12—5253!
. .3:

5:54. . is : :t...r

 

 

 

.aﬂavﬂiﬁ. Ex!
5.. i .

$4

c a
a;
9:. mm

3......
a. 3:1

6:. a. .

27...»! Lt... $3....
:5: . . .5. 1
s. . A 3.5th

 

 

. u 3.11.,»
4! .ﬁﬁi

:1. . .

1.... .

 

 

 

 

 

 

 

av .l 3.11.
1.5. f v! t
6.133.}!

THESIS

é—v
/4[r~ -.
.cL

This is to certify that the

thesis entitled

DYNAMIC FLOW STUDY IN A FLOW OSCILLATING SYSTEM USING
LASER DOPPLER VELOCIMETRY AND NUMERICAL SIMULATION

presented by
Xin Zhang
has been accepted towards fulﬁllment

of the requirements for

Master of Science degree in Mechanical Engineering

 

 

MM

Major professor

Date August 25,2000

 

0-7639 MS U i: an Afﬁrmative Action/Equal Opportunity Institution

 

 

 

 

 

usmy

Michigan State

University

 

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 GJCIRCIDmDmpes-ms

 

 

DYNAMIC FLOW STUDY IN A FLOW OSCILLATING SYSTEM USING LASER

DOPPLER VELOCIMETRY AND NUMERICAL SIMULATION

By

Xin Zhang

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Mechanical Engineering

2000

ABSTRACT

DYNAMIC FLOW STUDY IN A FLOW OSCILLATING SYSTEM USING LASER
DOPPLER VELOCIMETRY AND NUMERICAL SIMULATION
By

Xin Zhang

In this research, an air ﬂow oscillator was used to simulate the air flow through
an automotive induction system. The purpose of this research was to develop a better
understanding of the flow characteristics of the oscillating ﬂow in the test region by using
both an experimental approach and a numerical simulation.

In the experiment part, Laser Doppler Velocimetry (LDV) was used to measure
the velocity. Piezoresistive pressure sensor and ampliﬁer were used to measure pressure
distribution at different locations under different flow conditions. For the simulation part,
Gridgen was applied to create the grid and STAR-CD was used to predict the flow
characteristics under the same conditions. Both the experimental result and the STAR-CD
modeling result lead to a better understanding of the ﬂow condition in the ﬂow oscillator
system under different operating conditions. Also, comparison of the two different

approaches was discussed and analyzed.

ACKNOWLEDGMENT

There are many people to whom I would like to express my appreciation. Their
efforts and support help me complete this thesis. First, I would like to extend my sincere
gratitude to Dr. Harold Schock, my academic advisor, for giving me opportunities and
challenges of my graduate studies. I would not have been able to complete my studies
without his guidance and ﬁnancial support. Second, I would like to take the opportunity to
thank many people who helped me grow academically and professionally; Dr. Brereton
and Dr. Shih for being my committees and their great support, Dr. Keunchul Lee for his
expertise with the LDV system, Tom Stuecken for his assistance with everything from
repairing my test rig to helping me ﬁnd a tool, Dr. Yuliang Lin for teaching me a large
portion of grid generation, Shi Zheng for helping me use the STAR-CD software, and
Bobbie Slider and Jan Chappell, for assistance on a daily basis. I would also like to thank
my fellows at MSUERL, for their day to day guidance and their wonderful company;
Mark Novak, Edward Timm, Steve George, Stephen Yen, Mahmood Rahi, Kyle Judd,
Anthony Christie, Andrew Fedewa and Boon—Keat Chui. Finally, my brother Hui Zhang,

for his patience and support throughout my master's program.

iii

This thesis is dedicated to my parents
Yuanzong Zhang
&

Xiushu Zhao

For their love and support

iv

TABLE OF CONTENTS

LIST OF FIGURES .................................................... vii
LIST OF TABLES ..................................................... xii
LIST OF SYMBOLS AND ABBREVITIONS .............................. xiii
CHAPTER 1. INTRODUCTION ......................................... 1
l . 1 Motivation ....................................................... l
1.2 Problem Deﬁnition ................................................. 2
CHAPTER 2. EXPERIMENTAL TECHNIQUES AND FACILITIES .......... 5
2.1 Test Rig ......................................................... 5
2.2 Laser Doppler Velocimetry System .................................... 7
2.3 Pressure Measurement System ....................................... 17
CHAPTER 3. EXPERIMENTAL PROCEDURE .......................... 20
3.] Coordinate Systems ............................................... 20
3.2 LDV Measurement ................................................ 21
3.3 Pressure Measurement ............................................. 22
CHAPTER 4. EXPERIMENTAL RESULTS AND DISCUSSION ............. 24
4.1 LDV Velocity Measurements ........................................ 24
4.2 Pressure Measurements ............................................ 35
CHAPTER 5. NUMERICAL SIMULATION SETUP ....................... 39
5.1 Description of Problem ............................................. 39
5.2 Formulation of Problem ............................................ 4O

5.3 Numerical Method of Solution ....................................... 42

CHAPTER 6. NUMERICAL SIMULATION RESULTS AND DISSCUTION . . . 47

6.1 Velocity Results .................................................. 47
6.2 Pressure Results .................................................. 57
CHAPTER 7. COMPARISON .......................................... 65
7.1 Compare the CFD Velocity Results with LDV Results. ................... 65
7.2 Compare the CF D Pressure Results with Experimental Results ............. 75
CHAPTER 8. SUMMARY AND CONCLUSIONS .......................... 85
LIST OF REFERENCES ................................................ 87

vi

LIST OF FIGURES

Figure 1: Schematic of the Flow Oscillator System ........................................................... 3
Figure 2: Flow Oscillator Test Rig ..................................................................................... 5
Figure 3: LDV System Setup ............................................................................................ 13
Figure 4: Six-Jet Atomizer ................................................................................................ 15
Figure 5: FINDTM for Windows Version 1.4 Graphical User Interface ......................... 16
Figure 6: Pressure Measurement System Setup ................................................................ 19
Figure 7: The Local Coordinate System and the Three Measured Locations ................... 21
Figure 8: Velocity Proﬁle at Location 1 (2000rpm, Flow Rate 2) ................................... 24
Figure 9: Velocity Proﬁle at Location 2 (2000rpm, Flow Rate 1) ................................... 25
Figure 10: Velocity Proﬁle at Location 2 (2000rpm, Flow Rate 2) ................................. 26
Figure 11: Velocity Proﬁle at Location 2 (2000rpm, Flow Rate 3) ................................. 26
Figure 12: Velocity Proﬁle at Location 3 (2000rpm, Flow Rate 1) ................................. 27
Figure 13: Velocity Proﬁle at Location 3 (2000rpm, Flow Rate 2) ................................. 28
Figure 14: Velocity Proﬁle at Location 3 (2000rpm, Flow Rate 3) ................................. 28
Figure 15: Velocity Proﬁle at Location 2 (4000rpm, Flow Rate 1) ................................. 29
Figure 16: Velocity Proﬁle at Location 2 (4000rpm, Flow Rate 2) ................................. 30
Figure 17: Velocity Proﬁle at Location 2 (4000rpm, Flow Rate 3) ................................. 30
Figure 18: Velocity Proﬁle at Location 3 (4000rpm, Flow Rate 1) ................................. 31
Figure 19: Velocity Proﬁle at Location 3 (4000rpm, Flow Rate 2) ................................. 32
Figure 20: Velocity Proﬁle at Location 3 (4000rpm, Flow Rate 3) ................................. 32

Vii

Figure 21:
Figure 22:
Figure 23:
Figure 24:
Figure 25:
Figure 26:
Figure 27:
Figure 28:
Figure 29:
Figure 30:
Figure 31:
Figure 32:
Figure 33:
Figure 34:
Figure 35:
Figure 36:
Figure 37:
Figure 38:
Figure 39:
Figure 40:
Figure 41:
Figure 42:

Figure 43:

Velocity Proﬁle at Three Locations (2000 rpm, Flow Rate 1) ........................ 33

Velocity Proﬁle at Three Locations (2000 rpm, Flow Rate 2) ........................ 34
Velocity Proﬁle at Three Locations (2000 rpm, Flow Rate 3) ........................ 34
Pressure Drop at Three Different Locations (2000rpm, Flow Rate 1) ............ 35
Pressure Drop at Three Different Locations (2000rpm, Flow Rate 2) ............ 36
Pressure Drop at Three Different Locations (2000rpm, Flow Rate 3) ............ 36
Pressure Drop at Three Different Locations (4000rpm, Flow Rate 1) ............ 37
Pressure Drop at Three Different Locations (4000rpm, Flow Rate 2) ............ 38
Pressure Drop at Three Different Locations (4000rpm, Flow Rate 3) ............ 38
Schematic of Problem Studied ........................................................................ 39
Grid System Used ............................................................................................ 45
Velocity Proﬁle at Location 1 (2000rpm, Flow Rate 1) ................................. 47
Velocity Proﬁle at Location 2 (2000rpm, Flow Rate 2) ................................. 48
Velocity Proﬁle at Location 3 (2000rpm, Flow Rate 3) ................................. 48
Velocity Proﬁle at Location 1 (4000rpm, Flow Rate 1) ................................. 49
Velocity Proﬁle at Location 2 (4000rpm, Flow Rate 2) ................................. 49
Velocity Proﬁle at Location 3 (4000rpm, Flow Rate 3) ................................. 50
Velocity Proﬁle in the Region of Oriﬁce at 2000 rpm (CA=60) .................... 51
Velocity Proﬁle in the Region of Oriﬁce at 2000 rpm (CA=120) .................. 5]
Velocity Proﬁle in the Region of Oriﬁce at 2000 rpm (CA=180) .................. 52
Velocity Proﬁle in the Region of Oriﬁce at 2000 rpm (CA=240) .................. 52
Velocity Proﬁle in the Region of Oriﬁce at 2000 rpm (CA=300) .................. 53
Velocity Proﬁle in the Region of Oriﬁce at 2000 rpm (CA=360) .................. 53

viii

Figure 44:
Figure 45:
Figure 46:
Figure 47:
Figure 48:
Figure 49:
Figure 50:

Figure 51:

Figure 52:

Figure 53:

Figure 54:
Figure 55:

Figure 56:

Figure 57:

Figure 58:

Figure 59:

Figure 60:

Figure 61:

Figure 62:
Figure 63:
Figure 64:

Figure 65:

Figure 66:

Velocity Proﬁle in the Region of Oriﬁce at 4000 rpm (CA=60) .................... 54
Velocity Proﬁle in the Region of Oriﬁce at 4000 rpm (CA=120) .................. S4
Velocity Proﬁle in the Region of Oriﬁce at 4000 rpm (CA=180) .................. 55
Velocity Proﬁle in the Region of Oriﬁce at 4000 rpm (CA=240) .................. 55
Velocity Proﬁle in the Region of Oriﬁce at 4000 rpm (CA=300) .................. 56
Velocity Proﬁle in the Region of Oriﬁce at 4000 rpm (CA=360) .................. 56
Pressure Results for Crank Angle 10 - 60 Degree (2000 rpm) ....................... 57
Pressure Results for Crank Angle 70 — 120 Degree (2000 rpm) ..................... 58
Pressure Results for Crank Angle 130 ~ 180 Degree (2000 rpm) ................... 58
Pressure Results for Crank Angle 190 - 240 Degree (2000 rpm) ................... 59
Pressure Results for Crank Angle 250 - 300 Degree (2000 rpm) ................... 59
Pressure Results for Crank Angle 310 - 360 Degree (2000 rpm) ................... 60
Pressure Results for Crank Angle 10 - 60 Degree (4000 rpm) ....................... 61
Pressure Results for Crank Angle 70 - 120 Degree (4000 rpm) ..................... 61
Pressure Results for Crank Angle 130 - 180 Degree (4000 rpm) ................... 62
Pressure Results for Crank Angle 190 - 240 Degree (4000 rpm) ................... 62
Pressure Results for Crank Angle 250 - 300 Degree (4000 rpm) ................... 63
Pressure Results for Crank Angle 310 - 360 Degree (4000 rpm) ................... 63
Velocity Proﬁle Comparison at 2000 rpm (Location 1, Flow Rate 1) ............ 66
Velocity Proﬁle Comparison at 2000 rpm (Location2, Flow Rate 1) ............. 66
Velocity Proﬁle Comparison at 2000 rpm (Location 3, Flow Rate 1) ............ 67
Velocity Proﬁle Comparison at 2000 rpm (Location 1, Flow Rate 2) ............ 68
Velocity Proﬁle Comparison at 2000 rpm (Location 2, Flow Rate 2) ............ 68

ix

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:
Figure 85:
Figure 86:
Figure 87:
Figure 88:

