i
«€33

. u. c
3?

 

 

 

 

N33M4§

. g .
maﬁa".

 

 

.43 , an?!
. mluwrﬁmﬁ. .46.“

. “waif . ~11.) :
.- inn“? .

0851:...

..€..h.:}

c: ”3...... ~
.ubﬁﬁ.

 

 

 

 

.mﬁiw re,

'T'Hzisizq

67003

This is to certify that the

thesis entitled

THE EFFECT OF ACTIVE FLOW CONTROL
ON IN CYLINDER FLOW AND ENGINE PERFORMANCE

presented by

ANDREW M. FEDEWA

has been accepted towards fulﬁllment
of the requirements for

Mack—degree in Mechanical—Engineering

 

 

ajor professor

Date—May 2: 2002

 

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

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN REI‘URN 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 cJCIFIC/DateDuo.p65«p.15

 

The Effect of Active Flow Control on In Cylinder Flow and
Engine Performance

By

Andrew M. Fedewa

A THESIS

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

MASTER OF SCIENCE

Department of Mechanical Engineering

2002

ABSTRACT

THE EFFECT OF ACTIVE FLOW CONTROL ON IN CYLINDER
FLOW AND ENGINE PERFORMANCE

By

Andrew M. F edewa

Reducing the cycle-to-cycle variability present in stratiﬁed charge
engines is an important step in the process of increasing their efﬁciency. As a
result of this cycle-to-cycle variability, fuel injection systems are calibrated to
inject more fuel than necessary, to ensure the engine ﬁres on every cycle.
When the variability is lowered, the variation of work per cycle is reduced
and the lean operating limit decreases, resulting in increased fuel economy.

In this study an active ﬂow control device is used to excite the
intake ﬂow of an engine at various frequencies. The goal of this excitation is
to control the way in which vortices shed off of the intake valve, thus
lowering the cycle-to-cycle variability. This method of controlling ﬂow is
investigated through the use of three engines. First, a single cylinder
optically accessible engine is used to measure the in cylinder ﬂow ﬁeld.
Next, a single cylinder ﬁring engine with identical geometry to the optical
engine is used to measure the engine performance. Finally, a four cylinder
production engine is used to determine whether applying the flow control
device improves the work output of a production engine.

This study shows that the active ﬂow control device did help to
lower the cycle-to-cycle variability of the ﬂow ﬁeld; however, the reduction
did not translate to improved engine performance. Exciting the intake ﬂow
resulted in a reduction of velocity ms for many of the excitation frequencies
with a maximum reduction of approximately 25%. Applying the same active
ﬂow control to both the single cylinder ﬁring engine and the four cylinder

ﬁring engine resulted in no measurable increase in performance.

Acknowledgments

I would like to take this opportunity to thank the people who helped
me in this research effort. First, I would like to thank Dr. Harold Schock for
allowing me to be involved with the Engine Research Laboratory, acting as
my advisor, and providing guidance through each step of my research. Dr.
Giles Brereton and Dr. Tom Shih for serving on my advisory committee.
Tom Stuecken for his help in each phase of my research, from the design and
building of parts all the way to the ﬁnal editing of this thesis, and for his
friendship. Also, Ed Timm for his friendship and assistance in my
understanding of the method of data acquisition and processing for MTV.
My gratitude to the staff and students at the Engine Research Laboratory,
including Jan Chappell, Mark Novak, Anthony Christie, Boone Keat Chui,
Alvin Goh, Mahoomd Rahi, Andy Sasak, and Yuan Shen for their help on
several occasions. Finally, I thank my family, especially my parents Dale and
Sherry, not only for their support throughout this project, but for making my
entire collegiate experience possible, and for encouraging me every step of

the way.

iii

TABLE OF CONTENTS

List of Tables ................................................ vi

List of Figures ............................................... vii

List of Symbols and Abbreviations ................................ xi
CHAPTER 1

INTRODUCTION ............................................ 1

1 . 1 Motivation ........................................ 1

1.2 Previous Work ..................................... 3

1.3 Introduction to Molecular Tagging Velocimetry ........... 4
CHAPTER 2

Experimental Equipment ........................................ 7

2.1 Active ﬂow control device ............................ 7

2.2 Cylinder Head ...................................... 7

2.3 Firing Engine ...................................... 9

2.4 Optical Engine .................................... 11

2.5 1.9L Ford Engine .................................. 12

2.6 MTV Data Acquisition .............................. 13

2.7 Firing Engine Data Acquisition and Control ............. 15

2.7.1 Cosworth ................................... 15

2.7.2 SAKOR .................................... 15

2.7.3 RTCAM .................................... 16

2.7.4 ECM AFRecorder ............................ 16
CHAPTER 3

Experimental Procedure ....................................... 17

3.1 Optical Engine Assembly ............................ 17

3.2 Molecular Tagging Velocimetry ...................... 24

3.3 Firing Rig ........................................ 28

3.4 1.9L Ford ........................................ 32

iv

CHAPTER 4

Results and Discussion ........................................ 34
4.1 Molecular Tagging Velocimetry ...................... 34
4.2 Firing Rig ........................................ 48
4.3 1.9L Ford ........................................ 50
CHAPTER 5
Conclusions ................................................. 52
CHAPTER 6
Recommendations ............................................ 54
Appendices
Appendix A
Average Flowﬁelds for Various Crank Angles ................ 56
Appendix B
Disassembly of Firing Rig ................................ 59
Appendix C
Discussion of Indicated Mean Effective Pressure .............. 62
References .................................................. 65

List of Tables

Table l: Valvetrain speciﬁcations ................................ 9
Table 2: Single cylinder ﬁring engine speciﬁcations. ................ 10
Table 3: Single cylinder optical engine speciﬁcations. ............... 12
Table 4: Ford Engine Speciﬁcations ............................. 13

vi

List of Figures

Figure 1: Instantaneous ﬂow ﬁelds at 270 CAD for three consecutive cycles

at 1500 RPM. Cylinder head located at y = 0, cylinder walls
located at x = O and x = 7.95 ............................. 2

Figure 2: The natural flow (left) and excited ﬂow (right) from a circular jet.

................................................... 3
Figure 3: Undelayed Image of laser induced grid pattern. ............. 5
Figure 4: Delayed image of phosphorescing grid pattern. ............. 6
Figure 5: Speaker and ampliﬁer. ................................ 7
Figure 6: Cylinder head and cam carrier. .......................... 8

Figure 7: Cylinder head and ﬂat insert with centrally located spark plug hole.

................................................... 9

Figure 8: Single cylinder ﬁring engine. .......................... 10
Figure 9: Single cylinder optical test rig. ......................... 11
Figure 10: 1.9L Four cylinder Ford Engine ......................... 13
Figure 11: Xybion intensiﬁed CCD video camera and control box. ..... 14
Figure 12: Data recording and control equipment, from left: data storage

computer, volt meter (top), camera monitor, encoder box (top),

camera controller (middle), delay generator (bottom) and

computer monitor. ................................... 15

Figure 13: Lower End for Optical Engine with adaptor plates installed. . . 17

Figure 14: Reciprocating engine parts ............................ 18

vii

Figure 15:
Figure 16:

Figure 17:

Figure 18:
Figure 19:
Figure 20:
Figure 21:
Figure 22:
Figure 23:
Figure 24:

Figure 25:

Figure 26:

Figure 27:

Figure 28:

Figure 29:

Figure 30:

Schematic of counterbalancing shaft and bob weight. ....... 19
Counterbalance shaft and bob weight on balancing stand. . . . . 20
Crankshaﬁwithﬂywheelandbobweightattachedonbalancingstand.
.................................................. 21
Top half of optical engine. ............................ 22
Intake manifold (left) and exhaust manifold. .............. 23
Closed loop nitrogen/biacetyl ﬂow system. ............... 24
Original laser beam conﬁguration. ...................... 25
Modiﬁed laser beam conﬁguration ...................... 27
IMEP for the Firing Engine at 1500 RPM and 0 in. Hg. ..... 29
IMEP for the Firing Engine at 1500 RPM and 14.5 in. Hg . . . 29
Average IMEP for the Firing Engine At 1500 RPM and 14.5 in. Hg
.................................................. 30
IMEP for the Firing Engine At 1500 RPM and 14.5 in. Hg,
alternating between 1800 Hz excitation, and no force. ....... 31

IMEP for 4000 cycles of the Firing Engine at 1500 RPM and 14.5
in. Hg. ............................................ 32

IMEP for 2000 Cycles of the 1.9L Ford Engine at 1500 RPM and
14 in. Hg ........................................... 33

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and no excitation. (Day 1) ................ 36

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1500 Hz. .................. 37

viii

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:

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1600 Hz. .................. 38

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1700 Hz. .................. 39

