 

 

 

. . , . a. a s.
3;. . ,, . ‘ EL... .3:
on: . . , . m. ,9an

db

..
ﬂaw”: ,
2:.

A
\n 1:595...
31‘ .- 3.,

..‘. f...»

13-: '

i

 

 

 

 

 

 

 

 

 

A

.
:

 

 

I.“

 

 

 

1. . . Ik:.;...v.,~r~r . in... t

.. f ., . , .. ‘ ‘ .. . 1.9uzrvtu.§‘eﬁvnb§wq£pmrmr§

 

 

—<‘~—-.. -‘4

$0304 LIBRARY
MIL; “5.. State
University

 

 

 

 

This is to certify that the
dissertation entitled

ELASTOHYDRODYNAMIC MODELING AND
MEASUREMENT OF CYLINDER-KIT ASSEMBLY
TRIBOLOGICAL PERFORMANCE

presented by

BOON-KEAT CHUI

has been accepted towards fulﬁllment
of the requirements for the

PHD degree in MECHANICAL ENGINEERING

 

 

Wow

MajoVProfessor’s Signature

June 10, 2005

 

Date

MSU is an Afﬁrmative Action/Equal Opportunity Institution

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

“05128999

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2N5 cJClRC/DateDueJMd—pAs

 

 

ELASTOHYDRODYNAMIC MODELING AND
MEASUREMENT OF CYLINDER-KIT ASSEMBLY
TRIBOLOGICAL PERFORMANCE
By

Boon-Keat Chui

A DISSERTATION
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Mechanical Engineering

2005

ABSTRACT

ELASTOHYDRODYNAMIC MODELING AND MEASUREMENT OF
CYLINDER-KIT ASSEMBLY TRIBOLOGICAL PERFORMANCE

By
Boon-Keat Chui

Engine tribology is becoming a major focus in modern engine development
due to the rising demand for engines with lower energy loss and greater durability.
The cylinder-kit assembly of an internal combustion engine comprises some of
the most crucial components that affect the engine quality in the two areas
mentioned above. This dissertation features the computational modeling of the
engine cylinder-kit assembly tribology, and its implementation into an existing
engine simulation program, CASE, in particular the three-dimensional
elastohydrodynamic lubrication (EHL) system at the sliding interfaces of piston/

bore and piston ring/bore.

A numerical methodology is proposed to develop a three-dimensional
model of the top compression ring lubrication that accounts for the partially
ﬂooded condition coupled with the oil evaporative effect. This model adopts the
Reynolds equation, and Greenwood and Tripp equation to solve simultaneously
for the hydrodynamic and the asperity contact lubrication, satisfying both the force
equilibrium at ring/bore conforrnability and oil transport continuity. It is capable of
accounting for ﬁve distinctive conditions that can potentially exist at the ring/bore
lubrication interface. The ring/bore wear is also computed. Applying it in a speciﬁc

diesel engine problem has demonstrated its unique capability in capturing a

tribological phenomenon that fails to be captured by other approaches. A three-
dimensional piston EHL model is developed to Simulate the dynamics of a piston
that allows its structure to deform elastically within the cylinder bore over an
engine cycle. Using the ﬁnite element approach, the piston solid model is
automatically constructed, meshed, and used in the elastic deformation analyses
under the variation of boundary loads throughout the engine cycle. In the
computation of the hydrodynamic lubricant ﬁlm pressures, the 3-D ﬁnite element
domain is transformed locally into a 2-D domain over the entire piston Skirt
surface area. The calculated lubricant pressures on the thrust and anti-thrust
sides of the piston are then adopted as external loads for the entire piston EHL

analysis.

Experimental measurements for validating the computational results are
performed using a sapphire-bore optical engine system constructed under the
collaborative effort between MSU ARES and Mid Michigan Research. The laser-
induced-ﬂuorescent (LIF) technique is adopted in the oil ﬁlm thickness
measurement. The IR telemetric technique is employed in the measurement of
the piston ring pack gas pressure and the piston temperature using wireless
pressure transducers and thermocouples. Comparing and analyzing the
prediction and the measurement shows that this computational model is a very
practical CAE tool in the engine design and analysis. Further understanding of the
engine tribology has been gained, and recommendations for future improvement

of the computational and experimental techniques have been proposed.

To my Lord Jesus Christ.

To my wife, my daughter,
and my parents

ACKNOWLEDGEMENTS

This dissertation would not have been completed without the support and
assistance of many individuals. I am thankful to all who have contributed, in any
way, to the completion of this dissertation. Among them are a few that deserve

special appreciation.

I would like to thank my advisor Professor Harold J. Schock, for his guidance,
ideas, and supports throughout this work. His sharing of experience and advice

are particularly appreciated.

Special mention must be made to Andy and Jeremiah of Mid Michigan
Research, Tom and Ed from MSU Engine Research Lab for their technical

assistance.

I greatly acknowledge the aid and co-operation of Dr. Mikhail Ejakov from Ford
Motor Co. and of Dr. Lawrence Brombolich from Compu-Tec in the development

of computational engine system.

Finally, my greatest appreciation and love go to my family and friends. Special
thanks goes to my wife for her love, support and prayer. My admiration goes to my
daughter for the happiness she brings me. My sincerest thanks go to Jim and

Karen for their constant love and prayer.

TABLE OF CONTENTS

LIST OF TABLES .................................................................................................. ix
LIST OF FIGURES ................................................................................................. x
KEY TO SYMBOLS AND ABBREVIATIONS ...................................................... xvi
CHAPTER 1 ENGINE CYLINDER-KIT TRIBOLOGY MODELING ....................... 1
1.1 Introduction ...................................................................................... 1
1.2 Motivations and Objectives .............................................................. 2
1.3 Synopsis .......................................................................................... 4
1.4 Literature Review ............................................................................. 7
CHAPTER 2 COMPUTATIONAL ENGINE MODEL - CASE .............................. 19
CHAPTER 3 RING/BORE LUBRICATION AND WEAR ..................................... 23
3.1 Introduction .................................................................................... 23

3.2 Fundamentals in Tribology - Surface Measure and
Characteristics ............................................................................... 23
3.3 Wear Mechanism ........................................................................... 25
3.3.1 Adhesive Equation for Piston Ring/Bore Wear ................... 26
3.4 Piston Ring/Bore Lubrication ......................................................... 26
3.4.1 Hydrodynamic Lubrication .................................................. 27
3.4.2 Hydrodynamic Boundary Conditions .................................. 29
3.4.3 Boundary Lubrication .......................................................... 31
3.4.4 Ring Elastic Conformability in a Distorted Bore .................. 32
3.5 Stewed Lubrication Analysis for the Top Ring ............................... 33
3.5.1 Numerical Scheme ............................................................. 34
3.5.2 Flow Continuity ................................................................... 43
3.5.3 Load Equilibrium ................................................................. 44
3.6 Model Application .......................................................................... 46
CHAPTER 4 OIL EVAPORATION AND STARVED LUBRICATION ................... 54
4.1 Introduction .................................................................................... 54
4.2 Engine Oil Evaporation .................................................................. 55
4.3 Computational Methodology .......................................................... 59
4.3.1 Downward Stroke ................................................................ 60
4.3.2 Upward Stroke .................................................................... 61

vi

4.4 Model Application .......................................................................... 61
4.4.1 Fully Flooded and Various Partially Flooded
Lubrication Analyses ........................................................... 64
4.4.2 Temperature Effect on Bore Wear ...................................... 71
CHAPTER 5 PISTON DYNAMICS AND EHL SOFTWARE ............................... 79
5.1 Introduction ................................................................................... 79
5.2 Simpliﬁed Piston Model ................................................................. 80
5.3 Software Modules .......................................................................... 86
5.4 Three-Dimensional Finite Element Modeling ................................. 87
5.4.1 Constructing and Meshing a Piston Solid Model ................ 87
5.4.2 Load/Boundary Conditions ................................................. 87
5.5 Thermal and Static Analyses ......................................................... 89
5.6 Piston Skirt EHL Analysis .............................................................. 91
5.7 Equation of Motion ......................................................................... 92
5.8 Piston Impact ............................................................................... 101
5.9 Numerical Algorithm for Piston Skirt EHL and
Dynamic Analyses ....................................................................... 104
5.10 Preliminary Results and Discussions .......................................... 105
5.10.1 Piston Dynamics ............................................................... 106
5.10.2 Piston Skirt Elastohydrodynamic Lubrication ..................... 110

CHAPTER 6 COMPUTATIONAL VALIDATION - OPTICAL ENGINE SYSTEM 120

6.1 Introduction .................................................................................. 120

6.2 Laser-Induced Fluorescence Method .......................................... 120

6.3 Optical Engine System ................................................................ 121
6.4 Correlation between Oil Film Thickness and Film

Illumination Intensity .................................................................... 124

6.5 Procedure for Oil Film Thickness Measurement ......................... 128

6.6 Piston Ring Pack Pressure and Temperature Measurement ....... 129

6.7 Engine Test Runs ........................................................................ 130

6.7.1 Inter-Ring Gas Pressure Results ..................................... 130

6.7.2 Oil Film Thickness Measurement Results ........................ 159

CHAPTER 7 SUMMARY .................................................................................. 177

7.1 Partially Flooded Lubrication ....................................................... 177

7.1.1 Conclusions ...................................................................... 177

7.1.2 Recommendations ............................................................ 178

7.2 Oil Vaporization Effect on Partially Flooded Lubrication .............. 179

7.2.1 Conclusions ...................................................................... 179

7.2.2 Recommendations ............................................................ 180

7.3 Piston EHL Model ........................................................................ 181

7.3.1 Conclusions ...................................................................... 181

vii

7.3.2 Recommendations ............................................................ 181

7.4 Experimental Validation ............................................................... 182

7.4.1 Conclusions ...................................................................... 182

7.4.2 Recommendations ............................................................ 182

7.5 Closure ........................................................................................ 183
BIBLIOGRAPHY ............................................................................................... 185

viii

TABLE 1.

TABLE 2.

TABLE 3.

TABLE 4.

TABLE 5.

LIST OF TABLES

Ring/bore lubrication types at 348 degree CA ............................... 49
Engine speciﬁcations ..................................................................... 62
Optical system components and functions .................................. 121

Operating conditions for inter-ring gas pressure measurement .. 131

Operating conditions for oil ﬁlm thickness measurement ............ 162

FIGURE 1.
FIGURE 2.
FIGURE 3.
FIGURE 4.
FIGURE 5.
FIGURE 6.
FIGURE 7.
FIGURE 8.
FIGURE 9.
FIGURE 10.

FIGURE 11.

FIGURE 12.
FIGURE 13.
FIGURE 14.
FIGURE 15.

FIGURE 16.

FIGURE 17.

FIGURE 18.

FIGURE 19.

FIGURE 20.

LIST OF FIGURES

CASE structure .............................................................................. 21
Piston ring program structure ........................................................ 22
Piston program structure ............................................................... 22
Surface measurement and parameters ......................................... 24
Lubricant ﬁlm pressure and boundary conditions .......................... 30
Ring structural FEA and oil lubrication model [34] ......................... 32
Program ﬂow diagram ................................................................... 35
(a) Fully and (b) partially ﬂooded hydrodynamic lubrication .......... 40
(a) Fully and (b) partially ﬂooded mixed lubrication ....................... 41
Full boundary lubrication ............................................................... 42
Ring face discretization at (a) fully ﬂooded and

(b) partially ﬂooded condition ......................................................... 42
Ring surface coordinate system .................................................... 47
Oil ﬁlm distribution across top ring surface near TDC ................... 48

Oil ﬁlm thickness magnitude across top ring surface near TDC.... 48

Asperity contact and hydrodynamic sharing lubrication load ......... 50
Hydrodynamic pressure distribution across top ring surface

near TDC ....................................................................................... 52
Hydrodynamic pressure and front edge ﬁlm separation

point at three distinct bore locations .............................................. 52

Asperity contact pressure distribution across top ring surface

near TDC ....................................................................................... 53
Top piston ring in upward stroke .................................................... 57
Stribeck curve ................................................................................ 60

HGUREZL
HGUREZZ
HGURE23
HGURE24
HGURE25
HGURE26
HGUREZI

FIGURE 28.

HGUREZQ
HGURE3Q
HGURE3L
HGURE3Z
HGURE33
HGURE34
HGURE35
HGURE36
HGURE31
HGURE36
HGURE39
HGURE4Q

FIGURE 41.

FIGURE 42.

Program ﬂow chart ........................................................................ 63
Cylinder bore temperature distribution .......................................... 65
Bore wear (fully ﬂooded lubrication) .............................................. 67
Bore wear (partially ﬂooded, no evaporation) ................................ 68
Bore wear (partially ﬂooded, 2 microns supplied oil ﬁlm) .............. 68
Cyclic top ring oil ﬁlm thickness ..................................................... 69
Bore wear (partially ﬂooded, with evaporation) ............................. 70

Depletion of oil ﬁlm thickness at TRR under the

evaporative effect .......................................................................... 72
Cylinder bore wear with +10 degree C hot Spot temperature ........ 73
Cylinder bore wear with +25 degree C hot spot temperature ........ 73
Cylinder bore wear with +30 degree C hot Spot temperature ........ 74

Cylinder bore wear with +40 degree C hot spot temperature ........ 74
Cylinder bore wear with +50 degree C hot spot temperature ........ 76
Cylinder bore wear with +60 degree C hot spot temperature ........ 77
Cylinder bore wear with +75 degree C hot spot temperature ........ 77

Cylinder bore wear with +100 degree C hot spot temperature ...... 78

Temperature sensitivity of cylinder bore wear ............................... 78
Piston solid model constructed with commercial CAD tool ............ 82
Piston solid model constructed with CASE .................................... 82
Input piston geometry .................................................................... 83

Computed temperature distribution between complicated
and simpliﬁed Piston ...................................................................... 84

Computed distortion between complicated and simpliﬁed piston .. 85

xi

HGURE43
HGURE44
HGURE45
HGURE46
HGURE47
HGURE4&
HGURE49
HGURESQ
HGURESL
HGURESZ

FIGURE 53.

FIGURE 54.
FIGURE 55.
FIGURE 56.

FIGURE 57.

FIGURE 58.

FIGURE 59.

FIGURE 60.

FIGURE 61.

FIGURE 62.

Software modules for piston modeling .......................................... 86
Thermal boundary conditions ....................................................... 88
Constraints and unit pressure load ................................................ 88
Node-to-node temperature comparison ......................................... 90
Node-to-node distortion comparison ............................................. 90
Imposed loads and dynamics of piston during operation .............. 94
Motion modes ................................................................................ 95
Pin forces acting on piston .......................................................... 103
Piston impact with cylinder bore .................................................. 104
Cylinder gas pressure traces ....................................................... 107

Cyclic piston lateral movement at 1500 RPM and
3000 RPM 2 PSI Boost ................................................................ 109

Cyclic piston rotation at 1500 RPM and 3000 RPM 2 PSI Boost 109
Piston contact forces on thrust and anti-thrust sides .................... 110
Piston lubrication forces on thrust and anti-thrust sides ............... 111

Piston skirt thrust and anti-thrust oil ﬁlm pressure at
5 degree CA, 1500 RPM 2 PSI Boost .......................................... 113

Piston skirt thrust and anti-thrust side distortion due to oil
ﬁlm pressure at 5 degree CA, 1500 RPM 2 PSI Boost ................. 114

Piston skirt thrust and anti-thrust oil ﬁlm pressure at
5 degree CA, 3000 RPM 2 PSI Boost .......................................... 115

Piston skirt thrust and anti-thrust side distortion due to oil
ﬁlm pressure at 5 degree CA, 3000 RPM 2 PSI Boost ................. 116

Piston Skirt thrust and anti-thrust oil ﬁlm pressure at
365 degree CA, 3000 RPM 2 PSI Boost ...................................... 118

Piston skirt thrust and anti-thrust side distortion due to oil
ﬁlm pressure at 365 degree CA, 3000 RPM 2 PSI Boost ............. 119

xii

FIGURE 63.

FIGURE 64.

FIGURE 65.

FIGURE 66.

FIGURE 67.
FIGURE 68.
FIGURE 69.
FIGURE 70.
FIGURE 71.
FIGURE 72.
FIGURE 73.
FIGURE 74.
FIGURE 75.
FIGURE 76.
FIGURE 77.
FIGURE 78.
FIGURE 79.
FIGURE 80.
FIGURE 81.
FIGURE 82.
FIGURE 83.
FIGURE 84.

Optical engine equipped with sapphire cylinder .......................... 122
Sapphire cylinder provides optical access over entire

range of piston stroke .................................................................. 122
Optical engine system ................................................................. 123
LIF calibration apparatus (a) calibration stand, (b) micrometer,

and (c) LIF image of oil ﬁlm thickness ......................................... 125
Processed digital LIF image ........................................................ 126
LIF calibration curve .................................................................... 127
Optical engine in test cell ............................................................. 128
Piston equipped with IR telemetrics ............................................ 129
Detailed piston ring pack region .................................................. 131
Inter-ring gas pressure at 1000 RPM 2 PSI Boost ...................... 135
Inter-ring gas pressure at 1000 RPM 4 PSI Boost ...................... 136
Inter-ring gas pressure at 1000 RPM 6 PSI Boost ...................... 137
Inter-ring gas pressure at 1000 RPM 8 PSI Boost ...................... 138
Inter-ring gas pressure at 1500 RPM 2 PSI Boost ...................... 139
Inter-ring gas pressure at 1500 RPM 4 PSI Boost ...................... 140
Inter-ring gas pressure at 1500 RPM 6 PSI Boost ...................... 141
Inter-ring gas pressure at 1500 RPM 8 PSI Boost ...................... 142
Inter-ring gas pressure at 1500 RPM 15 mm Hg M.V. ................. 143
Inter-ring gas pressure at 2000 RPM 2 PSI Boost ...................... 144
Inter-ring gas pressure at 2000 RPM 4 PSI Boost ...................... 145
Inter-ring gas pressure at 2000 RPM 6 PSI Boost ...................... 146
Inter-ring gas pressure at 2000 RPM 8 PSI Boost ...................... 147

xiii

FIGURE 85. Inter-ring gas pressure at 2000 RPM 15 mm Hg M.V .................. 148
FIGURE 86. Inter-ring gas pressure at 2500 RPM 2 PSI Boost ...................... 149
FIGURE 87. Inter-ring gas pressure at 2500 RPM 4 PSI Boost ...................... 150
FIGURE 88. Inter-ring gas pressure at 2500 RPM 6 PSI Boost ...................... 151
FIGURE 89. Inter-ring gas pressure at 2500 RPM 8 PSI Boost ...................... 152
FIGURE 90. Inter-ring gas pressure at 2500 RPM 15 mm Hg M.V .................. 153
FIGURE 91. Inter-ring gas pressure at 3000 RPM 2 PSI Boost ...................... 154
FIGURE 92. Inter-ring gas pressure at 3000 RPM 4 PSI Boost ...................... 155
FIGURE 93. Inter-ring gas pressure at 3000 RPM 6 PSI Boost ...................... 156
FIGURE 94. Inter-ring gas pressure at 3000 RPM 8 PSI Boost ...................... 157
FIGURE 95. Inter-ring gas pressure at 3000 RPM 15 mm Hg M.V. ................. 158
FIGURE 96. Top ring oil ﬁlm thickness proﬁles measured over various

engine speeds at 320 degree CA ................................................ 160
FIGURE 97. Second ring oil ﬁlm thickness proﬁles measured over

various engine speeds at 320 degree CA ................................... 160
FIGURE 98. Oil ﬁlm thickness proﬁle at 500 RPM, 40 degree CA ................... 161
FIGURE 99. Oil ﬁlm thickness proﬁle at 1000 RPM, 320 degree CA ............... 161
FIGURE 100.0il ﬁlm thickness measured at 0 degree CA, 500 RPM .............. 163
FIGURE 101.0il ﬁlm thickness measured at 90 degree CA, 500 RPM ............ 164
FIGURE 102.0il ﬁlm thickness measured at 180 degree CA, 500 RPM .......... 164
FIGURE 103. Oil ﬁlm thickness measured at 270 degree CA, 500 RPM ......... 165
FIGURE 104.0i| ﬁlm thickness measured at 360 degree CA, 500 RPM .......... 165
FIGURE 105. Oil ﬁlm thickness measured at 450 degree CA, 500 RPM ......... 166
FIGURE 106.0il ﬁlm thickness measured at 540 degree CA, 500 RPM .......... 166

xiv

FIGURE 107. Oil ﬁlm thickness measured at 630 degree CA, 500 RPM ......... 167
FIGURE 108.0il ﬁlm thickness measured at 0 degree CA, 1500 RPM ............ 167
FIGURE 109. Oil ﬁlm thickness measured at 90 degree CA, 1500 RPM ......... 168
FIGURE 110.0il ﬁlm thickness measured at 180 degree CA, 1500 RPM ........ 168
FIGURE 111. Oil ﬁlm thickness measured at 270 degree CA, 1500 RPM ....... 169
FIGURE 1120" ﬁlm thickness measured at 360 degree CA, 1500 RPM ........ 169
FIGURE 113. Oil ﬁlm thickness measured at 450 degree CA, 1500 RPM ....... 170
FIGURE 114.0il ﬁlm thickness measured at 540 degree CA, 1500 RPM ........ 170

FIGURE 115. Oil ﬁlm thickness measured at 630 degree CA, 1500 RPM ....... 171

FIGURE 116.Cyclic top ring oil ﬁlm thickness at 500 RPM 2 PSI Boost ........... 173
FIGURE 117.Cyclic 2nd ring oil ﬁlm thickness at 500 RPM 2 PSI Boost .......... 173
FIGURE 118.Cyclic top ring oil ﬁlm thickness at 500 RPM 8 PSI Boost ........... 174
FIGURE 119.Cyclic 2nd ring oil ﬁlm thickness at 500 RPM 8 PSI Boost .......... 174

FIGURE 120.Cyclic top ring oil ﬁlm thickness at 1500 RPM 2 PSI Boost ......... 175
FIGURE 121.Cyclic 2nd ring oil ﬁlm thickness at 1500 RPM 2 PSI Boost ........ 175
FIGURE 122.Cyclic top ring oil ﬁlm thickness at 1500 RPM 8 PSI Boost ......... 176

FIGURE 123.Cyclic 2nd ring oil ﬁlm thickness at 1500 RPM 8 PSI Boost ........ 176

XV

Q

'53. “can:
‘b ‘6

Q'D<':>’®J_;®,

d

Eé

AQe'vap
At
(9(5)

¢r0d
¢rod

¢rod

KEY TO SYMBOLS AND ABBREVIATIONS

Thermal expansion coefﬁcient, Ring twist angle
Radius of curvature of asperity summit

Piston rotation angle

Piston angular velocity
Piston angular acceleration

Surface density of asperity peaks
Crank rotation angle

Wave length

Oil viscosity

Poisson's ratio

Density

Surface roughness

Shear stress
Wear rate, Piston tilt angle

Instantaneous area of exposure of cylinder bore oil ﬁlm in com-
bustion chamber

Incremental volume of vaporization of oil from the cylinder bore

Time of exposure of cylinder bore oil ﬁlm in combustion chamber
Standardized height distribution (Gaussian)
Flow factor

Connecting rod rotation angle

Connecting rod angular velocity

Connecting rod angular acceleration

Composite elastic modulus

Elastic modulus of ﬁrst and second surfaces
Lubrication load from asperity contact component

Load due to gas pressure behind ring

xvi

Flt
FLUB’FL
FR
F

tension

H
P

cylinder

P2

Tcylinder
U, V, W
W

a

h

h(x) ,h(x)*
hb

h *

min ’ min

Load due to hydrodynamic lubrication
Total lubrication load

Combined applied load

Load from ring tension

Material hardness

Cylinder gas pressure
2nd land gas pressure
Cylinder gas temperature

Velocity

Asperity contact load

Nominal clearance between two contact surfaces

Oil ﬁlm thickness across ring face proﬁle

Critical oil ﬁlm thickness at which full boundary lubrication occurs

Minimum oil ﬁlm thickness

Oil ﬁlm thickness at the location of ap/ax = 0
Supplied oil ﬁlm thickness

Wear coefﬁcient

Evaporative mass ﬂux of vaporized oil

Contact pressure from elastically deformed asperity
Hydrodynamic pressure

Lubricant ﬂow rate

Time

Velocity proﬁle across gap between two Sliding surfaces
Distance of ﬁlm separation point

Elastohydrodynamic lubrication

Bottom dead center

Top dead center
Top ring reversal position

xvii

a- , h, s,
”,1. PL 2, R

Top contact force (positive on thrust side, negative on antithrust
side)

Bottom contact force (positive on thrust side, negative on antithrust
Side)

Lubricant force on thrust side
Lubricant force on antithrust Side

Friction force at top contact point (positive on thrust side, negative
on antithrust Side)

Friction force at bottom contact point (positive on thrust side, nega-
tive on antithrust Side)

Lubricant friction force on thrust side

Lubricant friction force on antithrust side

Moment developed between piston pin and piston due to friction
Vertical component of combustion gas pressure load

Horizontal component of combustion gas pressure load

piston pin diameter

Coefﬁcient of friction at contact points

Mass of piston pin
Mass of piston
Mass of connecting rod

is

Mass moment of inertia of piston,1p

Mass moment of inertia of connecting rod

Geometrical dimension of piston assembly (refer to Fig. 48)

xviii

CHAPTER 1 ENGINE CYLINDER-KIT TRIBOLOGY MODELING

1 .1 Introduction

Knowledge in engine tribology is very important to the development of any
modern engines that involve mechanism of motions and contact of surfaces. What

is tribology? A deﬁnition according to Ludema [66] is as follows:

Tribology is the ‘ology’ or science of ‘tn'bein.’ The word
comes from the same greek root as tribulation. ’ A
faithful translation deﬁnes tribology as the study of

rubbing or sliding. The modern and broadest meaning

is the study of friction, lubrication and wear.

Indeed, engine tribology contributes to a very wide area of disciplines in
engine study, ranging from ﬂuid mechanics, heat and mass transfer, solid
mechanics, to engine kinematics and dynamics. The understanding of engine
tribology explains what happens at the frictional components of a reciprocating
internal combustion engine (bearings, piston assembly and valve train), or at the
rotational components (Shafts and bearing) of a turbo engine or an electric motor
in a hybrid engine, in which these components dictate predominantly the
efﬁciency of the engines. Engine tribology long has been researched by those
pioneering in engine development decades ago in an effort to look for more
efﬁcient engines that run smoother, faster and longer. In recent years, a
methodology that emerges to the focus of engine tribology research is the

computational method. The advanced computer has motivated many to employ

this methodology since it has been proven effective and useful. For instance, at
the availability of computationally systematized calculation methods, a signiﬁcant
amount of lead time and cost have been reduced for a new vehicle from the
conceptual phase to the production phase. Many alternative designs can be
evaluated through computational models before one is selected, and this means
that the selected design is often much more likely to be right and to perform better

in operations.

1.2 Motivations and Objectives

Despite many advantages of the computational method, it must be
integrated into a broader engineering strategy that allows a computational product
to be veriﬁed and gives conﬁdence to engineers, which is suggested by Parker
[81] as a principle in the veriﬁed predictive technique. A computational model
development often starts with experimental measurement, particularly where a
phenomenon of interest is intricate. From collected data, an insight into what is
actually happening is useful, if not crucial, in the proposition and construction of a
mathematical model. Then, the developed model can be run to simulate results to
be compared with other measurements, a procedure that often leads to
improvements to the model in an attempt to minimize the discrepancy between its
predictions and experimental observations. Likewise, gradual application of the
model to real design situations provides feedback from engine running
experience, which increases the reliability of its use and further allows its scope to

be extended.

Making use of the advantages of the computational method, this
dissertation attempts to establish a comprehensive engine simulation system that
has its predictions being veriﬁed or validated to some extent by experimental
observations. This dissertation focuses on the development, implementation and
application of engine tribological systems that simulates elastohydrodynamic
lubrication (EHL), friction and wear at the frictional cylinder-kit components (piston
skirt and ring) and a distorted cylinder bore under conscientious backing from
experimental observations. The ultimate goal is to produce a comprehensive tool
that computationally integrates the complex coupled systems in a cylinder-kit
system (piston, piston rings, cylinder bore, and connecting rod) in order that a
complete engine design analysis can be simulated efﬁciently in the aspect of
computational effort and time. Three speciﬁc major accomplishments achieved in

this dissertation are:

1.) Implemented the models of three-dimensional piston ring/liner wear,
and three-dimensional piston skirt elastohydrodynamic lubrication into an existing

engine Simulation program;

2.) Constructed and implemented a ring/bore lubrication model that
accounts for the partially ﬂooded analysis coupled with the oil vaporization

analysis; and

3.) Validated the predicted results with measurements obtained from an

optically accessible engine system.

1.3 Synopsis

An existing engine simulation program, CASE, under the sponsorship of
Mid Michigan Research, is used as a foundation for the development of the new
computational models of this dissertation. The CASE system is a computational
program used for the analysis and design of pistons, rings and other cylinder-kit
components. The existing CASE system allows for the computations of
phenomena such as piston ring pack gas ﬂow, ring/liner lubrication, piston and
ring dynamics for an entire engine cycle. Although it is a rather complete and
stand-alone program, it does allow new models to be implemented into it for better
and more accurate prediction of engine performance. More detailed background

of the CASE system is given in Chapter 2.

As piston rings slide against the bore liner in an engine, lubricated contact
occurs either in the regime of hydrodynamic, boundary or mixed. The boundary
contact induced from the asperity-to—asperity interaction between two surfaces
often causes surface deformation or wear. An improved piston assembly
lubrication and wear model has been developed and implemented into the CASE
system. This model couples the physics of the ring/liner conformation and
lubrication, and the effect of surface microstructures in the determination of piston
ring/liner wear. Potential lubrication conditions due to partially ﬂooded condition
are also simulated using a starved lubrication model which numerically derives
the hydrodynamic ﬁlm separation points through the satisfaction of global load
equilibrium and ﬂow continuity at the ring/bore interface. More detail of the

implementation and application of this model will be given in Chapter 3.

Starved or partially ﬂooded condition is commonly believed as a factor in
determining the piston ring/bore lubrication. The ring/bore interface is not always
ﬁlled with lubricant entirely - known as fully ﬂooded condition, despite being
assumed to be such in many early lubrication models. As a top compression ring
slides towards the bottom dead center (BDC), a lubricant ﬁlm is left behind on the
cylinder bore and exposed to the combustion chamber environment. Any heat and
mass transfer process, such as oil vaporization, will reduce the cylinder bore ﬁlm
thickness. As the top compression ring returns during compressing strokes
(towards TDC), the bore oil ﬁlm thickness becomes the supplied lubricant that
enters the leading edge of the ring face, and dictates ﬂooded condition at the
interface. Hence, there is a very close interaction between the heat and mass
transfer process and the lubrication process at the ring/bore interface. A model of
oil vaporization that accounts for local variation of boundary conditions, in
particular the bore temperature, distortion and hydrodynamic pressure across the
bore surface, is developed and implemented into CASE to account for the
temperature-induced starved condition at the piston ring/bore lubricated interface.

More detail of this model will be presented in Chapter 4.

Piston modeling using the computational approach has become a vital part
in today’s automotive engine development. In recent years, due to the
advancement of computer technology, more efforts have been seen in the
development of 3-D piston models, which are used in simulating the complicated
system of dynamics and tribology at the piston/cylinder-bore conjunction.

Chapter 5 presents a newly developed/stand-alone piston software that is capable

of automatically constructing a 3-D ﬁnite element (FE) piston model and solving
the coupled interaction between the piston skirt elastohydrodynamic and the
piston trajectory over the entire four-stroke engine cycle. The software is also
implemented with FEA solvers to solve for the thermal and mechanical distortion,
and the elastohydrodynamic lubrication of the piston. Comparison of some of the
computation results with another commercial software is also presented for the
purpose of validation. The application of this software is conducted over an
operating engine to demonstrate its capability in predicting the piston Skirt
lubrication and dynamics. More detail of its implementation and application will be

given in this chapter.

An experimental rig is constructed under the joint-effort between Mid
Michigan Research and Michigan State University to perform experimental
measurements of the cylinder-kit oil ﬁlm thickness, temperature and pressures.
The experiment rig includes an optically accessible engine and optic devices
constructed in line with the technique of laser-induced ﬂuorescence (LIF).
Measured and predicted results are compared and analyzed. More description of

this experiment will be provided in Chapter 6.