Figure 89:

Velocity Proﬁle Comparison at 2000 rpm (Location 3, Flow Rate 2) ............ 69

Velocity Proﬁle Comparison at 2000 rpm (Location 1, Flow Rate 3) ............ 70
Velocity Proﬁle Comparison at 2000 rpm (Location 2, Flow Rate 3) ............ 70
Velocity Proﬁle Comparison at 2000 rpm (Location 3, Flow Rate 3) ............ 71
Velocity Proﬁle Comparison at 4000 rpm (Location 2, Flow Rate 1) ............ 72
Velocity Proﬁle Comparison at 4000 rpm (Location 3, Flow Rate 1) ............ 72
Velocity Proﬁle Comparison at 4000 rpm (Location 2, Flow Rate 2) ............ 73
Velocity Proﬁle Comparison at 4000 rpm (Location 3, Flow Rate 2) ............ 73
Velocity Proﬁle Comparison at 4000 rpm (Location 2, Flow Rate 3) ............ 74
Velocity Proﬁle Comparison at 4000 rpm (Location 3, Flow Rate 3) ............ 74
Pressure Data Comparison at 2000 rpm (Location 1, Flow Rate 1) ............... 75
Pressure Data Comparison at 2000 rpm (Location 2, Flow Rate 1) ............... 76
Pressure Data Comparison at 2000 rpm (Location 3, Flow Rate 1) ............... 76
Pressure Data Comparison at 2000 rpm (Location 1, Flow Rate 2) ............... 77
Pressure Data Comparison at 2000 rpm (Location 2, Flow Rate 2) ............... 77
Pressure Data Comparison at 2000 rpm (Location 3, Flow Rate 2) ............... 78
Pressure Data Comparison at 2000 rpm (Location 1, Flow Rate 3) ............... 78
Pressure Data Comparison at 2000 rpm (Location 2, Flow Rate 3) ............... 79
Pressure Data Comparison at 2000 rpm (Location 3, Flow Rate 3) ............... 79
Pressure Data Comparison at 4000 rpm (Location 1, Flow Rate 1) ............... 80
Pressure Data Comparison at 4000 rpm (Location 2, Flow Rate 1) ............... 81
Pressure Data Comparison at 4000 rpm (Location 3, Flow Rate 1) ............... 81
Pressure Data Comparison at 4000 rpm (Location 1, Flow Rate 2) ............... 82

Figure 90: Pressure Data Comparison at 4000 rpm (Location 2, Flow Rate 2) ............... 82

Figure 91: Pressure Data Comparison at 4000 rpm (Location 3, Flow Rate 2) ............... 83
Figure 92: Pressure Data Comparison at 4000 rpm (Location 1, Flow Rate 3) ............... 83
Figure 93: Pressure Data Comparison at 4000 rpm (Location 2, Flow Rate 3) ............... 84
Figure 94: Pressure Data Comparison at 4000 rpm (Location 3, Flow Rate 3) ............... 84

xi

LIST OF TABLES

Table 1: Laser Optical Parameters (For green laser beams) .............................................. 14
Table 2: Pressure Measurement System ............................................................................ 17
Table 3: Three Different Flow Rates .................................................................................. 22
Table 4: Standard Control Parameter Settings for Transient PISO Calculations ............... 44

xii

LIST OF SYMBOLS AND ABBREVITION S

The following is a list of deﬁnitions of the main symbols used in this thesis:

BDC

CA

Nc

qi _

Re
TDC
TI

U(t)
(Tr)
UEA(9

u'..<0>

Area

Bottom Dead Center

Crank angle

Fringe spacing

Mechanical plus thermal energy per unit volume
Cycle number

Turbulent kinetic energy

Mixing length

Number of cycle

Pressure

Effective conduction in the ith direction
Reynolds number

Top Dead Center

Turbulent Intensity

Instantaneous velocity

Mean velocity

Ensemble averaged mean velocity

Ensemble average ﬂuctuation

xiii

Crank angle

Half angle between beams
Wavelength of the laser light
Air density

Effective shear stress

xiv

CHAPTER 1. INTRODUCTION

1.1 Motivation

Competition in the automotive industry has increased the need to develop
technology for superior engines. The automobile internal combustion engine has seen
much advancement over the past twenty years. Much of this advancement has been to
make engines more fuel efﬁcient and cleaner burning to preserve the environment and
bring rapid development to engine control and fuel injection systems. In order to have
precise computer controlled fuel injection, there must be a means to measure the mass of
air flowing into a vehicle’s engine, such as the modern Mass Air Flow Sensor (MAFS).
The Michigan State University Engine Research Laboratory has been previously involved
in characterization the ﬂows in a MAFS. A dynamic oscillating device was used to
evaluate the performance of the MAFS under pulsating air ﬂow conditions. The purpose
of this study was to get a better understanding of the flow characteristics of the oscillating
ﬂow in the test region using both experiment measurement and numerical Simulation.

MAFS's are now a standard part on nearly every production vehicle. They are
located downstream of an automobile’s air ﬁlter, and just upstream of the intake manifold
and fuel injection system, and their job is to measure the airflow into the engine to control
fuel delivery. There is some controversy, however, over how well certain MAFS's actually
measure air ﬂow into an engine. To fully understand the controversy, one must have some
understanding of the typical air ﬂow through the engines intake system, in particular how

the ﬂow passes the MAFS. First, nearly every engine is designed with a certain amount of

valve overlap, meaning that for a small crank angle duration, both the intake and exhaust
valves are held open. While the reasoning for this will not be addressed, it is important to
recognize that in many circumstances this will cause a small amount of air to ﬂow
backwards through the intake manifold before the fresh charge is drawn into the cylinder.
For many engine speeds this same phenomenon (although to a much smaller extent) will
be exhibited once the piston reaches bottom dead center (BDC), because the intake valve
is still open slightly past the piston’s BDC position. The ﬁnal result is flow oscillation
through the intake system, which may cause a discrepancy between the measured and

actual mass of air ﬂowing past the MAFS.

1.2 Problem Deﬁnition

To develop a better understanding the flow phenomena within the ﬂow
oscillator system, Laser Doppler Velocimetry (LDV) was conducted to measure the
velocity. A piezoresistive pressure sensor and ampliﬁer were also used to measure the
pressure distribution at different locations for different flow conditions.

Computational Fluid Dynamics (CFD) approach, STAR-CD as the solver was
used to predict the flow characteristics at the same conditions using Gridgen as the grid
generation tool. The adaptation of standard CFD modeling codes enables predicting the
major ﬂows patterns. Very small velocity gradients can be visualized by adjusting the grid
spacing to the geometrical location of interest.

The availability of a comparable amount of measured velocity and pressure
data has always been a problem in the veriﬁcation process of any modeling results. The
possibility to validate and improve the basic assumptions and boundary conditions of the

actual model is interdependent with the accuracy and availability of velocity and pressure

results measured in the same conditions. Since in the test region, only one component
velocity is dominant, the comparison was made only for the single component modeled

and measured results.

 

D = 2.446"
L = 20"
l = 4n

Piston Parameters:
L Bore = 2.85"

Te“ Region Stroke = 2.346"

\

Outlet to the Vacuum

 

 

 

Air Flow Oscillator

Adjustable Annular Throat

 

 

 

Figure 1: Schematic of the Flow Oscillator System

Figure 1 shows the schematic of the ﬂow oscillator system in this study. The
test region is consist of two 20 inches long clear acrylic round tubes with an optically clear
quartz round tube in the middle of them. A Single-cylinder engine was used to provide
oscillating airﬂow. All the measurements and simulations were conducted at two engine
speeds, 2000 rpm and 4000 rpm. The head assembly on the engine was removed and was
replaced with an adjustable ﬂow chamber. The ﬂow chamber was conﬁgured with four
outlets, which were connected to the vacuum accumulator tank by hoses. The ﬂow rate

can be increased or decreased by adjusting the ﬂow chamber.

CHAPTER 2. EXPERIMENTAL TECHNIQUES AND FACILITIES

2.1 Test Rig
The measurements were conducted on a test bench that consisted of a test
stand, single-cylinder flow oscillator, lO-hp DC electric motor, adjustable ﬂow chamber

and accumulator vacuum tank. The system is shown in Figure 2.

DC Motor

 

Vacuum Tank

 

Figure 2: Flow Oscillator Test Rig
The single-cylinder ﬂow oscillator was designed and built here at the MSU
Engine Research Laboratory (ERL). As stated earlier, an internal combustion engine (IC)

exhibits slight amounts of ﬂow reversal in the intake manifold, and the ﬂow reversal

effects are increased as the number of ﬁring cylinders decreases. To simulate this
phenomenon in a laboratory setting, a special test stand was built for laser diagnostics. The
ﬂow oscillator was designed to create small velocity oscillations in a flow with a given net
velocity. To achieve this, a setup was built consisting of a large vacuum pump, and a small,
single cylinder Suzuki engine (no head or valvetrain) and a DC motor. The vacuum pump
creates a net velocity past the ﬂow chamber (ﬂow velocity can be controlled with the
adjustable ﬂow modulator above the 4-stroke engine to simulate engines of different size,
number of cylinders, and operating speeds), with the reciprocating piston imbedding a
sinusoidal oscillation in the mean stream velocity. This oscillating air ﬂow is drawn
through a clear acrylic tube in which a quartz section. Located between the ﬂow oscillator
and the acrylic tube is a small section of aluminum honeycomb which serves two
functions: to straighten ﬂow and to prevent debris from falling onto the piston face.

The engine was driven by a 10—hp DC electric motor. Flows were examined
based on the crank angle, which is detected by an angle encoder installer directly on the
crankshaft. The encoder generates two series of 1024 square pulses per revolution. These
two series of pulses are offset a half pulse length from each other. The data acquisition
system can count 4 times the number of pulses from these two series of pulses. For this
study, 4096 pulses per revolution (360 crank angle degrees) were used, which represents
0.088 degrees of angular resolution. The encoder was set to trigger when the piston in the

engine was at Bottom Dead Center (BDC).

2.2 Laser Doppler Velocimetry System
2.2.1 Laser Doppler Velocimetry Technique Introduction

LDV has been used for non-intrusive ﬂow studies for many years and is one of
a very small group of technologies that can measure ﬂows within closed systems. It
requires no calibration under changing conditions and measures ﬂows, without affecting
the ﬂuid ﬂow. It uses only light and non-homogeneous scattering particle to obtain
velocity measurements.

Laser Doppler theory states that the velocity of a particle can be measured by
detecting the Doppler shift frequency of laser light scattered by the particle. The capability
of the seed particle to follow ﬂow characteristics was determined by determining the
sedimentation velocity calculating the distance the particle has dropped due to
gravitational forces in a ﬂow of a known velocity over a given distance. If this vertical
distance is less than 20% of the ﬁxed distance through the given distance, the particle
shows acceptable characteristics for following that ﬂow velocity.

LDV technique relies on the light scattered by scattering centers in the ﬂuid to
measure ﬂow velocity. In the following, the scattering centers will be preferred to as
particles, with the understanding that bubbles or anything else that has a relative refractive
index (compared to the ﬂuid) different from unity could be the source of scattered light.
The particles, whose velocities are measured, must be small enough (generally in the
micron range) to follow the ﬂow and large enough to provide signal strength adequate for
the signal processor to give velocity measurements. It should be noted that the signal
exists only when a detectable particle is in the measuring volume and, hence, is

discontinuous.

The dual-beam approach is the most common optical arrangement used for
LDV system for ﬂow measurement applications. The scattered light is collected by a
photomultiplier. The function of the photomultiplier is to convert bursts of light scattered
in the fringe pattern into an analog signal for interpretation by the data acquisition
equipment (IPA-750). The data acquisition equipment then combines the frequency of the
analog signals with a set of encoder values relating to the position of the impeller and
turbine. This related data set is stored in a PC connected to the data acquisition hardware.
The system which we were using is a back-scattering system, where the burst of light
scattered from the measurement volume are collected at a physical location in front of the
focal point. In back-scatter mode, the intensity of scattered light is weaker than scattered
light intensities using the forward-scattering method. Back-scatter data collection was
used in this experiment due to optical access constraints.

