Thesis Fitrian Imaduddin

A NOVEL MAGNETORHEOLOGICAL VALVE WITH MEANDERING FLOW
PATH STRUCTURE
FITRIAN IMADUDDIN
UNIVERSITI TEKNOLOGI MALAYSIA
Replace this page with form PSZ 19:16 (Pind. 1/07), which can be
obtained from SPS or your faculty.
We hereby declare that we have read this thesis and in our

opinion this thesis is sufficient in terms of scope and quality for the
award of the degree of Doctor of Philosophy
Signature
Name
Date
:
:
:
Associate Professor Ir. Dr. Saiful Amri Mazlan
Signature
Name
Date
:
:
:
Dr. Hairi Zamzuri
Replace this page with the Cooperation Declaration form, which can be
obtained from SPS or your faculty.
A NOVEL MAGNETORHEOLOGICAL VALVE WITH MEANDERING FLOW

PATH STRUCTURE
FITRIAN IMADUDDIN
A thesis submitted in fulfilment of the

requirements for the award of the degree of
Doctor of Philosophy
Malaysia-Japan International Institute of Technology

Universiti Teknologi Malaysia
JUNE 2015
ii
I declare that this thesis entitled A Novel Magnetorheological Valve with Meandering
Flow Path Structure is the result of my own research except as cited in the references.
The thesis has not been accepted for any degree and is not concurrently submitted in
candidature of any other degree.
Signature
Name
Date
:
:
:
Fitrian Imaduddin
iii
To my father, my mother, my wife and my brothers
iv
ACKNOWLEDGEMENT
I wish to praise the Almighty Allah the Most Gracious for the blessing and
strength that have been given to my life. My deepest gratitude goes firstly to my main
supervisor, Associate Professor, Ir. Dr. Hj. Saiful Amri bin Mazlan for his tremendous
support during my entire study. His intensive encouragement, enthusiasm and guidance
have made me able to pass through this process easier. I also would like to thank
my co-supervisor Dr. Hairi bin Zamzuri for his advice, interest and support to my
work. I must also express my gratitude to the Malaysia-Japan International Institute
of Technology (MJIIT) for the financial support provided during my study though the
MJIIT scholarship. I also would like to thank the Lembaga Pengelola Dana Pendidikan
(LPDP) for providing the additional incentive during the completion of my thesis
writing.
Appreciation is also given to the faculty members and colleagues in the Vehicle
System Engineering (VSE) research laboratory, especially Jamal, Yasser and Izyan
for helping me a lot during my earlier time in the university. Thanks also to Mr.
Hairullail, Madam Aishah and the remaining MJIIT staffs that have been supportive to
me during my study. Particular credit is also given to my Indonesian friends in UTM
Kuala Lumpur especially Ubaidillah and Burhanuddin for being my family abroad. I
would also like to thank Aizzat and the developers of the utmthesis LATEX project for
making the thesis writing process a lot easier for me. Special acknowledgement goes to
my previous supervisors Dr. Khisbullah Hudha and Dr. Gunawan Nugroho. They have
influenced me with the passion and love to the scientific research. I regret that I cannot
mention all the valuable names here, but I believe and pray that Allah will reward all
the good deeds that have been given to me.
Lastly, I would like to express my sincerest gratitude to my parents, Dr. Ahmad

Dahlan and Umi Sholichatin for unlimited love, support, trust and pray that have
brought me to this level. My two little brothers, Zamzam Ibnu Sina and Allahyarham
Ghilman Hunafa, for being such a good role model for me. Last but not least, my
dearest wife, Vivi Diawati, for the pray, patience, love and understanding that have
made me through this journey without any hesitations.
ABSTRACT
The development of a new Magnetorheological (MR) valve with meandering