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1800 Hz. (Day 2) ............ 40

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1900 Hz. .................. 41

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2000 Hz. .................. 42

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2100 Hz. .................. 43

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2200 Hz. .................. 44

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2300 Hz. .................. 45

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2400 Hz. .................. 46

Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2500 Hz. .................. 47
Average u and v RMS Velocities of Each Test Condition. . . . 48

Average IMEP Calculated From Three Wavelengths of Cycles For
the Firing Engine at 1500 RPM and 14 in. Hg .............. 49

Standard Deviation Calculated From Three Wavelengths of Cycles
For the Firing Engine at 1500 RPM and 14 in. Hg .......... 49

ix

Figure 44:

Figure 45:

Figure 46:

Figure 47:

Figure 48:

Figure 49:

Figure 50:

Figure 51:

Figure 52:
Figure 53:
Figure 54:

Figure 55:

Average IMEP for the 1.9L 4 Cylinder Ford Engine at 1500 RPM
and 14 in. Hg. ...................................... 50
Standard Deviation of IMEP for the 1.9L 4 Cylinder Ford Engine
at 1500 RPM and 14 in. Hg. ........................... 51
2000 cycle Average IMEP for the 1.9L Ford Engine at 1500 RPM
and 14 in. Hg. ...................................... 51
Average velocity ﬁeld for the crank angle of 90 degrees, 1500
RPM, wide open throttle. ............................. 56
Average velocity ﬁeld for the crank angle of 136 degrees, 1500
RPM, wide open throttle .............................. 56
Average velocity ﬁeld for the crank angle of 180 degrees, 1500
RPM, wide open throttle .............................. 57
Average velocity ﬁeld for the crank angle of 226 degrees, 1500
RPM, wide open throttle .............................. 57
Average velocity ﬁeld for the crank angle of 270 degrees, 1500
RPM, wide open throttle .............................. 58
Single cylinder ﬁring rig with head removed after testing. . . . 6O
Cylinder head from single cylinder ﬁring rig after testing. . . . 6O
Piston from single cylinder ﬁring rig after testing. .......... 61
Typical Pressure vs. Volume diagram for a four cycle spark
ignition engine. ..................................... 63

* Images in this thesis are presented in color.

ATDC
BBDC
BDC
BTDC
BWOD
CAD
DC

Hz
IMEP

MBWOD
MTV
PMEP
RMS
RPM
TDC

cm
in

List of Symbols and Abbreviations

Mm

After Top Dead Center

Before Bottom Dead Center
Bottom Dead Center

Before Top Dead Center

Bob Weight Offset Distance
Crank Angle Degrees

Direct Current

Mercury

Hertz

Indicated Mean Effective Pressure
Lieter

Measured Bob Weight Offset Distance
Molecular Tagging Velocimetry
Pumping Mean Effective Pressure
Root Mean Square

Revolutions Per Minute

Top Dead Center

Ultra Vilot

Centimeters

Inches

Millimeters

Wavelength

xi

CHAPTER 1

INTRODUCTION

1.1 Motivation
The existence of cycle-to-cycle variations in automotive engines is

a concern of engine designers. These variations result in a cyclic variation of
the in cylinder air and fuel ﬂows. Figure 1 gives an example of the measured
differences in the ﬂow ﬁeld for three consecutive cycles. These variations
compromise the tuned power output of engines operating at stoichiometric
conditions and place an undesirably high lean limit on fuel-lean stratiﬁed
charge engines. If this cycle-to-cycle variability could be reduced, the
variation in the air/ﬁiel ratio at the location of the spark plug could be
decreased resulting in more consistent initial conditions for the combustion
process. With such control, the variation of work per engine cycle would be
reduced and engines could be calibrated more accurately, thus lowering the

lean operating limit.

 

 

 

 

 

      

 

 

 

 

 

 

 

 

s 7 s 1 s .
I xi I I _ 2V1; M

m . . m L FUN“ m L \lilttl
o .i . o o \ I
1 . {til .6 1 s 1 in“: J
;Q\§.\\§§ . %§ 1% a
. . . \\\\§u\.\\\§l H I‘§\\\\I§ g: H
: ..\\.\\\Ioull . . a . ... .\.\\\. §\.\.\\..\\.\\ illu§ I///;II!J1\ 1
. 5 \lil\\.\\\\\l.t\\ s in.\.\.\\kitlu \ l

1 2 .\\\\\ 3 i .\ I////II
,\ . \Ilvsl\\\\\\\\\\\\\\ \ \\\\\\\\H/// \\ n
m . m cilianuvilmw§uv m u ..§\\\m __ Hi. A
v. .1. v. il .l\\\\\\\\\\ t. v. w...§::....ﬂ._xix . n
c l c ........ C: \II\\\\\\\.\\\ c \\.C\\\\\_./ .._. ..... i
. /, :tt\lt\\\\\\\ \ t ..._::___ .
.3 ,..:._:Si\\\\\\ xi .. .,.,.H,u...... 55:... . ..
.. / . __.:Q\\\\\\\\\\\\\\\\\\\\\\ ,77/ ....... z: :21 1 .

\ \\\\\\\\\\\:. ...... H _._~x\\\,\\\\\\\\\.\\\\\\\\\\\ ////////Ii\\\\\:::////.. .._
\\\\\\\\\ iii: _ _ I .2, . _:I§\\\\\\\\\\\\\\\ \\\\\ ///////I.\.\\\2: _ . 311,7 .
a 2 : .::\\\\\\\\\\ 2 777 \\§\-I,.,..... ,,,,, -
, . :\\\\\ I7I\\\\\il

; \\\\\\\ 3771 til/5755i .
xii- ..,.: Fiihhunnuuvmuwmwx L >7/inuwl2/II H
:,;//1/ \II/// 1 {\\\\\\\\\\\\ 1 ///III\\V|\ilII///II It.
— ,—. — _ ,—,,..—,...—.,,._,...—...._. _ _ P _ _ l

.. 2 a 4 o ... 4 1.. 4 .._. o 1. 4 a 4

> x h

 

 

 

 

Figure l: Instantaneous ﬂow ﬁelds at 270 CAD for three consecutive cycles

at 1500 RPM. Cylinder head located at y

and x

=0

0, cylinder walls located at x

7.95.

1.2 Previous Work
Before attempting to control the complicated ﬂow inside an engine

cylinder, an understanding of how to control the ﬂow in more simple
geometries must be reached. Several studies have been performed which
show that the natural ﬂow from round jets and two stream mixing layers can
be altered by exciting the ﬂow at the correct frequency[8][9][l4]. Results
from such a study by Koochesfahani and Stier [10] can be seen in Figure 2.
This ﬁgure illustrated the way in which exciting the ﬂow can alter the
location of vortex roll-up as well as the size of the vortices. Modifying these
properties of the vortices provides a means for controlling the amount of
turbulence and mixing downstream of the jet. When applied to an engine,
this provides a means for controlling the mixing of the fuel and air charge in

the engine cylinder.

 

Figure 2: The natural ﬂow (left) and excited ﬂow (right) from a circular jet.

Studies have also been done on the effect of exciting the intake ﬂow

on an engine model by Ambrose [1]. This study used a water analog model

of an internal combustion engine, which limited the data to the intake stroke
only, but promising results were obtained. Ambrose found that exciting the
ﬂow at certain frequencies resulted in signiﬁcant changes in the large scale
ﬂow features, including a maximum increase in circulation of 40% and a
maximum increase in kinetic energy of 30%. These values suggest that the
excitation could result in better mixing of the air/fuel charge in the cylinder.
They do not, however, reveal the effect of the forcing on the ﬂow patterns
late in the compression stroke, which is of particular interest since that is
when the onset of combustion occurs.

The effect of exciting the intake ﬂow of an engine on the
compression stroke was performed by Goh[6]. This work is of particular
interest since the measurements were made in a gas phase ﬂow as opposed to
the prior work on the water ﬂow model, allowing a view of late in the
compression stroke. This study showed that at 270 CAD the velocity rms
was reduced by 20%-30% at engine speeds of both 600 RPM and 1200 RPM
when the correct excitation frequency was used. With such encouraging
results, the next step is to determine how the reduction in velocity rms effects

combustion in a running engine.