The point of measurement is a volume determined by the half angle of the

beams (K) intersecting at the focal point and the wavelength of light used (A). This

volume is in the shape of an ellipsoid with the length governed by Equation 1. The height

of the ellipsoid is determined by Equation 2.

1 = —__4ﬂ‘ (l)

m rtDe-2 sin K

_ i (2)

m _ ItDe-2cosx

The measurement volume is thus deﬁned by Equation 3.

 

 

3
V = [(411) , 1 ] (3)

0:21:2 6((cosrc)2 - sin K)

Where f is the focal length, A represents the wavelength of the light used and

De'2 represents the diameter of the unfocused Gaussian laser beam measured at 1/e’2 of

the center line intensity. The frequency of the signal converted by the photodetector

increases in proportion to the half angle value. Therefore, the half angle can limit the range

of measurable velocities. Velocity measurements have been made from lttm /sec to 1000
m/s. This wide range of measurements is made possible, in part, by changing the half
angle to allow for measurement of the velocity range. The half angle generates a fringe
pattern when two beams of polarized laser light interfere with each other at the focal point.

As the beams of the unshifted and that of shifted beam cross, a fringe pattern
moving in a know direction, is formed. Shifting the frequency of a laser beam refers to
changing the frequency of the laser light by a known quantity either up or down.
Frequency shifting is commonly used in LDV measurements. The purpose for this is to
measure ﬂow reversal, to measure small component velocities in the presence of large
component velocities and to increase the range of velocities that can be measured by the
signal processor. The frequency of the signal collected by the signal processor is ﬁxed at
40MHz by a Bragg cell. This frequency is then down mixed to a value greater than twice
the Doppler frequency for the corresponding negative velocity. The direction of the fringe
motion is normally oriented in a direction opposing the mean positive ﬂow. Thus any
velocity measured that has a frequency less than the shift value is negative and those with
greater frequencies are positive. The formula use to resolve positive and negative velocity

measurements is represented by Equation 4.The velocity value used in Equation 4 is

determined by the result of Equation 5. The Doppler frequency is simply the frequency at
which the signal processor receives light burst signals from the incidence of seed particles

passing through the focal point.

 

. IVelocityI
= +
Frequency Shift Frequecy _ Fringe Spacing (4)
Velocity = Doppler Frequency - Fringe Space (5)

The data acquisition equipment has been pre-programed with the
characteristics of the laser beam and its shift parameters. The data acquisition equipment
calculates the fringe spacing using the wavelength of the laser light and the half angle

between the beams as seen Equation 6.

A

df = 23inK (6)

 

The velocity of the particle passing through the fringe pattern is then calculated
by multiplying the fringe spacing by the frequency of the Doppler shifted light scattered
by the particle. The frequency of Doppler shifted light is equal to the number of fringe
patterns scattered in a ﬁxed amount of time over a known distance and is measured in
Hertz (Hz). This information can then be used to infer direction and velocity directly by
noting the orientation of the shifted beam to the unshifted beam and their orientation in
Cartesian space. These calculations are carried out in the IFA 750 signal processor which
collects data from all the sensing equipment used. The information regarding the adjusted

frequency, representing the velocity of the particles through the measurement volume, is
then stored in a PC running TSI FINDTM data acquisition software. This software is used

to control and program the data acquisition equipment.

10

The laser used was a multi-frequency Argon-ion laser capable of producing
three wave lengths of light. The wave lengths of light are green at 514.5 nm, blue at 488
nm and violet at 476.5 nm. This experiment used the green wavelength. These
wavelengths are separated by a device called the Colorburst, manufactured by TSI. It is
composed of a set of optics used as beam splitters for certain wavelength of light. Once
separated, each distinct wavelength of laser light can be routed via ﬁber optics to the LDV
probe face where their orientation, on the face of the probe, determine which velocity
components they will measure.

The proper selection of particles can have a strong inﬂuence on the quality of
the data in the LDV system. For most experimentalists, the ﬁrst reaction is to rely on
particles naturally present in the ﬂow. This may be adequate in some of the applications,
e.g., forward scatter in water. In almost all other cases, it is generally desirable to seed the
ﬂow suitably to get adequate data rates. Several techniques give data in forward scatter,
Fluorescent particle work in backscatter. There is no universal solution but acceptable

particles and technique normally are available for a given application.

2.2.2 Ensemble Averaged Mean Velocity

An important characteristic of periodic ﬂow is ensemble-averaged mean
velocity. The statistical description of periodic ﬂow is conventionally used for engine
studies. Although the ﬂow visualization technique can determine bulk features, the
differences between features are often subtle and can not be described in a quantitative
way. Therefore quantitative analysis is important to understand the ﬂow characteristics in
a highly ﬂuctuating turbulent ﬂow like the one in the oscillating ﬂow system. One of the

important characteristics of turbulent ﬂow is that of its irregularity or randomness.

ll

Statistical methods are used to deﬁne such ﬂow ﬁeld. The quantities that characterize
turbulence are usually the mean velocity and the ﬂuctuation velocity about the mean, and

length and time scale. The local instaneous velocity can be deﬁned by Equation 7.

U(t) = 6+ u'(t) (7)

Here the mean velocity is the time average of U (I) . It is shown in Equation 8.
U = lim ii, U(t) (8)

The turbulent portion of the velocity is u'(t). For a steady ﬂow, the mean stands for the
simple time averaged value. In this steady, the mean value can vary signiﬁcantly from one

engine cycle to the next. Instantaneous velocity can be written as a function of cam angle

9 and cycle number instead of a function of time, which is Shown in Equation 9.

U(9, i) = (7(9, 1') + u'(0, i) (9)
The ensemble averaged velocity U 6 “(0) over No cycles can be deﬁned as:

Nc
aea(0) = NLCZUW, i) (10)
1

Where the NC is the number of cycles for which data is available. The ensemble average
velocity is only a function of cam angle since the cyclic variation has been averaged out.
The difference between the mean velocity in a particular cycle and the ensemble averaged

mean velocity over many cycles is deﬁned as the cycle by cycle variation in mean velocity:

i1(0, i) = U(0,i)—(-/ea(0,i) (11)

Thus the instaneous velocity can be split into three components:

12

me, i) = Emu), i) + {1(6), i) +u'(e, i) (12)
This is the most desirable equation to be used for the measuring velocity in the study. The
ensemble averaged ﬂuctuation u'm(0) is given by:

Nr‘

Vale) = $621419, 1') (13)

I
It includes all ﬂuctuations about the ensemble averaged mean velocity. Based upon
Equation (12) and (13), ensemble averaged mean velocity and mean ﬂuctuation were

calculated.

2.2.3 Laser Doppler Velocimetry System Overview

. .-_ l
’ 1 r i
. 1 :

FrcILency Shifter

.1 I -'_:i Ina; Photo— \lllillplltl
I‘ 1‘

Li

 

 

Figure 3: LDV System Setup

A single-component Laser Doppler Velocimetry (LDV) system was utilized in
determining the velocity proﬁles in this experiment. Elements making up the LDV include
an Argon-ion laser, beam collimator, Colorburst Multicolor Beam Separator (TSI), LV
Frequency Shifter, 4-beam ﬁber optic probe with transmitting and receiving optics,
photomultiplier, digital burst correlator (TSI, IFA 750), rotating machine resolver (RMR),
traverse mechanism (for moving and focusing the laser beams), and computer for data
collection and storage are also required for this experiment. The system is shown in Figure
3

The single-component LDV system is using the green beams (shifted beam and

unshifted beam). The laser optical parameter is shown is Table 1.

Table 1: Laser Optical Parameters (For green laser beams)

 

 

 

 

 

 

 

Laser Optical Parameters Green Laser Beams
Laser Wave Length (nm) 514.5
Beam Spacing (mm) 50
Focal Length (mm) 261
Half Angle K (degree) 5.52
Fringe Spacing (um) 2.674

 

 

 

 

A 20% mixture of propylene glycol and water mixture was atomized by a Six-

Jet Atomizer and used as laser light scattering particles. The diameters of the particles

generated by the atomizer are in the range of 0.6 to 1.5 um . These particles were supplied

to the measurement region without disturbing the airﬂow generated by the engine. The

Six-Jet Atomizer and the hose for inducting the seeding ﬂow were shown in Figure 4.

l4

 

Figure 4: Six—Jet Atomizer

2.2.4 Data Acquisition And Data Analysis System

FINDTM was used as the signal processing software for the LDV system.
FINDTM stands for Flow Information Display. It is a Windows-based data acquisition and
analysis software that is part of TSI'S integrated Laser Doppler Velocimetry (LDV) and
Adaptive Phase/Doppler Velocimetry (APV) systems. Comparing the old DOS version
system, it has a lot of advantages. FINDTM is easier to set up the experiment parameters
and the true multiple Real-time View Display gives more control to the experiment. The
post-processing work is made much easier by the Statistical Analysis tool. Statistical

results can viewed as tabular data, spreadsheet row column, or graphically using the built-

in graphics display. FINDTM for Windows Version 1.4 Graphical User Interface is shown

in Figure 5 below.

I’IUJ lnr WIHIIIIW'. mum” A. lwr I an.)
M W m

 
 

.IuII-umi III-\l I not” a..w
~ ' : ' »

   

 

36 1884

  

 

 

 

Figure 5: FINDTM for Windows Version 1.4 Graphical User Interface

FINDTM performs all hardware interfacing functions and runs the software
diagnostics to make sure the hardware is operation correctly. Since it operates in real time,
you can get instantaneous feedback on various parameter settings and make adjustments
before station the actual experiment.

FINDTM provides complete ﬂow statistics for every measurement point. It
calculates and displays all basic ﬂow statistics, including mean velocity, standard
deviation, turbulence intensity, shear stress, and correlation coefﬁcients for up to three

velocity components. Since raw data is accessible to other statistical programs, the user

can use it for any other type of calculations. Data can also be collected as a function of
time to provide time series information.

FINDTM has a built-in graphics display capability that can display many
different types of graphs. FINDTM uses an internal spreadsheet as a data viewer. Any data
in this spreadsheet can displayed in the graphics display program simply by selection (by
named column that is: Xposition, U-mean, etc.) what you want to display. Information
such as ﬂow ﬁeld proﬁles, Power Spectrum, and particle-size distribution are just a few of

the graphics lots that can be displayed.

2.3 Pressure Measurement System
2.3.1 Pressure Sensor and Ampliﬁer

For the pressure measurements, a Kistler piezoresistive pressure transducer
(Kistler Instrument Cooperation, Model 4053A2) and Kistler dual mode ampliﬁer (Kistler
Instrument Cooperation, Model 4618A0) were used. Details of the system used for the

pressure measurements are shown in Table 2.

Table 2: Pressure Measurement System

 

Description

 

Component/Speciﬁcation
81 unit English unit

 

 

 

Kistler Instrument Cooperation,

Pressure Transducer Model 405 3 A2

 

Kistler Instrument Cooperation,

 

 

 

 

 

 

 

Ampliﬁer Model 4618A0
Measurement Range 0 - 200 kparel 0 ' 2 bar rel
- Sensitivity 1.25 mV/kpa 250 mV/bar
Calibration Current 2 - 5 mA

 

17

Table 2: Pressure Measurement System

 

 

 

 

 

 

Description
Component/Speciﬁcation
SI unit English unit
Compensated Temperature _20 _ 70 C
Range
Ampliﬁer Supply +18 - 30V DC

 

 

 

 

2.3.2 Data Acquisition System

The Real Time Combustion Analysis Module (RTCAM) system collects all the
pressure vs. crank angle data from the engine. Using the piezoresistive pressure
transducer, along with the BEI optical shaft encoder (OSE), the system can display
pressure as a function of crank angle in real time, allowing effects of injection and ignition
timing changes to be easily observed. The pressure vs. crank angle data (for a speciﬁed
number of engine cycles) can also be permanently recorded in ASCII ﬁle format.

The Real Time Combustion Analysis Module acquires cylinder pressure and
other analog and digital data from single or multiple cylinder engines. Calculating the
parameters from the cylinder pressure data on-line while the engine is running is a key
future of real time analysis. This will reduce the volume of data required to be stored by
extracting information from the data and my storing the results. For this measurements,
twenty cycles of pressure data were collected when a trigger event occurred. The trigger
event was conﬁgured by the optical shaft encoder signal. Then the averaged value for each
crank angle was calculated by the system. The schematic of the pressure measurement

system can be seen in Figure 6.