flow path as a new approach to improve the MR valve performance is presented in
this research. The meandering flow path was formed by the arrangement of multiple
annular and radial channel so that the total effective area in an MR valve can be
increased without compromising the size and power requirement of the valve. The
main objective of this research is to explore the achievable pressure drop of the
MR valve with meandering flow path. This research was started with the concept
development where the meandering flow path structure is analytically modeled and
numerically simulated to predict and analyze the effect of variables involved. The
prediction results showed that the meandering flow path structure is able to increase
the achievable pressure drop of an MR valve significantly. The gap size analysis
showed that the size of annular gaps mainly contributed to determine the viscous
pressure drop component. Meanwhile, the field-dependent pressure drops were mainly
determined by the size of radial gaps. The prediction results of the concept was also
assessed and confirmed by the experimental work using a dynamic test machine. Based
on the experimental data, two hysteresis models, namely the polynomial model and
the modified LuGre model, were developed to model the hysteresis behavior. The
assessment results of the hysteresis models indicated that both model were able to
replicate the hysteresis behavior. However, the modified LuGre model, though 9.5%
less accurate than the polynomial model, was showing better consistency in a wider
range of input values. In general, the new concept contributes in the development of
a new type of MR valve that could achieve pressure drop nearly three times than the
annular, radial and annular-radial type MR valve.
vi
ABSTRAK
Pembangunan konsep baru injap reologi magnet (MR) dengan menggunakan

laluan aliran yang berliku-liku sebagai pendekatan baru untuk meningkatkan prestasi
injap MR dibentangkan dalam kajian ini. Laluan aliran yang berliku-liku dibentuk
melalui beberapa susunan saluran gegelang dan tebaran jejari secara berurutan
supaya jumlah kawasan yang berkesan di dalam injap MR boleh ditingkatkan tanpa
menjejaskan saiz keseluruhan dan prestasi injap. Tujuan utama kajian ini adalah untuk
meneroka kebolehcapaian nilai susutan daripada injap MR dengan menggunakan
laluan aliran yang berliku-liku. Kajian ini bermula dengan pembangunan konsep,
di mana injap dengan laluan aliran yang berliku-liku dimodelkan secara analitikal
dan disimulasikan secara berangka untuk meramalkan prestasi injap dan juga untuk
mengambil kira kesan pembolehubah yang terlibat. Keputusan simulasi menunjukkan
bahawa konsep injap dengan laluan aliran yang berliku-liku mampu meningkatkan
kebolehcapaian yang ketara dari segi nilai susutan tekanan daripada injap MR.
Berdasarkan kepada analisis saiz saluran telah dijalankan, hasil menunjukkan bahawa
saiz saluran gegelang lebih menyumbang kearah menentukan komponen kelikatan
dari susutan tekanan manakala komponen susutan tekanan akibat medan magnet
ditentukan terutamanya oleh saiz saluran dari tebaran jejari. Konsep ini turut dinilai
melalui kerja eksperimen menggunakan mesin ujian dinamik, yang telah mengesahkan
keputusan yang diramalkan oleh simulasi. Berdasarkan data eksperimen, dua model
histerisis, iaitu model polinomial dan model LuGre yang telah diubahsuai, telah
dibangunkan untuk mengilustrasikan tingkah laku histerisis injap MR. Keputusan
penilaian model histerisis menunjukkan bahawa kedua-dua model dapat mereplikasi
ciri-ciri histerisis daripada injap MR. Walau bagaimanapun, model LuGre yang
telah diubahsuai, walaupun 9.5% kurang tepat berbanding model polinomial, telah
menunjukkan konsistensi yang lebih baik dalam pelbagai ruang lingkup data masukan
yang lebih besar. Secara umumnya, konsep baru injap MR ini dapat memberikan
pendekatan baru dalam membangunkan sebuah injap MR yang dapat meningkatkan
kebolehcapaian susutan tekanan sehingga tiga kali ganda berbanding injap MR jenis
gegelang, jejari dan gegelang-jejari.
vii
TABLE OF CONTENTS
CHAPTER
TITLE
DECLARATION
DEDICATION
ACKNOWLEDGEMENT
ABSTRACT
ABSTRAK
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
LIST OF APPENDIX
PAGE
ii
iii
iv
v
vi
vii
x
xi
xiv
INTRODUCTION
1.1
Introduction
1.2
Motivation of Study
1.3
Research Objectives
1.4
Research Scope
1.5
Significance of Research
1.6
Outline of Thesis
1
1
2
6
6
7
7
LITERATURE REVIEW
2.1
Introduction
2.2
Magnetorheological Fluid
2.2.1
Composition of Magnetorheological
Fluid
2.2.2
Operational Modes of Magnetorheological Fluid
2.3
Magnetorheological Valve
2.3.1
Annular Magnetorheological Valve
2.3.2
Radial Magnetorheological Valve
9
9
9
10
11
14
15
18
viii
2.3.3
2.4
2.5
2.6
2.7
Combination of Annular and Radial

