LASERNET FINES WEAR DEBRIS ANALYSIS TECHNOLOGY:
APPLICATION TO MECHANICAL FAULT DETECTION
J. Reintjes1, J. E. Tucker1, S. E. Thomas2, A. Schultz1, L. L Tankersley3, C. Lu4,
P. L. Howard5, T. Sebok6 and C. Holloway6
1
Naval Research Laboratory, Washington, DC USA
Titan Corp, Crystal City, VA USA
US Naval Academy, Annapolis MD USA
Towson University, Towson, MD USA
5
PL Howard Enterprises, Newmarket, NH USA
6
Lockheed Martin NESS, Akron, OH USA
2
3
ABSTRACT. The operation of LaserNet Fines wear debris analysis technology is described.
Application to detection and identification of mechanical faults in diesel engines is presented. The
development of fault indicators directly from the LaserNet Fines analysis is described and results are
related to controlled laboratory experiments.
INTRODUCTION
Monitoring of wear debris in the lubricating oil of rotating and reciprocating
machinery is a well-established method of non destructive testing that has been used for
many decades to give machinery operators a measure of the state of wear in engines,
transmissions and drive trains, along with early warning of impending catastrophic
failures. Many laboratory analysis tests have been developed to allow assessment of
various wear debris characteristics in terms of machinery condition. Among the various
laboratory analyses conducted are tests to determine particle concentration, elemental
content, elemental composition, size and various shape and appearance characteristics.
Generally speaking, these tests can be time consuming, expensive and require human
expert interpretation. They also usually involve drawing of a sample of oil from the
equipment and sending it to a remote laboratory for analysis, a procedure that can involve
significant delay from the time the sample is drawn to the time results are received by the
operator. However, some of these tests, especially those involving microscopic
examination [1-7], can often give an excellent analysis of the onset of wear processes that
can ultimately lead to machinery failure. More immediate results can be obtained from online monitors, but on line detectors such as chip detectors and inductive monitors give
limited analysis of particle properties and can fail to give reliable early warnings of the
onset of excess wear conditions.
In this paper we describe the LaserNet Fines (LNF) wear particle analysis
technology [8-11]. This technology was developed for on line and on site analysis of wear
debris with the goal of providing more immediate, reliable and earlier detection of
mechanical faults without the need for expert human intervention. LNF is an optically
based wear particle analyzer that determines type, severity and rate of progression of
CP657, Review of Quantitative Nondestructive Evaluation Vol. 22, ed. by D. O. Thompson and D. E. Chimenti
2003 American Institute of Physics 0-7354-0117-9
1590
mechanical faults by measuring the size distribution, concentration and shape
characteristics of the wear particles. In addition to mechanical wear particles, LNF detects
andmechanical
identifies fiber,
oxide andthefreesize
water
concentration.
This capability
allows
faults abrasive
by measuring
distribution,
concentration
and shape
assessment
of
system
cleanliness
and
the
effects
on
system
health
of
specific
characteristics of the wear particles. In addition to mechanical wear particles, LNF detects
contamination
as theoxide
effect
offree
sandwater
on gear
wear and failure,
or the presence
and identifiestypes
fiber,(such
abrasive
and
concentration.
This capability
allowsof
filter
deterioration).
LNF has
similarities
analysis
in that it
assessment
of system
cleanliness
andto ferrography
the effects and
on microscopic
system health
of specific
contamination
types
(suchcharacteristics
as the effect ofof
sand
on gear wear
and failure,
or thebeyond
presencethese
of
makes
use of the
shape
individual
particles,
but goes
filter deterioration).
LNFquantitative
has similarities
ferrography
andcan
microscopic
thatonly
it
techniques
by developing
faulttoindicators
that
be used toanalysis
identifyinnot
use ofofa the
particles,
but goesapplicable
beyond these
themakes
presence
faultshape
but itscharacteristics
type and levelofofindividual
progression.
It is a broadly
debris
techniquestechnology
by developing
can be used
identify
not wetted
only
monitoring
that quantitative
can be usedfault
notindicators
only for that
the detection
oftofaults
in oil
the presence
a fault
but its
type and levelofofparticulate
progression.