18

 

 

i - I . Test Point

{ﬂ}

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Optical Shaft
DC Power Pressure
Supply Transducer Encoder
Ampliﬁer > RTCAM
System

 

 

 

 

 

 

 

Figure 6: Pressure Measurement System Setup

19

 

CHAPTER 3. EXPERIMENTAL PROCEDURE

3.1 Coordinate Systems

The LDV traverse table system is designed for translation in three dimensions.
It is equipped with stepper motors for automated translation with a maximum resolution of
1 pm. The local coordinate system is set by the LDV traverse table and requires a
reference point for all of its translational movement to be relative to. Setting the local
coordinate system is done by positioning the LDV laser focal point on the test stand or
converter itself and referencing the local coordinate system from that location. Both the
LDV measurement and the pressure measurement are only along the Z-direction of the
local coordinate system. It is also the main air ﬂow direction in the measurements.

The measurements were conducted on three different locations in the test
region. They were measured from the piston top in the ﬂow oscillator system when the
piston was at bottom dead center (BDC). The distances were 12 inches, 25 inches and 43
inches respectively. The origin was located in the center of the piston top surface. Since
the direction of most of the velocity measurements was toward the piston top because of
the offset ﬂow created by the vacuum pump, it is convenient to deﬁne the downward
direction as the positive direction of the velocity measurements.

The local coordinate system and the three locations can be seen in Figure 7.

20

 

 

LocatIoII 3

 

Positive Velocity

 

 

 

 

 

 

 

 

 

 

 

 

Direction
location 2
A V
43 incii Z
location I
25 mm
X Y
I J
12inc
/ Piston Top at BDC
V Y

 

 

 

 

Figure 7: The Local Coordinate System and the Three Measured Locations

3.2 LDV Measurement
3.2.1 Experiment Sequence

Prior to collection the LDV data, the system was turned on by following a
speciﬁed sequence of steps. First, the vacuum used to provide the offset airﬂow was turned
on and the atomized seeding particles were introduced into the airﬂow. The DC motor was
then activated and the engine was brought up to two different speeds, 2000 rpm and 4000
rpm respectively. The LDV system was then utilized to measure the ensemble averaged
velocity at three different locations. Once one ﬂow rate was ﬁnished, the air ﬂow was

adjusted to another ﬂow rate and the same procedures were repeated.

21

3.2.2 LDV Data Collection

Three different ﬂow rates and three different locations were measured under
two engine speeds, 2000 rpm and 4000 rpm respectively. The three locations are shown in
Figure 7. For location 1, only one ﬂow rate (Flow Rate 2) was measured because it was
very near the oscillator and the seeding ﬂow was really difﬁcult to get in there. The signal
and data rate was very bad down there. For location 2 and 3, data were collected at ﬁve
points across the test region (x=-20mm, - 10mm, 0mm, 10mm, 20mm). The three different
ﬂow rates are shown in Table 3. All data were collected in 360 crank angle degrees. The

encoder was set to trigger when the piston in the engine was at Bottom Dead Center

 

 

 

 

 

(BDC).
Table 3: Three Different Flow Rates
Velocity(m/s) Area(m2) Flow Rate(m3/s) Flow Rate(g/S)
Flow Rate #1 5.3536 0.00219 0.0117 14.1
Flow Rate #2 10.1544 0.00219 0.0223 26.7
Flow Rate #3 31.6382 0.00219 0.0694 83.2

 

 

 

 

 

 

 

3.3 Pressure Measurement

The pressure measurement was conducting under the same three ﬂow rates and
two engine speeds as the LDV measurement.

The test system was initiated by a speciﬁc sequence of steps. First, the vacuum
pump was turned on to create the offset ﬂow. Then the DC motor was activated and the
engine brought up to two different speeds, 2000 rpm and 4000 rpm respectively. The Real

Time Combustion Analysis Module (RTCAM) system was used as the data acquisition

22

system and collected the pressure data as the function of the crank angle. The shaft
encoder was set to trigger when the piston in the oscillator was at Bottom Dead Center
(BDC). Once one ﬂow rate was ﬁnished, the air ﬂow was adjusted to another ﬂow rate and

the same procedures were repeated.

23

CHAPTER 4. EXPERIMENTAL RESULTS AND DISCUSSION

4.1 LDV Velocity Measurements
4.1.1 Velocity Measurements at 2000 rpm

Figure 8 shows ﬂow velocities vs. crank angle at 2000 rpm for location 1 and
ﬂow rate 2. Because location 1 is the nearest measured point to the piston top in the test
region, the seeding ﬂow for the LDV measurement was difﬁcult to be provided to that
location. For higher engine speed, it was even more difficult. The results of these
measurements suggested that the velocity proﬁles were almost similar across the test

region. Image in this chapter of thesis are presented in color.

 

F |owRato#2-Location#1 -2000rpm

 

20

 

 

16 i _,

 

 

 

 

VeIOCIty (m/s)
m K?

_ ./

 

 

 

 

 

 

 

 

 

0 . .
0 60 120 180 240 300 360
Crank Angle

 

 

II; Center _- 10mm Right 20mm Right 10mm Left -— 20mm Left ]

 

 

 

Figure 8: Velocity Proﬁle at Location 1 (2000rpm, Flow Rate 2)

24

Figure 9, 10, 11 shows ﬂow velocities vs. crank angle at 2000 rpm for location
2 and ﬂow rate 1, 2 and 3, respectively. The velocity proﬁles are similar for three different
ﬂow rates. They all start at low velocity from crank angle at 0 degree, which piston moves
from the BDC, and the highest velocities are appearing at about 270 degree. And all
velocity curves are almost constants between 60 and 150 degrees. They are supposed to be
the bottom of the sine wave. The reason is the offset ﬂow created by the vacuum. When
the piston moves up from the BDC, the ﬂow is dragged back by the offset ﬂow. As we can
observe from the three plots, the velocity magnitudes increase as we increase the offset

ﬂow rate.

 

FiowRatu1-Locationn- 2000mm

20

 

 

 

16

121

 

\

\

 

 

Velocity (m/s)
oo

 

 

 

 

 

 

 

 

 

 

0 60 1 20 180 240 300 360

__ __ Crank Angle _____
L—_ Center — 10mm Right 20mm Right 10mm Left —-20mm Left ]

 

 

 

 

 

 

Figure 9: Velocity Proﬁle at Location 2 (2000rpm, Flow Rate 1)

25

 

 

FIowRatdZ-LocatioMZ- 2000mm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E
E.
E‘ \
§ _.
0
>
o i
0 60 120 180 240 300 360
___ __ __ _ Crank Anglo
!___—__ Center —-— 10mm Right 20mm Right 10mm Left — 20mm Lefﬂ
Figure 10: Velocity Proﬁle at Location 2 (2000rpm, Flow Rate 2)
FIowRateﬂ-Locationn- 2000mm
35
30
T;
g 25 «
E
8
3 20 4
>
15 «
10 . .
0 60 120 180 240 300 360

Crank Angle
i — CenIer —-- 10mmjight 20mm Right 10mm Left — 20mm Left J

 

 

 

 

Figure l 1: Velocity Proﬁle at Location 2 (2000rpm, Flow Rate 3)

26

 

 

Figure 12, l3, 14 shows ﬂow velocities vs. crank angle at 2000 rpm for
location 3 and ﬂow rate 1, 2 and 3, respectively. Comparing to the results at location 2, the
velocity proﬁles Show the same type of the curve and the characteristics for the highest

and lowest velocities are also similar.

 

FiowRatoM -Location#3- 2000mm

 

20 e e

 

 

 

 

12

 
 
 

 

 

Velocity (m/s)
on

- N

 

 

 

 

 

 

 

 

 

 

0 60 120 180 240 300 360

Crank Angie
L—_ 99mg —_10r_nm__RIgttt __ 20mm Right 10r_nm Left — 20mm Left]

 

 

 

 

Figure 12: Velocity Proﬁle at Location 3 (2000rpm, Flow Rate 1)

27

 

 

 

Velocity (m/s)

FlowRaten-Locationﬂ- 2000rpm

20

 

16‘

_a
N

 

 

 

 

 

 

 

 

 

 

 

O 60 120 180 240 300 360
Crank Angie _

i_;jCenter 170%? Qght 20mm Right 10mm Left — 20mm Eeﬁl

 

 

 

 

Figure 13: Velocity Proﬁle at Location 3 (2000rpm, Flow Rate 2)

 

 

Velocity (m/s)

FIowRate#3-Looation#3— 2000rpm

 

30

25»

 

N
O

 

 

 

 

 

 

 

.s
U!

 

.1»
O
t‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0 60 120 180 240 300 360

,--s.-_-___ __ Crank Angle “ , 7 _____
[.— Center ~ 10mm Right 20mm Right 10mm Left —— 20mm Lefti

 

 

 

 

Figure 14: Velocity Proﬁle at Location 3 (2000rpm, Flow Rate 3)

28

 

 

4.1.2 Velocity Measurements at 4000 rpm

Figure 15, 16, 17 shows ﬂow velocities vs. crank angle at 4000 rpm for
location 2 and ﬂow rate 1, 2 and 3, respectively. Comparing to the velocity proﬁles at the
same location at 2000 rpm, the curves look more like the sine wave. The reason for this is
the increasing frequency piston movements overcome the offset ﬂow created by the
vacuum.

For ﬂow rate 2 and 3, we can see the data collected by the LDV system have
some ﬂuctuations between crank angle 150 and 200 degree. This shows the ﬂow
characteristics at the time of changing the ﬂow direction when piston move across the

TDC.

 

FlowRateM -Location#2- 4000mm

4o .

30 IT ——I #3
./

20 I——— «t — - /

 

 

 

 

 

 

10 - LENA---

 

 

Velocity (m/s)

 

 

 

 

 

 

 

 

 

r'o
O
1»—

 

0 60 120 180 240 300 360

__ ,.L____-_._____ mﬁ Crank Angle
[ — Center —— 10m___rr_I_ Right 20mm Right 10mm Left —- 20mm Left ]

 

 

 

 

 

 

Figure 15: Velocity Proﬁle at Location 2 (4000rpm, Flow Rate 1)

O“

29

 

 

FlowRaten-Locationn- 4000mm

 

 

 

 

 

 

 

 

 

Velocity (m/s)

 

 

 

 

 

 

 

 

 

 

 

 

0 60 1 20 180 240 300 360

Crank Angle
l — Center — 10mm Right 20mm Right , 10mm Left — 20mm Left]

 

 

 

Figure 16: Velocity Proﬁle at Location 2 (4000rpm, Flow Rate 2)

 

 

FlowRateta-Locationn- 4000mm

 

 

 

 

 

 

.— _.__--_.(

 

 

Velocity (m/s)

 

 

 

 

 

 

 

 

 

 

 

 

 

O 60 120 180 240 300 360

Crank Angle
f— Center —~ 10mm Right 20mm Right 10mm Left —— 20mm LR]

 

 

Figure 17: Velocity Proﬁle at Location 2 (4000rpm, Flow Rate 3)

30

 

 

Figure 18, 19, 20 shows ﬂow velocities vs. crank angle at 4000 rpm for
location 3 and ﬂow rate 1, 2 and 3, respectively. Comparing to the velocity proﬁles at the
same location at 2000 rpm, the curves look more like the sine wave. The reason for this is
the increasing frequency piston movements overcome the offset ﬂow created by the
vacuum. When comparing the velocity proﬁles at location 2 at the same engine speed, the

velocity magnitude are almost the same.

 

F lowRate#1 ~Location#3- 4000mm

4o ._. ~—

 

 

 

 

 

 

 

Velocity (m/s)

 

 

 

 

 

 

 

 

 

 

 

 

 

0 60 120 180 240 300 360

. -_ _ Crank Angle
I— Center - 10mm Right 20mm Right . 10mm Left —— 20mm Lefﬂ

 

 

 

 

 

Figure 18: Velocity Proﬁle at Location 3 (4000rpm, Flow Rate 1)

31

 

 

FiowRateﬂ-Locationﬂ- 4000mm

Velocity (mls)

 

0 60 120 180 240 300 360

Crank Angle
— Center — 10mm Riqht 20mm Right . 10mm Left —- 20mm Lem

 

Figure 19: Velocity Proﬁle at Location 3 (4000rpm, Flow Rate 2)

 

 

FlowRateﬂG-Locatlonﬂ— 4000mm

Velocity (mls)

 

0 60 120 180 240 300 360
Crank Angie
— Center_ — 10mm Right 20mm Right 10mm Left —— 20mm Left

 

 

Figure 20: Velocity Proﬁle at Location 3 (4000rpm, Flow Rate 3)

32

 

 

4.1.3 Discussion

When the results from three different locations are plotted in one plot, the
velocity proﬁles for three different ﬂow rates show a lot of similarities. In order to Show
the overall phenomenon, only the velocities in the center points were compared. Figure 21,
Figure 22 and Figure 23 Show the velocity results at the center of the cross section of the
pipe for three different ﬂow rates when the engine was running at 2000 rpm.