Experimental Assessment Method for Magnetorheological Valve
Modeling Approach for Magnetorheological Valve
2.5.1
Steady-state Model
2.5.2
Dynamic Model
Utilization of Magnetorheological Valve
2.6.1
Linear Magnetorheological Damper
2.6.2
Rotary Magnetorheological Damper
2.6.3
New Magnetorheological-based Actuators
Summary of Chapter 2
20
20
22
23
27
30
31
35
39
41
MAGNETORHEOLOGICAL VALVE CONCEPT

3.1
Introduction
3.2
Design of Magnetorheological Valve
3.2.1
Conceptual Design
3.2.2
Design Consideration
3.2.3
Magnetic Simulation
3.3
Steady-state Modeling of Magnetorheological
Valve
3.4
Simulation of Magnetorheological valve Performance
3.5
42
42
42
43
45
47
EXPERIMENTAL ASSESSMENT
4.1
Introduction
4.2
Experimental Apparatus
4.2.1
4.2.2
4.2.3
Testing cell
4.3
Experimental Set-up
4.4
Experimental Results
4.4.1
Off-state and On-state Pressure Drop
Characteristics
4.4.2
Effect of Gap Size
4.4.3
Effect of Current Input Variation
62
62
62
62
64
67
68
71
50
54
61
71
75
76
ix
4.5
HYSTERESIS MODELING OF MAGNETORHEOLOGICAL VALVE

5.1
Introduction
5.2
Polynomial-based Hysteresis Modeling Approach
5.3
Modified LuGre-based Hysteresis Modeling Approach
5.4
Model Performance Comparison
5.5
CONCLUSIONS AND RECOMMENDATIONS

6.1
Conclusions
6.1.1
The New Magnetorheological Valve Concept
6.1.2
Gap Size Selection Effect
6.1.3
Experimental Assessment of Magnetorheological Valve Performance
6.1.4
Hysteretic Modeling of Magnetorheological Valve
6.2
Contributions of the Research
6.3
Open Problems and Recommendations for Future
Works
6.3.1
Pressure Tracking Control System
6.3.2
Other Open Problems
REFERENCES
Appendix A
4.4.4
Effect of Excitation Frequency Variation
79
80
82
82
82
87
90
95
96
96
96
97
97
98
99
100
100
103
105
120 129
LIST OF TABLES
TABLE NO.
3.1
3.2
3.3
4.1
4.2
5.1
5.2
5.3
5.4
5.5
TITLE
Materials selection of valve component for valve routing
List of MR valve parameter
Performance benchmarking between the proposed MR valve
concept and the counterparts
Typical properties and material compatibility of MRF-132DG
The variable arrangement of experimental test using
Shimadzu Fatigue Dynamic Test Machine
Correlation test results between the model coefficient a and
current input I
List of coefficients for the polynomial-based parametric MR
valve model
List of approximated function for different parameters
Comparison of relative error at 0.75 Hz frequency excitation
Comparison of relative error at 0.50 Hz and 1.00 Hz
frequency excitations
PAGE
46
53
60
63
71
85
87
90
91
94
xi
LIST OF FIGURES
FIGURE NO.
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
2.10
2.11
2.12
2.13
2.14
2.15
2.16
2.17
2.18
2.19
2.20
2.21
2.22
2.23
2.24
2.25
2.26
2.27
2.28
2.29
TITLE
Movement of magnetic particles in the MR fluid with and
without magnetic field
Shear mode
Valve mode
Squeeze mode
Magnetic Gradient Pinch mode
MR throttle valve
C-shaped pressure control valve
Three port MR valve
Double-coil annular MR valve
Basic structure of single stage radial MR valve
Two-way controllable radial MR valve
Annular-Radial MR Valve
Typical arrangement of constant flow assessment method
Typical arrangement of variable flow assessment method
Illustration of significant variables in an MR valve
Bouc-Wen model
Parametric hysteretic polynomial model
Artificial Neural Network model
Valve mode MR damper
Shear mode MR damper
External coil MR damper
MR damper with bifold valves
Bifold MR damper for high impulsive loads
Bifold MR damper for shock vibration mitigation
Basic structure of Bypass MR damper
Bypass MR damper for large scale seismic application
Vane type MR damper with arc valve
Vane type MR damper with outer coil valve
Vane type MR damper with inner coil valve
PAGE
10
12
13
13
14
16
16
17
18
19
19
20
21
22
25
29
30
31
32
32
33
33
34
34
35
35
36
37
38
xii
2.30
2.31
2.32
2.33
2.34
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
3.11
3.12
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
MR based bellow-driven motion control

