Diesel Engine Analysis Guide-107135

'LHVHO(QJLQH$QDO\VLV*XLGH
EPRI TR-107135s Electric Power Research Institute September 1997

R E P O R T S U M M A R Y
INTEREST CATEGORIES
Nuclear plant operations
and maintenance
Plant electrical systems and
equipment
Engineering and technical
support
Maintenance practices
KEYWORDS
Monitoring
Diagnostics
Predictive maintenance
Maintenance
Diesel Engines
Diesel Engine Analysis Guide
This guide provides a throrough background on diesel engine analysis
including combustion, vibration, and ultrasonic analysis theory. Inter-
pretation of results is also provided. This guide is intended to enable
nuclear utility personnel to make informed decisions regarding the
nature and use of diesel engine analysis, including how to set up an
effective program, how to establish analysis guidelines, how to make
use of the resulting data to plan maintenance, determine the causes
of off-design operating conditions, and warn of potential component
failures.
BACKGROUND Diesel engine analysis has not seen consistent widespread
use in the nuclear power industry, in part because utility personnel interested in
establishing an engine analysis program have not had a single comprehensive
source for technical guidance to establish and evaluate it. At many of the
locations where engine analysis has been used in the nuclear power industry, it
has been successful in identifying engine problems before these problems
progressed to catastrophic failures. For example, engine analysis has success-
fully identified bent connecting rods, cracked cylinder heads, scuffed cylinder
liners, leaking fuel injectors and valves, and collapsed valve lifters in nuclear
standby EDGs. In addition, engine analysis can and has been used successfully
to identify problems in diesel engines in other industries such as detonation,
cylinder load imbalance, faulty fuel injection, leaking valves and piston rings,
worn or scored cylinder liners, worn rocker arms, defective valve lifters, worn
cams, and damaged wrist pins.
Engine analysis techniques were first used to support engine design and
development efforts. Diesel engine analysis has been used in the marine, rail,
and pipeline industries for many years and has proven cost-effective and
successful. It has been used to tune engines, to improve fuel efficiency, to
support predictive maintenance methods, to assist with problem diagnosis, and
to support longer maintenance intervals.
OBJECTIVES
To provide a thorough discussion on the theory and practical aspects of
implementing combustion, vibration, and ultrasonic analyses on diesel engines
at nuclear power plants.
To enable nuclear utility personnel to make informed decisions regarding
combustion, vibration, and ultrasonic analysis of diesel engines, including how
to set up an effective program, how to establish analysis guidelines, how to
make use of the resulting data to plan maintenance, determine the causes of
EPRI
Powering Progress
off-design operating conditions, and warn of potential component
failures.
APPROACH The project team gathered and reviewed pertinent techni-
cal literature regarding the subject of diesel engine analysis. Information
and experience from utility personnel, consultants, and manufacturers
was condensed. The resulting document was reviewed by a technical
advisory group composed of experienced diesel engineers. This guide is
applicable to selected makes and models of diesel engines in nuclear
standby service in the United States and around the world.
RESULTS This guide includes a thorough background on diesel
engine combustion, vibration, ultrasonic analysis, theory, interpretation
of results, guidance on identifying specific modes of degradation, and
examples of specific engine combustion, vibration, and ultrasonic
analyses for diesel engines in nuclear standby service. In addition,
troubleshooting guidance and a discussion of acceptance guidelines
are reviewed. Also, engine analysis programs, equipment costs,
equipment selections, and setups are covered.
EPRI PERSPECTIVE Development of this guide was sponsored by
the Electric Power Research Institute Nuclear Maintenance Applica-
tions Center (EPRI/NMAC) to provide consistent, comprehensive, and
detailed guidance for EDG system engineers on diesel engine analy-
sis. The guide was developed in cooperation with the various EDG
owners groups, engine and analysis equipment vendors, and a utility
Technical Advisory Group (TAG).
PROJECT
WO 3814
EPRI Project Manager: James P. Sharkey
Nuclear Power Group
Contractor: MPR Associates
For further information on EPRI research programs, call EPRI
Technical Information Specialists at 415/855-2411.
TR-107135
Final Report
September 1997
Prepared by
MPR Associates
320 King Street
Alexandria, VA 22314
Principal Investigators
Christian Haller
Mark OConnell
Prepared for
Nuclear Maintenance Applications Center
1300 W.T. Harris Boulevard
Charlotte, North Carolina 28262
Operated by
Electric Power Research Institute
3412 Hillview Avenue
Palo Alto, California 94304
EPRI Project Manager
James P. Sharkey
Nuclear Power Group
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS REPORT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF WORK
SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI). NEITHER
EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING
ON BEHALF OF ANY OF THEM:
A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT
TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS
REPORT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH
USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTYS
INTELLECTUAL PROPERTY, OR (III) THAT THIS REPORT IS SUITABLE TO ANY PARTICULAR USERS
CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY
CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS REPORT OR ANY
INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS REPORT.
ORGANIZATION(S) THAT PREPARED THIS REPORT
MPR Associates
ORDERING INFORMATION
Price: $10,000
Requests for copies of this report should be directed to the Nuclear Maintenance Applications Center
(NMAC), 1300 W.T. Harris Boulevard, Charlotte, NC 28262, 800/356-7448. There is no charge for reports
requested by NMAC member utilities.
Electric Power Research Institute and EPRI are registered service marks of Electric Power Research Institute, Inc.
Copyright 1997 Electric Power Research Institute, Inc. All rights reserved.
iii
EPRI Licensed Mat erial
Important Notice Regarding This Document
Use of this document is voluntary.
It is not intended for regulatory or enforcement purposes.
It is offered for consideration and use by members of
EPRI/ NMAC.
Use of this document and its contents by anyone other
than those for whom it is intended is not authorized.
This document is based on consensus of the writers,
reviewers and contributors.
There may be other techniques or means of performing
the work or activities described here.
Questions concerning use of this material
should be directed to EPRIs
Nuclear Maintenance Applications Center (NMAC).
v
ACKNOWLEDGMENTS
EPRI/ NMAC would like to thank the following individuals and groups for their par-
ticipation during the development of this guide:
Kenneth Burr Southern Nuclear Operating Company
Ted Bronson Duke Engineering & Services
Ed Carr Liberty Technologies/ Beta
Gurpal Gill Southern Nuclear Operating Company
David Gustafson Commonwealth Edison Company
Stanford Harris Carolina Power & Light
Ray Howard Southern Nuclear Operating Company
Gary Hunt Houston Lighting & Power
Mike Knebel Entergy Operations
Dwain Lambert Commonwealth Edison Company
Bryan Long Liberty Technologies/ Beta
Mike Murdock Carolina Power & Light
Terence OBrien Commonwealth Edison Company
Larry Price Pacific Gas & Electric Company
James Wesley Southern Nuclear Operating Company
Donnie Williams Carolina Power & Light
Gary Yezefski Texas Utilities
Doug Stover PMC/ Beta
ALCO Owners Group
Cooper Bessemer Owners Group
ESI-EMD Owners Group
vii
CONTENTS
1 INTRODUCTION ............................................................................................................................. 1-1
1.1 Purpose ............................................................................................................................... 1-1
1.2 Background ......................................................................................................................... 1-1
1.3 Scope ................................................................................................................................... 1-2
1.4 Development ....................................................................................................................... 1-2
2 BACKGROUND .............................................................................................................................. 2-1
2.1 Traditional Approach to Engine Maintenance.................................................................. 2-1
2.2 The Integrated Maintenance Approach ............................................................................ 2-2
2.3 Integrated Maintenance Programs .................................................................................... 2-3
2.4 Engine Operability .............................................................................................................. 2-5
2.5 Degradation and Failure..................................................................................................... 2-7
3 ELEMENTS OF AN ENGINE ANALYSIS PROGRAM ................................................................... 3-1
3.1 Infrastructure ...................................................................................................................... 3-1
3.1.1 Management Support .............................................................................................. 3-1
3.1.2 Engine Analysis Equipment ..................................................................................... 3-1
3.1.3 Personnel ................................................................................................................. 3-3
3.1.4 Equipment Calibration and Maintenance Program .................................................. 3-3
3.1.5 Training .................................................................................................................... 3-4
3.2 Programmatic Issues ......................................................................................................... 3-4
3.2.1 Integration with Other Maintenance Programs ........................................................ 3-4
3.2.2 Parameters to Monitor ............................................................................................. 3-4
3.2.3 Monitoring Frequency .............................................................................................. 3-5
3.2.4 Standardized Data Collection Methods ................................................................... 3-5
3.3 Analysis ............................................................................................................................... 3-7
3.3.1 Trending ................................................................................................................... 3-7
3.3.2 Baseline Data .......................................................................................................... 3-7
3.3.3 First Time Engine Setup .......................................................................................... 3-7
3.3.4 Acceptance Guidelines ............................................................................................ 3-8
3.3.5 Case Histories ......................................................................................................... 3-8
3.3.6 Analysis Flow Charts ............................................................................................... 3-9
3.3.7 Feedback from Maintenance Personnel .................................................................. 3-9
viii
4 THEORY.......................................................................................................................................... 4-1
4.1 Combustion......................................................................................................................... 4-1
4.1.1 Two-Stroke Engines ................................................................................................. 4-4
4.1.2 Four-Stroke Engines ................................................................................................ 4-8
4.1.3 Four Phases of Combustion .................................................................................... 4-9
4.1.4 Engine Parameters ................................................................................................ 4-11
4.1.5 Measurement Methods .......................................................................................... 4-14
4.2 Combustion Analysis ....................................................................................................... 4-15
4.2.1 Combustion Curves ............................................................................................... 4-15
4.2.2 Combustion Characteristics ................................................................................... 4-22
4.2.3 Variability ............................................................................................................... 4-27
4.2.4 Interpretation of Combustion Signatures ............................................................... 4-28
4.2.5 Analysis Guidelines ............................................................................................... 4-28
4.2.6 Trending ................................................................................................................. 4-31
4.3 Vibration and Ultrasonic Analyses ................................................................................. 4-32
4.3.1 Sources of Engine Vibration and Ultrasonic Noise ................................................ 4-32
4.3.2 Characteristics of Engine Vibration and Ultrasonic Signals ................................... 4-33
4.3.3 Measurement Methods .......................................................................................... 4-34
4.3.4 Interpretation of Vibration and Ultrasonic Signatures ............................................ 4-34
4.3.5 Analysis Guidelines ............................................................................................... 4-38
4.3.6 Trending ................................................................................................................. 4-39
5 EQUIPMENT SELECTION, SETUP, AND USE .............................................................................. 5-1
5.1 Key Analyzer Characteristics ............................................................................................ 5-1
5.2 Cylinder Pressure Transducers ......................................................................................... 5-4
5.2.1 Pressure Transducer Characteristics ....................................................................... 5-4
5.2.2 Transducer Preparation ........................................................................................... 5-4
5.2.3 Transducer Calibration ............................................................................................. 5-5
5.2.4 Transducer Mounting ............................................................................................... 5-5
5.3 Vibration and Ultrasonic Accelerometers ........................................................................ 5-5
5.3.1 Accelerometer Characteristics ................................................................................. 5-5
5.3.2 Accelerometer Calibration ....................................................................................... 5-6
5.3.3 Accelerometer Mounting .......................................................................................... 5-6
5.4 Filtering and Signal Conditioning ..................................................................................... 5-9
5.5 Sampling Rate................................................................................................................... 5-10
5.6 Sources of Error ............................................................................................................... 5-10
5.6.1 Selection, Calibration, Preparation, and Mounting of Cylinder
Pressure Transducers ............................................................................................ 5-10
5.6.2 Indicator Passage Effects ...................................................................................... 5-10
5.6.3 Crank Angle Phasing and Correction .................................................................... 5-11
5.6.4 Load Drift ............................................................................................................... 5-15
5.6.5 Configuring the Analyzer ....................................................................................... 5-15
5.7 Preparing the Engine for Taking Data ............................................................................. 5-16
5.7.1 Start and Load the Engine ..................................................................................... 5-16
5.7.2 Lock the Governor ................................................................................................. 5-17
5.7.3 Set Engine Load .................................................................................................... 5-17
5.7.4 Record Initial Balance of Plant Data ...................................................................... 5-17
ix
5.8 Measuring Cylinder Pressures ........................................................................................ 5-18
5.8.1 Indicating Valve Extension ..................................................................................... 5-18
5.8.2 Check Cylinder Pressure Transducer Operation.................................................... 5-18
5.8.3 Reference Pressure ............................................................................................... 5-18
5.8.4 Record Cylinder Pressure...................................................................................... 5-18
5.9 Recording the Data ........................................................................................................... 5-20
6 ENGINE BALANCING AND TUNING ............................................................................................. 6-1
6.1 Purposes of Engine Balancing.......................................................................................... 6-1
6.2 What is a Balanced Engine?.............................................................................................. 6-2
6.2.1 Balancing Criteria .................................................................................................... 6-2
6.3 Engine Analysis Guidelines............................................................................................... 6-5
6.3.1 Maximum Engine Average Peak Firing Pressure and Maximum
Cylinder Peak Firing Pressure ................................................................................. 6-5
6.3.2 Maximum Cylinder-to-Cylinder Variation in Peak Firing Pressures.......................... 6-5
6.3.3 Variation of Cylinder Peak Firing Pressure Over a Number of Cycles ..................... 6-5
6.3.4 Engine Average Corrected Peak Firing Pressure Angle .......................................... 6-6
6.3.5 Maximum Range of Peak Firing Pressure Angles Across All Cylinders .................. 6-7
6.3.6 Engine Average Fuel Injection Timing ..................................................................... 6-7
6.3.7 Maximum Range of Fuel Injection Timing Across All Cylinders............................... 6-7
6.3.8 Maximum Range of Fuel Rack Settings Across All Cylinders at Full Load.............. 6-7
6.3.9 Allowable Range of Fuel Injection Nozzle Pop Pressures for New and
Used Fuel Injection Nozzles .................................................................................... 6-7
6.3.10 Maximum Cylinder Exhaust Gas Temperature ........................................................ 6-8
6.3.11 Maximum Range of Exhaust Gas Temperatures Across All Cylinders .................... 6-8
6.4 Factors Affecting Balance ................................................................................................. 6-8
6.4.1 Mechanical Variations or Manufacturing Tolerances of Engine Parts ...................... 6-9
6.4.2 Fuel Cetane Number ............................................................................................... 6-9
6.4.3 Engine Load............................................................................................................. 6-9
6.4.4 Fuel Rack Giggle ..................................................................................................... 6-9
6.4.5 Variations in Ambient Conditions ............................................................................. 6-9
6.5 Adjustments to the Fuel Injection System ....................................................................... 6-9
6.5.1 Fuel Injection Pump Timing ..................................................................................... 6-9
6.5.2 Fuel Delivery .......................................................................................................... 6-10
6.5.3 Inlet Valve Timing Adjustments .............................................................................. 6-10
6.6 Balancing Methods........................................................................................................... 6-11
6.6.1 Operating Factors Affecting Balance ..................................................................... 6-11
7 USING ENGINE ANALYSIS TO DETERMINE AND TROUBLESHOOT ENGINE
PERFORMANCE AND CONDITION .............................................................................................. 7-1
7.1 Integrated Maintenance Program...................................................................................... 7-1
7.1.1 Identifying Potential Problems ................................................................................. 7-1
7.1.2 Identifying and Characterizing the Failure ............................................................... 7-2
8 CASE HISTORIES .......................................................................................................................... 8-1
8.1 Power Cylinder Imbalance ................................................................................................. 8-2
8.2 Misfire .................................................................................................................................. 8-3
8.3 Piston Slap .......................................................................................................................... 8-4
8.4 Blowby/Valve Leak.............................................................................................................. 8-5
x
8.5 Unequal Lash Settings ....................................................................................................... 8-6
8.6 Excessive Exhaust Valve Lash .......................................................................................... 8-7
8.7 Faulty Fuel Injection ........................................................................................................... 8-8
8.8 Scored Liner ...................................................................................................................... 8-10
8.9 Port Bridge Problem......................................................................................................... 8-11
8.10 Improper Valve Seating .................................................................................................... 8-13
8.11 Bent Connecting Rod ....................................................................................................... 8-14
8.12 Burned Exhaust Valve ...................................................................................................... 8-16
8.13 Typical Alco 251 Cylinder Pattern ................................................................................... 8-17
8.14 Typical Cooper-Bessemer KSV Cylinder Pattern........................................................... 8-18
8.15 Typical EMD 645 Cylinder Pattern ................................................................................... 8-19
8.16 Typical FMOP 38TD 8-1/8: 12 Cylinder Pattern .............................................................. 8-20
8.17 Typical Enterprise DSRV Cylinder................................................................................... 8-21
8.18 Typical Colt Pielsick PC2 Cylinder Pattern..................................................................... 8-22
9 REFERENCES ................................................................................................................................ 9-1
xi
LIST OF FIGURES
Figure 2-1 Integrated Maintenance Approach ..................................................................................... 2-4
Figure 4-1 Piston Stroke: Top and Bottom Dead Center ...................................................................... 4-2
Figure 4-2 FMOP Engine: Inner and Outer Dead Center Arrangement ............................................... 4-3
Figure 4-3 Two-Stroke Engine Cycle ................................................................................................... 4-4
Figure 4-4 Loop Scavenging ................................................................................................................ 4-5
Figure 4-5 Uniflow Scavenging with Single Piston ............................................................................... 4-6
Figure 4-6 Uniflow Scavenging with Opposed Pistons ........................................................................ 4-7
Figure 4-7 Four-Stroke Cycle ............................................................................................................... 4-8
Figure 4-8 Cylinder Pressure Versus Crankshaft Angle..................................................................... 4-10
Figure 4-9 Rate of Heat Release Versus Crankshaft Angle ............................................................... 4-10
Figure 4-10 Two-Stroke Pressure Versus Volume Curve ..................................................................... 4-12
Figure 4-11 Four-Stroke Pressure Versus Volume Curve .................................................................... 4-12
Figure 4-12 Indicated Mean Effective Pressure ................................................................................... 4-14
Figure 4-13 Two-Stroke Engine Pressure Versus Crank Angle Curve ................................................. 4-16
Figure 4-14 Four-Stroke Engine Pressure Versus Crank Angle Curve ................................................ 4-17
Figure 4-15 Four-Stroke Log Pressure Versus Log Volume Curve ...................................................... 4-19
Figure 4-16 Polytropic Constant (n) ..................................................................................................... 4-20
Figure 4-17 First Derivative Curve ....................................................................................................... 4-21
Figure 4-18 Second Derivative Curve (dP
2
-d
2
) ................................................................................... 4-22
Figure 4-19 Four-Stroke Motoring Cylinder Pressure Versus Crank Angle.......................................... 4-25
Figure 4-20 Envelopes of Typical Vibration and Ultrasonic Patterns .................................................... 4-33
Figure 4-21 Nominal Four-Stroke Vibration and Ultrasonic Signatures ............................................... 4-36
Figure 4-22 Nominal Two-Stroke Vibration and Ultrasonic Signatures................................................. 4-37
xii
Figure 5-1 Typical Engine Pressure and Vibration Data Measured on a Cylinder Head Stud ............. 5-7
Figure 5-2 Typical Engine Pressure and Vibration Data Measured on the Cylinder Liner Flange ....... 5-8
Figure 8-1 Power Cylinder Imbalance .................................................................................................. 8-2
Figure 8-2 Misfire ................................................................................................................................. 8-3
Figure 8-3 Piston Slap ......................................................................................................................... 8-4
Figure 8-4 Piston Ring Blowby/Valve Leak .......................................................................................... 8-5
Figure 8-5 Unequal Valve Lash Settings .............................................................................................. 8-6
Figure 8-6 Excessive Exhaust Valve Lash ........................................................................................... 8-7
Figure 8-7 Normal Fuel Injection Vibration Pattern .............................................................................. 8-8
Figure 8-8 Faulty Fuel Injection Vibration Pattern ................................................................................ 8-9
Figure 8-9 Scored Cylinder Liner ....................................................................................................... 8-10
Figure 8-10 EMD 645 Vibration Patterns ............................................................................................. 8-11
Figure 8-11 Pressure and Vibration Plots for Cylinder with Port Bridge Problems .............................. 8-12
Figure 8-12 Vibration Pattern with Improper Valve Seating ................................................................. 8-13
Figure 8-13 Pressure Curves for Right Bank Cylinders ....................................................................... 8-14
Figure 8-14 Pressure Curves for Left Bank Cylinders ......................................................................... 8-14
Figure 8-15 Burned Exhaust Valve ...................................................................................................... 8-16
Figure 8-16 Typical ALCO 251 Cylinder Pattern .................................................................................. 8-17
Figure 8-17 Typical Cooper-Bessemer KSV Cylinder Pattern ............................................................. 8-18
Figure 8-18 Typical EMD 645 Cylinder Pattern .................................................................................... 8-19
Figure 8-19 Typical FMOP 38TD 8-1/8 Cylinder Pattern ..................................................................... 8-20
Figure 8-20 Typical Enterprise DSR 6-4 Cylinder Pattern ................................................................... 8-21
Figure 8-21 Typical Colt Pielstick PC2 Cylinder Pattern ...................................................................... 8-22
xiii
LIST OF TABLES
Table 1-1 Characteristics of Engines in U.S. Nuclear Service ............................................................ 1-3
Table 4-1 Existing Published Engine Analysis Guidelines................................................................ 4-30
Table 5-1 Engine Analysis Equipment Summary ............................................................................... 5-2
Table 6-1 Advantages and Disadvantages of Various Balancing Methods ........................................ 6-4
Table 7-1 Example of Simplified Engine Troubleshooting Chart ......................................................... 7-4
Table 8-1 Combustion Analysis Results for a Six Cylinder Engine..................................................... 8-8
Table 8-2 Combustion Report .......................................................................................................... 8-15
1-1
1
INTRODUCTION
1.1 Purpose
The purpose of this guide is to provide a thorough discussion on the theory and practi-
cal aspects of implementing combustion, vibration, and ultrasonic analyses on diesel
engines at nuclear power plants. This guide is intended to enable nuclear utility person-
nel to make informed decisions regarding combustion, vibration, and ultrasonic analy-
sis of diesel engines, including how to set up an effective program, how to establish
analysis guidelines, and how to make use of the resulting data to plan maintenance,
determine the causes of off-design operating conditions, and warn of potential compo-
nent failures.
1.2 Background
To date, diesel engine analysis has not seen consistent widespread use in the nuclear
power industry, in part because utility personnel interested in establishing an engine
analysis program have not had a single comprehensive source for technical guidance to
establish and evaluate it. At many of the locations where engine analysis has been used
in the nuclear power industry, it has been successful in identifying engine problems
before these problems progressed to catastrophic failures. For example, engine analysis
has successfully identified bent connecting rods, cracked cylinder heads, scuffed cylin-
der liners, leaking fuel injectors, collapsed valve lifters, and leaking valves in nuclear
standby emergency diesel generators (EDGs). In addition, engine analysis can and has
been used successfully to identify the following problems in diesel engines in other
industries:
Detonation
Cylinder load imbalance
Faulty fuel injection
Leaking valves and piston rings
Worn or scored cylinder liners
Worn rocker arms
Defective valve lifters
Worn cams
Damaged wrist pins
1-2
Engine analysis techniques were first used to support engine design and development
efforts. Diesel engine analysis has been used in the marine, rail, and pipeline industries
for many years and has proven cost-effective and successful in these industries. It has
been used to tune engines to improve fuel efficiency, support predictive maintenance
methods, assist with problem diagnosis, and support longer maintenance intervals.
1.3 Scope
This guide includes a thorough background on diesel engine combustion, vibration,
and ultrasonic analysis, theory, and interpretation of results, guidance on identifying
specific modes of degradation, examples of specific engine combustion, vibration, and
ultrasonic analyses for EDGs in nuclear standby service, troubleshooting guidance, and
a discussion of acceptance guidelines. In addition, engine analysis program and equip-
ment costs and equipment selection and setup are addressed.
This guide is applicable to selected makes and models of diesel engines in nuclear
standby service in the United States and around the world. These include:
ALCO model 251
Colt-Pielstick models PC2, PC2.3, and PC2.5
Cooper-Bessemer model KSV
Fairbanks Morse Opposed-Piston (FMOP) model 38TD 8-1/ 8
General Motors EMD model 645
Nordberg model FS-1316-HSC
SACM model UD 45
Cooper Enterprise (formerly TDI) models DSR and DSRV
Worthington model SWB12
Table 1-1 summarizes important characteristics of these engines.
1.4 Development
Development of this guide was sponsored by the Electric Power Research Institutes
Nuclear Maintenance Applications Center (EPRI/ NMAC) in order to provide consis-
tent, comprehensive, and detailed guidance for EDG system engineers on diesel engine
analysis. The guide was developed in cooperation with the various EDG owners
groups, engine and analysis equipment vendors, and a utility Technical Advisory
Group (TAG).
1-3
EPRI Licensed Material
T
a
b
l
e

1
-
1
C
h
a
r
a
c
t
e
r
i
s
t
i
c
s

o
f

E
n
g
i
n
e
s

i
n

U
.
S
.

N
u
c
l
e
a
r

S
e
r
v
i
c
e
M
a
k
e
A
L
C
O
C
o
l
t
-
P
i
e
l
s
t
i
c
k
C
o
o
p
e
r
-
B
e
s
s
e
m
e
r
C
o
o
p
e
r
-
E
n
t
e
r
p
r
i
s
e
E
M
D
F
a
i
r
b
a
n
k
s

M
o
r
s
e
N
o
r
d
b
e
r
g
S
A
C
M
W
o
r
t
h
i
n
g
t
o
n
M
o
d
e
l
2
5
1
P
C
2
,

P
C
2
.
3
,

P
C
2
.
5
K
S
V
D
S
R
D
S
R
V
6
4
5
O
p
p
o
s
e
d

P
i
s
t
o
n
3
8
T
D
8
-
1
/
8
F
S
-
1
3
1
6
-
H
S
C
U
D
4
5
S
W
B
1
2
C
o
n
t
i
n
u
o
u
s

P
o
w
e
r

R
a
t
i
n
g
(
k
W
)
1
7
5
0
2
6
0
0
4
0
0
0
5
5
0
0
3
5
0
0
8
8
0
0
2
1
0
0
2
8
5
0
2
7
0
0
3
0
0
0
3
5
0
0

-

4
0
0
0
2
7
0
0
B
M
E
P
2
0
1
2
6
5
1
9
4
2
1
4
2
2
5
1
9
9

-

2
2
7
W
o
r
k
i
n
g

C
y
c
l
e
4

s
t
r
o
k
e
4

s
t
r
o
k
e
4

s
t
r
o
k
e
4

s
t
r
o
k
e
2

s
t
r
o
k
e
2

s
t
r
o
k
e
4

s
t
r
o
k
e
4

s
t
r
o
k
e
4

s
t
r
o
k
e
#

o
f

C
y
l
i
n
d
e
r
s
1
6
/
1
8
1
2
/
1
4
/
1
6
1
6
/
2
0
8
/
1
6
/
2
0
1
6
/
2
0
1
2
1
6
1
6
1
2
?
B
o
r
e

(
i
n
)
9
1
5
.
7
4
1
3
.
5
1
7
9
-
1
/
1
6
8
-
1
/
8
1
3
.
5
9
.
4
5
S
t
r
o
k
e

(
i
n
)
1
0
.
5
1
8
.
1
1
1
6
.
5

(
m
a
s
t
e
r
)
1
7
.
1

(
l
i
n
k
)
2
1
1
0
1
0

(
e
a
c
h

p
i
s
t
o
n
)
1
6
.
5
8
.
6
6
A
s
p
i
r
a
t
i
o
n
T
u
r
b
o
c
h
a
r
g
e
d
T
u
r
b
o
c
h
a
r
g
e
d
T
u
r
b
o
c
h
a
r
g
e
d
T
u
r
b
o
c
h
a
r
g
e
d
G
e
a
r
-
D
r
i
v
e
n
T
u
r
b
o
c
h
a
r
g
e
d
G
e
a
r
-
D
r
i
v
e
n

B
l
o
w
e
r
/

T
u
r
b
o
c
h
a
r
g
e
d
T
u
r
b
o
c
h
a
r
g
e
d
T
u
r
b
o
c
h
a
r
g
e
d
T
u
r
b
o
c
h
a
r
g
e
d
P
i
s
t
o
n

A
c
t
i
o
n
S
i
n
g
l
e
-
A
c
t
i
n
g
S
i
n
g
l
e
-
A
c
t
i
n
g
S
i
n
g
l
e
-
A
c
t
i
n
g
S
i
n
g
l
e
-
A
c
t
i
n
g
S
i
n
g
l
e
-
A
c
t
i
n
g
O
p
p
o
s
e
d

P
i
s
t
o
n
S
i
n
g
l
e
-
A
c
t
i
n
g
S
i
n
g
l
e
-
A
c
t
i
n
g
S
i
n
g
l
e
-
A
c
t
i
n
g
C
y
l
i
n
d
e
r

A
r
r
a
n
g
e
m
e
n
t
4
5

V
4
5

V
4
5

V
I
n
l
i
n
e

(
D
S
R
)
4
5

V

(
D
S
R
V
)
4
5

V
I
n
l
i
n
e
4
5
V
5
0
V
V
?
C
o
m
p
r
e
s
s
i
o
n

R
a
t
i
o
1
1
.
5
1
3

(
P
C
2
)
1
1
.
5

(
P
C
2
.
5
)
1
0
.
6
1
6
1
3
.
5
#

o
f

I
n
t
a
k
e

V
a
l
v
e
s
2
2
2
2
0
0
1
2
2
#

o
f

E
x
h
a
u
s
t

V
a
l
v
e
2
2
2
2
4
0
1
2
2
C
o
n
n
e
c
t
i
n
g

R
o
d

T
y
p
e
M
a
s
t
e
r
M
a
s
t
e
r
A
r
t
i
c
u
l
a
t
e
d

M
a
s
t
e
r
-
L
i
n
k
A
r
t
i
c
u
l
a
t
e
d

M
a
s
t
e
r
-
L
i
n
k
M
a
s
t
e
r

(
F
o
r
k

&

B
l
a
d
e
)
M
a
s
t
e
r
M
a
s
t
e
r
A
r
t
i
c
u
l
a
t
e
d
M
a
s
t
e
r
-
L
i
n
k
M
a
s
t
e
r
C
o
n
n
e
c
t
i
n
g

R
o
d

L
e
n
g
t
h
(
i
n
)
2
1
M
a
i
n

4
1
A
r
t
i
c
u
l
a
t
e
d

4
0
.
9
3
8
T
i
m
i
n
g
:
I
V
O
/
I
P
O
I
V
C
/
I
P
C
E
V
O
/
E
P
O
E
V
C
/
E
P
C
F
u
e
l
8
0
.
0

B
T
D
C
3
7
.
5

A
B
D
C
5
7
.
5

B
B
D
C
6
0
.
0

A
T
D
C
2
5
-
2
7
.
5

B
T
D
C
8
0

B
T
D
C
4
0

A
B
D
C
4
5

B
B
D
C
5
0

A
T
D
C
2
7
-
2
9

B
T
D
C
7
5

B
T
D
C
2
5

A
B
D
C
6
0

B
B
D
C
4
0

A
T
D
C
2
2

B
T
D
C
4
5

B
B
D
C
4
5

A
B
D
C
7
7

B
B
D
C
1
1
9

B
T
D
C
5
0
-
1
1
0

B
T
D
C
4
0

B
B
D
C

t
o

2
0
A
B
D
C
5
0

B
B
D
C
E
V
C

6
5

A
T
D
C
#

o
f

N
u
c
l
e
a
r

P
o
w
e
r

P
l
a
n
t
s
9
9
1
6
1
0
3
1
2
2
3
2
2
N
u
m
b
e
r

o
f

E
n
g
i
n
e
s

i
n
U
.
S
.