Finally, the conclusions and recommendations of this present dissertation
will be given in Chapter 7 to summarize how each part is signiﬁcant to this present

work, with conclusions from the integration of all parts.

1.4 Literature Review

Many previous works have been committed tov’the study of the tribology at
the piston ring/bore and the piston/bore interfaces in internal combustion engines
because of the common belief that the tribological interaction at these two primary
interfaces accounts for a substantial inﬂuence on engine efﬁciency

[7.22.51.89,92,94].

Over the past few decades, much research has been conducted to study
and understand the lubricating mechanism at the piston ring/bore sliding interface.
This is a very challenging ﬁeld because an engine, in particular the IC engine, can
operate over a huge range of operating conditions. Fully ﬂooded or starved
conditions in hydrodynamic, boundary or mixed regimes can all potentially occur
in a running engine, and each posts a unique problem. Thus, researchers in
piston ring pack lubrication often conduct studies by imposing constraints through

limiting experimental operating ranges or through imposing speciﬁc assumptions.

Early diligent efforts to investigate the basic principles of the piston ring
lubrication, both experimental and theoretical, have been reported between the
year of 1925 and 1944, according to Grice [46]. Many early mathematical models
adopted the Reynolds equation, named after Osborne Reynolds. He derived this
equation from the full expression of hydrodynamics, known as the Navier—Stokes
equation [11,12,50,38], to solve for the hydrodynamic lubrication at the sliding
interaction between piston rings and cylinder bore. Many assumptions were made

to simplify their models, such as the use of the laminar and Newtonian ﬂow for the

lubricant and the smooth surface contact. Castleman found good correlation of his
theoretical prediction of piston ring lubrication with results of an experiment in
which the comparison of the oil ﬁlm thickness was taken at distances away from
the dead centers of piston stroke [15]. He concluded that the hydrodynamic theory
was sufﬁcient to represent the piston ring lubrication even without the boundary
lubrication analysis. Eilon and Sanders [27] also developed a hydrodynamic
theory of the piston ring, without the squeeze ﬁlm term, derived from their
experimental study of the lubrication between a single-ring piston assembly and a
cylinder. Their experimental apparatus consists of a stationary piston and a
reciprocating liner. Reasonably good agreement between the theoretical and

experimental results was claimed.

Without the squeeze ﬁlm effect, it is rather inadequate to represent the
piston ring lubrication at the dead centers of the piston stroke. Here, sliding
velocity is zero and surface metallic contact and the squeeze ﬁlm effect are
important. An improved hydrodynamic model was introduced by Furuhama [39]
along with his experimental effort [40], in which an apparatus like that of Eilon and
Sanders was adopted. He showed the importance of the inclusion of the squeeze
ﬁlm effect in the hydrodynamic theory, in particular at the TDC and BDC where the
piston ring velocity is zero. Furuhama et al. [41] later introduced another method
for measuring the oil ﬁlm thickness through the monitoring of the variation of ring
endgap. Furuhama’s ﬁnding was also supported by Lloyd’s work [65] in which he
solved the Reynolds equation numerically using a computerized system. The

advancement of calculation speed with the computer allowed him to conduct

design evaluation of the piston ring proﬁle on the cyclic lubrication performance.
His effort was followed by Baker et al. [5] in the investigation of the piston ring
design on the piston ring performance. They also recommended the importance
of including the non-uniform nature of circumferential ring loading and multi-ring

pack system for the piston ring lubrication modeling.

In contrast to early experiments that adopted the stationary ring and
reciprocating cylinder liner, various experimental approaches that conducted
direct measurement of piston ring in an operating engine have been reported in
the 1970’s. Wing et al. [127] reported a technique using inductive proximity
transducers and thermocouple in the measurement of oil ﬁlm thickness and
temperature of the piston rings, in order to investigate the thermal and vibrational
behavior of piston rings, which was considered a better approach for
understanding the piston ring lubrication mechanism. Hamilton et al. [48] also
reported a technique of oil ﬁlm thickness measurement using capacitance
transducers. Unlike Wing’s method, which had the measuring devices mounted
on the ring that caused instability in the signal output, Hamilton et al. [49] had the
capacitance transducers mounted at several locations on the cylinder liner to
measure the oil ﬁlm thickness of the ring pack that passes the locations. They
later introduced a numerical solution to the classical Reynolds equation assuming
Reynolds’ boundary conditions and fully ﬂooded condition [49]. When compared
with experimental measurements, the numerical solution overpredicted the oil ﬁlm
thickness, which led to the belief that the oil starvation could potentially exist in the

piston ring lubrication.

The rapid development of computerized technology in 1970’s has
motivated greater and more extensive efforts in piston ring tribological study. More
complicated analyses have been incorporated into the piston ring lubrication
model. Ting and Mayer [117,118,119] presented a comprehensive ring pack
model that simulates a more inclusive piston ring pack analysis, which includes
ring pack gas ﬂow dynamics using the oriﬁce volume method, mixed lubrication,
and adhesive cylinder bore wear. Ring elastic characteristics were also
considered. Ruddy et al. [96,97,98] also conducted a comprehensive study of
piston ring lubrication by including the effects of ring orientation due to twisting

and dynamics, and the thermal distortion and wear of ring groove.

Later works by Allen [1], Brown and Hamilton [9], Dowson et al. [23], and
Moore and Hamilton [75] suggested the importance of including the starvation
effect in piston ring lubrication as proven by their improved results in comparison
with the experimental measurements. Dowson et al. [23] extends the starved
lubrication analysis to a multi-ring pack computation. Their work showed how the
starved lubrication of a ring is related to the deposition of oil transport from the
preceding ring in a ring pack. They also concluded that because of the ﬂow
continuity of the inter-ring oil transport mechanism, the minimum oil ﬁlm thickness
experienced by all the ring must be similar under the criteria that the ring face

proﬁles did not have extreme difference.

A review by McGeehan [73] has suggested that hydrodynamic lubrication

predominantly contributes to the overall friction at the ring/bore interface except at

10

the TDC and BDC, which implies that mixed lubrication is basically not an
important factor to be considered in the ring/bore lubrication analysis. In spite of
that, many studies have recognized that surface roughness plays important roles
in lubrication, in particular in the regime of mixed lubrication as it has the same
order of magnitude as the average ﬁlm thickness. In efforts to account for the
global effect of surface roughness, an averaging technique has been adopted
because most engineering surfaces display random characteristics. Among these,
Patir and Cheng (PC) [82] have proposed an average ﬂow model using pressure
and shear ﬂow factors as correction terms in an equation derived from the
classical Reynolds equation to solve for slider bearing equation on two rough
surfaces that slide against each other. Greenwood and Tripp (GT) [45] derived a
theoretical model that describes the asperity contact of two nominally ﬂat rough

surfaces.

A mixed lubrication model for the piston ring was presented by Rohde [93],
which studied the sharing of frictional contact load between the hydrodynamic and
the asperity contact pressure. Adopting the PC-averaged Reynolds equation and
the GT asperity model, the effects of surface topography on lubrication and friction
were studied in detail. His results showed the dominance of asperity contact at
TDC and BDC, and that contributes to high friction loss. He concluded that the
piston ring power loss is dependent on the surface topography and lubricant
properties, in particular during low engine speeds. A more Speciﬁc application of
the PC and GT models was conducted by Ruddy et al. [98] on modeling a twin-

land oil control ring lubrication.

11

Another important aspect of the piston ring lubrication is the elastic
characteristics of the piston ring in cylinder bore. Dowson et al. [24] discussed the
necessity to consider both the rigid body motion and the rate of change of the
local elastic compression of piston rings in the squeeze ﬁlm contribution to
hydrodynamic action in the piston ring/bore interaction. Elastohydrodynamic
action was suggested as an important factor to the tribological performance of the
piston seal, particularly near TDC, in which it is believed to be able to improve the
oil ﬁlm thickness at TDC and reduce the friction. Sun [106] attempted to solve the
nonuniform contact problem of the ring/bore conforrnability using the elastic
theory incorporated with the thermal elasticity. He also [105] proposed a generic
computational solution for the Slider lubrication problems. His solution addressed
the complicated issues of the ﬂow continuity and the attachment/detachment of

the oil ﬁlm on the sliding interface.

The introduction of optical measurement technology with the application of
the laser beam has motivated several efforts in the measurement of piston ring oil
ﬁlm thickness. Ting in 1980 [120] published his ﬁrst effort in piston ring oil ﬁlm
thickness measurement using a laser ﬂuorescence technique, and demonstrated
the potential of this technique to achieve more precise measurement. Unlike the
conventional electronic method, the optical technique does not require any device
to be mounted on the piston ring or the cylinder liner and therefore reduces the
possibility of external disturbance on the piston ring dynamics and lubrication to
minimal. Another advantage is that it allows visual observation to be done

instantaneously during the measurement, and therefore allows more insight into

12

piston ring dynamics and lubrication. However, due to the pulsing effect in most
laser source, the calibration of the laser source needs to be done carefully in order
to achieve accurate measurement as reported by Hoult [52]. Richardson [90]
applied an improved optical technique, which is called the laser induced
ﬂuorescence (LIF) technique, in the measurement of piston ring oil ﬁlm thickness
along with his modeling effort in piston ring lubrication. Other efforts using LIF are
found in the works of Takiguchi [109], Nakayama et al. [78] and Suguru [103].
Other measurement techniques are also found in Liu et al. [64], who conducted
the measurement of the piston ring lubrication under motoring conditions using
Eddy Current Sensors which are mounted on the liner; and in Takiguchi et al.
[110], who adopted the capacitance method to measure a three-ring pack oil ﬁlm

thickness in an operating diesel engine.

The need to consider the circumferential variation in the ring/bore
interaction have been raised by McGeehan [74] and Ting [121]. McGeehan, in his
survey of the mechanical design factors that affect engine oil consumption,
pointed out the Signiﬁcance of bore distortion on oil transport along the ring/bore
interface. Ting’s review of piston ring tribology indicated that the ring/bore

conforrnability is an important issue to be considered in the piston ring lubrication.

Grice [46] developed a three-dimensional model of piston ring lubrication
that accounts for the circumferential variations that exist between a piston ring
and cylinder liner, in order to represent accurately the physics of the oil transport,

friction and lubrication ﬁlm thickness more correctly. His theoretical prediction was

13

compared with experimental measurement using an improved capacitance

technique that allows both the oil ﬁlm thickness and friction to be measured.

With the major advancement of computers in the 1990’s, computational
time is becoming a lesser issue in piston ring modeling. Hence, several
computational models of three-dimensional piston ring lubrication analysis have
been reported. Ma et al. [68,69,70,71,72] reported a comprehensive three-
dimensional model of piston ring lubrication that accounts for bore out-of-
roundness and starved condition. The full Reynolds equation is solved cyclically
using the ﬁnite difference method. The lubricant was assumed to be
incompressible, Newtonian and laminar. However, the model did not consider
surface roughness effect and secondary ring motion. They concluded that the
piston ring performance, energy loss and oil transportation are signiﬁcantly
affected by the bore shape and the ring lateral displacement. Their starved
lubrication analysis shows that the relative ring locations in a ring pack are crucial
in determining the oil availability to the rings of a ring pack, and starved condition
was observed on all compression rings throughout the engine cycle, in which only
10-40% of ring face is regularly covered with the full oil ﬁlm. In addition, the
increase of the bore temperature seems to be able to reduce the average power
loss and the net oil transport of the ring pack. A signiﬁcant degree of agreement
was claimed by Ma et al. in their comparison with other earlier measurement by
Brown and Hamilton [9], and Hamilton and More [49]. Tian et al.
[112,113,114,115,116] also presented a similar piston ring pack system that

displays similar capability as Ma et al., but with the addition of mixed lubrication

14

and more application efforts in production IC engines along with validation from

their experimental measurements.

Ejakov [34] developed a three-dimensional piston ring pack lubrication
model that accounts for load sharing between the hydrodynamic and asperity
contact pressure at the ring/bore interface. His earlier work in the three-
dimensional modeling of the piston ring twist [28,32] was incorporated with this
model along with a piston dynamic model [13] to illustrate the importance of
including ring/bore conforrnability, piston ring twisting motion, piston tilting motion
and bore distortion in the analyses of piston ring pack lubrication, gas ﬂow and

dynamics.

Despite the continuous effort in the three-dimensional modeling, the two-
dimensional theory is still being employed in several piston ring lubrication studies
[131,132,102,104] that usually combine the theoretical prediction and
experimental measurement because the two-dimensional theory is still reliable.
This is supported by many previous reports, provided the non-uniformability of the
ring/bore interaction is small. The efforts to include the non-Newtonian behavior of
the lubricant in the piston ring lubrication model were also reported by Taylor et al.
[111] and Tian et al. [113], in which the shear thinning effect on the lubricant

viscosity is considered in additional to the temperature effect.

The efforts in mixed lubrication lead to the implementation of the ring/bore
wear modeling [18,19,20,21,61,66,84,119]. A fundamental law of adhesive wear

discussed by J. F. Archard [2] has been adopted or modiﬁed in the calculation of

15

surface wear at the sliding interaction between the piston assemblies and the
cylinder bore. Chung et al. [21] developed a mathematical model to predict ﬁre
ring wear using a linear relationship between ring wear and the contact friction,
and compared the results with experimental measurement [101]. Pint et al. [89]
developed a model for 3-D piston ring wear accounting for surface topography on
the ring wear using Abott FireStone Curve (AFC) to account for surface change
affected by wear. Chui [20,21] conducted continuous cyclic analysis on piston ring
wear and its effect on oil consumption. Different piston ring wear measurement
techniques have been reported [6,53,67,101,123]. Among those is the radioactive
tracer technique that provides higher accuracy of ring wear measurement, which
has radioactive material implanted on the piston ring surface, and any amount
surface loss can be detected. Barkman [6], Treuhaft et al. [123] and Schneider et
al. [101] conducted the piston ring wear measurement using the radioactive tracer
technique. A general agreement on the predicted and the measured ring wear is
in the transient progression of the piston ring wear rate, in which the piston ring
wear rate begins with high value initially and decreases in an exponential trend,

and ultimately reaches a steady state value.

Compared with piston ring tribology, the efforts in piston tribology appears
to be very few. One reason is that the complexity to solve for the Reynolds
equation on the piston skirt surface area requires substantial amount of
computational power. The difﬁculty of solving for the three-dimensional solid

deformation is another reason.

16

Early piston models were developed primarily to investigate piston
dynamics and its impact on the cylinder bore that results in engine noise and
vibration, in particular the piston Slap phenomenon. The effects from the
lubrication and the piston Skirt elasticity on the piston dynamics were either not
considered or assumed insigniﬁcant [25,37,47,60,76,79,85,99,108,124].
Nevertheless, valuable insights and principles obtained from these studies
regarding piston dynamics have contributed to the later efforts that attempt to
couple lubrication and piston ﬂexibility into piston dynamics. By considering the
piston Skirt lubrication and ﬂexibility, the real physics of an operating piston can be
better represented. Knoll and Peekan [58], Li et al. [63], Goenka [44], Keribar [56]
introduced piston Skirt hydrodynamic lubrication models that solve for the two

dimensional Reynolds equation over the piston skirt surface.

The elastic characteristic of the thin piston skirt was ﬁrst considered in the
piston dynamic analysis by Li [62]. nltegrating piston skirt elastic behavior into
hydrodynamic lubrication has formed the elastohydrodynamic lubrication analysis,
which is considered by most of the recent efforts [26,57,59,80,129,133,134].
Among these, Oh et al. [80], Zhu et al. [133, 134] and Dursunkaya et al. [26] have
contributed to solving the elastic deformation of the piston using the ﬁnite element
method and its integration in the calculation of piston trajectory and lubrication as
functions of crank angle. The ﬁnite element piston analysis is solved by a
separate ﬁnite element solver, and the calculated coefﬁcient matrices or the
compliance matrices are input into the piston model of lubrication and dynamics in

order to reduce computational time. However, this approach may result in the lost

17

 

of crucial numerical information because of the transfer of the nodal information
between the 3-D FE model and the 2-D piston skirt lubrication model that each
posses different mesh resolution, particularly when there exist sharp gradients

between two nodes.

18

CHAPTER 2 COMPUTATIONAL ENGINE MODEL - CASE

The Cylinder-kit Analysis System for Engines (CASE) [13,14], under the
sponsorship of Mid Michigan Research, is used as the foundation for the
development of the new computational models proposed in this present work.
This engine simulator has been well established for years and consistently
undergoes reﬁnement at the availability of the advancement of engine technology
and knowledge. CASE can be used in the engine design and analysis tasks for
given engine conﬁgurations and operating conditions. The parametric study for
engine optimization can be performed at a relatively time and cost efﬁcient
manner by using CASE. CASE also has ﬂexibility for other computational routines
to be incorporated or integrated for more complicated engine simulation. The
reliability of CASE has been proven from its application and development in a
number of practical applications in the past [8,18,19,20,21,28,29,30,31,32,33,34,

35,84,85].

CASE contains coupled computational models of several important
systems in an internal combustion engine. These systems are the piston and
piston ring dynamics, heat and mass transfer system of engine working ﬂuids, the
solid mechanics, and the tribology at the relative sliding surface contacts - piston
skirt/liner and piston ring/liner, which are indeed some of the most crucial engine
systems that affect the overall engine performance, fuel and oil consumption, and
engine durability. The basic structure of CASE is shown in Fig. 1. TWIST, RING

and PISTON are three main CASE programs and the integration of them allows a

19

complete engine simulation even though each can be run independently for a
more speciﬁc engine analysis. The highlighted engine subsystems in Fig. 1 are
the focus of this dissertation, where improvement and implementation of these

models were conducted.

The RING and TWIST programs contain computational models for the
piston ring system while PISTON applies to the piston system. The piston ring
system, as shown in Fig. 2, includes the piston ring dynamics within piston
grooves, both axial and twisting behaviors under the inﬂuence of inertia force,
combustion gas pressure, and ring tension. It also includes the gas dynamics of
the combustion gas trapped within the piston ring pack, and the piston ring
tribology such as ring/liner lubrication, friction loss due to viscous lubricant and
asperities contact, and mechanical wear. The gas dynamics analysis is a two-
dimensional model that computes the average gas ﬂow across the inner piston
ring grooves and lands using the oriﬁce volume approach without considering the
circumferential variation at the ring/bore interface. However, when the ring twist
analysis is included, because of its three-dimensional nature that affects the gas
ﬂow path dimension axially and circumferentially, the gas ﬂow analysis can be
considered three-dimensional. The piston ring lubrication model is a three-
dimensional model that accounts for both axial and circumferential variations of
boundary conditions namely bore distortion, bore temperature, piston tilting and
ring twisting. In addition, it accounts for the mixed lubrication of the piston ring that

comprises the Reynolds hydrodynamic and asperity contact lubrication. The latest

20

developments contributed from this dissertation are three-dimensional ring/bore

wear, partially ﬂooded lubrication (oil transportation) and oil vaporization.

The piston system, as shown in Fig. 3, includes piston dynamics in the
engine cylinder under the inﬂuence of combustion gas pressure, inertia and
contact forces from friction and lubrication. In addition, it also includes piston solid

mechanics and piston skirt/liner tribology elastohydrodynamic lubrication.

CASE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_ Piston FE. Piston Piston Skill
P'STON Mode! DynamICSI TiiC‘L'DIOgv
_ A Ring Gas 0”
RING IDynamiA Dynamics Consumriiion
- TWIST

FIGURE 1. CASE structure

21

V
ELEMENT
ANALYSIS

FOR
PISTON
ELASTIC

—-

 

FIGURE 3. Piston program structure

CHAPTER 3 RING/BORE LUBRICATION AND WEAR

3.1 Introduction

The physical interaction between the piston rings and the cylinder bore of a
reciprocating engine is a tribological phenomenon that involves lubrication, friction
and wear of solid surfaces that slide against each other. Because the clearance
between the piston rings and the cylinder bore is typically small, often at the
length scale of micro or even nano level, the interaction of the surface
microstructures or asperities becomes a very important issue to be considered in
the study of cylinder-kit tribology. Despite the existence of lubricant, the ring/bore
interaction is not always ﬁlled completely with the lubricant and does not always
operate in the hydrodynamic lubrication regime. In contrary, it often operates in
the mixed lubrication regime and in the partially ﬂooded condition. A
computational model is presented in this chapter in an attempt to model this
particular tribological interaction. The computation involves the calculation of the
oil ﬁlm thickness, the hydrodynamic and the asperity contact pressures and the
wear at the ring/bore interaction at the possible existence of fully ﬂooded or
partially ﬂooded conditions. An analysis is also included to demonstrate the

usefulness of this model in the application of engine performance analysis.

3.2 Fundamentals in Tribology - Surface Measure and Characteristics

How is a surface measured? How is it represented numerically? Figure 4
shows the basic principle of a surface texture recorder and two commonly used
measured parameters - surface roughness and waviness. A real surface is often

complicated enough that two parameters are not sufﬁcient to characterize it

23

accurately. In order to represent a surface more correctly, additional statistical
parameters such as the probability density of asperities, the standard deviation of
the roughness height, the surface proﬁle across surface lay (distinctly directional
pattern due to machining) and asperity peak radius, are used. In many
applications, surfaces are often considered to have a symmetrical Gaussian
distribution in their microstructure (asperities), even though asymmetrical
characteristics do exist in common surfaces. A asymmetrical surface is more
complicated and involves additional surface parameters such as skewness -
measure of the symmetry of the proﬁle about the mean line, and kurtosis -
measure of shape (sharpness) of the amplitude distribution curve. The use of
statistical parameters, along with magnitudal parameters, allows a surface to be

represented mathematically for the tribological calculation.

 

 

    

   
  
 

I'\. O ‘
' I \‘\\’.’ ’ ‘2
I;’.’ ’, \\\\‘ o I...
\ O I
/ I. § ¢ 0,.

’.
.“‘."/‘ ’\o‘.’\.\
. Ix" ’9:
.l
/ ‘T§“o.‘.“\\l’.\\\\$e
J $31Q.’.:\. '\..:.\:~‘§ "
Do‘\-\ vv ‘00:“
“?O:.\:\\\‘\‘./A/‘ 4 . . "

 

 

   
   
  

 

’ U
.
’A‘.’A‘

t
A .

 

 

Roughness Waviness

FIGURE 4. Surface measurement and parameters

24

This equation requires statistical parameters of surface that can be
obtained from high precision surface measurement equipment [107]. Statistical
parameters can better represent a surface since most surfaces have complex and
randomly distributed geometrical structure. In engine studies, although most
interacting components begin with engineered surfaces, the tribological impact
after thousands of engine cycles may turn the surfaces into unpredictable forms.
The statistical parameters used in this equation are the combined surface
roughness o; the summit density n, and the estimated average asperity peak

radius of curvature ,B.

3.3 Wear Mechanism

When two surfaces come into a motion contact, a certain degree of wear is
expected to occur. There are three common types of wear mechanism - corrosive,
abrasive and adhesive. Corrosive wear occurs when there is a chemical
interaction of surface-to-5urface or the surface-to-contact medium at the surface
interaction. Abrasive wear occurs when a surface interaction results in the
existence of hard abrasive substances or particles (being able to be felt abrasive
to the ﬁngers) in a system, and the interaction causes scratching on a worn
surface. Adhesive wear occurs when a surface interaction produces none or
negligible abrasive substances and there is a tangential sliding of one clean
surface over another [66]. In adhesive wear, the existence of oxides and adsorbed
substances are usually ignored. Due to the complexity of accounting for chemical
interaction as well as the particle-to-particle and particle-to-surface interaction,

corrosion and abrasion are not considered in the calculation of ring/bore wear.

25

3.3.1 Adhesive Equation for Piston Ring/Bore Wear

According to Archad, the wear rate of two surfaces that slide against one
another is proportional to the sliding speed and normal contact load and inversely

proportional to the surface hardness. Eq. 1 shows the wear equation

 

WV)

w=k(; (n

where the wear coefﬁcient, k, is determined empirical solution. In the application
to the sliding interaction at piston rings and cylinder bore conjunction, the applied

load, W0, is the boundary contact force at the interaction; the sliding Speed, V, is

the magnitude of the piston reciprocating velocity, and the material hardness, H,
is the combined hardness of the interacting surfaces. The ﬁnding of all the terms

in Eq. (1) for a piston ring/bore wear analysis is rather a straightforward process

except the applied load term, Wu. The ﬁnding of Wa requires the solving of the

elastohydrodynamic mechanism between piston rings and cylinder bore, which is

presented in the following section.

3.4 Piston Ring/Bore Lubrication

Piston rings operate under a wide range of lubrication conditions. In most
cases, the lubrication can be categorized into the regime of hydrodynamic, bound-
ary and the mixture of the two, according to the popularly adopted Stribeck Dia-
gram. However, in the application of a reciprocating piston assembly system, two
potential oil transport conditions can emerge at the ring/bore interface, resulting in

a total of ﬁve lubrication conditions, which are

26

1.)Fully ﬂooded hydrodynamic,
2.)Partially ﬂooded hydrodynamic,
3.)Fully ﬂooded mixed,
4.)Partially ﬂooded mixed, and
5.)Full boundary.

The chance of occurrence for each lubrication condition depends on the
relative magnitude difference between surface roughness and ring/bore

clearance, and the sufﬁciency of oil for lubrication.

3.4.1 Hydrodynamic Lubrication

At the piston ring/bore sliding interface, oil ﬁlm that separates the contact
surfaces is generated by pressures arising from the shearing of a viscous
lubricant during their relative sliding motion. This kind of pressure is often deﬁned
as hydrodynamic pressure. One of the most extensively used formulas to solve for
the ring/bore hydrodynamic lubrication is the Reynolds equation. The classical

Reynolds equation is as follows:

a h 6p 3 h 6p ah h
3x(:d +dz(paz) _ 6Uax +6Waz +12V (2)

Consider the nature of piston ring/bore lubrication. The axial velocity of the
ring, U, is much higher than the rotational velocity, W; and at the ring/bore

clearance, the circumferential dimension is much larger than the radial dimension.

27

Thus,
8p 8p (3)

_.»__

3x 32

This unique circumstance allows the ring/bore lubrication to be considered

as a quasi-1 D problem, which can be represented by the following equation:

dh .
—a—h—QI3 = 6Ua—h-+ 12 ’2’” (4)

 

 

" = V is the squeeze ﬁlm motion.

In spite of its proven reliability, the classical Reynolds equation is
inadequate to address the effect of surface microstructure on hydrodynamic
pressure development because the surface roughness is not being considered.
For this reason, efforts to include surface roughness effect have been carried out
[82,83,122], among which the modiﬁed Reynolds model developed by Patir and
Cheng [82] has been widely adopted in the effort of modeling piston assembly
lubrication. The governing equation for the model is actually a modiﬁed version of
the classical Reynolds equation but is deﬁned in terms of Shear and pressure ﬂow
factors, which are functions of the ratio between surface roughness and nominal
clearance between two surfaces. It can be applied into the partially lubricated
interaction between two rough surfaces when the effect of roughness is most

crucial [82]. The partial differential equation is

3 a,» dh .
a h 3p _ 3h_ 5 mm
3;(¢.7,-;,— ‘6U5; 6U0'5;+127T ‘5)

28

where according to Zhu et al. [133], the simpliﬁed pressure and stress ﬂow factors

are
c». = [1-(‘912 (6)
A = 5,; (7)

3.4.2 Hydrodynamic Boundary Conditions

In general, the solution of piston ring/bore lubrication can be obtained using
one of three types of pressure boundary conditions: 1.) Sommerfeld, 2.)Half-
Sommerfeld, and 3.)Reynolds conditions. The boundary conditions of each type

can be expressed as the following respectively:

p=pll at x=O

8
p=p2 at x=b ()

p=p1 at x=O
p=p2 at x=b (9)
p=0 if p(x)<0

p=p and —=0 at x=0
1 dx (10)

p=p2 at x=b

The boundary conditions are illustrated in Figure 5.

29

Despite some shortcomings [119,90], Half-Sommerfeld boundary
conditions are the simplest to apply [90] and are widely used for obtaining a

reasonable hydrodynamic load estimation [119].

   

"\\\\\\\\\\\}\\\t\\\\\\\\‘

 

(a) Full-Sommerfeld Boundary Conditions

   

‘\\\\\\\\\\}\\‘\\\\\\\\\

 

(b) Half-So

B

merfeld Boundary Conditions

    

W\\\\\\\\}x\\\\\\\\\\‘

 

(c) Reynolds Boundary Conditions
FIGURE 5. Lubricant ﬁlm pressure and boundary conditions

30

3.4.3 Boundary Lubrication

Carefully deﬁned surface microstructure data is particularly important in the
computation of the boundary lubrication that occurs at the sliding contact between
two surfaces. The theoretical model describing the asperity contact of two
nominally ﬂat rough surfaces has been proposed by Greenwood and Tripp [45].
Even though the deformation of asperities can be both elastic and inelastic, it is
noted that plastic deformation is usually permanent after a surface has been run-
in, and it is reasonable to assume that an elastic effect is likely to be experienced
predominantly by surfaces during steady state operation. The nominal pressure
per unit area carried by asperities at the contact derived from the model is

expressed as:

 

 

 

pa(h)— —- 1—6——1ﬁ“(ol3n) E ,[3’F5(3) (11)
where
EC = (12)
[1—v12:1—v22)
.. §
2
F 5( 3) = TEEN—K 3) <1><§>d§ (13)
2 Rh
5

Integrating the nominal pressure in Eq. 11 over the surface in the sliding

direction results in the load carried by the asperities:

W=-1—1-—6ﬁ“(oﬁn ) E J33) [F53)d <14)

31

3.4.4 Ring Elastic Conformability in a Distorted Bore

In the piston ring/bore wear modeling, the primary step involved is solving a
non-linear elastic ring conforrnability inside a distorted cylinder bore. The details
of this computation can be found in the mixed lubrication model implemented in
RING [13]. The piston ring is discretized into a 91-node 3-D ﬁnite element beam
model (see Fig. 6), and the displacement of each node relative to the bore surface
is solved iteratively without violating the Spatial constraints within the piston
groove and distorted bore under the tensional load from the piston ring stiffness,
the gas pressure load from behind the piston ring, coupled with the hydrodynamic
ﬁlm pressure load from the lubricant and the contact load from the elastically

deformed asperities.

 

FIGURE 6. Ring structural FEA and oil lubrication model [34]

32

Convergence of the solution is reached when global and local force
equilibrium is achieved at each node with a predicted nodal displacement or
clearance between the node and the cylinder bore. Under the assumption of fully
ﬂooded ring pack lubrication, the clearance is also the same as the oil ﬁlm

thickness.

3.5 Starved Lubrication Analysis for the Top Ring

For a multi-ring piston assembly, the fully ﬂooded assumption is applied to
all piston rings underneath the top ring. In comparison to other piston rings, the
top compression ring is one of the least lubricated components in the cylinder-kit
assembly of an internal combustion engine due to its direct interaction with
combustion gas and extreme temperature environment. When the top
compression ring slides towards the top dead center, the only supplied oil to the
front of the ring face is a very thin layer of oil ﬁlm retained on the cylinder bore,
which is left from previous downward strokes. This thin oil layer is often
insufﬁcient to provide lubrication across the whole ring/bore interface. A common
approach in modeling a partially ﬂooded lubrication is through the ﬁnding of the
ﬁlm attachment location on the ring face where the hydrodynamic interaction
begins to develop so that the criteria of load equilibrium and ﬂow continuity are
satisﬁed. This approach is found to be useful for a hydrodynamic lubrication
model. However, for mixed lubrication, which comprises hydrodynamic and
asperity contact components, the common approach needed to be modiﬁed. This
paper introduces a modiﬁed computational approach that is capable of addressing

ﬁve possible lubrication conditions encountered by the top ring during its upstroke

33

motion. The three-dimensional feature of this model allows a variation of
boundary conditions to be considered across bore circumference and axial

direction.

This present work introduces a computational methodology to deal with the
problem of the mixed ring/bore lubrication by modifying an existing three-
dimensional lubrication model of CASE [13]. The existing model assumes that the
top compression ring is always operating under a fully ﬂooded condition. This
assumption will not be appropriate in the analysis for severe engine operating
conditions such as high engine speeds and loads when ring/bore metallic contact
and oil starvation is more likely to occur. To account for the potential occurrence of
partially ﬂooded condition at the top ring/bore interaction due to ﬁlm breakdown
during its upstroke movement, a numerical scheme is introduced. This numerical
scheme has the capability to account for ﬁve distinct lubrication conditions at the
top ring/bore interface. For secondary rings in a multi-ring pack system, the fully

ﬂooded assumption is applied.