The three locations are near the piston, in the middle of the pipe and at the end
of the pipe respectively.For all three plots, we can the velocities increase from the ﬁrst
location to the second location and decrease. This trend are shown in all the plots. The
offset ﬂow created by vacuum is the main reason for this. Without the vacuum, the ﬂow is
supposed to be up and down. With the vacuum, the ﬂow is dragged backward when it is

about to go up.

 

Velocity Proﬁle at Three Locations (2000 rpm, Flow Rate 1)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20
—Location 1
16 . "Location 2
M
Location 3 // A”
12 ~—~ /
f’
E i
e r
e e e — — 1
8 t
g /
4 \ / /
0 ”WM/
0 60 tip 1%0 2410 3(50 ago
4 _ I I l i
Crank Angie

 

 

 

Figure 21: Velocity Proﬁle at Tl'tree Locations (2000 rpm, Flow Rate 1)

33

 

 

Velocity Proﬁle at Three Locations (2000 rpm, Flow Rate 2)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20 1
— Location 1
16 ~ -- Location 2
A Location 3 \
w ‘1 ——4
E, 12 ,
3‘ ‘\
.6 ,
2 it “
§ 8 R N
4 - I M #4
. l
0 60 120 180 240 300 360
Crank Angle

 

Figure 22: Velocity Proﬁle at Three Locations (2000 rpm, Flow Rate 2)

 

 

Velocity Proﬁle at Three Locations (2000 rpm, Flow Rate 3)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

30 —— —— ——
— Location 1
26 ,. -- Location 2
A Location 3
U)
5, 22
.é‘
3 K
a) 18 , —«~-- ————~—
> N I
|
14 -L-__ _:\b‘+ _‘
to I
o 60 120 180 240 300 360
Crank Angle
Figure 23: Velocity Proﬁle at Three Locations (2000 rpm, Flow Rate 3)

34

 

 

4.2 Pressure Measurements
4.2.1 Pressure Measurements at 2000 rpm

Figure 24, Figure 25 and Figure 26 show the pressure drop at 2000 rpm at three
different locations for three different flow rates. Although pressure measurements were
only taken at three locations, it can show the pressure drop trend throughout the test
region. Because the geometry is the straight pipe and the end of the pipe is always open to
air at the atmosphere pressure.

The plot which shows pressure data vs. crank angles at each location is similar

to the sine wave. The pressure starts at 14.7 psi, which is the atmosphere pressure.

 

Oscillating Flow Pressure Wave (2000 rpm at ﬂow rate #1)

155- - 15

 
  

<71
1

Pressure (DSI)

145~

 

 

 

    

0 100 200 300 1 locatIon
Crank Angle

 

 

 

Figure 24: Pressure Drop at Three Different Locations (2000rpm, Flow Rate 1)

35

 

 

OSCIIIanng Flow Pressure Wave (2000 rpm at ﬂow rate #2)

155-

 

    

Pressure (psi)

.3

1:

01
l

 

 

 

    

0 100 200 300 1 location
Crank Angie

 

Figure 25: Pressure Drop at Three Different Locations (2000rpm, Flow Rate 2)

 

 

OSCIIIatIng Flow Pressure Wave (2000 rpm at ﬂow rate #3)

 

155-

Pressure (psi)

 

 

 

   

1__00 200 300 1 location

Crank Angle

 

Figure 26: Pressure Drop at Three Different Locations (2000mm, Flow Rate 3)

36

 

 

4.2.2 Pressure Measurements at 4000 rpm

Figure 27, Figure 28 and Figure 29 show the pressure drop at 4000 rpm at three
different locations for three different ﬂow rates.

Comparing to the results at 2000 rpm, the pressure magnitudes are larger and
the curves of pressure data vs. crank angle at different locations are similar to the Sine

anCS .

 

Oscnlating Flow Pressure Wave (4000 rpm at ﬂow rate #1)

Pressure (DSI)

 

 

 

O 100 200 300
Crank Angie

Iocahon

 

 

 

Figure 27: Pressure Drop at Three Different Locations (4000rpm, Flow Rate 1)

37

 

 

.s
O)

_.
A

Pressure (pSI)

 

       

0 _.
100 200 300 1 location
Crank Angle

 

Figure 28: Pressure Drop at Three Different Locations (4000rpm, Flow Rate 2)

 

 

Oscﬂianng Flow Pressure Wave (4000 rpm at ﬂow rate #3)

20-
19—V
18d

   

a ...
o: w
I I

PSI)

Pressure

 

 

 

0 100 200 300
~ Crank Angie

location

 

Figure 29: Pressure Drop at Three Different Locations (4000rpm, Flow Rate 3)

38

 

 

CHAPTER 5. NUMERICAL SIMULATION SETUP

5.1 Description of Problem

A schematic diagram of the problem investigated is shown in Figure 30.
Basically, it involves the ﬂow (air) in a straight pipe of round shape cross section with
smooth walls. Because this problem is axisymmetric and the most important
characteristics of the ﬂow is the pressure and vertical velocity component, a two-

dimensional analysis is used. The diagram shows the section selected for the simulation.

 

 

 

 

 

 

r/ W
T R , r = 1.223"
est egIon R = 1.425"
\ x D = 1.723"
L = 50"
Ll = 4.55"
Center Line L2 L2 = 44.9"
\ L
2
Vacuum
(Offset Flow) i
X
. . LI
Cylinder Region -
. ii

 

 

 

 

 

 

Figure 30: Schematic of Problem Studied

39

The simulation commences at bottom dead centre (BDC) and continues for a
time interval corresponding to 360 of crank revolution. The ﬂuid within the cylinder is
assumed to be air, initially at room temperature and pressure. The ﬂow is assumed to be
turbulent and driven entirely by the motion of the piston and the pressure difference
created by the vacuum. The pressures on the top of the pipe are held constant (at ambient

conditions) throughout the simulation.

5.2 Formulation of Problem
5.2.1 Governing Equation
The governing equations for this problem were taken to be the ensemble-

averaged conservation equations of mass (continuity), momentum (compressible Navier-

Stokes), and total energy, closed by a high Reynolds number K — 8 model.

92 L92 in: Bow-
Bt+8x+ay+az ‘0 (I)

 

 

 

 

 

a pa 8 puu a puv a puw 2)
3; RV +5 mm *7, pvv +3; pvw (
M: W“ WV WW
I, ‘ r I ‘
+t x
a” *" ,2. . ,
— 8x Tyx 8y p +1”
Tax TZY J
8 a * * *
a—:+—a-;(e+p )u+§a§(e+p )v+-§:(e+p )w (3)
= 3Emma +Txyv+1xzw—qx)

a
+ 3(1),“: + I), v + “Cyzw — gy)

y

+—a—(1: u+t v+t w— )
az zx :y :z q:

40

Where

p“ = p+ get (4)
5.2.2 Turbulent Model

The turbulent model used in this study is high Reynolds k - a model. It is the
standard model, in which the high (turbulent) Reynolds number forms of the k and 8

equations are used in conjunction with algebraic ‘Law of the wall’ representations of ﬂow,

heat and mass transfer for the near wall region. The initial turbulent kinematic energy and
8 levels were deﬁned before the simulation. The initialization of the turbulent intensity I

at the inlet is 0.05 and the mixing length l is 0.0047 In (0.1D). Those quantities are used to

calculate the initial TKE and 8 levels.

k = 1.5><(Ux1)2 (5)
1.5
0.75 k '

e = Cu X_l (6)

For this problem, the Reynold number is 2 x 105. The Reynold number was

calculated at the main stream in the test region.

5.2.3 Boundary Conditions and Initial Conditions

The boundary and initial conditions for the governing equations and turbulent
modes are as follows.

At the pipe inlet, the pressure an temperature as well as the turbulent quantities
(turbulent intensity TI and mixing length l) were speciﬁed. The pressure was deﬁned as

the atmosphere pressure and the temperature was deﬁned as the room temperature 293k.

41

At the pipe outlet, an average pressure was imposed for three different ﬂow rates. At the
walls of the pipe, the no-slip and adiabatic conditions were imposed.

There are three types boundary conditions deﬁned in STAR-CD. The inlet and
outlet were deﬁned as the pressure boundaries. At the pipe inlet, i.e. at the top of the pipe,
the pressure was deﬁned as the atmosphere pressure and the turbulence quantities were
speciﬁed. At the pipe outlet, i.e. where the offset ﬂow passed through, the vacuum
pressures were deﬁned for three different ﬂow rates. The vacuum pressures were -4 kpa, -
20 kpa and -40 kpa respectively (all relative to atmosphere pressure).The symmetry
boundaries were deﬁned to reduce the size of the computational mesh by placing the
boundary along the planes of geometrical and ﬂow symmetry. The pipe surface and the
cylinder surface are deﬁned as the wall boundaries. The wall was set to ‘no slip’ and
adiabatic. The wall roughness was set to be standard.

The initial conditions of this transient problem are: room temperature and
pressure. The piston starts from the BDC. At the last step of each run, the results were

written into a restart ﬁle. Then for the next run, the analysis starts from the restart ﬁle.

5.3 Numerical Method of Solution
5.3.1 Introduction to STAR-CD

Solutions to the government equations described in the previous section were
obtained by using the STAR—CD version 3.10, developed by Analysis and Design
Applications Company, Ltd. STAR-CD is one of the leading multi-purpose therrnoﬂuids
analysis codes. It is designed for engineers and scientists requiring a versatile, efﬁcient
and robust software package capable of modeling ﬂuid ﬂow, heat transfer, mass transfer

and chemical reaction in industry and the environment.

42

STAR—CD is a self-contained package incorporating: mathematical models of a
wide range of thermoﬂuids phenomena, powerful solvers enabling full complex-geometry
capabilities and ﬂexible pre-processing and post-processing facilities tailored for both
expert and novice users. The code can also be readily integrated with other computer-
aided engineering software for any of the following functions: surface modeling, mesh
generation, mesh and results display and provision of STAR boundary information to
structural analysis codes. This is achieved by internal translators between the STAR-CD
and external ﬁle formats, including PATRAN, IDEAS, ANSYS, GRID3D, NASTRAN,
STL AND IGES. STAR-CD has been ported to a wide range of computers comprising
PCs, workstations, minicomputers, mainframe machines and super computers. Optimized

versions have also been developed for vector, parallel and vector/parallel architectures.

5.3.2 Solution Algorithm

STAR-CD currently incorporates three different implicit algorithms, namely:
SIMPLE, P180 and SIMPISO. Since P180 is applicable to both transient and steady-state
calculations and is particularly suitable for the former, for which it has been shown to be
considerably more efﬁcient than iterative methods, PISO was chosen as the algorithm used
in this study.

PISO performs at each time (or iteration) step, a predictor, followed by a
number of correctors, during which linear equation sets are solved iteratively for each
main dependent variable. The decisions on the number of correctors and inner iterations
(hereafter referred to as ‘sweeps’, to avoid confusion with outer iterations performed as
part of the steady-state solution mode) are made internally on the basis of the splitting

error and inner residual levels, respectively, according to prescribed tolerances and upper

43

limits. The default values for the solver tolerances and maximum correctors and sweeps

are given in Table 4.

Table 4: Standard Control Parameter Settings for Transient PISO Calculations

 

 

 

 

 

 

Variable
Parameter
Velocity Pressure Thrbulence Enthalpy Concentration
501”“ 0.01 0.001 0.01 0.01 0.01
tolerance
Sweep limit 100 1000 100 100 100

 

 

 

 

 

Correction limit = 20

 

Pressure correction relaxation factor = 1.0

 

 

 

Corrector step tolerance = 0.25

 

The remaining key parameter in transient calculations with P180 is the size of
the time increment 5t. This is normally determined by accuracy considerations and may
be varied during the course of the calculation. The step should ideally be of the same order

of magnitude as the smallest characteristic time 5’c for convection and diffusion, i.e.