1.3 Introduction to Molecular Tagging Velocimetry
The in cylinder ﬂow measurements for this study were performed

using Molecular Tagging Velocimetry (MTV). MTV is a non-intrusive
optical technique for measuring the ﬂow in a plane. The tracer (in this case
Biacetyl) is "tagged" by the pulse of a UV laser (7»: 308 ) by way of two sets
of intersecting lines forming a grid and an image (see Figure 3) is taken of the

ﬂuorescing biacetyl with an intensiﬁed CCD camera. This is done several

times and the images are averaged to obtain an undelayed image. To obtain
the delayed images, the process is repeated, but this time the image is
recorded after the laser pulse, but still within the phosphorescence lifetime of
the molecules (see Figure 4). The time between images is known and the
distance each line intersection has traveled can be measured giving a velocity
vector for each intersection point. This type of measurement technique has
been used successfully by many authors, including Stier and Koochesfahani
[11][12], Hill and Klewicki [7], Gendrich and Koochesfahani [3], and
Gendrich, Koochesfahani, and Nocera [4]. With this technique, many
quantities including vorticity can be determined; however, for this study the

quantity examined is the velocity ﬁeld, and the ms of the velocity.

 

Figure 3: Undelayed Image of laser induced grid pattern.

 

Figure 4: Delayed image of phosphorescing grid pattern.

CHAPTER 2

Experimental Equipment

2.1 Active flow control device
For this experiment, the active ﬂow control device used was a

Speco model SRD-100 horn driver. It is a 16 ohm speaker rated at 100 Watts
RMS, with a frequency response of 250 - 8,000 Hz. This type of speaker was
chosen over a standard audio speaker based on its robust design. The only
modification made to the speaker was the addition of a .040" hole in the
center of the diaphragm; this was done to allow the pressure behind the
diaphragm to be equal to the manifold pressure, thus allowing the speaker to
use its full range of motion even under, throttled conditions. To power the
speaker, a Haﬂer model P1000 ampliﬁer was used to amplify the sine wave

produced by a Agilent model 33120A function generator.

 

Figure 5: Speaker and ampliﬁer.

2.2 Cylinder Head
In order to study the ﬂow features as well as the engine’s

performance, both an optical engine and a ﬁring engine were constructed
using the same induction system and cylinder head design. The cylinder

head was designed at Michigan State University, and includes a removable

cam carrier and combustion chamber insert. The goal of this system is to
enable the change of combustion chamber geometry and spark plug location
without necessitating the construction of entirely new cylinder heads. The
cam carrier houses dual overhead cams, which actuate the valves in the four
valve head and the insert houses the spark plug and a thermocouple for
monitoring head temperature. The speciﬁcs of the valvetrain can be found in
Table 1. Detailed drawings of the head and cam carrier can be found in
George [5].

Since the intent of this study is to determine the effect of the active
ﬂow control device and not the combustion chamber geometry, a cylinder
head insert was designed to produce a combustion chamber with a ﬂat top

and a centrally located spark plug.

 

Figure 6: Cylinder head and cam carrier.

 

Figure 7: Cylinder head and ﬂat insert with centrally located spark plug hole.

Table 1: Valvetrain speciﬁcations

 

 

 

 

Intake Exhaust
Open 12 BTDC 38 BBDC
Close 8 ATDC ATDC
Maximum Liﬁ 8.7 mm 8.9 mm

 

 

 

 

 

2.3 Firing Engine
The base for the ﬁring engine is a model 503 AVL research engine.

The single cylinder engine block uses a removable cylinder, which enables
the compression ratio to be varied by changing the thickness of the spacer
between the block and cylinder. Coupled to the AVL ﬂywheel is a 200hp
General Electric DC electric dynamometer. The function of the
dynamometer is twofold; it ﬁrst acts as a power source for motoring and
starting the engine, and then as a brake for absorbing the power produced.

Attached to the front of the crankshaft is a 360 pulse per revolution shaft

angle encoder which is used for spark and fuel injector timing as well as data

acquisition.

 

Figure 8: Single cylinder ﬁring engine.

Table 2: Single cylinder ﬁring engine speciﬁcations.

 

 

 

 

 

 

 

 

Bore 79.5 mm

Stroke 100 mm

Displacement 496 cm3
Compression Ratio 9.0 : 1

Connecting Rod Length 198 mm

 

The robust design of this engine block is not the only advantage to
using it as the base for the ﬁring test rig. It also has an externally accessible
oil system enabling the head and cam carrier to be lubricated. Also, the AVL
block incorporates an external drive pulley which runs at cam speed,

providing a source for powering the dual overhead cams.

2.4 Optical Engine
The base for the optical engine is a Hatz model IDSIZ single

cylinder diesel engine. This engine is chosen based on several of its features.
It uses two counterbalancing shafts in conjunction with the crankshaft
counterweight to produce a smooth operation. The crankshaft can be stroked
and rebalanced allowing for different engine conﬁgurations to be used. Also,
an adjustable oil pick up system allows the engine to be motored in either the
clockwise or counterclockwise direction. Finally, the cylinder and head are

separate, removable pieces.

 

Figure 9: Single cylinder optical test rig.

Table 3: Single cylinder optical engine speciﬁcations.

 

 

 

 

 

 

 

 

Bore 79.5 mm

Stroke 100 mm

Displacement 496 cm3
Compression Ratio 9.0 : 1

Connecting Rod Length 151 mm

 

2.5 1.9L Ford Engine ‘
In addition to the two single cylinder engines, a 1.9L four cylinder

Ford engine was also used, see Figure 10. This engine is a 1995 model year
and was produced for use in the Ford Escort. Testing was done on this engine
in an attempt to determine if any immediate improvements could be made by
applying this type of active ﬂow control to a production engine. The engine
was coupled to the same dynomometer as the single cylinder ﬁring engine
and also used the same style shaﬁ angle encoder. The only modiﬁcations
made to the engine were the addition of the active ﬂow control device to the
number one cylinder runner on the intake manifold, and the insertion of a
pressure transducer into cylinder number one. The speciﬁcations for this

engine can be seen in Table 4.

12

 

Figure 10: 1.9L Four cylinder Ford Engine.

Table 4: Ford Engine Speciﬁcations

 

 

 

 

 

 

Bore 82 mm

Stroke 88 mm
Displacement 1859 cm3

Compression Ratio 9.1 : 1
Connecting Rod Length 13.62 mm

 

 

 

2.6 MTV Data Acquisition
To record the images of the MTV grids, a Xybion model ISG-3 50

intensiﬁed CCD video camera, as seen in Figure 11 was used. "MTV", a
program developed by Inﬁnite Delta Corporation was used to store the

images. Although MV-lOOO software has been used successfully in the past

for the recording of MTV grids, the software from Inﬁnite Delta Corporation
was used based on two distinct advantages. First, using "MTV" guarantees
that an image will be stored for every consecutive engine cycle, whereas the
MV-lOOO software does not. Also, using "MTV" enables the entire frame to
be stored. In contrast, the MV-lOOO software only allows a ﬁeld, or every

other line of the frame, to be stored.

 

Figure 11: Xybion intensiﬁed CCD video camera and control box.

The equipment used to store the images and control the camera can
be seen in Figure 12. An Agilent model 34401A multimeter is used to
monitor the camera gain, and act as a reference for image intensity. A JB- 1 00
timing control box is used to control the crank angle at which the images are
recorded, and a Stanford Research Systems model D6535 delay generator is
responsible for controlling the camera gate, and the time delay between laser
ﬁring, and image recording. A Linux based Pentium 11 computer is used to

store the images.

 

Figure 12: Data recording and control equipment, from left: data storage
computer, volt meter (top), camera monitor, encoder box (top), camera
controller (middle), delay generator (bottom) and computer monitor.

2.7 Firing Engine Data Acquisition and Control
In order to control the ﬁring engines and log the desired

performance data, four separate systems must be used.

2.7.1 Cosworth
There are two separate devices which make up the Cosworth

control system, the IC 5460 and IC 5580. The IC 5460 is a microprocessor-
based engine control system. Based on the input from the optical shaft
encoder, it controls the spark timing, as well as the timing and duration of the
fuel injector. The spark signal from the IC 5460 is sent to the IC 5580, a

distributorless ignition system, which provides the spark to the engine.

2.7.2 SAKOR
The SAKOR system has the capability to function as a data

acquisition system, but is used in this test for engine control and monitoring.
The primary use is for the control of the engine temperature of the single

cylinder ﬁring rig. Based on the input from a thermocouple, located in the

cylinder head insert, the SAKOR system operates a pneumatic proportional
control valve to maintain a constant cylinder head temperature. It also serves
as a monitor for the head temperature, oil temperature, horsepower and

torque for both the single cylinder ﬁring rig and the 1.9L Ford engine.

2.7.3 RTCAM
Produced by Real Time Engineering, the Real Time Combustion

Analysis Module (RTCAM) system was used for logging the cylinder
pressure data. RTCAM uses the input from the optical shaft encoder, as well
as an Optrand model Q312940-Q optical pressure transducer, located in the
cylinder head, to record the cylinder pressure value at each crank angle. This
pressure versus crank angle data can be seen in real time as the engine is
running. The soﬁware is also capable of calculating indicated mean effective
pressure, which is the desired performance parameter for the test. This is not
done in real time, but is instead calculated and stored by the software at the

completion of each test.