It is a broadlylevels
applicable
debris
machinery
but of
also
for the
determination
contamination
in hydraulic
monitoring
technology that
can be used type,
not only
for the
detection
of faults in oil wetted
systems
and identification
of contaminant
guiding
remedial
actions.
machinery but also for the determination of particulate contamination levels in hydraulic
systems andFINES
identification
of contaminant type, guiding remedial actions.
LASERNET
TECHNOLOGY
LASERNET FINES TECHNOLOGY
The LNF analysis process is illustrated in Fig. 1. A sample of fluid, either a bypass
from the The
mainLNF
flowanalysis
for anprocess
on lineis system
or ina Fig.
bottle
from
a manually-sampled
illustrated
1. sample
A sample
of fluid,
either a bypass
system,
is
illuminated
with
a
laser
diode
and
images
of
entrained
objects
are recorded on
from the main flow for an on line system or a bottle sample from a manually-sampled
an system,
electronic
camera.
LNF
measures
the
size
distribution
of
particles
from
µm to above
is illuminated with a laser diode and images of entrained objects are4 recorded
on
100anµm,
and
provides
classification
based
on
shape
features
for
objects
larger
than
20 µm.
electronic camera. LNF measures the size distribution of particles from 4 Jim
to above
Analysis
of and
particles
in this
size rangebased
is useful
for the
early for
detection
mechanical
100 jLim,
provides
classification
on shape
features
objectsoflarger
than 20 faults
|im.
in Analysis
diesel and
turbine
engines,
turbochargers
and
gearboxes.
This
range
of
sizes faults
is also
of particles in this size range is useful for the early detection of mechanical
relevant
to particulate
levels inand
hydraulic
and fuel
Examples
in diesel
and turbine contamination
engines, turbochargers
gearboxes.
This systems.
range of sizes
is alsoof
images
of different
typescontamination
of particles and
theirin implications
mechanical
are
relevant
to particulate
levels
hydraulic andforfuel
systems. difficulties
Examples of
images
of different
types of particles
and made
their implications
forof
mechanical
shown
in Fig.
2. Quantitative
records are
of the amount
each typedifficulties
of particle are
as a
shown of
in size,
Fig. 2.
Quantitative
recordsmap
are of
made
of the amount
each20
type
particle
as a
function
along
with an image
all particles
largerofthan
µm.ofThese
records
function
size, along
an image map
of allfrom
particles
than 20 providing
|iim. These the
records
allow
fault of
signatures
to with
be developed
directly
the larger
LNF results,
basis
fault signatures
to be developed
directly
fromdetermination
the LNF results,
providing theactions,
basis
forallow
root cause
analysis, definition
of fault
severity,
of maintenance
root cause
analysis,
definition of fault severity, determination of maintenance actions,
andforremaining
useful
life prognostics.
and remaining useful life prognostics.
SAMPLE ANALYSIS
WEAR MODE
WEAR MODE
CLASSIFICATION
CLASSIFICATION
SAMPLE ANALYSIS
MACHINE
MACHINE
CONDITION
CONDITION
ASSESSMENT
ASSESSMENT
MAINTENANCE
MAINTENANCE
ACTION
ACTION
Sev ere slidin g wear
40 0
Full Fluid Flow
Sampled Fluid
Flow
Lens
4x Mag
TV Rate Camera
Image Processing
Shape Classification
concentration (ml-
35 0
30 0
25 0
20 0
15 0
10 0
50
0
20-25
25-50
50-100
>1 00
size range (µm)
Laser Diode
Wear
Debris
FIGURE
1. Schematic
illustration
operationofofLaserNet
LaserNetFines
Finesand
and its
its application
application to
FIGURE
1. Schematic
illustration
of of
thethe
operation
to fault
fault analysis.
analysis.
1591
WMM?M®im*
Fatigue spall. Excess load or
end of life
c
Air bubble. Possible leak in hydraulic
system.
Cutting wear. Shaft
misalignment or secondary
damage from debris buildup or
sand contamination.
Water bubble. Free water
Severe sliding wear. Oil film
failure or inadequate lubrication.
Fiber. Artifact in bottle sample
analysis, filter deterioration, composite
bearing wear.
Sand. Contaminant, source of
secondary damage.
FIGURE 2. Images of objects detected and identified by LaserNet Fines.
LNF can be configured as an off line batch sample analyzer or an on line wear or hydraulic
cleanliness monitor. First and second generation batch processor models are in
commercial production by Lockheed Martin. An on-line system is undergoing laboratory
test and evaluation.