3.5.1 Numerical Scheme

Developing a computational model for the top piston ring/bore lubrication
requires the solving of multiple systems prior to it, although in reality, all these
systems operate almost simultaneously. For this reason, several existing routines
of CASE are employed, which are the gas dynamics and the fully ﬂooded ring/
bore lubrication. The ﬂow chart in Figure 7 shows the fundamental structure of the

piston ring/bore lubrication system.

34

 

@put: Geometries & Operating Conditi@

l
E’iston Ring Pack Gas Dynamics]

Conformability of Piston Ring

Pack in Distorted Bore:

Generate Ring Stiffness Matrix and
Ring/Bore Clearance
I

@ﬁmate Initial Minimum on Film Thickneg
J

 

 

 

 

 

 

 

 

1
Ring/Bore Lubrication:
Fully Flooded Mixed Lubrication Analysis

 

 

 

[Check Forge Balance]

G“)-

gore Oil Film Thickness in Bore Spatial Domainﬂ
T
@eck h and ho during upstrokeg

 

 

 

 

 

 

 

 

<0.5h?

 

[ Generate New Film Proﬁle and hm," from h ]

Partially Flooded Mixed Lubrication Analysis

for Top Compression Ring:
Determine Minimum Oil Film Thickness for Pure Boundary
Lubrication, hb , using Iterative Procedure

Pure Boundary Lubrication:
Asperity Contact Dominates

Fh:0" FleL
Estimate Initial Film Separation
Posrtion on Ring/Liner Front Edge, x,

Solve for Hydrodynamic and Asperity]

 

 

 

 

 

V

 

 

 

 

 

 

Contact Pressures Calculate New h from
Rescaled Hydrodynamic Pressure Proﬁle

 

 

 

 

 

EZheck Force Balancg
{€30
+l
' it3 Check Flow Continuig

la.-.»

 

an.»
GED

FIGURE 7. Program flow diagram

 

 

35

The input for the model consists of engine geometrical parameters,
material properties and boundary conditions of a studied engine analysis.
Because this is a three—dimensional model, the variation along the circumference
of a piston assembly such as piston secondary motion (tilt), bore asymmetric
distortion and temperature, piston ovality, and ring twisting motion can be included
in the input. The gas ﬂow of the piston ring pack coupled with piston ring dynamics
is ﬁrst computed to generate ring pack pressures for all piston grooves and lands
for an entire engine cycle. The data in every crank angle increment are stored to

serve as boundary conditions for other analyses in successive simulation loops.

In a new engine cycle loop, the ring/bore conforrnability analysis is
conducted. Each piston ring is modeled as a three-dimensional FEA beam model
with 91 nodes along the circumference. A standard FEA method is employed to
obtain the ring stiffness and displacement of a piston ring in a distorted bore, ﬁrst
under dry lubrication, meaning that each piston ring node is only loaded with ring
tension from input and previously calculated groove pressure forces [13,34]. The
ring stiffness is due to the structural deﬂection of the ring as it is ﬁtted into a
distorted bore. Then, at the same time interval, the fully ﬂooded lubricated ring/
bore interaction is conducted where the lubricant ﬁlm loads due to hydrodynamics
and asperity contact are generated. The dry and lubricated ring/bore interactions
are coupled to obtain the effective clearance between piston rings and cylinder
bore using an iterative procedure. Under the fully ﬂooded assumption, the

effective clearance iS the same as the oil ﬁlm thickness. The effective clearance is

36

obtained when the nodal force equilibrium state at the ring/bore interface is

achieved by satisfying

where
FR : Fgas+Ftension (16)
and
FR=Fh+Fa+Fr (17)

If the computational time step falls within the piston upstroke duration
(compression and exhaust strokes) the partially ﬂooded lubrication for the top ring
will be performed. During downstroke, the top ring is assumed to have sufﬁcient
supplied lubricant from the inter-ring region to operate in a fully ﬂooded condition.
During its upward movement, the lubricant supplied to the top ring/bore interface
can be obtained either from an input of ﬁlm thickness of the cylinder bore, or from
the stored ﬁlm thickness left from the top ring in its previous downward stroke. The
ability to include measured bore ﬁlm thickness is particularly useful in the
application for the optical engine system constructed under the collaborative effort
between MSU and MMR, at the Michigan State University Engine Research
Laboratory, where ring/bore oil ﬁlm thickness over the entire bore surface can be
measured using the Laser—Induced Fluorescence (LIF) method. When the top

compression ring slides towards top dead center, it can encounter ﬁve possible

37

lubrication conditions. If the supplied oil ﬁlm thickness, 71, is less than twice the oil
ﬁlm thickness, ho, which is obtained from the initial fully ﬂooded hydrodynamic

pressure solution, the partially ﬂooded analysis will be performed; otherwise, the

fully ﬂooded analysis computed previously applies. If the oil ﬁlm thickness at the
top ring/bore interface, h(x), is much higher than the combined surface
roughness, Ra, the asperity pressure may be reduced to zero and the

hydrodynamic component becomes the sole lubrication load. Figure 8 Shows the

schematic representation of fully and partially ﬂooded hydrodynamic lubrications.

If h(x) is comparable with Ra, the mixed lubrication condition will occur since both

hydrodynamic and asperity contact components are involved in sharing the
lubrication load. Figure 9 shows the schematic representation of fully and partially

ﬂooded mixed lubrication.

In dealing with the partially ﬂooded mixed condition as opposed to the

partially ﬂooded hydrodynamic condition, an additional parameter is needed to

determine if pure boundary lubrication exists. As the new oil ﬁlm proﬁle h(x)*
across the ring/bore interface is now a function of h in the partially ﬂooded
condition, one must obtain a critical minimum oil ﬁlm thickness, hb, at which
asperity contact completely dominates in the lubrication load. The value of hb is
obtained using an iterative procedure by solving Eq. (11) until load equilibrium is
satisﬁed according to Eq. (15). When the new minimum oil ﬁlm thickness, hmmt‘,

is less than h b, pure boundary lubrication is assumed, as shown in Figure 10. The

38

oil ﬁlm thickness will be set to h b in order not to violate Eq. (15). It is to be noted

here that when surface roughness is involved in a lubrication analysis, the oil ﬁlm
thickness deﬁnition is no longer referred to as the thickness of lubricant, but the
nominal clearance between two surfaces, which is the combination of lubricant

thickness and asperity height.

When the partially ﬂooded lubrication occurs, a solution needs to reach
numerical convergence for two constraints: the lubricant ﬂow continuity in the oil
transport mechanism, and the load equilibrium in the elastohydrodynamic
interaction at each node of the top ring. For this reason, every cross section of
piston ring face across the circumference is discretized into 101 nodes as shown
in Fig. 11. This implies that at every time interval the computational task will
involve a total number of 91 x 101 nodes. The moving grid method is applied on

the meshing of the ring oil ﬁlm proﬁle in which the ﬁrst node is always at the
estimated separation point, xl , of the oil ﬁlm from the ring face. It is to be noted
here that the ﬁlm proﬁle at the ring/bore interface is dependent on the ring face
proﬁle and its orientation under the inﬂuence of ring twisting (0i) and piston tilting
(w) motion, and oil ﬁlm is assumed to attach constantly to the cylinder bore. The

101-node resolution is essential for the solving of Reynolds hydrodynamic

pressure through Eq. (5) as well as for the ﬂow continuity analysis.

39

 

 

 

 

 

 

 

   

No ‘
Asperity .
Contact
vX
'7
7 E _ x
Film <>g F < 0-5ho
Separation
A Location < h(X) > Re
0| Phr P 1%
Zn 3 PG :
’ 5
3 3 No
T I Asperity I
> Contact '
< _,
V IIX vX

 

 

 

(b)

FIGURE 8. (a) Fully and (b) partially ﬂooded hydrodynamic lubrication

40

j]
"T

CYLINDER BORE

Film
U Separation
A Location

   
 

 

 

 

 

(b)

b b
FLUB- Ph P,3
Xi
75 < 0.5ho
h(x) ~ R,3
P,, Pk
ix, i. '
9}? 21>
0L”.
3' 0x ﬁ
:j . I
[x x

 

FIGURE 9. (a) Fully and (b) partially ﬂooded mixed lubrication

41

   
   

b
F...=[ P.
0

 

 

F < 0.5ho

h(x) < R,
. Phr \ P‘!»
No LE:
wgagic (2:?)
Pressure ‘:>

 

 

 

 

 

 

 

 

 
 

» ..... ~ “VFW???” " W
W‘ L L L LL L L L L L L LL L L LLLL (D)?t W5 .Lwﬁﬁﬁiﬁswew >375?
WWVKWA yBCAAAAA WNW A A A A‘A‘mﬁ
FIGURE 11. Ring face discretization at (a) fully ﬂooded and (b) partially
ﬂooded condition

 
 

42

3.5.2 Flow Continuity

To ensure ﬂow continuity, lubricant ﬂow across the top ring/bore front
interface is treated as the combination of Couette and Pouiseille ﬂows. Consider
Figure 8 that shows a schematic drawing of hydrodynamic lubrication, and a
common Reynolds hydrodynamic pressure distribution along the ring face that

can be obtained through the numerical solving of Eq. (5). If the piston ring and the
bore are treated as two bodies sliding at a relative velocity of U, at the point
where ap/ax = 0, only Couette ﬂow exists, and the ﬁlm thickness is designated
as ho at the location. A general Couette ﬂow equation as shown in [77] is

= ._y_ Lip 2_
u Uh(x)+2u(8x)(y mm (18)

In the case of ap/ax = 0 at h 0, the equation above simply reduces to

_ y
u — U — (19)
110
The average ﬂow past ho can be represented by
U
q 2 5’10 (20)

Assuming steady oil ﬁlm that enters the ring/bore front edge, designated as

h, the average ﬂow is given as

21 = Uh (21)

Since c} = q for continuous ﬂow, it results in

h (22)

Without considering the details of how the oil transport evolves across the
ring/bore interface, the ﬂow continuity in this numerical scheme is ensured
through the satisfaction of the condition in Eq. (22). The condition in this equation
is also used to check for oil sufﬁciency for the development of hydrodynamic

pressure across the entire face of the piston ring. If the lubricant ﬁlm on the
cylinder bore 72 is greater than or equal to 0.5110, a fully ﬂooded condition will

exist. Otherwise, sufﬁcient lubricant will not be available to the ring to develop fully
the Reynolds hydrodynamic pressure across its face. This is referred to as
lubricant starvation or partially ﬂooded condition. An assumption used in the
model for the simpliﬁcation of computation is that the second land between the top
and the second rings is always in ﬂooded condition. The mass conservation
scheme at the trailing edge can be included in the later improvement of this

model.

3.5.3 Load Equilibrium

The insufﬁciency of the supplied oil to the ring/bore interface will result in
the reduction of oil ﬁlm thickness in the partially ﬂooded condition. In order to
ensure load equilibrium in a starved condition, the following procedure is

employed:

44

1. Consider that when lubricant starvation occurs, ho should be set

twice the 7: value according to Eq. (22) in order to ensure oil ﬂow

continuity.

2. In the numerical solution of Equations (5) and (11), lower
minimum oil ﬁlm thickness will result in the increase of the
lubrication force (FL), which is greater than the applied force (FR). In
order to reduce the lubrication force, one way is through the
correction of the hydrodynamic component by adjusting the location
at which the hydrodynamic pressure begins (ﬁlm separation point at
ring face). The location is moved from the leading edge of the ring to
a location some distance from the leading edge (x=x,) and the value
of x, is determined by adjusting it until Eq. (15) is satisﬁed. This
adjustment will not affect the asperity component because the
asperity contact equation is indeed only a function of the ring/bore
clearance regardless of whether it is fully ﬂooded or partially

ﬂooded.

3. The new ﬁlm proﬁle h(x)* will be rescaled beginning with x1, and

the location of the maximum pressure where ap/ax = 0 is again
determined. This location will be different from where it was

originally calculated as illustrated in Figures 8 and 9.

45

4. The new minimum oil ﬁlm thickness is adjusted by 6h within

bounded criteria of pure hydrodynamic and pure boundary

lubrication, and once again, h is set twice the 72 value according

to Eq. (22). The process in steps 2 and 3 is repeated until Equations

(15) and (22) are satisﬁed simultaneously.

3.6 Model Application

In order to demonstrate the potential of this present model in Simulating
various lubrication conditions, it is applied to a hypothetical engine operating
condition. A single-ring piston assembly system utilizing the measured cylinder
pressure trace of an engine, which operates at 2900RPM rated running condition,
is selected for this analysis because it provides unique boundary conditions at a
particular time interval for the instantaneous emergence of various lubrication
conditions across the ring circumference. The impact from piston secondary
motion (tilting) is also included to create an asymmetrical ring/bore interaction
across ring circumference. The selected top compression ring has a barrel face
proﬁle, and the cylinder bore oil ﬁlm thickness exposed to the combustion
chamber is maintained at 2.5 micron. This bore oil ﬁlm thickness value will serve
as the supplied oil to the front edge of ring/bore interface during piston upward
strokes. The ring surface coordinate system shown in Figure 12 is employed in
the plotting of the results. The ring circumferential direction begins from the thrust
side, and the normalized axial direction begins from ring edge that faces the

combustion chamber.

46

The results shown in Figures 13 to 18 are captured at crank angle of 348
degree when the ring is approaching TDC during the compression stroke. Figure
13 illustrates the ﬁlm attached/detached characteristics at the ring surface. The
partially ﬂooded condition is observed across the ring front edge. At about 78,
136, 220 and 284 degrees across ring circumference, the partially ﬂooded
condition extends across the entire ring width. Observation from the oil ﬁlm
thickness result in Figure 14 shows that these areas are where the oil ﬁlm
detaches from the ring face, and minimum lubrication is maintained merely by the
surface roughness, which deﬁne the full boundary condition. On top of that, there
are four regions fully ﬁlled with lubricant at about the circumferential degrees of
60, 120, 240 and 300. The circumferential variation of the lubrication condition is
due to the piston tilting effect as well as the conformed ring tension pressure
pattern. It is also noted that the variation of the ﬁlm attached/detached proﬁle
follows a rather symmetrical pattern across ring circumference, which illustrates

the symmetrical effect from the piston tilting motion about the piston pin axis.

‘
.................
o "'
0"

I 'e
.........
e

u
O Q ‘
......

'''''
. s“.

I
.u’ 1" .O-II‘ . .
- aaaaa
" nuge"..'-’ c -..".'-o‘
0.. .I~
.0. ".
a u
0' n.
O I
..0 ~..
’.' 0'
o

       
  
  

L Ring Circumferencem": .L
EG

Normalized
Width

FIGURE 12. Ring surface coordinate system

47

Normalized Ring Width

mm

It

It“

m: Oil Filled Region

      

SSS

SEPARATION P

I

w
(0
'—

N
O)
‘—

 

ROFILE

J "‘ggzaﬁﬁt‘ﬁgypwi < ”fair 5
it
. , 3 _

Ring Circumference DEG

33 Dry Region
AT 348 CAD

O
(D
0')

FIGURE 13. Oil ﬁlm distribution across top ring surface near TDC

Nonmllzed Rlng
Width 0

  
    

80

Ring Clrcumterence DEG

 

on Film Thickness m

I 2.60E-02-280E-02
I 2 40E-02-2.60E-02
I 2.20E-02-2.40E-02
I 2 00E-02-2 2OE-02
I 1,80E-02-2.00E-02
I 1 60302-1 80E-02
I 1 40E-02-‘l 60E-02
I: 1 20502—1 40E-02
D 1,00E-02-120E—02
Cl 8 00E-03—1OOE-02
I 6.00E-03-8.00E~03
I 4.00E-03—6 DOE-03
I 2.00E-03-4 ODE-03
I 0 00E+00-2 ODE-03

 

FIGURE 14. Oil ﬁlm thickness magnitude across top ring surface near

TDC

48

 

Figure 15 shows how the hydrodynamic and asperity contact pressures
Share the role in supporting the lubrication load at 348 degree CA. Except at the
full boundary regions, (about circumferential degree of 78, 136, 220 and 284), the
total lubrication load is divided near-equally between asperity contact and
hydrodynamic pressures. At the full boundary regions, asperity contact fully
supports the lubrication load. From the oil ﬁlm separation proﬁle in Figure 13, one
can observe that there are three lubrication conditions existing simultaneously at
the top ring/bore interface at 348 degree CA. Table1 gives a summary of this
observation. Not shown in the results are the fully and partially ﬂooded
hydrodynamic conditions, which are similar to mixed condition except without the
contribution of asperity contact to the lubrication load when the oil ﬁlm thickness is
an order higher than the asperity height. The present analysis can be easily
altered by using a higher value of the bore oil ﬁlm thickness as an input in order to

simulate pure hydrodynamic conditions.

TABLE 1. Ring/bore lubrication types at 348 degree CA

 

 

 

 

 

 

 

Condition Region across Circumference (DEG)
Fully Flooded Mixed 52-68, 108-128, 232-252, 292-308
Partially Flooded Mixed 0-48, 88-104, 144-216, 256-272, 312-360
Full Boundary 72-84, 132-140, 220-228, 276-288

 

49

 

 

LUBRICATION LOAD SHARING O/OAspgmy Contact
at CA 348 DEG (Near TDC during Compression Stroke) — . - %Hydrodynamic

1°” ’ H D D Fl

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

80% i'
a
C
'C
0
0:)
“u 60% F
B
O
o A." .
& .
o h .
g 405 '
3 '.
8 l
O. .
20% v I
l
0% '
0 60 1 20 180 240 300 360

Bore Circumference DEG

FIGURE 15. Asperity contact and hydrodynamic sharing lubrication load

Figures 16 and 17 Show the hydrodynamic pressure distribution on the ring
surface at 348 degree CA. At this crank angle, the hydrodynamic pressure
gradient across the ring width is large due to high compression pressure in the
combustion chamber. Breakdown of the hydrodynamic pressure is found at the
full boundary lubrication region as mentioned in the previous section. Three
hydrodynamic pressure proﬁles are selected from Fig. 16 and plotted in Fig. 17.
They are typical Reynolds hydrodynamic pressure proﬁles. It is noted here that a
half-Sommerfeld boundary condition is adopted, in which the front edge pressure
is speciﬁed with the cylinder pressure and the trailing edge is with the sump
pressure. Figure 17 illustrates more clearly the ﬁlm separation effect on the

development of hydrodynamic pressure. At the circumferential location of 0 and

50

24 degrees, the normalized ﬁlm separation points (x ,/b) at the ring front edge are

0.05889 and 0.03043 (partially ﬂooded). The separation point reaches the ring
front edge at 52 degree (fully ﬂooded). In numerical computation, the separation
points mark the ﬁrst integration point for the calculation of the lubrication load due
to hydrodynamic pressure. Thus, only the ring/bore region that is fully ﬁlled with oil
ﬁlm (not ﬁlm separation) will be accounted for in sharing the lubrication load.
However, the ﬁlm separation condition will not affect the asperity contact load
calculation because asperity contact pressure is a function of the effective ring/
bore clearance instead of the hydrodynamic oil ﬁlm, according to Eq. (11). Figure
18 shows the asperity contact pressure distribution across the ring surface. The
asperity contact pressure peaks at the full boundary lubrication region and
concentrates around the axially central area across the ring face. This is because
the barrel proﬁle of the ring face provides gradually increased clearance away
from the middle of the ring surface in the axial direction. In Spite of a tiny change
of the ring face proﬁle across ring width, (on the order of 0.05 micron per nodal
point; see Fig. 11), the change of the clearance-induced asperity contact pressure
is quite large near the central area across the ring width. This large change is
caused by the non-linear characteristic of the function used in the calculation of

the asperity contact pressure as given by Eq. (11).

51

Hydrodynamic
Pressure kPa

 
  
 
     
 
       
      
 
 

  

I9.DUE+03—1.05E+04
I7.SDE+03-9.CICIE+03
I6.UUE+U3-7.5UE+O3
I4.SUE+03-5.UDE+03
D3.DUE+03-4.50E+Cl3
El 1 .50E+D3-3.UDE+O3
DODGE-{i134 .SDE+03

 

‘ ., Ring Circumference DEG

0.60 . .

Normalized Ring 0 80
Wk“ 1.00

FIGURE 16. Hydrodynamic pressure distribution across top ring surface

.- s

.
s

------------

 

Hydrodynamic Pressure (kPa)

Front Edge Separation Points
:1 0.05889 at Circ. 0 deg
I 0.030432 at Circ. 24 deg
e 0.0 at Circ. 52 deg

     
 

HYDRODYNAMIC PRESSURE PROFILE
- - - - Ring Circumference Location. 0 DEG
\ — ~ - Ring Circumference Location, 24 DEG
7'1. - - -Ring Circumference Location. 52 DEG
it). — Ring Proﬁle

..\'

   

 

A i I J I I

0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10
Normalized Ring Width

FIGURE 17. Hydrodynamic pressure and front edge film separation point
at three distinct bore locations

52

 

I9.00E+D4-1.05E+05
I7.50E+04-9.00E+04
IB.DUE+D4-7.50E+04
I4.50E+O4-6.00E+04
D3.0UE+D4-4.SUE+U4
D1.50E+04-3.00E+04
IU.DUE+I]J-1.SUE+IJ4

   
   
 

Asperity Pressure kPa

 

 

 

4.50E+04

    
 
 
  
 

120
80

0.40 L ‘ - ' 40 Ring Circumference DEG

0.60 L ‘1
Normalized Ring 0.80 ‘
Width 1.00

FIGURE 18. Asperity contact pressure distribution across top ring
surface near TDC

53

CHAPTER 4 OIL EVAPORATION AND STARVED LUBRICATION

4.1 Introduction

The piston ring pack in an internal combustion engine does not always
operate with sufﬁcient lubrication. The causes of lubricant insufﬁciency can be
attributed to the hostile combustion environment, improperly engineered surface
microstructure, poorly designed ring face proﬁle, insufﬁcient lubricant wetting
system, excessive temperature-induced oil consumption [3,18,19,28,125,] etc.;
and the consequences such as wear and scufﬁng can signiﬁcantly shorten the life
of engines. For a multi-ring pack system, the top compression ring is known to be
the most impacted ring due to its excessive thermal and mechanical stress and

poorly lubricated state [113,114].

Many efforts that include partially ﬂooded analysis to account for better
physical representation of the piston ring/bore lubrication have approached it from
the aspects of ﬂuid mechanics [13,43,102,113]. A common approach in modeling
the partially ﬂooded lubrication is through imposing the inlet oil ﬁlm thickness as
the boundary condition at the ring/bore conjunction, and ﬁnding the location of the
oil ﬁlm attached/detached edge in order to satisfy both ﬂow continuity and force
equilibrium. In many studies, the inlet oil ﬁlm thickness for the top compression
ring during its upward stroke is either an input from measured data, or from the oil
ﬁlm layer left on the cylinder bore from its previous downward stroke, or from a
constant value slightly higher than the surface roughness. These studies seldom
accounted for the thermal effect from the temperature of the cylinder gas and bore

wall on the oil ﬁlm layer, which is exposed to the environment in the combustion

54

chamber. Bore temperature is accounted for mostly in the oil viscosity calculation
while the cylinder gas temperature is not taken into account in the piston ring/bore
lubrication. However, it is believed that the combustion gas and bore temperature
do play greater roles in inﬂuencing the piston ring/bore lubrication through the
evaporative effect on the piston ring pack oil transport system [4]. This model
attempts to simulate a more realistic condition faced by the top compression ring
by incorporating heat and mass transfer process in the computational analysis of

partially ﬂooded piston ring/bore lubrication.

4.2 Engine Oil Evaporation

In most applications, oil consumption due to lubricant evaporative process
is inevitable due to high pressure and temperature gradient conditions in the
combustion chamber [3,18,28,125]. Even though the top compression ring usually
has very high sealing pressure that controls the combustion gas from escaping
into the engine crank case, and helps in keeping the oil ﬁlm thickness at a minimal
level to minimize oil loss, the ulbricant can still ﬁnd paths into the combustion
chamber. As a surface is never smooth, the peak-and-valley proﬁle of a lubricated
surface provides potential venue for the lubricating oil to retain. In the cylinder-kit
assembly, a thin layer of oil is retained on the cylinder bore surface during the
downward stroke of the top ring. This oil layer can become an important
lubrication supply for the top compression ring during its return trip. However, this
oil layer is also subjected to the hostile environment that encourages rapid heat
and mass transfer. If the oil evaporative rate is high, the time interval between two

successive strokes then provides signiﬁcant exposure time for the oil layer to

55

evaporate, and results in a severe oil starvation condition for the top compression

ring.

The oil evaporation model included in this present work is an extended
effort from the authors' previous work [18,19], but with modiﬁcation to incorporate
a three-dimensional bore temperature distribution; thus accounting for local
variation of the heat transfer process across the bore circumferential and axial
directions. Single-specie oil is assumed in the model due to limitations of oil
chemical properties data available (11-Methylnonacosane -C30H62 is used in the
calculation of the oil vaporized pressure). The oil evaporation process is a function
of mass ﬂux, engine oil density, and the area of cylinder bore being exposed to the
combustion chamber. The instantaneous evaporation rate at a local cylinder bore

area can be calculated using the equation:

 

_ evap
AQevap _ (23)

The cylinder bore temperature is a key determining factor in the process of
oil evaporation [3,18,125]. Consider the thin oil ﬁlm thickness left on the cylinder
bore, which is on the order of microns; the ﬁlm surface temperature is very close
to the cylinder bore temperature. The ﬁlm surface temperature is a key factor in
the formation of oil vapor for the evaporation process to occur [126]. Hence,

increasing bore temperature will speed up the evaporation process.

Not only does the oil evaporation process affect engine emissions, it also

inﬂuences the lubrication of the piston ring pack, speciﬁcally of the top

56

compression ring. As illustrated in Figure 19, the oil layer at the front of the top
ring during its upward movement serves as an oil supplier for its lubrication. This
oil layer plays a role in assisting with the building up of hydrodynamic pressure
that prevents metallic contact and scuffing. When being exposed to the
environment of the combustion chamber, the oil ﬁlm thickness begins to deplete,
causing the top compression ring to be more susceptible to encounter mixed

lubrication or even dry lubrication during its upward movement.

 

Tcylinder

P cylinder Phydrodynamic
Top Piston
Ring ’ '

FL
FR M—
_V_...
U

 

 

FIGURE 19. Top piston ring in upward stroke

The oil ﬁlm thickness encountered by the top compression ring during its

upward movement determines the boundary condition for the lubrication regime

57

the contact interaction will experience. The well recognized Stribeck Curve in
Figure 20 illustrates this point. At sufﬁcient oil supply, the oil ﬁlm thickness h is an

order of magnitude higher than the surface roughness Ra resulting in the surface

interaction at hydrodynamic regime. When the amount of lubricant supplied to the

contact decreases, h will decrease to an order of magnitude comparable with R0,

and as a consequence, interactions between the asperities of the surfaces occur,

resulting in mixed or boundary lubrications. As a result, friction and wear increase.

In fact, the piston ring/bore contact always operates in mixed lubrication
condition due to the fact that the oil ﬁlm thickness between piston ring and bore is
on the same order of magnitude as the surface roughness in most engine

operating conditions, particularly near TDC and BDC. At the piston ring/bore

interface, as Shown in Figure 19, both the hydrodynamic pressure, Ph, and

asperity contact pressure, P0 are sharing the lubricating reaction load in

balancing the applied load due to the gas pressure behind ring groove, the ring
tension and the ring stiffness due to ring/bore confomiability. The integration of
the hydrodynamic and asperity contact pressures over the ring face results in the
lubrication reacting load. However, in the partially lubricated condition, there exists
ﬁlm separation near the front edge of the ring face, and this results in a smaller
surface area available for the development of hydrodynamic pressure. This also
implies that the asperity contact interaction may play a greater part in the
lubricating load if the minimum oil ﬁlm thickness is comparable with the surface

roughness. The distance from the front edge of the ring to the separation point is

58

denoted as x], and it is dependent on the amount of lubricant supplied to the

contact, h. In the computational approach, xl can be determined numerically

while solving for hydrodynamic and asperity contact pressures using the Reynolds

and asperity contact equations. Convergence is achieved when, both locally and

globally, FL is equal to FR and the ﬂow continuity is also satisﬁed. The solution,

which consists of hydrodynamic and boundary components, can serve as input for
further calculation of friction and wear at the piston ring/bore interface. The piston
ring/bore wear model is integrated in the lubrication model, and it is a direct
application of Archard’s equation [2] for mechanical adhesive wear between two

surfaces.

4.3 Computational Methodology

The cylinder-kit analysis of an engine system consists of two basic
components: cylinder gas dynamics and tribology. In the application of this model,
CASE is used to provide essential data for the calculation of this model. Referring
to the ﬂow chart in Fig. 21, the input speciﬁcations are the geometrical dimensions
and the operating conditions of an engine. Within CASE, the structure of its
lubrication system is rearranged to accommodate the oil evaporation model. The
lubrication computation is divided into two time domains, downward stroke and
upward stroke. The oil evaporation model is employed as a function to compute
oil supply for the piston ring/bore interface under the effect of oil evaporation
during upward stroke. Using the gas dynamic model of CASE, the boundary and

initial conditions such as the piston ring pack pressure, piston and piston ring

59

motion are calculated and adopted as the input for the computation of this
modiﬁed lubrication system. The computational cycle of a four-stroke engine
system begins with the intake stroke (0 - 180 degree crank angle (CA)) and ends

with the exhaust stroke (540 - 720 degree CA).

 

 

 

Boundary Lubrication

 

Mixed Lubrication

 

Friction

       

Clearance L
Surface Roughness

Viscosity RH

Relative Velocity Hydrodynamic Lubrication
Contact Load

Sommerfeld Number, Z(nN/L)
FIGURE 20. Stribeck curve

r23 ”5"

 

4.3.1 Downward Stroke

During the downward stroke (intake and power strokes), the piston ring
pack is assumed to operate under the fully ﬂooded condition. At each time
interval, the minimum oil ﬁlm thickness along the circumference of the top
compression ring is recorded into the cylinder bore spatial domain, and treated as

a layer of oil ﬁlm retained on the bore wall. The spatial domain of the cylinder bore

60

is represented by a system of nodes, where each node contains lubrication
information such as surface roughness, bore distortion and temperature, retained
oil ﬁlm thickness, heat transfer coefﬁcients, etc. At successive time intervals after
the passing of the top compression ring, nodal oil ﬁlms that are exposed to the
combustion chamber will begin their evaporative process. During this phase, the
nodal oil ﬁlm thickness will deplete as a result of oil loss calculated using Eq. (23).
The evaporation process continues as long as the node is exposed to the

combustion chamber.

4.3.2 Upward Stroke

After the piston reaches bottom dead center (BDC) and returns in upward
motion (compression and exhaust strokes), the oil ﬁlm thickness of the nodes that
were exposed to the combustion chamber will become the oil supply to the ring/
bore interface. The amount of oil encountered by the interface will be used in the
determination of partially ﬂooded mixed lubrication.—-Those nodes that have
entered the piston ring/bore interaction will then be replenished with new oil ﬁlm

thickness data during the coming downward stroke.

Over an entire engine cycle, the spatial domain of the cylinder bore records
the cumulative wear data, which is a resultant phenomenon due to partially

ﬂooded lubrication under the inﬂuence of oil evaporation.

4.4 Model Application

In general engine application, there exists a vital relation between the ring/

bore lubrication and oil consumption [53]. This model is applied in a diesel engine

61

analysis to demonstrate the importance of oil evaporation effect in the analysis of
partially ﬂooded lubrication, and to study and understand a phenomenon that
could potentially exist in an operating diesel engine. Modern diesel fuel injection
systems are able to operate over a very broad range and this allows fuel injection
events to be timed throughout the compression/expansion process. It is well
demonstrated that there are potential adverse consequences in certain areas of
fuel injection operation, some of which can cause increased cylinder bore wear.
The phenomenon of “bore wall fuel wetting” is a well-known example. Given
concerns around the potential for wear-inducing interactions between the fuel
injection plumes and the cylinder bore wall, a particular interaction is being
studied and analyzed in this work. The interaction shows what impact an imposed

local heating of the bore wall can have on the cylinder bore wear.