2.7.4 ECM AF Recorder
The air/fuel ratio was monitored using the AF Recorder 2400G

produced by ECM. This system uses a heated sensor to determine the
amount of Carbon Monoxide, Hydrogen, and Oxygen present in the exhaust.
From this information, the air/fuel ratio is calculated and reported as lambda,
which is deﬁned as the inverse of the equivalence ratio, or air/fuel ratio
divided by the stoichiometric air/fuel ratio.
(’3)
F

(g) Stoichiometric

k:

 

16

CHAPTER 3

Experimental Procedure

3.1 Optical Engine Assembly
The ﬁrst step in modifying the Hatz engine was to create a ﬂat

surface on which to mount the head and quartz cylinder supports. An adaptor
plate was constructed and attached to the front of engine, while a second
adaptor plate was welded to the ﬂywheel housing, see Figure 13. With the
two adaptor plates attached, the engine block was ﬁxtured in a mill and the
plates machined to ensure a ﬂat surface on which to mount the optical engine

components.

 

Figure 13: Lower End for Optical Engine with adaptor plates installed.

To provide optical access, a quartz cylinder was mounted above the

standard Hatz engine cylinder. This is accomplished using a piston extension

which is bolted to the top of the Hatz piston. On the top of the extension
rides a second piston with the same dimensions as the piston in the engine
which is being duplicated. This piston also houses a quartz window to
provide optical access into the cylinder from below the piston. Torlon rings
provide the sealing between piston and the quartz cylinder.

The addition of the second piston, extension, rings and quartz
window add a large amount of mass to the reciprocating assembly of the Hatz
engine, which makes rebalancing necessary. The ﬁrst step is to add up the
entire reciprocating mass of the rig. This includes both pistons, all six rings,
quartz piston insert, wrist pin, snap rings, piston extension, screws and light

end of the connecting rod See Figure 14.

 

Figure 14: Reciprocating engine parts

With the reciprocating mass known, the next step is to balance the

counterbalancing shafts. To do so, a specially made adjustable bob weight is

 

used as seen in Figure 15. To determine the location of the bob weight, the

following calculation is used.

H

 

 

 

 

 

 

 

 

.— _ _ d — —_ -— — — 1P ‘ — — ‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

’ ‘ t
Figure 15: Schematic of counterbalancing shaft and bob weight.
1. Bob weight (g mm) = 1/4 reciprocating mass (g) * crank radius (mm)
2. Bob weight offset distance (BWOD) (mm) = Bob weight (g mm) /
43.48g
3. Measured bob weight offset distance (MBWOD) = BWOD (mm)
+19.97
With the bob weight adjusted to the proper location, it is attached to
the counter balancing shafts one at a time, and placed in the balancing rig as

shown in Figure 16. Mallory metal, which is approximately twice the density

of the original counterweight material is then added to the counterweight to

19

balance out the weight of the bobweight until the assembly can be rotated in

the rig without having a tendency to fall to one side.

 

Figure 16: Counterbalance shaft and bob weight on balancing stand.

Next, the crankshaft had to be balanced. To do so, a second bob
weight of adjustable mass was used. To determine the proper amount of
mass to be added, the following procedure is used.

4. Rotating mass = mass of the large end of the connecting rod

5 Reciprocating mass = same as above

6. Crank shaft bob weight = 1/2 reciprocating mass + rotating mass
7 Mass to be added = crank shaft bob weight - 520 g

The mass is then added to the bob weight and it is attached to the
rod journal of the crankshaft. Just as with the counterbalancing shafts, the
crankshaft is placed in the balancing rig see Figure 17, and mallory metal is

added until it balances out the mass of the bobweight.

20

‘3?

mg .2 a“
“rte-51‘ .- 1'
.. . if“ . ,

 

it

Figure 17: Crank shaft with ﬂywheel and bobweight attached on balancing
stand.

The optical engine is topped off with the custom built aluminum
cylinder head. Two steel supports were fabricated to hold the head in place
atop the quartz cylinder. A belt drive system is used to drive the dual over
head cams. This is accomplished by running a belt to a jack shaft with a
pulley twice the size of the crankshaft pulley to make the reduction from
engine speed to cam speed. Then, a second belt is used to drive the camshafts
at a one to one ratio with the jack shaft. Not only does it serve the purpose of
reducing the rotational speed, but the jack shaﬁ also routs the belts to the side

of the engine, allowing for better optical access to the cylinder.

21

 

Figure 18: Top half of optical engine.

Custom intake and exhaust manifolds were built for use with the
MSU cylinder head, see Figure 19. Both manifolds were constructed using
one inch steel tubing, and both were designed to angle upward to provide
better optical access of the quartz cylinder. The intake manifold did have to
incorporate some special additions. First, since the same manifold would be
used on the ﬁring engine, fuel injectors needed to be installed to provide fuel.
Also, a means for attaching the active ﬂow control device had to be
incorporated. This was accomplished by creating a "Y" in the intake and

attaching a threaded adaptor for the speaker.

22

 

Figure 19: Intake manifold (left) and exhaust manifold.

Since MTV was to be performed on the optical engine, a special
closed loop ﬂow system had to be built. (See Figure 20). Since a mixture of
Nitrogen and Biacetyl must be supplied to the engine, a system was designed
and constructed to rout the exhaust back into the intake of the engine. If the
exhaust were to be fed directly into the intake, however, it would cause a
pressure pulse at the intake each time the exhaust valves were opened. This
is avoided through the use of two plenums. A large rectangular plenum is
used on the intake side of the engine to provide a mass of nitrogen at a steady
pressure each time the intake valves are opened. This plenum also serves as
the mixing chamber for the Biacetyl and nitrogen. By ﬁlling the bottom with
liquid biacetyl, it provides a large surface area over which the biacetyl can
evaporate into the nitrogen. Once the mixture is expelled from the cylinder,
it travels to the exhaust plenum. This is a cylindrical plenum with two
internal bafﬂes which serve to disperse the mixture throughout the plenum

and reduce the exhaust pulsations traveling through the plenum. After

23

leaving the exhaust plenum, the mixture travels through a throttle to provide
the desired manifold vacuum. From the throttle, the mixture travels back into

the intake plenum to start the process again.

 

 

 

 

 

 

(r ‘— ‘ 5.72:3: ' a
I
\

 

Intake Plenum

éggtﬁ JQIOOIQJJ

Throttle ____
quuld Blacetyl [I I

 

 

 

 

 

 

 

 

 

 

 

 

Figure 20: Closed loop nitrogen/biacetyl ﬂow system.

3.2 Molecular Tagging Velocimetry
Before ﬁring the engine, work was ﬁrst done to determine the in

cylinder ﬂow features using the optical engine. It was decided that all
measurements would be taken at 1500 RPM and wide open throttle. Based
on previous work on a similar conﬁguration by Goh [6], it was predicted that
active ﬂow control would produce the greatest reduction in cycle to cycle
variability between 1800 Hz and 2200 Hz. Therefore, measurements were to
be taken in 100 Hz increments from 1500 Hz to 2500 Hz. Also based on
previous work at MSU’s Engine Research Laboratory, 200 undelayed images
were averaged to provide a reference grid and 500 delayed images were

needed to provide a repeatable average ﬂow ﬁeld.

24

Experiments began with a laser setup, as shown in Figure 21. Such
a setup was used for several previous experiments at MSU and has
traditionally worked well; however, this setup would prove troublesome for
the current study. The biggest problem was with the Biacetyl ﬁlm produced
by this engine. Although the intake manifold designed for this engine served
the purpose of incorporating a speaker, it had the unfortunate side effect of
causing the evaporated biacetyl to condense. This liquid biacetyl would
eventually make its way into the cylinder, and form a ﬁlm on the top of the
piston. The droplets would act as tiny lenses and cause the incoming laser
light to diverge after passing through them. By the time the light had traveled
to the top of the cylinder, it was no longer in the form of precise grid lines,
but instead, appeared as a haze. With no distinct lines, no intersections were

available for meaningful analysis.

C)

    

Figure 21: Original laser beam conﬁguration.

25

 

Aside from the downfall of unclear images, there were also other
disadvantages associated with the laser beam setup. Bringing lines through
the piston meant that only the center of the cylinder could be studied since the
piston itself would block beams near the cylinder walls. Also, bringing in
beams horizontally makes measurement at the top of the cylinder near the
spark plug difﬁcult since parts of the engine outside the cylinder block these
beams.