MR hydraulic power actuation system
Actuation with embedded Terfenol-D pump
MR based link manipulator
4/3 way directional MR valve (citation)
Concept assessment sequence of the new MR valve concept
Basic concept of MR valve with meandering flow path
Approximation of yield stress as a function of magnetic flux
density, B
Two-dimensional axisymmetric model of the MR valve in
FEMM
Flux lines and contour of magnetic field of the MR valve in
FEMM
Magnetic flux density along MR fluid flow path for 0.5 mm
gap size with respect to various current input
Gaps zone in MR valve with multiple annular and radial gaps
Dimension and variables of MR valve
Estimation of achievable pressure drop of MR valve with 0.5
mm gap size
Percentage of pressure drop contribution from each zone (a)
viscous (b) field-dependent at 1 A current input
Effect of gap size on the pressure drop (a) viscous (b) fielddependent at 1 A current input
Comparison of operational range between various gap
configurations
B-H curve of the MRF-132DG
Field induced yield stress of the MRF-132DG
Exploded view of the MR valve prototype
Failure of the bolt-locking mechanism to withstand internal
pressure
Modification and comparison of the MR valve prototype(a)
Exploded view of the MR valve design (b) Fabricated
prototype of MR valve
MR valve installation in the testing cell
Experimental arrangement schematic for the MR valve testing
MR valve testing set-up using in-house test machine
Testing cell installation in the Shimadzu Fatigue Dynamic
Test Machine
39
39
40
40
41
44
45
48
48
49
50
52
54
55
55
57
59
63
64
65
65
66
67
69
70
71
xiii
4.10
4.11
4.12
4.13
4.14
4.15
4.16
5.1
5.2
5.3
5.4
5.5
5.6
6.1
6.2
Comparison of measured and theoretical off-state peak

pressure drop at various flow rates for 0.5-0.5 mm (annularradial) gaps configuration
Comparison of measured and theoretical on-state peak
pressure drop at various current inputs and flow rates for 0.50.5 mm (annular-radial) gaps configuration
Comparison of measured peak pressure drop in various gap
size combinations
Comparison of the MR valve dynamic range for each gap size
combinations
The pressure dynamics of MR valve at various current input
for 0.5-0.5 mm (annular-radial) gaps configuration, (a) 0.50
Hz (b) 0.75 Hz (c) 1.00 Hz
The trend of peak pressure drop at various current input for
0.5-0.5 mm (annular-radial) gaps configuration
The pressure dynamics of MR valve at current input of 1A at
various frequency excitation for 0.5-0.5 mm (annular-radial)
gaps configuration
The difference between MR damper model and MR valve
model excitation
Trend of the normalized coefficient values at the positive flow
acceleration (lower loop) to the variations of current input (a)
a6 , a5 and a4 (b) a3 , a2 , a1 and a0
Trend of the normalized coefficient values at the negative flow
acceleration (upper loop) to the variations of current input (a)
a6 , a5 and a4 (b) a3 , a2 , a1 and a0
Trend of estimated parameters with respect to current input
Comparison between the test data and the model results for
various current input, (a) 0.3 A (b) 0.6 A (c) 0.9 A
Comparison between the test data and the model results for
current input of 1.0 A at various frequency excitations, (a)
0.50 Hz (b) 1.00 Hz
Basic structure of pressure tracking control of MR valve
Simulation results of pressure tracking control under various
functions as reference, (a) Sinusioidal (b) Pulse (c) Saw-tooth
72
73
76
77
78
79
80
83
85
86
90
92
93
100
102
xiv
LIST OF APPENDIX
APPENDIX
A
TITLE
CAD Drawings
PAGE
120
CHAPTER 1
INTRODUCTION
1.1
Introduction
Magnetorheological (MR) fluid is one of the fluids in the class of field