N
u
c
l
e
a
r

S
e
r
v
i
c
e
2
6
1
9
3
6
2
0
8
2
4
7
8
2
4
S
p
e
e
d

(
r
p
m
)
9
0
0
5
1
4
6
0
0
4
5
0
9
0
0
9
0
0
5
1
4
1
2
0
0
C
y
l
i
n
d
e
r

D
i
s
p
l
a
c
e
d
V
o
l
u
m
e

(
i
n
3
)
6
6
8
3
5
2
8
2
3
6
2

(
m
a
s
t
e
r
)
2
4
4
3

(
l
i
n
k
)
4
7
6
7
6
4
5
1
0
3
7
2
3
6
2
6
0
7

(
m
a
s
t
e
r
)
2-1
2
BACKGROUND
2.1 Traditional Approach to Engine Maintenance
Emergency diesel generators in nuclear standby service utilize large, medium-speed
diesel engines driving 17508800 kW generators. Many of these diesel engines were
designed in the 1950s for service in marine propulsion, natural gas transmission, and
railroad locomotive applications. The base-loaded applications for which the engines
were designed involve long periods, as long as several weeks or months, of continuous
operation. EDGs in nuclear standby service, however, usually operate no more than
150200 hours per year. These hours of operation are usually accumulated during
periodic surveillance testing required by the U.S. Nuclear Regulatory Commission (U.S.
NRC) to demonstrate that the engines will start in the event of an emergency.
The engines are normally maintained in a standby condition with jacket water and
lubricating oil systems maintained at keep-warm temperatures of 100F170F to aid in
starting and to reduce wear during starting. Since surveillance tests last only about 14
hours each, the ratio of run time to number of starts for engines in nuclear standby
service is usually much higher than that for engines in commercial service (approxi-
mately 1 hour of operation per start for nuclear standby engines versus hundreds or
thousands of hours per start for commercial engines).
Commercial operators of diesel engines in marine propulsion, natural gas transmission,
railroads, and other base-loaded applications perform engine maintenance at scheduled
intervals since their diesel engines are base-loaded. This maintenance usually consists of a
series of engine teardowns and overhauls during which the engine and subsystems are
refurbished with new components after 8,00018,000 hours of operation or after fixed
intervals of time. It is difficult or impossible to predict wear out and failures, and a rela-
tively large amount of degradation is allowed to occur before maintenance actions are
required. In order to ensure that the statistical distribution of failures is addressed, compo-
nents are replaced early if a high confidence level is required. Only the complete failure of a
component would cause operators to deviate from the scheduled maintenance approach.
When engines were sold for use at nuclear power plants, the engine vendors developed
maintenance recommendations for nuclear standby engines based on their experience
with engines in base-loaded operation. When these maintenance intervals were devel-
oped, the fixed-time intervals were typically adopted without regard to the number of
engine operating hours anticipated during the intervals.
The result in some cases was that annual engine teardown requirements resulted in
component replacement after only 150200 hours of operation. Thus, components were
2-2
regularly replaced just following or during break-in. Reliability was therefore challenged
by maintenance-induced and infant mortality failures. Most U. S. nuclear power plants
have Technical Specification or license commitments to follow vendor-recommended
maintenance requirements for their EDGs. As a result, many utilities have had little or no
flexibility to create a maintenance program that reflects their experience, needs, and
operating histories. At the same time, nuclear utilities have been under tremendous
pressure to control maintenance costs while improving maintainability and reliability.
2.2 The Integrated Maintenance Approach
Condition-based, predictive, and performance-based maintenance methods have been
developed as an alternative to calendar-based maintenance. Condition-based maintenance
methods include visual inspections or dimensional measurements of components to
ascertain if wear or physical conditions are within the allowable limits. However, the
degree of engine disassembly required to take adequate, accurate measurements presents
the same maintenance-induced problems, downtime, and labor costs as a full engine
rebuild. Non-intrusive visual inspections of some components are possible. In any case,
condition-based methods cannot detect all failures or wear on-line before the failure occurs.
The premise of performance-based maintenance is that engine and generator degrada-
tion will be reflected in the performance of the EDG. Therefore, any wear or degrada-
tion that does not result in decreased performance is considered acceptable. Perfor-
mance-based maintenance is very effective at eliminating what could be called unneces-
sary maintenance. However, engine performance degradation must proceed to where it
is noticeable before any corrective action is identified. Thus, maintenance activities
cannot effectively be planned without experiencing periods of degradation or unsched-
uled unavailability.
Both condition-based and performance-based methods use single snapshots of EDG
condition. Experience demonstrates that a single snapshot of performance does not
produce a complete assessment of EDG condition. Predictive maintenance methods use
essentially the same data obtained from condition- and performance-based maintenance
methods, except that trending of analysis data is added. Data from inspections, perfor-
mance monitoring, and testing are trended over time.
Change analysis is used to detect differences in the data. It is normally assumed that
changes are a result of wear or degradation. If sufficient historical experience is avail-
able, the rate of any downward trends can be estimated, and the point in time at which
maintenance will be required can be predicted.
By looking forward, utilities can more readily plan maintenance activities. This can have
significant benefits where replacement parts have a long lead time or when minimal EDG
downtime is available to complete the maintenance. However, two potential downsides
are possible with predictive maintenance. First, non-fleet operators (those with one or
only a few engines) rarely have experience with all possible degradation modes. Even
fleet operators may not combine experience from multiple engines to base maintenance
predictions upon. Therefore, the nature of trends may be difficult to determine (that is, the
2-3
data and trends are apparent, but the meaning is not). Second, even small changes in
trends are typically evident. This can result in unwarranted conclusions that any changes
in trends are causes for remedial actions (that is, any unknown data pattern is taken as a
cause for concern).
The idea of integrated maintenance was developed to capitalize on the advantages of
each method while minimizing the downsides. Specifically, integrated maintenance
presumes that effective condition-based, performance-based, and predictive-based
maintenance tools (test equipment, procedures, manuals) and data (historical experi-
ence, trends, acceptance criteria) are in place.
Given this, conditions (wear or degradation) that are indicated by one test, inspection,
performance, or monitoring method are verified by one or more additional techniques.
Additional techniques or data can be used to further refine the diagnosis. In this way,
more precise and focused maintenance decisions can be made. Predictions regarding
degradation rate and EDG operability can be more accurate. This is largely due to the
fact that degraded machinery exhibits symptoms of component degradation rather than
component degradation itself. Another advantage of this method is that it forces the
cause of a failure to be assessed and recorded for future reference.
Some of the motivations for implementing an integrated maintenance program for
nuclear standby EDGs are to:
Increase EDG reliability by identifying and correcting degradation before failures
occur
Increase availability and reliability of EDGs through reduction of unnecessary or
intrusive maintenance actions
Reduce EDG maintenance costs by focusing maintenance efforts on known or sus-
pected problems
Most nuclear power plants already perform one or more elements of integrated mainte-
nance programs on EDGs, although few capitalize on the benefits of integration. For
example, many stations already analyze engine lube oil to determine both the condition
of the oil and the condition of the engine, and many plants regularly perform internal
borescopic inspections of EDGs.
2.3 Integrated Maintenance Programs
In addition to engine combustion, vibration, and ultrasonic analyses, several other
advanced techniques for evaluating engine condition and performance have been
developed and implemented by a wide variety of diesel engine operators. Some of these
techniques include:
Non-intrusive inspections of engine internal components using borescopes, video
probes, or thermographs
Exhaust gas emissions analysis
2-4
Trend line analysis of engine operating parameters
Periodic testing of engines or components
Jacket water chemical analysis
Lube oil chemical, ferrographic, and spectrographic analysis
Fuel oil chemical analysis
Fuel oil consumption measurement
Engine analysis is most effective when used in combination with one or more of these
additional techniques in an integrated maintenance program. The premise of such a
program is that these individual methods are useful in providing insight into particular
degradation mechanisms, but in only a few cases will individual methods positively
identify the specific component prior to a failure.
With an integrated maintenance program in place, conditions indicated by one test,
inspection, performance, or monitoring method are investigated using one or more
additional techniques (see Figure 2-1). In this way, more precise and focused mainte-
nance decisions can be reached. As many of the potential causes as possible are elimi-
nated using the additional non-intrusive methods, thereby reducing the amount of
engine disassembly necessary to find the cause of the problem. The symptoms of the
degradation are used to find the cause of the problem rather than searching for the
degraded component. This method forces the root cause of a failure to be determined
and recorded for future reference.
No
No
Yes
Yes
Symptom
Identify
Possible
Causes
Identify
System
Identify
Component
Schedule for Next
Maintenance
Window
Fix or
Replace
Action
Required
Now?
Other
Components
Likely to be
Affected?
Done
Fix or
Replace
Testing,
Monitoring,
Inspections,
Sampling
Figure 2-1
Integrated Maintenance Approach
2-5
For example if operating logs indicate low exhaust temperatures for one cylinder of an
engine, some or all of the following actions might be taken:
Check for a faulty or out-of-calibration thermocouple.
Measure cylinder peak firing pressure and peak firing pressure angle to check for
proper combustion.
Perform ultrasonic and vibration analyses of the cylinder to check fuel injection and
valve timing.
Perform a borescopic inspection of the cylinder to identify damaged or leaking
valves or plugged fuel injector nozzle holes (check the pattern on the piston crown).
Check the fuel injector nozzle opening (pop) pressure.
Monitor the crankcase vacuum to check for piston ring blowby.
Each of these additional techniques can be used to investigate potential causes of the
low exhaust temperature.
2.4 Engine Operability
For operators of commercial diesel engines, engine analysis results are simply a mainte-
nance tool used to help determine when engine maintenance is required and to trouble-
shoot potential problems. However, nuclear power plants are required to periodically
demonstrate the operability of EDGs and to maintain them in a state of readiness. As a
result, questions may arise regarding engine operability when engine analysis results
indicate that engine degradation has occurred or might have occurred.
Some plant personnel have expressed concerns that if operating limits or guidelines are
established, regulators will use them to question EDG operability or to force plants to
declare EDGs inoperable whenever the limits or guidelines are exceeded. It appears that
some plants have delayed or avoided the use of engine analysis due in part to this concern.
Some plant personnel advocate the use of engine analysis as an additional maintenance
tool to be used to enhance and focus existing maintenance capabilities and believe that
analysis results should not be used for operability decisions.
The Cooper-Bessemer Owners Group (CBOG) implemented a revised maintenance
program for Cooper-Bessemer Model KSV engines in nuclear standby service in 1992
that relies on engine analysis and performance monitoring. The revised maintenance
program was endorsed by Cooper-Bessemer and was issued as an update to the vendor
recommended maintenance program. An engine analysis guide has been developed
and published by CBOG as part of the revised maintenance program.
CBOG utilities gather engine analysis data for engineering and maintenance informa-
tion purposes. When a parameter does not meet the agreed-upon guideline (see Section
6), plant personnel take appropriate maintenance action, which may include scheduling
2-6
specific maintenance actions immediately or during the next scheduled outage, or
increasing monitoring until the problem can be fixed.
To date, none of the CBOG stations have declared an engine inoperable or entered an
EDG limiting condition of operation (LCO) based solely on engine analysis results.
Conditions that could affect reliability of engines are generally scheduled for corrective
maintenance. Several off-design conditions have been found through the use of engine
analysis, allowing utilities ample time to prepare for the necessary maintenance actions
during the next scheduled outage.
Most vendors for engines in nuclear standby service have not established limits for
many of the parameters recorded during engine analysis. Typically, the only limits that
exist are for engine balance parameters, such as cylinder peak firing pressures and
exhaust gas temperatures. Reasons for this include:
Engine balance criteria are typically based on an engine structural limitations, such
as cylinder head or cylinder head stud loads, turbocharger or piston crown melting
temperatures, and other structural limitations.
Engines used in nuclear standby service were designed before the availability or
widespread use of modern engine analysis equipment, and it was not anticipated
that operators would have the need for detailed engine performance criteria.
Interpretation of some engine analysis data may be considered qualitative due to the
nature of the data and how it is recorded.
Engine manufacturers might not be familiar with engine analysis techniques.
With these considerations in mind, one possible use of engine analysis results for oper-
ability considerations is as follows:
Engine analysis guidelines that are based on engine manufacturers engineering
design bases could be considered operating limits that should not be exceeded and
corrective actions should be taken if they are exceeded. For example, engine firing
pressure limits are typically based on the mechanical limits of the power assembly
and should not be exceeded. The limits on the balance of the firing pressures be-
tween cylinders are usually based on limiting excessive engine vibration and also
should not be exceeded.
Engine combustion, vibration, and ultrasonic analyses produce some results that are
qualitative in nature and that must be interpreted to reach any conclusions regard-
ing engine condition. Unlike engine balance, which is accomplished to ensure that
certain criteria are met, combustion, vibration, and ultrasonic analyses are used to
identify, localize, or troubleshoot known or suspected problems that might have
been identified during engine balancing. In addition, combustion, vibration, and
ultrasonic analyses can be used to focus intrusive maintenance activities. For ex-
ample, the results could be used to identify the worst power cylinders for additional
inspections during normal planned maintenance activities.
2-7
2.5 Degradation and Failure
All components have finite lives. In addition, most adjustments tend to drift somewhat
over time for a variety of reasons. One of the challenges of engine analysis is the ability to
distinguish between normal (anticipated) and abnormal (unanticipated) degradation, as
well as between acceptable and unacceptable degradation. To support these concepts,
methods have been developed for formulating acceptance guidelines, as well as methods
of analysis for parameters such as cylinder-to-cylinder variation, cycle-to-cycle variation,
and test-to-test variation. These are discussed in more detail in Sections 4 and 5.
3-1
3
ELEMENTS OF AN ENGINE ANALYSIS PROGRAM
3.1 Infrastructure
3.1.1 Management Support
As with any new program at a nuclear power plant, the first and most important ele-
ment necessary for the establishment of a successful engine analysis program is the
support of plant management. This support involves the commitment of utility re-
sources, personnel, time, and funding to the development and maintenance of the
program. The program will likely require the cooperation of several plant departments
to run the engine and collect, analyze, trend, and interpret the data. It will likely take
several engine analysis runs before personnel acquire the skills, techniques, and experi-
ence necessary to take full advantage of engine analysis as a maintenance tool.
Plant management should also recognize that it might be some time before any payoff is
seen from engine analysis in terms of a reduction in intrusive maintenance or the dis-
covery of an impending failure, although several utilities have had very good early
successes that largely paid for the entire engine analysis program investment.
3.1.2 Engine Analysis Equipment
In order to perform engine analysis of a diesel engine, an engine analyzer is needed. The
capabilities, features, and cost of engine analyzers vary from manufacturer to manufac-
turer and from model to model, just as they do with most equipment. Since engine ana-
lyzer vendors generally do not rent or lease engine analysis equipment, two options are
available for nuclear power stations wanting to establish an engine analysis program.
The first option is to establish an in-house engine analysis program, where utility per-
sonnel collect, trend, and analyze the engine analysis data. In this case, purchase of
engine analysis equipment is necessary. The cost of portable engine analyzer equip-
ment, capable of measuring and recording cylinder pressure, vibration signatures, and
ultrasonic signatures as a functions of crankshaft angle, currently ranges from approxi-
mately $25,000$70,000.
Table 5-1 includes a summary of available engine analysis equipment, including cost
and capabilities of the various analyzers. In addition, continued funding will be re-
quired for equipment repairs and upgrades, training classes, software, and any neces-
sary calibration equipment.
3-2
The second approach available is to contract for engine analysis service. Many of the
engine analysis equipment vendors and several consulting companies offer contract
engine analysis services for fees ranging from $4,000$10,000 per engine. This cost
typically includes data analysis and a final report that documents engine maintenance
recommendations. One caution to this option is that it is important to determine where
and how often a contractor has successfully performed engine analysis on the particular
engine make and model installed at the nuclear power plant. A contractors lack of
familiarity with off-design operating conditions could result in a great deal of wasted
time and effort.
Many of the engine analysis equipment vendors will provide demonstrations of their
equipment for station personnel. The demonstration sometimes includes analysis of one
EDG, thus providing an introduction to engine analysis
At first glance, a quick comparison of the costs associated with each of these options
might indicate that contracting for engine analysis services is the cheaper option. How-
ever, the number of engines to be analyzed and the frequency of the analysis must also
be considered in the financial assessment. For utilities with a small number of tests to be
performed, contracting for engine analysis services might be cost-effective, although the
ability to quickly address emergent problems may be lost. For utilities with more tests
to be performed, it is likely that purchase of engine analysis equipment will be more
cost-effective.
If contract services are chosen, it is recommended that a long-term relationship be
established with one company due to the emphasis on trending and specific engine
experience necessary for a successful engine analysis program.
If the decision is made to purchase engine analysis equipment, the station personnel
responsible for developing the engine analysis program are faced with selecting the
optimal equipment from the wide variety of systems offered. However, with the ongo-
ing efforts to limit nuclear power plant operating and maintenance costs, financial
limitations heavily influence equipment purchases. The types of equipment available
range from portable engine analyzers, which can be easily carried and operated by one
person, to large systems of permanently mounted test instrumentation capable of re-
cording dozens of parameters. Ideally, the engine analyzer chosen should have the
following characteristics:
Portability
Digital storage of data
Data export capability
Good human factors
Support calibration
Built-in trending and analysis of stored data
Low maintenance
Measurement or calculation of all parameters of interest
3-3
3.1.3 Personnel
Whether engine analysis is performed by utility personnel or by contract personnel, it is
recommended that the utility dedicate personnel to the engine analysis program, on
either a full-time or part-time basis, and allow these personnel the time needed to de-
velop the data collection and analysis skills required for successful engine analysis.
Station management will endorse the engine analysis program and station personnel
will be in control of the program only if station personnel are knowledgeable in engine
analysis.
It will likely take several engine analysis runs and training sessions for these personnel
to become comfortable gathering data and performing analyses, and initial program
implementation will likely require a significant amount of engineering effort. As per-
sonnel become familiar with engine analysis techniques, the time required for data
collection and analysis can be expected to decrease.
The first time engine analysis is performed on an EDG at a nuclear power plant, it can
be expected to take up to 8 person-hours to complete the data collection and 20 person-
hours to analyze the data, assuming no major anomalies are found in the engine. After
personnel are experienced with the techniques, these times can be expected to decrease
to approximately 6 and 8 person-hours respectively.
3.1.4 Equipment Calibration and Maintenance Program
Engine analysis equipment must be calibrated and maintained in order to ensure the
accuracy of the data collected. Because the data will be used to support EDG mainte-
nance and operating decisions, it is critical that data collection errors be minimized. An
effective calibration program identifies equipment problems before the equipment is
used to collect data, helping to avoid misdiagnosis of EDG problems. Both the data
collection unit and the pressure transducers and accelerometers require calibration.
In addition to regular periodic calibration of engine analysis equipment, calibration
checks should be performed following data collection to ensure that instruments have
not drifted.
Maintenance of engine analysis equipment is crucial in order to ensure that scheduled
engine analysis runs are not delayed by equipment failures. Many of the engine analysis
equipment vendors offer extended warranties on their equipment. In addition to cover-
ing necessary repairs or replacement, these warranties may include hardware and
software upgrades for a specified period of time.
Obviously, if engine analysis services are contracted, equipment calibration and mainte-
nance are the responsibility of the contractor. However, station personnel should ensure
that the contractor arrives at the plant with calibrated and operable equipment. There
have been instances where engine analysis tests have been delayed due to missing,
failed, or out-of-calibration equipment.
3-4
3.1.5 Training
Most engine analysis equipment vendors and many independent contractors offer
engine analysis training. Two types of training are typically given. The first is training
on the use of the particular engine analysis equipment, including how to set up, install,
and use the equipment and its associated software. This training is generally given by
the equipment vendor and much of it is specific to the make and model of the engine
analysis equipment used. Actual engine analysis, is typically not addressed. The second
type of training available is training on the interpretation and analysis of the data col-
lected. This type of training is typically available from both engine analysis equipment
vendors and independent contractors.
It is important to obtain training specific to the type of engine to be analyzed. Engine
analysis methods vary significantly between natural gas and diesel engines and be-
tween two-stroke and four-stroke engines.
3.2 Programmatic Issues
3.2.1 Integration with Other Maintenance Programs
As discussed in Section 2, engine analysis is most effective when it is used in combination
with other non-intrusive maintenance or inspection techniques, such as lube oil analysis.
However, it takes a combined effort by several departments at a nuclear power plant to
utilize the inspection techniques and draw conclusions from the data. In order for an
integrated maintenance program to be successful, it must have the backing of plant man-
agement to allocate the resources and ensure cooperation among the various departments.
3.2.2 Parameters to Monitor
With modern equipment, it is possible to monitor hundreds or even thousands of pa-
rameters. However, just because it is possible to do something does not mean that it is
appropriate to do so. Diesel engine performance parameters that should be monitored
include those that:
Are relatively easy to monitor
Can have potential responses developed
Can indicate and warn of anticipated failures
Can be used to develop or calculate additional performance parameters that are
candidates for monitoring
Suggested parameters for monitoring in a diesel engine analysis program include:
Cylinder pressures (compression and firing) versus crankshaft position
Vibration signatures versus crankshaft position
Ultrasonic signatures versus crankshaft position
Engine data such as auxiliary system temperatures and pressures
3-5
Most modern digital engine analyzers are capable of recording cylinder pressure and
vibration and ultrasonic signatures as functions of crankshaft position. Auxiliary system
temperatures and pressures are typically recorded for EDGs at nuclear power stations
as part of the EDG operating logs. However, several stations now have the capability to
monitor and record these parameters on installed computer-based monitoring systems.
3.2.3 Monitoring Frequency
In addition to being capable of monitoring thousands of parameters, modern equip-
ment is also capable of monitoring parameters on a near-continuous basis. Once again,
however, just because something is possible does not mean that it is appropriate. In-
stead, the monitoring frequency for both engine analysis and other maintenance meth-
ods should be determined by the lead time required to evaluate the information and
respond to any potential incipient failures with sufficient time to prevent equipment
damage. As a result, the periodicity will vary from site to site, depending on the mainte-
nance methods used, the expertise of station personnel, and the manpower available.
Stations that have established engine analysis programs, typically perform engine
analysis two or three times per fuel cycle. Engine analysis is performed following out-
age work to confirm EDG performance at the beginning of the cycle. In plants with an
18-month or 24-month fuel cycle, engine analysis is often performed again at the mid-
point of the cycle. Finally, engine analysis is performed a few months before each refu-
eling outage both to confirm engine condition and to identify any components requiring
maintenance or inspection during the upcoming outage.
The post-maintenance engine analysis should be completed and problems corrected
prior to performing surveillance tests that could lead to valid start or load failures. In
this case, engine analysis is used to show that outage maintenance was performed
properly and that the engine was reassembled correctly. In addition, engine analysis
can be used to investigate or diagnose specific problems that occur at other times dur-
ing the operating cycle.
3.2.4 Standardized Data Collection Methods
Since much of the analysis of engine analysis data consists of trending and comparing
the data, it is critical that variations in measured engine parameters due to external
factors be minimized. External factors include:
Equipment calibration
Data collection techniques
Differences in equipment
Changes in engine load
Environmental factors.
Equipment calibration was discussed in Section 3.1.4 and is addressed in detail later in
this guide.
3-6
Differences in analysts techniques have the potential to affect engine analysis results
and, possibly, the conclusions drawn from the results. For example, each analyst might
take vibration and ultrasonic signatures at a different location on the cylinders. As a
result, the vibration and ultrasonic signatures likely are not directly comparable, and
incorrect conclusions might be drawn regarding the condition of one or more cylinders.
Therefore, it is recommended that vibration and ultrasonic signatures be taken in the
same location on each cylinder and during each engine analysis run.
To date, there is no industry standard for data collection, processing, and display. Some
engine analyzers may process the recorded data differently than others do. Many of the
analyzers filter and process cylinder pressure signals and vibration and ultrasonic signa-
tures before recording or displaying the data. If the equipment used on a particular en-
gine is changed from one analysis run to the next, the changes in the data might be
masked by changes in the data processing performed by the engine analyzer. Should it be
necessary to change the type of analyzer used from one engine analysis run to the next,
the filtering and processing performed by both analyzers should be well understood.
Variations in engine load directly affect individual cylinder loads. As a result, any
change in engine load during the engine analysis run affects the data collected. In order
for cylinder-to-cylinder trending to be effective, the engine load during data collection
for each cylinder must held constant. In addition, for trending to be effective from one
engine analysis run to the next, total engine load must be the same for each run.
One method to ensure that engine load is constant is to lock the governor. When the
governor is locked, the load limit knob on the governor actuator is used to minimize
load swings. Use of the load limit knob prevents the fuel racks from moving unless the
EDG undergoes a complete load rejection or overspeeds; it does not disable any of the
safety features of the EDG. However, it might not be appropriate to lock the governor at
all sites.
If the governor cannot be locked, operators must pay close attention to generator load
during engine analysis testing to minimize load swings. If the governor is locked during
engine analysis, the load limit knob must be returned to its original setting at the con-
clusion of the data collection.
Variations in engine inlet air temperature from season to season, or even from day to
night, affect engine analysis data. Peak firing pressures and exhaust temperatures are
higher when ambient air temperatures are higher. These changes can complicate data
analysis and, in some cases, make a well-balanced engine appear to be imbalanced.
Therefore, it can be useful for stations to establish methods to adjust engine analysis
results for varying environmental conditions.
One method of adjustment is through the use of engine performance maps. A thorough
discussion of engine performance maps is outside the scope of this guide; however,
more details can be found in References 1 and 2.
3-7
3.3 Analysis
3.3.1 Trending
Much of the analysis of engine analysis data consists of trending results from one cylin-
der or engine to another or from one test to another. As a result, one of the necessary
skills of an engine analyst is that of interpreting changes. Engine analysts spend a good
portion of their time comparing newly recorded engine analysis data to previously
recorded data in an effort to identify signs of engine degradation or faults.
This is particularly true for analysis of vibration and ultrasonic signatures. Vibration
and ultrasonic signature analyses are qualitative processes where comparisons are
made between different signatures. Changes in timing and amplitude of engine events
are possible indications of changes in the condition of engine components. In addition,
the appearance of new events or the disappearance of previously seen events are also
possible indications of changes in engine material condition.
3.3.2 Baseline Data
Most of the manufacturers of diesel engines used in U.S. nuclear service have not pub-
lished or released typical or expected values for the parameters measured during en-
gine analysis. As a result, individual nuclear power stations or diesel owners groups
have to determine what the baseline or normal values for these parameters are. This can
be done by performing engine analysis on an engine that is known to be well-balanced
and in good mechanical condition.
Baseline engine analysis data is necessary as the starting point for parameters that are
trended and for vibration and ultrasonic analysis that is comparison based. If data for a
well-running engine are not available, it might not be possible to identify a poorly
running engine.
It should be noted that baseline data collected on one engine might not be applicable to
another engine of the same make and model due to configuration differences or differ-
ences in rated load. Name plate data and operating data for both engines should be
compared prior to using baseline engine analysis data from one engine on another
engine.
3.3.3 First Time Engine Setup
Before data that can be considered baseline data can be collected, the engine must be in
good operating condition and well-balanced. As a result, it might be necessary to bal-
ance or tune the engine prior to running it for the purpose of collecting baseline data.
The state of engine balance can be checked by performing engine analysis.
If the analysis shows that the engine is well-balanced, this data can be considered baseline
data. If it turns out that the engine is not well-balanced, it must be balanced prior to the
collection of baseline data. Section 5 of this guide discusses engine balancing.
3-8
3.3.4 Acceptance Guidelines
For some performance parameters monitored during engine analysis, numerical limits
are appropriate and are either available or can be developed. These numerical accep-
tance guidelines are used as performance criteria to determine when engine mainte-
nance or adjustments are needed. Parameters for which numerical acceptance guide-
lines are appropriate include:
Engine average peak firing pressure
Maximum range of cylinder average peak firing pressures
Engine average peak firing pressure angle
Maximum range of peak firing pressure angles
Engine average fuel injection timing
Maximum range of engine fuel injection timing
Maximum range of fuel rack settings
Fuel injector nozzle pop pressures
Maximum cylinder exhaust gas temperature
Maximum range of exhaust gas temperatures
While several of the engine vendors have published desired or expected values for
some of these parameters, only Cooper-Bessemer has established guidelines for all of
them. For other engines, analysis guidelines must be developed. Potential sources of
analysis guidelines include:
Engine manufacturer design data
Baseline data collected on a well-running (balanced) engine
Diesel engine owners group development
Outside consultant development
Nuclear power station development
More details on the development of engine analysis guidelines are contained in Section
4 of this guide.
3.3.5 Case Histories
A library of case histories is a particularly useful tool for the engine analyst. Case histo-
ries are examples of data taken during engine analysis runs of engines with known
faults. The purpose of the data is to illustrate analysis results that are indicative of
specific modes of engine degradation or poor engine analyzer setup. While none of the
case histories exactly match results for a given engine, the characteristics of the curves
due to the specific mode of degradation are identified in the case histories and can be
compared to data collected during engine analysis runs to identify similarities to assist
in the identification of the specific engine fault.
3-9
If engine analysis identifies an engine fault that is subsequently corrected, the engine
analysis data that identified the fault should be added to the library of case histories. If
a fault occurs that was not identified beforehand by engine analysis, the engine analyst
should review previously collected engine analysis data to determine if the fault pro-
vided indications of its presence, but was not detected. Again, any data that are found
to provide warning of the fault should be added to the library of case histories.
Sharing case histories among stations with the same make of diesel engine is an easy
and cost-effective way to gain the benefits of experience without suffering the associ-
ated failures directly. If one station suffers an engine failure that provided warning of
its occurrence through engine analysis, other stations with similar engines need not
suffer the same failure if the case history is made available to them.
In addition to their benefit for future fault identification, case histories provide a written
record of the successes and capabilities of an engine analysis program.
3.3.6 Analysis Flow Charts
Analysis flow charts consist of step-by-step instructions on how to perform engine
analysis. Generally, they are specific to each make and model of diesel engine. The
analysis flow charts are not intended to take the place of station procedures, but are
intended to assist operators and system engineers by ensuring that the analysis and
evaluation of engine data is thorough and repeatable and that any troubleshooting steps
are accomplished in an efficient and logical order.
It is recommended that the individual diesel owners groups develop or fund develop-
ment of analysis flow charts for their specific engine make and model to ensure that the
stations take the same general approach to engine analysis. This will assist in making
engine analysis results transferable from one station to another.
3.3.7 Feedback from Maintenance Personnel
When engine analysis is performed, the analyst should make an effort to be present for
the data collection to be aware of any anomalies or difficulties experienced in the data
collection effort. If the analyst is aware of conditions during the engine analysis run, it
can eliminate later questions regarding the data collected.
In addition to witnessing the data collection, the analyst should make an effort to get
feedback from station maintenance personnel regarding problems experienced with the
engine, both during engine analysis and during normal standby or operating condi-
tions. When the experience of the maintenance personnel is combined with the engine
analysis data, it can provide insight to engine faults that would otherwise be missed.
4-1
4
THEORY
4.1 Combustion
The diesel engine is a reciprocating internal combustion engine. Each mechanical cycle
of a diesel engine involves intake, compression, expansion, and exhaust of a fuel-air
mixture. Engines are classified by the type of mechanical working cycle used. Two
working cycles are used in diesel engines for U.S. nuclear standby service: the two-
stroke cycle and the four-stroke cycle. Engines that use a two-stroke cycle are called
two-stroke engines. Engines that use a four-stroke cycle are called four-stroke engines.
A stroke is defined as the motion of the piston between the points of minimum and maxi-
mum cylinder volume, referred to as top dead center (TDC) and bottom dead center
(BDC), respectively (see Figure 4-1). Each stroke consists of 180 of crankshaft rotation. At
TDC and BDC, the piston reverses direction and is, for an instant, motionless.
4-2
Stroke
TDC
BDC
Bore
Bore
TDC
0
270 90
180
BDC
TDC
0
270 90
180
BDC
Cylinder at TDC Cylinder at BDC
Figure 4-1
Piston Stroke: Top and Bottom Dead Center
4-3
The Fairbanks Morse Opposed-Piston (FMOP) engine has two crankshafts that are geared
together so that the lower piston reaches inner dead center (IDC) or outer dead center
(ODC) several degrees before the upper piston. This difference is referred to as the crank
lead. As a result of the crank lead, the positions of minimum and maximum cylinder
volume in a FMOP engine occur just after the lower piston has passed through IDC or
ODC, but before the upper piston has passed through IDC or ODC (see Figure 4-2).
Outer Dead
Center
Injection
Nozzle
Opening
Lower Crank
Lead
Lower
Piston
Exhaust
Ports
Combustion
Space
Inlet
Ports
Upper
Pistons
Inner Dead
Center
Lower
Piston
Figure 4-2
FMOP Engine: Inner and Outer Dead Center Arrangement
4-4
4.1.1 Two-Stroke Engines
In a two-stroke engine, the mechanical cycle is completed in two strokes of the piston or
one revolution (360) of the crankshaft. The two strokes are (see Figure 4-3):
A compression stroke occurs when the piston compresses the cylinder contents as it
moves from BDC to TDC (or from ODC to IDC for an FMOP engine), during which
the inlet ports and exhaust ports/ valves are closed. Shortly before the end of the
stroke, fuel is injected and combustion begins.
A power or expansion stroke occurs when the high-pressure products of combustion
drive the piston from TDC to BDC (or IDC to ODC for an FMOP engine). Before the
piston reaches BDC, the inlet ports and exhaust ports/ valves are opened to exhaust
the combustion products and take in a new charge of fresh air. The cycle then re-
peats.
Compression Stroke
Scavenging Compression
Power Stroke
Power Exhaust
Figure 4-3
Two-Stroke Engine Cycle
4-5
Because combustion occurs once per crankshaft revolution in a two-stroke engine, air
intake and exhaust flow must also be completed during each revolution. The exchange
of exhaust gas from one combustion event for a charge of fresh air for the next combus-
tion event must take place during a short time period. As a result, two-stroke engines
normally use some type of forced induction in order to sweep out or scavenge the ex-
haust gases from the combustion chamber. Several gas exchange, or scavenging pro-
cesses are used in two-stroke diesel engines. These include:
Loop Scavenging (see Figure 4-4). In a loop-scavenged engine, the inlet and exhaust
processes are controlled by ports in the cylinder wall. A single piston covers and
uncovers the ports, allowing intake and exhaust to occur. While loop scavenging is
the simplest scavenging method, it has several disadvantages. The symmetrical
timing leads to mixing of the inlet air and combustion products, reducing the purity
of the air charge. None of the engines used in nuclear standby service in the U.S. use
loop scavenging.
Intake
Port
Exhaust
Port
Figure 4-4
Loop Scavenging
4-6
Uniflow Scavenging with Single Piston (see Figure 4-5). In the uniflow-scavenged,
single-piston engine, air intake is controlled by ports in the cylinder wall, and the
exhaust is controlled by camshaft-operated exhaust valves. The advantages of this
system are that the inlet air tends to displace, rather than mix with, the combustion
products, and the timing of the inlet and exhaust processes can be adjusted to pre-
vent loss of the fresh air charge to the exhaust. The EMD 645 is a uniflow-scavenged,
single-piston engine.
Intake
Port
Intake
Port
Exhaust
Valve
Exhaust
Valve
Figure 4-5
Uniflow Scavenging with Single Piston
4-7
Uniflow Scavenging with Opposed Pistons (see Figure 4-6). In a uniflow-scavenged,
opposed-piston engine, the inlet ports are controlled by one piston and the exhaust
ports are controlled by the other piston. This system has the same advantages as the
uniflow-scavenged, single-piston engine but is more complicated mechanically. Its
advantage is the high specific power output that can be achieved from each cylinder
with two pistons. The FMOP engine is a uniflow-scavenged, opposed-piston engine.
Exhaust
Port
Exhaust
Port
Lower
Piston
Upper
Piston
Inlet
Port
Inlet
Port
Figure 4-6
Uniflow Scavenging with Opposed Pistons
4-8
4.1.2 Four-Stroke Engines
In a four-stroke engine, the mechanical cycle is completed in four strokes of the piston
or two revolutions (720) of the crankshaft. These four strokes are (see Figure 4-7):
An intake stroke occurs when a charge of fresh air is forced into the cylinder as the
piston moves from TDC to BDC. The inlet valves open shortly before the intake
stroke begins and close shortly after it ends.
A compression stroke compresses the air charge as the piston moves from BDC to
TDC. Toward the end of the compression stroke, fuel is injected into the cylinder
and combustion begins.
A power or expansion stroke occurs when the high-pressure products from combus-
tion drive the piston from TDC to BDC. The exhaust valves open before the end of
the power stroke.
An exhaust stroke occurs when the piston moves from BDC to TDC. There is typi-
cally some residual pressure remaining in the cylinder at the end of the power
stroke, so the exhaust stroke begins with an initial rush of combustion products past
the exhaust valves. The inlet valve opens toward the end of the exhaust stroke, and
the exhaust valves close shortly after the exhaust stroke ends. The cycle then repeats.
Intake Stroke Compression Stroke Power Stroke Exhaust Stroke
Figure 4-7
Four-Stroke Cycle
Because combustion occurs every other revolution of the crankshaft in a four-stroke
engine, the remaining revolution can be used for gas exchange. The ALCO 251, Colt-
Pielstick PC2 series, Cooper-Bessemer KSV, Cooper-Enterprise DSR and DSRV, SACM
UD45, and the Worthington SWB12 are four-stroke engines.
4-9
4.1.3 Four Phases of Combustion
For discussion purposes, the combustion process in a diesel engine is typically divided
into four phases [3]. The four phases are described below and shown in Figures 4-8 and
4-9, which are plots of the cylinder pressure and rate of heat release as functions of
crank angle. The plots shown are typical of both two- and four-stroke engines.
The ignition delay phase is the time between the start of fuel injection and the start
of combustion. Fuel is injected into the cylinder at a temperature that is necessarily
below its auto-ignition temperature of approximately 950F [4]. During the ignition
delay phase, the injected fuel is atomized, vaporized, mixed with air, and raised to
its auto-ignition temperature.
When the fuel has reached its auto-ignition temperature, combustion occurs and the
rapid combustion phase begins. This phase is characterized by a rapid rise in pres-
sure due to the rapid combustion of the fuel that mixed with air and reached auto-
ignition conditions during the ignition delay phase. In addition, some of the fuel that
enters the cylinder during this phase reaches ignition conditions and burns during
this phase. Heat released by the burning fuel produces increased cylinder pressure.
This pressure acts on the surface of the piston, which creates torque on the crank-
shaft. The rapid combustion phase ends when the peak firing pressure is reached.
The third stage is the mixing-controlled combustion phase. The premixed fuel sup-
ply has been consumed during the rapid combustion phase, and the rate of combus-
tion during the mixing-controlled combustion phase is controlled by the rate of fuel
injection and fuel-air mixing. This phase ends when fuel injection ends.
The final stage is the late combustion stage. In this stage, heat release continues into
the expansion stroke as unburned and partially burned fuel burn when they come
into contact with oxygen. However, cylinder pressure declines rapidly as piston
motion increases the cylinder volume.
In a typical diesel engine, approximately 80% of the heat release occurs during the rapid
and mixing-controlled combustion phases [3].
4-10
1600
1400
1200
1000
800
600
400
200
0
C
y
l
i
n
d
e
r