In order to make the desired measurements, a different laser grid
had to be designed. The modiﬁed laser beam setup can be seen in Figure 22.
Beams enter the cylinder from each side aiming upward on a 30 degree angle.
This results in an intersection angle of 120 degrees which still allows the
intersection point to be found accurately [3]. Such a setup has multiple
advantages; it allows measurements to be taken all the way to the sides as
well as to the top of the cylinder. Most importantly, the beam intersections
are still distinguishable when the piston becomes coated with biacetyl. Of
course, this design is not without ﬂaws. Since the beams are angled upward,
the region directly above the piston cannot be measured. Still, the beneﬁts

far outweigh the disadvantage.

26

V/
I
I
t
t
\
A

-9V‘5‘r-
v.9”o:
vooooo.
.3.o.:.:.,.
'.’o 0‘

///}

Figure 22: Modiﬁed laser beam conﬁguration

Ideally, all of the testing would take place in one session; however,
the combination of torlon piston rings, and biacetyl cause the cylinder walls
to become too cloudy for accurate data analysis before all 12 conditions
could be tested. This means the rig would be run until the images become too
cloudy, then the rig would be disassembled and cleaned. In the ﬁrst run ﬁve
conditions were completed, which included a no force, and 1500Hz through
1800 Hz. After cleaning the rig, the 1800 Hz through 2500 Hz conditions as
well as another no force condition were run. Although not all of the tests
were run consecutively, the no force condition, and 1800 Hz condition were
run in each test providing a way to ensure the repeatability of the experiment.

Once all of the data had been collected, it was then processed using

software developed at MSU’s Turbulent Mixing and Unsteady Aerodynamics

 

Laboratory. The software serves the purpose of ﬁnding the grid intersections
in the undelayed images, and correlating them with the appropriate
intersections in the delayed images. The details of the correlation process
can be found in Gendrich and Koochesfahani [3] and a discussion of the

accuracy of the plotting process can be found in Cohn and Koochesfahani

[2].

3.3 Firing Rig
With the ﬂow ﬁeld data collected, it was time to test the effect of

active ﬂow control on the ﬁring engine. The intake plenum, manifold, and
speaker from the MTV rig were ﬁtted to the ﬁring rig to ensure the in
cylinder ﬂow ﬁelds would be the same. The initial test conditions were
chosen as 1500 RPM, wide open throttle, with forcing from 1500 Hz to 2500
Hz. The spark timing and injector timing were set to 18 degrees before TDC
and 100 degrees before TDC respectively. The injector pulse width was
manually adjusted to maintain lambda approximately equal to 1.

Indicated Mean Effective Pressure (IMEP) was the measurement of
performance for the running engines in this study, the deﬁnition of which can
be found in Appendix C. The IMEP for each of 400 consecutive engine
cycles was recorded and averaged to determine the average work output per
cycle. However, due to the engine’s inability to run smoothly at this
condition no meaningﬁil results could be obtained. Figure 23 shows the large
ﬂuctuations in IMEP between cycles at the 1500 RPM, wide open throttle
condition. Further investigation showed that throttling the engine to 14.5"
Hg produced a much smoother running engine, as seen in Figure 24.

Therefore, subsequent testing was performed at 14.5" Hg.

28

180 ~

160 *

 

 

i
l ‘.
; : V ,’_.[ .
14o -h,.n.
. . I‘ I .
. t
' I II

120

100

IMEP

80

60

40

20 a

100

80

60

IMEP

40

20 —

 

 

 

 

 

 

 

 

 

 

 

0 50 100 150 200 250

Cycle

300 350 400

Figure 23: IMEP for the Firing Engine at 1500 RPM and 0 in. Hg.

 

'f’v" In“!

A JV“! ¢M¢MA c.l\(4

ﬁM‘M-«WA'AAVW N WWMWA WV“ uf‘t‘,~v [IV VNN/v

 

”‘7”VVVVV“ ‘W‘ MN N/V’V-H"

 

 

 

 

 

 

 

 

 

 

 

200 250

Cycle

100 150

300 350 400

Figure 24: IMEP for the Firing Engine at 1500 RPM and 14.5 in. Hg

Before completing the entire set of test conditions, it was important

to ﬁrst determine whether meaningful data could be gathered at the throttled

29

condition. To test this, a simple test sequence was used, 1000 cycles of IMEP
would be recorded with 1800 Hz excitation then 1000 cycles with no
excitation force. This sequence was then repeated 2 more times, so as to
determine the repeatability of the results. Figure 25 shows that although the
runs with active ﬂow control appear to produce a higher IMEP, they are not
very consistent. The unaveraged data- shown in Figure 26 reveals an
interesting trend; it would appear that the data resembles a broken sine wave.

77.5 —~- ., 7 , 7 77*»A77 innﬂ rwriwv-ﬁiv ~ ,

 

 

 

 

Average IMEP

 

 

 

i‘ a g a s :2
§ 3 § 3. g g.
2 2 2

Figure 25: Average IMEP for the Firing Engine At 1500 RPM and 14.5 in.
Hg

30

85 7 _ ,,wf,i,__ ._V — , ‘ .777 , w . ._ ——— W , , H A” ‘ iiiiiﬁ—ﬁerﬁ-mn—a

 

 

 

65 e

 

 

 

60

 

0 1000 2000 3000 4000 5000 6000 7000 8000
Cycle

Figure 26: IMEP for the Firing Engine At 1500 RPM and 14.5 in. Hg,
alternating between 1800 Hz excitation, and no force.

To investigate this apparent sine wave, a test run was made with no
active ﬂow control. The IMEP of 4000 consecutive cycles was recorded and
plotted in Figure 27. It can be seen that the data does in fact take the form of
a sine wave. Closer inspection reveals that the wavelength of the sine wave
is approximately 750 cycles, or 1 minute since the engine is running at 1500
RPM. Since the dynamometer controller is an electronic device operating at
60 Hz it is assumed that the sine wave is somehow linked to the way that the
controller is making adjustments to the dynamometer. This discovery
accounted for the inconsistency in the 1000 cycle averages since the sampled
cycles may contain two peaks and one valley, or two valleys and one peak,

depending on when the data logging is initiated.

31

 

 

 

IMEP

 

 

 

so ,h______n___m____*A4_H*Ak,~___WW

 

 

 

55 J, WM,“ A —+f1*v——————

 

 

50

 

f T

0 500 1000 1500 2000 2500 3000 3500 4000 4500
Cych

Figure 27: IMEP for 4000 cycles of the Firing Engine at 1500 RPM and 14.5
in. Hg.

Data was then collected for 3000 cycles, and from that data, three
wavelengths of data were averaged. This was done for both the non forced
and 1800 Hz excited conditions three times and these results were compared.

Upon completion of the testing, the engine was disassembled and

inspected. The ﬁndings of this inspection can be found in Appendix B.

3.4 1.9L Ford
Although the single cylinder engine makes for a good way to

compare ﬂow ﬁeld information and corresponding engine performance, it
may not show how active ﬂow control would perform in an automotive
application. To investigate this, the active ﬂow control device was attached
to a 1.9L 4 cylinder Ford engine. As with the single cylinder ﬁring rig, the

parameter of interest was IMEP.

32

Step one would be to determine whether the sine wave present in
the data from the single cylinder engine would also be present in the data
from the four cylinder. IMEP for 2000 consecutive cycles without forcing
was recorded and plotted in Figure 28. It can be seen from these plots that
the sine wave was no longer present. Although the exact reason is unclear, it
is possibly the result of the smoother operation of a four cylinder as opposed
to a single cylinder engine.

With the sine wave no longer an issue, the testing proceded. IMEP
for 500 consecutive cycles were recorded for the no force condition as well as
the 1500 Hz to 2500 Hz excitation conditions at 100 Hz intervals. The results

for these tests can be found in chapter 4.

 

100

90

 

80

 

 

 

 

7o _.._ ___ ___..._-__._ N- ___.___

6 o w———-——

 

IMEP

so 4- — “___-

 

 

 

40 ~» 7 ----~—-— --—-- «

 

 

30 4» ——

 

 

2 0 ~» ————————— - — -- ~ -——

 

 

 

1o -____ -- W—- W

 

 

 

 

0

 

j

0 200 400 600 800 1000 1200 1400 1600 1800 2000
Cycle

Figure 28: IMEP for 2000 Cycles of the 1.9L Ford Engine at 1500 RPM and
14 in. Hg.

33

CHAPTER 4

Results and Discussion

4.1 Molecular Tagging Velocimetry
The ﬂow ﬁelds calculated from the measurements made in the

optical engine for the crank angle of 270 degrees can be seen in Figure 29
through Figure 40. In these plots, The axes are in centimeters, cylinder head
is located at Y = 0 and the cylinder walls are located at X = 0 and X = 7.95.
The intake valves are located on the left side and the exhaust valves on the
right. The vectors represent the average velocity and the color represents the
rms value of the u and v components of velocity. The unforced ﬂow ﬁelds
for the intake and compression strokes at various crank angles between 90
and 270 degrees are shown in Appendix A, Figure 47 to Figure 51.