DIESEL ENGINE CLASS STUDY
As an example of the ability of LNF to find and identify specific mechanical faults,
a study was made of 32 medium speed diesel engines. The engines were put into service at
different times, but were all subject to the same operational profile. A class study of this
type provides a snapshot of the diesels over their life, allowing the establishment of normal
and abnormal operation.
Total particle concentration for the engines is shown in Fig. 3, and concentration of
particles in the different wear categories are shown in Fig. 4. The majority of the engines
showed consistently low particle concentrations, allowing the establishment of a baseline
for normal wear. However, significantly elevated particle concentrations were found on 6
engines. Of the 6 engines that showed elevated particle concentrations, one (#18) showed
significantly higher concentrations than the other engines in all LNF categories (fatigue,
cutting, sliding and non-metallics). This engine was singled out for detailed study. During
the study it was made known to us that a complete failure of a piston wrist pin had
occurred in one of the engines (#14). Our initial sample was taken for that engine after the
cylinder assembly was replaced. Subsequently a sample was obtained from that engine
that had been taken 14 months prior to detection of the failure.
Engines with elevated total particle concentrations but low concentrations of
particles larger than 20 microns are indicated by open symbols in Figs. 3 and 4. This
characteristic can indicate the presence of faults such as grinding, that do not produce large
particles or of heavy sooting. In either case, it is indicative of a second type of fault, not
related to the severe wear seen on the engines with elevated large-particle concentrations.
1592
concentration (ml-1)
Total Particles
25000
Total Particles
20000
•
15000
o o
10000
V*
5000
%« **«*x±* » « V %»*»*
0
0
5
0
10
5
15
10
20
15 Number
20
Diesel
25
30
25
35
30
35
Diesel Number
FIGURE
3. 3.
Total
particle
concentrations
forforthe
FIGURE
Total
particle
concentrations
thesetsetofofmedium
mediumspeed
speeddiesels.
diesels.
Severe sliding Wear
Severe sliding Wear
600
Cutting Wear
120-
i:
500
400
300
•§nn
m
200
"Inn
80
$
•
60
°
40
I
«g»»Mtity»tt*..o. •?••?•?•••?,??. .
0
5
5
10 10
15 15
20 20
25 25
30 30
&
35 35
°
2
°
0
0 D
Diesel
Number
Diesel
Number
•
4
20
0
+
o*
ttTMi»tf»ttfr..».r,.»t»%»»tttt..l
5 5
1010
1515
2020
2525
30
30
25
30
Diesel
DieselNumber
Number
Fatigue
wear
Fatigue
wear
Oxides
Oxides
160
1200
140
concentration (ml-1)
1000
concentration (ml-1)
•
6
1
®
|
100nn
0
•
100
*— 100-
concentration (ml-1)
concentration (ml-1)
Cutting Wear
120
«
^800
800
120
s
600
•+
100
80
•
1^ 60 60
400
©
C
200 200
0
0
0
10
15
20
25
30
o
J20 »•
ti'tttt'ttttt* - t* ,» t t.ttt,tt
5
40
0 00
35
5
10
15
20
35
Diesel Number
Diesel Number
FIGURE
LNF
wear
class
analysis
dieselengines
enginesshowing
showingparticle
particleconcentrations
concentrationsfor
foreach
each
FIGURE
4. 4.
LNF
wear
class
analysis
forfor
thethesetsetofofdiesel
engine
in the
various
wear
types.Top
Top
left:severe
severesliding
slidingwear;
wear;top
topright:
right:cutting
cuttingwear;
wear; bottom
bottomleft:
left:fatigue
fatigue
engine
in the
various
wear
types.
left:
wear;
bottom
right:
non-metallics.
wear;
bottom
right:
non-metallics.
Validation
LNF
Results
Validation
of of
LNF
Results
analysis
consistingofofoptical
opticalmicroscope
microscopeexamination
examinationand
andscanning
scanningelectron
electron
AA
lablab
analysis
consisting
microscope
energy
dispersive
x-ray
(SEMEDX)
measurements
was
made
of
the
debris
microscope energy dispersive x-ray (SEMEDX) measurements was made of the debris
samples from engines 18 and 14 and also of the failed parts from engine #14.
samples from engines 18 and 14 and also of the failed parts from engine #14.