The basic engine speciﬁcations and operating condition are given in

 

 

 

 

Table 2.
TABLE 2. Engine speciﬁcations
Engine Model 5.9L I6 Diesel Engine
Bore Diameter 102.2 mm
Engine Operating Condition 2900RPM Rated
Ring Pack Two Compression Rings and One

Oil Control Ring

 

Surface Roughness:
Top Ring 0.08 micron
Bore Wall 0.33 micron

 

 

 

 

62

 

Input:

Engine System
Conﬁguration

I

 

 

 

Downward
Stroke?

 

F lily
Flooded
Mixed
Lubrication

I

[StoreT Oil Film

 

Thickness Le
By Top R'ng

 

 

 

 

[Oil Evalporatiorﬂ

I

 

Upward
Stroke?

 

Input:d
Evaporate

Oil Filmd
Thickness
Partialy n]

 

Flooded
Mixed
Lubrlisetio

 

 

J

FIGURE 21. Program ﬂow chart

In this application, the cylinder bore temperature is an external input to the

model. The original bore temperature is obtained from experimental measurement
of a diesel engine operation. Under the assumption that the fuel injection plumes

in an operating diesel engine are additional heat sources that contribute to the

63

increase of local bore wall temperature, the original bore temperature proﬁle is
modiﬁed through the imposing of higher than surrounding temperature at ﬁve
distinct locations near TRR on the cylinder bore wall as illustrated in Figure 22. In

this example, the peak temperature of the ﬁve distinct locations (being referred as

hot Spots in later discussion) is increased from 180°C to 280°C.

Two discussions of the application will be presented. The ﬁrst discussion is
on the comparison between various lubrication models and how their assumptions
affect the bore wear results. The second discussion is on the sensitivity of the

bore wear result to the temperature of the hot spots, in which the peak
temperature of the ﬁve hot spots will be increased by 10°C, 25°C, 30°C, 40°C,

50°C, 60°C, 75°C and 100°C.

4.4.1 Fully Flooded and Various Partially Flooded Lubrication Analyses

Under the same operating conditions and bore temperature, the bore wear
results shown in Figures 23-25 and 27 demonstrate that results can vary under
different lubrication assumptions. The bore temperature used in simulating these

results is the same as the example shown in Figure 22. The hot spot peak
temperatures are speciﬁed about 100°C higher than their surroundings. The
engine thrust side is at 0° on the bore circumference and the hot spots are at 35°,

105°, 180°, 250° and 325° from the thrust side at the same axial location. The
axial position of the hot Spots is around the top ring reversal (TRR) location, which

is about 9.28 mm in actual scale or 7.75 units in the nondimensional scale in

Figure 22.

64

Cylinder
Bo

  

Combustion
Plumes 7.x 7g

 

 

 

Bore Temoerature Distribution. de. C

 

Nondimensional Axial Location

 

270 360

1 80
Circumference Location DEG

FIGURE 22. Cylinder bore temperature distribution

In all cases, as Shown in Figures 23-25, and 27, higher local bore wear is
predicted near the top, the second and the oil ring reversal at both TDC and BDC
as is agreeable with ﬁndings from other works [53,67,119]. Among the results, the

fully ﬂooded lubrication analysis predicts the lowest bore wear magnitude near

65

TRR. The fully ﬂooded analysis assumes that the top compression ring/bore
conjunction is operating at fully ﬂooded condition throughout the entire engine

cycle.

Figure 24 provides the bore wear result from the partially ﬂooded
lubrication analysis without the consideration of oil evaporative effect. In this
analysis, the oil ﬁlm layer, which is left by the top ring during downward movement
and exposed to the combustion chamber environment, remains the constant and
is used on the successive upward stroke as the input for the partially ﬂooded
lubrication calculation at the piston ring/bore interface. It shows only a slight

increase of bore wear over the result in Figure 23.

Figure 25 shows the bore wear result from the partially ﬂooded analysis
that assumes a constant oil ﬁlm thickness (2 microns) encountered by the top ring
during its upward stroke. The value of 2 microns, which is about 6 times the bore
wall surface roughness, is chosen because the top ring/bore interface is operating
within the range of about 2 micron of oil ﬁlm thickness over the entire cycle as
shown in Figure 26. This analysis predicts substantially higher bore wear near the

TRR along the entire bore circumference.

Without considering the evaporative effect, the results in Figures 23 - 25
fail to capture the greater effect of hot spot temperature on the bore wear. Careful
examination of these ﬁgures shows that there is a small degree of bore wear
increase around the hot spots, which is caused by the oil viscosity reduction in

high temperature. According to the common rule of oil viscosity, temperature

66

increase will result in the decrease of oil viscosity. The direct impact of the
decrease of oil viscosity on the lubrication calculation is the decrease of
hydrodynamic pressure around the hot Spot area. As a result, the oil ﬁlm around
the hot spot area will be reduced when the ring conforms to the cylinder bore,
which in turn raises the asperity contact pressure in the lubrication load sharing,
resulting in the increase of wear around the hot spot areas. It is also known that
lower viscosity will result in less squeezing force from the oil around ring reversal
and thus greater asperity contact force is generated to support the applied load on

the ring in the radial direction [113].

BORE WEAR DISTRIBUTION (+100 C Hot Spot Temperature Full Flooded
7 1:1 3. 5015-10-4. 005-10
I 3.00E-10-3.50E-10
I 2.50E—10—3.00E—10
I 2.00E-10-2.50E-10
El1.50E-10-2.00E-10
D1.00E-10-1.50E-10
I 5.00E-11-1.00E-10

  
 
 
 
  
 
  
 
     
 

4.00E-10
3.50E-10
3.00E-10

 

   

2 2 50E-10 I 0.00E+00-5.00E-11
E .
g 2.00E-10
B 1 505-10
1 00E-10
360
5.00E-1 1 300
0 00E+00
Bore
Circumference
O
‘0 8 6° DEG

Nondimensional Axial Location

FIGURE 23. Bore wear (fully ﬂooded lubrication)

67

BORE WEAR DISTRIBUTION (+100 C Hot Spot Temperature
Starved. No Evaporation) D 3. 50E-10-4. ODE-10

I 3.00E—10-3.50E-10
I 2.50E-10-3.00E-10
I 2.00E-10—2.50E-10
EI1.50E-10-2.00E-10

    
 
 
   
  
  
   
  

4.00E—10

3.50E-10

 

u 1 0015-10-1 sue-10
300510 I 5.00E-1 14.00510

2 2 505-10 I 0.00E+00-5.00E-11

2

3 2 00E-10

3 1 5015-10

180
120

   

Circumference
DEG
w 0
Nondimensional Axial Location a: 0

FIGURE 24. Bore wear (partially ﬂooded, no evaporation)

     
 
 
 
 
 
 
    

BORE WEAR DISTRIBUTION (+100 C Hot Spot Temperature
2 um Supplied Oil Film) -' . El 3. 50E-10—4. DOE-10

I 3.00E-10-3.50E-10
I 2.50E-10-3.00E-10
I 2.00E-10-2.50E-10
1 .50E-10-2.00E-10
Cl 1.00E-10-1.50E-10
I 5.00E-11-1.00E-10

 

2 2 505-10 l0.00E+00-5.00E-11
2
§ 2 00E-10
3 1 50E-10
1 00E-10
360
5.00E-11 300
0 00E+00
Bore
Circumference
DEG
3 a o
Nondimensional Axial Location 0

FIGURE 25. Bore wear (partially ﬂooded, 2 microns supplied oil ﬁlm)

68

300 .... a... .. ,. .. _.,,,__.._ wmwgmm _--___ ________. ._ ,
Circumferential-Averaged Top Ring Oil Film Thickness

//L‘\\ \

 

N N
8 8
L

\\
/ “I
/
<\
\ \
\\\
8
\
\
I/

 

 

.1.
C
O

 

Oil Film Thickness MM
211 .
o

 

0.50

 

0.00 1 4 1 1 1 1

O 90 1 80 270 360 450 540 630 720
Crank Angle DEG

FIGURE 26. Cyclic top ring oil ﬁlm thickness

 

By including the evaporative effect in the partially ﬂooded lubrication
calculation, the result in Figure 27 shows a signiﬁcant difference from previous
results. Distinctive pockets of high wear are observed at the hot spots while the
surrounding areas along bore circumference at TRR have about 25% less wear.
This implies that the local effect of bore temperature variation on the bore wear is

captured in this analysis.

69

BORE WEAR DISTRIBUTION (+100 C Hot Spot Temperature)
E1 3.50E-10-4.00E-10
I 3.00E—10—3.50E-10
2.50E—10-3.00E-10
I 2.00E-10-2.50E-10
D1.50E-10-2.00E-10
E1 1.00E-10-1.50E-10
I 5.00E-11-1.00E-10
El 0.00E+005.00E-11

 

 
 
 
  
 
  
   
  
 
 
  

4.00E-10
3.50E-10
3.00E-10

 

 

2.50E-10
2.00E-10
1.50E-10
1.00E-10

Wear MM

360
5.00E-1 1 300
0.00E+OO . 240
Bore
Circumference
DEG

Nondimensional Axial Location 00 8 §
FIGURE 27. Bore wear (partially ﬂooded, with evaporation)

The importance of including the evaporation analysis can be explained by
studying the evaporation rate of oil ﬁlm in the bore spatial domain. Figure 28
shows the change of oil ﬁlm thickness along the bore circumference at TRR over
an entire engine cycle. At 0 and 360 degree CA, the nodal oil ﬁlm thicknesses at
TRR are replenished with oil left from the top ring during its downward movement.
As time advances, the nodal oil ﬁlm thicknesses across bore circumference
experience depletion in magnitude as a result of evaporation. The depletion rate is
particularly notable at the hot spots. With the temperature 100°C higher than their
surroundings, hot spots cause rapid evaporation of oil ﬁlm. During the intake/

compression process, there is no oil ﬁlm left at hot spots after a few degrees of

70

crank angle while the non-hot spot locations experience a relatively slow
evaporation rate. This implies that when the top compression ring returns to TRR
at the end of the compression stroke, dry lubrication will be encountered at the hot
spots while hydrodynamic or mixed lubrication is still likely to be encountered at
the surrounding areas, which still have remaining oil ﬁlm. This also means that
bore wear is going to be much higher at the hot spots than the surrounding area.
Another observation from Figure 28 is that the evaporation rate at the hot spots is
slower during the power/exhaust stroke process than during the intake/

compression stroke. This observation agrees with other literature [3,125].

4.4.2 Temperature Effect on Bore Wear

In order to investigate the sensitivity of bore wear to the magnitude of the
hot spot temperature, simulations have been conducted using eight sets of bore

temperature distribution, in which each has different peak temperature at the hot
spots. The peak temperatures of the hot spots are set 10°C, 25°C, 30°C, 40°C,
50°C, 60°C, 75°C and 100°C higher than their surroundings. The original

temperature at the hot spots is about 180°C. The location of the hot spots is
similar to those of Figure 22. The bore wear results under these conditions are

shown in Figures 29 - 36 respectively.

71

Depletion of Oi Film Thickness at TRR due to Evaporation
720 W" '— __ w—‘T

 

 

630

540

ES
c» is
o 01
o o

Crank Angle D

 

0 90 1 80 270 360
Bore Circumference DEG
FIGURE 28. Depletion of oil ﬁlm thickness at TRR under the evaporative
effect
From the comparison of these results, one can observe that the increase of
the hot spot temperature by 10°C does not produce any wear pockets (Figure 29).
But the wear pockets at hot spots begin to emerge when the hot spot temperature
is raised by 25°C (see Figure 30). From 25°C to 40°C increment (Figures 31 - 32),

the wear increase at the hot spots remains relatively small (about 3% of total wear

at TRR), but the area of wear pockets does increase.

72

 
 

e
E] 3.50E—10—4.00E-10
I 3.00E-10-3.50E-10
I 2.50E-10—3.00E-10

I 2.00E-10—2.50E-10
3.50E-10 I1.50E-10—2.00E—10
U1.00E-10—1.50E—10
I 5.00E-1 1-1 DOE-1O
I D.DOE+OO-5.00E—11

   
    
    
 
   
   
   
   
 

Wear MM
N
o
o
r."
3

Circumference

DEG

  
  
 
   
 
 
    
  

  
 

7 El 3.50E—10—4.00E-10

I 3.00E-10-3.50E-10
I 2.50E-10-3.00E-10
I 2.00E-1 0-2.50E-1 0
El 1.50E-10-2.00E-10
El 1.00E-10-1.50E-10
I 5.00E—11-1.00E-10

4.00E-10

  

3.50E-10

  

  

3.00E-1O

 
 

2 2,505.10 .o.oos+oes.ooe-11
2 1
if, zoos-10
B 15015-10
1.00E-10
360
5.005-11 300
000500

   
 
 

180 Bore
120 .
60 Circumference
DEG
. . . . °° 8 o
Nondimensional Axual Locatlon 2

FIGURE 30. Cylinder bore wear with +25 degree C hot spot temperature

73

  

BORE WEAR DISTRIBUTION (+30 C Hot Spot Temperatur

y D 3.50E-10-4.00E-10
I 3.00E-10—3.50E-10
I 2.50E-10-3.00E—10
I 2.00E-1 0-2.SOE-1 O
D1.50E-10—2.00E—10
D1.00E-10-1.50E-10
I 5.00E-11-1.00E-10

  
  
   
 
      
      
  

4.00E—10

  

3.50E-10

 
 

3.00E-10

 

   

2 2.50E—10 I0.00E+00-5.00E-11
2 .
g 2.00E—10
3 15015-10
1.00E-10
5.00E-11
000E+00
12380 Bore
Circumference
DEG

 
  
   
       
   
 
    
    

D 3.50E-10-4.00E-10
I 3.00E-10-3.50E-10
I 2.50E—10—3.00E-10
I 2.00E-10-2.50E-10
El 1.50E-10-2.00E-10
El 1 .00E-10-1.50E-10
I 5.00E-11-1.00E-10
I 0.00E+00-5.00E-11

  

 
 

Wear MM
.N
O
0
II"
3

180 Bore

Circumference
°° 8
Nondimensional Axial Location 8

FIGURE 32. Cylinder bore wear with +40 degree C hot spot temperature

74

As the hot spots’ temperature is increased to 230°C (50°C increase),

spikes of wear are observed at hot spots (Figure 33). The magnitude of wear

pockets jumps up by about 16%. From 50°C, 60°C, 75°C to 100°C increments

(Figures 33, 34, 35 and 36), the magnitude of wear pockets increases gradually to

about 25%. The increase of the wear pocket area is also more distinct after 50°C
hot spot temperature increment. The observation of the change of hot spot wear
with respect to temperature increase is summarized in Figure 37 where the non-

hot spot wear is included for comparison.

This observation may indicate that there exists a range of temperature at
the hot spots at which local bore wear pockets may emerge at a substantial rate.
One possible reason may be due to the nonlinear relation between the
evaporation rate and the bore temperature (oil vapor pressure is an exponential
function of ﬁlm surface temperature [126]). Hence, the temperature range for

substantial increase of the bore wear pocket may fall between the increment of

50°C and 100°C margin. Another possible reason may be due to the exponential
relation between ﬁlm temperature and viscosity. The oil viscosity decreases
exponentially as its temperature increases. The nonlinear temperature effect on
oil viscosity will result in the reduction of hydrodynamic pressure in a nonlinear
pattern, which ultimately causes the increase of asperity contact interaction and

wear.

75

A useful conclusion from this observation is that the hot spot temperature
at the cylinder bore wall needs to be controlled so that it does not go 50°C higher
than the surrounding area in order to prevent undesirable wear pockets on the
cylinder bore. In practice, this analysis may be useful in providing guidelines to the

design of a modern diesel engine system that has the potential to encounter this

phenomenon.

e
D 3.505—10—4.00E—10
I 3.00E-10—3.50E-10
2.50E-10—3.00E—10
I 2.00E-1 0-2.50E-1 0

BORE WEAR DISTRIBUTION (+50 C Hot Spot Temperatur
El 1.50E-10-2.00E-1O
El 1.00E-10-1.50E-10

’ I5.00E-11-1.00E-10

0.00E+00-5.00E-11

   
    
  
 
 
   

 

 

4.00E-10

  

  

3.50E-10

      
   
     
   

  

3.00E-10

 
 

2.50E-10
2.00E-10

Wear MM

1.50E—10
1.00E—10

360
5.00E-1 1 300
0.00E+00
180
Bore
o 120 Circumference
Vt O 60
"7 DEG

Nondimensional Axial Location

FIGURE 33. Cylinder bore wear with +50 degree C hot spot temperature

76

  
   
    
 
      
      
 

BORE WEAR DISTRIBUTION (+60 C Hot Spot Temperature

1 y El 3.50E-10—4.00E—10
1!.
I

I 3.00E—10-3.50E-10
I 2.50E-1 D-3.00E-1 0
I 2.00E-10-2.50E-10
1.50E-1 0-2.00E-1 0
D1.00E—10—1.50E-10
I 5.00E-11-1.00E-10
I 0.00E+00-5.00E-11

   

 

Wear MM
N
'o
O
r.“
8

Bare

Nondimensional Axial Location

FIGURE 34. Cylinder bore wear with +60 degree C hot spot temperature

    
   
       

BORE WEAR DISTRIBUTION (+75 C Hot Spot Temperature

1:: 350510400510
’ I 300515350510
I 2.50510-300510
I 2005-10.250510
21150515200510
1:: 100510450510
I 5.00511-100510

4.00E—10

   
 
 

  

3.50E-10

     
      
  
  
  
  

  

3.00E—10

2 2505.10 _ I0.00E+00-5.00E-11
E > , ., ,
5 2.00E-10
3 1.50510
1.00E—10
360
5.00E-11 .. 300
0.00E+00
Bore
Circumference
DEG
3 a
Nondimensional Axial Location 8

FIGURE 35. Cylinder bore wear with +75 degree C hot spot temperature

77

  

BORE WEAR DISTRIBUTION (+100 C Hot Spot Temperatu

I: 350510-400510
I 300510350510
I 2.50510-300510
I 200515250510
1: 150510200510
[1100515150510
I 5.00511-100510
I 0,005+00-500511

   
    
       
   
    
     

J
1

 

Wear MM
N
b
O
r."
8

180

Bore
120 Circumference
DEG
. . . . °° e o
Nondimensional Ax1al Location 3

FIGURE 36. Cylinder bore wear with +100 degree C hot spot temperature

 

 

 

 

 

 

 

 

 

3.90E-10 —7 ~~ w W ,,, ,, — , , W 75. ,,._. , 1e _, ~ ,W -
Wear vs. Temperature
3.70510 / ;
3.50E-10
Z -— Hot Spot Wear
E 3 30510 -'~— Non Hot Spot Wear
E .
h
8 /
3 3.10E-10 ,
2'905'10 +0°c +10°c +25°c +30°c +40°c +50°c +60°c +75°c +1oo°c
2.70E—10
2.50E-10 - . . . 1
180 200 220 240 260 280

Hot Spot Temperature °C

FIGURE 37. Temperature sensitivity of cylinder bore wear

78

CHAPTER 5 PISTON DYNAMICS AND EHL SOFTWARE

5.1 Introduction

In an internal combustion engine, the piston is the primary component that
transfers combustion energy from the fuel to the translating work at the crank
shaft that drives the whole engine to perform useful tasks. Hence, the design of a
piston is very crucial in determining the efﬁciency of an engine’s performance.
Piston quality is often due to its optimal geometrical design, body mass and
material selection that ensure minimal energy loss due to inertia, heat transfer and
surface interaction with the cylinder bore. The piston also should not compromise
its structure strength against fatigue failure. Nevertheless, to derive an optimal set
of design parameters is often a very challenging and difﬁcult job because there
are many design parameters that affect the piston function. Hence, computational
tools become very handy in piston design because of the time and cost

effectiveness.

Taking advantage of today’s high computer speed, this software, as part of
the CASE [13] system, has integrated a piston ﬁnite element model into the
coupled model of the piston dynamics and tribology so that the same ﬁnite nodal
information computed from the 3-D FE model is used in the solving of the piston
skirt elastohydrodynamics and the piston trajectory. The time-dependent local
bore temperature seen by each node during the piston reciprocating motion is
used in determining the local viscosity at the piston skirt; hence it accounts for
more realistic lubrication calculation that can be inﬂuenced by the viscosity

variation across the piston skirt surface due to non-uniform bore temperature.

79

5.2 Simpliﬁed Piston Model

Figure 38 shows the solid model of a production piston that contains
detailed geometrical information. Constructing this solid model typically requires
many hours of skillful computer-aided design (CAD) work. To conduct thermal and
static analyses using the model also requires an enormous amount of
computational time and effort. Hence, it is not feasible if one is to conduct a typical
four-stroke engine cyclic analysis using the model for parametric design study. A
simpliﬁed piston version (Fig. 39) is used to develop a piston software that can be
time-efﬁcient in performing a cyclic piston performance analysis, beginning with
the construction of the solid model, the meshing of the model, to the ultimate
calculation of the coupled analysis of elastohydrodynamic lubrication and
dynamics. A major concern in the model simpliﬁcation is to decide on the degree
of geometrical simpliﬁcation so that principal structural characteristics of the
piston are not lost. On the other hand, the geometrical input for the simpliﬁed
model should have reasonable ﬂexibility that allows the software to be applied in
most production pistons for the piston performance analysis. Hence, in the
preliminary effort of this software development, the simpliﬁed piston solid model

contains the following parameters (refer to Fig 40):

1. Piston diameter

2. Piston height

3. Piston pin axis locations
4. Average boss thickness

5. Average crown thickness

80

6. Piston skirt inside radius
7. Piston skirt radius center
8. Piston skirt height

9. Piston skirt width

10. Piston skirt tab dimensions

The above parameters can be considered as the principal parameters that
deﬁne the piston of most internal combustion engines. An important parameter to
be included later is the piston pin hole. Currently the piston pin is merged into the
average piston pin boss section, and the pin axis locations are designated by four
ﬁnite element nodes. These four nodes allow inertia loads to be assigned
appropriately, and the non-symmetrical effect of piston movement due to the

piston pin offset is preserved.

In order to investigate the structural and thermal behavior of the simpliﬁed
piston model with respect to the actual production piston model, thermal and static
analyses are conducted. The results are shown in Figures 41 and 42. It is
observed that the magnitude of the calculated piston skirt surface temperatures
are comparable between the two models, and the maximum temperature at the
top of piston crown differs by about 4%. The calculated piston distortions are also
in the same order of magnitude, but the distortion gradient toward the center of
skirt surface on the actual model is higher than on the simpliﬁed model. This
discrepancy is due to the absence in the simpliﬁed model of the pin hole and the
pin boss, which extend to the bottom of the top piston crown. It is believed that the

discrepancy will be reduced signiﬁcantly once those features are implemented

81

into the simpliﬁed piston. At this preliminary stage, considering that the distortion
characteristics on the piston skirt are similar from the qualitative point of view, this
simpliﬁed piston model will be adopted for the computation of the coupled analysis

of piston elastohydrodynamic lubrication and dynamics.

“News

mu
. “EEK“:
' ‘ mi:

_§:‘.

 

FIGURE 38. Piston solid model constructed with commercial CAD tool

 

FIGURE 39. Piston solid model constructed with CASE

82

 

XCG

 

 

PDIA

 

I— YCG

 

 

 

PHT
SI

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H—PITIA —

 

7////////////////

 

 

 

ﬁ\\\\\\\\\\\\\

CTHJ

 

 

M\\\\\\\\\\

 

 

 

 

FIGURE 40. Input piston geometry

83

’7
Maxaoss .

——- 306 640
- 281 037

- 255 533

- 229 930

,. . 204 427
it 178 873
153 320
12.7 767
102 213
76.660
51 107
25 553
o 00

 

 

   

Min,242 94

   

FIGURE 41. Computed temperature distribution between complicated and
simpliﬁed Piston

84

x ES
40

 

FIGURE 42. Computed distortion between complicated and
simpliﬁed piston

85

5.3 Software Modules

This piston modeling software comprises several modules integrated
together for a comprehensive simulation of piston performance as shown in Fig.
43. The preprocessing module allows the input of piston assembly geometrical
parameters, operating boundary conditions and material properties, and other
simulation control data needed for a cyclic piston analysis. This software is
capable of taking in the non-uniformly spatial bore temperature and distortion

information to compute the piston skirt elastohydrodynamic lubrication.

  
    
   
  

     

 

 

V
ELEMENT
OF PISTON ANALYSIS
MODEL FOR
PISTON

   
 

 

ELASTIC
, i .;.T

   

FIGURE 43. Software modules for piston modeling

86

5.4 Three-Dimensional Finite Element Modeling

5.4.1 Constructing and Meshing a Piston Solid Model

The ﬁrst computation module is to automatically construct and mesh a
piston ﬁnite element model similar to Fig. 39. The meshing of a piston solid model
is made using two type of elements, hexahedral (8 nodes) and pentahedral
(6 nodes). Speciﬁc loads and boundary conditions are imposed on the model in

order to carry out the ﬁnite element analysis in the next module.

5.4.2 Load/Boundary Conditions

Input data of the ambient temperatures and convection coefﬁcients are
distributed into four regions on the piston: 1) crown top surface 2) piston ring pack
region 3) piston skirt region 4) under crown region. Published works present these
thermal data obtained through measurement or numerical calculation
[55,86,91,100,128,130]. Figure 44 shows the speciﬁcation of temperature with

respective heat convection coefﬁcient on the open surface of elements.

The piston FE model is constrained at its four nodes that specify the piston
pin axis locations to zero displacements; and at its center node at the top of crown
to 1 degree of freedom (DOF) along vertical axis. A unit pressure is applied
uniformly over the top surface of piston crown, normal to the face of elements
(refer to Fig. 45). A unit acceleration with respect to its global Cartesian
coordinates is also speciﬁed uniformly over the whole model to account for inertia-

induced distortion.

87

 

FIGURE 44. Thermal boundary conditions

Constraint -—> 2
Unit Pressure Load ‘ "’

 

/

FIGURE 45. Constraints and unit pressure load

88

5.5 Thermal and Static Analyses

The next module consists of the ﬁnite element analysis solvers for the
thermal and static analyses. It is used to solve for the temperature distribution and
the distortion of the piston model. The piston distortion due to thermal load is
computed in this module. Assuming steady state thermal conditions, the thermal
distortion computed in this module will be repeatedly used throughout the entire
cyclic analysis. However, the distortion due to the inertia and the cylinder gas
pressure are transient. In order to save computational time, the approach
suggested by Zhu [133,134] is used to calculate numerically the distortion due to
inertia and cylinder gas pressure. The inﬂuence coefﬁcient matrices, deﬁning the
piston skirt deformation caused by unit cylinder gas pressure and unit
gravitational acceleration, is computed and stored for the next module. The
module involves the calculation of piston dynamics coupled with the piston’s

elastohydrodynamic behavior inside the cylinder bore.

For the purpose of verifying the accuracy of calculation, a piston FE model
is exported to a commercial software, COSMOS GeoSTAR, for similar thermal
and static analyses. The computation from the built-in solvers are validated to be
accurate as proven in the node-to-node comparison of the temperature and

distortion results, as shown in Figures 46 and 47.

89

9.00502
8.00E+02
7.00502

u_ 6.00E+02

é 5.00502

g 4.00502

'9 3.00502
2.00502
1.00502
000500

I
III

mlédaiféﬁér—amré 0516551156}.

 

—CASE RESULT
|— - -COSMOS RESULT 1

 

 

 

 

 

 

 

 

 

0

500 1 000 1 500 2000

Node Number

FIGURE 46. Node-to-node temperature comparison

8.00E-01
7.00501
6.00E-01

(I)

g 5.00501

.5 4.00501

5

E 3.00501

0
2.00501
1.00501
o.005+oo

 

1
Nodal Distortion Comparison I

 

 

—CASE RESULT L

 

]- - -COSMOS RESULT

 

 

 

 

 

 

 

0

 

500 2500

1500
Node Number

1 000 2000

FIGURE 47. Node-to-node distortion comparison

90

5.6 Piston Skirt EHL Analysis

At every degree of crank angle, the input cylinder gas pressure and the
calculated piston acceleration are multiplied with the inﬂuence coefﬁcient matrices
to obtain the resultant displacements at each ﬁnite element node along with the
steady-state thermal distortion calculated from the previous module. On top of
that, the distortion due to the hydrodynamic lubrication pressure load needs to be
calculated. Because of the non-uniformity of the hydrodynamic lubrication
pressure that impact the piston skirt surface on the thrust and anti-thrust sides,
the 3-D static analysis for the piston distortion due to the hydrodynamic load
needs to be conducted at every crank angle. Therefore, the effective piston
distortion is due to loads from the cylinder gas pressure acting vertically on the
piston crown, the inertial load acting vertically at the piston pin axis, the thermal
load due to temperature variation over the piston body, and the lubricant

hydrodynamic pressures acting radially over the area of skirt/bore conjunction.