The ﬁrst parameter of interest is the velocity ﬁeld. The most
notable detail is the strangely out of place vectors found on the lower
diagonal lines of the sample area in several of the conditions, namely no
force, 1700 Hz, 2000 Hz, 2200 Hz, 2300 Hz and 2400 Hz. These vectors
seem to imply high velocities in directions contradictory to the vectors
around them. However, these vectors are more likely a result of errors
ﬁnding the intersections along the lowest diagonal lines due to the piston
partially blocking the beam. Therefore, these vectors will be ignored
throughout the rest of the discussion.

With these vectors discarded, one can see that the rest of the ﬂow
ﬁeld is almost identical for each of the conditions. This implies that the

active ﬂow control applied to the engine had essentially no effect on the

34

average in cylinder ﬂow ﬁeld. However, it does not mean that the active
ﬂow control had no effect on the in cylinder ﬂow.

Looking at the u and v RMS velocities reveals a different trend.
One can note that in all of the cases tested with the exception of the 1700 Hz,
2200 Hz, and 2300 Hz cases, all of the excitation frequencies produce a
reduction in RMS velocities, as compared to the non forced condition. In
order to see this more clearly, the RMS velocities from each point were
averaged over the entire plane and plotted in Figure 41, which clearly shows
the difference. The largest improvement can be seen at 1900 Hz with a
reduction of approximately 25 percent from the non forced condition. This
implies that the air/fuel ratio and velocities will be more constant in the
region surrounding the spark plug, providing a more consistent flame
initialization. Also of interest is the fact that both u and v RMS are reduced

by approximately the same amount for any given condition.

35

 

 

 

 

 

 

 

   

 

 

 

Average Velocitylu' Field

.
t
-1:
-2:—
-3:—
>. Z
4r
-5:—
6:-

I-JllllllllllllLlllllllllllllllllllllllll'
'70

X
Average Velocitylv' Field

-1:-
-2:—
-3:—
>. I
4Z-
-5:—
62-

-7-1llI'IlllllllIlllllllllLllLLLLJllJLlJlll

( 1 2 3 4 5 6 7 8

X

 

 

 

Figure 29: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and no excitation. (Day 1)

36

 

   

 

 

 

 

 

 

 

 

 

 

Average Velocitylu' Field
.-1:—
-2:—
-3:—
>. Z
4.-
-5:—
6:-
-lllllllllllllllllllllllllllllll|l|l|ll
'70 4
X
Average Velocitylv' Field
#-
-1:—
-2:—
-3:—
>. I
4:-
-5:—
e:-
-7-lllllllll|lllIllIIIIIlllllllllllllllIll]
1 2 3 4 - 5 6 7 8
X

 

 

 

Figure 30: Average Velocity/u’Iv’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1500 Hz.

37

 

 

 

 

 

 

 

 

 

 

 

 

Average Velocitylu' Field
-1:—
.2}
.
-3:—
>. Z
4r
-55.
6:-
_7-111111111LLIlllllllllllllllllllllllllll
ll 1 2 3 4 5 7
X
Average Velocity/v' Field
-1:—
-2:—
-3:—
>. I
42-
-5;
.6;
I 130
_ 100
_7-llIllllllllllllJlJlllllllJlIIlllllllllll
( 1 2 3 4 5 6 7 8
X

 

 

 

Figure 31: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1600 Hz.

38

 

 

 

 

 

 

   

 

 

 

 

Average Velocitylu' Field
-1 :.
-2 E.
-3 :—
>. Z
4 .-
-5 _—
6i-
_7 I I I I l I I I l I I l I I I I I I l I l I I l I I I I l I I I I l I I I L
1) 1 2 3 4 5
X
Average Velocitylv' Field
-1 :—
-2 :- ii
- m
: Ill
-3 7 ?
> i i? 5"
.4 I— j!
r :1.
: gs:-
-5 .- ii
-6 -_- "
-7 —I I l I I I I I I l I l I l I I I I l l I I I l i I I I I I l I I I I I I I I J
0 1 2 3 4 5 6 7 8
X

 

 

 

Figure 32: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1700 Hz.

39

 

Average Velocitylu' Field

-1

-2

-3
>

-4

-5

.3

 

_7IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
0 1 2 3 4 5 6 7

 

Average Velocity/V Field

_7 I I I l I I I I I I I I I I I I I I I I I I I I I I I I I I I I l I I I I I I I
1 1 2 3 4 5 6 7 8
X

 

Illl'lllllllll'lllllllllllIIV'IIII

 

 

 

 

 

 

Figure 33: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1800 Hz. (Day 2)

4o

 

 

 

 

 

 

 

         

 

 

 

 

Average Velocitylu' Field
-1 :—
-2 :—
-3 :—
>. Z
4 _-
-5 :—
6 E
— J I 11 I l I I I l I I l I 1 I I I I I l I I I l l I I I l I I I I I I l I
'70
it
Average Velocitylv' Field
: 10 mls
- L v'
1 - ‘ 700
I 670
_ _ 640
-2 " 610
Z a 550
— I 520
'3 _- I 490
>' - H' 430
L 400
‘4 _ E; 370
.. 8"- 340
I 3.. 310
-5 - 5 280
3 1 4 250
Z I'.‘
-6 I-
-7 I‘I I I I I I I I I I I I I I I I I I I I l l 1 ]_LI I l 1 I I I I l I I I l l I
1| 1 2 3 4 5 6 7 8
x

 

 

 

Figure 34: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 1900 Hz.

41

 

Average Velocitylu' Field

-1

-2

-3
h

.4

-5

_7 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
0 1 2 3 4 5 6 7

J
O

§§q la
‘0

  

“E
a
O

§§§§§§§§

§§§§§§§

 

 

téé

X

 

Average Velocity/V Field

 

 

 

 

 

 

A

y
a. A a a. ;
lIIIllIIfIIrTIIIII'IITIIIrjIﬁIIIr
N
(D
a)
J "'-.< . ‘ .- - w. d
‘c' "4', 1 °
N“U&§0I thaw
sa§s§8§se§§a§as‘~ 12
m

' 160

130

I 100
_7 LIJIJIIIIIIIIILIIIIIIIIIIIIIIIIIIIIIIII
7 8

4 5
X

 

 

 

Figure 35: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2000 Hz.

42

 

   

 

 

 

 

 

 

  

 

 

 

 

Average Velocity/u' Field
: 10 III/S
L ‘ “'
'1 r- 700
1’ 670
C , 640
-2 i- . 610
i' 580
I ,7 550
- 520
-3 _- '_ ‘ 490
_ - 460
> - 430
; 400
'4 ~ 370
' 340
Z ‘ 310
-5 - zoo
‘ , 250
I 220
.. 190
-6 - 160
I 130
_ 100
_I I I I I I I I I I l I I J I I I I I I J I I l I I I L I I I I l l I I I I I
'70
x
Average Velocity/V Field
-1 L
C
-2 :—
-3 E.
> i
4 :-
-5 E
«s:-
C
_7 r1 1 I I I I I I I I I I I I I I I l I I I l I I I I I I I I I I I I I I I I I I
u 1 2 3 4 5 6 7 8
X

 

 

 

Figure 36: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2100 Hz.

43

 

Average Velocitylu' Field

5

§§e|1

2S

 

ITITIIV[TIT—rTI'I‘l’TTIIITI‘I’W—TIIITIII

 

§§§§§§§§§§§§§§§§§

 

 

 

 

_7 IlIIlIIIIlllIIIIIIIIIIIIIIIIIIILIIIIII
1 2 3 4 5 6 7
X

A

 

Average Velocitylv' Field

 

I
(A)
[III '7'] VIII l'rTllllllll
I l I l I l

 

 

 

llIIIIIllIllIIIIIJIIJIIIIIIIIIIIIIIIIIII

4 5 6 7 8
X

 

 

A
.III]

 

 

 

Figure 37: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2200 Hz.

44

 

    
 

 

 

 

 

 

 

 

 

 

 

 

Average Velocitylu' Field
: 10 mls
-1 :—
:
'2 L- 3‘;
- III
: In
-3 L. I
. I.
>. C 2‘
-4 ' 1'-
: PE
_ all
1.!
-5 r I:
-6 :- 160
I 130
_ 100
-7 _J I l I I I l l I I I I I I I l I I I I I I I I I I I I I I I I I I I I L J L
(I 1 2 3 4 5 6 7
X
Average Velocitylv' Field
: 10 mls
- '_ V,
1 _ 700
I , 670
- 41‘. 640
-2 _ 610
: ' 580
- 550
_ 520
'3 1’ _ 490
> - ‘ 430
’ 400
'4 - 370
r 340
7 310
-5 - 280
. 250
C 220
_ ~. 190
-6 _- 160
L 130
L 100
_7 PI l I I I I I l l I l l l I I I I I l I I I L I I I I I LI I I I I I l I I I I
(l 4 5 6 7 8
X

 

 

 

Figure 38: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2300 Hz.