responsive material [1, 2], that has sensitive rheological properties to magnetic field
[37]. The development of the fluid, together with the progressing research in the
understanding of its behavior, has convinced researchers and engineers that MR fluid
is a promising material for future applications [810]. This is because of their adaptive
force capacity and their inherent ability to provide a simple, fast and robust interface
between electronic controls and mechanical components. The fluid was first introduced
in Rabinows Magnetic clutch in 1948 [11] and has gained in popularity since entering
the automotive market. MR fluid is very responsive to magnetic field, with an estimated
response time of less than 10 milliseconds [12], and requires relatively low power to
operate. The advantages of MR fluid have created great interest in MR based device
development in a wide range of applications.
One of the most popular devices that utilized the unique characteristics of
MR fluid is MR damper [13], which has been commercially available for highend passenger vehicles as a semi-active suspension or adjustable suspension [14].
The working principle of an MR damper is basically similar to a conventional
viscous damper which employs flow restriction concept to generate damping. The
flow restriction in a conventional viscous damper is normally generated by an orifice
channel which act as the valve. Since the gap of the orifice channel is fixed, the flow
restriction that can be generated by the valve of the conventional viscous damper is
also fixed. The MR dampers use different approach by employing MR fluids as its
working fluid and an MR valve in its flow restriction mechanism. Although the gap
size of the channel in an MR valve also can be fixed, the magnetic field strength in the
flow channel of the MR valve can be regulated [15]. Therefore, the flow of MR fluid
2
that pass through the MR valve can be controlled without having to modify the gap
size of the channel. On the other hand, it can be said that the performance of the MR
valve to generate flow restriction highly determines the overall performance of the MR
damper.
Considering the importance of MR valve, many designs of MR valve have