P
r
e
s
s
u
r
e

(
p
s
i
)
0 60 120 180 240 300 360 420 480 540 600 660 720
Crank Angle (degrees)
Ignition Delay
Start of Injection
Start of Combustion
Rapid Combustion Phase
Mixing-Controlled
Combustion Phase
Late Combustion Phase
Figure 4-8
Cylinder Pressure Versus Crankshaft Angle
Ignition
Delay
Period
SOI
Rapid Combustion Phase
Mixing-Controlled
Combustion Phase
Late
Combustion
Phase
R
a
t
e

o
f

H
e
a
t

R
e
l
e
a
s
e
340 350 360 370 380 390
Figure 4-9
Rate of Heat Release Versus Crankshaft Angle
4-11
4.1.4 Engine Parameters
The engine output torque drives the generator. Engine torque is normally measured
with a dynamometer, which is a mechanical, electromagnetic, or hydraulic load at-
tached to the engine crankshaft. The power delivered by the engine to the dynamometer
can be measured. This power is known as the brake power (P
b
) because the dynamometer
acts as a brake during the test. The brake power can be divided by the engine speed (for
two-stroke engines) or by one-half the engine speed (for four-stroke engines) to calcu-
late the brake work per cycle.
Cylinder pressure data can be used to calculate the power transfer from the cylinder gas
to the piston. The energy released by the combustion of the fuel in the cylinder is con-
verted to work through the action of gas pressure on the pistons. By definition, work is
produced when a force (F) acts through a distance (x).
W Fdx =
The force (F) is produced by cylinder pressure acting on the piston surface. Substituting
pressure (P) times piston surface area (A) for the force (F) and the change in cylinder
volume (dV) divided by piston area (A) for the distance traveled (dx) yields
W Fdx PA
dV
A
PdV = = =

Using the above equation, cylinder pressure data can be used to calculate the work
performed by the gas on the piston. For a complete mechanical cycle, the work per-
formed in the cylinder is given by
W PdV
i
=
Eq. 4-1
If cylinder pressure is plotted as a function of cylinder volume, the area inside the curve
represents the indicated work. For a two-stroke engine, where the pressure versus
volume curve is a single loop (see Figure 4-10), it is quite easy to determine the indicated
work (W
i
). For a four-stroke engine, where the diagram consists of two loops (see Figure
4-11), the area in the upper loop is the positive work from compression and combustion
while the lower loop represents the pumping work to exhaust the combusted gases and
draw in a fresh charge of air. The pumping work is always negative on normally aspi-
rated engines, but may be zero or positive on highly turbocharged diesel engines if the
inlet manifold pressure is greater than the exhaust manifold pressure.
4-12
1800
1441
1082
722
363
4
0 20 40 60 80 100
Swept Volume (%)
P
r
e
s
s
u
r
e

(
p
s
i
)
Figure 4-10
Two-Stroke Pressure Versus Volume Curve
1600
1400
1200
1000
800
600
400
200
0
P
r
e
s
s
u
r
e

(
p
s
i
)
0 500 1000 1500
Volume
2000 2500 3000
Figure 4-11
Four-Stroke Pressure Versus Volume Curve
4-13
The work performed by the gas on the piston is known as the indicated work (W
i
). The rate
at which all indicated work is done is known as the indicated power (P
i
), and is given by

P
WN
n
i
=
where
W is the indicated work calculated from Equation 4-1
N is the engine speed
n is the number of crankshaft revolutions per power cycle
(2 for 4-stroke engines and 1 for 2-stroke engines).
Indicated power is usually given in units of horsepower and referred to as indicated
horsepower (IHP).
The difference between the brake power and the indicated power represents the friction
power. The friction power includes the power necessary for gas exchange, power used in
overcoming friction between engine components, and the power necessary to drive
engine auxiliaries such as pumps. The ratio of the brake power to the indicated power is
called the mechanical efficiency of the engine:

=
P
P
b
i
The terms brake and indicated are used to describe many other engine performance
parameters. When discussing engine performance parameters, it must always be under-
stood whether a given parameter is a brake or indicated value.
The indicated mean effective pressure (IMEP) is defined as the indicated work divided by
the cylinder volume displaced by the piston during each cycle. It is a measure of torque
per unit displaced volume. IMEP is also equal to the constant pressure that, if applied
only during the entire power stroke, would yield work equal to the work of the com-
plete cycle (see Figure 4-12). It is calculated as

IMEP
W
V V
W
V
i i
=
=
2 1
Delta V is the cylinder volume swept by the piston (equal to the piston stroke times
cylinder cross-sectional area). V
2
is the maximum cylinder volume (at BDC). V
1
is the
minimum cylinder volume (at TDC).
4-14
Rearranging and substituting Equation 4-1,

IMEP
P n
N V
i
=
Eq. 4-2
Enclosed
Areas are
Equal
Constant
Pressure
During Power
Stroke
1500
1250
1000
750
500
250
0
P
r
e
s
s
u
r
e
0 500 1000 1500 2000 2500 3000
Volume
Figure 4-12
Indicated Mean Effective Pressure
Brake mean effective pressure (BMEP) can be calculated similarly using brake power.
4.1.5 Measurement Methods
Historically, engine manufacturers have used cylinder pressure and volume data in
engine research and design, in particular to determine the work done in the cylinder per
cycle using Equation 4-2. Peak pressures in large, slow-speed engines could be mea-
sured using pressure gages and check valves. Engine indicators were used to plot cylin-
der pressure versus cylinder volume curves for one mechanical cycle. These plots be-
came known as indicator diagrams or indicator cards. An indicator used a piston and
cylinder connected to the engine so that the engine cylinder pressure acted on the indi-
cator piston. The indicator piston was connected to a pencil that recorded the cylinder
pressure on a card that was given a reciprocal motion proportional to the movement of
the engine piston. However, both peak pressure gages and engine indicators are only
useful for slow speed engines.
Instruments that measure the pressure at only one point during the mechanical cycle
are also available. By determining the pressure at many points throughout the cycle, a
4-15
complete history of the cylinder pressure during the cycle can be created. However,
because it takes several hundred cycles to create a single pressure history, this method
was less accurate because the instantaneous cylinder pressure varies from one cycle to
the next. As reviewed in Section 6, variations in peak cylinder pressures of up to 2%
from one cycle to the next are not unusual in diesel engines.
Fortunately, the development of fast acting pressure transducers has allowed combus-
tion analysis techniques to be improved and used on faster running engines. These
transducers rely on materials that convert mechanical deformation into electrical sig-
nals. The most common types in use for engine combustion analysis are resistive
(strain-gage), capacitive, and piezoelectic transducers. On most engines, the transducer
is connected to an indicator valve, which is connected to the cylinder by an indicator
passage machined through the cylinder head. A detailed discussion of pressure trans-
ducer characteristics and use is included in Section 5.
Cylinder volume is a function of engine geometry and the position of the piston in the
cylinder. Piston position can be calculated using engine geometry data and crankshaft
position or angle. The crankshaft position information comes from a reference signal
derived from the crankshaft. There are two methods commonly used to obtain the crank
angle position. The first method is to use a once-per-revolution pulse that is taken from
an interruption device. An interruption device uses some anomaly that is produced
once with each revolution of the shaft. This interruption is usually positioned at the
TDC of the number one piston for reference purposes. The anomaly can be anything
from a reflective piece of tape to a high or low spot on the flywheel. This once-per-
rotation interruption is used to create an electrical pulse to indicate each time that point
passes a pickup or sensor.
The once-per-turn method has one potentially significant disadvantage. With only one
trigger per revolution, the intermediate angles are interpolated assuming constant
rotational velocity. Therefore, if the engine does not turn a consistent angular velocity,
P- data accuracy will be affected. To overcome this uncertainty, it is preferable to use a
shaft encoder. This type of crankshaft position device provides a pulse train at fixed
angular increments in addition to the once-per-turn pulse required for absolute posi-
tioning. Commonly there is one pulse per degree, resulting in 360 reference points per
revolution.
The most common method of measuring crank angle is through the use of optical or
magnetic encoders. Many engine analyzers are supplied with an encoder tailored to the
analyzer.
4.2 Combustion Analysis
4.2.1 Combustion Curves
Several different combustion curves can be created from the cylinder pressure versus
crankshaft angle data. The characteristics and features of these curves can be used to
evaluate engine condition.
4-16
4.2.1.1 Cylinder Pressure Versus Crank Angle Curve (P-)
The starting point of diesel engine analysis is the cylinder pressure versus crank angle,
or P-, curve. The P- graph displays the pressure in the combustion chamber as the
crank shaft rotates through one complete mechanical cycle of 360 (2-stroke engine) or
720 (4-stroke engine).
For a given engine, the nominal pressure versus crank angle curve can be established.
To assess engine performance, new cylinder patterns are evaluated and compared with
the nominal patterns. From the P- curves, information such as the start of combustion,
rate of pressure rise, peak firing pressure, expansion pressure, and terminal pressure
can be baselined, evaluated, and trended.
Figure 4-13 shows a typical P- curve for a two-stroke turbocharged diesel cylinder. At
the left edge of the plot, the piston is at bottom dead center (BDC) with both intake and
exhaust ports/ valves open. At this point, the exhaust blowdown is nearing completion,
and a fresh charge of air is being drawn into the cylinder. As the piston moves toward
top dead center, first the intake ports and then the exhaust ports/ valves are closed.
Once both the inlet and exhaust are closed, pressure in the cylinder begins to rise as the
piston motion compresses the air trapped in the cylinder.
1800
1440
1080
720
360
0
0 90 180 270 360
Fuel
8.00g
P4 VT4
Intake
Exhaust
p
s
i
Figure 4-13
Two-Stroke Engine Pressure Versus Crank Angle Curve
At approximately 2030 before top dead center (TDC), fuel injection begins. Following
the ignition delay, combustion begins followed by a rapid increase in cylinder pressure
4-17
to the peak firing pressure (P
max
). Between the start of combustion and the point at
which P
max
occurs, the piston reaches TDC and changes direction to begin the power
stroke. The pressure produced by combustion drives the piston toward BDC, and at
approximately 100 after TDC (ATDC), the exhaust valves/ ports open and exhaust
blowdown begins to remove the combustion products from the cylinder.
At approximately 140 ATDC, the inlet ports are opened to draw a fresh charge of air
into the cylinder. Typically, the ports and/ or valves are arranged such that when both
the inlet and exhaust are open at the same time, the inlet air flow assists in the removal
of the combustion products by pushing them towards the exhaust ports/ valves.
Figure 4-14 shows a P- curve for a typical four-stroke cylinder. Starting at the left edge
of this plot, the piston is at BDC with the intake valves open. The intake stroke has just
completed, but the intake valves remain open for a time as the piston moves toward
TDC. When the intake valves close, compression begins and cylinder pressure begins to
rise. At approximately 2030 before TDC, fuel injection begins. After the ignition
delay, combustion begins, followed by a rapid pressure rise to P
max
at an angle of
Pmax
.
The cylinder reaches TDC and reverses direction during this pressure rise. This begins
the power or expansion stroke.
1600
1400
1200
1000
800
600
400
200
0
C
y
l
i
n
d
e
r

P
r
e
s
s
u
r
e

(
p
s
i
)
0 60 120 180 240 300 360 420 480 540 600 660 720
Start of Combustion
Compression Reference
Pressure
Peak Firing
Pressure Angle
Peak Firing Pressure
Expansion Reference
Pressure
Figure 4-14
Four-Stroke Engine Pressure Versus Crank Angle Curve
4-18
Near the end of the power stroke, the exhaust valves open, and the piston moves from
BDC to TDC during the exhaust stroke to remove the combustion products from the
cylinder. The inlet valves open toward the end of the exhaust stroke so that the incom-
ing inlet air can assist in the removal of the combustion products. During the intake
stroke, the piston moves from TDC to BDC, drawing in a fresh charge of air. During the
time that the inlet valves are open, cylinder pressure remains constant at the air inlet
manifold pressure.
4.2.1.2 Cylinder Pressure Versus Volume Curve (P-V)
The pressure versus volume curve, or P-V curve, is another form of presenting mea-
sured cylinder pressure. In this case, the horizontal axis is the volume in the cylinder. It
can be plotted as either total volume or swept volume and in absolute volume or vol-
ume relative to BDC volume. The piston is at TDC at the left of the plot and BDC at the
right (see Figures 4-10 and 4-11).
Pressure-volume curves are derived from P- curves by converting crank angle to
cylinder volume using engine geometry information. Most engine analyzers make this
calculation automatically using the engine geometry data entered into the analyzer.
From Reference 3, for a simple piston, cylinder, and connecting rod system, (see Figure 4-1):

V V r R R c c ( ) [ ( )( cos sin )] = + + 1 1 1
1
2
2 2
Eq. 4-3
where
V
c
= cylinder clearance volume
r
c
= compression ratio
R = ratio of connecting rod length to crank radius
= crank angle ( = 0 at TDC, 180 at BDC)
P-V curves for two- and four-stroke engines are significantly different. As shown in
Figure 4-10, two-stroke P-V curves consists of a single loop since the cycle is completed
in one revolution of the crankshaft. The lower portion of the loop is the compression
stroke and the upper portion is the power stroke. The maximum pressure reached on
the P-V curve is the same as the maximum pressure shown on the P- curve for the
same cycle. As shown in Figure 4-11, four-stroke P-V curves contain two loops since it
takes two revolutions of the crankshaft to complete the cycle. During each cycle, the
piston passes through TDC and BDC twice. The upper loop consists of the compression
and power strokes while the lower loop includes the exhaust and intake strokes.
As discussed in Section 4.1.4, the work produced in the cylinder is equal to the area
enclosed by the P-V curve.
4-19
4.2.1.3 Log Cylinder Pressure Versus Log Volume Curve (Log P-Log V)
Log P versus log V curves are sometimes plotted using cylinder pressure versus crank
angle data, respectively. An example of a log P versus log V curve for a four-stroke
engine is shown in Figure 4-15. Studies have shown that the compression and expan-
sion processes are closely approximated by a polytropic process where:
pV
n
= constant
where
n = 1.25
en
1.35 [3]
10,000
1000
100
10
L
o
g

P
r
e
s
s
u
r
e

(
p
s
i
)
0.01 0.1
Log Volume (%)
1 10
Figure 4-15
Four-Stroke Log Pressure Versus Log Volume Curve
4-20
On a log P versus log V curve, the compression and expansion processes plot as straight
lines with slope equal to -n (see Figure 4-16). The start of combustion occurs at the point
where the compression curve departs from the straight line. Other deviations from the
straight line slope can be an indication of leakage during the compression or exhaust
processes.
5
4
3
2
1
0
-1
-2
-3
-4
-5
P
o
l
y
t
r
o
p
i
c

C
o
n
s
t
a
n
t
-90 -75 -60 -45 - 30 -15
0 15 30 45 60 75 90
Compression
Stroke
Expansion
Stroke
Figure 4-16
Polytropic Constant (n)
4-21
4.2.1.4 First Derivat ive Curve (dP-d)
The first derivative curve is a plot of the rate of pressure change versus crankshaft angle
(dP-d). Figure 4-17 shows a first derivative curve for a typical diesel engine. The main
characteristic of interest is the maximum pressure rise rate (MPRR), which occurs during
the rapid combustion phase. Ideally, the rate of heat release, and therefore the MPRR,
should be as rapid as possible without causing uncontrolled combustion (detonation or
knock) that can damage the engine. For any set of circumstances, there will be some
maximum pressure rise rate that is optimum. High rates of pressure rise can lead to
increased engine stresses and vibration, including crankshaft vibration and bending [5].
The first part of the first derivative curve represents the compression stroke. Cylinder
pressure is rising during the compression stroke due to the piston motion compressing
the cylinder contents. When fuel is injected into the cylinder, the rate of pressure rise
decreases slightly due to the heat transfer to the injected fuel. After a short delay (the
ignition delay), the fuel ignites causing the cylinder pressure to rise rapidly. Following
the rapid combustion phase, the rate of pressure rise decreases, cylinder pressure
reaches a peak, and then starts to decline as piston motion continues to increase the
cylinder volume. Note that in some cases, it can be difficult to determine the start of
injection and start of combustion accurately from first derivative curves. In these cases,
the second derivative curve might be useful (see Section 4.2.1.5).
60
45
30
15
0
-15
-30
-45
-60
R
a
t
e

o
f

C
h
a
n
g
e

(
p
s
i
/
d
e
g
r
e
e
s
)
0 60 120 180 240 300 360 420 480 540 600 660 720
Maximum Pressure
Rise Rate
Figure 4-17
First Derivative Curve
4-22
4.2.1.5 Second Derivat ive Curve (dP
2
-d
2
)
Another type of combustion curve that can be plotted is the second derivative curve
(dP
2
-d
2
). This curve represents the rate of curvature of the pressure versus crank angle
curve. The point at which combustion begins can be determined from this curve (see
Figure 4-18) [6].
8
6
4
2
0
-2
-4
-6
-8
0 60 120 180 240 300 360 420 480 540 600 660 720
S
e
c
o
n
d