2
SIC = min(%, Ear—L) (7)

Here, U and F are a characteristic velocity and diffusivity, respectively, and SL is a mean
mesh dimension. Typically, it is possible to operate with St = 50 SIC and still obtain

reasonable temporal accuracy. Values signiﬁcantly above this may lead to errors and
numerical instability, whereas smaller values will lead to increased computing times.
During the course of a calculation, the limits given in Table 4 may be reached,

in which case messages to this effect will be produced. This is most likely to occur during

the start-up phase but is nevertheless acceptable if, later on, the warnings either cease
entirely or only appear occasionally, and the predictions look reasonable. If, however, the
warnings persist, corrective actions should be taken. The possible actions are: reduction in
time step, increasing in the sweep limits, changing the value of pressure correction under-

relaxation, or setting the corrector step tolerance to a lower value.

5.3.3 Grid System Used

 

 

 

   

Pipe Mesh
Center Line

Vacuum

Cylinder Mesh

 

 

 

Figure 31: Grid system Used

The domain of the problem was replaced by a single-block grid system that
was assembled by patching together two blocks of structured grids. It is shown in Figure

31. The two blocks are cylinder mesh and the pipe mesh in the test region. The total

number of deﬁned cells is 30654. The mesh was created by Gridgen. Gridgen is a software

system for the generation of 3D grids.

5.3.4 Solution Procedures

Firstly, run the problem without piston movement and observe the streamlines
in the duct, especially the velocity proﬁle in the oriﬁce.

Then close the vacuum part and set boundary and initial conditions as
previously discussed and run the ﬁrst cycle. Then run the Simulation several cycles until
the periodical results were achieved.

After that, open the vacuum part and slowly decrease the vacuum pressure to
run the problem under different ﬂow rates. For each case, several runs were needed to

reach the periodical results.

46

CHAPTER 6. NUMERICAL SIMULATION RESULTS AND DISSCUTION

6.1 Velocity Results
6.1.1 Velocity Proﬁle for Different Flow Rates at 2000 rpm

From Figure 32 to Figure 37, velocity proﬁles at 2000 rpm at three different
locations and three different ﬂow rates are shown. Since the volume of post data ﬁles are
huge, only the cells at the same locations were investigated. The velocity proﬁles were
plotted vs. crank angle at 20 degree interval. The comparison to the experimental results

will be discussed in the next chapter.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

STAR“
PROSTAR 3.10
VEL. COMP W
W8
20 40 60 80 100 120 2000
1800
"- 1600
: 14 00
. I s ' I I : iii
i I i I ' ' 5000
. I I i 3383
00000600
MO 160 180 m 220 240
i i L i Z
260 280 300 320 340 360 i x

Figure 32: Velocﬁyﬁﬁle at Loication 17(2000rpm, Flow Rate 1)

47

 

 

SAR“

 

 

 

 

 

 

 

 

 

A i l i I i I ) PROSTAR 3.10
-‘ I I . i V '
VEL COMP w
20 40 60 80 100 120
25 00
22 50
20 DO
— 1750
— 15.00
, 12.50
10.00
5000
2 500
00000900
[40 I60 180 200 220 2 0
260 280 300 320 340 360

 

 

 

L.

Figure 33: Velocity Proﬁle at Location 2 (2000rpm, Flow Rate 2)

 

 

20

H

40

80

120

5 AP.‘

PROSTAR 3.10

 

I40

I60

180

200

220

 

 

260

 

280

 

 

300

H
H

320

 

ii
ii

 

340

360

 

 

 

 

 

Figure 34: Velocity Proﬁle at Location 3 (2000rpm, Flow Rate 3)

48

6.1.2 Velocity Proﬁle for Different Flow Rates at 4000 rpm

 

 

 

 

 

 

 

 

 

 

 

0 "40 60 80 I 120
ii
ii
I!

I40 160 180 200 220 240
260 28ft 7 300 U 32 U 340 360

 

 

 

S Aﬁ‘

PROSTAR 3.10

 

Figure 35: Velocity Proﬁle at Location 1 (4000rpm, Flow Rate 1)

 

_J

 

 

ll

80

3 AR“

PROSTAR 3.10

 

ii
II

140 I60

180

220

 

 

 

 

2 280

 

300

320

 

 

340

 

360

 

 

L—.-

IFigure 36: vase}; Profile 2.; LooRoha (4000rpm, FlowIRiatEIZ)

49

illll

l
l

0 0000E~00

 

;._i

 

 

 

 

 

 

 

 

 

 

20 40 60 80 100 120
I I 5. i
I i ‘I g
1 I 1"
I
140 160 I80 200 i7 220 240
264) 280 H 30(il 32C 3 360

 

 

 

Sim-I“

PROSTAR 3.10

 

 

Figure 37: Velocity Proﬁle at Location 3 (4000rpm, Flow Rate3)ir

6.1.3 Velocity proﬁles in the region of oriﬁce (outlet of the pipe)

The velocity proﬁles in the region of the oriﬁce are affected by both piston
movement and the vacuum pressure difference. It is important to Show the ﬂow patterns in

this region. Because the proﬁles in this region will inﬂuence the results in the upstream

calculations.

From Figure 38 to Figure 43, the velocity proﬁles near the region of oriﬁce
with function of the crank angle are shown, when engine was running at 2000 rpm and
ﬂow rate 1, i.e. the vacuum pressure was set at 4kpa below the atmosphere pressure.

From Figure 44 to Figure 49, the velocity proﬁles near the region of oriﬁce

with function of the crank angle are shown, when engine was running at 2000 rpm and

ﬂow rate 1.

50

6 Cl 0
0 0000Eo00

Z

L.

 

 

 

 

E E

m D
a I . m
l. N I. 0 X m. N w 0 X
3 G w 0 3 G N no.
m m. m m m; m

a .
T . wmmwmwwmmmm Z v. r wwmmmmmmmmm Z
S W I 0 DDDDDD 0000 "M m m,” 00000000000
0 a .L 196765 3210 O C .t19076543210.
R .. R 3
s p m _- s p m t _

m
: ‘

  
 

 

 

.3» /

/ .. ///

/////iii \\ ///

’/////ii \‘~///

:,/////i \:,//
\‘.////// iiiiiii l\\\\\\\\\\~’//////f/
iii\‘\\\\\\\\~,zrz/l_ iiiiiii \\\\\\\\\‘.//////f
ii’lill‘\‘\\\\\\\\\‘,znrllm iiiiiii \\\\\\\\\.4’////fl
iii-litttt\\\\\\\\\\\\\.,, It 2 iiiiiii \\\\\\\\_.,////li
iti’ltt\\\\\\\\\\\\\a.. ,/ L iiiiiii \\\\\\\\..,///III
iilIIIIIiIII\1\\\\\\\\\\... I iiiiiiii \\\\\\L., IIIIII
IIIIIII’I’IIII‘II\‘\I\\\\\\\\\\ ..... 7 iiiiiiiii \\\\\~_a Ill},
IIIIIIIIII’I‘lll‘l‘t‘\\\\\\ ....... 7 iiiiiiiiii \\ \ x ‘ . a I III!
IIIIIIIIIEIIIIII\\\\\\\ \ " I'liiti'tiltll.|\\\\ \ x . / I I If
I'IIIII‘I'I'I“I*\“‘\ \ i I'tl'll'tl‘l‘l'l‘l“\\\ \ t llllll
Iiittllittt. 3 iiiiliiitttt.. -iiii

i

 

 

Figure 38: Velocity Proﬁle in the Region of Oriﬁce at 2000 rpm (CA=60)

 

 

 

51

 

Figure 39: VelocityProﬁle in the Region of Oriﬁce at 2000 rpm (CA=120)

 

 

 

 

    

 

SJAIT‘
PROSTAR310
VELOCITVMAGNFI'UDE
1000
tltiitx iii
60.00
l\\\\\\\\ \\ ;3333
l i\\\\\\\\\ :3333
ll \\\\\\\\\\ VI —,000
lll\\\\ \\\\\\\\x\~ > — 000005.00
ll\\\\\\\\\\\\ g
lll\\\\\\\\\\—*”'7 \
lit\\\\\\\\\~’/
l\\\\\\ \\\\\\ ’//
\\\\\ \\\\\\\\ I/
\l\\\ \\\\\\\\ //
|| \ - -2 II 2
‘ L.

 

 

 

 

 

3m?“

PROSTAR 3.10

VELOCITYMAGNITUDE

l
|
m
a
o
o

illll“

1
0 0000E~00

 

4.4..

 

 

 

 

 

 

 

 

 

 

 

’/////’%’

I L.

 

 

 

 

FigtIre 41: Velocity Proﬁle in the Region of OriﬁcIa—at 2000 rpm (CA=240)

52

 

0
0

o
E
0
0
0
0
0

PROSTAR 3.10

2.5

VE LOCFW MAGNITUDE

    

:\V\\\\\\\ \ \

|\\\\\\\\\\\ \ \

 

 

. .
‘ c
\ \
\ x
\

tllI‘I‘AI‘\\\\\ \

"I‘\\\\\\\Q

 

\
\
\

.Illt“\\\\\\\\\\ \.
\
\

 

i‘i‘h“\\‘\‘\\ \ \.

 

 

I-|||\‘\L‘\\\\I\ .\ ti

 

 

 

I‘I‘IIII.

IIIII'IIII.

lr‘l'IIIIIt .

 

 

 

 

360)

777!7-—_i

Figure 42: Velocity Proﬁle in the Region of Oriﬁce at 2000 rpm (CA=300)

0 0000E000

PROSTAR 3.10

VELOCI'W MAGNITUDE

53

 

 

lIIIIIIEICNIS‘\\\\\ \

 

 

 

i§\\\\\i \ \\

 

 

 

 

Figure 43. Velocity Profile in the Region of Oriﬁce at 2000 rpm (CA

 

 

 

 

 

 

 

U
do.
w
T
s m

E.
D.
U_.
.H
N.
m”
m
C
m
w

  

 

\ \ \\\\\\\\W

\ \ \ \

 

I/I/II

;Il/”

 

__u

60)

_

A
C
(
m
m.
4
t
a
c
C
m
f
0
n
.m
g
e
R
e
L h
t
.m
.mic
H
m
P
.m.
d
V
m
u
.We
F

 

 

 

m
sm

m
m
m
m

  
 

M/S
-,:,‘: VIU‘IL‘ 1.,1: .‘r‘iii‘

rrrrrrrr

llllllllll

lllllllll

rrrrrrrrr

lllllllllll

t I I f

x
L
.

I
I

I / / / l/l/l/I/f’l‘f"

/ / / /////l/l/l/ll‘iull
I I / /I//l/l/’//‘f‘.i"
I I/ll/fl/l/fl‘i‘itif
I I I I’ll/(I’lf;
I I [I’ll/ff};
I I III/flitti/‘f‘i-li
i i filliliiilitl

i i fittllitltl

' fIIOI'lrrII

 

 

Figure’45: Velocify Proﬁle in the Region of Oriﬁce at 4000 rpm (CA=120)

 

 

54

 

AP."

PROSTAR 3.10

VELOCFI'VMAGNITUDE

 

 

 

 

N
A
o
o

12 00
0 00005000

x / / //////”W

I

r I / z ////’/’W

- Cytfrdrrf

 

---.—.—,.,".——

 

A---.—-"yy

 

 

 

 

SAR“

PROSTAR 3.10

VELOCITYMAGNITUDE

 

N
h
D
D

 

 

.—-.._

 

 

 

 

 

 

 

 

 

 

 

;

 

 

Figure 47: Velocity Proﬁle’in—the Region ofOriﬁee atA4000 rpm (CA=240)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SAP."

PROSTAR 3.10

VELOCIWMAGNITUDE

N
A
o
t:

12 00
0 0000Eo00

 

 

 

 

 

 

 

 

 

 

 

 

ill
lll\\\\
.::::M
\t\\\\\\
1\\\\\\\\
111\\\\\\\\
lll\\\\\\\\
lll\\\\\\\\
ttl\\\\\\\\
ilt\\\\\\\\
lii\\\‘\‘\\
||1\\‘\"1j

 

SAR“

PROSTAR 3.10

VELOCITYMAGNITUDE

N
A
D
1:)

12 00
0 0000Eo00

L.

