. The task of classification is to find a rule, which, based on external observations, assigns an object to one of several classes. A lot of techniques have been proposed to carry out these tasks. Some of them are artificial neural networks, genetic programming, and very recently by support vector machines. Traditional neural network and genetic programming approaches have suffered difficulties with generalization, producing models that can over-fit the data. As a powerful machine learning approach for classification problems, support vector machine is identified to have good generalization ability. An approach is proposed based on the support vector machine technique to detect and classify multiple gear-fault conditions, which can be further generalized to the complete health monitoring of machines. In this work, samples of vibration signals of gearbox casing are recorded under various monitoring conditions and formulated a procedure for classifying expected/predicted data. The specific characteristics of the vibration spectrum that are associated with common gear damage conditions are well known. Two approaches for feature extractions are adopted for analyses in frequency domain. The first method involved peak extractions from the vibration signatures, and the second one was based on obtaining statistical characteristics of the same such as skewness, kurtosis, RMS, etc. It is observed that the second method gave better classification on tuning the ratio of the numbers of training and test data sets to an optimum level. The effectiveness of this method of feature extractions are applied to time domain analysis as well. Cases of 0% errors during classifications are found in 2 out of 3 instances. Overall classification efficiency of SVMs is compared with and is found to be better than that of other learning algorithms.
The pump is an essential technical device used in almost all major sectors. For the undisturbed running of different sectors, the failure of the pumping system should be prevented. The failure of the pump causes loss of production, therefore, loss of revenue. These failures should be detected at an early stage to avoid catastrophic failure. Many traditional methods, such as vibration analysis and motor current signature analysis, can detect the fault after the failure occurs, but artificial intelligence (AI)-based techniques can identify the failures more efficiently to predict the fault early. Among various AI-based methods, machine learning (ML) and deep learning (DL)-based methods are the most useful techniques with widespread applications. This paper presents a comprehensive review of studies on various faults in centrifugal pumps and their identification by various traditional and ML-based techniques.
Structural health monitoring and assessment (SHMA) is exceptionally essential for preserving and sustaining any mechanical structure’s service life. A successful assessment should provide reliable and resolute information to maintain the continuous performance of the structure. This information can effectively determine crack progression and its overall impact on the structural operation. However, the available sensing techniques and methods for performing SHMA generate raw measurements that require significant data processing before making any valuable predictions. Machine learning (ML) algorithms (supervised and unsupervised learning) have been extensively used for such data processing. These algorithms extract damage-sensitive features from the raw data to identify structural conditions and performance. As per the available published literature, the extraction of these features has been quite random and used by academic researchers without a suitability justification. In this paper, a comprehensive literature review is performed to emphasise the influence of damage-sensitive features on ML algorithms. The selection and suitability of these features are critically reviewed while processing raw data obtained from different materials (metals, composites and polymers). It has been found that an accurate crack prediction is only possible if the selection of damage-sensitive features and ML algorithms is performed based on available raw data and structure material type. This paper also highlights the current challenges and limitations during the mentioned sections.
In recent years, bearing fault diagnosis has been a research hotspot. In order to improve the reliability of acoustic fault diagnosis, this paper combines Cross Wavelet Transform (XWT) and complementary ensemble empirical mode decomposition (CEEMD) to extract bearing fault features from acoustic signals. Finally, the time-domain features and spectral centroid are input into the SVM for fault classification. The results show that the proposed method can effectively improve the reliability of acoustic fault diagnosis.