D
e
r
i
v
a
t
i
v
e

(
p
s
i
/
d
e
g
r
e
e
s
) Start of
Combustion
Figure 4-18
Second Derivative Curve (dP
2
-d
2
)
4.2.2 Combustion Characteristics
An experienced analyst can draw conclusions regarding engine condition comparing
the combustion curves of the cylinders to one another. Prior to the advent of digital
engine analyzers, this is how engine analysis was done. However, digital engine analyz-
ers permit a large number of data to be reduced to a number of key characteristics that
can be extracted or calculated from the collected data. By evaluating and comparing
these key characteristics and engine performance, the mechanical condition can be
assessed to a significant degree.
4-23
4.2.2.1 Cylinder Peak Firing Pressure (P
max
)
Cylinder peak firing pressure is the maximum pressure in the cylinder during a me-
chanical cycle (see Figures 4-13 and 4-14). Peak firing pressure is a function of a large
number of variables including:
Compression ratio
Engine speed
Cylinder leakage and heat transfer
Fuel injection timing, quantity of fuel injected
Intake air temperature and pressure
Fuel properties
Because of the potential for cycle-to-cycle variation in peak firing pressure in a cylinder,
P
max
should be evaluated on a statistical basis from a number of cycles (200 cycles is
considered a valid sample) in each cylinder and across a number of cylinders by deriv-
ing parameters such as:
The average peak firing pressure in each cylinder over a number of cycles
The maximum peak firing pressure which occurred in each cylinder over a number
of cycles
The minimum peak firing pressure which occurred in each cylinder over a number
of cycles
The standard deviation of peak firing pressures over a number of cycles
The difference between the overall engine average peak firing pressure and the
average peak firing pressure for each cylinder
The maximum pressure rise rate that occurred in each cylinder over a numberof
cycles
4.2.2.2 Cylinder Peak Firing Pressure Angle (
Pmax
)
Peak firing pressure angle,
Pmax
, is the angle at which the peak firing pressure occurs
(see Figures 4-13 and 4-14). It is usually reported as the number of degrees after top
dead center (ATDC) at which the peak occurs. If the peak firing pressure occurs too
early, excessive loads, stresses, and temperatures can result. If the peak firing pressure
occurs too late, combustion energy will be lost and mechanical efficiency will be low.
This angle must be correct to attain the best performance from a cylinder. The peak
firing pressure angle for each cylinder should be compared to the engine average angle.
If a cylinder
Pmax
varies significantly from the engine average, an off-design condition is
indicated and should be investigated. Peak firing pressure angle is strongly dependent
on fuel injection timing.
4-24
4.2.2.3 Combust ion St art Angle
Combustion start angle is the angle at which pressure starts to rise measurably due to
heat release from combustion of the fuel (see Figures 4-13 and 4-14). At this point,
pressure in the cylinder increases beyond that which would occur due to compression
only. The combustion start angle can be determined from first or second derivative
curves. The second derivative curves have been reported to give a more accurate deter-
mination [6].
4.2.2.4 Expansion Reference Pressure
Some engine analyzers use the power stroke to evaluate cylinder condition. Expansion
reference pressure is the cylinder pressure at a pre-determined crank angle during the
power expansion stroke (see Figures 4-13 and 4-14). This allows cylinder pressure
comparisons from one cylinder to the others. This measurement must be used with
some discretion. Expansion reference pressure is affected by engine load, speed, and
other operating variables. When any of these factors change, the expansion reference
pressure will also change. Expansion reference pressure should primarily be used to
evaluate the spread in expansion reference pressures across the entire engine. Compari-
son of the expansion reference data from measurement sets having different operating
conditions is not recommended.
4.2.2.5 Terminal Pressure
Terminal pressure is the residual pressure remaining in the cylinder at BDC at the end
of the power stroke (see Figures 4-13 and 4-14). In the case of a two-stroke scavenged
engine, it is equal to the scavenge air pressure. In this case, the terminal pressure pro-
vides information to evaluate the scavenging blower and the force of the air purging the
cylinder. In a four-stroke engine it is equal to the pressure in the exhaust port. If the
cylinder pressure does not drop properly, it indicate exhaust restriction that prevents all
the exhaust gas from leaving the cylinder. Exhaust restriction prevents a correct fuel/ air
mixture for the next cycle. Alternately, the cause could be delayed exhaust valve or port
opening.
4.2.2.6 St art of Fuel Inject ion
Because it has no immediate affect on the cylinder pressure, the start of fuel injection
cannot be detected from the P- curve. However, it can normally be detected from the
ultrasonic signal trace. Injection timing will affect the start of combustion, MPRR, peak
firing pressure, and peak firing pressure angle.
4.2.2.7 Compression Reference Pressure
Some engine analyzers use the initial rising portion of the compression stroke to evalu-
ate cylinder compression performance. Compression reference pressure refers to the
4-25
cylinder pressure at a pre-determined angle during the compression stroke before
combustion occurs. The point where compression reference pressure should be mea-
sured varies from engine to engine due to differences in engine design. Engine analysis
equipment vendors can recommend suitable crank angle positions. Note that the com-
pression reference pressure is not the highest compression pressure produced by the
cylinder and should not be compared with the true compression pressure specified by
the engine manufacturer.
Almost any angle prior to the start of combustion can be selected as the reference point.
Usually, an angle just before the start of injection is used. The extracted pressure at this
angle can be used for trending and comparison purposes. It should be noted, however,
that differences in engine speed, intake air temperature, and intake air pressure will
affect the compression reference pressure. If a problem affects cylinder compression, the
change can be detected.
4.2.2.8 Peak Compression Pressure
Cutting off the fuel to one cylinder and recording the cylinder pressure during the
mechanical cycle can provide indications of several engine problems. A cylinder that
has been cut off from fuel is said to be motoring. Similar to the peak firing pressure in a
firing cylinder, the peak compression pressure is the maximum pressure reached during a
mechanical cycle in a motoring cylinder. The magnitude of the peak compression pres-
sure will vary with inlet manifold temperature and pressure, and it will depend on the
mechanical condition of the cylinder components. Figure 4-19 shows a pressure versus
crank angle curve for a motoring four-stroke cylinder.
600
500
400
300
200
100
0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180
Theta (degrees)
P
r
e
s
s
u
r
e

(
p
s
i
a
)
Figure 4-19
Four-Stroke Motoring Cylinder Pressure Versus Crank Angle
Note that this curve is nearly symmetrical around TDC. However, the peak cylinder
pressure actually occurs slightly before TDC. The angle between TDC and the peak
4-26
compression pressure is known as the thermodynamic loss angle (TLA). The pressure in
a motoring cylinder is affected by the following:
Intake air temperature and pressure
Volume change caused by piston motion
Heat transfer losses from the compressed air to the engine
Gas leakage past piston rings and ports/ valves
During most of the motoring cycle, the effects of piston motion dominate. However, as
the piston nears TDC it slows, and at TDC it is motionless. During the compression
stroke just prior to TDC, the pressure and temperature of the cylinder contents are
reaching their highest values. As a result, heat transfer and leakage, both of which
reduce cylinder pressure, are at maximum. The rise in pressure due to piston motion is
less than the heat transfer and leakage as the piston approaches TDC, causing the pres-
sure to begin to decrease before the piston reaches TDC. Peak compression pressure
would only occur at TDC if there were no heat transfer or leakage from the cylinder. All
engines experience some heat transfer and leakage loss and therefore have peak com-
pression pressures located slightly before TDC.
The peak compression pressure measured during motoring will depend on:
Heat loss from the cylinder
Initial cylinder temperature and pressure
Engine speed
Valve timing
Piston ring blowby
Port/ valve leakage
In addition, on turbocharged engines, load has a significant affect on indicated peak
compression pressure, mainly due to the higher initial cylinder pressure.
The combustion characteristics and parameters discussed in this section can be deter-
mined directly from engine analysis data. Other characteristics, such as indicated horse-
power and indicated mean effective pressure (IMEP), can be derived from the combus-
tion curves.
In addition to quantifying combustion characteristics from each cylinder, these charac-
teristics should be evaluated across all cylinders. The engine average and spread for
each should be determined. Spread can be used to determine if a cylinder abnormality
exists. If there is a consistent problem in a cylinder, it will affect every stroke of the
piston. If only data from that cylinder were evaluated, problems could be masked.
Comparison to other cylinders can help identify a problem. When a cylinder parameter
differs substantially from the engine average, it is a strong indication of a potential
problem.
4-27
4.2.3 Variability
Engine analysis data will vary from cycle to cycle, from cylinder to cylinder, and from
test to test. There are a number of reasons for these variations and a large part of engine
analysis is determining when this variation is due to real degradation of the engine and
when it is due to other factors.
4.2.3.1 Cycle-t o-Cycle Variabilit y
The mixing of fuel and air in the cylinder prior to and during combustion is a random
process. As a result, the combustion process and the parameters that are measured
during engine analysis vary from cycle to cycle. The variation is random [7]. The cycle-
to-cycle variability of peak firing pressure in a well-tuned diesel engine typically has a
standard deviation on the order of 1%4% [8]. Thus, some variation is normal and
expected. However, if the variability of one cylinder is much greater than the others, it
could be an indication of a problem.
To more accurately describe the condition of any given cylinder, multiple measure-
ments of combustion parameters should be made. The average of a number of measure-
ments is a better estimate of the true average than is any single measurement, since the
standard deviation of a series of measurements decreases as the number of measure-
ments increases.
The potential for substantial cycle-to-cycle variability complicates analysis. Simply
analyzing one P- curve could provide incorrect indications regarding engine condition.
Instead, engine analysis should be based on several samples and statistical measures.
Most engine analyzers collect data from many combustion events and use these data to
determine the average peak firing pressure, the average deviation (a quantitative mea-
sure of cyclic dispersion), the maximum peak firing pressure, and the minimum peak
firing pressure. Other attributes can be derived similarly.
Whenever there is a fuel injection, air or mixing problem, cycle to cycle comparison is
an excellent anomaly detection tool. While comparing many P- curves from a cylinder
manually is tedious and time consuming, many analyzers are capable of recording
several cycles worth of data, selecting key pressure points and generating statistics such
as average, maximum, minimum, and standard deviation of the peak firing pressures,
and generating a typical P- trace to display based on the statistical data. These statistics
are a powerful analysis tool.
4.2.3.2 Cylinder-t o-Cylinder Variabilit y
In an ideal world, the performance of all of the cylinders of a given engine would be the
same and cylinder-to-cylinder variation in combustion parameters would be zero.
However, some cylinder-to-cylinder variation will exist due to variations in fuel rack
settings, fuel injection timing, mechanical variations in engine parts, and other effects.
Any outliers in comparisons of data from cylinder to cylinder suggest the need for
further action to tune the cylinder or to take corrective actions to fix degraded parts.
4-28
4.2.3.3 Test -t o-Test Variabilit y
Changes in atmospheric conditions will affect engine analysis data. Because it is un-
likely that atmospheric conditions will be the same from one test to the next, it must be
recognized that the data can vary for reasons that have nothing to do with engine con-
dition. However, many characteristics of the engine analysis curves and data should
remain the same. Changes in test conditions should not affect the relative magnitudes of
engine analysis data from cylinder to cylinder. For example, if one cylinder of engine
has always had the highest peak firing pressure, changes in atmospheric conditions will
not be responsible for it suddenly having the lowest peak firing pressure in a test. In
addition, the general shapes of the various combustion, vibration, and ultrasonic curves
should not vary from test to test unless engine degradation has occurred.
4.2.4 Interpretation of Combustion Signatures
There are two concerns when interpreting the combustion curves and parameters de-
scribed above:
1. Have any engine design limits been exceeded?
2. Are there any indications of problems?
Assuming that engine design limits exist, the first concern is relatively straight forward
to evaluate. The measured data can be directly compared to the limits. If any limits are
violated, corrective actions can be taken. Addressing the second concern requires judg-
ment and experience. Where fixed limits do not exist, the data from each cylinder can be
compared to other cylinders and against previously measured baseline data. Differ-
ences found in these comparisons may provide indications of adjustments which need
to be made or corrective actions which need to be taken.
If a problem is found or suspected, the data and equipment set up should be examined
to ensure that the problem is real and not the result of false data.
4.2.5 Analysis Guidelines
Most engine vendors do not have bases or guidelines for most of the data recorded during
engine analysis. Peak firing pressures and exhaust gas temperatures are typically the only
guidelines related to engine analysis used by vendors. Table 4-1 summarizes the available
analysis guidelines for the engines covered by this guide. These values are taken from
engine vendor documents, EDG owners group publications and individual nuclear power
plant manuals. As can be seen, guidelines for most of the parameters of interest in engine
analysis do not exist. As a result, individual nuclear power plants or EDG owners groups
will likely have to develop guidelines using an engineered approach.
In general, two types of analysis guidelines are used:
Guidelines that are quantitative and are based on the engine manufacturers design
basis or on previously recorded operating data
4-29
Guidelines that are based on data that is qualitative in nature and must be inter-
preted to make conclusions regarding engine condition
The engineered approach to developing analysis guidelines involves a combination of
the following methods or sources of information:
Operating limits published by the engine vendor
Baseline data collected on a properly tuned engine in good operating condition
Engine design basis, if available
Industry good-practice
Engineering judgment
Appropriate parameters for numerical analysis guidelines include:
Maximum engine average peak firing pressure
Maximum cylinder peak firing pressure
Variation of cylinder peak firing pressure over a number of cycles
Maximum cylinder-to-cylinder variation in peak firing pressures
Engine average corrected peak firing pressure angle
Maximum range of peak firing pressure angles across all cylinders
Engine average fuel injection timing
Maximum range of fuel injection timing across all cylinders
Maximum range of fuel rack settings across all cylinders at full load
Allowable range of fuel injection nozzle pop pressures for new fuel injection nozzles
Allowable range of fuel injection nozzle pop pressures for used fuel injection nozzles
Maximum cylinder exhaust gas temperature
Maximum range of exhaust gas temperatures across all cylinders
A detailed discussion on the development of analysis guidelines for these parameters is
contained in Section 5 of this guide.
Guidelines for vibration and ultrasonic signatures are extremely difficult to quantify.
Rather than being compared to numerical limits, vibration and ultrasonic data are best
used by comparing them to previously recorded signatures for the same cylinder or from
other cylinders. However, it must be stressed that these comparisons are qualitative in
nature. As a result, vibration and ultrasonic analyses should only be used to direct or
guide further inspections, analyses, or testing. It is recommended that EDG operability or
maintenance decisions not be based on vibration or ultrasonic analyses alone.
4-30
T
a
b
l
e

4
-
1
E
x
i
s
t
i
n
g

P
u
b
l
i
s
h
e
d

E
n
g
i
n
e

A
n
a
l
y
s
i
s

G
u
i
d
e
l
i
n
e
s
C
a
t
e
g
o
r
y
T
y
p
e
E
n
g
i
n
e

M
a
k
e
/
M
o
d
e
l
A
L
C
O

2
5
1
C
o
l
t
-
P
i
e
l
s
t
i
c
k
P
C
2
,

P
C
2
.
3
,
P
C
2
.
5
F
a
i
r
b
a
n
k
s
M
o
r
s
e
O
p
p
o
s
e
d
P
i
s
t
o
n
3
8
T
D
-
8
-
1
/
8
C
o
o
p
e
r
-
B
e
s
s
e
m
e
r

K
S
V

1
T
D
I
(
E
n
t
e
r
p
r
i
s
e
)

D
S
R
a
n
d

D
S
R
V
E
M
D
6
4
5
N
o
r
d
b
e
r
g

H
S
-
F
S
-
1
3
1
6
S
A
C
M
U
D
4
5
W
o
r
t
h
i
n
g
t
o
n

S
W
B
1
2
F
u
e
l

i
n
j
e
c
t
i
o
n
t
i
m
i
n
g
E
n
g
i
n
e

A
v
e
r
a
g
e
2
5

B
T
D
C
2
9
3
1
B
T
D
C
2
2

B
T
D
C
M
a
x
i
m
u
m

r
a
n
g
e

o
f

a
l
l
c
y
l
i
n
d
e
r
s
N
R
F
u
e
l

r
a
c
k
s
e
t
t
i
n
g
R
a
n
g
e

(
m
m
)

0
.
5

m
m

o
f
e
n
g
i
n
e

a
v
g
.
F
u
e
l

i
n
j
e
c
t
o
r
p
o
p

p
r
e
s
s
u
r
e
P
e
r

c
y
l
i
n
d
e
r

[
n
e
w
]
(
p
s
i
g
)
3
,
7
0
0
4
,
0
5
0
3
,
4
5
0
3
,
5
5
0
1
,
8
0
0
1
,
9
0
0
3
,
7
0
0
3
,
8
0
0
P
e
r

C
y
l
i
n
d
e
r

[
u
s
e
d
]
(
p
s
i
g
)
3
,
2
0
0
3
,
5
5
0
3
,
4
0
0
3
,
7
0
0
C
y
l
i
n
d
e
r

p
e
a
k
f
i
r
i
n
g

p
r
e
s
s
u
r
e
M
a
x
i
m
u
m

e
n
g
i
n
e
a
v
e
r
a
g
e

(
p
s
i
g
)
1
,
2
7
0

(
9
0
0
R
P
M
)
1
,
6
9
0
1
,
4
0
0
M
a
x
i
m
u
m

r
a
n
g
e

o
f

a
l
l
c
y
l
i
n
d
e
r
s

(
p
s
i
g
)
1
5
0
7
0

a
l
e
r
t
1
0
0

a
c
t
i
o
n
1
6
0
8
0
1
1
0
S
t
a
n
d
a
r
d

d
e
v
i
a
t
i
o
n
o
v
e
r

m
a
n
y

c
y
c
l
e
s
(
p
s
i
g
)
C
o
r
r
e
c
t
e
d
f
i
r
i
n
g

p
r
e
s
s
u
r
e
a
n
g
l
e
E
n
g
i
n
e

a
v
e
r
a
g
e
1
0
1
4
A
T
D
C
M
a
x
i
m
u
m

r
a
n
g
e

o
f

a
l
l
c
y
l
i
n
d
e
r
s

o
f
e
n
g
i
n
e

a
v
g
.
E
x
h
a
u
s
t

g
a
s
t
e
m
p
e
r
a
t
u
r
e
M
a
x
i
m
u
m

s
i
n
g
l
e
c
y
l
i
n
d
e
r

(
F
)
8
5
0
1
,
1
0
0
1
,
1
0
0
M
a
x
i
m
u
m

r
a
n
g
e

o
f

a
l
l
c
y
l
i
n
d
e
r
s

(
F
)
1
5
0

(
1
6
c
y
l
i
n
d
e
r
)
4
0
0

(
1
8
c
y
l
i
n
d
e
r

m
u
l
t
i
-
p
i
p
e
)
1
2
6

a
l
e
r
t
1
5
0

a
c
t
i
o
n
<

2
0
0
U
l
t
r
a
s
o
n
i
c

a
n
d
v
i
b
r
a
t
i
o
n
s
i
g
n
a
t
u
r
e
s
N
u
m
e
r
i
c
a
l

v
i
b
r
a
t
i
o
n

a
n
d

u
l
t
r
a
s
o
n
i
c

s
i
g
n
a
t
u
r
e

a
c
c
e
p
t
a
n
c
e

c
r
i
t
e
r
i
a

h
a
v
e

n
o
t

b
e
e
n

e
s
t
a
b
l
i
s
h
e
d

o
r

i
d
e
n
t
i
f
i
e
d

f
o
r

a
n
y

m
o
d
e
l

o
f

E
D
G

i
n

n
u
c
l
e
a
r

s
t
a
n
d
b
y
s
e
r
v
i
c
e
.

H
o
w
e
v
e
r
,

c
o
m
p
a
r
a
t
i
v
e

g
u
i
d
e
l
i
n
e
s

a
r
e

a
v
a
i
l
a
b
l
e
.

W
h
e
n

t
h
i
s

d
a
t
a

i
s

c
o
l
l
e
c
t
e
d
,

i
t

i
s

u
s
e
d

f
o
r

t
r
e
n
d
i
n
g

i
n
d
i
v
i
d
u
a
l

c
y
l
i
n
d
e
r
s

a
n
d

f
o
r

c
o
m
p
a
r
i
n
g
c
y
l
i
n
d
e
r
s

t
o

o
n
e

a
n
o
t
h
e
r
.

N
R

=

N
o
t

r
e
p
o
r
t
e
d

1
.

F
r
o
m

C
o
o
p
e
r
-
B
e
s
s
e
m
e
r

M
o
d
e
l

K
S
V

E
m
e
r
g
e
n
c
y

D
i
e
s
e
l

G
e
n
e
r
a
t
o
r

E
n
g
i
n
e

A
n
a
l
y
s
i
s

G
u
i
d
e
l
i
n
e
s
,

R
e
v
i
s
i
o
n

1
,

p
u
b
l
i
s
h
e
d

b
y

t
h
e

C
o
o
p
e
r
-
B
e
s
s
e
m
e
r

O
w
n
e
r
s

G
r
o
u
p
.
4-31
4.2.6 Trending
The fact that EDGs are usually only run for a few hours per month does not mean that
trending cannot be performed. In fact, it means that trending must be performed in
order to identify potential problems. However, the engine analyst must be aware that
outside factors can affect engine analysis data. If these factors are not recognized and
their effects are not considered or corrected, erroneous conclusions could be made
regarding engine condition. The two factors that will vary between engine analysis runs
and that can affect results are engine load and environmental conditions.
Individual cylinder loads are directly proportional to total engine load. If engine load
varies from one engine analysis run to another, parameters such as peak firing pressure,
IMEP, and fuel rack settings will vary. As a result, trending one run against another
could indicate that the material condition of the engine has changed when, in fact, only
the load carried by the engine has changed. If engine load varies during one engine
analysis run, the individual cylinder data will be difficult to compare to one another.
The effects of engine load changes must be accounted for before data trending will
provide reliable information.
Additionally, changes in ambient air conditions have resulted in changes in engine
balance and analysis data at several stations. Changes in temperature, pressures, and
humidity can change combustion parameters sufficiently to make an engine balanced
under different environmental conditions appear unbalanced. References 1 and 2 pro-
vide guidance on development of engine performance maps and variations in engine
performance due to ambient environment changes. Individual nuclear power plants can
generate engine performance maps and use them for trending purposes.
In addition, effects such as equipment calibration, accelerometer location and mounting
method, and changes in signal processing can affect the usefulness of data for trending
purposes.
Any of the combustion parameters discussed in Section 4.2.2 can be trended over time.
However, the decision of which parameters to trend should be based on the potential
benefits of trending each of the parameters, the potential for each of the parameters to
drift over time, and the capabilities of the engine analyzer used. Recommended param-
eters for trending include:
Cylinder peak firing pressures
Cylinder peak compression pressures
Terminal pressures
Expansion reference pressures
Compression reference pressures
Cylinder power (IMEP or IHP) spread
Exhaust temperature spread
4-32
4.3 Vibration and Ultrasonic Analyses
In traditional vibration analysis of rotating equipment, such as pumps, turbines, and
generators, vibration amplitude is measured as a function of frequency. This type of
analysis is referred to as frequency-domain analysis. Many EDG components can also
be analyzed using frequency-based analysis. A large number of references cover vibra-
tion analysis in the frequency domain and, as a result, it will not be discussed further in
this guide. This section is devoted to vibration analysis in the time domain. In time-
domain analysis, the vibration amplitude is plotted as a function of time (or crankshaft
angle). Many of the most common diesel engine problems can be detected by time-
domain analysis, including leaking valves, valve timing errors, cylinder liner scuffing,
and fuel injection problems. It is important to note that frequency-domain vibration
analysis has not been demonstrated to be useful for detecting these problems.
Nuclear standby EDGs in the U.S. operate at 60 Hz generator output frequency. The
large, medium-speed diesel engines that are prime movers for the generators operate at
speeds of 450 RPM to 1,200 RPM with combustion events occurring at 3.75 Hz to 10 Hz
in four-stroke engines and 7.5 Hz to 20 Hz in two-stroke engines. Vibrations caused by
combustion-related events are typically at much higher frequencies than engine operat-
ing speed and are broad band in frequency. Therefore, combustion vibration sources are
best analyzed by filtering out the low frequencies corresponding to engine speed and
performing the analysis in the time domain. Using this technique, the timing of differ-
ent combustion events in the mechanical cycle can be compared in each cylinder or
from cylinder to cylinder. The vibration signal is measured and plotted as a function of
time (or crank angle). This type of plot provides indications of the timing of combustion
events (crank angle) and the relative strength of events (vibration amplitude).
Ultrasonics refer to relatively high frequency vibrational energy that is present in oper-
ating machinery. There is no precise definition of ultrasonics. In this guide, frequen-
cies above 15 kHz are considered to be in the ultrasonic range. The principles of ultra-
sonic analysis are the same as those of vibration analysis.
4.3.1 Sources of Engine Vibration and Ultrasonic Noise
There are many sources of vibration and ultrasonic noise in a reciprocating engine. The
reciprocating motion inherently introduces vibrational forces due to the inertia of recip-
rocating parts. Rotating components also vibrate. In addition, mechanical impacts (for
example, valves landing on seats) and high velocity gas flows (for example, exhaust
blowdown) produce high frequency signals. Signals from these sources can be mea-
sured and used to diagnose engine problems by comparing the measured signatures to
expected or previously measured baseline signatures of a healthy or normal engine.
The ultrasonic or high frequency vibration energy of interest originates from several
sources:
Mechanical impacts, such as valve closure or piston slap
High velocity gas flow, such as exhaust blowdown or piston ring blowby
4-33
Mechanical roughness or friction
Direct combustion-induced energy
All these events are expected in a new engine.
4.3.2 Characteristics of Engine Vibration and Ultrasonic Signals
Three general categories of vibration and ultrasonic noise are of interest in time-domain
analysis: gas flow, mechanical impacts, and rubs. Conditions that cause vibration or
ultrasonic signals produce one or more of these types of signals.
Gas flow vibration or ultrasonics usually occurs at high frequencies. As gas flows, it
creates a high frequency whistle effect. It is normal for this to occur in all gas flows. The
presence of this gas flow during times when no gas should be flowing is an indication
of a leak. A high gas vibration or ultrasonic amplitude during times gas should be
flowing freely is an indication of a possible obstruction.
Reciprocating mechanical components such as pistons, connecting rods, valves, fuel
linkages, and piston bushings produce mechanical impacts. A knock occurs when
mechanical clearance increases in a bushing pin or bearing. Reciprocating actions create
forces that switch from one direction to the other, resulting in impact or knock.
Rubs occur when there is inconsistent (non-linear) contact throughout the movement of
a reciprocating or rotating component. Piston liner rub is the most common form of this
type of vibration in reciprocating machinery. When a cylinder liner has a rough or bad
spot, the vibration increases each time the piston passes that point.
Engine events that produce vibration or ultrasonic signals typically occur once per
mechanical cycle and are characterized by certain shapes or patterns when plotted
versus time or crankshaft angle. Envelopes of the characteristic shapes are shown in
Figure 4-20.
Valve
Opening
Gas
Flow
Mechanical
Impact
Combustion
A
m
p
l
i
t
u
d
e
Figure 4-20
Envelopes of Typical Vibration and Ultrasonic Patterns
4-34
4.3.3 Measurement Methods
Engine and component vibration and ultrasonic signatures are typically measured using
accelerometers attached to the component of interest. For measurement of combustion-
related vibration and ultrasonic signals, the accelerometer is normally mounted on the
cylinder head. On the FMOP engine, which does not have cylinder heads, the accelerom-
eter is typically attached to the cylinder liner. Piezoelectric transducers are typically used
because they are capable of measuring the high frequencies associated with combustion
events. The selection and mounting of accelerometers are two of the most important
decisions in vibration and ultrasonic analyses. See Section 5 for more detail.
4.3.4 Interpretation of Vibration and Ultrasonic Signatures
Vibratory motion can be described in terms of displacement, velocity, or acceleration.
Displacement is the distance traveled by the point of interest in a specific direction. If
displacement in the vertical direction is measured, the vibrating point will move a peak
distance down and a peak distance up. This total distance traveled from top to bottom
is referred to as peak-to-peak displacement.
Velocity is another characterization of vibratory motion. As the vibrating point reaches
the upper peak of the displacement, it stops, reverses direction, and begins the down-
ward motion. As it accelerates downward, it picks up speed, until it reaches the point of
zero displacement. At this point, it begins to gradually slow down until it stops at the
bottom peak of the displacement. Now this process repeats in the upward direction.
The peak speed or velocity happens at the mid-point of displacement.
Acceleration is another way of describing vibratory motion. Maximum or peak accelera-
tion occurs at the points of peak displacement. Vibratory acceleration is commonly
expressed as a ratio to gravitational acceleration. For example, 2.0 g acceleration means
twice gravitational acceleration.
4-35
When a vibration measurement point is set up for an engine, the decision must be made
to measure displacement, velocity, or acceleration. Each has advantages. It is desirable
to select the amplitude unit that will provide the best tracking of component condition.
Generally, displacement provides the best amplitudes at low frequencies from 1 Hz
10 Hz. Velocity provides good condition information for general machinery producing
frequencies from 10 Hz100 Hz. Acceleration amplitudes are most sensitive to changes
in high frequencies from 100 Hz and higher. Most engine analyzers measure vibration
and ultrasonic signals using acceleration signals only; therefore, changing to velocity or
displacement units is not a supported option.
Vibration transducers are sensors that convert motion into proportional electrical sig-
nals. Selection of these sensors requires proper frequency range, temperature range, and
mounting considerations. The most common transducer used in the collection of vibra-
tion and ultrasonic signals from machinery is the accelerometer.
Proper attachment of the transducer is very important. The function of transducer is to
the turn the mechanical motion into an electrical signal. To do this, the transducer must
be attached so that it moves with the vibrating surface.
Selecting the point to attach the transducer for the vibration measurement is critical. It is
desirable to select a location that will provide the most reliable and consistent engine
condition information. Measurements should be made at the same point each time to
take advantage of amplitude trending.
4-36
Nominal vibration and ultrasonic patterns from a four-stroke engine are shown in
Figure 4-21. The pattern can best be interpreted by examining angular ranges.
From TDC to exhaust valve opening, direct combustion energy is usually visible.
Problems such as piston slap, piston ring blowby, and cylinder liner scoring can be
detected in this region.
From exhaust valve opening to TDC, only the exhaust blowdown can usually be
seen. Opening of the exhaust valve is normally not visible. A mechanical impact ear
TDC could be due to excessive wrist pin or connecting rod bearing clearance.
Between TDC and BDC, a mechanical impact is usually visible when the exhaust
valves close. Problems with the exhaust valve train could lead to early or late closing
or high or low amplitude of the closing signal.
Between BDC and TDC, a mechanical impact is usually visible when the intake
valves close. Again, problems with the valve train could lead to early or late closing
or high or low signal amplitude. Usually, the start of fuel injection is visible shortly
before TDC. A gas flow pattern in this region is an indication of piston ring blowby
during the compression stroke.
2000
1500
1000
500
0
-360 -180 0 180 360
5.00g
5.00g
Ultrasonic
Vibration
Intake
Exhaust
Intake
Closure
Exhaust Closure
Fuel
Exhaust
Opens
Intake
Opens
Injection Event
Pressure
Fuel/Combustion
Exhaust
Closure
Intake
Closure
Exhaust Blowdown
Figure 4-21
Nominal Four-Stroke Vibration and Ultrasonic Signatures
4-37
Nominal vibration and ultrasonic patterns from a two-stroke engine is shown in Figure
4-22.
From TDC to exhaust valve/ port opening, direct combustion energy should be
visible. Problems such as piston ring slap, piston ring blowby, and cylinder liner
scoring may be seen in this region.
Normally, between exhaust valve/ port opening and BDC only exhaust blowdown
can be seen. If there are intake or exhaust ports, ring clip can be seen as the compres-
sion rings enter the port areas.
Between BDC and exhaust valve/ port closure, exhaust valve closure will be de-
tected as a mechanical impact if the engine is equipped with exhaust valves. Again,
problems with the valve train can lead to early or late closure, or high or low signal
amplitude. Exhaust port closure will not be visible.
Between exhaust valve/ port closure and TDC, the start of fuel injection can usually
be seen. A gas flow pattern in this region is an indication of piston ring blowby
during compression.
Intake
Exhaust
RPM = 898.2
HP = 463.7
PFP Angle = 19.3
3.00g
P1 VT4
1440
1157
874
591
308
25
0 90 180 270 360
p
s
i
Figure 4-22
Nominal Two-Stroke Vibration and Ultrasonic Signatures
4-38
4.3.5 Analysis Guidelines
As stated earlier, all engines vibrate regardless of how well they are designed, as-
sembled, and maintained. Threshold vibration levels above which an engine can be
considered to be in need of corrective actions (bad condition) have been recommended
by various sources. For example, Reference 9 provides recommended vibration limits
and suggested actions for diesel generators. However, these threshold levels are usually
not based on scientific analysis and prediction.
Interpretation involves of vibration and ultrasonic signatures involves pattern recogni-
tion (verifying the presence of expected events at the correct time and with acceptable
amplitude) and checking for the absence of expected events. The issue of amplitude is
problematic; it depends not only on the energy in the event sensed, but also on the
transmission path, sensor characteristics, and mounting method. Obtaining consistent
and meaningful amplitudes in ultrasonic analysis is more difficult than for conventional
low frequency vibration analysis. Fortunately, engine faults usually produce large
amplitude changes and minor changes due to technique are less significant.
Ultrasonic signals will provide indications of such events as valve closings, fuel injec-
tion, exhaust blow down, piston rings passing over inlet and exhaust ports, and direct
combustion noise. In addition, similar events in adjacent cylinders can be picked up by
the accelerometer. This interference from other cylinders is known as cross talk. Any
other significant components in the pattern indicate potential defects.
The most commonly used method for monitoring engine vibration is to measure or
determine baseline vibration signatures for an engine known to be in good condition
and to monitor changes in these signatures over time. Changes in these signatures are
an indication of changes in the material condition of the engine. Example of changes
include:
Increase or decrease in vibration amplitude
Appearance of new vibrations or disappearance of previously seen vibrations
Changes in vibration frequency (seen on amplitude versus frequency plots)
Changes in timing of events (seen on amplitude versus crank angle plots)
Numerical analysis alert guidelines are usually not used for vibration analysis due to
the difficulty in determining meaningful numerical limits. Instead, trending and com-
parison are used to monitor changes in vibration patterns over time.
It is recommended that, due to the subjective nature of the data, maintenance decisions
not be made based on vibration or ultrasonic data alone. Instead, vibration and ultra-
sonic data should be used in combination with other analysis and monitoring data.
Ultrasonic signatures are usually band filtered. Higher frequency bands have higher
sensitivity to gas flows, such as piston ring blowby. Lower frequency bands have more
4-39
sensitivity to mechanical impacts, such as valve closing and piston slap. This function is
performed by most engine analyzers and is not a user changeable feature.
4.3.6 Trending
Vibration patterns should be compared to previously measured vibration signatures for
the same cylinder, other cylinders of the same engine, and from other engines of the
same type. Significant variations in the amplitude or timing of the signatures can pro-
vide indications of problems. For example, an increase in the amplitude of the vibration
for a valve closure event indicates that the valve is impacting the valve seat harder than
it was previously. This can indicate that the valve seat is damaged or that the valve
lifter has collapsed. A decrease in the amplitude of this event could indicate that the
valve seat is damaged or the valve clearance needs to be adjusted. The timing of the
valve closure event could also indicate a problem with the valve train. For example, if
the valve closes earlier than expected, the valve clearance could be out of adjustment or
the valve lifter could be collapsed.
In addition, the appearance of new vibration patterns or the disappearance of previously
seen patterns can also provide indications of problems. Piston slap will typically show up
a vibration spike in the area around TDC and could be an indication of excessive bearing
clearances, improper piston ring loading, detonation, or uneven combustion.
5-1
5
EQUIPMENT SELECTION, SETUP, AND USE
Like most other analysis methods, the usefulness and accuracy of the results are largely
dependent upon the measurement equipment and its use. This section provides guid-
ance regarding equipment selection and use.
5.1 Key Analyzer Characteristics
There are currently many companies manufacturing engine analysis equipment. Fea-
tures and capabilities of these systems vary greatly. We recommend that the end user
review the available equipment and request an on-site demonstration of one or more
systems before purchasing. In several cases, utilities have purchased engine analyzers
but have stopped using them due to dissatisfaction with features or ease-of- use. It
should be noted that should you choose not to purchase an analyzer, some consider-
ation should still be given to the analysis equipment used by the contractor.
Engine monitoring systems, which are typically permanently installed and that do not
record cylinder pressure and vibration signatures, are not included in this discussion as
they are not true engine analyzers. This review covers hand held, portable engine analy-
sis equipment specifically designed to record and analyze engine signatures as well as
PC-based engine analysis systems. Table 5-1 summarize the features and capabilities of
these devices. The equipment vendors provided the data in Table 5-1. The table pro-
vides a summary of the recorded parameters and engine analysis methods available for
each make and model. Table 5-1 is by no means a complete list of all available equip-
ment. This technology is changing rapidly; readers are advised to check for updated
information.
5-2
T
a
b
l
e