45

 

    

 

 

 

 

 

   

 

 

 

 

 

Average Velocitylu' Field
: 10 mls
- I. u.
1 - 700
: 670
. :4 640
-2 - 610
: 580
_ 550
520
‘3 [- ‘ 490
>0 - j 430
_ t m
‘4 — ‘ 370
. 3‘0
I 310
-5 - 250
_ a 250
I 220
- 190
-6 - ‘ 160
I 130
~ 100
_7 ’- J_l IJJ—L I I I I I I I I l I l I I I I I I I l I I I I I I I I I I I l I
U 1 2 3 4 5 6 7
X
Average Velocitylv' Field
-1 E.
-2 :—
-3 :—
>. I
4 .
i-
-5 L-
-s :-
_7 I-I I I I l l I I I I I I I I l I I l I I I I IALI I I I I I l l l I I I I I I
1 2 3 4 5 6 7 8
X

 

 

 

Figure 39: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2400 Hz.

46

 

Average Velocitylu' Field

-1

-2

-3
>.

.4

-5

 

 

 

 

 

 

 

-6
IllIIIIIIIllIIIIIIIIIIIIIIIJIIIIIIIIIJI
'7' 2 3 4 6
X
Average Velocitylv' Field
-1:—
-25.
:
-3u—
>. I
4:-
-5_F
-6;
Z 130
_ 100
_7y-IIIIIIIIIIIIIIILIJIILIIIIIIIIIIIIIIIIIll
n 1 2 3 4 5 6 7 8

 

 

 

X

 

 

 

Figure 40: Average Velocity/u’/v’ Fields for the crank angle of 270 at 1500
RPM, 0 in. Hg and excited at 2500 Hz.

47

Average RMS Velocity

 

 

 

1900 Hz
2000 Hz
2100 Hz
2200 Hz
2300 Hz
2400 Hz
2500 Hz

noforoeday1
nofomeday 2
1500Hz
1600Hz
1700 Hz
1800 Hz day 1
1800 Hz day 2

Figure 41: Average 11 and v RMS Velocities of Each Test Condition.

4.2 Firing Rig
The ﬁnal results of testing on the single cylinder ﬁring engine can

be seen in Figure 42. Since measurements taken at wide open throttle
showed little promise, it was decided to simply concentrate on the non forced
condition and 1800 Hz, the condition predicted to be the best. It appears that
the active ﬂow control produces very little difference in the average IMEP.
In fact, there is more difference between the two non forced runs than
between the between no force and 1800 Hz. Furthermore, no evidence exists
in the standard deviation plot in Figure 43, to suggest that the cycle-to—cycle

variability of IMEP was improved.

48

 

 

 

Aver-go HEP
N
o

 

 

 

notoreeﬂ
1800Hz£1
Noiorcell
1800Hzﬁ2

Figure 42: Average IMEP Calculated From Three Wavelengths of Cycles For
the Firing Engine at 1500 RPM and 14 in. Hg.

2.5 , - - -———-M ~~A~~m 77—77777777777

 

Q g ‘a‘é g
§ § 5 s

Figure 43: Standard Deviation Calculated From Three Wavelengths of Cycles
For the Firing Engine at 1500 RPM and 14 in. Hg

49

4.3 1.9L Ford
The results of the experiments on the 1.9L 4 cylinder Ford engine

can be seen in Figure 44 and Figure 45. It is evident that the active ﬂow
control has little effect on the IMEP of this engine. In fact, the difference
between the no force condition and 1800 Hz, the best case, is less than one
percent. Likewise, the standard deviation, although relatively small, is
essentially the same for each of the measured conditions.

To further investigate the small difference between the non forced
case and the case of 1800 Hz excitation, a second test was run. This time,
2000 cycles worth of IMEP data would be stored for each set alternating
between no force and the best case, 1800 Hz. The results of this run can be
seen in Figure 46. From this plot we again see very little difference between
conditions. In fact, the difference between two data sets of the same
condition is just as large as the difference between conditions, implying the
differences are due simply to experimental error.

90

as 77 7777 777 7 77 7 7 *7

IIIIHII

ﬁ‘

75 *7

 

   

notorce #1
1500 Hz
1600 Hz
1700 Hz
1800 Hz
1900 Hz
2000 Hz
2100 Hz
2200 Hz
2300 Hz
2400 Hz
2500 Hz

no force #2

Figure 44: Average IMEP for the 1.9L 4 Cylinder Ford Engine at 1500 RPM
and 14 in. Hg.

50

2 W”. . ., ,._

 

1.9

 

1.8

 

1.7

1.6

 

 

 

W Mm

1.2 *

1.1< V.

1.5 *-
1.‘ 4

1.3 -

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

nature.”
1500Hz‘
600
1700Hz

Figure 45: Standard Deviation of IMEP for the 1.9L 4 Cylinder Ford Engine

90 _.

 

Average IMEP
0
O
I

at 1500 RPM and 14 in. Hg.

 

 

g; g i
2 g 2 §

Figure 46: 2000 cycle Average IMEP for the 1.9L Ford Engine at 1500 RPM

and 14 in. Hg.

51

CHAPTER 5
Conclusions

Two nearly identical engines have been designed and built to study

the effects of active ﬂow control on the in cylinder ﬂuid motion, and in turn,

the effect of the ﬂuid motion on the performance of a running engine. Also, a

third engine, a production 1.9L four cylinder Ford, was used to determine

what effects the active ﬂow control would have on an actual automotive

engine. The main conclusions from the study are as follows:

1.

The use of active ﬂow control does affect the RMS velocities.
Reductions occur at many of the excitation frequencies, the maximum
being approximately 25 percent at 1900 Hz.

The average ﬂow ﬁeld in the tumble plane studied are not affected by
the use of active ﬂow control.

The combustion chamber geometry does not allow the generic ﬁring
engine to run properly at 1500 RPM and wide open throttle, making
attempts to improve IMEP through active flow control useless.
Active ﬂow control applied to the generic ﬁring engine at 1500 RPM
and 14.5 in. Hg showed no improvement in the average or standard
deviation of IMEP.

When applied to the 1.9L 4 cylinder Ford engine, active ﬂow control
showed no evidence of improving performance.

Although the RMS of the velocities can be reduced through the use

of active ﬂow control, no positive effects are evident in the engine’s

performance. This suggest that engine performance cannot be signiﬁcantly

improved simply by lowering the rrns of velocity. Therefore, there is no

52

reason to pursue further experiments with this type of active ﬂow control

device.

53

CHAPTER 6
Recommendations

Before further investigation of this type of active ﬂow control, a more
robust and powerful excitation device should be developed, one
possibility would be to use a linear motor to drive a diaphragm in the
intake manifold.

Investigate the effect of introducing the extication through the valve
itself similar to the water analog model study by Ambrose[l] rather than
through the use of an upstream diaphram as in this study.

Future attempts to compare ﬂow ﬁeld data with running engine data
should be done with a combustion chamber design more suitable for 3 SI

gasoline engine than the ﬂat head design used in this study.

54

Appendices

55

Appendix A

Average Flowﬁelds for Various Crank Angles

 

 

 

 

 

 

QODegrees
50mls
w 44:::::::/ _*
‘1':§€/:/,// 5:':::::::%
:__:; 1;: ::::‘.I:j
ﬁi?’ s‘kiil‘3' :5
'2." ///7 *\\\\:'2: . .1
//// ~\\\§...2 .,
/// \\\\~ - ,,
¢y \§\\‘555 C:
3— \§§,.. ..
\w“ ‘
\\|'
555 I
>. x': ;
-4- 5: I
::'.;\
i. ‘5‘
5_ w; l
-6-
L 11541; I 1 ll
7 2 4 6 8
X

 

 

Figure 47: Average velocity ﬁeld for the crank angle of 90 degrees, 1500

 

 

 

 

 

 

 

 

 

 

 

 

 

RPM, wide open throttle.
136Degrees
T‘fx .lﬁféééiiéééé “mls
57/7; :EEKILEEEEE‘EE
'2?/// tt~r§§wjzi :5
: / “Mir-.212:
.3,— / \ng. :
- \§\\;'.:
. \\‘ .z
>. \i‘ I:
4' l5;::
5— 55552‘?
H\\
‘6:-
W7éiié;
X
Figure 48: Average velocity ﬁeld for the crank angle of 136 degrees, 1500
RPM, wide open throttle

56

 