By treating the skirt/bore interface as a slider bearing and assuming
incompressible lubricant, the 2-D Reynolds equation, Eq. (24) can be applied to
solve for the hydrodynamic lubrication over the piston skirt/bore surface both on

thrust and antithrust sides using the ﬁnite element approach [133,87,88].

ah3ap _a_h__3ja_p_ ah ah
a—(Ea 5705-.) 6%. +1253 54>

91

5.7 Equation of Motion

The static and dynamic loads that act on the piston as functions of crank
angle are shown in Fig. 48. These loading factors are affecting, individually as
well as collectively, the orientation and impact of the piston at every crank angle.
The moment and forces of the piston induced from the loading factors are
computed at every increment of computation over a complete cycle of engine
analysis. The piston moves along the cylinder bore in one of several modes as
shown in Fig. 49. The piston can come into contact with the bore at a top contact
point or a bottom contact point on either the thrust side or antithrust side of the
piston. If the piston is in contact with the bore at both contact points, this is
denoted as Mode 1 (two-point contact). If neither contact point is in contact with
the bore, this is denoted as Mode 4 (free contact). In Mode 2 (high-point contact),
the piston is in contact with the bore at the top contact point, but not the bottom
contact point. The piston is then free to pivot (rotate) about the top contact point.
In Mode 3 (low-point contact), the piston is in contact with the bore at the bottom

contact point, but not the top contact point. The piston is then free to pivot (rotate)

about the bottom contact point. In order to determine the contact forces (F1 and

F2 ), the piston lateral acceleration (£2) and the piston rotational acceleration (,3),
moments are summed about the crank pin location (Pt. 1) and the piston pin

location (Pt. 2). Rotational moments and acceleration (,8) are considered positive

clockwise. Forces and lateral piston acceleration (552) are considered positive to

92

the right. Summing moments about Pt. 1 for the piston and the connecting rod

loads gives:

— P,,(L2 + u) --PH(Ll +1) + F,(Ll + a) +F2(Ll -b) + F3(Ll + d) (25)
— F4(Ll — h) + JT,uF1(L2 + u + x2 + JSTR) + JBpF2(L2 + u + x2 +JSBR)
.m3g(L2_e+u)+m3y'3(1.2—e+1.)—"23):"3uI +c)+JTF-,(L2+u+R)
+JTF8(L2 + “‘R)—m2;2(L1) +"7252042) Tmrodsza—ILBTIde’ = 0

Summing moments about Pt. 2 for the piston and the connecting rod loads

gives:

—- Pv(u) —P”(f) + F1(a)—F2(b)+ F3(d) —F4(h) +JTﬂF1(u + x2 + JSTR) (26)
+JB,uF2(u + x2 +JSBR) +JTF7(R + u) —JTF8(R — u) + m3g(e — u)
—m3);3(e—u)—m3£3(c)—-Mpm—I3ﬁ = 0

Velocity direction coefﬁcients are denoted as J, and J3. Side contact
coefficients are denoted as JST and J53. Jsr = 1 if top contact is on the thrust side
and Jsr = —1 if top contact is on the antithrust side. Coefﬁcients are similar for

bottom contact friction forces. Additional sign conventions are:

1. Crank angle (6) - positive clockwise
2. Connecting rod angle (¢) - positive counterclockwise

3. Piston axial location from top of deck (y) - positive down
4. Piston axial velocity ()3) - positive down

5. Piston axial acceleration ()3) - positive down

93

top contact

0int
/

 

 

 

F.
bottom

 

 

 

 

 

 

 

FIGURE 48. Imposed loads and dynamics of piston during operation

94

 

 

 

 

 

 

 

 

 

 

 

FIGURE 49. Motion modes

By introducing the following new terms:

JV=JT=JB

x3 = xuz+ﬂ(c)
i3 = 17+,B(e—c)

Wpis : m3g

Equation (25) can be rewritten as

—1>,(L2 +u)-—PH(LI +1) +F,(Ll +a) “’2“: —b)+F3(Ll +d)
—F4(Ll — h) +JV,uFl(L2 + u + x2 +JSTR) +JV,uF2(L2 + u +x2 +JSBR)
—(L2—e+u)+m31y“+B(e—u)1(L2—e+u)—m3ix”3 +ﬁ(c)1<L. +c)
+ JVF7(L2 + u + R) + 4,1580.2 + u —R) -— m2x2(L,) + m2y2(L2)

+ m..d<s2)i1—I,...ﬁ+ 1...?» = 0

OT

95

(27)
(28)
(29)

(30)

(31)

(11,)Fl + (15(2))?2 + (1513);?2 + (H4),"6 = (H5)

and Equation (26) can be rewritten as

— mu) — P,,(/) + Fl(a)—F2(b) + F3(d) —F4(h) + JVpFl(u + x2 + JSTR)
+ JVyF2(u + x2 + 153R) + JTF7(R + u) —JTF8(R — u) + Wm
— m3ly" + Re — u)1(e— u) - "13052 + ﬁtcnic) —M,,,-,, —I,,,-.23 = 0

(e-u)
or

(”01:1 + (117)152 + (Hg);2 + (119),?! = (Hm)

where

HI = Ll +a+JV,u(L2+u+x2+JSTR)
H2 = Ll—b+JV,u(L2+u+x2+JSBR)
H3 = —m3(Ll+c)—m2(Ll)

H4 = -m3(L2-e+u)(€—u)‘m3(Ll +C)(C)—Ipis

2 to
”5 = PV(L2+u)+PH(LI +f)_[1r0d+mrod(s )I¢‘F3(L1+d)
+ Wpl.5.(L2 -—e+ u)—m3))(L2 —e+ u) —mz)3(L2)

H6 = a+JV,u(u+x2+JSTR)
H7 : —b+JV#(u+x2+JSBR)

HR = —m3(c)
2 2
H9 = —m3(e—u) —m3(c) —lp,-s

Hlo = Pym + PH(/) —F3(d) + F4(h) —JVF7(u + R) +JVF8(R — u)

— Wpl-S(e — u) + m3):(e -— u) + Mp,"

96

(32)

(33)

(34)

(35)
(36)
(37)

(38)

(39)

(40)
(41 )

(42)

(43)

(44)

Referring to Fig. 49, the boundary conditions for each motion mode can be

expressed as following:

Mode 1:
£2=0and,'6=0 (45)

Mode 2:
F2 = 0 and 552 = —a,B (46)

Made 3:
F, = 0 and £2 = M (47)

Made 4:
F, = 0 and F2 = 0 (48)

For Mode 1, solving Equations (32) and (34) with the boundary conditions
of Equations (45) yields,
_ H 5” 7 ‘HzH 10

F' _ 11,1517—1512151,5 (49)

 

_ H5H6_HIHIO

 

For Mode 2, solving Equations (32) and (34) with the boundary conditions

of Equations (46) yields,

' H,(— H80 + H9) -H6(- H3a +H4)

 

(51)

97

‘H1H10TH5H6

)6 = _H,(_H80+H9)+H6(—H3a+H4) (52)

 

For Mode 3, solving Equations (32) and (34) with the boundary conditions

of Equations (47) yields,

 

(53)

*H2H10+H5H7

ﬂ = —H2(H8b+H9)+H7(H3b+H4) (54)

 

For Mode 4, solving Equations (32) and (34) with the boundary conditions

of Eq. (48) yields,

 

 

- H H —H H
M = 5 9 4 10 (55)

The piston motion is ready to be solved with all the known values of the

input piston geometry and the calculated external loads except with the unknown

moment, M

W, developed due to the friction between the piston and the piston

pin. Referring to Fig. 50, Mp,” can be determined through the ﬁnding of the pin

force, A , ﬁrst. The direction of Mp," depends on the relative angular velocity of the

piston, [6, and the connecting rod 0. An iterative approach is adopted to

determine the lateral and the axial components of the pin force, denoted as A,

98

and A2 respectively. Summing piston forces in the lateral (x) and axial (y)

directions yields

+PII—F3+F4 = 0

+m3}33+—PV—m3g+F7+F8 = O

Rearranging Eq (57) yields
A, +Jp,u,)A2

or can be in the form of
A1 + C1142 = C2
Rearranging Eq (58) yields

or can be in the form of
C3A, + A 2 = C4
where

p

99

(57)

(53)

(59)

(60)

(61)

(62)

(63)

C4 = —J,,;z(Fl +F2)—m)73+PV+m38 (66)
"F7_F8

Solving for A, and A 2 results in

A = _—_C'C4‘C2 (67)
' C,C3—l

Then, the friction moment Mp,” can be determined as follows:

# d 2 2
Mp," = Jp-%’,/A, +212 (69)

With all equations needed for the piston dynamics derived at every crank

angle, the piston contact forces (F, and F2) and the piston accelerations (£2 and

,B) are determined from the moment relationships derived previously in Equations
(49) to (56) by initializing Mp,” equal to zero. The moment Mp," can now be

calculated using Eq. (69) and used in determining the new contact forces and the

piston accelerations as well as new values for A, and A2 using Equations (67)

and (68). The process is repeated until convergence is achieved.

100

The piston accelerations (£2 and ,8) and the contact forces (F, and F2)
are determined at every crank angle throughout a four stroke cycle. The lateral

velocity and position of the piston are calculated by integrating £2 and ,6 over the

appropriate time step. The location of the top and bottom of the piston skirt is
continually monitored so that the mode of the piston motion can be determined at

every crank angle.

5.8 Piston Impact

When the piston impacts the cylinder bore, the kinetic energy of the piston
at impact is determined under the assumptions that the velocity of the point on the
piston at which impact occurs is reduced to zero (rebound does not occur); and if
the piston is free in the bore prior to impact, then the angular velocity of the piston
after impact is determined according to the principle of conservation of angular

momentum.

Piston angular velocity after impact may be determined as follows.

Consider the system just prior to impact, with the piston having a horizontal
velocity (223) and angular velocity (,8). The mass at point 2 (piston pin location) is

me, and is equal to the piston pin mass (m2) plus the equivalent connecting rod

mass located at point 2. The mass (meq) has velocity :22. The angular momentum

just before impact with respect to the point (P) of impact (see Fig. 51) is given by

Ha = m3x3(b+c)+13)8+ meqxzb (70)

101

After impact the velocity of the piston at P is zero, and the piston rotates

about the point of contact at the new angular velocity (,8’ ). Thus the angular

momentum after impact is given by
Ha' = m3x3'(b + c) + [3,3’+ meqxz'b

but from Fig. 51, one has

)(Tz' b3 and 21"3 = (b+C)).6'

Then

Ha' = I3B+meqbzﬂ+m3(b+c)2,3'

Applying the principle of conservation of angular momentum,

gives

_ m3x(b + c) + 13B + meqizb

 

[3 + m3(b + c)2 + meqb2

Eq. (75) deﬁnes the angular velocity of the piston after impact.

(71)

(72)

(73)

(74)

(75)

As a measure of the intensity of each impact, the change in kinetic energy

of the piston is calculated from

Kinetic Energy Lost at Impact = KE, — K152

where

102

(75)

.2 .2 .2 .2
KE, = 0.5(m3x3 +m3y3 +meqx2+l3j3 )

= Kinetic energy of piston before impact

 

 

(77)
K152 = 0.501238%, + m3)»; + meqiia + 1,133)
= Kinetic energy of piston after impact
(78)
where the subscript a refers to the velocities after impact.
top
P contact
I " point
.. P -2 *X
F "‘23: F '
f .
msiz
1F, "‘39 F ‘
_F._ M .__5_
JinpA? ..... pm .A.1....
J :

 

 

    

I; ______
bottom '
F,I contact/J IF“
point

FIGURE 50. Pin forces acting on piston

103

 

 

FIGURE 51. Piston impact with cylinder bore

5.9 Numerical Algorithm for Piston Skirt EHL and Dynamic Analyses

The lubricant ﬁlm thicknesses and pressures (on the thrust and antithrust
sides of the piston at the end of the crank angle increment) are calculated

according to the following procedure:

1. Skirt distortions due to lubricant pressures are determined using nodal
lubricant pressures that exist at the beginning of the crank angle increment.
Additional skirt distortions are added to the lubricant pressure distortions.
These include skirt proﬁle (including ovality) and skirt thermal distortions which

remain constant over the four-stroke cycle.

104

2. Skirt distortions due to combustion pressures are determined by multiplying
the gas pressure distortion vector (inﬂuence coefﬁcient) by the actual pressure

occurring at the current crank angle.

3. Skirt distortions due to piston inertia loads are determined by multiplying the
inertia distortion vector (inﬂuence coefﬁcient) by the piston acceleration at the

current crank angle.

4. Gas pressure and inertia distortions are combined with all other distortion
vectors resulting in a skirt contour at the end of the current crank angle

increment.

5. The skirt contour and lateral piston location in the bore and piston tilt are
then used to determine lubricant ﬁlm thickness on the thrust and antithrust
sides of the piston. With the axial velocity of the piston at the end of the current
crank angle increment known, the lubricant pressures at the end of the current

crank angle increment are determined.

6. The resultant loads and shear forces as well as gas pressure loads and

inertia loads are computed at the current crank angle.

7. The procedure is repeated for all crank angles.

5.10 Preliminary Results and Discussions

Simulations using this piston software are conducted for a boosted-

motored four-stroke engine at two operating conditions: 1500 RPM 2 PSI Boost

105

and 3000 RPM 2 PSI Boost. The intake throttle is regulated with boosted
pressures in order to imitate the ﬁring pressure trace. Some basic input data are

listed as follows:

Nominal piston diameter 88.95 mm
Nominal cylinder bore diameter 91.00 mm
Stroke 90.00 mm
Length of connecting rod 169.20 mm
Piston height 50.10 mm
Piston skirt height 27.00 mm
Average pin boss thickness 13.50 mm
Average crown thickness 12.10 mm
Piston skirt tab height 3.50 mm
Piston skirt width 65.00 mm
Piston skirt tab width (top) 40.00 mm
Piston skirt tab width (bottom) 29.00 mm

Bore distortion and other non-symmetric boundary conditions are not
considered in the present analysis so that the piston performance under the effect
of skirt elastohydrodynamic lubrication can be studied more clearly. The cylinder
gas pressure traces for the two operating conditions are plotted in Fig. 52 as
functions of crank angle. The intake stroke start at 0 degree crank angle and the

exhaust stroke ends at 720 degree crank angle.

5.10.1 Piston Dynamics

Figure 53 shows the solution of the cyclic piston trajectory at 1500 RPM
2 PSI Boost and at 3000 RPM 2 PSI Boost condition. The antithrust side is at the

positive upper bound and the thrust side is at the negative lower bound in the

106

plots. The upper and lower bounds are the calculated effective clearance
allowable for the piston to move laterally taking into consideration the piston and
bore distortion, piston skirt proﬁle and piston ovality. More vigorous piston motion
is observed at the 3000 RPM 2 PSI Boost condition than at the 1500 RPM 2 PSI
condition. The drastic direction change of the pin movement in particular near 360
degree crank angle (CA) signiﬁes the event of high piston impact. The change of
piston velocity at TDC and BDC (0, 180, 360, 540, 720 degree CA) and the
change of piston acceleration at midstroke (90, 270, 450, 630 degree CA) seem to
initiate changes in the piston pin lateral position. This is understandable, as it is
known that the shift in velocity direction affects the balance of piston momentum,

and the change in acceleration direction affects the piston load equilibrium.

Boosted Motored Cylinder Gas Pressure at 1500 RPM 2PSI
Boost and 3000 RPM 2PSI Boost
450.0 ~

400 o — ,4, —1500RPM2PSI Boost
" ‘ -~-~3000RPM2PSI Boost
350.0 5

300.0 -
250.0 —
200.0 —
1500 ~
100.0 ~
50.0 —
0.0 ‘ 1 '1
0 90 180 270 360 450 540 630 720
Crank Angle (degree)

 

 

 

 

 

Pressure (PSI)

 

 

FIGURE 52. Cylinder gas pressure traces

107

Figure 54 illustrates the cyclic piston tilting motion. Higher engine speed is
seen to cause more piston rotational motion. It is also observed that the piston
rotates at the highest magnitude of allowable angle (constrained by the piston/
bore clearance) during 3000 RPM throughout the whole engine cycle, but a small
degree of piston rotation is seen at some crank angles during the 1500 RPM
operation. This implies that during high engine speed, piston momentum is
sufficiently high to permit a complete change of piston orientation, from one

motion made to another opposite mode.

With higher kinematics at higher engine speeds, one can expect to see
greater contact forces impacting the piston as shown in Fig. 55, which shows the
contact forces that take place at the top and bottom contact points. Positive
magnitude designates the contact point at the thrust side, and negative at the
antithrust side, for both top and bottom contact points. The results again show that
the engine speed contributes to the increase of contact forces due to more
dynamic piston motion at higher speed. The contact force experienced by the

piston at 3000 RPM is about four times higher than at 1500 RPM.

108

1.20 k,

0.80-E::

.0

&

O
l

Piston Trajectory at 1500 RPM and 3000RPM 2PSI Boost

3 I.
.' I
I. .
.11 -

 

 

-1.20 — .,

FIGURE 53.

9.00E-02

7.00E-02

5.00E-02

3.00E-02

1 .00E-02

-3.00E-02

Rotation Angle (Rad)

-5.00E-02

-7.00E-02

-9.00E-02

 

-1 00502 t

 

 

 

 

Crank Angle (degree)
b1500RPM2BT - - - -3000RPM2BT

Cyclic piston lateral movement at 1500 RPM and 3000 RPM
2 PSI Boost

 

 

 

Piston Rotation at 1500RPM and 3000RPM 2PSI Boost

 

 

Crank Angle (degree)

 

| —1500RPM2BT - - - -3000RPMZBT |

 

FIGURE 54. Cyclic piston rotation at 1500 RPM and 3000 RPM 2 PSI Boost

109

Piston Contact Force at 1500 RPM 2PSI Boost

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5.0051-05 5" .__,-.__ WW W , WWW-“j
A.5,oos+os 450 540 630 720
z ?
34.00506 - 2
g-isoewe - ;
Ll. ?

4905106 ' Top Point(F1)

450906 _ ——Bottom Point(F2) I

-3.00E+06 .

Crank Angle (degree)
. Piston Contact Force at 3000 RPM 2PSI Boost

2.00E+06 -; i

0.00500 ‘ , . W
£200,906 1) 90 180 270 360 150 540 630 720
§400E+06 ~
63600506 - %

8 DOS +06 ------ Top Point (F1)

- — Bottom Point (F2)

4.00507 55 ,_ . ~. . -

WCrank Angle (degree)
FIGURE 55. Piston contact forces on thrust and anti-thrust sides

5.10.2 Piston Skirt Elastohydrodynamic Lubrication

Figure 56 gives the lubrication forces on the thrust and anti-thrust sides.
The lubrication forces are computed from the integration of hydrodynamic
lubrication pressure over the bearing area of thrust and anti-thrust sides. Higher
lubrication force is observed at 3000 RPM than at 1500 RPM due to the fact that
Reynolds hydrodynamic pressure is proportional to the sliding speed at two
contact surfaces. In addition, with the lubrication analysis coupled with the piston
elasticity and dynamics, one can also expect higher hydrodynamic pressure due

to the squeeze ﬁlm effect, which is a function of the rate of radial surface

110

movement either due to the piston lateral motion or the skirt radial elastic

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

deformation.
Piston Lubrication Force at 1500 RPM 2PSI Boost

5.00E+06 —~ '7

0,005+00 ,, j ,, .. w r . . I
A ' 90 180 270 “I 3601450 540 630 720
55.00806 -
3 I
54.00907 - I
“- é

4.50907 ~ """ Thrust(F3) I

——Anti-Thrust (F4) i
-2.00£+07 W- .
Crank Angle (degree)
Piston Lubrication Force at 3000 RPM 2PSI Boost

2.005+07

0.00rs+00 r r, a . .. - V 23' r . .
$400907 _ 90 180 270 360 «I50 540 630 7I0
$400907 .
3 -6.00E+O7 -

—————— Thrust (F3) g

- . , I

we“ —AntI-Thrust(F4) I

-1.00E+08 L~~~—--—«—— - ---——— -—:— ~ - a I

 

 

Crank Angle (degree)
FIGURE 56. Piston lubrication forces on thrust and anti-thrust sides

Figure 57 gives the magnitude of contours of hydrodynamic lubrication
pressures computed at 5 degree crank angle during the 1500 RPM 2 PSI Boost
analysis. Imposing these pressures on the 3-D piston FE model results in the
distortion of the piston skirts as shown in Figure 58. Positive distortion represents
the expansion of the body structure and negative distortion means contraction.
This also means that positive distortion will result in smaller skirt/bore clearances

and vice versa. The different pressure proﬁles on the thrust and anti-thrust sides

111

are due to the difference in skirt/bore clearance as a result of distortion from
pressure, inertia and thermal loads. Without considering the non-symmetric
inﬂuence from bore distortion or temperature, symmetric results about the axis

plane normal to the piston pin are expected.

Figures 59 and 60 give the magnitude of contours of hydrodynamic
lubrication pressures and resultant distortions computed at 5 degree crank angle
during the 3000 RPM 2 PSI Boost analysis. Similar distribution proﬁles of
hydrodynamic pressure and distortion are observed at the two cases. Comparing
the magnitude of hydrodynamic distribution, one can observe that the 3000 RPM
engine speed contributes to better hydrodynamic lubrication than the 1500 RPM

condition.

112

Skirt Oil Film Pressure-Thrust Side

  
  

 

[IDES

’5‘
“i
2 0.08»
3
E s
i 0.04\-~":
E .
E E
5 U.02~..-"'
E
x
m

0:

15

Skin Circumferential Node Skirt Axial Node

Skin Oil Film Pressure-Anti—Thrust Side

8

   
 
   
 

on
D
1

Skirt Oil Film Pressure (kPa)
El ’5

 

_5
mo
\I

Skirt Circumferential Node Skirt Axial Node

FIGURE 57. Piston skirt thrust and anti-thrust oil film pressure at 5 degree
CA, 1500 RPM 2 PSI Boost

113

Oil Film Pressure Induced Skirt Distortion-Thrust Side

-2~.

Skirt Distortion (mm)

 

 

Skirt Circumferential Node Skirt Axial Node

Oil Film Pressure Induced Skirt Distortion-Anti-Thrust Side

x10.

  
 

Skirt Distortion (mm)

_5
I

 

O
V

15 ..

Skirt Circumferential Node Skirt Axial Node

FIGURE 58. Piston skirt thrust and anti-thrust side distortion due to oil
ﬁlm pressure at 5 degree CA, 1500 RPM 2 PSI Boost

114

Skirt Oil Film Pressure-Thrust Side

 

  
  
 
 

0.2..
’c? .
3

a .....
E 01.

E

E

6 D.%~

.E

x

(O

 

d
010
V

Skirt Circumferential Node Skirt Axial Node

Skirt Oil Film Pressure-Anti-Thrust Side

Skirl Oil Film Pressure (kPa)

 

 

Skirt Circumferential Node Skirt Axial Node

FIGURE 59. Piston skirt thrust and anti-thrust oil ﬁlm pressure at 5 degree
CA, 3000 RPM 2 PSI Boost

115

Oil Film Pressure Induced Skirt Distortion-Thrust Side

Skirt Distortion (mm)

 

 

Skirt Circumferential Node Skirt Axial Node

Oil Film Pressure Induced Skirt Distortion-Anti-Thrust Side

U.B\..,.~:~"U'I

I3
0')
I

.‘3
a.
I

Skirt Distortion (mm)

.D
M
1

 

 

Skirt Circumferential Node Skirt Axial Node

FIGURE 60. Piston skirt thrust and anti-thrust side distortion due to oil
ﬁlm pressure at 5 degree CA, 3000 RPM 2 PSI Boost

116

An interesting observation is seen at 365 degree CA during the 3000 RPM
2PSI simulation, as illustrated in Figures 61 and 62. At 365 degree CA, the piston
movement is observed to be at a very vigorous stage because of the impact from
the peak pressure, as shown in Figures 53 and 54. The piston top is tilting away
from the anti-thrust side, resulting in a divergent surface proﬁle which acts like a
scrapper piston ring. As a result, no hydrodynamic pressure is generated on the
anti-thrust side of the piston skirt. However, the hydrodynamic pressures
generated at the thrust side, applied in the piston 3-D model, produce distortion
on the whole piston structure. Hence, distortion is seen at both thrust and anti-

thrust sides of the piston skirt.

117

Skirt Oil Film Pressure-Thrust Side

B

.a
(II
I

  

Skirt Oil Film Pressure (kPa)
a

 

Skirt Circumferential Node Skirt Axial Node

Skirt Oil Film Pressure-Anti-Thrust Side

Skirt OiI Film Pressure (kPa)

 

 

Skirt Circumferential Node Skirt Axial Node

FIGURE 61. Piston skirt thrust and anti-thrust oil ﬁlm pressure at 365
degree CA, 3000 RPM 2 PSI Boost

118

Oil Film Pressure Induced Skirt Distortion-Thrust Side

Skirt Distortion (mm)

 

   
  

Skirt Circumferential Node Skirt Axial Node
Oil Film Pressure Induced Skirt Distortion-Anti-Thrust Side

x 10'

DB~ .

.0
U)
1

.0
A
l

Skirt Distortion (mm)

P
M
.’

 

Skirt Circumferential Node Skirt Axial Node

FIGURE 62. Piston skirt thrust and anti-thrust side distortion due to oil
ﬁlm pressure at 365 degree CA, 3000 RPM 2 PSI Boost

119

CHAPTER 6 COMPUTATIONAL VALIDATION - OPTICAL ENGINE SYSTEM

6.1 Introduction

The engine performance predictions using the computational models
presented in previous chapters are compared with experimental measurements
obtained from experiments conducted under the collaborative effort between the
Engine Research Laboratory of Michigan State University and Mid Michigan
Research, an Okemos engineering consulting company. The USA. Army
sponsors this effort [36]. Basic principles of the experiments are presented in this

chapter.

This experimental rig is an optical system consisting of an engine equipped
with an optically accessible full-length cylinder made of sapphire. The optical
engine is positioned in a test rig equipped with laser optical devices so that high
precision oil ﬁlm thickness measurement using the laser-induced ﬂuorescence
technique (LIF) can be made during engine operation. For the piston temperature
and the piston ring pack gas pressure measurements, telemetric technology is
adopted. Here a piston is installed with wireless micropressure transducers and
thermocouples for measuring the gas pressure and piston temperature at piston

ring grooves and lands.

6.2 Laser-Induced Fluorescence Method

Laser-induced ﬂuorescence (LIF) has been widely adopted in the
measurement of piston assembly oil ﬁlm thickness. The laser induced

ﬂuorescence is the wave emission from atoms or molecules because of their

120

excitation to higher energy levels due to the absorption of laser radiation. This
method is selected because the ﬂuorescence detection has the advantage of
higher sensitivity and is non-intrusive. It is a spectroscopic method for sampling
neutral or charged atomic or molecular species at their fundamental or excited
states. When applied in the optical engine equipped with the full-length sapphire
cylinder bore, the LIF method can be used to measure the oil ﬁlm thickness
directly at any crank angle and location; thus it provides a complete visual

observation on the piston assembly oil ﬁlm thickness.

6.3 Optical Engine System

Figure 63 shows the optical engine. The optical properties of the sapphire
cylinder allow oil ﬁlm measurement to be made at all possible directions along
circumferential and axial directions (Fig. 64). It is mounted on a non-ﬁring
production engine structure, which is driven by a motor. Figure 65 shows the
photo and the schematic of the optical system. The optical system consists of a
CCD camera, an Excimer laser, laser optics and the optical engine. These are
aligned precisely in a test rig such that the laser beam is directed to the sapphire
cylinder for the measurement of the oil ﬁlm thickness at desired locations. Table 3
gives a summary of the functions of components in the optical engine system.

TABLE 3. Optical system components and functions

 

 

 

 

 

 

 

Component Function

Excimer Laser Provides laser beam source at wavelength of
about 308 nm

Laser optics Guide and direct beam projection

CCD(charge-coupled Captures illuminated images of oil ﬁlm thickness

device) Camera

 

121

 

 

 

 

Component Function

Filter Provides narrow bandwidth for light waves of
about 505nm to pass through

OpticTabIe Provides guided positions for optical components

Data Acquisition Device Processes image data

 

 

 

 

 

F"
.

   
     

TDC I 1 Mid Stroke“ ‘ BDC

FIGURE 64. Sapphire cylinder provides optical access over entire range
of piston stroke

122

 

Sapphire
(Cylinder

 

 

 

 

 

 

 

 

 

    

Laser D (I
Fme’ OPTICAL
kNGINE
Acquisition___C.'Iﬂ‘ILA—"'91e

 

 

 

FIGURE 65. Optical engine system

123

6.4 Correlation between Oil Film Thickness and Film Illumination
Intensity

Prior to oil ﬁlm thickness measurement for an operating optical engine, the
relationship between the thickness and the brightness of the illuminated oil ﬁlm
must be known. For this reason, a LIF calibration technique is developed. Making
use of the sensitivity of LIF, the numerical relationship between ﬂuorescence
intensity and oil ﬁlm thickness is determined. The calibration is conducted using
the static calibration method on a calibration stand, shown in Figure 66(a). The
accuracy of this method is accomplished by using a micrometer with a resolution
of 1 micron as shown in Figure 66(b). The micrometer is used to hold a known
thickness of oil against the sapphire window in the test stand, thus allowing
ﬂuorescence intensity to be measured at various oil ﬁlm thicknesses. To measure
the oil ﬁlm thickness, the area between the micrometer head and the sapphire is
ﬁrst ﬂooded with SAE 5W30 engine oil. The area is then illuminated with
ultraviolet light (308 nm wavelength) from an Excimer laser, which results in the oil
ﬂuorescing at a wavelength of approximately 505 nm. A 505 nm narrow band ﬁlter
is ﬁtted onto a CCD camera in order to capture the images that only contain
information from the ﬂuorescing oil ﬁlm with no interference from the original
308nm light wave reﬂection. An example of a LIF image of the calibrated oil ﬁlm
thickness is seen in Figure 66(c) where the bright, round area is the oil trapped

between the micrometer and the sapphire.

124

 

FIGURE 66. LIF calibration apparatus (a) calibration stand, (b) micrometer,
and (c) LIF image of oil ﬁlm thickness

With the known oil ﬁlm thickness from the micrometer and the recorded
ﬂuorescence intensity, the correlation between these two data can be determined.
To do so, the image captured needs to be transformed into digital form - pixel
format, in order for the correlation to be completed. Figure 67 shows two
examples of post-processed pixel intensity images for 32 microns and 106
microns respectively. As illustrated here, the LIF method is extremely capable in
capturing the ﬁlm thickness variation at the micron level. In an operating engine,
the oil ﬁlm thickness at the ring/bore interface is also at an order of microns. Thus,
the LIF method has proven to be effective in the oil ﬁlm thickness measurement

for the optical engine.

125

Data Captured at 32 micron Thickness Pixel 'ntenSitY

 

 

60 90 120 150

Data Captured at 106 micron Pixel Intensity
24o

 

60 90 120 150
FIGURE 67. Processed digital LIF image

Images in this dissertation are presented in color.

126

Figure 68 shows a result of LIF calibration, where a measurement of oil ﬁlm
thickness on the calibration stand was taken at seven different oil ﬁlm thicknesses
regulated using the micrometer. The oil ﬁlm thickness is observed to have an
exponential relationship with its image pixel intensity. However, under normal
operating conditions, the ring/bore oil ﬁlm thickness usually does not go beyond
50 micron. Hence, the linear approximation from the correlation curve can be

used as the reference for oil ﬁlm thickness measurement of the optical engine.

100

 

8
l
l
I.
i.
i
i.
ol

 

 

. k

 

D
.\,
.‘

 

O) «I
O
\O

 

Pixel Intensity
8

 

A
O

 

0)
O

 

N
O

 

1O

 

 

l

0 50 100 150 200
Oil Film Thickness (urn)

FIGURE 68. LIF calibration curve

 

127

6.5 Procedure for Oil Film Thickness Measurement

The oil ﬁlm thickness measurement for the optical engine under running
conditions is performed using the following two-step procedure. First, the engine
is rolled away and the calibration stand is stationed in its place. The LIF calibration
is carried out in the way described in previous section. After the calibration is
completed, the engine is rolled back to the exact same location, where the stand
was, for oil ﬁlm thickness measurement. High precision translation of optical
engine position is achieved using a test stand which is equipped with devices for
small horizontal displacement. Figure 69 shows the optical engine that is installed

on a test stand in a test cell.

 

FIGURE 69. Optical engine in test cell

128

6.6 Piston Ring Pack Pressure and Temperature Measurement

The piston of the optical engine is equipped with state-of-the-art
technology, IR telemetrics, to measure gas pressures at each of its grooves and

lands, and its body temperature, as shown in Figure 70.

 

FIGURE 70. Piston equipped with IR telemetrics

Six pressure transducers are mounted on the piston with one transducer

each at the second and third lands, top ring groove and oil ring groove, and with

129

two transducers at the second ring groove with one at the thrust side and another
at the antithrust. Five thermocouples are mounted on the piston with one
thermocouple each at the second and third lands, and all the three piston ring
grooves. The measuring devices are wireless; magnetic induced electricity is
generated from the interaction between the coil on the piston and a stationary
base on the engine body. Digital signals are acquired from the transducers and

thermocouples during engine run time.

6.7 Engine Test Runs

Engine tests are conducted to measure the performance of the motored
optical engine over a range of operating conditions. Even though the optical
engine does not operate with fuel combustion, the cylinder gas pressure is
regulated or boosted at the intake throttle to imitate the ﬁring engine gas pressure

trend.

6.7.1 Inter-Ring Gas Pressure Results

During the inter-ring gas pressure measurement, the sapphire cylinder
bore is replaced with a cast-iron cylinder bore so that the measurement can be
taken over a greater range of operating conditions. Cast-iron is used because it
can withstand higher impact from the piston dynamics during high speed, and the
inﬂuence of the material of the cylinder bore on the gas dynamic is almost
negligible. Selected operating conditions for the inter-ring gas pressure

measurement are given in Table 4.

130

TABLE 4. Operating conditions for inter-ring gas pressure measurement

 

 

 

 

 

 

 

Speed Intake Pressure Condition

1000 RPM 2 PSI Boost, 4 PSI Boost, 6 PSI Boost, 8 PSI Boost

1500 RPM 2 PSI Boost, 4 PSI Boost, 6 PSI Boost, 8 PSI Boost,
15 mm Hg Manifold Vacuum

2000 RPM 2 PSI Boost, 4 PSI Boost, 6 PSI Boost, 8 PSI Boost,
15 mm Hg Manifold Vacuum

2500 RPM 2 PSI Boost, 4 PSI Boost, 6 PSI Boost, 8 PSI Boost,
15 mm Hg Manifold Vacuum

3000 RPM 2 PSI Boost, 4 PSI Boost, 6 PSI Boost, 8 PSI Boost,
15 mm Hg Manifold Vacuum

 

 

 

Figures 72 to 95 show the measured inter-ring pressures in comparison to
the CASE simulated results. Gas pressures at the top ring groove, the piston
second land, the second ring groove, and the third piston land, as illustrated by

Figure 71, are compared between the prediction and measurement.

The overall result comparison shows that the prediction of the ring pack
pressures from the computational models is in relatively good agreement with the
measurement except at the third piston land pressures and the third ring groove

pressures (not presented).