The optical microscope examination confirmed that the oil samples from engines
The optical microscope examination confirmed that the oil samples from engines
18 and 14 contained large concentrations of metal chips of various sizes and shapes in
18 and 14 contained large concentrations of metal chips of various sizes and shapes in
agreement with LNF analysis. A high-resolution microscope image of debris from engine
agreement with LNF analysis. A high-resolution microscope image of debris from engine
1593
#18 (Fig. 5) shows the presence of flakes, slivers and curled cutting wear confirming the
#18
(Fig. 5)Fines
shows
the presence
of type.
flakes, slivers and curled cutting wear confirming the
LaserNet
analysis
of debris
LaserNet The
FinesSEMEDX
analysis ofanalysis
debris type.
of the failed parts from engine #14 showed that the wrist
The SEMEDX analysis of the failed parts from engine #14 showed that the wrist
pin was
steel with chromium and nickel components, while the bearing sleeve was mild
pin was steel with chromium and nickel components, while the bearing sleeve was mild
steel with no chromium or nickel. The debris from the failed engine showed pieces
steel with no chromium or nickel. The debris from the failed engine showed pieces
matching the wrist pin and others matching the sleeve. All of the particles in the oil
matching the wrist pin and others matching the sleeve. All of the particles in the oil
samples from #18 and #14 matched only the bearing sleeve material, not the wrist pin.
samples from #18 and #14 matched only the bearing sleeve material, not the wrist pin.
Theseresults
resultsconfirm
confirm
that
LaserNet
Fines
is indeed
measuring
heavy
when
These
that
LaserNet
Fines
is indeed
measuring
heavy
wearwear
when
it it
occurs,
and
is
not
reporting
it
in
engines
without
heavy
wear.
occurs, and is not reporting it in engines without heavy wear.
LNFFault
FaultSignatures
Signatures
LNF
AnalysisofofthetheLaserNet
LaserNet
Fines
records
indicates
quantitative
distributions
Analysis
Fines
records
indicates
thatthat
the the
quantitative
distributions
of
particle
concentration
by
size
in
the
various
wear
categories
can
be
combined
to give
of particle concentration by size in the various wear categories can be combined to give
a a
signatureforforfault
faultpresence
presence
and
severity
(Figs.
6 and
distributions
engine
signature
and
severity
(Figs.
6 and
7). 7).TheThe
distributions
for for
the the
engine
withnormal
normalwear
wearshow
showlow
low
particle
concentration,
with
a steeply
declining
concentration
with
particle
concentration,
with
a steeply
declining
concentration
withparticle
particlesize
sizeforforthetheindividual
individual
wear
categories
(Fig.
engine
particle
with
wear
categories
(Fig.
7). 7).
ForFor
engine
#18,#18,
the the
particle
sizedistributions
distributionsarearesignificantly
significantlydifferent,
different,
being
skewed
to the
larger
particle
size
being
skewed
to the
larger
particle
sizessizes
100 µm
I ^————
500500
jimµm
FIGURE 5. Magnified images of debris from engine #18 showing fatigue flakes, sliding slivers and cutting
FIGURE 5. Magnified images of debris from engine #18 showing fatigue flakes, sliding slivers and cutting
curls.
curls.
Severe sliding wear
Fatigue wear
Fa tigue wea r
70
t
350
400
300
200
100
20-25
20-25
25-50
50-103
see rang e Jjm)
25-50
50-100
size rang e (µm)
>100
>100
60
300
250
200
0
150
100
50
20^25
0
concentration (ml-1
500
0
C utting we ar
400
600
concentration (ml-1
concentration (ml-1
i
Cutting wear
Se v e re sliding we a r
20
50
40
30
S: LI
20
10
25-50
50-100
size range (|jn)
20-25
25-50
50-100
si ze range (µm )
>100
>100
2D-25
0
25-50
20-25
FIGURE 6. LNF particle distributions for engine #18 showing signature for heavy wear.
FIGURE 6. LNF particle distributions for engine #18 showing signature for heavy wear.