 

180 Degrees

_ \\ x xx \ \\ \\\\\\\\\\\tt(l/// \
B\\\\ \ \ \ \ \\\\\\\\\\\\\\.\I\I/ I
( \\\\| u - - t \ \\\\\\\\\\\\\\\|/ /

_\\\\\\ v . . . , s \\\\\\\\\\\\\\\\I

_\\\\\\\.._._e\\\\\\\\\\\\\\\\
~\\\\\\\._.._........._.--\\
_\\\\\\\_::::,,,zzzzz:::

..\ x ~ ~ _ . , (Ix/Illlllll/l/Ill/Il

. . , , , I 1 I /l/l/////////////////

 

. l 4 a
I u 1 Illnllll/l//////////////////

. I Jill/ll/I///// // ////////l
. . e . . , l//////////x / z/ / ////////
xx x : /////////////////////////
“\ x \ _ / ///////////////////////
D x . _ z / /////////////////////

_ \ \ :I/////////.Iulllllllll

.-—....—....—.7.......—_..e——.rL

1 2 3 4 5

>

6

 

 

 

Figure 49: Average velocity ﬁeld for the crank angle of 180 degrees, 1500

RPM, wide open throttle

 

 

226 Degrees

   

s
m
5
2
:7\\\\\\\\:
.\\\\\\\\\\\§§‘l

 

~\\\\\\II|I§
\\\\\\\\\\\\\\§i
\\\\\\\\\\\\\\§§
~\\\\\\\\\\\\\\§§
. \ \ x x \ \\\\\\\\\\\\«\§\\
..... . t \ \\\\\\\\\\\\\\\\u\\\\\\
x. . _ _ _ _ . \\\\\\\\\\\\\\\\\\\\\\\
illzr—___\\\\\\\\\\\\\\\\\\\\\\
115::__:::::::\\\\\\\
lI//////———_____~.~.\_~\\\\\\\\

._—._..arr/r/zzrr/rz/IIIIII/rrz

___ ....... arr/z/z//////////////
~\.\ ..(///,,///zr///////////
.\\\\\ .«IIII/V,’/’////////////
\\\\\\.I/l/Iza../r/,zrz/////I/

 

 

,\\\\\\_z////,, ,,,,,, arz/z/zrl

\\\\\\..,/:......,,,,,:,/z

\.\\\. ...:.,.,...,.,,,////

.._.._.—._.__....—,,»»__.r5_pr5r
4 o. no 4 6 6 7

 

 

Figure 50: Average velocity ﬁeld for the crank angle of 226 degrees, 1500

RPM, wide open throttle

57

 

 

270 Degrees

 

 

 

 

 

 

_._~‘~\\\\\\\\\VVVVV€dlllll \
~\~._~\\\\\\\\\\\\\V\V“"’
___.__.\\\\\\\\\\\\\\\\\ll

 

4 X

 

_..__._‘\\‘\\\\\\\\\\\\\l
.~._.__.____.~\\\\\ \\\\\
v 2

.\‘~_.._.r.__.\eextxxi
.~e\\_,_./,r_.: \\\\\\
.‘\\\._.,... .........
.~_‘.__..: .........
,e\.___: __________
,\:.::. ........ /
. .—.. ..—., ..-. ..—. .— .

4 9.. A. 4 6 7

 

 

Figure 51: Average velocity ﬁeld for the crank angle of 270 degrees, 1500

RPM, wide open throttle

58

Appendix B

Disassembly of Firing Rig

Upon completion of the tests the single cylinder ﬁring rig was
disassembled and inspected. The combustion chamber was found to contain
a surprisingly large amount of carbon deposits. Inspection of the piston
would reveal that the heavy carbon was a result of the failure of the oil
control ring as well as the lower compression ring. Also of note was the
scoring of the cylinder walls and the piston skirt. Although the exact cause of
the excessive wear is not known, there are two known possible causes. The
ﬁrst is that when the injectors were initially being calibrated, possibly too
much fuel was injected when the engine was not ﬁring. This could have
washed the walls of oil, thus causing the piston to overheat. The second
possibility is that the piston skirt may have protruded too far out of the
bottom of the cylinder bore at BDC. The cylinder and head combination has
been used for previous experiments at the Michigan State University Engine
Research Laboratory; however, for this experiment the compression ratio was
much lower, requiring a larger shim at the bottom of the cylinder. The
difference in shim thickness was less than 1 centimeter, but this may have
allowed the piston to extend far enough out of the bottom of the bore to rotate
about the piston pin and place excessive force on the piston skirt/cylinder

wall interface, thus overheating the piston.

59

 

Figure 53: Cylinder head from single cylinder ﬁring rig after testing.

60

 

Figure 54: Piston from single cylinder ﬁring rig after testing.

61

Appendix C

Discussion of Indicated Mean Effective Pressure

The performance for the ﬁring engines in this study is measured in
terms of Indicated Mean Effective Pressure (IMEP). IMEP relates to the

work output per unit swept volume and is deﬁned as:

indicated work output per cylinder per mechanical cycle

IMEP: .
swept volume per cylinder

 

[13] This can also be seen graphically in the pressure verses volume plot in
Figure 55. The graph contains two loops, the top which represents the work
done on the piston by the combustion process and the bottom which
represents the pumping work performed by the piston to draw fresh charge
into and expel spent charge out of the cylinder. The area of the top loop
(shaded in blue) is the IMEP. The area of the lower loop is known as the
Pumping Mean Effective Pressure, and can be subtracted from the IMEP to
get the net work done on the piston per cycle or net IMEP. In this study,
however, all of the data is in terms of gross IMEP since the available

equipment made for PMEP values of questionable accuracy.

62

Pressure

 

 

 

 

 

Volume

Figure 55: Typical Pressure vs. Volume diagram for a four cycle spark
ignition engine.

63

References

64

References

Ambrose, G, "A Study of the Effects of Exciting the Intake Flow in an
Internal Combustion Engine Model." Master’s Thesis, Michigan State
University, 1997

Cohn, R. K., and Koochesfahani, M. M., "The Accuracy of Remapping
Irregularly Spaced Velocity Data onto a Regular Grid and the
computation of Vorticity." Experiments in Fluids, pp. 861-869, 2000

Gendrich, C. P., and Koochesfahani, M. M., "A Spatial Correlation
Technique for Estimating Velocity Fields Using Molecular Tagging
Velocitrimity (MTV)." Experiments in Fluids, Vol. 22, pp. 67-77, 1996

Gendrich, C. P., and Koochesfahani, M. M., and Nocera, D. G,
"Molecular Tagging Velocitrimity and Other Novel Applications of a

New Phosphorence Supermolecule." Experiments in Fluids, Vol. 23, pp.
361-372, 1997

George, S., "Test Stand Design for a High Compression, Direct-Injection
Spark-Ignition Methanol Fueled Engine and Results of Various Engine
Conﬁgurations." Master’s Thesis, Michigan State University, 2000

Goh, A. C. H., "Active Flow Control for Maximizing Performance of
Spark-Ignited Stratiﬁed Charge Engines." Master’s Thesis, Michigan
State University, 2001

Hill, R. B., and Klewicki, J. C., "Data Reduction Methods for ﬂow
Tagging Velocity Measurements." Experiments in Fluids, Vol. 20, pp.
142-152, 1991

Hussain, A. K. M. F., and Zaman, K. M. B. Q., "The ’Preferred Mode’ of

the Axisymmetric Jet." Journal of Fluid Mechanics, Vol. 110, pp. 39-71,
1981

65

10.

ll.

12.

13.

14.

Koochesfahani, M. M., and MacKinnon, C. G. "Inﬂuence of Forcing on
the Composition of Mixed Fluid in a Two-Stream Shear Layer.” Physics
of Fluids A, Vol. 3. pp. 1135-1142. 1991

Koochesfahani, M. M., and Stier, 8., "Molecular Tagging in Gas Phase
and its Application to Jet F lows." ASME Fluids Engineering Devision
Summer Meeting, FEDSM’97, 1997

Stier, B. and Koochesfahani, M. M., "Molecular Tagging Velocitrimity
(MTV) measurements in Gas Phase F lows." Experiments in Fluids, 26

(4), pp. 297-304, 1999

Stier, B. and Koochesfahani, M. M., "Whole Field MTV Measurement
in a Steady Flow Rig Model of an IC Engine." SAE Paper 980481, 1998

Stone, R., "Introduction to lntemal Combustion Engines." Third Edition,
SAE International, 1999

Zaman, K. M. B. Q., and Hussain, A. K. M. F., "Vortex Pairing in 3
Circular Jet Under Controlled Excitation. Part 1. General Jet Response."
Journal of Fluid Mechanics, Vol. 101, part 3, pp. 449-491, 1980

66