Fault detection is a critical step for machine condition monitoring and maintenance. With advances in machine learning technologies, automated faulty condition identification can be achieved by training an artificial intelligent (AI) model if sufficient data is available. In most practical applications, it is unlikely that enough data from faulty cases are available for supervised training of AI models for anomaly detection. This work is based on training data that are exclusively constituted of healthy signals (i.e., semi-supervised) and aims to develop an automated algorithm capable of identifying any abnormal mechanical behaviour captured by vibration measurements. A deep learning method with long short-term memory (LSTM) architectures combined with a one-class support vector machine (SVM) is used to separate abnormal data from normal vibration signals collected during an endurance test of a reduction gearbox and helicopter test flight data measured by multiple sensors situated at different locations of the aircraft. For the gearbox dataset, a detailed understanding of the physical mechanisms by which different types of faults (gear wear and bearing faults) affect the vibration signal led to the design of two anomaly detection architectures: i) an LSTM regression + one-class SVM for detecting new deterministic components introduced by gear faults; and ii) a two-step LSTM regression + one-class SVM for detecting new random components caused by bearing failures. For the helicopter dataset, which does not contain consecutive time-series, we show that the LSTM regression is not advantageous, and a better performance can be achieved by a simpler one-class SVM outlier detection based on statistical features. This work contributes to the field of machine condition monitoring by introducing a novel two-step LSTM configuration for removing deterministic components associated with dominant gear signals in the first step and removing the ‘residual deterministic’ components associated with varying gear signals in the second step.
Although acoustic emission testing has traditionally been used to detect flaws in structures, it has also been shown to be effective for monitoring rotating machinery. Large-scale systems based on modified acoustic emission monitoring are now proving the value of quantifying and tracking machine ″noise″ as a leading indicator for preventive maintenance. The design concepts and operation of these systems are discussed, and a number of case history examples are presented to illustrate this application of acoustic monitoring.
Predictive maintenance, or PDM, using vibration analysis is a technique which has been available for many years. Initially PDM gained most usage in the Power Generation and Petrochemical industries. Generally speaking, in those industries, machines rotate at speeds above 1200 rpm and rarely much below 600rpm.
This paper investigates the condition of gearbox bearings using stress wave sensors which respond to very high frequencies. Accelerameter measurements were analysed in parallel to the stress wave sensors in order to provide comparison between the two methods. The bearings were first examined individually on a special test rig, allowing an assessment in isolation away from the contaminating gearbox noise. Various ‘intentional’ faults were introduced and changes in the signature examined, using techniques based in the time and frequency domains, and thus enabling detection of such faults. The bearings were then fitted into a gearbox and special detection techniques were used for extracting their signatures from the gearbox noise. All the ‘intentional’ faults were identified and fault diagnosis using the stress wave sensor method is hence assessed.
The technology of Acoustic Emission (AE) is based upon the detection of naturally generated high-frequency elastic waves within materials and is most widely known as a non-destructive testing technique having an ability to globally monitor structures. In this role, AE is widely used in the aerospace and petrochemical industries for testing critical structures. In addition to those AE sources associated with defect growth, AE sensors are also extremely sensitive to a plethora of other energy loss mechanisms such as impacts, friction, turbulence and cavitation. Since these are precisely the energy loss mechanisms which are associated with the degradation of the mechanical condition of a machine, it follows that AE has great potential as a Condition Monitoring (CM) tool. In particular, the high signal-to-noise ratio (SNR) which AE offers, compared to lower frequency vibration techniques, opens the way for a more direct measure of machine faults. However, past research in this field has been dogged by the variability in the response of AE detection systems and fragmented by the diversity of signal processing techniques employed. These factors have restricted the evolution of AE as a CM tool from the research laboratory to the industrial shop-floor. Recent developments in AE instrumentation have resulted in very simple but effective means of carrying out such monitoring in the industrial environment. In this paper, some of these developments are briefly described and the application of AE as a CM tool is illustrated by way of actual examples.
Although high-frequency (Acoustic Emission or AE) and low frequency (Vibration Monitoring or VM) techniques are both widely used in industry for Condition Monitoring (CM) purposes, there are a wide range of opinions held on the similarities and differences of the two techniques as well as their relative merits. This paper seeks to clarify these issues by way of a brief review of the subject of monitoring the elastic wave activity generated in machinery in order to ascertain information on machine condition. The various arguments are simply presented and illustrated where appropriate by laboratory data.
Acoustic emissions can be used to indicate mechanical faults before problems have developed to the point where increased vibration or temperature levels can be detected. Signal conditioning is needed to compensate for normal acoustic noise and discriminate contributions to the characteristic frequency spectrum caused by defects.
An overview of the detection of machinery distress and degradation through monitoring the high frequency content of the accompanying acoustic signature.