 

 

 

 

 

 

 

 

 

i \ Combustion Chamber
f T Top Ring Groove
1' J 1 Second Land
i I Second Ring Groove
1‘ Third Land
l PISTON Third Ring Groove

 

 

\ Crank Case
\ ORE

FIGURE 71. Detailed piston ring pack region

131

The comparison of the top ring groove pressures for all operating
conditions, except 1000 RPM 8 PSI Boost, shows that the error is within about
5%. Good correlation is observed in both quantitative and qualitative
comparisons. This observation should be expected due to the direct application of
the measured cylinder gas pressure as a boundary condition for the gas ﬂow
calculation at regions surrounding the top piston ring - ﬁrst groove and second
land. For the second land pressures, 15 of the 24 cases show that the comparison
is within about 5% error and good correlation is also observed in both quantitative
and qualitative comparisons for these cases. The lesser agreement at the second
land pressures compared to the top ring groove pressures are due to the fact that
the calculation is now not relying on the measured cylinder gas pressures, but
rather on the calculated top groove pressure and the calculated second land
pressures as boundary conditions. For the second ring groove pressures, 11 of
the 24 cases show good correlation in both quantitative and qualitative
comparisons. However, for the third land pressures, only 3 of the 24 cases show

good correlation.

The discrepancy can be attributed to some degree of shortcoming in both
the experiment and the computational model. The telemetric technology used in
the experiment has a major problem of noise disturbance on the measured
pressures and temperatures. Filtering technique is currently being used to ﬁlter
out the high frequency data, which are assumed to be from the external noise.
However, the ﬁltering process has a very high potential to eliminate useful

information at the same time. For instance, the pressure spikes due to the ring

132

ﬂuttering are predicted at some cases, but the measurement does not display the
spikes even though a similar trend of pressure traces between the prediction and
measurement is observed during the occurrence of the spikes (refer to Figures

85,86,87,92,93,94).

On the other hand, the model of the inter-ring pack pressures in CASE is
based on the oriﬁce-volume method, which solves for the ﬂow pressures across
all piston grooves and lands simultaneously using the iterative method. The
calculated pressure of a control volume is very dependent on the pressures at the
upper and lower boundaries. While the pressure at the top boundary (top ring) is
the measured cylinder gas pressure, the pressure at the bottom boundary (oil
control) is the crank case pressure, which is assumed to be at constant
atmospheric pressure (measurement is not available). The constant pressure
assumption is not so appropriate in this application as the measure pressure at
the oil control ring region does show a sinusoidal pressure trace pattern as a
function of crank angle. The current gas dynamics model needs to be improved in
order to account for the cyclic variation of crank case pressure. As a result, good
correlation is observed at the region closer to the combustion chamber, but the
discrepancy is seen to increase across the piston ring pack region towards the

engine crank case.

Several useful observations regarding piston ring dynamics are given here.
Spikes on the pressure traces are observed at some cases, implying the ring

ﬂuttering phenomenon, or instability of the ring dynamics within the ring groove. It

133

is known that there is a mutual dependent relationship between the piston ring
dynamics and the inter-ring gas ﬂow dynamics. The piston ring relative positions
in ring grooves affect the pressure variation across the piston ring pack as the
axial and radial movement of piston rings within piston grooves creates variable
paths for the cylinder gas to ﬂow from one piston land or piston groove to another.
On the other hand, pressure variation also has countereffect on piston ring
dynamics. The pressure gradient between the top and the bottom of a piston ring,
coupled with its inertia, will dictate its position equilibrium within the groove. The
ﬂuttering observed in some cases is believed to be from the second and third ring
since no pressure ﬂuctuation is observed on the top ring groove. The instability of
the piston ring dynamics is observed to increase as the engine speed increases.
This implies that the gas pressures on the top and the bottom of a ring and its
inertia have mutually comparable effects on the ring dynamics during high speed
conditions. In contrast, for the top ring, the pressure on the top of the ring (cylinder
gas pressure) is much higher than on the bottom of the ring, resulting in much
stable and predictable piston ring axial motion. A more severe degree of pressure
ﬂuctuation is observed at the third piston land, which implies that the oil control

ring probably experiences the least stability condition.

134

“moom an. N SEN. 92:. an 0.59.35 mam 9.2.53:— .N~. mung".

 

 

 

 

 

 

 

 

‘—
39.83 992 £55
own 08 o3 one 08 EN 8? cm 0
é..l\ - ..
// Snug?
.35 E 85an . /{ N.
voczwmezl
58m .wa a 2am 08.. 6.5805 “ES 35. -
8283 292 €20
own can own omv own cum 2: 8 o

 

. _ . q

 

 

 

mw<o E 25360 .
35302 I

.moom _wn_ w Sam 82. 9:82.“. use; 988m

 

 

 

 

i

O
F

O
N

In
N

o O o
Vt 0') N
(isd) aJnSSSJd

O
10

oo

‘52
(led) einsseid

$9.33 0.92 x55

can can own on? own EN 03 cm o

 

_ . _ _

 

 

1.

Xx
\

 

 

 

 

on.» ~oaﬁéc$ 3.5302. . .. <
35 B 323.3
3.5.9.32 II

.woom 5a a 2am Door 05.89:... 0390 3E accoow -

 

 

 

8233 665‘ £80
own 8w 3.0 one own EN 03 om o

 

. . . _ _ s _

 

 

 

 

 

$5 .5 858.3
35235. I

“woom an N 2am coo... 2385 3090 as”. no... ;

 

 

 

 

O o O O o
V m N 1-
(isd) aJnSSSJd

(isd) einsseid

can
own
00v

135

moon .2 v 2%. 82 a 2:32.. mam 32.3.... .2. $50.”.

3233 293. EEO
ovn one oon onm o? oo

_ . 4

own cum

 

 

E‘s-'11

DID

/ 5...... 2
mm<o E “.3238. /J.._ N

6053mm”: I
.88 an 4 sam 82 2385 9.3 2.5.

 

3983 292 EEO
owe omo o3 o9q oom o3 o? oo

_ _ _ a _ _ 1 I

 

 

i

 

 

 

 

 

$5 3 328.8 .
beammoi I <

.88 an; 2%. 89 2885 23 28mm

 

O
.—

0 IO
N ‘-
(isd) sinsseid

IO
N

O O O O
V (‘0 N I-
(isd) ainssaid

O
I!)

oo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$9er 292 €20

own one 96 on v own o8 8— oo o

3% 32:55; 85232 . .
$5 3 85323
_ Oohamaoi I
763 5% SE”. 89 238i 9020 new 9685 L
$233 29.4. EEO
one omo own omv 8n EN om: oo o
$5 .3 858.8
85322 I _I <
zoom 5% Sam 83 93me 985 SE nob ;

O O O O O O O
(D In Vt ('0 N ‘-
(isd) einsseid

2
136

0
on

000
one
Nv-v-

own

(isd) einsseid

C800
to I00
VVMM

88m 5.. o 2%. 82 8 288... 8a 9:32... .2. manor.

30503 293 EEO

own onw own omv own 2N on V cm 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.— I—I a 4 I1 dI u a o
m
_
.— 0? H
m
with... a
m
ow (
88 .3 88.3.8
85.9.8.2 I mm
_
. $0.0m ﬁn. m Sam 89 238...". ES PE... on
3933 0.92 x520
own omo 03 one oom osm o? oo o
m A I.‘ 1 I— d _ 4 o
_
_ 9
\ I ON H
A <... 0mm
.. . a.
A \ OV(
w .
88 3 88.8.8 . / .. on
vasomozl <
_ Econ .wn o 2mm 82. 2:89.“. 95.. ocgow . oo

8933 0.93 EEO
own one own one own cum oo— om o

< q a <

”HT"?

 

/ <...
s ,.
w .
4 .
x . .38 825.23 8.882 . . ..
I P 88 .5 88.8.8
. < _i 3.382 I

 

 

 

 

 

 

 

 

 

.mOom .ma 0 SEE Door Sowmoi im>ooL0 oEm ucoomm L

$233 29?. x55
owe o8 own omv own cum o? oo o

_ _ _ _ _

 

 

 

 

88 .3 8.23.8 . C
thnmnmrz I

.88 an. w 2%. 89 288... 2,85 as”. 8.. -

 

 

O
1-

O
N

O
‘1'

O
In

com

com

8v

com

0
('7
(red) einsseid

(isd) einsseid

137

.88 .8. a six 82 8 2:82.. 8.. 9.32... .2. mane...

3233 o.oc< x520
owe o8 oen one own ouw o2. om

. a .

 

— . . .

 

 

I!)

O
‘-

 

,"2
(rsd) einssaid

O
N

 

‘0
N

 

 

88 .3 88.3.8
02382 I

32.5 an m Emu. coop .2385 9.3.8.2...

 

$0.38 292 3.5.0
one own oen on. on

. _

own 08 oen

. . .I _

 

. on

O
('0

 

O
V

 

(isd) einssaid

O
In

 

 

.88 .3 88.3.8

3.385. I

«moon 5... a 21m 08.. 2.6on “.5... ocooom

 

 

8

 

.2

can one oen o

$283 292 EEO
ne 08 EN oow oo o

. . _ . a I .

 

 
 

on.» 32.5-2.3 3.38.2 . . . .
88 .3 8.23.8
3.38.2 I

.8095an 2am coo... 9:32.”. ,o>oo...0 3E .ucooom -

  
   

 

$0.33 292. 3.5.0
own o8 oen one cow oR o9 oo o

. _ .

 

_ _ . _ _

 

 

 

88 .5 88.3.8 . ..
qummoﬁl ., .

 

 

.. .
zoom .3 o 2%. com 9.88.5 m>oo.0 9.2". no... -

O O O
V 6') N
(isd) ainsseid

O
In

oo

0
O
N

O
D
C")

(isd) sinsseid

O
O
V

138

 

«moon .mn. N Emm— oomF «a 2.5me man mat-.2... .2. mmngu.

.0963. 29.... €80
own o8 oen one own cum o3 oo o

— _ _ . _ 4 . —

 

 

 
  

 

  

88 .3 88.3.8 .
3.385. I

Zoom .9". N 2%. oonw £38»... 9.3 E2... ‘ %

 

 

.8503 0.9.... .520

own 08 oen one own EN 09 8 o

 

_ . _ < _ . . .

 

 

 

 

é .
.
N
88 .3 Ensuﬂ g
3.38.... I :

  

 

 

.88 .8 N .2... 8m. 9:82. 9.3 288

o

a 2 2
(gsd) amssajd

In
N

o to o to O
0') N N '- '-
(lsd) aJnSSSJd

.858. 292. .520
o3 one oen one own o8 oo. oo o
. . . . . . o

 

 

 

 

 

 

       
 

 

 

 

 

_

_

.

I .. mm
(x .3.“ 3255.2. 8.38... . . .. oe

M . _ $8 .3 86.8.8
“ . 3.38211. we
.. 38m 5.. a 2%. con. 0.38.... 88.0 9.5 2.80% on

.853. o.o..< €20
own ono oen one cow ohm o2 om o
_ . o

H. _ _ .7 . ._ .on

 

 

 

 

(gsd) SJnSSSJd

    

.
_
.
.

 

88 .5 85.3.8 .
3.3.8:. I

< 8m
. 38m .8 .0. 2%. 82 £38... 28.0 8E 8.. 8e

 
   

 

139

“moon .9... V Emm— oomw «a gamma...“ mum mot-.35 .2. mung...

.8508 0.92 x55
ow. one oen one own onm o? oo

 

W . J . . . . .

 

 

 

 

W .moom .ma e .2n.w. cone 2:82.... 95.. 2......

.832... o.o..< EEO
one own oum oo— om

_ 4‘ .

own omo oen

_ a . . .

 

 

 

 

 

 

 

85 3 86.3.8

 

3.385. I.

.88 an. e 2%. 8m. 938x. 9.3 9.88

 

 

ID

0
.—

O In
N v-
(gsd) amssaJd

IO
N

on

8 8 8
gsd) amssaJd

8 8

ows

one

.0283 was... .520
oen one own ohm oo. oo o

 

. _ _ u .

 

- -5--."

 

 

 

 

.. 88 .3 88.3.8 \

 

 

.88 .m. e 2%. 82 £38.“. M98.0 9.x 9.88 l

ono

8.385. ll

 

 

 

 

.853. o_o..< €20
oen one own ohm o9 8 o

 

l1 ‘ .

. _ J88'!

 

_/\

 

.

 

.

 

 

 

 

    

.
.
.
.
.

 

mm<0 .3 085.3060 .

 

35802 I:

 

 

.moom.ma w SEW. cone 3:89.... m>oow0 new. no... l

O
M
(ysd) amssajd

O
V

oo

140

88m .9. o 2%. 82 a 238... man 8.33... .8. $50.".

.0953. o.o:< €90

 

 

 

 

 

 

 

 

 

 

 

 

ow. omo own onv own onu 8.. oo o
__
_
. Z .
. .. N .
.5
m
$5 3 83.3.8 _
vasomoél _
.38 En m in”. 9on— 9.38%. 55.. 2E... .
.0203. 2oc< 3.5.0
own one ovn onv 08 ohm cop oo o
ﬂ‘ _ < _ _ - q -
ol-

 

//

 

é...

 

 

 

 

$5 .3 8.23.8 .

8.382 I

 

 

.88 .9“. o 2%. 82 238.... 9.3 88w

If) 0
1- 1-

O
N

O O
t") N
(gsd) amssaJd

C)
V

(ng) amssajd

o

 

 

 

 

 

  

 

 

 

 

 

 

.8503 0.92 .55
.2 com ovn onv own o8 oo. oo o
i .311
4..
n.
. 83 335.92. 8.38.2 , . ..
n $5 .5 88.3.5 t
ﬂ oesmuozl
.38 .mn. o 2%. can. 928.0... @620 ocﬁ uconow
8233 0.92. x520
own omn ovn onv oom o5 oo— oo o

4 _ 1 q .
V

 

 

 

 

 

$5 3 85.3.8 -

 

<

_ .moom an; En”. comrwswmﬂnwgﬁw 9E. no.7 ._

8.38.2 I

 

O
V?

00..

com

com

(gsd) aJnSSSJd

(gsd) amssaxd

141

88. .9. a six 82 8 288... 98 8.32... .2. $50.”.
89608 0.92 .280 8283 292 €80
o oms o8 own on». own o3 o9 oo o

CNN 25 own omv own cum cap 8
. _ . _ o . . q _

O

 

. . <

 

J

 

 

J”
J

 

O
1-

I
“,‘2
(gsd) emssajd
{'11
r‘““‘“::'2:‘-g-
IO 0 (0 N
(lSdYSJnSSOJd

”‘5:

 

 

 

/‘ _
4
N:
K9:

0

“‘3‘
“T‘h
i
o
N

 

.35 5255.5. 8.882
88 .3 88.3.8
8:385.

 

O
(D

  

IO
N

 

 
  

8.385. -
.moom .ma m. 29% com r oSmen 9.3. 9...... on

 

       

 

. n .88 .8 w 5...... 8m. 982.. 88.0 PE 888 E

8.89 992 8.0 .383 0.92 8.0
8.. 8” EN 8. 8 o

own 08 own onv com EN on" ea c own one own

 

 

 

 

 

 

u. .[

o _ \

s T

n

m .. . . 08
w, . .

B. .

( 8..

 

 

O
In
C:
O
8

  

88 .3 8.23.8 -
USJmNmS. .\

    

O
CO

 

 

 

0
Is

_ 895 .ma m 2%.. con— «toga. 95H 280w %

88m .9“. o 2%. con. 2:82.”. 260.0 9.5. och . ooo

(gsd) aJnssaJd

142

.>..... m: E... m. 2...”. 82 8 2:32. man 8.32:. .8 mane...

.853. 29.... .85

can omo ovm omv own EN 99 cm o

 

_ a _ . . _ u .

     

  

ww<o 3 3.2330 .9
00.32.02 I.

 

  

 

 

 

 

 

 

 

 

r \r/
m ..>.2 o: E89. .29”. com. 2685 2:... BE... .
.8608 o.u:< €80
on. one own one can Eu 8.. cm
7 — u . - _ < _

 

 

 

.
.
.
_ FIN}
.

 

_. .7:

8..

 

 

$5 .3 8.23.5 ,
3.335. I

 

 

 

.>..2 a: sea 2%.. 8m. 23.8.. ES 283

 

 

ow

m—

(gsd) aJnsseJd

om

or

In
.-

O
N
(gsd) amssaJd

mm

on

mm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.3...va 292 x520
on» cam own on». own 8h . ow? om o
24.....- a
.3.» 83.5.2... 8.335.. . . .
mm<o .5 88528 ,
T1 90.50602]!
_ ”PE 9.. Eﬁm .4. Eu”. . pomp Wﬁmmomhrgmoﬁmsm wmoomm
8233 0.9.... €90
own 08 own 09. own EN 8. om 0
j _ — _ A u _ — v
.
T
1 .
_[ 004:. 3 3.23.8

 

 

 

u

   

8.3me I

  

 

.32 a: 22m. 2%. 82 288... 28.0 8.x no:

“‘70
v—v-

to O IO 0
0') c') N N
(gsd) emssajd

O
V

Jcl

143

.88 .m... N .28. 88 8 238.... m8 8....2... .8 maze...

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

$0.qu 292. .280
om. one own onv own ohm oo— oo o
a T _ _ _ _ u _ o
. m
_ o_. H
1 a 3151.. . m _. m
. . A 44:1 m
, . a 8,Kb
, 88 3 88.3.8 . . mm
USJva—z I.
.83.. an N 3E”. 88 £88.... 9.5.. v.2... L on
3233 v.92 €20
oms o8 own omv 8m osm o3 om o
1 4‘ 4 . _ _ 4 a o
.
W or
om
/, .8 . .
.. 8 m
n
7... o. ..
/ \ m
. on
88 3 8.8.3.8 VN
8.382 | 6. cm
88m .9. m 2am ooow £885 9:... 9.88

 

on

0

Nu 8w

A858. 0.9.... €80
own omv own ohm oo— oo o

 

4 - 4‘ 4 4 _

 

O
N

 

 

 

O O
V ('1
(gsd) eJnssaJd

 

 

 

p... .N. .8.“ 8.5.8... 8.88.2 . . ..
J. 88 .3 88.3.8 . on

 

 

 

 

 

 

 

 

 

 

 

d _ 3.38le
.88 .m... m 5...”. Son Bowman. m$8.5 3E. 2.88 cm
3238 o.o..< VEEO
own one own omv own 05.0. o? om o
- ~ u _ o
o9
_ cow
.

V 8m

. 88 .3 8.23.8 4 Sq
3.38.2 I

.88 .9. w 2%.. ooow 288k. 98.0 9.5. nob - 8m

(gsd) aJnssaJd

144

.38. .m. e s... 88. .5 2:32. «.2. 9....25 .3 meow...

 

 

 

 

 

 

 

    

 

 

3233 v.92 €20
om. one o3. on.~ own osw oo— oo
4.2;. §\
vezmmmil
. ‘ 68m .mn v Ena‘ooon 23mm... 9.3 9......
$239 0.9.... €20
own omo oen one cow ohm oo. oo o

 

 

 

 

 

.88 .m. v .2... 88 2:32. E... 288

O to O In
N v- 1-
(gsd) amssaJd

IO
N

on

O O
n N
(gsd) amssaJd

O
V

 

.

on

322.9 m.oc< €20
om. one own omv oom ohm oo. oo o
M / .. .22“ 325.3... 8.38.2 . . 4 .
. $5 .3 85.3.8
., < 3.332 I
. .88 5. v 2am Doom 2:32. 98.0 95. 9.98m
3288 292 €20
own one o3 omv oom oR oo. oo o

 

 

 

 

 

 

 

  

 

 

   

 

 

. _ . .

 

 

 

  

mm<o 232230.50 .

 

 

  

Banana: I

  

 

38.. .m. .V s... 8 238.. 28.0 9... 8.. .

(gsd) SJnSSOJd

cow

com

com

(gsd) amssemd

oov

com

145

.88 .m. o 2.2 Sea .a 23.2. as. 9.....2... .8 mane...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8239 2.92 €20
on. one o3 o9 o8 oR o? oo o
02:32... I
. moon .9. 0 22¢ ooow 2.332.. 95... 9...: -
$233 0.9.... €20
own o8 oen om.~ own cum o2. 8 o
.I 1 T 4 I I. _ I.
/ M.
V .
..
./(\.
$5 .3 8.23.8
3.2322 I
.80.... .m. o Eam ooow 2.382.. 9a.. 9.08m .

IO

0
'-

O
N

‘0
N

2
(gsd) aJnssaJd

(gsd) aJnsseJd

8239 w.o..< €20

own one own omv 8.... EN 09 ea c

 

 

 

 

 

 

 
 

.2... 335...... 8.38.2 . . . ,
$5 .3 8.23.3
. 2233.2 I

 

 

I «woom 5.. 0 En”. ooow 233...... 960.0 95.. 9.80m

.0239 293 €20
ow. o8 ovn onv ooo o5 oo. oo o

. _ .

/ x
. .
E

3.38: I d

.88 .0... m 2%. ooom 23mm... 98.0 95. a2. .

 

 

 

 

 

 

 

 

O
,—

C
N

8 8
(ng) amssajd

0
IO

0 O
O O
V ('7

§

80

(ysd) ems-591d

146

.88 .0. 0 .2... 88 .a 2:82. a... 9.....2... .8 02:0...

 

 

 

 

 

 

 

  

 

 

80509 0.92. €20
cu. one owe omv oen o8 oe. om
.
I; a m h}
.f _. .2 2.1.x
0..
ix
ww<0 .3 3.23030 r.
30.3muozl
_ .008 an; Enﬂooow 0.5000,.necﬁ 9......
80503 0.9.... €20
ow. one 03 omv oen 2h on. em o
.

 

 

 

 

008 3 8.23.8 ,

 

002.5005. I

.. .moom _wa e 2%... ooow 0500.0... 9.0. 9.80m

 

 

.

IO

0
v-

0 In
N I-
(gsd) amssaJd

In
N

on

O O o O
V ('0 N ‘-
(gsd) aJnSSSJd

8

cm

 

 

 

 

 

 

 

 

 

 

80509 0.92 €20

om. one ove oev oen ohm oe. oo o
/ 4......
.

w .2... 32.2.2.2. 02:80: . . . .

N 8858.238 .

4‘ 8.5302I

.88 .02 0 SEE ooow.2:mm0..n. 0320 92”. 9.80m

30509 0.92 €20
own one own or... oen o5 oe. oo o

 

 

 

 

MM
a“

 

 

88 .3 8.23.8 ,
8.382'

.88 .0. m 2... 8K. 2:28.. 28.0 Em .8 .

 

O O
N v-

0
(V)
(gsd) aJnssaJd

147

§ § 2 a
(gsd) amssajd

O
0
IO

ooe

.>.s. m: E... 3 2mm 88 .a 2:32.. mum 9.....2... .3 mane...

 

 

 

 

 

 

 

 

 

 

.0233 292 .520 .8503 292 x520
own one ovm omv own ohm o? oo o own one 96 omv coo ohm o2 oo o
. _ . q _ _ . o _ _ _ J‘ n < _
m
.8.» 5252...... 8.385. . . . .
E d $5 .3 2.53.8 «
J uoSmaoEl
or a.
m.
if; m
.3. \‘ mw M.
Q ..

 

 

20.5 .3 022323
33322 I

>2 2. 5.5.21”. 88 2:32.. .23 3.: L 8

 

.858. 0.92 .55

omv own onm 8F om o own one own omc own 2N cs 8 o

 

 

$233 292 €80

 

 

«

_ _ _ _ o . Axum“ _ _ . _J‘_ _ E

 

 

 

 

 

 

 

 

,/
4?"
8
(Isd) SJnSSSJd

 

 

 

mw<0 >9 35.30.00 .

. “.2335. l .
.>,s. 9.. 5E9 2a.”. ooow 9....»me DE»... 988% 2 on

r
K. mm . M
.

 

. $5 3 3.28.8 .
om .8532... I

 

 

.>..2, o... EEmv 2am. omen 2owmen. 98.0 9...”. .ucoomw.

 

 

 

 

>2 a: 2...? 2%. 88 2:82.. 98.0 Em 8. -

or

(gsd) aJnssaJd

OOOO
(OVN

O O O
N 0 co
1- v-
(gsd) aJnssaJd

COO
CCD?
1-1-1-

148

.88 .m. u 5... 8mm 3 2:32. «an 9.....2... .2. maze...

 

 

 

 

 

 

 

 

 

 

 

(gsd) eJnssaJd

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.228... 292 2.20 .028... 292 2.20
om. 08 9... om...V 8.. Eu 8. 8 o om. 08 9... 8.. 8m 2m 8. 8 o
M . q . a a _ . O .ﬂ 4 . ~ 4 . q u C
. . m
e
o. H
. n
m.
8 ( .V
a . . 82235.22. 3.332.... mm
. .96 3 83.3.8 .
.28 .3 323.8 s
3.382.! r... em no.3uo2l ov
.88 .mn. 5... 8mm 2:82. 2.3 2...... . on zoom. .mnﬂinmoomwmswmm... 38.0 9.3288. me
.3303 0.92 €80 .353. 292 €20
om. one owe o9~ oen o3 oer oo o ow... one own one oen onw of oo o
.q u q d _ 4 q a o q u .- u o
. o. 8.
H
om a com
S
n
m
on M 8m
.3 a 3...... e .3... gm”? < s.
02:30.2.l
.886. «2.183 2:32. Ea... 2.8mm . om . .88 .m... s... comm 2:82.985 a... .o» . com

149

(gsd) emssaJd

«moon in. Q 2mm oomN .5 9:5me man mar—£3:— .~.e wKDUE

.338. 292 .520
om. one owe oev oen ohm on. on o

. . a . . . . _

 

 

   

  

85 .6 88.3.8 .
023305. I

 

 

. moon .9. v 2%. comm 2:82.. 9.... v.2... . L
.853. 0.92 €20

own one owe oev oen CNN oep om o

. .

 

  

 

II—

  

mw<o .3 3.23.8 .
8.382 I

n.
.
.._.

    

 

88.... .m. v s... 88 288.. 8... 888 -

(gsd) amssajd

0..

om

on

(gsd) aJnssaJd

ov

on

.88... 2...... .85
ow. .8

~ _ _ I4 . _ a .

eve omv oen EN 8 V cm 0

 

 

 

 

. . .2... 8.5.8.... 8.8.8: . . . .
F 88 .3 8.23.8
v0.38... I

 

. .88 .w. e s... .8. 2:82.286 9... 888

.353. v.92 €20
om. one o3 on.~ oen oR

on P on o

.

 

88 .3 88.3.8 , <

 

3.3.8.2 I
.88 .m. v 2... 88 2:82. 28.0 9... .8 .

 

O
N
(gsd) eJnsseJd

O
('0

O o
O O
N 1-

O
O
(‘0

oov

ooe

ooe

(lSd) amssaJd

150

zoom .9. o 2%. 83 .a 2:82... 3.... 9.32:. .8 $59".

 

 

 

 

.358. 29;. £55
on. one own one own ohm on. on o
M q _ _ q 4 4 .
En?! .\ 3! i.
. Ar/ . W JR}

 

.I
$3 .3 3.93.3 , i

3.335. I \

M I .Moom .mn m 5%. com“ ﬂamman. 33 EE-

 

.mo.uo_u. £92. .380
own ono ovm omv own ohm on. om

 

 

. _ _ q < . _

 

 

 

 

$5 .3 3.23.3 .
023322 I

., .woom 5a m 21m comm 2:395 mom. ‘mcooow

 

 

.

l0

0
v-

0 In
N r-
(gsd) amssajd

on

or

ow

on

(ysd) amsseJd

ow

on

owh com 03.

.858. 292 x55
8.. 98 OR 8. 8 o

 

. 4

 

 

 

 

 

32» 33.5.5.2... 3.5302 . . . .
mw<o >0 8.93050 It

 

 

(J

35302 I

 

7.00m .mﬂm SEW. comm 2:305 0520 95» ucooom

 

.352. 0.92 x520

own omo ovm onv com CNN 09 om o

 

 

 

 

 

Zoom 5... 0 2mm comm 9:89... 9020 ocﬁ och L

(gsd) aJnssaJd

O

09

§

(gsd) amssajd

com

8
v

§

com

151

88m .9. a six 88 a 233:. £8 8.22:. .8 $59“.

.3629. 29...... xch

 

 

 

 

 

 

 

 

 

  

   

 

om. o8 own onv own EN oo. oo o
.
, 88 3 3.23.8 . .
vousmmmEI
.83 5a. m Eamoonm 28mm...“— ES .25... .
.858. 0.92 2:80
om. ono ovn on.~ own o5 on. on

.

 

u . . 1 _ .

 

  

88 .3 3.238
35322 I

60.5 an; SEW. oonw 2835 new. racoww

 
   

0..

@—

 

O
N
(gsd) amssajd

O
('9

V

(gsd) amssajd

own

.

.

.

.0968. .292 2:80

and own omv own cum 02 cm 9

. _ n 11 . 4

 

 

 

 

 

.3; .858... 3.33: I , . I
88 3 3.23.8
0238:. I

    

 

  

 

.88 5n n 2am oonw @3me 98.0 BR E88
.0058. 29.4. 2:90

o8 own onv own cum on: cm 0

 

 

 

 

 

 

 

 

O O O O
V m N ‘-
(gsd) aJnssaJd

O
In

em

8 8 8 8
(.0 to V G)
(gsd) aJnssaJd

O
O
N

152

.>.E a: EE 2. 2mm comN um 0.59.3.5 mum met-23:. .5 maﬁa."—

 

   

.858. 0.92. 2:90
on. ono own onv own ohm o? 8 o
M96 3 882330 . . _ 4 _ _
gzuaozl

 

 

 

 

53?.

 

 

. {a

.32 9.. 8:. n. 5%. obnm mEmmoE new. EEF

.858. 0.92 2:90
own one own one oon ca o? oo

 

 

.I . _ . _ 4 4 .

 

 

 

  

 

 

/ . x
f . ..
88 3 3.23.8 I . :

 

 

. uoSmmoEI _ m
.>.2 2. E... n. 2%. 88 23mm... 9.3 2.83

 

 

L

2
(gsd) amsseJd

IO
,—

(nsd) aJnssaJd

on

on

 

 

.858. 292 2:80
on. one own onv own cum on. oo o
8.23585323383 _ _ 4 _
88 .5 3.23.8
Baaaozl

 

 

 

 

 

 

.3... 2. E... 2 saw. 0me 233.3 820 8.2 8688

.858. 292 2:80

own onm own onv own cum oo— 8 o

 

 

 

 

 

 

 

88 2.. 3.23.8 .
005285— I

 

.-,- 1,—.- - V..-

.>.2 5.: EE. n. 2am oonm @589... 88.0 3E ooh L

(gsd) amssaJd

8 8 8 8 8 8 8
1- ‘- .-
(gsd)aJnssaJd

8

car

153

.88 .9. N 2:: 88 .a 2:82: man 9.32... .2... $59".

.0293. 05:... 2:90
own one ovm or... oen onw on. on. o

. . q _ _ . . In

 

 

 

 

 

2%

.
.

Hf!
9}“ .1
, 4 .
_a|| r.

 

 

.

.

. .moem .ma «Sam ooon 8:83: Em... PE... ‘
.0053. 23.... 2:80

o3. one ove oev oen cum on. 0..

_ _ .

o

In 0
N N
(gsd) emssaxd

88

O
N

 

 

8
(psd) amssaJd

 

O
V

 

3.38.2 |
.moom 5n N 2am coon Somme... “Ea. ucooow

 

8

 

own one ovm omv oen EN

.333. m_o:< 2:90
oe. on. o

_ _ _

 

. _ _ _

 

 

 

 

 

 

 

.8.» .3255. 338.2 . . , .
38 3 8.23.8
35232 I

. .,
.,,..

,T W .
.

 

on

.88 an. N. 2:288 2:32: £86 8.: 28$ . mm

.358. m_o:< 2:90
own one eve oev oen ohm oe. oo o

d . _

 

 

 

 

 

 

  

mw<0 nugﬂauﬁo .
veammezl

    

 

.88 .9. N 2%. 88 933.: 98.0 2.: 8. .

cop

§ 8
N
(gsd) aJnssaJd

O
O
9

8m

(ng) aJnSSSJd

154

zoom .2 v 5%. 88 a 2:82.. «an 3E3... .3 $59.“.