1594
50-100
size range (\m)
25-50
>100
50-100
si ze range (µm )
>100
Fatigue wear
Severe sliding wear
10
Fatigue wear
concentration
(ml-1 (ml-1
concentration
concentration (ml-1
concentration (ml-1
1012
810
68
46
24
0
2
0
20-25
20-25
20-25
I
25-50
50-100
25-50
50-100
size range (µm )
25-50
50-100
size
range (jjii)
Severe sliding wear
10
8
10
8
6
6
4
4
2
2
0
>100
>100
0
>100
Cut ting we ar
Cutting wear
10
Severe sliding wear
concentration (ml-1
concentration (ml-1
Fatigue wear
12
8
6
4
Cut ting we ar
8
6
4
2
2
0
20-25
20^25
20-25
size range (µm )
25-50
50-100
2550
50-100
size range (µm)
25-50
50-100
size range (|jm)
>100
>100
20-25
20-25
0
>100
20-25
size range (µm)
25-50
50-100
25-50
50-100
size range (µm )
25-50
50-100
size range
(jjm)
>100
>10D
>100
size range (µm )
FIGURE7.7.LNF
LNFparticle
particledistributions
distributionsfor
foranan engine
enginewith
withnormal
normaloperation
operation
FIGURE
FIGURE 7. LNF particle distributions for an engine with normal operation
(Fig.6).
6).The
Therelative
relativeamounts
amountsofofparticles
particlesininthe
thevarious
variouswear
wearcategories,
categories,along
alongthe
thesize
size
(Fig.
(Fig. 6). Theprovide
relativethe
amounts
of particles
the
variouspresence
wear categories,
along
the
size
distributions
information
neededin
interpret
andseverity
severity
thefault.
fault.
distributions
provide the
information
needed
totointerpret
presence and
ofofthe
distributions
providequantitative
the information
needed
to
interpret
and severity ofofthe
fault.
LNF
thusprovides
provides
analysis
thatcan
can
beused
usedpresence
forrecommendation
recommendation
remedial
LNF
thus
quantitative
analysis
that
be
for
of remedial
remedial
LNF
thus
provides
quantitative
analysis
that
can
be
used
for
recommendation
of
action. Furthermore,
Furthermore,the
theLNF
LNFfault
faultsignatures
signaturesare
arerobust
robustagainst
againstmaintenance
maintenanceactions
actionssuch
such
action.
action.
Furthermore,
the
LNF fault
signatures
are robust
against
maintenance
actionsnot
such
as
topping
off
of
oil
supplies
because
they
are
based
on
relative
distributions,
just
asastopping
off
oil
not just
just
topping
offofofconcentrations.
oil supplies
supplies because
because they
they are
are based
based on
on relative
relative distributions,
distributions, not
absolute
particle
absolute
particle
concentrations.
absolute particle concentrations.
EarlyFault
Fault Detection
Early
Early FaultDetection
Detection
Followingthis
this analysisititwas
was discoveredthat
that a samplefrom
from engine#14
#14that
thathad
had
Following
Following thisanalysis
analysis it was discovered
discovered that aa sample
sample from engine
engine #14 that
had
been
drawn
14
months
prior
to
detection
of
the
failure
was
available.
LNF
analysis
of
that
been
of that
that
beendrawn
drawn1414months
monthsprior
priortotodetection
detectionof
ofthe
thefailure
failure was
was available.
available. LNF
LNF analysis
analysis of
sample showed
showed an abnormal signature
signature in the fatigue
fatigue andsliding
sliding categories (Fig.
(Fig. 8),
8),
sample
sample showed anan abnormal
abnormal signature inin the
the fatigue and
and sliding categories
categories (Fig. 8),
indicativeofofan
anearly
earlystage
stageofofthe
themechanical
mechanicalfault
faultthat
thateventually
eventuallyresulted
resultedinintotal
totalfailure
failure
indicative
indicative of an early stage of the mechanical fault that eventually resulted in total failure
the
wrist
pin.
This
demonstrates that
that LNF
LNF has
has the
the capability
capability for
for providing
providingaaa
ofof
for
providing
ofthe
the wrist
wrist pin.
pin. This
This demonstrates
demonstrates
that
LNF
has
the
capability
considerable
advanced
warning
of
potentially
serious
wear
conditions.
considerable
considerableadvanced
advancedwarning
warningof
ofpotentially
potentially serious
serious wear
wear conditions.
conditions.