Figure 49: veioeity Proﬁieiu The Region of Oriﬁce at 4000 rpm (CA=360)

56

 

 

In Figure 49, there is a vertex near the oriﬁce. The reason for this is the
inﬂuences of the piston movements and the vacuum pressure. Since the velocities in the
pipe reach to the maximum values at about 300 crank angle, the drag force created by the
piston plays an important role. Also, comparing the velocity proﬁles in the region of
oriﬁce between 2000 rpm and 4000 rpm, the piston movement plays a more important role

in the latter one.

6.2 Pressure Results
6.2.1 Pressure Results for 2000 rpm.
From Figure 50 to Figure 55 the pressure wave development in the ﬁrst cycle

are shown when engine is running at 2000 rpm.

 

 

i!i!::’
PROSTAR310

PRESSURE
RELATRE
PA

15 0
0 0000E¢00

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 50: Pressure Results for Crank AngTe IK60 Degree (2000 rpm)

57

 

 

 

SIAR‘

i

PROSTAR 3 10

PRESSURE
RELATIVE
PA

W
row
air

QC?
DD

 

4 0 0
0.0000900

 

 

 

 

 

 

 

 

 

 

 

Figugl: Pram Results forCrEA—ngle 70 - 120 Deg‘rgeTZIOODTrp-rﬁ)

 

 

 

SIAP."
I

i' PROSTAR 3.10

irr

PRESSURE
RELATIVE
PA

‘ Agrarian-.1

 

 

a-‘i‘rRErr’ﬁf'ﬁp‘s’aff‘I‘thb-f!

 

 

 

 

 

if L

 

 

 

 

 

Figiuiei52: Pressure Results IfOr CrankArTgle 130 - 180 Degree (2000 rpm)

58

 

 

 

STAR"

PROSTAR310

PRESSURE
RELATNE
PA

' 34 0
gr -4200

Tgﬂi’lffnkt
at
or
o
c:

 

 

 

 

 

 

 

 

 

 

ifli L,

 

 

 

 

 

Figure 53: PressureResults for Crank Angle 190 - 240 Degree (2000 rpm)

 

 

 

SARV

PROSTNR310

PRESSURE
RELATNE
PA

nun“
U)
D
D
O

 

 

 

 

 

 

 

 

 

 

I L

 

 

Figure 54: Pressure Reisults for Crank AngTe 250300 Degree (2000 rpm)

59

 

 

 

syAﬁ?‘
it
PROSTAR 3 10

PRESSURE
RE LAT IVE
PA

 

 

 

 

 

 

 

 

 

 

 

rifw

Figure 55: Pressure Results f0; Crank Angle E510 - 360 BegﬁreeaOORO rpm~)

 

 

 

 

 

From Figure 50 to Figure 55, we can see how the pressure wave developed in
the ﬁrst cycle. This can be demonstrated easily by the numerical simulation. On the
contrary, it is very difﬁcult for the experimental approach to be applied when the engine

just turned on.

6.2.2 Pressure Results for 4000 rpm
From Figure 56 to Figure 61 the pressure wave development in the ﬁrst cycle

are shown when engine is running at 4000 rpm.

60

 

 

sm‘

PROSTARSJU
PRESSURE
RELKTNE
PA

0 OOOOEoOU
-50 ,U

 

 

 

 

 

 

 

 

 

Illﬁg

lle

 

 

 

 

 

Figure 56: Pressure Results for Crank Angle 10 — 60 Degree (4000 rpm)

 

 

Aﬁ‘

PROSTAR3JU

PRESSURE
RELATNE
PA

 

 

 

 

 

 

 

 

 

 

 

 

i .J' . , . z
5"” L x

 

 

 

Figure 57: Pressure Results for Crank Angle 70 — 120 Degree (4000 rpm)

61

 

 

 

 

 

 

 

a Li:

a).

B’

 

I" fat: .6:

 

 

s‘mi‘
E

PROSTAR 310

PRESSURE
RELATIVE
PA

 

___;

 

Figure 58: Pressure Results for Crank Angle 130 - 180 Degree (4000 rpm)

 

 

 

1.3!,‘r’i1‘ Uh

 

 

 

 

.,

:.M‘N,z~'—

 

i

 

 

 

 

 

 

 

a

‘

PROSTAR 3‘ l 0
PRESSURE
RELATIVE
PA

0 UOOOEﬂO
8 U 0

 

L

EU; 59: Pressure Results for CrankA-ngle 190 - 240 Degree (4000 rpm)

62

 

 

 

SITAR“
2'
(a.

".1 1.‘ a PROSTAR 3.10

wen-”w

PRESSURE
RE LATNE
PA

0 OOODE‘OO
—500 0

 

 

 

 

 

 

 

 

 

 

 

 

 

trill L

 

 

 

Figure 60: Pressure Results for Crank Angle 250 - 300 Degree (4000 rpm)

 

 

 

 

r“ " " STAﬁ‘
:

PROSTAR 3.10

PRESSURE
RELATIVE
PA

(a we
I I
N
o
a
o

 

 

 

 

 

 

 

Figure 61: Pressure Results for CrankiAnigle 310 :360 Degree (21000 rpm)

 

 

63

 

The periodical results at the same locations where the measurements were

taken will be compared in the next chapter.

64

CHAPTER 7. COMPARISON

7.1 Compare the CFD Velocity Results with LDV Results.
7.1.1 Comparison at 2000 rpm

From the vast amount of data from the simulation, only a few major ﬁndings
are discussed in this chapter. In order to compare the results of simulation and LDV, we
should choose the cell or cells in the grid at the same location and the comparison must be
in the same conditions, i.e. same engine speed and ﬂow rate. All velocity results are
plotted as the function of the crank angle. The three locations are 12 inches, 25inches and
43inches from the piston top surface. Since the post data ﬁles in STAR-CD recorded data
every 10 crank angle degrees, the results from CFD in the ﬁgures are plotted every 10
crank angle degrees. Because the size of the post ﬁle would be huge if the data were saved
during the simulations for every crank angle during the runs.

From Figure 62 to Figure 64, the comparisons are shown at the ﬁrst flow rate.
At all three locations, the predicted and experimental results show the same trend lines.
There is a phase lag between two plots in each ﬁgure. It is about 30 to 40 degree. The
largest discrepancy is between 30 and 120 crank angle when piston moves from BDC to
TDC in Figure 63. The predicted results are almost twice as the measured ones. The
velocity magnitude differences between two results for all three ﬁgures are less than 10
percent for the maximum values. The velocity magnitude differences between two results

for all three ﬁgures are about 50 percent for the minimum values.

65

 

 

Velocity (mls)

.-
M

Q
- A
r

Comparison at Location#1 Flow Rate#1 2000rpm (LDV vs CFD)

//

 

 

  

 

 

 

 

 

 

 

 

180 240 300 360

Crank Angle

60 120

 

Figure 62: Velocity Proﬁle Comparison at 2000 rpm (Location 1, Flow Rate 1)

 

 

Velocity (mls)

_A
N

on

0

gm m w l
l Os

.1;

Comparison at Location#2 Flow Rate#1 2000rpm (LDV vs CFD)

 

 

l f AV
\ I l /
W t

0 60 120 180 240
Crank Angle

\
\

 

 

 

 

 

 

 

 

300 360

 

Figure 63: Velocity Proﬁle Comparison at 2000 rpm (Location2, Flow Rate 1)

66

 

 

 

Comparison at Location#3 Flow Rate#1 2000rpm (LDV vs CF D)

16-

 

— LDV + CFD

1: i __ / \\\
Z5 A

0 60 120 180 240 300 360
Crank Angle

 

 

 

 

 

 

 

Velocity (mls)

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 64: Velocity Proﬁle Comparison at 2000 rpm (Location 3, Flow Rate 1)

From Figure 65 to Figure 67, the comparisons are shown at the second flow
rate. In all three ﬁgures, the predicted and experimental results show the same trend lines.
There is a phase lag between two plots at location 1 and 3. It is about 30 to 40 degree. The
velocity magnitude differences between two results for all three ﬁgures are less than 10
percent for the maximum values in location 1 and 3. For location 2, the difference is about
30 percent. The velocity magnitude differences between two results for all three ﬁgures

are about 40 percent for the maximum values for location 2 and 3.

67

 

 

Comparison at Location#1 Flow Rate#2 2000rpm (LDV vs CFD)

— LDV + CFD

Velocity (mls)

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 65: Velocity Proﬁle Comparison at 2000 rpm (Location 1, Flow Rate 2)

 

 

Comparison at Location#2 Flow Rate#2 2000rpm (LDV vs CFD)

 

 

 

 

 

 

 

 

 

 

 

 

25
—-LDV
20 77A
g 15 7, J W—
a
.9 .l
E 10 \‘hjbv-
5 \v
o , , .
o 60 120 180 240 300 360

Crank Angle

 

Figure 66: Velocity Proﬁle Comparison at 2000 rpm (Location 2, Flow Rate 2)

68

 

 

 

Comparison at Location#3 Flow Rate#2 2000rpm (LDV vs CFD)

20 -— —-1~——
— LDV + CFD

,6 _ /" L
/

 

 

 

 

 

 

 

Velocity (m/s)

 

 

 

 

 

 

 

 

 

 

0 60 1 20 1 80 240 300 360
Crank Angle

 

 

 

Figure 67: Velocity Proﬁle Comparison at 2000 rpm (Location 3, Flow Rate 2)

From Figure 68 to Figure 70, the comparisons are shown at the third ﬂow rate.
The velocity magnitudes for this one are much larger than the previous two flow rates. In
all three ﬁgures, the predicted and experimental results show the same trend lines. There is

also a phase delay between two plots at all locations. It is about 40 degree.

69

 

 

Comparison at Location#1 Flow Rate#3 2000rpm (LDV vs CFD)

30 aye __,_ ‘1" - 777 -- 1» WL",TT ,,
—LDV +CFD ‘ i ‘

  

Velocity (mls)

 

 

 

 

180 240 300 360
Crank Angle

 

Figure 68: Velocity Proﬁle Comparison at 2000 rpm (Location 1, Flow Rate 3)

 

 

Comparison at Location#2 Flow Rale#3 2000rpm (LDV vs CFD)

—- LDV + CFD

Velocity (mls)

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 69: Velocity Proﬁle Comparison at 2000 rpm (Location 2, Flow Rate 3)

70

 

 

 

Comparison at Location#3 Flow Rate#3 2000rpm (LDV vs CFD)

 

 

 

30 —-—~—-—~#—T————_ —-—
— LDV + CFD

26 -
j/

22 «e—ﬂ— ,_s /f'
18 k 1'
5g-

14

 

 

 

Velocity (m/s)

 

 

 

 

 

 

 

 

1O

 

0 60 120 180 240 300 360
Crank Angle

 

 

 

Figure 70: Velocity Proﬁle Comparison at 2000 rpm (Location 3, Flow Rate 3)

7.1.2 Comparison at 4000 rpm

From Figure 71 to Figure 76, the comparisons are shown at 4000 rpm at two
different locations (location 2 and location 3) and at three different ﬂow rates.

In all six ﬁgures, the predicted and experimental results show the same trend
lines. In Figure 75 and 76, the predicted results show some negative values between ‘60

and 180 degree. That means the velocity direction is away from the piston.

71

 

 

Comparison at Location#2 Flow Rate#1 4000rpm (LDV vs CFD)

Velocity (mls)

— LDV + CFD

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 71: Velocity Proﬁle Comparison at 4000 rpm (Location 2, Flow Rate 1)

 

 

Comparison at Location#3 Flow Rate#1 4000mm (LDV vs CFD)

Velocity (mls)

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 72: Velocity Proﬁle Comparison at 4000 rpm (Location 3, Flow Rate 1)

72

 

 

 

 

Velocity (mls)

 

Comparison at Location#2 Flow Rate#2 4000rpm (LDV vs CFD)

— LDV + CFD

60 120 180 240 300 360
Crank Angle

 

Figure 73: Velocity Proﬁle Comparison at 4000 rpm (Location 2, Flow Rate 2)

 

 

Velocity (mls)

 

Comparison at Location#3 Flow Rate#2 4000mm (LDV vs CFD)

—- LDV + CFD

60 120 180 240 300 360
Crank Angle

 

Figure 74: Velocity Proﬁle Comparison at 4000 rpm (Location 3, Flow Rate 2)

73

 

 

 

 

Comparison at Location#2 Flow Rate#3 4000rpm (LDV vs CFD)

 

 

Velocity (mls)

 

 

 

 

 

 

Crank Angle

 

Figure 75: Velocity Proﬁle Comparison at 4000 rpm (Location 2, Flow Rate 3)

 

 

Comparison at Location#3 Flow Rate#3 4000rpm (LDV vs CFD)

Velocity (mls)

'—LDV +CFD

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 76: Velocity Proﬁle Comparison at 4000 rpm (Location 3, Flow Rate 3)

74

 

 

7.2 Compare the CF D Pressure Results with Experimental Results
7.2.1 Comparison at 2000 rpm

The results’ unit from the pressure transducer measurements was changed from
pa to psi and then compared to the CFD in this section. The comparisons at 2000 rpm are
shown from Figure 77 to Figure 85. The simulation results are compared with
experimental data at every location and at each ﬂow rate.