 

 

 

 

 

 

 

 

 

 

 

(gsd) amssaJd

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(gsd) aJnssaJd

 

 

 

 

 

 

 

 

 

 

8233 £92 €90 $233 29.2 x520
ems one own 09. com EN cm. 8 o own one own one cow EN 09 cm 0
u 1 u . q _ a 4 o a d a a a q 4 o
m m
3
Ma
r M aw
m. / M w W m .,
_ _ W a
3 a in W... A35 a
$3 3 25.3.8 .‘ M . LI
uoSaooﬁl < on m w .8528: on
,wmoom .mn. .v Eaﬂ‘ooom mammmi new». 35 % mm .88 .mA V 55m. 08.". .913»an @520 BE ncooom‘ L mm
8238 29? £55 3233 292 v.55
emu cam 03 omv com EN 09 om o 03. cam 03 one cow 2m 8w om o
*1 q _ q a q u q o _ u a o
f o. m OS
on H
cow
_ m
_ m R
T {A [ omw: \ com
mw<o 3 8.28.8 . m. w ov .35 3 35.3.8 ‘ oov
02332 .I .5, 85322 .l
2335, o>020 951 no... ; com

 

 

.88 .3 V 2%. 08m 2385, 23 Beam ; 8 a .88 .9; v 2%. z ,

155

zoom .2 o 2%. 88 a 2:32.. «no 322:. .8 $59”.
3993 £92 €80

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$233 292 x580
omu o8 oen one oom 2h o9 oo o
A S \n/rqqk
4 a
$5 3 8.23.8 , x
2.5822 I
mono Lawn. ‘mAEn—mkooomQPmMmE New... 35 -
3233 292 €80
own one oen one oon ohm o2 oo o
_ / om b
M I. N
o w
< x
o R. ...._._.. ... m.
$5 3 2.28.8 T
Uganmozl K ..
.

 

 

 

 

38m an; 530000 9:39,...— ucs ucoomw

 

L

to O In
N N '-
(gsd) amssajd

O
0')

mm

0
N

O
m

(gsd) amde

0
v

on

0N5

08

oem

one

own

2N

on F om o

 

d

_

u

_ —

 

 

 

 

 

:00
F‘—

 

If)
N

 

No-1..“p ‘ .. -

m V, ,ﬁ

.moom..ma m SEE coca £olmmmohsm>o€0 05m accomm‘ ;

0mm

oen

 

8
(gsd) amssaJd

 

 

82m $2,:an 8.38.2 . . . .

$5 3 2.3.3.8
3532,. I

100
(000

 

320mg 292 xcEO

one

com

EN

09 cm o

 

 

 

 

$5 3 853.8
35322 .I

 

 

 

omoom 5a m 2am coon 8:3a 9005 95... och L

 

 

O
V

O O O

O O O

('0 N 1'-
(gsd) amssaJd

O
O
‘3'

com

com

156

.38 .m. m 5%. 88 a 2:82.. man 2:22... .em mane...
30.608 292 .280 $933 292 £55
own own oen one own ohm on. on o own omo oen one on» oR o? oo o
- u < u u a o a _ u _ . u A o

 

_ _

 

 

 

 

j
ﬂ
?
a;

 

 

 

/
Q;
to
N

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

. Kr. \ Al , .AoEm «mainogv 35822. on
29.0 3 8.23.8 . 1 mw<o 3 35.3.8
ouSwamEl on 1 . _ 3.38211 mm
.
600m in m Enm ooom‘ﬂsmmmmn ucmg PEP - nm ~woom .mm m SEE coon 23¢on w>oo.0‘o..‘_m,ucooow oe
$0.93 292 x520 80528 292. £55
own 0mm oen one 98 EN 09 om o own one oen one 08 CNN 03 om o
.— . _ . . . _ . O .
, 2
.1: _ “w u.
. 1:.
m , .. .. m
M A. ,. :8
$5 3 8.28.8 oe $5 3 85.3.8 (C
35322 I . “.2382 ll com
.88. 6.. o 5...... 88 2:32.. 2.3 2820, - on . 38m .9. a 2.... 88 2:32.. 2.8.0 Em 2: - 8 .

157

.>._.._ m... E... n. .2... 88 .u 2:32: 22. 9.....25 .3 $50...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.8503 v.92. x220
own omn oen one own onw of on o
Fi
_
. 3201.
umozamo§l
.
, .22 oz. E... ‘nw 2am coon 0.3.8.5 9.3 EE...
.3523 2.92 .55
ow“ omn oen one can ohm oo— oo o
_ 1 _ . 4 a _ .
_
.
_
.
.
. ..
.1 I: N
. .
_ J. ..
_ I p u.
. l g
mw<o .3 0293.3 .
neommmﬁl I! “l\({

   

.>.2 a: e... m. 2...”. 88 2:82.. 2.... 2.820,

 

.

O
,—

‘2
(gsd) amssajd

If) C If)
v- v-

Q
N

(gsd) aJnssaJd

.9233 v.92. .520
own omo oen one own onm oo— oo o

 

 

 

 

 

 

$2.». $25.52;; “55305. . ,
, . mm<0 .3 9823050
_ . 3.335. I

  

 

 

_ .>.—2 DI Fe m_ SEE 08h 2:139... 98.0 @558le

$29.3 2.92 x220
own one oen one own osm on. on o

.. .
. ..
:

8.3302 :1 4
>5 ow. ££,,n_ 55m obom 2:32.... 9.8.0 9.... n8. .

 

 

 

  

 

 

O 10 o If)
N 1- v-
(gsd) amssajd

IO
N

on

O
(O

8
(lSd) eJnssaJd

cm?

of

158

6.7.2 Oil Film Thickness Measurement Results

During the oil ﬁlm thickness measurement using the LIF method, the
sapphire cylinder bore is used. Currently, the highest engine speed is only limited
to 1500 RPM because of the concern over the ductility of the sapphire in response
to thermal load induced from the piston ring/bore interaction. An infrared camera
is being used to monitor the temperature distribution of the sapphire cylinder to
ensure that it is not overheated during engine operation. Figures 96 and 97
show the oil ﬁlm thickness measured at the top compression ring and the
second compression ring respectively over three engine speeds - 500, 750 and
1000 RPM. The measurement is taken at 320 degree crank angle (CA) (40
degree CA before top dead center, compression stroke). The measurements are
consistent with the general observation that higher speed leads to thicker
hydrodynamic oil ﬁlm. The ring proﬁle is used for reference. Comparison with the
prediction is made at 500 RPM 40 degree CA and 1000 RPM 320 degree CA for
the top compression ring oil ﬁlm thickness, as shown in Figures 98 and 99.
Without taking into consideration the roughness edge of the ring face on the
predicted result, the difference of the mean oil ﬁlm thickness between the
prediction and the measurement are 32.3% and 28.5% for the 500 RPM
40 degree CA and 1000 RPM 320 degree CA respectively. Considering the
degree of complication involved in the calculation as well as the measurement of

oil ﬁlm thickness, the error is considered relatively small.

159

500 RPM
750 RPM
1000 RPM
RING

Ring Axial Location (mm)

 

0 5 10 15 20 25 30
Film Thickness (um)

FIGURE 96. Top ring oil ﬁlm thickness proﬁles measured over various
engine speeds at 320 degree CA

—o
h

 

+ 500 RPM
+ 750 RPM

+ 1000 RPM
— RING

N

 

 

—A

 

O
CO

C)
3.

Ring AXIaI Location (mm)
D
G’)

O
N

 

O

0 10 20 30 40 50 60
Film Thickness tum)

FIGURE 97. Second ring oil ﬁlm thickness proﬁles measured over various
engine speeds at 320 degree CA

160

 

Top Ring Oil Film ProﬁIe

_ _.
- , _-’-
I.)
’_.

    
 

 

" Measured
4+ Predicted

 

 

 

 

 

A L l L

0 2 4 6 8 10 12 14 16 18

 

Oil Film Thickness (um)

FIGURE 98. Oil ﬁlm thickness proﬁle at 500 RPM, 40 degree CA

 

Top Ring Oil Film Proﬁle

1000 RPM 320 CAD _ ﬂ

   
 

 

" Measured
4+ Predicted

 

 

 

‘i-Jﬁu

 

 

 

0 2 4 6 8 1‘0 l2 1A4 16 1A8
on Film Thickness (um)
FIGURE 99. Oil ﬁlm thickness proﬁle at 1000 RPM, 320 degree CA

161

A systematic series of test runs is conducted for cyclic oil ﬁlm thickness
measurement. The selected operating conditions for the oil ﬁlm thickness

measurement are given in Table 5.

TABLE 5. Operating conditions for oil ﬁlm thickness measurement

 

 

 

Speed Intake Pressure Condition
500 RPM 2 PSI Boost, 8 PSI Boost, 15 mm Hg M.V.
1500 RPM 2 PSI Boost, 8 PSI Boost, 15 mm Hg M.V.

 

 

 

 

The measured oil ﬁlm thickness results are presented in Figures 101 - 115.
The oil ﬁlm thickness is measured over the region that covers the top and second
rings at 500 RPM and 1500 RPM under three intake pressure conditions (15 mm

Hg M.V. (or equivalent of 7 PSI Vacuum), 2 PSI Boost and 8 PSI Boost).

The detailed lubrication proﬁle obtained from measurement does shed
some light regarding the partially ﬂooded assumption made for the piston ring
pack lubrication modeling. The measured proﬁles given in Figures 101 - 115 show
that the oil transport at one ring is not relying solely on another ring. There is oil
trapped between two rings that contributes to the inter-ring oil ﬂow. At 270 and
630 degree CA, when the piston ring pack is moving upwards, the oil ﬁlm left from
the top ring is going to be a source of supplied oil for the trailing second ring. At 90
and 450 degree CA, the top ring becomes the trailing ring. A commonly used
assumption in the theoretical modeling of partially ﬂooded analysis is that the oil
left from the leading ring is solely the supplied oil for the trailing ring, and the
trapped oil between them is ignored, in order to established a inter-ring oil

transport model. This assumption is not appropriate at least in the application of

162

this motored and boosted optical engine system, as shown from the measurement
where the trailing ring is observed to have thick oil ﬁlm cumulation in its front edge
for all the cases. Therefore, the assumption used in the partially ﬂooded model of
this dissertation, which considers fully ﬂooded conditions for the rings below the

top ring, is shown to be valid in this application.

Oil Film Thickness Standard Top Ring, 500 RPM, TDC Intake

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I I I I I T I I I I

 

0 5 10 15 20 25 3O 35 40 45 50
Microns

--~~- 7 PSI Vacuum - - - -2 PSI Boost ----8 PSI Boost

—Top Compression Ring —Second Compression Ring

 

 

 

FIGURE 100. Oil ﬁlm thickness measured at 0 degree CA, 500 RPM

163

Oil Film Thickness Standard Top Ring 500 RPM 90 Degrees

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

~. . ::.:: 2'21‘rv ; -.
.1: - L: TOP Compression Ring

 

 

 

 

 

r r T I l r l

 

 

 

0 20 25 30 35 4O 45 50
Microns
----7 PSI Vacuum - - ~ 2 PSI Boost “*8 PSI BOOSt

—Top Compression Ring —Second Compression Ring

 

FIGURE 101. Oil ﬁlm thickness measured at 90 degree CA, 500 RPM

Oil Film Thickness Standard Top Ring, 500 RPM, BDC Intake
Stroke

 

 

 

 

 

 

 

 

 

 

 

 

I l I r ---I I I I I l

 

 

 

0 5 10 15 20 _25 3O 35 4O 45 50
Microns
- ----- 7 PSI Vacuum - - - -2 PSI Boost ----8 PSI Boost

—Top Compression Ring —Second Compression Ring

 

FIGURE 102. Oil ﬁlm thickness measured at 180 degree CA, 500 RPM

164

 

 

9“ FI'mTWCkr‘eSS 33.023"?! TEP. 359909 R?” $996.9”? .

 

 

 

 

 

 

 

 

 

“5...- T pCompressr n Ring
" ----- . ...... KW.-.” "'
“:5:

 

_ .?.LW—=—g;~. . _. I.
. f / Second CompressiOn Ring

’ I I F i

 

 

 

 

0 5 1O 15 20 25 30 35 40 45 50
Microns
7 PSI Vacuum - - - -2 PSI Boost —-—8 PSI Boost
—Top Compression Ring —Second Compression Ring J

 

FIGURE 103. Oil ﬁlm thickness measured at 270 degree CA, 500 RPM

Oil Film Thickness Standard Top Ring, 500 RPM, TDC

 

 

 

 

 

 

 

 

 

 

 

Compression Stroke
-__-~~~v~—I __ . . . -- “7.-.”; :7; .. : '1 '51-7771"‘*j
Top Compression Ring I
ISM“ ..........
C: Second Compression Ring
II;&;%.H;,__ 7..
""" <1...
0 5 10 15 20 25 30 35 40 45 50
Microns
7 PSI Vacuum - - - -2 PSI Boost ~- 8 PSI Boost
--Second Com ession Rin

 

—To Com ression Rin
FIGURE 104. Oil ﬁlm thickness measured at 360 degree CA, 500 RPM

165

Oil Film Thickness Standard Top Ring, 500 RPM, 450 Degrees

 

I
- t
u D ° '

 

 

nun—.—

5 Q Top Compression Ring

 

huh—E. -,

J __,

 

. ..\.~e-.~.~.~.....~~r-\'~~- M ~-‘

Mu.“

 

 

 

 

 

 

 

 

0 5 1O 15 20 , 25 30 35 40 45 50
Microns

 

 

7 PSI Vacuum - - - -2 PSI Boost ---—-8 PSI Boost
—Top Compression Ring —Second Compression Ring

 

FIGURE 105. Oil ﬁlm thickness measured at 450 degree CA, 500 RPM

Oil Film Thickness Standard Top Ring, 500 RPM, BDC Expansion
Stroke

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ .-,..-r~--h~-~ .-- _ q - _ ..—.' -2. '” .. “H; ' "' " -. ..--..--- -,- -~~--' ~~- -~
g Top Compression Ring
1‘1- . ,. _..’..,_..._ c. ‘. ; - . - ”."L H ' -----,.......-.-..- if“
,rf/ Second Compression Ring i
..------,.-.:2» ----- '" ~ ,M---
<3“... '- ___________________
I I I I ----- i‘ I I l I l

 

Microns

 

 

W7 PSI Vacuum - - - -2 PSI Boost --—--8 PSI Boost
-—Top Compression Ring —Second Compression Ring

 

FIGURE 106. Oil ﬁlm thickness measured at 540 degree CA, 500 RPM

166

 

 

9.-.". Film Thickness Standard Top Ring, 500 RPM, 630 Degrees _

 

.-_
--.
0.-
--
-D'.
,-
o-.-

. . L< Top Compression Ring i

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

ii {
‘. g g .5’:
Second Compression Ring
0 5 10 15 20 mag”. 30 35 4o 45 50

 

7 PSI Vacuum - - - -2 PSI Boost ---8 PSI Boost
—-Top Compression Ring —Second Compression Ring

FIGURE 107. Oil ﬁlm thickness measured at 630 degree CA, 500 RPM

 

 

Oil Film Thickness Standard Top Ring, 1500 RPM, TDC Intake

 

 

 

 

 

 

 

 

 

 

 

 

 

Stroke
Top Compression Ring
Second Compression Ring
0 5 10 15 20 25 30 35 4O 45 50
Microns
----- 7 PSI Vacuum - - - '2 PSI Boost ---8 PSI Boost

-—Top Compression Rin4g ---Second Compression Ring

 

FIGURE 108. Oil ﬁlm thickness measured at 0 degree CA, 1500 RPM

167

 

 

 