Validation
the
LNF
Fault
Signature
Validation
Validationofof
ofthe
theLNF
LNFFault
FaultSignature
Signature
Tests
have
been
carried
outinin
inlaboratory
laboratorycontrolled
controlledenvironments
environmentstotoobtain
obtainLNF
LNF
Tests
obtain
LNF
Testshave
havebeen
beencarried
carried out
out
laboratory
controlled
environments
signatures
for
specific
mechanical
wearprocesses
processesin
invarious
variousstages
stagesofofdevelopment.
development. The
The
signatures
development.
The
signaturesfor
forspecific
specificmechanical
mechanicalwear
wear
processes
in
various
results
show
that,
for
severe
sliding
test
conditions,
the
fault
signature
described
above
results
above
resultsshow
showthat,
that,for
forsevere
severe sliding
sliding test
test conditions,
conditions, the
the fault
fault signature
signature described above
develops
the
LNF
sliding
category after
after sliding
slidingseizure
seizureevents.
events.(Fig.
(Fig.9)9)No
Nosimilar
similar
develops
similar
developsinin
inthe
theLNF
LNF sliding
sliding category
category
after
sliding
seizure
events.
development
evident
the fatigue
fatigue category,
category, confirming
confirming the
the ability
ability ofof LNF
LNFto
development
LNF
toto
development isis
is evident
evident inin
in the
the
fatigue
category,
confirming
distinguish
different
mechanical
wearconditions.
conditions.(Fig.
(Fig.10)
10)
distinguish
distinguishdifferent
differentmechanical
mechanicalwear
wear
conditions.
(Fig.
Fatigue
wea
Fatigue
Fatiguewear
wea
Severesliding
slidingwea
wea
Severe
wear
Severe
6060
Cutting
wear
Cutting
wear
Cutting
wear
60
60
6060
50
50
5050
40
40
4040
3030
30
30
3030
2020
20
20
2020
1010
10
10
1010
00
00
5050
4040
20-25
20-25
20-25
25-50
50-100
25-50
50-100
25-50
50-100
size
range
(µm)
size
range
(µm)
size
range
(Mm)
>100
>100
>100
00
20-25
20-25
25-50
25-50
50-100
50-100
size
sizerange
range(µm)
(µm)
>100
>100
20-25
20-25
25-50
25-50
50-100
50-100
>100
>100
size
range
(µm)
size
range
(µm)
FIGURE8.8.
8.LNF
LNFanalysis
analysisof
from
engine
FIGURE
LNF
analysis
of oil
oil sample
sample from
from engine
engine 14
14taken
taken14
14months
monthsprior
priortoto
tofailure
failuredetection
detectionshowing
showing
FIGURE
14
taken
14
months
prior
failure
detection
showing
earlydetection
detectionof
ofheavy
heavywear.
wear.
early
detection
of
heavy
wear.
early
1595
Sliding Onset of seizure
80
70
70
concentration(ml-
concentration (ml
Sliding Pre-seizure
80
Sliding Pre-seizure
60
50
40
60
C
30
10
30,-
10
30
i § "
0
20
25
50
size (um)
100
Sliding During seizure
concentration (ml
20
-2
0
80
size
concentration (ml
Sliding During seizure
60
20
25
50
size (um)
100
Sliding Post seizure
80
(urn)
70
50
I 40
30
20
10
3U
i!«
8"
30
-[:
20
I ™
50
m 7°
40
t
Sliding Onset of seizure
60
0
70
60
50
40
30
20
10
0
20
25
50
100
20
25
50
size (um)
size (um)
100
FIGURE 9. LNF analysis of wear particles from a controlled laboratory severe sliding test showing
development of the abnormal signature in the severe sliding wear category following a sliding seizure event.
FIGURE 9. LNF analysis of wear particles from a controlled laboratory severe sliding test showing
development of the Fatigue
abnormal
signature in the severe sliding wear category following
sliding
seizure event.