5
-
1
E
n
g
i
n
e

A
n
a
l
y
s
i
s

E
q
u
i
p
m
e
n
t

S
u
m
m
a
r
y
V
e
n
d
o
r
M
o
d
e
l
D
e
s
c
r
i
p
t
i
o
n
M
e
a
s
u
r
e
d

I
n
p
u
t
s
O
u
t
p
u
t

V
a
l
u
e
s
B
u
i
l
t
-
i
n
A
n
a
l
y
s
i
s
C
a
p
a
b
i
l
i
t
i
e
s
U
n
i
q
u
e

F
e
a
t
u
r
e
s
C
o
m
m
e
n
t
s
V
e
n
d
o
r

C
o
n
t
a
c
t
I
n
f
o
.
L
i
b
e
r
t
y
T
e
c
h
n
o
l
o
g
i
e
s
/
B
e
t
a
R
e
c
i
p
-
t
r
a
p
(
R
T
-
9
2
4
0
)
H
a
n
d
-
h
e
l
d

e
n
g
i
n
e
a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n

u
l
t
r
a
s
o
n
i
c
,
t
e
m
p
e
r
a
t
u
r
e
s
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,

P
-
,
P
-
V
,

v
i
b
r
a
t
i
o
n

b
o
t
h
f
r
e
q
u
e
n
c
y

t
i
m
e
d
o
m
a
i
n

u
l
t
r
a
s
o
n
i
c
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s

w
i
t
h

P
C
s
o
f
t
w
a
r
e
D
a
t
a

c
a
n

b
e

e
x
p
o
r
t
e
d
t
o

P
C

f
o
r

a
n
a
l
y
s
i
s
u
s
i
n
g

B
e
t
a

W
i
n
d
o
w
s
9
5
/
N
T

a
n
a
l
y
s
i
s
s
o
f
t
w
a
r
e
.

A
l
s
o
c
a
p
a
b
l
e

o
f

m
o
n
i
t
o
r
i
n
g
c
o
m
p
r
e
s
s
o
r
s

a
n
d
r
o
t
a
t
i
n
g

e
q
u
i
p
m
e
n
t
.
B
a
t
t
e
r
y

p
o
w
e
r
e
d
.
T
w
o

h
o
t
-
s
w
a
p
p
a
b
l
e
b
a
t
t
e
r
i
e
s
,

5
h
r
s
.
/
p
a
i
r
.

4
8
6
-
5
0
p
r
o
c
e
s
s
o
r
.

1
0

M
B
D
a
t
a

s
t
o
r
a
g
e
.
L
i
b
e
r
t
y
T
e
c
h
n
o
l
o
g
i
e
s
/
B
e
t
a

1
-
8
0
0
-
9
5
0
-
B
E
T
A
D
i
e
s
e
l

T
r
a
p
1

c
h
a
n
n
e
l
+

c
h
a
n
n
e
l
H
a
n
d
-
h
e
l
d

e
n
g
i
n
e
a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s

w
i
t
h

P
C
s
o
f
t
w
a
r
e
D
a
t
a

c
a
n

b
e

e
x
p
o
r
t
e
d
t
o

P
C

f
o
r

a
n
a
l
y
s
i
s
u
s
i
n
g

B
e
t
a

D
O
S
-
b
a
s
e
d

a
n
a
l
y
s
i
s
s
o
f
t
w
a
r
e
B
a
t
t
e
r
y

p
o
w
e
r
e
d
B
e
t
a

t
r
a
p
H
a
n
d
-
h
e
l
d
b
a
l
a
n
c
i
n
g

a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
P
m
a
x
,

P
m
a
x

s
t
a
t
i
s
t
i
c
s
N
o
n
e
D
a
t
a

c
a
n

b
e
d
o
w
n
l
o
a
d
e
d

t
o

P
C
P
e
a
k

f
i
r
i
n
g
p
r
e
s
s
u
r
e

b
a
l
a
n
c
i
n
g
o
n
l
y
.

B
a
t
t
e
r
y
p
o
w
e
r
e
d
.
N
K
F
/
C
A
R
M
A
C
A
R
M
A

8
D
P
o
r
t
a
b
l
e

P
C
-
b
a
s
e
d
e
n
g
i
n
e

a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s
I
n
c
l
u
d
e
s

a
n
a
l
y
s
i
s
s
o
f
t
w
a
r
e
-
-
N
K
F

E
n
g
i
n
e
e
r
i
n
g
7
0
3
-
3
5
8
-
8
6
0
0
P
M
C
/
B
e
t
a
B
e
t
a

2
5
0
P
o
r
t
a
b
l
e
o
s
c
i
l
l
o
s
c
o
p
e
-
b
a
s
e
d
e
n
g
i
n
e

a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
,

o
t
h
e
r
i
n
s
t
r
u
m
e
n
t

c
h
a
n
n
e
l
s
N
o
n
e
D
i
g
i
t
a
l

d
a
t
a

s
t
o
r
a
g
e
.
D
a
t
a

c
a
n

b
e

e
x
p
o
r
t
e
d
t
o

P
C
.
-
-
P
M
C
/
B
e
t
a
7
1
3
-
8
2
1
-
4
1
7
5
B
e
t
a

3
6
0
P
o
r
t
a
b
l
e
o
s
c
i
l
l
o
s
c
o
p
e
-
b
a
s
e
d
e
n
g
i
n
e

a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
,

o
t
h
e
r
i
n
s
t
r
u
m
e
n
t

c
h
a
n
n
e
l
s
.
P
m
a
x

a
n
d

P
m
a
x

c
a
n
b
e

r
e
a
d

f
r
o
m

s
c
r
e
e
n
.
N
o
n
e
N
o
n
e
N
o

d
a
t
a

s
t
o
r
a
g
e
c
a
p
a
b
i
l
i
t
y
C
S
I
/
W
i
n
d
r
o
c
k
M
o
d
e
l

6
1
0
0
E
n
g
i
n
e

m
o
n
i
t
o
r
i
n
g
s
y
s
t
e
m
U
p

t
o

2
5
0

i
n
p
u
t
s
,
i
n
c
l
u
d
i
n
g

c
y
l
i
n
d
e
r
p
r
e
s
s
u
r
e
,

c
r
a
n
k
a
n
g
l
e
,

e
n
g
i
n
e
s
y
s
t
e
m

p
r
e
s
s
u
r
e
s
a
n
d

t
e
m
p
e
r
a
t
u
r
e
s
,
v
i
b
r
a
t
i
o
n

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
.

O
t
h
e
r
s
d
e
p
e
n
d

o
n

i
n
p
u
t
s

a
n
d
c
a
n

i
n
c
l
u
d
e

a
i
r
/
f
u
e
l
r
a
t
i
o
,

l
u
b
e

o
i
l
c
o
n
d
i
t
i
o
n
,
e
m
i
s
s
i
o
n
s
.
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s
,
c
o
m
p
a
r
i
s
o
n
s
v
e
r
s
u
s

b
a
s
e
l
i
n
e
s
O
n
l
i
n
e

m
o
n
i
t
o
r
i
n
g

v
i
a
P
C

w
o
r
k
s
t
a
t
i
o
n
P
e
r
m
a
n
e
n
t
l
y
m
o
u
n
t
e
d

e
n
g
i
n
e
m
o
n
i
t
o
r
i
n
g

s
y
s
t
e
m
W
i
n
d
r
o
c
k

I
n
c
.
4
2
3
/
5
3
9
-
5
9
4
4
M
o
d
e
l

6
3
0
0
H
a
n
d
-
h
e
l
d

e
n
g
i
n
e
a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s
D
i
g
i
t
a
l

d
a
t
a

s
t
o
r
a
g
e
.
D
a
t
a

c
a
n

b
e

e
x
p
o
r
t
e
d
t
o

P
C
.
B
a
t
t
e
r
y

p
o
w
e
r
e
d
5-3
T
a
b
l
e

5
-
1
E
n
g
i
n
e

A
n
a
l
y
s
i
s

E
q
u
i
p
m
e
n
t

S
u
m
m
a
r
y

(
c
o
n
t
i
n
u
e
d
)
V
e
n
d
o
r
M
o
d
e
l
D
e
s
c
r
i
p
t
i
o
n
M
e
a
s
u
r
e
d

I
n
p
u
t
s
O
u
t
p
u
t

V
a
l
u
e
s
B
u
i
l
t
-
i
n
A
n
a
l
y
s
i
s
C
a
p
a
b
i
l
i
t
i
e
s
U
n
i
q
u
e

F
e
a
t
u
r
e
s
C
o
m
m
e
n
t
s
V
e
n
d
o
r

C
o
n
t
a
c
t
I
n
f
o
.
L
i
b
e
r
t
y
T
e
c
h
n
o
l
o
g
i
e
s
/
B
e
t
a
R
e
c
i
p
-
t
r
a
p
(
R
T
-
9
2
4
0
)
H
a
n
d
-
h
e
l
d

e
n
g
i
n
e
a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n

u
l
t
r
a
s
o
n
i
c
,
t
e
m
p
e
r
a
t
u
r
e
s
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,

P
-
,
P
-
V
,

v
i
b
r
a
t
i
o
n

b
o
t
h
f
r
e
q
u
e
n
c
y

t
i
m
e
d
o
m
a
i
n

u
l
t
r
a
s
o
n
i
c
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s

w
i
t
h

P
C
s
o
f
t
w
a
r
e
D
a
t
a

c
a
n

b
e

e
x
p
o
r
t
e
d
t
o

P
C

f
o
r

a
n
a
l
y
s
i
s
u
s
i
n
g

B
e
t
a

W
i
n
d
o
w
s
9
5
/
N
T

a
n
a
l
y
s
i
s
s
o
f
t
w
a
r
e
.

A
l
s
o
c
a
p
a
b
l
e

o
f

m
o
n
i
t
o
r
i
n
g
c
o
m
p
r
e
s
s
o
r
s

a
n
d
r
o
t
a
t
i
n
g

e
q
u
i
p
m
e
n
t
.
B
a
t
t
e
r
y

p
o
w
e
r
e
d
.
T
w
o

h
o
t
-
s
w
a
p
p
a
b
l
e
b
a
t
t
e
r
i
e
s
,

5
h
r
s
.
/
p
a
i
r
.

4
8
6
-
5
0
p
r
o
c
e
s
s
o
r
.

1
0

M
B
D
a
t
a

s
t
o
r
a
g
e
.
L
i
b
e
r
t
y
T
e
c
h
n
o
l
o
g
i
e
s
/
B
e
t
a

1
-
8
0
0
-
9
5
0
-
B
E
T
A
D
i
e
s
e
l

T
r
a
p
1

c
h
a
n
n
e
l
+

c
h
a
n
n
e
l
H
a
n
d
-
h
e
l
d

e
n
g
i
n
e
a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s

w
i
t
h

P
C
s
o
f
t
w
a
r
e
D
a
t
a

c
a
n

b
e

e
x
p
o
r
t
e
d
t
o

P
C

f
o
r

a
n
a
l
y
s
i
s
u
s
i
n
g

B
e
t
a

D
O
S
-
b
a
s
e
d

a
n
a
l
y
s
i
s
s
o
f
t
w
a
r
e
B
a
t
t
e
r
y

p
o
w
e
r
e
d
B
e
t
a

t
r
a
p
H
a
n
d
-
h
e
l
d
b
a
l
a
n
c
i
n
g

a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
P
m
a
x
,

P
m
a
x

s
t
a
t
i
s
t
i
c
s
N
o
n
e
D
a
t
a

c
a
n

b
e
d
o
w
n
l
o
a
d
e
d

t
o

P
C
P
e
a
k

f
i
r
i
n
g
p
r
e
s
s
u
r
e

b
a
l
a
n
c
i
n
g
o
n
l
y
.

B
a
t
t
e
r
y
p
o
w
e
r
e
d
.
N
K
F
/
C
A
R
M
A
C
A
R
M
A

8
D
P
o
r
t
a
b
l
e

P
C
-
b
a
s
e
d
e
n
g
i
n
e

a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s
I
n
c
l
u
d
e
s

a
n
a
l
y
s
i
s
s
o
f
t
w
a
r
e
-
-
N
K
F

E
n
g
i
n
e
e
r
i
n
g
7
0
3
-
3
5
8
-
8
6
0
0
P
M
C
/
B
e
t
a
B
e
t
a

2
5
0
P
o
r
t
a
b
l
e
o
s
c
i
l
l
o
s
c
o
p
e
-
b
a
s
e
d
e
n
g
i
n
e

a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
,

o
t
h
e
r
i
n
s
t
r
u
m
e
n
t

c
h
a
n
n
e
l
s
N
o
n
e
D
i
g
i
t
a
l

d
a
t
a

s
t
o
r
a
g
e
.
D
a
t
a

c
a
n

b
e

e
x
p
o
r
t
e
d
t
o

P
C
.
-
-
P
M
C
/
B
e
t
a
7
1
3
-
8
2
1
-
4
1
7
5
B
e
t
a

3
6
0
P
o
r
t
a
b
l
e
o
s
c
i
l
l
o
s
c
o
p
e
-
b
a
s
e
d
e
n
g
i
n
e

a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
,

o
t
h
e
r
i
n
s
t
r
u
m
e
n
t

c
h
a
n
n
e
l
s
.
P
m
a
x

a
n
d

P
m
a
x

c
a
n
b
e

r
e
a
d

f
r
o
m

s
c
r
e
e
n
.
N
o
n
e
N
o
n
e
N
o

d
a
t
a

s
t
o
r
a
g
e
c
a
p
a
b
i
l
i
t
y
C
S
I
/
W
i
n
d
r
o
c
k
M
o
d
e
l

6
1
0
0
E
n
g
i
n
e

m
o
n
i
t
o
r
i
n
g
s
y
s
t
e
m
U
p

t
o

2
5
0

i
n
p
u
t
s
,
i
n
c
l
u
d
i
n
g

c
y
l
i
n
d
e
r
p
r
e
s
s
u
r
e
,

c
r
a
n
k
a
n
g
l
e
,

e
n
g
i
n
e
s
y
s
t
e
m

p
r
e
s
s
u
r
e
s
a
n
d

t
e
m
p
e
r
a
t
u
r
e
s
,
v
i
b
r
a
t
i
o
n

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
.

O
t
h
e
r
s
d
e
p
e
n
d

o
n

i
n
p
u
t
s

a
n
d
c
a
n

i
n
c
l
u
d
e

a
i
r
/
f
u
e
l
r
a
t
i
o
,

l
u
b
e

o
i
l
c
o
n
d
i
t
i
o
n
,
e
m
i
s
s
i
o
n
s
.
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s
,
c
o
m
p
a
r
i
s
o
n
s
v
e
r
s
u
s

b
a
s
e
l
i
n
e
s
O
n
l
i
n
e

m
o
n
i
t
o
r
i
n
g

v
i
a
P
C

w
o
r
k
s
t
a
t
i
o
n
P
e
r
m
a
n
e
n
t
l
y
m
o
u
n
t
e
d

e
n
g
i
n
e
m
o
n
i
t
o
r
i
n
g

s
y
s
t
e
m
W
i
n
d
r
o
c
k

I
n
c
.
4
2
3
/
5
3
9
-
5
9
4
4
M
o
d
e
l

6
3
0
0
H
a
n
d
-
h
e
l
d

e
n
g
i
n
e
a
n
a
l
y
z
e
r
C
y
l
i
n
d
e
r

p
r
e
s
s
u
r
e
,
c
r
a
n
k

a
n
g
l
e
,
v
i
b
r
a
t
i
o
n
,

u
l
t
r
a
s
o
n
i
c
P
m
a
x
,

P
m
a
x
,

d
P
,
I
M
E
P
,

I
H
P
,
P
-
,

P
-
V
,

v
i
b
r
a
t
i
o
n
,
u
l
t
r
a
s
o
n
i
c
H
i
s
t
o
r
i
c
a
l

d
a
t
a
t
r
e
n
d
i
n
g

a
n
d
a
n
a
l
y
s
i
s
D
i
g
i
t
a
l

d
a
t
a

s
t
o
r
a
g
e
.
D
a
t
a

c
a
n

b
e

e
x
p
o
r
t
e
d
t
o

P
C
.
B
a
t
t
e
r
y

p
o
w
e
r
e
d
5-4
5.2 Cylinder Pressure Transducers
5.2.1 Pressure Transducer Characteristics
Most engine analyzers are provided with pressure transducers tailored for the specific
analyzer. However, some analyzers do provide for the use of a user-selected pressure
transducer and amplifier. A comprehensive discussion of all of the issues related to
cylinder pressure transducer selection is beyond the scope of this guide. Transducers
are available in many different types, each with unique advantages and disadvantages.
References 1022 provide guidance on pressure transducer characteristics and selection.
If use of user-selected pressure transducers and amplifiers is desired, the following
characteristics should be considered:
LinearityA linear transducer delivers an output signal strictly proportional to
input pressure. Linearity is the deviation from strict proportionality.
SensitivityThe sensitivity of a transducer determines how large a change occurs in
the output signal for a given change in the input pressure.
DampingDamping describes how quickly the output signal of a transducer settles.
Dynamic responseThe transducer must be capable of measuring rapid pressure
changes associated with combustion.
HysteresisTransducer hysteresis is the maximum difference in output signal for
the same measured input pressure when that pressure is approached from above
and below.
Mounting characteristicsThe transducer must be able to be mounted within the
space available and using methods that will not result in additional errors.
Cooling requirementsMany transducers require external cooling to prevent dam-
age or drift.
Each of these characteristics can be discussed with the transducer or engine analysis
equipment manufacturers.
5.2.2 Transducer Preparation
Cylinder pressure transducers are exposed to rapid (for example, 300 times per minute
in a 600 RPM 4-stroke engine) thermal transients. Many transducers are adversely
effected by these temperature transients. References 14, 18, 22, and 23 discuss the ac-
tions necessary to minimize the effects of the thermal cycles. Specifically, some vendors
recommend that the face of the transducer be coated with a thin layer of RTV for flush
mounted transducers. Most EDGs in nuclear service have relatively long indicator
passages, which reduce the magnitude of the thermal cycles. The manufacturer of the
analyzer or transducer should be consulted before the cylinder pressure transducers are
prepared. Also, it should be ensured that any vendor-recommended cooling is always
available during data collection.
5-5
5.2.3 Transducer Calibration
Calibration of the transducer and associated amplifiers contributes directly to the accu-
racy of the engine balancing and analysis data. Calibration instructions should be pro-
vided with the engine analyzer. However, experience shows there are three areas re-
garding calibration that merit special note.
1. Calibration should include all components of the engine analyzer including the
amplifier, cables, read-out devices, and mechanical mounting hardware. This end-
to-end calibration is the only way to account for all system variances. The system
should be recalibrated following change out of any component.
2. Some pressure transducers react to dynamic pressure only; static pressures decay
rapidly. This is particularly true of piezoelectric or crystal transducers. Special
dynamic pressure calibration devices are available to calibrate these transducers
accurately. However, static calibration devices, such as dead weight testers, can be
used with success as long as the operator is aware of the dynamic effects and the
transducer vendor supports static calibrations.
3. Traditional cylinder pressure transducers are somewhat fragile. They should be
checked periodically during use and any time the transducer is dropped, over-
heated, or the transducer mounting is changed.
Transducer calibration requirements vary by vendor and historical experience with
specific devices. However, the calibration should always be spot checked following
balancing to ensure that the transducer has not drifted. Cylinder pressure transducers
typically cannot accurately measure low or absolute pressures. Therefore, many engine
analyzers support direct or indirect input of a reference pressure as the baseline for the
pressure transducer every revolution. Usually, the cylinder pressure at BDC (before the
compression stroke for four-stroke engines) is assumed to be equal to the pressure of
the inlet manifold [24]. This correction should always be used when it is supported by
the engine analyzer. Failure to input a reference pressure will allow the pressure read-
ing to drift from cycle to cycle.
5.2.4 Transducer Mounting
Most pressure transducers are mounted in an adapter that facilitates the connection to
the cylinder test connection. Forces exerted on the transducer housing can pre-load the
transducer and cause an offset in the output. For this reason, it is recommended that the
transducer be mounted in the adapter prior to calibration. Further, the transducer
calibration should be checked each time it is mounted in the adapter.
5.3 Vibration and Ultrasonic Accelerometers
5.3.1 Accelerometer Characteristics
Accelerometers transform mechanical vibration and ultrasonic signals into electrical
signals proportional to the acceleration. Most accelerometers use piezoelectric materials
5-6
as the sensing element. Accelerometers should be used to measure signals well below
their natural frequencies. Usually, the operating frequency range is limited to approxi-
mately ten percent of the natural frequency [25]. The frequency response of an acceler-
ometer is the range of frequencies that can be measured without significant error in the
output signal due to resonance of the accelerometer. The high end of the frequency
response for an accelerometer is typically defined as the frequency at which the increase
in vibration amplitude due to resonance of the accelerometer is 5%. Note that the usable
frequency range of an accelerometer is not limited by its frequency response. However,
the error in vibration amplitude must be considered when measuring signals with the
frequency components greater than the frequency response of the accelerometer [26].
The frequency response of an accelerometer is usually provided by the accelerometer
manufacturer.
5.3.2 Accelerometer Calibration
Accelerometers are calibrated to determine their sensitivity. Accelerometers should be
checked prior to every use from front-to-back, including cables and the amplifier.
Hand-held calibration devices that generate a calibrated vibration (frequency and
magnitude) are available from a number of vendors.
5.3.3 Accelerometer Mounting
The method of mounting an accelerometer effects its frequency response. The objective
is to use a mounting technique that provides a flat response within the frequency range
of interest. Several different methods of mounting accelerometers, along with their
effect on frequency response, are described below in order of preference [27]:
Stud mountingThe accelerometer is mounted to the measurement surface using a
threaded stud. Of all the mounting methods, stud mounting provides the best fre-
quency response. However, convenient stud mounting locations are not present on
many diesel engines.
Adhesive mountingThe accelerometer is directly cemented to the measurement
surface or a pad mounting is cemented to the surface. A pad mounting includes a
stud for attaching the accelerometer. The frequency response of an accelerometer
mounted with adhesive is almost equivalent to a stud mounting. This is the recom-
mended method for vibration analysis of engines where stud mounting locations are
not available.
Magnetic mountingA magnet is attached to the base of the accelerometer. The
magnet is then attached to the measurement surface. The frequency response of an
accelerometer with a magnetic mount is acceptable for mounting accelerometers in
temporary locations.
Hand-held probeThese probes are typically cylindrical and end in a point which is
held against the measurement surface. Although hand-held probes are very easy to
use, they are not recommended because their frequency response is usually much
worse than other mounting methods.
5-7
Accelerometers with a magnetic mount have been found to produce acceptable vibra-
tion measurements if adhesive-mounted studs cannot be used. The vibration data
shown in Figures 5-1 and 5-2 were measured using a Kistler Model 8622 accelerometer
with a magnetic mount. The manufacturers listed frequency response for the acceler-
ometer is 1.00 Hz to 8.00 kHz. This frequency response is reported for a stud mounted
accelerometer. The manufacturer indicated that the frequency response of the acceler-
ometer would decrease when using a magnetic mount. However, if the accelerometer is
mounted on a relatively flat, smooth surface such as a cylinder head stud, and a suffi-
ciently powerful mounting magnet is used, the frequency response of the accelerometer
might not degrade significantly due to the magnetic mount.
1600
1400
1200
1000
800
600
400
200
0
-180
-90 0 90 180 270 360 450 540 630
-10.0
-5.0
0.0
5.0
10.0
A
c
c
e
l
e
r
a
t
i
o
n