Oil Film Thickness Standard Top Ring, 1500 RPM, 90 Degrees
.1”)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Microns

~~~~~~ 7 PSI Vacuum - - - -2 PSI Boost —-~8 PSI Boost
—Top Compression Ring —Second Compression Ring

FIGURE 109. Oil ﬁlm thickness measured at 90 degree CA, 1500 RPM

 

 

 

Oil Film Thickness Standard Top Ring, 1500 RPM, BDC Intake
Stroke

1
l
nnnnnnn
.............. I
............... I

A

 

Top Compression Ring

 

 

 

 

 

 

 

 

....-- I
W ‘i
I
I
m ‘qu—[Q'1'1’F «n--.--.“--..r-----—.~ r.».-— v.. ..

I
Second Compression Ring ,
I

  

 

 

 

 

Microns

 

 

7 PSI Vacuum - . - '2 PSI Boost ---8 PSI Boost
—Top Compression Ring —Second Compression Ring

 

 

FIGURE 110. Oil ﬁlm thickness measured at 180 degree CA, 1500 RPM

168

Oil Film Thickness Standard Top Ring, 1500 RPM. 270 Degrees

 

 

 

 

 

 

 

 

 

 

 

 

l I I l I I I n I I

0 5 10 15 20 25 30 35 40 45 50
Microns
7 PSI Vacuum - - - -2 PSI Boost ...3 PSI Boost

 

—Top Compression Ring —Second Compression Ring

FIGURE 111. Oil ﬁlm thickness measured at 270 degree CA, 1500 RPM

Oil Film Thickness Standard Top Ring, 1500 RPM, TDC

 

 

 

 

 

 

0 5 10 15 20 _25 30 35 40 45 50
MICI’OI‘IS

. 7 PSI Vacuum - - - -2 PSI Boost —---8 PSI Boost

--To Com ression Rin —Second Com ression Rin

 

FIGURE 112. Oil ﬁlm thickness measured at 360 degree CA, 1500 RPM

169

 

,_._.__ _...‘

Oil Film Thickness Standard Top Ring, 1500 RPM, 450 degrees

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- . . . .7. ................ ’. ; . :‘r‘r ....... ". ‘1
Top Compression Ring 1
~-—..___.,-___’_ M
“Ra... _ - T '- 11:12“- , " ‘. ,,,,, . I
W
- --’ y.“ i
- - '1’ g w"W“n,”mm-...---..—~..~-»—--»-w""”"5!m ---------- —----
Second Compression Ring
-W ...- -- .. _ I
l l I I %
0 25 30 35 4O 45 5O
Microns
7 PSI Vacuum - - - .2 PSI Boost ---8 PSI Boost

 

—Top Compression Ring —Second Compression Ring

 

FIGURE 113. Oil ﬁlm thickness measured at 450 degree CA, 1500 RPM

Oil Film Thickness Standard Top Ring, 1500 RPM, BDC
Expansion Stroke

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

I C Top Compression Ring
R— ,
L.
I
Second Compression Ring
a...
l l l l l l l I l 1
0 5 1O 15 20 25 30 35 40 45 50
Microns
7 PSI Vacuum ' - - -2 PSI Boost ....3 PSI Boost

—Top Compression Ring —Second Compression Ring

 

FIGURE 114. Oil ﬁlm thickness measured at 540 degree CA, 1500 RPM

170

 

Oil Film Thickness Standard Top Ring, 1500 RPM, 630 degrees

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I I n I I" I I r I I

 

0 5 10 15 20 .25 30 35 40 45 5O
Microns
7 PSI Vacuum - - - -2 PSI Boost --—8 PSI Boost

 

 

—Top Compression Ring —Second Compression Ring

FIGURE 115. Oil ﬁlm thickness measured at 630 degree CA, 1500 RPM

 

Figures 116 - 123 show the cyclic variation of oil ﬁlm thickness of the
measurement and of the prediction using the new lubrication model. High
discrepancy is observed at the comparison especially at 500 RPM for both the top
and second piston rings (Figures 116 to 119). For the top ring oil ﬁlm thickness at
1500 RPM (Figures 121 and 123), a small degree of similarity in the cyclic
variation characteristic is observed except at the BDC, where the model under-
predicts the oil ﬁlm thickness. The sources of error are under investigation for both
the experiment and computation. It is to point out here that the amount of data
comparison at this stage is rather inadequate to derive conclusive observation.

More measurements and comparisons are recommended. The resolution of the

171

measured oil ﬁlm thickness needs to be increased in order to capture the detail of
cyclic oil ﬁlm variation. At this stage, only 8 measurements are made over a four-
stroke engine cycle, at crank angles of 0, 90, 180, 270, 360, 450, 540 and 630
degree (720 is the same as 0 degree) for the preliminary comparison with the
predicted results. In addition, the technique to determine measured minimum oil
ﬁlm thicknesses and ring edges needs to be improved. The ring edges need to be
located more precisely from the recorded LIF images so that the oil ﬁlm thickness

proﬁle across ring faces can be determined.

172

Minimum Film Thickness at the Top Ring Face
500 RPM 2PSI Boost

N

 

01 0)

I31
//
///,/

1;
/
\\\\
,8
/

Film Thickness (micron)
w

 

 

 

’ \
\ / \ // \\
_ 3‘: ............. _ Z .
"a. I I \
1 L \x \
l J .. \ V
0 L J o 1
O 90 180 270 360 450 540 630 720
Crank Angle

 

—B— 2PSI Boost(Measured) -- - - 2PSI Boost(Predicted)

 

 

 

FIGURE 116. Cyclic top ring oil ﬁlm thickness at 500 RPM 2 PSI Boost

Minimum Film Thickness at the 2nd Ring Face
500 RPM 2PSI Boost

 

 

 

 

6 ~ ~— —

2 C1 5,7,1 - ﬂ

.2 2

a \

g3 '- \ z“ \\/

ﬁ \ / -'

£2 " ‘ // /

.1 _ / a,
0 1 l: 1 i

O 90 180 270 720
Crank Angle

 

 

—B— 2PSI Boost(Measured) ---- 2PSI Boost(Predicted)
FIGURE 117. Cyclic 2nd ring oil ﬁlm thickness at 500 RPM 2 PSI Boost

 

 

173

Minimum Film Thickness at the Top Ring Face
500 RPM 8PSl Boost

 

 

 

 

10

e 8 ‘
2
'§
V 6 ' //’A\ .
g [\I/ \ M
g 4 - \A\ / I
l- \\ / i
E . ..
E 2 [— /,r-" i.x.,..r' \‘ \K I”, , -- ._ farm, ~.. ‘

0 1 I: 1 ”any"; 1 t 1 j

0 90 180 270 360 450 540 630 720
Crank Angle

 

 

—a— 8PSI Boost(Measured) ---- 8PS| Boost(Predicted)
FIGURE 118. Cyclic top ring oil ﬁlm thickness at 500 RPM 8 PSI Boost

 

 

Minimum Film Thickness at the 2nd Ring Face
500 RPM 8PSI Boost

 

 

 

6 .. _ _ __
I
If ‘ (.r'“ N I
A5 " ’ .‘.. 3' i
C .' I
.2 A j '. A
E 4 E /A\\ A ._ ';/‘i
7; \ \ f/ .' f. E
m \ \, / 2 . / - l
2 3 h \ / K '- \ A; / '. g
x \' / / \.'\ / '. i
E \ / .- \ / 1 ',
- \ .1
*- 2 b \/ -' \a/ 13
u-1 ~ Ii
1‘
- - 4.
O L x I l I r 1 4

0 90 1 80 270 360 450 540 630 720

Crank Angle
—a— 8PSI Boost(Measured) ---- 8PS| Boost(Predicted)
FIGURE 119. Cyclic 2nd ring oil ﬁlm thickness at 500 RPM 8 PSI Boost

 

 

 

 

174

Minimum Film Thickness at the Top Ring Face

1500 RPM 2PSI Boost

5 -W _-.__..___ -. In______--._ WWW,
/’ \\ l

 

or
\q
\\
//
121’ /

Film Thickness (mmmn)
(A)
\\
\

N
"\

.C<
/

h

/
‘\
/.

 

 

:11."

1 z

0 . J a L . . . l
0 90 1 80 270 360 450 540 630 720

Crank Angle
-e-— 2PSI Boost(Measured) ---- 2PSI Boost(Predicted)
FIGURE 120. Cyclic top ring oil ﬁlm thickness at 1500 RPM 2 PSI Boost

 

 

 

 

Minimum Film Thickness at the 2nd Ring Face
1500 RPM 2PSI Boost

 

 

 

9 __

8 _ r, ..... ~..,‘ ff“, . I
57 -
U 1
E5 “ . I
”’5 ~ - '- 131 ~
(0 6K H/Q. .' I. x/ \' . '- m
a) _ , ,
E 4 _ V ?\\\ -/,‘.’R\ // :‘\\ ’r: ;
E : \\ ,/ ‘3 \ , .5 / Z:
l- 3 “ 3 \ ,// '. \ / .' /
g :' 2 \3' r \ ti
E2 T- ‘ .-.. ”I,

1 ~

O 1 1 1 L 1 1 1 J

0 90 1 80 270 360 450 540 630 720
Crank Angle

 

—B— 2PSI Boost(Measured) ---- 2PSI Boost(Predicted)
FIGURE 121. Cyclic 2nd ring oil ﬁlm thickness at 1500 RPM 2 PSI Boost

 

 

 

175

Minimum Film Thickness at the Top Ring Face
1500 RPM 8PSI Boost

 

(n
f

.5
. - /
\>

 

Film Thickness (micron)
w
\
\“1
J’
</
"1\
'1
“x
*1

 

 

 

2 1
AK,
1 i
l
0 1 1 1 1 1 1 1 l
O 90 180 270 360 450 540 630 720
Crank Angle

 

 

 

+ 8PSI Boost(Measured) ---- 8PSI Boost(Predicted)
FIGURE 122. Cyclic top ring oil ﬁlm thickness at 1500 RPM 8 PSI Boost

 

Minimum Film Thickness at the 2nd Ring Face
1500 RPM 8PSI Boost

 

I" i... ...-o u. l
L Jr a... ,f" ‘3. g
I 2 ...! a" l

mm.o

OI
I

 

Film Thickness
00 h
l/]
/

J

N
f

 

 

 

 

1
O 1 1 1 1 1 1 1
O 90 180 270 360 450 540 630 720
Crank ArLLle
—es— 8PS| Boost(Measured) - - - - 8PSl Boost(Predicted)

 

 

 

FIGURE 123. Cyclic 2nd ring oil ﬁlm thickness at 1500 RPM 8 PSI Boost

176

CHAPTER 7 SUMMARY

Theoretical models have been developed to address the engine tribological
phenomena that occur at the conjunction between the cylinder bore and the
piston skirt, and between the cylinder bore and the piston rings. These models
were implemented in a computational engine simulation system, CASE, and its
capabilities were demonstrated in simulating comprehensive engine
performances that couple the analyses of ring/bore wear, oil consumption, blowby
and backﬂow, and piston dynamics/elastohydrodynamics (EHL). This chapter
provides the conclusions and recommendations of each part of this work, and
comments on the importance of integrating them together in order to develop an

effective computational tool for engine study.
7.1 Partially Flooded Lubrication

7.1 .1 Conclusions

A new piston ring/bore lubrication computational model has been
developed, replacing an existing model that only accounts for the fully ﬂooded
condition. As was demonstrated here, this model now accounts for a wider range
of lubrication conditions that could exist in a piston ring/bore system. Five distinct

lubrication regimes are accounted for by this model:
1. Fully ﬂooded hydrodynamic
2. Partially ﬂooded hydrodynamic

3. Fully ﬂooded mixed

177

4. Partially ﬂooded mixed, and

5. Full boundary (metal-to-metal).

The three-dimensional feature of this model has the ability to account for
asymmetrical variation of ring/bore wear and physical boundary conditions. Such
variations are commonly found in operating engines due to phenomena such as

piston secondary motion (tilt), bore distortion, and bore temperature distribution.

7.1 .2 Recommendations

Even though the new lubrication model satisﬁes both the lubricant ﬂow
continuity and force equilibrium at piston/ring bore interaction, some results show
abrupt or "step-wise" change from one node to another due to the limited
resolution and computational simpliﬁcations. Therefore, future recommendations
for this model will be to include higher nodal resolution particularly across the ring
circumference, or to include detailed numerical ﬂow continuity analysis across the
entire ring surface that also involves the trailing edge ﬁlm attach/detach
interaction. This partially ﬂooded lubrication model should be ultimately
implemented into a multi-ring piston assembly system for a more realistic engine
simulation. Considering the high dynamic ring/groove interaction, especially in
diesel engines, the wear phenomenon at the ring/groove region needs to be

included.

178

7.2 Oil Vaporization Effect on Partially Flooded Lubrication

7.2.1 Conclusions

A coupled model of oil vaporization and partially ﬂooded mixed lubrication
has been presented along with its application on a modern diesel engine system.
In its application on a diesel engine system, unique results have been predicted
by this model in comparison with other piston ring lubrication models. Results with
CASE demonstrate its uniqueness and the importance of including oil evaporative
analysis in partially ﬂooded analysis. Simulations successfully show the
appearance of local bore wear pockets induced from locally concentrated higher-
than-surrounding bore temperature, which potentially represents the wear-
inducing interactions between the fuel injection plumes and the bore wear. The
application is further conducted in temperature-sensitive bore wear analysis in
order to determine the range of critical temperature that will cause excessive bore
wear. Observation from the analysis shows a drastic increase of the bore wear

pocket at a small degree of temperature rise at the critical temperature range

(about 50°C increase from the surrounding). This ﬁnding is very critical in the
development of diesel engine injection spray systems, considering the potential of
temperature-induced high wear from the injection plumes on the cylinder bore that

can disrupt the entire engine performance and structure strength.

Examination of the results also offers useful information regarding the

factors that contribute to the thinning of oil ﬁlm. The ﬁrst analysis uses the same

bore temperature distribution where hot spot temperature is about 280°C (100°C

179

higher than surrounding area). Here, all results, with and without the oil
evaporation analysis, show a different degree of wear increase at hot spots,
indicating the increase of local asperity contact pressure. The increase of asperity
contact implies that ring/bore surfaces are interacting with a thinner oil ﬁlm.
Without evaporative inﬂuence, the thinning of oil ﬁlm thickness is minimal because
it is attributed merely to the oil viscosity reduction at high temperature. By
including evaporation analysis, the result indicates substantial oil ﬁlm thickness
loss due to evaporation. The dominant effect of evaporation on oil ﬁlm loss at high
temperature, in comparison to viscosity reduction, implies that a piston ring/bore
lubrication model with oil evaporation analysis can better capture the tribological
behavior of an engine operating in extreme conditions. One sees the same
behavior in most modern diesel engines, which have a higher compression ratio

than gasoline engines.

7.2.2 Recommendations

The evaporation analysis can be improved to represent the physics of the
heat transfer system more closely. For example, one can include transient heat
conduction analysis on the heat source and bore temperature interaction and
perform continuous engine cycle analysis to verify if the prediction from the
evaporation-coupled analysis is realistic in comparison with practical engine
operation. The evaporation analysis can also be extended to account for
composite oil instead of single-specie oil if oil data are available. For the diesel
engine application, experimental effort is necessary to verify the predicted results

on the temperature-induced bore wear pocket.

180

7.3 Piston EHL Model

7.3.1 Conclusions

A stand-alone model for the three-dimensional piston simulation has been
developed to be used in the study of piston design and analysis. This model
improves on other piston models (seen in publication) that usually do not have all
built-in ﬁnite element tools for a complete piston analysis to be conducted
automatically. The present model is considered a more advanced and efﬁcient
CAE tool: it comprises built-in modules that automatically construct the piston
ﬁnite element solid model, solve for the thermal and static analysis, and simulate
the entire cyclic analysis of piston dynamics coupled with piston skirt

elastohydrodynamic lubrication.

Preliminary testing on a production engine demonstrates the capability of
this model in performing a comprehensive piston study. The simulation results
show that cylinder gas pressure load and engine speed play signiﬁcant and
interdependent roles in determining piston trajectory and skirt elastohydrodynamic

lubﬁcaﬁon.

7.3.2 Recommendations

A comparison of the ﬁnite element analysis results, using the simpliﬁed
piston model and the actual model, shows that the simpliﬁed model must be
improved in order to capture the physics more closely, for example including the
piston pin holes and a more detailed pin boss geometry. Othenivise, the piston

model can be extended to have the capability of including an actual piston

181

geometry rather than the simpliﬁed geometry if computational time is not an issue
in a particular analysis. On top of that, the multi-mode lubrication wear model,
which has been implemented in the ring model, needs to be extended to the
piston skirt lubrication calculation to account for metal-to-metal contact and piston

skirt wear.

7.4 Experimental Validation

7.4.1 Conclusions

Experiments have been conducted for the comparison of the piston ring
pack gas pressures and the oil ﬁlm thickness predicted using the computational
models. Predictions agree reasonably well with the oil ﬁlm thickness and inter-ring
gas pressure measurements in the optical engine. The experimental data also
shows that the fully ﬂooded condition is a valid assumption for piston rings below
the top compression ring as it is seen that a substantial amount of oil is trapped at

the inter-ring region, suppling oil for full lubrication on those piston rings.

7.4.2 Recommendations

The discrepancy of the inter-ring gas pressure near the bottom of the
piston ring pack can be improved by incorporating the input of measured
crankcase pressures or a new model that can estimate the crankcase dynamic
pressures. This will provide accurate boundary conditions for the computation of

the inter-ring gas pressures.

182

The current measurements of oil ﬁlm thickness over an engine cycle are
only conducted at eight different degree crank angles (0,90,180,270,360,450,540
and 630). As a result, any conclusive observation with respect to the prediction is
difﬁcult to make. More measurement need to be acquired for further validation of

the new lubrication model.

7.5 Closure

The developments of several engine tribological models, from theory to
computational module, have been presented in this dissertation. Even though
each model represents an important subsystem of an engine, proper integrating of
these tribological models into a main engine system, along with other subsystems
of gas dynamics, heat transfer and structural mechanics and dynamics, proves to
be a more crucial step in arriving a comprehensive engine simulation system that
can better represent a real engine performance. Hence, every subsystem is
essentially dependent on other subsystem for it to be applicable in the engine
analysis. The computational efﬁciency, accuracy, and the applicability of the
improved CASE system have been validated through various practical engine
applications. The uniqueness of this developed engine system can be

summarized as follows:

1. its advanced technology in predicting three-dimensional ring/bore wear
(from literature survey: it is the only known model in the engine research

ﬁeld capable of doing this),

183

2. its advanced technology in integrating major built-in modules (ﬁnite
element and dynamic) into a single simulation so that a complete
comprehensive piston EHL performance can be performed

automatically,

3. its capability of simulating ﬁve lubrication regimes at the ring/bore

interface,
4. its capability of simulating four piston kinematic modes, and

5. experimental veriﬁcation using the optical engine system.

The robustness and validation of the new engine system CASE potentially
allows it to be a very crucial tool in the engine design, particularly when it is
integrated into an optimization system. Consider its computational efﬁciency in
which an engine simulation can be conducted at a relatively short time (about 15
minutes per engine operating condition on a 2.4 GHz Pentium 4 PC), a parametric
study to arrive an optimized conﬁguration of engine system for desired
performance output, or an engine performance study over a range of engine
speeds and loads, can be accomplished within a period of days. This essentially
implies that the application of CASE can help in reducing a signiﬁcant amount of
time and cost in the conventional design process (usually takes months or years)
and contribute to the ultimate production of modern engines with higher durability,

higher efﬁciency and lower emission.

184

6.)

7.)

a.)

BIBLIOGRAPHY

Allen, D.G., Dudley, B.R., Middleton, V. and Parker, DA, 1975, “Prediction
of Piston Ring-Cylinder Bore Oil Film Thickness in Two Particular Engines
and Correlation with Experimental Evidence,” Proc IMechE Conference on
Piston Ring Scufﬁng, Paper No. 073/7 5. -

Archad, J.F., 1953, “Contact and Rubbing of Flat Surfaces,” J. of Applied
Physics, Vol. 24, No. 8, pp. 981-988.

Audette, W.E. Ill, and Wong, V.W., 1999, “A Model for Estimating Oil
Vaporization from the Cylinder Liner as a Contributing Mechanism to
Engine Oil Consumption,” SAE Paper 1999-01-1520.

Audette, W.E. Ill and Wong, V.W, 2000, “Oil Transport Along the Engine
Cylinder Liner - Coupling Vaporization and Film Thickness Analysis,” 12th
International Colloquium Tribology, Tribology 2000 - Plus, January 11th -
13th, 2000, W. J. Bartz (publisher), Vol. 1.

Baker, A.J.S., Dowson, D. and Strachan, P., 1973, “Dynamic Operating
Factors in Piston Rings,” International Symposium of Marine Engineers,
Paper 256, Tokyo, Japan.

Barkman, DA, 1992, “ An Evaluation of the Parameters Involved in
Developing A Radiotracer Based Wear Diagnostics System,” MS. Thesis,
Mechanical Engineering Department, Michigan State University, East
Lansing, MI, USA.

Bishop, l.N., 1960, “Effect of Design Variables on Friction and Economy,”
from Engine Oil Effects on Vehicle Fuel Economy - Progress in Technology,
Series No. 27, SAE Paper No. 812A.

Brombolich, L.J., 1988, “Effect of Cylinder Distortion and Piston Ring
Design on Oil Consumption and Friction Losses in Automotive Engines,”
Contract No. DE-AC02-8GCE-90236, Report to US Department of Energy.

185

9.)

1o.)

11.)

12.)

13.)

14.)

15.)

16.)

17.)

18.)

19.)

Brown, SR. and Hamilton, GM.,1977, “The Partially Lubricated Piston
Ring,” J. Mech. Eng. Sci., Vol. 19, No. 2, pp. 81-89.

Bush, AW. and Gibson, RD, 1980, “The Effect of Surface Roughness and
Elastic Deformation in Hydrodynamic Lubrication - A Perturbation
Approach,” Surface Roughness Effects in Hydrodynamics and Mixed
Lubrication, ASME Publication, pp. 51-73.

Cameron, A., 1966, Principles of Lubrication, John Wiley and Sons Inc.
New York.

Cameron, A., 1971, Basic Lubrication Theory, Harlow, Longman.

CASE Theoretical Manual, 2003, Mid Michigan Research, Okemos, Ml.

CASE User’s Manual, 2003, Mid Michigan Research, Okemos, Ml.

Castleman, RA, 1936, “A Hydrodynamic Theory of Piston Ring
Lubrication,” Physics, Vol. 7, pp. 364-367.

Chow, PL. and Saibel, EA, 1978, “On the Roughness Effect in
Hydrodynamic Lubrication,” Trans. of the ASME, Vol. 100, pp. 176-180.

Christensen, H., 1972, “A Theory of Mixed Lubrication,” Proc. IMechE. Vol.
186. 90421-430.

Chui, B.K., 2001, “Computational Analysis of Oil Consumption and Piston
Ring Wear,” MS. Thesis, Mechanical Engineering Department, Michigan
State University, East Lansing, MI, USA.

Chui, B.K. and Schock, H.J., 2002, "Computational Analysis of Piston-Ring
Wear and Oil Consumption for an Internal Combustion Engine,”
Proceedings of the 2002 Fall Technical Conference of the ASME Internal
Combustion Engine Division, September 2002, New Orleans, Louisiana,
Paper No.: lCEF2002-529, pp. 461-469.

186

20.)

21.)

22.)

23.)

24.)

25.)

26.)

27.)

28.)

Chung, Y.S., 1992, “Development and Analysis of a Fire Ring Wear Model
for a Piston Engine,” Ph.D. Thesis, Mechanical Engineering Department,
Michigan State University, East Lansing, MI, USA.

Chung, Y.S., Schock, H.J., and Brombolich, L.J., 1993', “Fire Ring Wear
Analysis for a Piston Engine,” SAE Paper 930797.

.Cleveland, AB. and Bishop, IN, 1960, “Several Possible Paths to
Improved Part-load Economy of Spark-ignition Engines,” SAE Paper 150A.

Dowson, D., Economou, P.N., Ruddy, B.L., Strachan, P.J., and Baker, A.J.
S., 1979, “Piston Ring Lubrication - Part 2. Theoretical Analysis of a Single
Ring and a Complete Ring Pack,” Energy Conservation Through Fluid Film
Lubrication Technology: Frontiers in Research and Design, Winter Annual
Meeting of ASME, pp. 23-52.

Dowson, D., Ruddy, B.L., and Economou, RN, 1983, “The
Elastohydrodynamic Lubrication of Piston Rings,“ Proc. R. Soc. Lond., Ser.
A, Vol. 386, pp. 409-430.

Dursunkaya, Z. and Keribar, R., 1992, “Simulation of Secondary Dynamics
of Articulated and Conventional Piston Assemblies,” SAE Paper 920484.

Dursunkaya, Z., Keribar, R. and Ganapathy, V., 1994, “A Model of Piston
Secondary Motion and Elastohydrodynamic Skirt Lubrication,” ASME J. of
Trib., Vol. 116, October, pp. 777-785.

Eilon, S. and Saunders, O. A., 1957, “A Study of Piston-Ring Lubrication,”
Proc. Inst. Mech. Engrs., Vol. 171, pp. 427-443.

Ejakov, MA, 1998, “Ring Pack Behavior and Oil Consumption Modeling in
IC Engines,” Ph.D. Thesis, Mechanical Engineering Department, Michigan
State University, East Lansing, MI, USA.

187

29.)

30.)

31.)

32.)

33.)

34.)

35.)

36.)

37.)

Ejakov, M.A., Schock, H.J., Brombolich, L.J. and Williams, R.L., 1996,
“Comparison Between Predicted and Measured Inter-Ring Gas Pressures
and Analysis of Factors Affecting Them,” DAR 96—00858, Ford Design
Analysis Report, November.

Ejakov, M. A., Hascher, H. G., Schock, H. J., Brombolich, L. J., Trinker, F.
H., and LoRusso, J. A., 1996, “Simulation Analysis of Ring Pack Behavior
in a Deactivated Cylinder,” ICE-Vol. 27-2, 1996 Fall Technical Conference,
Vol. 2, ASME.

Ejakov, M. A., Schock, H. J., Brombolich, L. J., Carlstrom, C. M., and
Williams, R. L., 1997, “Simulation Analysis of Inter-Ring Gas Pressure and
Ring Dynamics and Their Effect on Blow-by,” ICE-Vol. 29-2, 1997 Fall
Technical Conference, Vol. 2, ASME Paper No.: 97-lCE-50.

Ejakov, M.A., Schock, H.J. and Brombolich, L.J., 1998, “Modeling of Ring
Twist for an IC Engine,” SAE Paper 982693

Ejakov, M.A., Diaz, AR. and Schock, H.J., 1999, “Numerical Optimization
of Ring-Pack Behavior,” SAE Paper 1999-01-1521

Ejakov, MA, 2001, "Modeling of Axial and Circumferential Ring Pack
Lubrication,” ICE-Vol 37-3, 2001 Fall Technical Conference, Vol. 3, ASME
Paper No. 2001-lCE—433.

Ejakov, M.A., Yeager, DA, and Kayupov, M., 2003, “Reliability and
Robustness Analysis of Engine Ring-Pack Performance,” SAE Paper
2000-01-1221.

Fedewa, A. and Schock, H.J., 2004, “Tribological Phenomena for
Advanced Diesel Engines,” Contract No. DAAE07-03-C-L021, Report to
US Army Tank Automotive Command, November 15.

Fielding, B.J. and Skorecki, J., 1969-1970, “Identiﬁcation of Mechanical
Sources of Noise in a Diesel Engine: Sound Originating from Piston Slap,”
Proc. Inst. Mech. Eng., Vol. 184, Part I, No. 46, pp. 859-871.

188

38.)

39.)

4o.)

41.)

42.)

43.)

44.)

45.)

46.)

47.)

Fuller, DD, 1978, Theory and Practice of Lubrication for Engineers, 2nd
Ed., John Wiley & Sons, New York.

Furuhama, 8., 1959, “A Dynamic Theory of Piston-Ring Lubrication (1st
Report: Calculation),” BuI. of JSME, Vol. 2, No. 7, pp. 423-428.

Furuhama, S., 1960, “A Dynamic Theory of Piston-Ring Lubrication (2nd
Report: Experiment),” Bul. of JSME, Vol. 3, No. 10, pp. 291-297.

Furuhama, S., 1961, “A Dynamic Theory of Piston-Ring Lubrication (3rd
Report: Measurement of Oil Film Thickness),” BuI. of JSME, Vol. 4, No. 16,
pp. 744-752.

Furuhama, S., Takiguchi, M. and Tomizawa, K., 1981, “Effect of Piston and
Piston Ring Designs on the Piston Friction Forces in Diesel Engines,” SAE
Paper810977.

Glidewell, J., and Korcek, S., 1998, “Piston Ring/Cylinder Bore Friction
Under Flooded and Starved Lubrication Using Fresh and Aged Engine
Oils,” SAE Paper 982659.

Goenka, PK. and Meemik, PR, 1992, “Lubrication Analysis of Piston
Skirts,” SAE Paper 920490.

Greenwood, J.A., and Tripp, J.H., 1971, “The Contact of Two Nominally
Flat Surfaces,” Proc. Inst. Mech. Engrs., Vol. 185, pp. 625-632.

Grice, N., 1990, “An Experimental and Theoretical Study of Piston Ring
Lubrication,” Ph. D. Thesis, Lancashire Polytechnic in collaboration with
B. P. Research International, UK, October.

Grifﬁths, W.J. and Skorecki, J., 1964, “Some Aspects of Vibration on a
Single Cylinder Diesel Engine,” J. Sound Vibration, Vol. 1, No. 4, pp. 345-
364.

189

48.)

49.)

50.)

51.)

52.)

53.)

54.)

55.)

56.)

57.)

Hamilton, GM. and Moore, S.L., 1974, “The Lubrication of Piston Rings.
First Paper. Measurement of the Oil Film Thickness between the Piston

Rings and Liner of a Small Diesel Engine,” Proc. Inst. Mech. Engrs., Vol.
188. PP. 253-261.

Hamilton, GM., Moore, S.L., 1974, “The Lubrication of Piston Rings.
Second Paper. Comparison between Measured and Calculated
Thicknesses of the Oil Film Lubricating Piston Rings,” Proc. Inst. Mech.
Engrs., Vol. 188, pp. 262-268.

Hersey, M., 1966, Theory and Research in Lubrication; Foundations for
Future Developments, John Wiley 8 Sons Inc., New York

Heywood, J.B., 1988, Internal Combustion Engine Fundamentals,
McGraw-Hill, Inc, New York

Hoult, D.P. et al., 1988, “Calibration of Laser Fluorescence Measurement
of Lubricant Film Thickness in Engines,” SAE Paper 881587.

lshizuki, Y., Sato, R., and Takase, K., 1981, “Effect of Cylinder Liner Wear
on Oil Consumption in Heavy Duty Diesel Engines,” SAE Paper 810931.

Jeng, Y.R., 1992, “Friction and Lubrication Analysis of a Piston-Ring Pack,”
SAE Paper 920492.

Kajiwara, H., Fujioka, Y. and Negishi, H., 2003, “Prediction of
Temperatures on Pistons with Cooling Gallery in Diesel Engines using
CFD Tool,” SAE Paper 2003-01-0986.

Keribar, R. and Dursunkaya, Z., 1992, “A Comprehensive Model of Piston
Skirt Lubrication,” SAE Paper 920483.

Keribar, R., Dursunkaya, Z., and Ganapathy, V., 1993, “An Integrated
Design Analysis Methodology to Address Piston Tribological Issues,” SAE
Paper930793.

190

58.)

59.)

60.)

61.)

62.)

63.)

64.)

65.)

66.)

67.)

Knoll, 6D. and Peeken, H.J., 1982, “Hydrodynamic Lubrication of Piston
Skirts,” J. of Lubrication Technology, Vol. 104, pp. 504-509, October.

Knoll, G, Peeken, H., Lechtape-Gruter, R. and Lang, J., 1994, “Computer
Aided Simulation of Piston and Piston Ring Dynamics,” from Heavy Duty
Engines: A Look at the Future, ASME 1994 Conference, ICE-Vol. 22, pp.
301-309.

Laws, A.M., Parker, DA. and Turner, B., 1973, “Piston Movement in Diesel
Engine,” Paper No. 33, CIMAC Conference, Washington DC, pp. 809-
831.

Lee, Y.Z. and Ludema, KG, 1990, “The Shared-Load Wear Model in
Lubricated Sliding Scufﬁng Criteria and Wear Coefﬁcients,” WEAR,
Vol. 138. PD. 13.

Li, OH, 1986, “Thermoelastic Behavior of an Aluminum Diesel Engine
Piston,” SAE Paper 860136.

Li, D.F., Rohde, SM, and Ezzat, HA, 1983, “An Automotive Piston
Lubrication Model,” ASLE Transactions, Vol. 26, No. 2, pp. 151-160.

Liu, S.H., Li, J., Zhou, L.B., Rong, W., 1996, “An Experimental Investigation
of the Oil Film Lubricating Piston Rings,” SAE Paper 961912.

Lloyd, T., 1968, “The Hydrodynamic Lubrication of Piston Rings,” Proc.
Inst. Mech. Engrs, Vol. 183, pp. 28-34.

Ludema, KC, 1996, Friction, Wear, Lubrication, A Textbook in Tribology,
CRC Press, New York.

Ma, 2., Henein, N.A., Bryzik, W., and Glidewell, J., 1998, “Break-In Liner
Wear and Piston Ring Assembly Friction in a Spark-Ignited Engine,” STLE
Preprinted No. 98-AM-3D-1, Presentation at 53rd STLE Annual Meeting,
Detroit, MI.

191

68.)

69.)

7o.)

71.)

72.)

73.)

74.)

75.)

76.)

Ma, M-T, Smith, E.H. and Sherrington, I., 1995, “A Three-Dimensional
Analysis of Piston Ring Lubrication. Part“ 1: Modeling,” Proc. Inst. Mech.
Engrs., Part J, Vol. 209, pp. 1-14.

Ma, M-T, Smith, E.H. and Sherrington, l., 1995, “A Three-Dimensional
Analysis of Piston Ring Lubrication. Part 2: Sensitivity Analysis,” Proc. Inst.
Mech. Engrs., Part J, Vol. 209, pp. 15—27.

Ma, M-T, Smith, E.H. and Sherrington, I., 1996, “Implementation of an
algorithm to Model the Starved Lubrication of a Piston-Ring in Distorted
Bores; Prediction of Oil Flow and Onset of Gas Blow-by,” Proc. Inst. Mech.
Engrs., Part J, Vol. 210, pp. 29-44.

Ma, M-T, Sherrington, l. and Smith, E.H., 1997, “Analysis of Lubrication
and Friction for a Complete Piston-Ring Pack with an Improved Oil
Availability Model. Part 1: Circumferentially Uniform Film,” Proc. Inst.
Mech. Engrs., Part J, Vol. 211, pp. 1-15.

Ma, M-T, Smith, E.H. and Sherrington, l., 1997, “Analysis of Lubrication
and Friction for a Complete Piston-Ring Pack with an Improved Oil
Availability Model. Part 2: Circumferentially Variable Film,” Proc. Inst.
Mech. Engrs., Part J, Vol. 211 , pp. 17-27.

McGeehan, J.A., 1978, “Literature Review of the Effects of Piston Ring
Friction and Lubricating Oil Viscosity on Fuel Economy,” from Engine Oil

Effects on Vehicle Fuel Economy - Progress in Technology, Series No. 27,
SAE Paper 780673.

McGeehan, J.A., 1979, “A Survey of the Mechanical Design Factors
Affecting Engine Oil Consumption,” SAE Paper 790864.

Moore, S.L. and Hamilton, GM, 1978, “The Starved Lubrication of Piston
Rings in a Diesel Engine,” J. of Mech. Engrg. Sci, Vol. 20, No. 6, p. 345.

Munro, R., and Parker, A., 1975, “Transverse Movement Analysis and Its
Inﬂuence on Diesel Piston Design,” SAE Paper 750800.

192

77.)

78.)

79.)

80.)

81.)

82.)

83.)

84.)

85.)

86.)

Munson, B.R., Young, DR and Okiishi, T.H., 1998, Fundamentals of Fluid
Mechanics, 3rd Ed. John Wiley & Sons, Inc., New York.

Nakayama, K., Seki, T., Takiguchi, M., Someya, T., and Furuhama, S.,
1998, “Effect of Oil Ring Geometry on Oil Film Thickness in the
Circumferential Direction of the Cylinder,” SAE Paper 982578.

Oguri, A., Totii, H., Hosokawa, T. and Nakada, M., 1988, “Development of
Analysis Model for Piston Movement,” SAE Paper 885031.

Oh, K.P., Li, OH. and Goenka, PK, 1987, “Elastohydrodynamic
Lubrication of Piston Skirts,” ASME J. of Tribology, Vol. 109, pp. 356-362.

Parker, DA, 1989, “The Tribology of Automotive Components:
Development of Veriﬁed Predictive Design Techniques,” Fourth BP/IMechE
Tribology Lecture, Part D, J. of Automobile Engineering.

Patir, N. and Cheng, HS. 1978, “An Average Flow Model for Determining
Effects of Three-Dimensional Roughness on Partial Hydrodynamic
Lubrication,” ASME J. of Lubrication Technology. Vol. 100, January, pp.
12-17.

Patir, N. and Cheng, HS. 1979, “Application of Average Flow Model to
Lubrication Between Rough Sliding Surfaces,” Vol. 101, April, pp. 220-230.

Pint, S., and Schock, H.J., 2000, “Design and Development of a Software
Module for Analysis of Three Dimensional Piston Ring Wear,” SAE Paper
2000-01 -0920.

PISTON DYNAMICS PROGRAM, 1991, Compu-Tec Engineering,
Chesterﬁeld, Missouri.

Ramchandani, M. and Whitehouse, ND, 1976, “Heat Transfer in a Piston
of a Four Stroke Diesel Engine,” SAE Paper 760007.

193

87.)

88.)

89.)

90.)

91.)

92.)

93.)

94.)

95.)

Reddi, MM, 1969, “Finite Element Solution of the Incompressible
Lubrication Problem,” J. of Lubrication Technology, July, pp. 524-533.

Reddi, MM. and Chu, T.Y., 1970, “Finite Element Solution of the Steady-
State Compressible Lubrication Problem,” J. of Lubrication Technology,
July, pp. 495-503.

Richardo, H. and Hempson, J.G.G, 1968, “The High Speed Internal
Combustion Engine,” Blackie, London, 5th Ed., pp. 139-141.

Richardson, DE, 1990, “The Development and Implementation of
Theoretical and Experimental Methods for Studying Oil Films in Engine
Cylinders,” Ph.D. Dissertation, University of Wisconsin, Madison,
Wisconsin.

Roehrie, MD, 1978, “Thermal Effects on Diesel Engine Pistons,” SAE
Paper780781.

Rogowski, AR, 1961, “Method of Measuring the Instantaneous Friction of
Piston Rings in a Firing Engine,” SAE Paper 379F.

Rohde, SM, 1980, “A Mixed Friction Model for Dynamically Loaded
Contacts with Application to Piston Ring Lubrication,” Surface Roughness
Effects in Hydrodynamics and Mixed Lubrication, ASME Publication, pp.
19-50.

Rosenberg, R.C., “General Friction Considerations for Engine Design,”
SAE Paper 821576.

Ryan, J.P., Wong, V.W., Lyon, R.H., Hoult, D.P., Sekiya, Y., Kobayashi, Y.,
and Aoyama, S., 1994, “Engine Experiments on the Effects of Design and

Operational Parameters on Piston Secondary Motion and Piston Slap,”
SAE Paper 940695.

194

96.)

97.)

98.)

99.)

100.)

101.)

102.)

103.)

Ruddy, B.L., Dowson, D., Economou, RN. and Baker, A.J.S., 1979, “Piston
Ring Lubrication - Part III: The Inﬂuence of Ring Dynamics and Ring Twist,”
Energy Conservation Through Fluid Film Lubrication Technology: Frontiers
in Research Design, Winter Annual Meeting of ASME, December, ASME
Publication, New York, pp. 191-215.

Ruddy, B.L., Parsons, B., Dowson, D. and Economou, RN, 1980, “The
Inﬂuence of Thermal Distortion and Wear of Piston Ring Grooves Upon the
Lubrication of Piston Rings in Diesel Engines,” Proc. 6th Leeds-Lyon
Symp. on Tribology: Thermal Effects in Tribology, Bury St., Edmunds, UK,
pp. 84-94.

Ruddy, B.L., Dowson, D. and Economou, RN, 1981, “A Theoretical
Analysis of the Twin-Land Type of Oil-Control Piston Ring,” J. Mechanical
Engineering Science, Vol. 23, No. 2, pp. 51 -62, 1981 .

Sander, W., Steidle, W. and Wacker, E., 1979, “Piston Movement and Its
Inﬂuence on Noise of Automotive Engines,” SAE Paper 790272.

Saugerud, O.T., Sandsmark, N. and Veritas, D., 1979, “Strength Analysis of
Thermally Loaded Engine Components by Two-Dimensional and Three-
Dimensional Finite Element Models,” SAE Paper 790820.

Schneider, E.W., Blossfeld, OH, and Balnaves, MA, 1988, “Effect of
Speed and Power Output on Piston Ring Wear in Diesel Engine,” SAE
Paper880672.

Stanley, R., Taraza, D., Henein, N. and Bryzik, W., 1999, “A Simpliﬁed
Friction Model of the Piston Ring Assembly,” SAE Paper 1999-01-0974.

Suguru, S., Masanao, S, Takiguchi, M., Ushijima, K. and Aoyama, S.,
2002, “Development of a Small LIF System for Oil Film Thickness
Measurement,” Proceedings of the 2002 Fall Technical Conference of the
ASME Internal Combustion Engine Division, September 2002, New
Orleans, Louisiana, Paper No.: lCEF2002-527, pp. 437-443.

195

104.)

105.)

106.)

107.)

108.)

109.)

110.)

111.)

112.)

113.)

Sui, PC. and Ariga, S., 1993, “Piston Ring Pack Friction and Lubrication
Analysis of an Automotive Engine Using a Mixed Lubrication Model,” SAE
Paper931937.

Sun, DC, 1990, “A Subroutine Package for Solving Slider Lubrication
Problems,” ASME J. of Tribology, Vol. 112, January, p. p. 84-91.

Sun, DC, 1991, “A Thermal Elastica Theory of Piston-Ring and Cylinder-
Bore Contact,” ASME J. of Applied Mechanics, Paper No. 91-APM-8.

Surface Texture 2000, Handbook of Wagner’s Gage Application Training,
Orion, Michigan.

Suzuki, T., Fujimoto, Y., Ochiai, Y., and Fujimura, l., 1987, “A Numerical
Study on Piston Slap in Diesel Engine,” JSME Transactions, Series B, Vol.
53, No. 492, pp. 2610-2618.

Takiguchi, M., Nakayama, K., Furuhama, S., and Yoshida, H., 1998,
“Variation of Piston Ring Oil Film Thickness in an Internal Combustion
Engine,” SAE Paper 980563.

Takiguchi, M., Sasaki, R., Takahashi, l., lshibashi F., Furuhama, S., Kai, R.,
and Sato, M., 2000, “Oil Film Measurement and Analysis of a Three Ring
Pack in an Operating Diesel Engine,” SAE Paper 2000-01-1787.

Taylor, R.I., Brown, M.A., Thompson, OM. and Bell, J.C., 1994, “The
Inﬂuence of Lubricant Rheology on Friction in the Piston Ring Pack,” SAE
Paper 941981.

Tian, T., Noordzij, L.B., Wong, V.W. and Heywood, J.B., 1996, “Modeling
Piston-Ring Dynamics, Blowby, and Ring-Twist Effects,” ASME 1996 Fall
Technical Conference, ICE-Vol. 27-2, Vol. 2, pp. 67-80.

Tian, T., Wong, V.W. and Heywood, J.B., 1996, “A Piston Ring-Pack Film
Thickness and Friction Model for Multigrade Oils and Rough Surfaces,”
SAE Paper 962032.

196

114.)

115.)

116.)

117.)

118.)

119.)

120.)

121.)

122.)

Tian, T., Rabute, R., Wong, V.W. and Heywood, J.B., 1997, “Effects of
Piston-Ring Dynamics on Ring/Groove Wear and Oil Consumption in a
Diesel Engine,” SAE Paper 970835.

Tian, T. and Wong, V. W., 1998, “Modeling the Lubrication, Dynamics, and
Effects of Piston Dynamic Tilt of Twin-Land Oil Control Rings in Internal
Combustion Engines,” ASME 1998 Fall Technical Conference, Paper No.
98-lCE-130, pp. 9-21.

Tian, T., Wong, V.W. and Heywood, J.B., 1998, “Modeling the Dynamics
and Lubrication of Three Piece Oil Control Rings in Internal Combustion
Engines,” SAE Paper 982657.

Ting, LL. and Mayer, J.E., 1973, “Piston Ring Lubrication and Cylinder
Bore Wear Analysis. Part I - Theory,” ASME J. of Lubrication Technology,
Vol. 96, pp. 305-314, Paper No. 73-Lub-25.

Ting, LL. and Mayer, J.E., 1973, “Piston Ring Lubrication and Cylinder
Bore Wear Analysis. Part II - Theory Veriﬁcation,” ASME J. of Lubrication
Technology, Vol. 96, Paper No. 73-Lub-27.

Ting LL, 1980, “Lubricated Piston Rings and Cylinder Bore Wear,” ASME
Wear Control Handbook, p. p. 609-665, ASME Publication, New York.

Ting, LL, 1980, “Development of a Laser Fluorescence Technique for
Measuring Piston Ring Oil Film Thickness,” J. of Lubrication Technology,
Vol. 102, pp. 165.

Ting, LL, 1985, “A Review of Present lnfon'nation on Piston Ring
Tribology,” SAE Paper 852355.

Tripp, J.H., 1983, “Surface Roughness Effects in Hydrodynamic
Lubrication: The Flow Factor Method,” ASME J. of Lubrication Technology,
Vol. 105, July, pp. 458-465.

197

123.)

124.)

125.)

126.)

127.)

128.)

129.)

130.)

131.)

Treuhaft, M.B., Iddings, F.A., Boyd, GA. and Sprague, SR, 1994, “The
Use of Radioactive Tracer Technology in Studying Lubricant Chemistry to
Enhance Bearing and Ring Wear Control in an Operating Engine,” SAE
Paper941982.

Ungar, E. E. and Ross, D., 1965, “Vibration and Noise Due to Piston Slap
in Reciprocating Machinery,” J. Sound Vibration, Vol. 2, pp. 132-146.

Wahiduzzaman, S., Deribar, R. and Dursunkaya, Z., 1992, “A Model for
Evaporative Consumption of Lubricating Oil in Reciprocating Engines,”
SAE Paper 922202.

Wilhoit, RC, and Zwolinski, B.J., 1971, “Handbook of Vapor Pressures
and Heat of Vaporization of Hydrocarbons and Related Compounds,”
Texas A&M University, College Station, Texas, USA.

Wing, RD, and Saunders, O., 1972, “Oil Film Temperature and Thickness
Measurements on the Piston Rings of 3 Diesel Engine,” Tribology and
Combustion Engines Groups, Proc. Inst. Mech. Engrs., Vol. 186.

Wochni, G, 1979, “Determination of Local Heat Transfer Coefﬁcients at the
Piston of a High Speed Diesel Engine by Evaluation of Measured
Temperature Distribution,” SAE Paper 790834.

Wong, V.W., Tian, T., Lang, H., Ryan, J.P., Sekiya, Y., Kobayashi, Y., and
Aoyama, S., 1994, “A Numerical Model of Piston Secondary Motion and
Piston Slap in Partially Flooded Elastohydrodynamic Skirt Lubrication,”
SAE Paper 940696.

Wu, H. and Chiu, G, 1986, “A Study of Temperature Distribution in a Diesel
Piston - Comparison of Analytical and Experimental Results,” SAE Paper
861278.

Yun, J.E., Chung, Y.C., Chun, SM. and Lee, K.Y., 1995, “An Application of
Simpliﬁed Average Reynolds Equation for Mixed Lubrication Analysis of

Piston Ring Assembly in an Internal Combustion Engine,” May, SAE Paper
952562.

198

132.)

133.)

134.)

Zhang, Y., Chen, G. and Li, B., 1999, “Two Dimensional Numerical Analysis
of Piston Ring Lubrication of an Internal Combustion Engine,” SAE Paper
No. 1999-01-1222.

Zhu, D. Cheng, H.S., Arai, T. and Hamai, K., 1992, “A Numerical Analysis
for Piston Skirts in Mixed Lubrication - Part I: Basic Modeling,” ASME J. of
Tribology, Vol. 114, July, pp. 553-562.

Zhu, D. Hu, Y., Cheng, H.S., Arai, T. and Hamai, K., 1993, “A Numerical
Analysis for Piston Skirts in Mixed Lubrication - Part II: Deformation
Considerations,” ASME J. of Tribology, Vol. 115, January, pp. 125-133.

199

   

11,1,11,11,11,1,1,,