Fatigue a
Onset
of seizure
Pre-seizure
80
70
concentration(ml-1
concentration (ml-1
80
Fatigue Eire-seizure
60
80
I
50
70
40
60
i
30
50
20
40
Fatigue Onset of seizure
60
50
40
30
20
10
10
& 30
0
J 20
20- —
10
70
0
20
25
50
size (um)
20
100
100
Fatigue Post seizure
80
80
70
70
concentration (ml-1
concentration (ml-1
Fatigue During seizure
25
50
size (um)
60
50
40
30
20
10
0
Fatigue Post seizure
60
50
40
30
20
10
0
20
25
50
size (um)
100
20
25
50
size (um)
100
FIGURE 10. LNF analysis of wear particles from controlled laboratory severe sliding test showing a normal
25
50
25
50
signature in the fatigue wear
following the severe sliding seizure.
size category
(urn)
size (urn)
FIGURE 10. LNF analysis of wear particles from controlled laboratory severe sliding test showing a normal
LNF fault
signature
for sliding
heavyseizure.
wear in the operating equipment
signatureThis
in theresult
fatigueties
wearthe
category
following
the severe
to fundamental mechanical wear processes.
This result ties the LNF fault signature for heavy wear in the operating equipment
to fundamental mechanical wear processes.
1596
ACKNOWLEDGEMENTS
The authors acknowledge the support of the Office of Naval Research.
REFERENCES
1. T. P. Sperring, J. Tucker, J. Reintjes, A. Schultz, C. Lu and B. J. Roylance," Wear
Particle Imaging and Analysis - a contribution towards monitoring the health of
military ships and aircraft," International Conference on Condition Monitoring, pp.
539-546, University of Wales, Swansea, UK, April 1999.
2. B. J. Roylance and G. Pocock, "Wear Studies through Particle Size Distribution,"
Wear, 90, pp. 113-136 (1983).
3. A. Albidewi., A. R. Luxmore, B. J. Roylance, and G. Wang, "Determination of
Particle Shape by Image Analysis-the Basis for Developing an Expert System," in
"Condition Monitoring '91," M. H. Jones, J. Guttenberger and H. Brenneke, eds.,
Pineridge Press, Swansea, UK, 1991, p. 411.
4. B. J. Roylance and S. Raadnui, "The morphological attributes of wear particles - their
role in identifying wear mechanisms," Wear 175, 115 (1994).
5. B. J. Roylance, I. A. Albidewi, A. R. Luxmoore, A. L. and Price, "The Development
of a Computer-Assisted Systematic Particle Analysis Procedure - CASPA," Lub. Eng.,
48, pp. 940-946 (1992).
6. B. J. Roylance, I. A. Albidewi, M. S. Laghari, A. R. Luxmoore and F. Deravi,
"Computer-Aided Vision Engineering (CAVE) - Quantification of Wear Particle
Morphology," Lub. Eng., 50, pp. 111-116 (1994).
7. T. G. Barraclough, T. P. Sperring and B. J. Roylance, "Generic-based Wear Debris
Identification - the First Step Towards Morphological Classification", International
Conference on Condition Monitoring, University of Wales, Swansea, UK April 1999.
8. J. E. Tucker, J. Reintjes, A. Schultz, C. Lu, L. L. Tankersley, P. L. Howard, T. Sebok,
and C. Holloway, "LASERNET Fines Shipboard Wear Debris Monitor", Technology
Showcase 2000, JOAP International Conference, Condition, Mobile AL, April 3-6,
2000.
9. J. E. Tucker, A. Schultz, C. Lu, T. Sebok, C. Holloway, L. L. Tankersley , T.
McClelland, P. L. Howard and J. Reintjes, LaserNet Fines Optical Wear Debris
Monitor," International Conference on Condition Monitoring, pp. 445-452, University
of Wales, Swansea, UK, April 1999.
10. J. E. Tucker, J. Reintjes, T. L. McClelland, M. D. Duncan, L. L. Tankersley, A.
Schultz, C. Lu, P. L. Howard, T. Sebok, C. Holloway, and S. Fockler, "LaserNet Fines
Optical Oil Debris Monitor," 1998 JOAP International Condition Monitoring
Conference, pp. 117-124 April, 1998, Mobile AL.
11. J. Reintjes, R. Mahon, M. D. Duncan, L. L. Tankersley, J. E. Tucker, A. Schultz, V. C.
Chen, C. Lu, T. L. McClelland, P. L. Howard, S. Raghavan and C. L. Stevens, "Real
Time Optical Oil Debris Monitors," Proceedings of 51st Meeting of the Society for
Machinery Failure Prevention Technology, April, 1997, pp. 443-448, Virginia Beach,
VA.
1597