(
g
)
C
y
l
i
n
d
e
r

P
r
e
s
s
u
r
e

(
p
s
i
)
Figure 5-1
Typical Engine Pressure and Vibration Data Measured on a Cylinder Head Stud
5-8
2000
1500
1000
500
0
-360 -180 0 180 360
5.00g
5.00g
p
s
i
Figure 5-2
Typical Engine Pressure and Vibration Data Measured on the Cylinder Liner Flange
Vibration measurements are typically taken on the engine structure (that is, cylinder
liner flange or cylinder head stud). The location of the accelerometer has a significant
effect on the measured vibration signal. In general, mounting the accelerometer closer
to the source of vibration will produce a stronger vibration signal. Placing the acceler-
ometer at one location can give a strong signal for exhaust valve closure while another
location can give a strong signal for piston impacts. It is desirable, however, to use a
location where all the various vibration signals can be measured conveniently.
When first taking vibration readings on an engine, various accelerometer locations
should be used to identify the best location for measuring a clean signal. As an example,
Figure 5-2 is a pressure and vibration trace for an engine measured from the cylinder
liner flange. If the vibration trace of Figure 5-2 is compared to the vibration trace of
Figure 5-1, it can be seen that the vibration signature is much stronger when measured
from the cylinder head stud and, therefore, the cylinder head stud would be the pre-
ferred location.
Successive measurements on a cylinder should always be taken at the same location
(with the same accelerometer and mounting) such that comparisons between the data
are consistent. In addition, since it is useful to compare data from different cylinders,
measurements should be taken at the same location on each cylinder. For most diesel
engines, it has been found that the cylinder head studs provide a good location for
5-9
identifying mechanical impacts in the engine. Note that the same cylinder head stud
must be used for every cylinder, since different head studs produce different vibration
signatures. For the Fairbanks Morse Opposed Piston engine, it is recommended that the
accelerometers be mounted on the cylinder liner next to the fuel injector.
5.4 Filtering and Signal Conditioning
Engine analyzers with built-in vibration measurement capabilities often have internal
filtering and signal processing. Figure 5-1 shows an unfiltered vibration signal from a
Cooper-Bessemer Model KSV engine. It is important when using an engine analyzer
with vibration measurement capabilities, that the user understand the filtering and
signal processing that is performed by the analyzer, if any. For example, the BETA
RECIP-TRAP engine analyzer has vibration measurement and analysis capabilities.
Figure 5-2 is an example of a vibration signature of a Cooper-Bessemer Model KSV
engine taken with the BETA RECIP-TRAP. The vibration signature looks similar to the
signature shown in Figure 5-1. However, the RECIP-TRAP displays a filtered and pro-
cessed vibration signal, rather than the raw vibration signal. The same filtering and
signal processing must be used for vibration data to be compared from one session to
another.
The measured vibration signal contains vibration from other sources besides the desired
source wave form. Signal processing is required to remove the unwanted components
of the vibration signal and recover, as best possible, the source vibration signal. Various
types of signal processing can be used. Two examples are listed below:
Band-pass filteringFiltering is used to remove unwanted frequency components of
the vibration signal. The high-pass filter removes frequencies associated with the
engine speed and the low-pass filter removes frequencies above the mounted fre-
quency response of the accelerometer. The low-pass filter should be determined
based on the frequency response of the accelerometer and the sampling frequency.
Time-domain (synchronous) averagingData points one period (engine cycle) apart
are averaged. This type of averaging removes components of the signal that are not
linked to the engine period (that is, random noise). This method of filtering is typi-
cally not used by engine analyzers.
Signal processing can be performed as the measurements are being recorded or after
measurements are taken (post-processing). If the analyzer supports post-processing, it is
recommended that filtering be accomplished after the data is acquired and recorded.
The use this method allows the user to refer back to raw data if questions arise. All
frequencies above half the sampling frequency range of the accelerometer should be
removed from the signal. For example, if the sampling frequency is 50 kHz then a low-
pass filter should be used to remove all frequencies above 25 kHz.
Until a sufficient base of data is built up for the engines at a particular station, it is
recommended that vibration be recorded for individual engine cycles (no averaging) if
your analyzer supports post-data acquisition averaging. Data can be taken for multiple
5-10
cycles on the same cylinder and time-domain averaging can be performed as a part of
post-processing.
5.5 Sampling Rate
The sampling rate used to acquire vibration data should be determined based on the
frequency components associated with vibration signal. The sampling rate should be a
minimum of twice the maximum frequency components of the vibration signal to
ensure that signal aliasing is avoided.
5.6 Sources of Error
The usefulness of engine analyzer data for analysis and balancing is highly dependent
on the accuracy of the data. The potential for error exists at every step of the engine
analysis process; from setup of analysis equipment to review of analysis data. Many of
the calculated engine performance parameters, such as IMEP and IHP, are very sensi-
tive to small data acquisition errors. While EDG engine analysis is not intended to be a
research project, it is prudent to keep data acquisition errors as low as practical. Poten-
tial sources of error are discussed in this section and include:
Improper selection, calibration, preparation, and mounting of cylinder pressure
transducers
Effects of the cylinder pressure transducer indicator passage
Phasing errors between the cylinder pressure measurement and the crankshaft
position
Load drift during data collection
Analyzer setup
The analyzer measures cylinder pressure and crank position and uses these two vari-
ables to calculate other engine performance parameters of interest. In order for the
analyzer to make these calculations correctly, the correct engine geometry information
such as cylinder bore and piston stroke must be entered.
5.6.1 Selection, Calibration, Preparation, and Mounting of Cylinder Pressure
Transducers
Selection, calibration, preparation, and mounting of cylinder pressure transducers are
discussed in Section 5.2.
5.6.2 Indicator Passage Effects
On most large diesel engines, internal cylinder pressure is recorded by an indicator
passage machined into the cylinder head (or cylinder liner for FMOP engines). These
passages range in length from several inches to a few feet. An indicator valve, or Kiene
5-11
valve, is mounted to the end of the indicator passage by a cast iron boss. Pressure trans-
ducers capable of recording pressures inside the cylinder typically cannot be mounted
to the Kiene valve without the use of an extension. The resulting configuration (a long,
narrow passage) interferes with accurate dynamic cylinder pressure measurements. For
this reason, engine research has usually been performed with the cylinder pressure
transducer mounted flush with the cylinder head flame face. However, this is not prac-
tical for field measurements of cylinder pressure. The effects of the indicator passage,
indicator valve, and extensions are well documented. References 28 through 36 discuss
the effects:
Delays of the cylinder pressure signal, inducing a significant phase shift between the
actual peak pressure angle and the measured peak pressure angle
Damping of the cylinder pressure signal, reducing the magnitude of the signal
Resonance effects due to indicator passage geometry that could cause a standing
pressure wave. Any resonance could result in pressure oscillations and indicated
peak pressures that are much higher than what actually occur in the cylinder.
These effects are present to varying degrees in all indicator passages. The phase shift
between crank angle and cylinder pressure will typically be the most significant effect.
The most accurate method of ensuring proper phasing between the TDC reference mark
and engine cylinder pressure is by using the thermodynamic loss angle (TLA) method
during a compression motoring test on the reference cylinder. This method will also
accommodate for the effect of the indicator passage.
Many analyzers allow the operator to input the known pressure delay or phase shift
manually into the analyzer, and the analyzer will then make all the necessary correc-
tions. Otherwise, the crank angle timing must be adjusted to offset the time delay or
phase shift. This can be accomplished by mechanically shifting the crank angle mea-
surement by the known delay angle.
5.6.3 Crank Angle Phasing and Correction
Many sources [24,25,28,3035] address the sensitivity of correct phasing of crank angle
and cylinder pressure. Clearly, errors in crank angle measurement directly offset
timing measurements on the P- diagram, such as
Pmax
and the start of combustion
(SOC); while other crank angle insensitive parameters such as P
max
are unaffected.
Calculated parameters, in particular IMEP and IHP, are significantly sensitive to phas-
ing errors between cylinder pressure and crank angle. Reference 24 reports sensitivities
as high as 5% per degree error for full-load IMEP and 40% per degree for motoring
IMEP.
The literature recommends that the crank angle (and TDC) be accurately measured to
within 0.1 of crank angle for activities that involve studying engine combustion. How-
ever, this value is likely to be unrealistic for non-laboratory settings. Instead, it is recom-
mended that the total error on TDC and crank angle be limited to approximately 0.5
5-12
using reasonable methods, and that measurements of the total error be made to ensure
the total error remains in the acceptable range.
There is more than one method for accurately determining the true TDC for each cylin-
der and ensuring the cylinder pressure is phased correctly with the crank angle. The
timing marks on the engine flywheel are the traditional method for determining TDC.
However, flywheel marks can be mismarked or painted over. Therefore, an alternative
method for determining TDC should be used at least once to verify the TDC readings.
The most accurate method for determining TDC is the thermodynamic method. Also, it
is typically the most time effective method. The mechanical method was historically
used prior to the development of electronic engine analysis methods. However, the
mechanical method is very time consuming. Further, the results of the mechanical
method must be corrected for the effects of the indicator passage, which is frequently
not a straightforward effect, whereas the thermodynamic method is self-correcting.
Thermodynamic methodThe location of peak pressure in an engine cylinder with no
combustion must fall slightly before TDC. This phenomenon, which has been referred
to as thermodynamic loss angle (TLA), [29] can be explained by considering the three
factors acting to change the cylinder pressure in a motoring cylinder:
Volume change caused by cylinder motion
Heat transfer
Gas leakage
During the majority of the motoring engine cycle, piston motion is the dominant effect.
But as the piston approaches TDC, its velocity slows. At the instant the piston is at TDC,
the piston is motionless. During the approach to TDC, the pressure and temperature of
the cylinder gases are reaching their highest points. Consequently, leakage and heat
transfer, both of which serve to reduce cylinder pressure, assert their greatest influence.
The rise in pressure due to compression is overwhelmed by leakage and heat transfer as
the piston approaches TDC, causing the pressure to begin decreasing before the piston
reaches its peak location. Only during isentropic (that is, quasi-static with no heat transfer
or leakage) compression would peak pressure occur precisely at TDC. All engines are
subject to at least some heat transfer and leakage loss and therefore have peak pressure
located slightly before TDC. Noise or indicator passage errors could cause measured
pressure histories to appear to contradict this fact. Independent observations have given
some insight to the proper location of peak monitoring pressure for ensemble averaged
pressure histories. Thermodynamic lag angle has been reported to be inversely propor-
tional to engine speed [29] and to be on the order of one degree [39]. However, these
results cannot be considered universal because the leakage and heat transfer will vary
from engine-to-engine (and from cylinder to cylinder on a given engine).
If leakage and heat-transfer rates were known exactly, a first law of thermodynamics
model could predict the exact location of the peak compression pressure. Unfortunately,
these quantities are difficult to predict accurately enough to use such a model to estab-
lish the location of TDC. The first law of thermodynamics can be used to write an
5-13
expression for the polytropic constant that depends on engine speed, engine geometry,
trapped mass, pressure, heat transfer, and leakage [38].
( )
n
w
Q
PV
m
m
V
l l
dV
d
+ +
1
]
1

1
1
&
&
Eq. 5-1
where
n = polytropic constant
= specific heat ration
w = engine frequency
&
Q
l
= rate of heat loss
P = cylinder pressure
V = cylinder volume
&
m
l
= rate of blowby mass loss
m = mass of cylinder contents
dV
d
= rate of change of cylinder volume
This equation shows that near TDC, the polytropic constant changes rapidly as the rate
of change of cylinder volume approaches zero. Because the rate of change of cylinder
volume changes sign, the polytropic constant goes from negative infinity to positive
infinity as the piston moves through TDC. Expanding Equation 5-1 by using a Taylor
series and simplifying the resulting expression by neglecting the higher order terms, an
expression for the polytropic constant as a function of crank angle can be obtained:
n () = K
P
1
_
,
Eq. 5-2
where
n = polytropic constant
K = constant
P
= angle of peak pressure
= crank angle
The values of K and
P
are dependent on the constants of the first order terms in the
Taylor series, which approximate the heat-transfer and leakage terms. These constants
would be more difficult to evaluate explicitly than the leakage and heat transfer terms
that they approximate. However, the measured pressure data can be used to calculate
the polytropic constant at two particular crank angles,
1
and
2
, using the following
equation.
n =
( )
( )
ln
ln
P P
V V
1 2
2 1
5-14
1
and
2
should be less than 10 to ensure that the small angle assumption used in the
derivation of Equation 5-2 is valid. These values can then be substituted into Equation
5-2 to solve for K and
P
.