All the comparisons at 2000 rpm show the same trend lines. All the pressure
waves have the similar shape to the letter w. The differences between the measured and
predicted pressure data are all less than 10 percent, which is much better than the

comparisons of the velocity.

 

Comparison at Location#1 Flow Rate#1 2000 rpm (Experiment vs CFD)

 

 

15.5 7 _
— Expenment

+ CFD
' /'

14.75

 

 

_a
01

 

 

 

Pressure (psi)

0 60 120 180 240 300 360
Crank Angle

14.5 -

 

14.25 -

 

 

 

 

 

 

 

 

 

14

 

 

 

 

Figure 77: Pressure Data Comparison at 2000 rpm (Location 1, Flow Rate 1)

75

 

 

Comparison at Location#2 Flow Rate#1 2000 rpm (Experiment vs CFD)

— Experiment

+ CFD

Pressure (psi)

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 78: Pressure Data Comparison at 2000 rpm (Location 2, Flow Rate 1)

 

 

Comparison at Location” Flow Rate#1 2000 rpm (Experiment vs CFD)

— Experiment
+ CFD

Pressure (psi)

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 79: Pressure Data Comparison at 2000 rpm (Location 3, Flow Rate 1)

76

 

 

 

 

Comparison at Location!“ Flow Rate#2 2000 rpm (Experiment vs CFD)

15.5 ”ieiww—

*-

15.25

14.75

Pressure (psi)

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 80: Pressure Data Comparison at 2000 rpm (Location 1, Flow Rate 2)

 

 

Comparison at Location#2 Flow Rate#2 2000 rpm (Experiment vs CFD)

15.57 re. .. ——~r W ~ MA# — ,_-.__
—Experiment

+ CFD
15.25

 

 

 

 

 

Pressure (psi)

 

 

 

14.5 , ,

 

 

 

 

14.25

 

 

_ Crank Angle

 

Figure 81: Pressure Data Comparison at 2000 rpm (Location 2, Flow Rate 2)

77

 

 

 

 

 

Comparison at Location#3 Flow Rate#2 2000 rpm (Experiment vs CFD)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

15.5 «—'--————---- »-—- ——— ——————-- —— ——
—Experiment
+CFD .
15.25 . , ____E
E 15 A
a, l
5 .
14 75
i . “—"“1'—.“ ‘i-“‘
14.5 « ———e
14.25 i
0 60 120 180 240 300 360
Crank Angle

 

Figure 82: Pressure Data Comparison at 2000 rpm (Location 3, Flow Rate 2)

 

 

15.25

15

Pressure (psi)

14.5

14.25

14

Comparison at Location#1 Flow Rate#3 2000 rpm (Experiment vs CFD)

 

-— Experiment I

+ CFD

 

 

14.75 '.-

-<

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O 60

120 180 240 300 360

Crank Angle

 

Figure 83: Pressure Data Comparison at 2000 rpm (Location 1, Flow Rate 3)

78

 

 

 

 

15.25

15

14.75

Pressure (psi)

14.5

14.25

Comparison at Location#2 Flow Rate#3 2000 rpm (Experiment vs CFD)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- -____---_ T“ --_ __
— Experiment
+ CF 0
r I
l
l
l
0 60 120 180 240 300 360
Crank Angle

 

Figure 84: Pressure Data Comparison at 2000 rpm (Location 2, Flow Rate 3)

 

 

15.25

15

14.75

Pressure (psi)

14.5 *4

14.25

Comparison at Location#3 Flow Rate#3 2000 rpm (Experiment vs CFD)

 

 

-- Experiment

+ CFD

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60 120 1 80 360

_ Crank Angle

240 300

 

Figure 85: Pressure Data Comparison at 2000 rpm (Location 3, Flow Rate 3)

79

 

 

7.2.2 Comparison at 4000 rpm

The comparisons at 4000 rpm are shown from Figure 86 to Figure 94. The
simulation results are compared with experimental data at every location and at each flow
rate.

Both experimental and simulated results have the similar shape of a sine wave.
The results for location 3 are better matched because it is near the inlet of the pipe, where

the pressure always keeps at the atmosphere pressure.

 

Comparison at Location#1 Flow Rate#1 4000 rpm (Experiment vs CFD)

 

 

— Experiment + CFD

 

\

 

 

 

/_.

 

Pressure (psi)

 

 

 

 

 

 

 

 

 

0 60 120 180 240 300 360
Crank Angle

 

 

 

 

Figure 86: Pressure Data Comparison at 4000 rpm (Location 1, Flow Rate 1)

80

.“‘.'!£n‘-'

 

 

 

Comparison at Location#2 Flow Rate!" 4000 rpm (Experiment vs CFD)

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

ZOT—mr”’ l "
. l l
— Experiment + CFD} [
1a 1. i.,,,ﬁ_,,,,,,,_l,‘_
:2: 16— , 7,
2
a
g / ,,
a.
10
0 60 120 180 240 300 360
CrankAngle

 

Figure 87: Pressure Data Comparison at 4000 rpm (Location 2, Flow Rate 1)

 

 

Comparison at Location#3 Flow Rate!" 4000 rpm (Experiment vs CFD)

Pressure (psi)

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 88: Pressure Data Comparison at 4000 rpm (Location 3, Flow Rate 1)

81

 

 

 

 

Comparison at Location#1 Flow Rate#2 4000 rpm (Experiment vs CFD)

Pressure (psi)

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 89: Pressure Data Comparison at 4000 rpm (Location 1, Flow Rate 2)

 

 

Comparison at Location#2 Flow Rate#2 4000 rpm (Experiment vs CFD)

20 , - ~ ,W___-.E ~~———~-~-*‘:

— Experiment + CFD

  

 

 

 

Pressure (psi)

 

 

 

 

 

 

 

 

0 60 1 20 1 80 240 300 360
Crank Angle

 

Figure 90: Pressure Data Comparison at 4000 rpm (Location 2, Flow Rate 2)

82

 

 

 

 

Comparison at Location#3 Flow Rate#2 4000 rpm (Experiment vs CFD)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20 « 1
- — Experiment + CFD[ ,
l
18
l
"E 16 »
2
a [EBA-‘5’“ I I
2 14 - 7 7»
a.
l
12 ,
10
0 60 120 180 240 300 360

Crank Angle

 

Figure 91: Pressure Data Comparison at 4000 rpm (Location 3, Flow Rate 2)

 

 

Comparison at Location#1 Flow Rate#3 4000 rpm (Experiment vs CFD)

Pressure (psi)

 

0 60 120 1 80 240 300 360
Crank Angle

 

Figure 92: Pressure Data Comparison at 4000 rpm (Location 1, Flow Rate 3)

83

 

 

 

 

Comparison at Location#2 Flow Rate#3 4000 rpm (Experiment vs CFD)

 

1g __ _. __-_______...______

 

18 « —Experiment +CFD»

 

 

 

17

 

 

 

 

 

Pressure (psi)

 

 

 

 

 

 

 

 

 

 

 

 

0 60 120 180 240 300 360
Crank Angle

 

Figure 93: Pressure Data Comparison at 4000 rpm asocation 2, Flow Rate 3)

 

 

Comparison at Location#3 Flow Rate#3 4000 rpm (Experiment vs CFD)
19
18 -Experiment + CFD
17
16

15

Pressure (psi)

14

13

12

 

1 1
O 60 120 180 240 300 360

Crank Angle

 

Figure 94: Pressure Data Comparison at 4000 rpm (Location 3, Flow Rate 3)

84

 

 

CHAPTER 8. SUMMARY AND CONCLUSIONS

The goal of this research was to develop a better understanding of the ﬂow
characteristics of the oscillating ﬂow in the test region using both an experimental
approach and a numerical simulation.
Laser Doppler Velocimetry (LDV) was used to measure the velocity proﬁles at
three different locations and three flow rates. The test rig was operated at two engine
speeds, 2000 rpm and 4000 rpm. The piezoresistive pressure sensor and ampliﬁer were
used to measure the pressure at three different locations under the same conditions.
Computations were performed to examine the velocity proﬁles and pressure in
the oscillating flow system for different conditions. The simulation commences at BDC
and continues for a time interval corresponding of 360 degree of crank revolution.
Different engine Speeds and different ﬂow rates were investigated in the simulation.
The following conclusions can be drawn based on the results of this research.
1. The velocity proﬁles at the three different locations in the test region was measured
at three different flow rates and two different engine speeds. The single component
LDV system was proved to be a feasible method for measuring ﬂow characteristics
in an oscillating ﬂow.

2. The pressure measurements were conducted at the three different locations in the test
region at the same conditions as the test conducted for the LDV measurements. The
results plotted as a function of crank angle show how the pressure wave moves in

this test assembly.

85

 

The velocity proﬁles in the region of the oriﬁce as a function of crank angle were
investigated in the simulation. During the 2000 rpm runs, the vacuum pressure at the
oriﬁce made the flow all the way down from the top of the pipe without
recirculation. As the engine speed increased to 4000 rpm, a recirculation region
formed near the oriﬁce.

Direct comparison of velocity and pressure results was analyzed under the same
conditions. The 2000 rpm high flow rate data showed a 50% maximum difference
from the theoretical velocity data and a 30% maximum difference from the
theoretical pressure data.

The 4000 rpm high ﬂow rate data showed a 60% maximum difference from the

theoretical data and a 30% maximum difference from the theoretical pressure data.

86

 

[1]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

LIST OF REFERENCES

Peyret, R. and Taylor, T. D. [1983]. Computational Methods for Fluid Flow,
Springer-Verlag, New York.

Lee, K., Novak, M. and Schock, H., 3-D LDV Measurement of In-Cylinder Air
Flow in a 3.5L Four-Valve SI Engine, SAE Paper No. 950648, 1995

Hwang, K., Lee, K., Mueller, J ., Stuecken, T. and Schock, H., Dynamic Flow Study
in a Catalytic Converter Using Laser Doppler Velocimetry and High Speed Flow
Visualization, SAE Paper No. 950786, 1995.

Hascher, H., Kasser, J., Novak, M., Lee, K. and Schock, H., An Evaluation of
Turbulent Kinetic Energy for the In-Cylinder Flow of a Four-Valve 3.5L SI Engine
Using 3-D LDV Measurements, SAE Paper No. 970793, 1997.

Kasser, J., Hascher, H., Novak, M., Lee, K. and Schock, H., Tumble and Swirl
Quantiﬁcation within a Motored F our-Valve SI Engine Cylinder Based on 3-D LDV
Measurements, SAE Paper No. 970792, 1997.

Watanabe, H., Kurahashi, T. and Kojima, M., Flow Visualization and Measurements
of Torque Converter Stator Blades Using a Laser Sheet Lighting Method and a Laser
Doppler Velocimetry, SAE Paper No. 970680, 1997.

Hascher, H.G., SChock, H.,Avanessian, 0., Novak, J ., Comparison of Modeled and
Measured 3-D In-Cylinder Charge Motion Throughout the Displacement of a Four-
Valve 81 Engine, SAE Paper No. 2000-01-0000, 2000.

Stephens, M.A., Shih, T. I.-P., Flow and Heat Transfer in a Smooth U-Duct with and
without Rotation, JOURNAL OF PROPULSION AND POWER, Vol. 15, No.2,
March-April 1999.

Heywood, J., lntemal Combustion Engine Fundamentals, McGraw-Hill, 1988.

Shapiro, A.H., The Dynamics and Thermodynamics of Compressible Fluid Flow,
The Ronald Press Company, 1983.

87

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIII

lll|illllllllllllllllllllllllll11111111111