Equation 5-2 can be compared to the measured polytropic
constant at each crank angle near TDC. This process can be iterated with incremental
changes to the TDC phasing of the measured pressure until the measured polytropic
exponent versus crank angle is closest to the relationship predicted by Equation 5-2.
This method is accurate to within approximately 0.3 of crank angle [38].
Mechanical Method of Checking/ Setting TDCMost diesel engines are provided with
timing marks inscribed on the engine flywheel or barring gear. These timing marks
provide a reference for TDC for each cylinder. Typical factory-provided timing marks
are accurate to within 1. However, this timing mark setting accuracy can and should
be improved to support accurate engine balancing. For engines with single-acting
pistons a dial indicator with a long extension is required to determine TDC accurately
using the mechanical method. In addition, a bushing or guide to center the dial indica-
tor accurately in the fuel injection port of the cylinder head should be fabricated.
An initial estimate of TDC should be made by using the existing timing marks and
aligning the pointer with the TDC mark on the flywheel when the piston is at this
position. This initial setting is only approximate because the piston vertical position in
the cylinder changes very slowly with crank angle near TDC. This makes the exact TDC
difficult to establish. A more accurate determination of TDC is made by utilizing the
more rapid change in piston position with crank angle that occurs at approximately 40
on either side of TDC. With the dial indicator tip contacting the top of the piston crown,
the engine is rotated about 60 from the approximate TDC in either direction. It is then
rotated back toward TDC and stopped at 40 from TDC (as indicated by flywheel mark-
ings). A dial indicator reading is taken. The engine is now rotated 60 past TDC in the
other direction and then rotated back toward TDC until the dial indicator reading is the
same as the one taken at 40 BTDC on the other side. The crank angle at that position is
read. By taking both readings as the piston is moved toward TDC, the effects of changes
in mechanical clearances are minimized. TDC is the crank angle at the midpoint be-
tween equal dial indicator readings. The engine is rotated until this midpoint is aligned
with the pointer, setting the piston at minimum cylinder volume. The TDC mark is then
prescribed with the pointer. The above procedure might have to be repeated to refine
the readings. Additionally, it could be done for each cylinder.
The precision of the mechanical method for checking TDC is dependent on the precision
of crank angle readings taken from the flywheel. The degree markings machined on the
flywheel are assumed to be spaced accurately. Any errors in readings are then due to
the width of the pointer, the width of the flywheel degree markings, and parallax. This
mechanical method can be used to determine TDC to within 0.10.25.
TDC for each cylinder can be determined in this fashion and marked on the flywheel for
each EDG at a station. The baseline of the engine timing marks need only be performed
once.
5-15
Once TDC is established using this technique, fuel injection, and inlet and exhaust valve
timing can be checked by the refined flywheel timing marks. Further, the flywheel
timing marks can now be used to set up the crank angle probe or signal pick-up con-
nected to the engine analyzer. The method of accurately calibrating the crank angle
transducer is dependent upon the specific analyzer involved. The analyzer vendors
instructions should be followed in this area.
Note that crank angle, as measured at the flywheel, will vary due to torsional vibrations
or alternating twist in the crankshaft. These effects are small and do not effect the re-
sults of engine balancing and analysis.
5.6.4 Load Drift
Often, engine load will drift during the data acquisition process, resulting in variations
in cylinder pressure. To preclude this, it is recommended that the governor be locked to
preclude fuel rack movement, if possible. The governor should only be locked if the
implications of this action are understoodthat is, it might make the EDG inoperable
and if it needs to be locked due to load drift or changes. See Section 5.7 for additional
information.
5.6.5 Configuring the Analyzer
The cylinder pressure transducer calibration should be checked both before and after
engine data is recorded. In addition, checks to ensure TDC and crank angle position are
properly aligned with the crank angle transducer are appropriate each time the ana-
lyzer is used. Errors in crank angle and TDC of only a few degrees will skew the calcu-
lations of IHP and IMEP, as well as the measurements of
Pmax
. As a result, this data will
not be comparable to the baseline engine data to support diagnostics, if necessary.
The cylinder pressure transducer must be calibrated in accordance with utility calibra-
tion programs and based on operating experience. It is a good idea to check the opera-
tion of the transducer prior to starting and warming the engine to ensure that the ana-
lyzer is functional. A dead-weight tester provides the most accurate static calibration of
the cylinder pressure transducer. Some utilities have used high pressure gas in conjunc-
tion with a calibrated pressure gauge. Both of these calibration methods are static and
must be used with caution when using piezoelectric pressure transducers. These trans-
ducers are best for measuring dynamic pressures such as cylinder pressure in a recipro-
cating engine. Some companies manufacture dynamic pressure transducer calibration
machines, but they are expensive and their use is not necessary. The transducer should
be calibrated over the entire range of expected pressure values.
Crank angle and TDC can be checked by barring the engine in the direction of normal
rotation and ensuring that the TDC measurement coincides with the TDC mark on the
flywheel for the cylinder of interest. Typically, TDC for other cylinders is entered as an
offset to the TDC value of the cylinder for which TDC is measured directly.
5-16
There are a number of methods of generating crank angle signals. These include mag-
netic pickups operating off of crankshaft, camshaft, or flywheel gear teeth, an optical
pickup using reflective tape on the crankshaft or flywheel, and optical encoders rigidly
attached to the crankshaft. All engine analyzers are supplied with some type of crank
angle signal generator. It is desirable to have both a once per revolution signal to refer-
ence the TDC position, usually referred to as the TDC pulse, and a more frequent signal,
usually occurring every degree or every few degrees. It is possible to use a only a once
per revolution signal with the analyzer interpolating all of the crank angles between
each TDC position. However, if the crankshaft rotational speed is not constant during
each revolution, this will introduce some error into the engine analysis parameters that
are dependent on crank position.
Once the proper TDC location has been determined, the next step is to install the crank
angle measurement device so that the crank angle measurement will be properly
phased with the mechanical TDC. The indicator passage will delay the cylinder pres-
sure signal. As a result, the actual peak firing pressure will occur before the measured
peak firing pressure. If the mechanical method of phasing TDC is used, this delay must
be accounted for in the analyzer setup. Some engine analyzers allow input of an indica-
tor passage delay angle and correct the crank angle signal accordingly. If the analyzer
does not allow for input of a delay angle, the crank angle signal generator will have to
be manual offset to account for the delay. If the thermodynamic method of phasing
TDC is used, the indicator passage delay is already accounted for because the indicated
cylinder pressure is used to phase TDC rather than the actual cylinder pressure or
piston position.
5.7 Preparing the Engine for Taking Data
A number of steps must be accomplished prior to performing engine analysis. These
include: starting and loading the engine, allowing it to warm up to operating tempera-
ture, and setting the engine load.
5.7.1 Start and Load the Engine
The engine should be started and operated at rated load in accordance with utility and
vendor procedures. The engine should then be operated at full load for at least one hour
prior to taking data to ensure that the engine is at steady-state.
In many cases, the generator load is indicated in the engine bay as well as being re-
peated in other locations, such as the main control room. Often, differences in observed
readings between various power meters for a given diesel generator exist. For this
reason, it is desirable to select a single power meter for each engine that will be used for
all future engine analysis run for that engine. Further, the power meter readings should
be checked against the generator voltage, current, and power factor readings. In this
way, the same load will be used for all tests, ensuring that test results are comparable to
the baseline data for that engine.
5-17
Engine analysis data should be collected at or near the load at which the engine is
expected to operate. Note that the rated load per cylinder in terms of brake mean effec-
tive pressure (BMEP) might vary between sites for the same model engine due to differ-
ent rated loads or engine configurations. As a result, engine analysis data from different
sites and engines might not be directly compatible.
5.7.2 Lock the Governor
Often the load on the diesel generator will vary somewhat due to changing plant and
electrical grid conditions. Changes in overall engine load will produce changes in indi-
vidual cylinder loads. To minimize load swings, the load should be set by locking the
governor by the load limit knob on the governor actuator, if possible. Locking the
governor is accomplished by raising the load of the diesel generator (by the governor
speed setting) known to some value slightly greater than the final desired test value.
The diesel generator load will control the speed; therefore, the diesel generator should
be synchronized to the grid before locking the governor so that adequate speed control
is maintained. This method essentially prevents the fuel racks from moving unless the
diesel experiences a load rejection or over-speeds.
None of the safety features of the engine are disables by this technique. However, lock-
ing the load opens the formerly closed-loop speed control system. The diesel generator
should not be loaded by plant hotel loads (unless synchronized to the grid) during
engine analysis. The station operations department and appropriate engineering per-
sonnel should be consulted prior to locking the governor. If it is not appropriate to lock
the governor at your site, special attention must be paid to load swings during the test.
Finally, the load limit setting must be returned to its original value once testing is com-
pleted.
5.7.3 Set Engine Load
Establishing nearly the same load every time the engine analysis is performed is impor-
tant from the standpoint of data trending. Many utilities have experienced relatively
large discrepancies between local power meters and the power meters located in the
main control room. It is recommended that operations consistently choose a single
meter from the purpose of engine loading.
5.7.4 Record Initial Balance of Plant Data
Once the engine attains steady-state operating temperatures, the engine auxiliary sys-
tem temperatures and pressures should be recorded. The purpose of this data is to
document the operating conditions of the engine in case questions arise later. This data
will assist in diagnosing engine problems if any are identified, and to compare against
data taken at the end of the test to ensure operating conditions have not changed.
5-18
5.8 Measuring Cylinder Pressures
The actual measurement of cylinder pressure versus crank angle (and cylinder volume)
is straightforward. The actions needed to record cylinder pressure will vary among
engine analyzers. Documentation supplied with the engine analyzer should be con-
sulted as to the actual procedure. The following provides some general guidelines for
cylinder pressure measurement.
5.8.1 Indicating Valve Extension
If any adjustments have been made for the indicator passage, indicating valve, and
valve extension, check the adjustment by performing a motoring test on one or more
cylinders to ensure that the TLA is approximately 1 BTDC. In any case, the length and
type of the valve extension used should be recorded.
5.8.2 Check Cylinder Pressure Transducer Operation
Cylinder pressure transducers are available in a large number of types. Many of these
devices are relatively fragile and tend to drift with changes in temperature. Few can
withstand more than one hundred hours of engine operation before replacement is
required. Because of this, the cylinder pressure transducer should be checked before
and after the pressure readings have been taken. Some utilities require that the cylinder
pressure transducer be calibrated or checked more or less frequently, such as once per
engine bank. Note that the pressure readings should be rerecorded if the transducer
goes out of calibration during the readings. Ensure that any necessary cooling methods,
such as cooling air directed at the fittings or transducer cooling water systems, are used
during the calibration.
5.8.3 Reference Pressure
The inlet air manifold pressure at rated load should be recorded. This pressure should
also be used as the reference pressure during data acquisition. The inlet air manifold
pressure is assumed to be the same as the cylinder pressure with the piston at BDC
(during the exhaust stroke for four-stroke engines). The reference pressure is used to
correct dynamic pressure transducers, such as piezoelectric devices, that do not mea-
sure static pressures.
5.8.4 Record Cylinder Pressure
Specific instructions for recording cylinder pressure are available from each engine
analyzer vendor. However, there are a few technical issues that should be highlighted.
The indicator passage should be cleared prior to recording cylinder pressure. This is
accomplished by opening the indicating valve fully without the cylinder pressure
transducer attached to the valve extension for several engine cycles or until all debris
has been discharged. The purpose of this is to discharge any carbon or other products of
5-19
combustion that might have become trapped in the indicator passage. The presence of
foreign material will cause erroneous readings if it is not blown out. The indicator valve
must be closed prior to attaching the cylinder pressure transducer.
Some cycle-to-cycle variation in combustion characteristics for a given cylinder is nor-
mal. Because of this, cylinder pressure should be recorded for approximately 200 cycles
to obtain an average reading. If the analyzer is not capable of recording 200 cycles,
record as many as possible. Some variation in cycle-to-cycle peak firing pressures are
expected [33,35,36]. Averaging numerous cycles eliminates many of these effects as well
as reducing the effect of signal noise. Many engine analyzers provide statistics on cycle-
to-cycle variation in cylinder pressure. These statistics are important and should be
recorded. Examples include standard deviation, combustion quality index, and combus-
tion variation index. All are based on standard deviation calculations. Because there is
not an industry-wide standard for combustion variation, standard deviation will be
used throughout this manual.
The P-V and P- curves for each cylinder should be recorded at this time. (In most cases,
cylinder vibration and ultrasonic signatures are also recorded concurrently.) If the
engine analyzer is capable, both individual and average P- curves should be recorded.
Given the choice of plotting one cycle or a composite, which represents the average of
several cycles, the engine average (composite) is preferred. The composite reading
provides the data on the average cylinder combustion characteristics.
Cylinder pressure data is taken for each cylinder during the balancing process. The
order in which the cylinder pressure data is recorded is not important. However, if the
balancing process is interrupted for any reason, such as to make adjustments to the fuel
injection system and the engine is shut down, then the entire balancing process must be
repeated for all cylinders. Adjusting the load carried by one cylinder changes the load
carried by all of the cylinders.
A brief inspection of each curve should be performed at the time the data is taken.
Abnormal cylinder pressure data could be caused by:
A cylinder pressure transducer that has drifted or malfunctioned
A loose connection between the cylinder pressure transducer and the indicating
valve extension
A blocked indicator passage
A malfunction of the engine analyzer
An actual combustion anomaly
The cylinder data should be retaken if any anomalies are present to ensure that equip-
ment malfunctions are not the reason for the anomalies.
After the cylinder pressure data has been recorded for each cylinder and the operator is
satisfied that the data is accurate, then the cylinder pressure transducer calibration
should be checked to determine if any drift in the transducer has occurred. The cylinder
pressure transducer should not drift more than 15 psi from the beginning of the
5-20
balancing procedure to the end of the balancing procedure. This amount of drift repre-
sents about one percent of typical full load peak firing pressures. This value is only a
rule of thumb, but it should be recognized that it still could represent as much as 10% of
the allowable variation between all cylinders.
5.9 Recording the Data
As soon as the cylinder pressure data for all of the cylinders has been recorded and the
data is determined to be satisfactory for analysis (cylinder pressure transducer calibra-
tion checks, etc.), then the remaining balance of plant data and cylinder exhaust gas
temperatures should be recorded. The difference between the start of test and end of
test balance of plant data should be calculated and checked against acceptable limits.
The purpose of this data is to determine if any engine parameters have changed during
the data collection process, including load, which could adversely skew the results of
the test.
The data that is derived from the P-V and P- curves, such as P
max
and
max
, should be
recorded. The derived
max
should be corrected for the delay introduced by the indicator
passage and the indicator valve extension, unless the correction was made in the engine
analyzer during the data acquisition process.
6-1
6
ENGINE BALANCING AND TUNING
Ideally, engine balancing is the process of tuning an engine so that all of its cylinders
produce the same amount of power for a given EDG load. An imbalanced engine will
allow:
One or more cylinders to hog the load
Excessive vibration and crankshaft stresses
Accelerated wear and degradation
Increase the possibility that engine design limits will be exceeded
A balanced engine maintains:
Equal load levels in each cylinder
Minimizes engine vibration and wear
Reduces stresses on internal engine components
Reduces exhaust gas emissions
Much of the data recorded for engine balancing when engine load is constant for all
data taken is also used for combustion analysis. This section covers engine balancing
only. Section 7 of this guide addresses when to shift from engine balancing to engine
troubleshooting and diagnostics.
6.1 Purposes of Engine Balancing
The two main purposes of engine balancing are:
1. Tune the engine to some nominal state specified by the engine vendor for which the
vendor has considered such parameters as performance and fuel oil consumption
2. Balance the average power within each cylinder so as to minimize the engine vibra-
tion and dynamic stresses on the engine components
To accommodate both purposes, the recommended analysis guidelines for parameters
used to balance the engine normally consist of two parts:
1. The magnitude of each parameter to ensure the nominal state of tune is achieved
2. A span or range about the engine average value to ensure the balance (or variation
between cylinders) is satisfactory
6-2
In addition, some cycle-to-cycle variation in peak firing pressures and associated com-
bustion parameters is normal and expected in diesel engines due to random variations
in the combustion process [7,8]. The amount of this cycle-to-cycle variation might also
be indicative of the condition and state-of-tune of the engine. Therefore, balancing
criteria for some parameters should include limits on cycle-to-cycle variation in com-
bustion parameters for a given cylinder.
6.2 What is a Balanced Engine?
In a perfectly balanced engine, each cylinder would produce identical pressure versus
crank angle curves and, therefore, would produce the same amount of power. How-
ever, our inability to precisely control the amount of fuel injected and the timing of the
injection for each cylinder prevents perfect balancing. An attainable goal, however, is to
ensure that the pressure curves for all cylinders have the same characteristics and the
power per cylinder is nearly equal.
The characteristics of the pressure curves for each cycle in each cylinder are determined
by the combustion process in the cylinder. A deviation from the normal or expected
pressure curve indicates a problem in the combustion process, in the mechanical com-
ponents of the engine, or in the collection of the data. Identifying and diagnosing com-
bustion anomalies falls within the scope of combustion analysis and are discussed in
Section 7.
In many cases, combustion anomalies are small and indicative of normal engine degra-
dation. Engine balance can sometimes be considered satisfactory with this type of
degradation. When engine balance cannot be achieved, combustion analysis is used to
identify, diagnose, and direct corrective actions.
6.2.1 Balancing Criteria
Indicated cylinder power can be calculated from the P- data recorded during engine
combustion analysis. However, as discussed in Section 5, cylinder power calculations
are very sensitive to small phasing errors between cylinder pressure and crank angle.
Even a perfectly balanced engine can exhibit large cylinder-to-cylinder variations in
calculated power if small phasing errors are present.
As a result, although equalization of cylinder power is the primary objective of balanc-
ing, itself should be a secondary measurement. Instead, measured data that is indicative
of power should be used for balancing and combustion analysis to confirm that the
characteristics of the cylinder pressure curves are within specifications. If the mechani-
cal condition of two cylinders are approximately equal and the same amount of fuel is
injected into each cylinder at the same time in the mechanical cycle, the resulting P-
curves will have the same shape. As a result, if the timing and magnitude of the peak
firing pressure of two cylinders can be equalized, the power produced by each cylinder
will be approximately equal. The cylinder peak firing pressure (P
max
) and the peak firing
6-3
pressure angle (
Pmax
) represent the coordinates of a single point on the P- curve. These
are the two key variables that should be used to determine engine balance.
A range on the engine average values for both parameters should be imposed to ensure
engine structural and performance limits are satisfied. Cylinder power measurements
(if they are available) should be used as a secondary confirmation of cylinder balance.
However, it should be recognized that the margin of error for cylinder power calculated
from P- data is much larger than the margin of error for the P- data itself.
Different sources recommend the use of different variables for engine balancing. Cylin-
der peak firing pressure, fuel rack position, fuel injection pump timing, exhaust gas
temperatures, cylinder peak firing pressure angle, indicated horsepower, and IMEP
have all been mentioned as parameters that should be equalized from cylinder to cylin-
der in order to balance an engine. The benefits and drawbacks of the various balancing
methods are discussed in Table 6-1.
6-4
Table 6-1
Advantages and Disadvantages of Various Balancing Methods
Balance Parameter Advantages Disadvantages
Cylinder peak firing
pressure
Avoids problems associated with
phasing crank angle with piston
position. Easy, minimal
equipment required.
Most likely will not result in a well-
balanced engine if peak firing pressure
angles or mechanical conditions vary
from cylinder to cylinder.
Fuel rack position Does not require special
instrumentation and can be
accomplished quickly.
balanced engine due to variations in
fuel injection pump timing. Also, this is
an open-loop balancing because only
an input to the combustion process is
controlled.
Fuel injection pump
timing
Does not require special
instrumentation and can be
accomplished quickly.
balanced engine due to variations in the
fuel rack. Also, this is an open-loop
balancing because only an input to the
combustion process is controlled.
Exhaust gas
temperatures
Does not require special
instrumentation.
Temperature is not a good indicator of
cylinder power. Actually, it leads to poor
power balance. Because the amount of
fuel delivered to each cylinder is
different, the amount of power produced
in each cylinder will be different. In
addition, exhaust manifold effects
(standing waves, cylinder interactions,
and cylinder location) can introduce
significant exhaust temperature
variations in some engines even though
cylinder power balance might be
relatively good.
Cylinder peak firing
pressure angles
When used in combination with
peak firing pressures, results in
proper timing and value of peak
of P- curve. If curves are similar
in shape, it will result in good
balance.
Requires special instrumentation,
training, and time.
Indicated
horsepower or
indicated mean
effective pressure
Equalizing indicated horsepower
from all cylinders results in a
balanced engine.
Difficult to equalize indicated
horsepower across many cylinders.
Small TDC phasing errors can lead to
large errors in indicated horsepower.
Balancing just peak firing pressures will not necessarily ensure a well-balanced engine.
The timing of the peak is just as important as the value of the peak. If the peak occurs
too early, the compression stroke work transfer from the piston to the cylinder gases
increases because cylinder pressure is higher during the compression stroke. This re-
sults in reduced engine efficiency. If the peak is delayed, it occurs later in the expansion
stroke and is reduced in magnitude (due to the increased cylinder volume). This results
in reduced expansion stroke work transfer from the cylinder gases to the piston, also
reducing engine efficiency. The optimum peak firing pressure occurs when these two
opposing trends just offset each other. Maximum engine efficiency occurs when the
peak firing pressure angle is about 15 ATDC [5,38].
6-5
Perfect balance is unlikely to be attainable on any engine. Small differences in engine
component dimensions and conditions will result in finite limits on engine balance.
6.3 Engine Analysis Guidelines
In order to evaluate the state-of-tune of an engine, engine analysis guidelines are
needed. The parameters for which numerical analysis guidelines are appropriate were
discussed in Section 4. Here, the methodology for developing the appropriate values for
the guidelines is developed and discussed. Each of the parameters for which numerical
guidelines should be developed are discussed, too.
6.3.1 Maximum Engine Average Peak Firing Pressure and Maximum Cylinder
Peak Firing Pressure
Analysis guidelines for maximum engine average and maximum cylinder pressures are
usually based on the structural limits of engine components such as cylinder heads,
cylinder head studs, connecting rods, and connecting rod bearings. Most engine ven-
dors have published operating limits for either maximum engine average peak firing
pressure or maximum cylinder peak firing pressure. If available, these limits should be
used as the analysis guidelines. A guideline for only one of these two parameters is
sufficient, because if a limit for engine average peak firing pressure exists, the maxi-
mum cylinder firing pressure is controlled by the limit on the maximum range of peak
firing pressures.
6.3.2 Maximum Cylinder-to-Cylinder Variation in Peak Firing Pressures
Published guidelines on the maximum cylinder-to-cylinder variation of peak firing
pressures are rare. Where they exist, they are typically based on limiting engine vibra-
tion and crankshaft stresses. In addition to the normal, random cycle-to-cycle combus-
tion variation, the tolerance should include allowances for:
Instrument error
Variations in engine load during data collection
Variations in fuel rack settings between cylinders
Differences in swept volumes between cylinder banks (for engines with articulated
master-link connecting rods)
For operators of commercial engines, a total allowance (maximum to minimum) of 5%
is typical [9]. For engines in nuclear standby service, a total allowance of 10% is consid-
ered acceptable.
6.3.3 Variation of Cylinder Peak Firing Pressure Over a Number of Cycles
Published guidelines on the variations of peak firing pressure in a single cylinder over a
number of cycles are also rare. Because the mixing of fuel and air in the cylinder prior to
6-6
and during combustion and combustion itself are random processes, combustion and
parameters that are measured during engine analysis vary from cycle to cycle on a
random basis [7]. Large variations from cycle to cycle are an indication of poor combus-
tion performance (detonation or misfire) or engine component problems (inconsistent
fuel injection equipment performance, blowby, and so on). Limits on cycle-to-cycle
variation should be expressed in terms of the standard deviation of the cylinder peak
firing pressure over a number of cycles. The cycle-to-cycle variability of peak firing
pressure typically has a standard deviation on the order of 1%2% over 50 cycles [8].
6.3.4 Engine Average Corrected Peak Firing Pressure Angle
Engine specific guidelines on the engine average corrected peak firing pressure angle
are also rare. However, maximum IMEP and, therefore, maximum mechanical effi-
ciency will occur with corrected peak firing pressure at about 15 ATDC [5,38]. In this
case, piston motion during the rapid combustion phase will be small and will not
greatly effect the cylinder pressure. Peak firing pressure angle is strongly dependent on
fuel injection timing. Early injection results in long ignition delays due to the lower
cylinder temperature and pressure when injection begins. The inward piston motion
and rapid combustion combine to give high rates of pressure rise.
If injection timing is advanced far enough, the peak firing pressure can occur prior to
TDC leading to reduced mechanical efficiency. If the delay is long enough, a large
quantity of fuel can burn during the rapid combustion phase, leading to an excessive
rate of pressure rise and high peak firing pressure with resulting high stresses and
vibration. This is known as detonation or knock. Similarly, late injection timing also
results in long ignition delay and high rates of pressure rise. However, in this case,
piston motion is increasing the cylinder volume, leading to greatly reduced peak firing
pressures and increased peak firing pressure angles.
Some of the important considerations when establishing guidelines for the engine
average peak firing pressure angle include:
The angle should be as close as practicable to the angle that gives the highest IMEP.
This will result in the most efficient engine operation. However, with increasing
emphasis being placed on engine emissions, the desired angle might need to be
retarded to reduce undesirable emissions [40,41].
The angle must be large enough to avoid the possibility of detonation, which could
damage the engine.
The guideline should also allow sufficient margin for drift such that detonation will not
occur. It should be noted that there will be one corrected peak firing pressure angle that
will be best for mechanical efficiency and fuel economy and another that will be best for
exhaust emissions. The actual angle chosen will depend on the situation at each nuclear
power station.
6-7
6.3.5 Maximum Range of Peak Firing Pressure Angles Across All Cylinders
Guidelines on the range of peak firing pressure angles across all cylinders should be
strict, with a range of no greater than 24. This is to ensure that proper power balance
can be achieved.
6.3.6 Engine Average Fuel Injection Timing
Fuel injection pump timing is set to default values by engine manufacturers through the
use of timing marks on the engine flywheel. The actual timing varies from model-to-
model and from engine to engine in order to meet peak firing pressure limits and load
requirements. It is often not published, but is mechanically set through the design of
engine components. The pump timing is set with the engine shut down using, for ex-
ample, timing windows (ALCO 251) or fuel injector timing gauges (EMD 645). How-
ever, this timing is static fuel injection timing.
The actual dynamic fuel injection timing, also known as start of injection (SOI), can vary
from the static timing by as much as several degrees due to camshaft windup and
variations in fuel injection nozzle pop pressure. Camshaft windup is the twist imparted
to the camshaft because the camshaft is driven from one end, but has loads along its
entire length. Camshaft windup results in retarded fuel injection timing at the end of
the engine opposite the camshaft drive. As a result, equalizing the fuel injection timing
for all cylinders will not necessarily result in the same start of injection for all cylinders.
For this reason, it is recommended that
Pmax
be used for the timing characterization of
engine balance.
6.3.7 Maximum Range of Fuel Injection Timing Across All Cylinders
The design range of fuel injection timing across all cylinders is zero. However, due to
tolerance in engine parts and wear, it is not possible to set the fuel injection timing
exactly the same for all cylinders. In addition, the effects of camshaft windup must be
accounted for. As a result, some differences between cylinders will exist. However, the
differences should be kept small enough that they do not mask other problems. A range
of 4 around the engine average fuel injection timing is considered appropriate.
6.3.8 Maximum Range of Fuel Rack Settings Across All Cylinders at Full Load
Some engine vendors have published limits for fuel rack balance. These limits are typi-
cally quite tight and are in the order of 1%2% of full load fuel rack position. Because
imbalances in the quantity of fuel injected into the engine lead directly to power imbal-
ances, fuel rack balance should be maintained within the specified tolerances.
6.3.9 Allowable Range of Fuel Injection Nozzle Pop Pressures for New and Used
Fuel Injection Nozzles
Most engine vendors have established limits for fuel injection nozzle pop pressures.
Because the pop pressure will have an effect on fuel injection timing and duration, it is
6-8
prudent to control nozzle pop pressures within the vendor recommended bands to
prevent undesirable variations in fuel injection timing or fuel delivery. Adjustments to
nozzle pop pressure are relatively easy to make when the nozzles are removed for
testing.
6.3.10 Maximum Cylinder Exhaust Gas Temperature
Limits for maximum cylinder exhaust gas temperatures have been published by most
engine vendors. These limits are typically based on the melting temperatures of engine
components such as pistons, valves, exhaust manifolds, and turbochargers. If the engine
vendor has published an exhaust temperature limit, it should be used as the engine
analysis guideline. If a limit is not available, the analysis guidelines should be estab-
lished at a value below the melting temperature of engine components. The guidelines
should include an allowance for instrument error.
6.3.11 Maximum Range of Exhaust Gas Temperatures Across All Cylinders
Most engine vendors have published limits for the maximum range of exhaust gas
temperatures across all cylinders. With most engines, if all of the cylinders are doing the
same amount of work, the cylinder exhaust temperatures should be closely matched. As
a result, prior to the general availability of engine analysis equipment, exhaust gas
temperatures were used to check engine power balance. While this method is accept-
able for some enginesfor engines that do not exhibit a strong relationship between
load and exhaust gas temperaturesit is less than adequate. An engine that has been
power-balanced using engine analysis could have large cylinder to cylinder exhaust
temperature variation due to standing waves in the exhaust manifold, cylinder interac-
tions through the exhaust piping, or other effects.
Parameters for which numerical limits are not appropriate and that, instead, should be
trended against past results for the same cylinder and against other cylinders to provide
indications of engine condition:
Combustion start angle
Expansion reference pressure
Terminal pressure
Compression reference pressure
Cylinder vibration signature
Cylinder ultrasonic signature
6.4 Factors Affecting Balance
In addition to factors that operators and mechanics can adjust to affect engine balance,
such as fuel rack position, fuel injection pump timing, and fuel injection nozzle pop
6-9
pressure, there are a number of factors outside of their control that can affect engine
balance or engine performance.
6.4.1 Mechanical Variations or Manufacturing Tolerances of Engine Parts
At least one station has experienced difficulty in balancing an engine due to differences
in valve pushrod lengths from one cylinder to another. In addition, differences in fly-
wheel timing mark locations, connecting rod lengths, cam lobe profiles, fuel injection
pump calibrations can all affect engine analysis results.
6.4.2 Fuel Cetane Number
If engine analysis runs are conducted with fuels with different cetane numbers, the
results will be affected. Higher cetane numbers will cause the fuel to ignite more
quickly in the cylinder, resulting in shorter ignition delays and increasing the possibility
of detonation. Because the fuel storage tanks at nuclear power stations are large and
fuel turnover is slow, this will likely only be a factor when the entire fuel inventory is
replaced. However, it is noted that most nuclear power stations only have requirements
for minimum cetane number in their Technical Specifications.
6.4.3 Engine Load
As discussed in Section 5, changes in engine load during a test or from test to test will
change engine analysis and balance results. In order to ensure adequate engine balanc-
ing and engine analysis data trending, engine analysis should be performed at the same
engine load each time. One way to minimize load changes during engine analysis is to
lock the governor.
6.4.4 Fuel Rack Giggle
Fuel rack giggle will affect the amount of fuel delivered to the cylinders.
6.4.5 Variations in Ambient Conditions
As discussed in Section 5, changes in ambient conditions from test to test will affect
engine analysis results.
6.5 Adjustments to the Fuel Injection System
6.5.1 Fuel Injection Pump Timing
Both fuel injection timing and the quantity of fuel injected into the cylinder can be
adjusted to tune the fuel injection system and balance the engine. Fuel injection timing
determines the dynamic fuel injection and, therefore, determines both the peak firing
pressure angle (assuming normal combustion occurs) and the peak firing pressure.
6-10
Advancing the fuel injection timing shifts P
max
towards top dead center. The rapidly
expanding products of combustion expand into a relatively smaller volume, increasing
P
max
. Retarding fuel injection has the opposite effect. On most engines, fuel injection
timing can be adjusted for each cylinder individually.
Fuel injection pump timing is fixed by the mechanical configuration and adjustments to
the camshaft and fuel injection system. However, SOI might deteriorate due to degrada-
tion in the fuel injection pump or nozzle tip holder. In particular, SOI is advanced with
decreasing nozzle holder pop pressures and delayed by increasing pop pressures.
Changes in SOI will have a direct impact on P
max
.
Pmax
is strongly related to fuel injection timing and fuel rack position. Changes in
Pmax
correlate directly to changes in the nominal fuel injection timing. Therefore, when
adjustments to
Pmax
are required, and the fuel racks are balanced (precluding further
adjustments), they are affected by changing the fuel injection timing. However,
Pmax
could degrade independently of fuel injection timing if SOI is affected by other faulty
fuel injection system components.
Pmax
must not be too close to TDC or else excessive peak firing pressures and combus-
tion instability will occur.
Note also that on some engines, adjustments to the individual fuel rack positions also
adjusts the fuel injection timing at less than full load due to the shape of the helix on the
fuel injection pump plunger. While fuel rack adjustments are normally made for fuel
delivery purposes, the effect on timing should also be considered.
6.5.2 Fuel Delivery
The fuel rack position on each fuel injection pump controls the volume of fuel delivered
to each cylinder. In addition, at low- and mid-level loads on some engines, the fuel rack
position also effects the fuel injection timing. In any case, equal amounts of fuel must be
injected into each cylinder to balance the power produced. As a result, relatively tight
tolerances are placed on the cylinder-to-cylinder fuel rack variation.
6.5.3 Inlet Valve Timing Adjustments
For engines with inlet values, proper inlet valve timing directly impacts the effective
cylinder compression ratio. Some engine operators have noticed improvements in
engine balance after correcting inlet valve timing. It is strongly recommended that the
inlet valve timing be checked and corrected (if necessary) for all cylinders prior to
implementing a rigorous engine analysis program.
6-11
6.6 Balancing Methods
6.6.1 Operating Factors Affecting Balance
In addition to the variables listed in Section 6.4, several other factors affect engine bal-
ance. However, it is not possible to completely compensate for these effects. Engine
imbalance can be attributed to:
Inlet and exhaust manifold standing waves
Cylinder interaction from the exhaust manifold
Flow losses
Mechanical variations in engine components
Fuel pump calibration errors
Valve clearances
Gas leakage past piston rings and ports/ valves
In addition, changes in atmospheric conditions (temperature, pressure, and humidity)
during tests or between tests will result in variations in the recorded data. However,
there is no fixed threshold at which the extent of imbalance prevents an engine from
operating and producing power, although engine performance will degrade with in-
creasing imbalance. As a result, this guide recommends three types of engine balancing:
1. Static balancing (set-up) of the engineThis method is used to establish the nominal
settings of fuel injection pump timing, fuel rack settings, and fuel injector nozzle
pop. This method should be used when the engine becomes so far out of adjustment
that the operator determines that it is easier to start over, or when a complete engine
overhaul is accomplished and the original balanced engine fuel injection timing
cannot be replicated. Calibrated fuel injection pumps are recommended when set-
ting up an engine using this method. Static balancing is mechanical only. No com-
bustion analysis is conducted during static balancing.
Because the static setup does not account for such factors as minor piston ring or
port/ valve leakage, camshaft windup, or other small mechanical differences be-
tween cylinders, static set-up of an engine might not result in a balanced engine
when the engine is run. Static balancing is usually only required once for the entire
engine, but can be used whenever questions arise regarding the settings on the
engine.
2. Dynamic balancing of the key engine analysis parameters (P
max
and
Pmax
)This
method recognizes the variations in cylinder mechanical components, such as leak-
age past ports/ valves and rings, fuel injection pump calibration differences, cam-
shaft windup, and other minor component variation or degradation. It is used to
balance the engine quickly so that performance will be satisfactory for one nuclear
power plant fuel cycle. This method could involve adjustments to the fuel rack, fuel
6-12
injection pump timing, or nozzle replacement. In essence, an imbalance of fuel is
used to compensate for normal engine degradation and to obtain a balanced engine.
This method might not be applicable to all engines.
3. Fine-tuning to optimize engine performanceFine-tuning ensures that the engine is
precisely balanced, all cylinders receive equal fuel, and all balancing performance
parameters are within like-new specifications. Fine-tuning should be used during
each fuel cycle, if possible. An engine optimized using this method should accom-
modate normal degradation of the fuel injection and other systems as well as com-
pensate for camshaft windup. This balancing method is recommended following
major engine overhauls, or after one nuclear power plant fuel cycle of operation
without fine-tuning.
Using these three methods, the operator will have consistent methods to set up the
engine following extensive maintenance in an initial state that minimizes further bal-
ancing, to quickly balance the engine when time is limited, and to optimize perfor-
mance when the time is available.
Balancing or tuning the engine so that it just meets the established analysis guidelines
leaves no margin for performance parameter drift over time. Therefore, it is recom-
mended that the engine be initially balanced or tuned to a state better than the analysis
guidelines to account for drift that will occur over time. Recommended balancing or
tuning goals should be established for dynamic balancing and fine-tuning, which have
tighter allowance bands than the analysis guidelines.
7-1
7
USING ENGINE ANALYSIS TO DETERMINE AND
TROUBLESHOOT ENGINE PERFORMANCE AND
CONDITION
7.1 Integrated Maintenance Program
The goals of an engine maintenance program are to maximize the availability and the
reliability of the engine. As discussed in Section 2, an integrated maintenance program
uses elements of preventive-, condition-, and performance-based maintenance pro-
grams to achieve these goals. A number of techniques for evaluating engine condition
and performance have been developed and used by operators of diesel engines. Among
these techniques are:
Operating log review
Periodic engine testing
Non-intrusive visual inspections by borescopes, videoprobes, and thermography
Visual inspections of disassembled parts
Exhaust gas analysis
Combustion analysis
Vibration and ultrasonic signal analysis
Lube oil, fuel oil, and cooling water chemical analysis
Most nuclear power stations already use several of these techniques in evaluating the
condition and performance of nuclear standby EDGs. For example, all stations are required
to perform periodic surveillance testing of EDGs, and most stations sample and analyze
lubricating oil on a periodic basis. As discussed in Section 2, many of these techniques are
most effective when used in combination with one or more of the others. Frequently, one
technique will provide an indication of a problem, but will not provide enough informa-
tion to isolate the problem to a single component. In these cases, other techniques will
frequently permit positive identification of the source causing the problem.
7.1.1 Identifying Potential Problems
The first step to solving any problem is to identify that a problem exists and what it is.
However, this is not always as straightforward as it might seem to be. In some cases, the
problem could cause only subtle changes in performance. In others, prior maintenance
actions might have been taken to correct the problem, but have actually masked it until
7-2
a more serious condition becomes evident. For example, at one station a cracked cylin-
der head on an EDG allowed cooling water to leak into the combustion chamber, caus-
ing low firing pressures. Contract personnel performing maintenance work on the
engine increased the fuel rack on that cylinder to increase the firing pressures to within
the normal range. However, that cylinder remained low relative to the remaining cylin-
ders. The cracked head was not found until lube oil samples showed increased amounts
of chemicals used for jacket water system corrosion control.
Had the cause of the low firing pressures been investigated when it was first noted, a
great deal of effort to find the source of the leak could have been avoided. Each engine
problem will cause its own set of symptoms. It is up to the engine analyst to diagnose
the problem based on the symptoms.
Typically, one of the techniques discussed in Section 7.1 will first provide indications of
an engine fault. For the head crack example discussed above, combustion analysis
would have shown low peak firing pressures for the cylinder, lube oil analysis would
have showed contamination by jacket water chemicals, or engine logs would have
showed an increased fuel rack position for the cylinder and decreasing water level in
the jacket water system expansion tank. Which technique shows the first indication is
often only a matter of timing.
7.1.2 Identifying and Characterizing the Failure
If a failure cannot be identified from the initial indication, the next step is to attempt to
identify and characterize the degraded or failed component using one or more of the
additional inspection techniques. In order to accomplish this, it is best to have a plan of
attack worked out based on all the possible failures that could cause the symptom or
symptoms discovered. Additional tests or inspections should be performed to confirm
or eliminate suspected causes of the problem. The most likely causes of the problem
should be investigated first. It is helpful to have a table or flow chart of typical or ex-
pected symptoms for the each suspected failure being investigated.
Most engine vendor technical manuals include some form of troubleshooting chart.
However, they tend to be rather qualitative approaches to problems, and not based on
actual measured data that might be available to the engine analyst. The table or flow
chart generated to solve a specific problem should identify how the problem will mani-
fest itself; that is, which inspection techniques will identify the problem and how the
results of each inspection will be effected.
Table 7-1 depicts a table for combustion, vibration, and ultrasonic analyses. The left
most column contains the observed potential symptoms and some of the potential
problems that could cause it. The other columns contain other engine analysis param-
eters that could be affected by each of the problems. For example, if the peak firing
pressure for a cylinder is low, potential causes include:
Improper fuel rack setting
7-3
Retarded fuel injection timing
Late inlet valve closure
Valve or port leakage
Piston ring blowby
High fuel injection nozzle pop pressure
Bent connecting rod
Decreased ignition delay (caused by fuel property changes for example)
Water leakage into the cylinder
7-4
T
a
b
l
e

7
-
1
E
x
a
m
p
l
e

o
f

S
i
m
p
l
i
f
i
e
d

E
n
g
i
n
e

T
r
o
u
b
l
e
s
h
o
o
t
i
n
g

C
h
a
r
t
C
o
m
b
u
s
t
i
o
n

A
n
a
l
y
s
i
s
V
i
b
r
a
t
i
o
n

A
n
a
l
y
s
i
s
U
l
t
r
a
s
o
n
i
c

A
n
a
l
y
s
i
s
O
t
h
e
r
A
n
a
l
y
s
i
s
S
y
m
p
t
o
m
/
C
a
u
s
e
P
e
a
k
F
i
r
i
n
g
P
r
e
s
s
u
r
e
P
e
a
k

F
i
r
i
n
g
P
r
e
s
s
u
r
e
A
n
g
le
S
t
a
r
t

o
f
C
o
m
b
u
s
t
i
o
n
E
x
p
a
n
s
i
o
n
R
e
f
e
r
e
n
c
e
P
r
e
s
s
u
r
e
T
e
r
m
i
n
a
l
P
r
e
s
s
u
r
e
C
o
m
p
r
e
s
s
i
o
n
R
e
f
e
r
e
n
c
e
P
r
e
s
s
u
r
e
S
t
a
r
t

o
f
I
n
j
e
c
t
i
o
n
I
M
E
P
C
o
m
p
r
e
s
s
i
o
n
P
r
e
s
s
u
r
e
M
a
x
i
m
u
m
P
r
e
s
s
u
r
e
R
i
s
e

R
a
t
e
F
u
e
l
R
a
c
k
P
o
s
i
t
i
o
n
N
o
z
z
l
e
P
o
p
P
r
e
s
s
u
r
e
T
i
m
e
D
o
m
a
in
F
r
e
q
u
e
n
c
y
D
o
m
a
in
T
i
m
e
D
o
m
a
in
F
r
e
q
u
e
n
c
y
D
o
m
a
in
L
O
W

P
E
A
K

F
I
R
I
N
G

P
R
E
S
S
U
R
E
L
o
w

F
u
e
l

R
a
c
k
S
e
t
t
i
n
g
L
o
w
L
o
w
L
o
w
L
o
w
R
e
t
a
r
d
e
d

F
u
e
l
I
n
j
e
c
t
i
o
n

T
i
m
i
n
g
L
o
w
R
e
t
a
r
d
e
d
R
e
t
a
r
d
e
d
R
e
t
a
r
d
e
d
L
o
w
L
o
w
X
L
a
t
e

I
n
l
e
t

V
a
l
v
e
T
i
m
i
n
g
L
o
w
L
o
w
L
o
w
L
o
w
X
V
a
l
v
e

L
e
a
k
a
g
e
L
o
w
L
o
w
L
o
w
L
o
w
X
X
P
i
s
t
o
n

R
i
n
g
B
l
o
w
b
y
L
o
w
L
o
w
L
o
w
L
o
w
L
o
w
L
o
w
X
X
H
i
g
h

F
u
e
l

I
n
j
e
c
t
i
o
n
N
o
z
z
l
e

P
o
p
P
r
e
s
s
u
r
e
L
o
w
R
e
t
a
r
d
e
d
H
ig
h
B
e
n
t

C
o
n
n
e
c
t
i
n
g
R
o
d
L
o
w
L
o
w
L
o
w
L
o
w
L
o
w
D
e
c
r
e
a
s
e
d

I
g
n
i
t
i
o
n
D
e
la
y
L
o
w
A
d
v
a
n
c
e
d
A
d
v
a
n
c
e
d
L
o
w
F
u
e
l
A
n
a
l
y
s
i
s
W
a
t
e
r

L
e
a
k
a
g
e
L
o
w
7-5
For each of these potential problems, the effects on other engine analysis parameters are
identified in Table 7-1. For example, if blowby is the cause, the expansion and compres-
sion reference pressures and peak compression pressure would be expect to be low and
indications of blowby should be seen on the vibration and ultrasonic signal traces for
the cylinder. In addition, the maximum pressure rise rate and IMEP for the effected
cylinder might also be lower than the other cylinders. If retarded fuel injection is the
cause, the peak firing pressure angle and start of combustion will be delayed, the indi-
cated mean effective pressure and maximum pressure rise rate will be reduced, and the
late nozzle opening event would be visible in the ultrasonic signal trace for the cylinder.
It is highly recommended that each station or EDG owners group develop troubleshoot-
ing tables or flow charts for their particular EDGs, or, at the very least, keep records of
failures that are found.
Detection of engine problems using engine combustion, vibration, and ultrasonic analy-
ses is a skill that takes time to develop. For the first several engine analysis runs, it is
best to place the emphasis on identifying major anomalies such as:
Detonating or misfiring cylinders
Cylinder power imbalance
Cylinder liner scoring or scuffing
Blowby
Leaking valves or ports
In some cases, engine faults will not be detected prior to the failure of a component. In
these cases, it is important to review engine analysis data from the prior test runs to
determine if any signs of the impending failure were present but not detected.
Finally, the use of engine analysis as part of an integrated maintenance includes several
key elements:
1. An understanding of the difference between symptoms and causes of failures is
required. It is beneficial to construct a flow chart or table of typical or historical
symptoms for the specific equipment that you are operating. This effort can be
rather large and can benefit tremendously from consideration of the experience of
other similar equipment or identical equipment installed at other locations. A coop-
erative effort is this area by EDG owners group would be beneficial.
2. A database or record of prior failures and their symptoms is required to provide a
guide during the diagnosis of symptoms.
3. The operator must understand the results of any tests, trending, inspections, analy-
sis, etc. before they are accomplished. This is perhaps the most important point. This
point cannot be over-emphasized. Performing numerous tests and inspections,
recording and trending data, etc. without some understanding of the meaning of the
data and the appropriate actions to take in the event of any deviations is of little
7-6
value. Understanding the tests, inspections, and analysis is important to accomplish-
ing such procedures successfully. In addition, without this insight, the operator
might inadvertently question good results and schedule unnecessary maintenance,
or could miss indications of impending failure.
8-1
8
CASE HISTORIES
This section includes sample engine analysis data for many of the diesel engines in
nuclear standby service, in addition to case histories of typical problems and failures.
The case histories include:
Power cylinder imbalance
Misfire
Piston slap
Piston ring blowby or leaking valves/ ports
Unequal valve lash setting
Excessive exhaust valve lash
Faulty fuel injection
Cylinder liner scoring or scuffing
Port bridge problem
Improved valve seating
Bent connecting
Burned exhaust valve
This section also includes plots of typical cylinder pressure and vibration traces for the
following engines:
ALCO model 251
Cooper-Bessemer model KSV
EMD model 645
Cooper-Enterprise model DSRV
Fairbanks Morse Opposed Piston model 38TD 8-1/ 8
Colt-Pielstick model PC2
8-2
8.1 Power Cylinder Imbalance
The plot (see Figure 8-1) shows the P- curves for all cylinders of a 20-cylinder engine
superimposed. In this engine, the spread in peak firing pressures is 10%, which is re-
garded as the maximum acceptable spread.
2000
1501
1002
504
5
0 90 180 270 720 630 540 450 360
Speed
AVE 601
HIGH 602
LOW 600
P1L P6L P9L P8L P7L 10L P5L P2L P3L
P4L
10R P5R P2R P3R P4R P1R P6R P9R P8R P7R
Angle (degrees)
p
s
i
Figure 8-1
Power Cylinder Imbalance
8-3
8.2 Misfire
A misfiring cylinder would be indicated as shown above on the engine analyzers manu-
factured by Liberty Technologies/ Beta. The vertical portion of the icon next to the peak
of each pressure curve represents the minimum and maximum peak firing pressures
which occurred in each cylinder over a number of cycles (see Figure 8-2). The lower end
of the vertical bar for cylinder 5R, which extends to the compression pressure, indicates
that the cylinder had at least one complete misfire during the cycles analyzed.
2000
1500
1000
500
0
0 90
180 270 360 450
540 630 720
p
s
i
Angle (degrees)
Speed
AVE 600
HIGH 601
LOW 599
P5R P3R P1R P9R P7R
P10R P2R P4R
P6R P8R
Figure 8-2
Misfire
8-4
8.3 Piston Slap
Figure 8-3 shows the vibration traces for several cylinders of the same engine. The
traces are arranged so that they are all in phase in this plot. Piston slap shows up as a
sharp mechanical impact occurring at about the angle of peak firing pressure as shown
for cylinder 2L (second trace from the top on the left side).
2 Stroke Full Scale: 4.00
Ref: TDC
ME/9
P1L
VT4
247
ME/9
P2L
VT4
246
ME/9
P3L
VT4
245
ME/9
P4L
VT4
246
ME/9
P5L
VT4
248
ME/9
P1R
VT4
247
ME/9
P2R
VT4
247
ME/9
P3R
VT4
247
ME/9
P4R
VT4
245
ME/9
P5R
VT4
246
Figure 8-3
Piston Slap
8-5
8.4 Blowby/Valve Leak
Leakage past rings or valves will be detected as a gas flow signature during the com-
pression process before the beginning of fuel injection. In Figure 8-4, such a pattern can
be seen beginning prior to the -36 position. Of course, the leakage continues during the
injection and combustion processes, but the signal is usually obscured by other energy.
If significant leakage is occurring, the compression reference pressure for the cylinder
will be low.
Intake
Exhaust
RPM = 449.8
HP = 81.5
PFP Angle = 14.0
Ign. Adv P9L = 24.0
Ign. Adv P9R = 23.0
16.00g
P9 VT4
591
512
433
355
276
197
-178 -107 -36 36 107 178
p
s
i
Fuel
Figure 8-4
Piston Ring Blowby/ Valve Leak
8-6
8.5 Unequal Lash Settings
In engines with two intake and exhaust valves, a pattern similar to that shown in Figure
8-5 will often be seen, where the two intake or exhaust valves (exhaust valves are
shown in the middle of the figure) are closing at slightly different times. This is due to
unequal lash settings or unequal wear in components. Some difference is permissible,
but that shown is probably unacceptable.
4 Stroke Full Scale: 6.00 Ref: TDC
ME/5
P7L
VT4
600
Figure 8-5
Unequal Valve Lash Settings
8-7
8.6 Excessive Exhaust Valve Lash
Excessive valve lash causes valve dynamics to deviate from design. The most observ-
able effect is that the impact amplitude at closure starts to increase, and the closure
becomes earlier (see cylinder 10L in Figure 8-6). Eventually, an impact would also be
seen at the valve opening and the valve opening will occur late.
Ref: TDC
ME/7
P1L
VT4
602
ME/7
P2L
VT4
600
ME/7
P3L
VT4
602
ME/7
P4L
VT4
600
ME/7
P5L
VT4
600
ME/7
P6L
VT4
600
ME/7
P7L
VT4
600
ME/7
P8L
VT4
600
ME/7
P9L
VT4
602
ME/7
10L
VT4
602
Figure 8-6
Excessive Exhaust Valve Lash
Possible causes include:
A collapsed lifter
A bent pushrod
A worn rocker arm bushing
A valve lash misadjustment
8-8
8.7 Faulty Fuel Injection
Table 8-1 shows the combustion analysis results for a six cylinder engine. Note that P6 indi-
cated horsepower is low compared to other cylinders, even though its peak firing pressure is
close to engine average. This would happen if the fuel injection duration was short.
Table 8-1
Combustion Analysis Results for a Six Cylinder Engine
Cylinder Speed
(rpm)
I HP Press
at 20
BTDC
(psi)
Comb.
Start
deg.
ATDC
Max.
Ri se
Rat e
psi / deg.
Peak Firing Pressures Peak
Firing
Pressure
Angl e
ATDC
Di ff.
psi
at
180
ATD
C
Ave
(psi)
Dev
(psi)
Max
(psi)
Mi n
(psi)
Del ta
(psi)
P1 396 629 408 LO 0 LO 42 HI 1276 4 1295 1249 18 16 24
P2 394 616 425 2 HI 33 1202 3 1234 1178 -57 18 23
P3 397 632 419 0 LO 39 1257 4 1280 1236 -2 16 23
P4 398 629 419 0 LO 33 LO 1231 3 1253 1211 -28 16 25
P5 398 666 444 HI 0 LO 39 1332 4 1351 1301 74 16 26 HI
P6 401 597 422 0 LO 35 1253 4 1277 1229 -6 16 22 LO
Eng: 397 3769 423 0 37 1259 4 30 16 24
Sprd: 2% 11% 9% 2 deg. 23% 10% 27% 2 deg. 17%
The vibration patterns reveal the start and end of fuel injection. The closeup of the firing
pressure and vibration curves for a typical cylinder is shown in Figure 8-7. The start and
end of injection can be seen clearly.
Fuel
RPM = 395.1
HP = 615.6
PFP Angle = 16.0
6.00g
P2 VT4
1342
1074
805
537
268
0
-63 -30 3 36 70 103
p
s
i
Figure 8-7
Normal Fuel Injection Vibration Pattern
8-9
In cylinder P6, the fuel cutoff is less obvious and seems earlier than in other cylinders.
Cylinder Injection Begins Injection Ends Duration
(BTDC) (ATDC) (deg)
P1 8.0 20.1 28.1
P2 7.5 21.5 29.0
P3 6.3 20.9 27.2
P4 6.6 20.6 27.2
P5 6.3 22.0 28.3
P6 6.7 18.0 24.7
The table above compares the fuel injection duration for each cylinder in the last col-
umn. Note that the duration for cylinder P6 is three to four degrees shorter than for the
other cylinders. The suspected cause is a leaking fuel pump. Figure 8-8 depicts a faulty
fuel injection vibration pattern.
Fuel
RPM = 399.1
HP = 596.9
PFP Angle = 14.0
6.00g
P6 VT4
1342
1074
805
537
268
0
-63 -30 3 36 70 103
p
s
i
Figure 8-8
Faulty Fuel Injection Vibration Pattern
8-10
8.8 Scored Liner
Comparing all the vibration patterns, the most significant anomalies appear on P2R and
P3R (see Figure 8-9). Because these two cylinders are adjacent to each other, it is important
to check for cross talk. The anomalous components seen on P2R actually are from P3R.
ME/9
P1R
VT4
601
ME/9
P2R
VT4
599
ME/9
P3R
VT4
601
ME/9
P4R
VT4
601
ME/9
P5R
VT4
599
ME/9
P6R
VT4
599
ME/9
P7R
VT4
601
ME/9
P8R
VT4
599
Figure 8-9
Scored Cylinder Liner
In P3R, the mechanical impact events occur at several points in the 720 cycle and are
found symmetrically placed with respect to TDC. This is consistent with liner deteriora-
tion, such that piston rings cause repeated impacts.
8-11
8.9 Port Bridge Problem
The vibration patterns in Figure 8-10 are from an EMD 645 engine, with intake ports
and exhaust valves. Note how the P2L intake port closure event shows up as a large
mechanical impact. Normally the piston and rings should travel upward across the
ports with little if any vibration energy.
Ref: TDC
ME/5
P1L
VT4
848
ME/5
P2L
VT4
845
ME/5
P3L
VT4
845
ME/5
P4L
VT4
845
ME/5
P5L
VT4
848
ME/5
P6L
VT4
848
ME/5
P7L
VT4
848
ME/5
P8L
VT4
848
Figure 8-10
EMD 645 Vibration Patterns
8-12
The implication is that there is some type of blockage or barrier. One possibility is that
there might be carbon buildup in the ports. Another possibility is that the port bridges
have worn in such a way that there is a ridge impeding the ring travel. Figure 8-11
shows the P- and vibration plots for cylinder 2L.
Cyl# P2L
Intake
Exhaust
Fuel
RPM = 845.5
1800
1442
1084
725
367
9
0
90 180 270
360
8.00g
P2L VT4
p
s
i
Figure 8-11
Pressure and Vibration Plots for Cylinder with Port Bridge Problems
8-13
8.10 Improper Valve Seating
The cylinder P9L at intake valve closure has a pattern typical of that occurring when a
valve fails to seat properly, which can be seen in Figure 8-12. There is a leak past the
valve for some degrees until the pressure inside the cylinder forces the valve to seat.
This condition can exist because of a worn guide, a bent valve, or a damaged seat.
ME/9
P1L
VT4
602
ME/9
P2L
VT4
602
ME/9
P3L
VT4
600
ME/9
P4L
VT4
600
ME/9
P5L
VT4
602
ME/9
P6L
VT4
602
ME/9
P7L
VT4
602
ME/9
P8L
VT4
600
ME/9
P9L
VT4
600
ME/9
10L
VT4
600
Figure 8-12
Vibration Pattern with Improper Valve Seating
8-14
8.11 Bent Connecting Rod
The plot of the P- curves for this engine shows good balance between cylinders (see
Figure 8-13) except for P6L (see Figure 8-14), which is low by a significant amount.
Another important observation is that all cylinders are firing consistently, including
P6L, as shown by the tight vertical dimensions of the statistical icons.
2000
1500
1000
500
0
0 90 180 270 360 450 540 630 720
Speed
AVE 601
HIGH 602
LOW 600
P5R P3R P1R P9R P7R
10R P2R P4R P6R P8R
p
s
i
Angle (degrees)
Figure 8-13
Pressure Curves for Right Bank Cylinders
2000
1500
1000
500
0
0 90 180 270 360 450 540 630 720
Speed
AVE 601
HIGH 602
LOW 600
p
s
i
P1L P9L P7L P5L P3L
P6L
P8L P10L P2L
P4L
Angle (degrees)
Figure 8-14
Pressure Curves for Left Bank Cylinders
8-15
Table 8-2 shows that P6L firing pressure is 317 psi below engine average. Another
important piece of information is that the P6L compression reference pressure is 465 psi,
compared to about 560 psi in other cylinders. The immediate cause of the low firing
pressure is clearly the low compression pressure.
Low compression pressure means that the minimum combustion chamber volume in
this cylinder is too large or that a compression leak is occurring. The vibration patterns
show no evidence of leak. The increased combustion chamber volume could result from
an incorrect piston installed, a worn wrist pin, or a bent connecting rod. Inspection
revealed the latter.
Table 8-2
Combustion Report
DIESEL GENERATOR
Unit Name: Model: KSV-20T
Location: Unit Mfr: COOPER-BESSEMER Serial No:
Stroke: 4 Marker Correction Angle: .0 degrees Engine runs clockwise and is articulated
Cyl Speed MEP IHP Comp
Ref
Press
Comb
Start
Max
Rise
Rate
Peak Firing Pressures
AVE DEV MAX MIN DELTA
PFP
Angle
Term
Ref
Press
Exhaust
Ref
Press
rpm psi ihp @ 703
deg
deg
ATDC
psi/deg psi psi psi psi psi ATDC psi psi
1L 602 245 455 597 HI 70 1609 10 1663 1549 -29 10 125 54
2L 600 238 441 591 72 1547 9 1600 1502 -91 10 122 54
3L 600 270 500 562 64 1741 11 1804 1676 103 10 131 56
4L 600 253 468 582 64 1611 8 1668 1562 -27 10 125 55
5L 600 256 473 581 64 1626 8 1678 1588 -12 10 132 56
6L 602 249 462 465 LO -6 HI 58 LO 1321 12 1392 1262 -317 14 134 62 HI
7L 600 254 470 546 60 1585 10 1638 1536 -53 12 128 56
8L 600 264 489 575 64 1695 9 1759 1643 57 10 127 53
9L 600 264 488 575 70 1660 10 1709 1598 22 12 134 58
10L 600 259 481 585 68 1650 10 1717 1581 12 10 132 55
1R 600 262 471 574 68 1647 9 1702 1594 9 12 129 55
2R 600 246 442 556 73 1633 10 1699 1567 -5 10 119 53
3R 602 258 463 586 75 LO 1675 11 1736 1597 37 10 127 57
4R 602 259 463 570 67 1693 12 1768 1617 55 10 129 55
5R 602 247 444 580 66 1642 10 1709 1595 4 12 118 LO 51
6R 602 283 508 536 73 1710 14 1789 1645 72 12 135 HI 58
7R 602 260 466 580 74 1680 10 1738 1609 42 10 122 53
8R 602 261 468 562 67 1689 9 1738 1637 51 10 126 58
9R 600 268 479 561 60 1649 9 1709 1603 11 12 129 57
10R 602 271 487 560 -34 LO 68 1698 14 1790 1632 60 12 127 50 LO
Eng: 601 258 9419 566 -20 67 1638 10 53 11 128 55
Sprd: 0% 17% 14% 23% 28 deg 25% 26% 59% 4 deg 13% 22%
NOTES:
1. The peak firing pressure (PFP) is the highest pressure in the cylinder during the cycle.
2. PFP statistics are based on up to 64 cycles. Ignition statistics are based on 20 cycles.
3. DEV means the average deviation of each cycle from the cylinder average. It is used to determine how consistently a cylinder fires.
4. DELTA is the difference between the cylinder and engine average PFP and indicates cylinder balance.
5. PFP Angle is the angle ATDC at which the PFP occurred.
8-16
8.12 Burned Exhaust Valve
Leakage past valves or rings shows up in vibration patterns as a rounded, gas flow-type
pattern (see Figure 8-15). The amplitude will be related to the pressure in the cylinder.
Injection and combustion occurring near TDC tend to obscure any such leakage pattern.
Therefore, the best place to look is before the start of injection. In the case of Figure 8-15,
the vibration energy can be seen starting to increase well before TDC as compression
pressure increases. This was caused by an exhaust valve leak.
Intake
Exhaust
RPM = 449.8
HP = 81.5
PFP Angle = 14.0
Ign. Adv P9L = 24.0
Ign. Adv P9R = 23.0
591
512
433
355
276
197
-178 -107 -36
36 107 178
P9 VT4
16.00g
p
s
i
Figure 8-15
Burned Exhaust Valve
8-17
8.13 Typical Alco 251 Cylinder Pattern
Fuel
Intake
Exhaust
2000
1500
1000
500
0
-360
-100
0
100 360
10.00g
P8R VT5
10.00g
P8R VT4
p
s
i
Figure 8-16
Typical ALCO 251 Cylinder Pattern
8-18
8.14 Typical Cooper-Bessemer KSV Cylinder Pattern
Fuel
Intake
Closure
Exhaust Closure
Intake
Opens
Exhaust
Opens
Intake
Exhaust
2000
1500
1000
500
0
-360 -180
0 180
360
Vibration
5.00g
Ultrasonic
5.00g
Injection Event
Pressure
Intake
Closure
Exhaust
Closure
Fuel/Combustion
Exhaust Blowdown
p
s
i
Figure 8-17
Typical Cooper-Bessemer KSV Cylinder Pattern
8-19
8.15 Typical EMD 645 Cylinder Pattern
Fuel
Intake
Exhaust
RPM = 899.6
1800
1440
1080
720
360
0
0 90 180 270 360
P4 VT4
8.00g
p
s
i
Figure 8-18
Typical EMD 645 Cylinder Pattern
8-20
8.16 Typical FMOP 38TD 8-1/8: 12 Cylinder Pattern
Intake
Exhaust
1440
1157
874
591
308
25
0 90 180 270 360
P1 VT4
3.00g
RPM = 898.2
HP = 463.7
PFP Angle = 19.3
p
s
i
Figure 8-19
Typical FMOP 38TD 8-1/ 8 Cylinder Pattern
8-21
8.17 Typical Enterprise DSRV Cylinder
Fuel
Intake
Exhaust
RPM = 449.5
HP = 650.0
PFP Angle = 12.9
8.00g
P7L VT4
1440
1155
870
585
300
14
0
180 360 540 720
p
s
i
Figure 8-20
Typical Enterprise DSR 6-4 Cylinder Pattern
8-22
8.18 Typical Colt Pielsick PC2 Cylinder Pattern
Fuel
RPM = 514.1
Intake
Exhaust
8.00g
P9 VT4
1800
1446
1091
737
383
28
0 180
360
540 720
p
s
i
Figure 8-21
Typical Colt Pielstick PC2 Cylinder Pattern
9-1
9
REFERENCES
1. A. Golverk, The Method for Development of a Diesel Engine Universal Perfor-
mance Map, Paper No. 941928 SAE.
2. W. Lowe, The Effect of Ambient and Environmental Atmospheric Conditions,
Imech Proceedings, 184:PT.3P, 1969070, pp. 1824.
3. J.B. Heywood. Internal Combustion Engine Fundamentals. McGraw-Hill, 1988.
4. E.L. Keating. Applied Combustion. Marcel Dekker, 1993.
5. C.F. Taylor. The Internal-Combustion Engine in Theory and Practice Volume 2: Combus-
tion, Fuels, Materials, Design. Revised Edition, MIT Press, 1985.
6. P. Reddy, et al, Evaluation of Combustion Parameters in Direct Injection Diesel
EnginesAn Easy and Reliable Method, Paper No. 930605, Society of Automo-
tive Engineers, 1993.
7. D. Kouremenos, C. Rakopoulos, and K. Kotsos. A Stochastic-Experimental Inves-
tigation of the Cyclic Pressure Variation in a DI Single-Cylinder Diesel Engine,
June 3, 1992.
8. C. Chayangkura, R. Latorre, and D. Charnews, Experimental Study of Diesel
Engine Cycle-to-Cycle VariationPart I: Analysis of Cycle-to-Cycle Cylinder
Pressure Variation, Journal of Ship Research. Vol. 33, No. 3 (September 1989).
9. C. Jones. Diesel Plant Operations Handbook, McGraw-Hill, 1991.
10. A. Krawczak and V. Zacharias, Analysis of Reciprocating Machinery, P/PM
Technology. (February 1993).
11. C.L. Nachtigal, et al. Instrumentation and Control Fundamentals and Applications,
John Wiley & Sons, Inc., 1990.
12. U. Rsli and P. Wolfer, Piezoelectric Transducers Measure Pressures, Technische
Akademie Esslinge, Kistler Instruments #20.1273, Federal Republic of Germany
(March 1986).
13. Dr. A. Wenger, Stability of Pressure Transducers, Technisches Messen. Vol. 12,
pp. 447454 (1986).
14. H. Mller, B. Winter, and K. Rohde-Brandenburger, Comparative Experiments
with Piezoelectric Pressure Transducers to Determine Their Suitability for Use in
Combustion Engines, MTZ Motortechnische Zeitschrift. Stuttgart (1986).
15. Permanent Cylinder Pressure Monitoring of Large-Bore Diesel Engines. IN6.9149Q,
Kistler Instrument Corp., September 1992.
9-2
16. M. Bennett, A Case for Low Impedance. K20.240, Kistler Instrument Corpora-
tion, Undated.
17. W. Mahr, Measuring and Processing Charge Signals from Piezoelectric Transduc-
ers, Technisches Messes. pp. 149158 (1982).
18. U. Dbendorfer and P. Wolfer, State-of-the-Art Pressure Measurements in Com-
bustion Engines, ISA Transactions. Vol. 24, No. 2, Instrument Society of America
(1985).
19. R. Kail and W. Mahr, Piezoelectric Measuring Instruments and Their Applica-
tions, Messen und Prfen. Federal Republic of Germany (1984).
20. H.W. Hugli, Frequency Response of Piezoelectric Transducers: A Practical Ap-
proach. K20.260, Kistler Instrument Corporation, Undated.
21. W.P. Kistler, The Piezotron Concept as a Practical Approach to Vibration Mea-
surement. K20.250, Kistler Instrument Corporation, Undated.
22. R.V. Fisher and J.P. Macey, Digital Data Acquisition with Emphasis on Measur-
ing Pressure Synchronously with Crank Angle, SAE 750028, Automotive Engi-
neering Congress and Exposition, Detroit, MI (February 2428, 1975).
23. J.A. Robison, M.D. Behrens, R.G. Mosher, and J.M. Chandler, Investigating Com-
bustion Phenomena in Unmodified Engines, SAE National Fuels and Lubricants
Meeting, Cleveland, OH (November 8, 1957, SAE Transactions, Volume 66, 1958).
24. D.R. Lancaster, R.B. Krieger, and J.H. Lienesch, Measurement and Analysis of
Engine Pressure Data, SAE 750026, Automotive Engineering Congress and Expo-
sition, Detroit, MI: February 24-28, 1975.
25. A. Crawford and S. Crawford, The Simplified Handbook of Vibration Analysis, Volume
1: Introduction to Vibration Analysis Fundamentals, Computational Systems, Inc.,
1992.
26. H. Hugli, Frequency Response of Piezoelectic Transducer: A Practical Ap-
proach, Kistler K20.260, Kistler Instrument Corporation, Undated.
27. ISO Standard 5348 Mechanical Vibration and ShockMechanical Mounting of
Accelerometers, 1987.
28. B.G. Prasad, H. Shiva, and L. Frederick, A Study of Indicator Passage Error
Correction Schemes, ASME 91-ICE-9, Energy-Sources Technology Conference
and Exhibition, New Orleans, LA (January 2024, 1991).
29. P.V. Puzinauskas and C.M. Beeson, Power-Cylinder Performance Testing on
Large-Bore Engines, (September 1991).
30. W.L. Brown, Methods for Evaluating Requirements and Errors in Cylinder Pres-
sure Measurement, Paper No. 670008, Society of Automotive Engineers (1967).
31. J.D. McCullough, Engine Cylinder Pressure Measurements, SAE Annual Meet-
ing, Detroit, MI (January 15, 1953, SAE Transactions, Volume 61, 1953).
9-3
32. P. Belardini, C. Bertoli, F. Corcione, S. Diana, and G. Police, Error Analysis in
Indicated Pressure and Heat Release Calculations in Direct-Injection Diesel En-
gines, International Symposium on Automotive Technology and Automation
(September 1984).
33. F.R. Caldwell and E.F. Fiock, Some Factors Influencing the Performance of Dia-
phragm Indicators of Explosion Pressures, Paper No. RP1368, National Bureau of
Standards (March 1941).
34. P.G. Bradley and D. Woollatt, Correction of Errors in Indicator Diagrams: Pas-
sage Effects, The Engineer. pp. 511515 (March 1968).
35. E.I.T. Buchholz, Investigation of Pressure Transducer Adapter Dynamics,
Purdue Compressor Technical Conference, pp. 275282 (1978).
36. W. Ullrich, The Influence of Error in Finding TDC on the Determination of Mean
Indicated Pressure, MTZ Motortechnische Zeitschrift. pp. 389392 (1983).
37. A.L. Randolph, Cylinder-Pressure-Transducer Mounting Techniques to Maxi-
mize Data Accuracy, International Congress and Exposition, Detroit, MI (Febru-
ary 26March 2, 1990), Paper No. 900171, Society of Automotive Engineers.
38. C.R. Ferguson, C.F. Warner, R.W. Gottschlich, and S. Wahiduzzaman, Using an
Optical Proximeter to Phase Cylinder Pressure to Crankshaft Position, Paper No.
870478, Society of Automotive Engineers (1987).
39. J.H. Lienesch and M.K. Krage, Using Microwaves to Phase Cylinder Pressure to
Crankshaft Position, Paper No. 790103, Society of Automotive Engineers (1979).
40. S. Kohketsu et al, Technology for Low Emission, Combustion Noise, and Fuel
Consumption on Diesel Engine, Paper No. 940672, Society of Automotive Engi-
neers (1994).
41. T. Nagai et al, Investigation on Reducing NO
x
Emission of Medium-Speed Diesel
Engines, Fuels, Combustion, and Lubrication. ICE Volume 11, pp. 2938 (1990).

Diesel Engine Analysis Guide-107135

Uploaded by

Copyright:

Available Formats

Diesel Engine Analysis Guide-107135

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Diesel Engine Analysis Guide-107135

Uploaded by

Copyright:

Available Formats

What is the purpose of this guide?

What types of diesel engine analysis are discussed in the guide?

'LHVHO(QJLQH$QDO\VLV*XLGH

EPRI TR-107135s Electric Power Research Institute September 1997

You might also like