been proposed. One of the earliest design of stand-alone MR valve was proposed in
Kordonski et al. [16] which later elaborated by Gorodkin et al. [17]. In the literatures,
annular MR valve designs with optimize-able geometry and controllable MR fluid flow
resistance were provided. A simpler concept of annular MR valve was proposed by
Yokota et al. [18], which consisted of annular flow channel and electromagnetic coil
installed in adjacent to the flow channel. The works were improved by Yoshida et al.
[19] by proposing a three-port annular MR valve using permanent magnet. In the same
time, a meso-scale (less than 25 mm outer diameter) annular MR valve were proposed
by Yoo and Wereley [20] using internal double coils with counter flux direction. While
the advancement of annular type MR valve were continuously explored, Wang et
al. [21] started to discuss about the radial type MR valve for the large-scale seismic
bypass damper configuration. The benefit of radial MR valve over annular MR valve
in terms of pressure drop rating as well as the benefit of external bypass MR valve
configuration was compared in the literature. Performance assessments of MR valve
were also performed by Grunwald and Olabi [22] through the performance analysis
of the annular and orifice type MR valve. The discussions of MR valve type were
extended by Ai et al. [23] and Wang et al. [24] through an MR valve design with both
annular and radial flow path. In their design, both type of resistance channel were used
in an MR valve to increase the on-state resistance force while maintaining valve size
and power consumption. In order to make an MR valve more applicable to retrofit
general hydraulic applications, Yoo and Wereley [25] introduced the installation of
multiple MR valves in H-bridge configuration to actuate a hydraulic cylinder. The work
then followed by John et al. [26] with the embedded version of H-bridge MR valve and
by Salloom and Samad [27] with the introduction of 4/3 way MR valve design.
1.2
Motivation of Study
MR damper for semi-active vehicle suspension systems are among the most
popular and commercially successful MR fluid devices [2834]. In general, vehicle
suspension system can be divided into three categories; passive suspension system,
semi-active suspension system, and active suspension system [35]. Passive suspension
3
system is the common suspension system installed in most vehicle nowadays
which typically consists of spring and damper in parallel configuration. Semi-active
suspension system is similar with passive suspension system but the stiffness of the
component (spring and/or damper) can be controlled to suit the desired ride or handling
performance [36, 37]. Active suspension system, on the other hand, is the suspension
system with the involvement of active actuators such as hydraulic [38], pneumatic [39]
or electro-mechanic [40,41], which could provide external force to the suspension. MR
dampers are usually implemented as a semi-active device to retrofit hydraulic dampers
to enhance passive suspension performance. Enhancement of suspension performance
is feasible since the performance limitations of passive suspension system occurred due
to a fixed stiffness value of the spring and damper. In this case, MR damper, in contrast
to conventional linear hydraulic damper, has the capability to change its damping
stiffness by varying the magnetic field strength inside the damper. Together with
embedded control system, MR dampers have gained popularity and proved its potential
to enhance the performance of suspension systems. Aside of dampers, other types of
MR devices have been developed to meet other automotive application demands such
as engine vibration suppressors [4245], seat suspensions [4649], brakes [5053] and
clutches [5457].
According to the location of the valve, the MR damper can be divided into two
groups, the MR damper with internal valve and the MR damper with bypass valve.
The MR damper with internal valve typically has MR valve embedded in the piston
of the damper, similarly with the configuration of the valve in a conventional viscous
damper. This configuration is the most common valve installation in an MR damper
since it is neat and compact. However, the internal valve configuration is not without
setback. The disadvantages of internal valve configurations are mainly in the space
limitation of MR valve installation, the complexity of wiring and in the risk of thermal
build-up from the immersed valve. The MR valve integration to the piston is the main
reason why the construction of the MR damper with internal valve can be really neat
and compact. However, since the available space inside the cylinder is very limited
and the MR valve requires sufficient space for electromagnetic coil and magnetization
channel, the performance range of the damper is very narrow. Moreover, since the
coil is embedded with the piston, the common way of wiring installation is normally
made through the conduit along the rod, which made it prone to leakages and tends
to be costly for fabrication. On the other hand, the heat dissipation, as a result of
kinetic energy conversion into heat, can be more severe in an MR damper than in a
conventional viscous damper because the magnetically altered damping stiffness will
definitely increase the heat dissipation. In the case where the MR valve is immersed in
the MR fluid, the heat dissipation from the MR valve will have to disperse to the MR
4
fluid first, which responsible in the increase of fluid temperature, before eventually
released to the environment. The experimental observation conducted by [58] reported
that the temperature rise of MR fluid in an MR damper after 400 s of operation at
current input of 2 A and frequency excitation of 6 Hz have caused the damping force to
degrade in about 38%. However, the same experiment observed that less degradation
can be achieved if the MR damper is properly finned, whereas increase the thermal
release to the environment.
The practice in the other type of MR damper, known as the bypass MR

damper, is not embedding the valve in its piston since the construction uses no
fluid channel in the piston [59]. In the bypass MR damper, the fluids flow between
chambers through the bypass channel outside the cylinder where MR valve is installed.
Therefore, the valve in the bypass MR damper configuration is easier to be installed and
maintained since the construction of the main cylinder is similar with the structure of a
conventional hydraulic cylinder. The bypass MR damper is also less prone to over-heat
because the valve is already located outside the cylinder. Various types of MR valve
also can be implemented in an MR damper with bypass configuration because the valve
size is no longer constrained by the cylinder size nor the piston size. However, the
existence of bypass channel and MR valve outside the cylinder are obviously making
the bypass construction not as neat and compact as the damper with internal valve.
The MR damper with bypass configuration is also difficult to be installed in spaceconstrained applications since the bypass damper requires more room than the damper
with internal valve. With these characteristics, the bypass configuration is normally
implemented in the large scale MR damper with high energy dissipation [6062].
Despite the advantages and disadvantages of each MR damper structure,

the technological advancement of the MR valve, as the heart of the MR damper
performance, is not as extensive as the advancement of the MR damper. Regardless
the types of MR damper, most of them are still using the same MR valve concept. The
only differences are probably the size, coil configuration and/or MR fluid types. Most
of MR dampers are employed with annular type MR valve as one of the most popular
types of MR valve [29, 46, 49, 6368]. The annular MR valve is the first generation
of MR valve that utilized the annular channel as the effective area. The effective area
is the area where the MR fluids are exposed to magnetic flux perpendicularly to the
flow direction. There are several variants of annular MR valve that has been proposed
by researchers [1620, 69], but the main concept is basically similar. The annular MR
valve is popular because it is simple to be manufactured and has a high ratio between
the on-state and off-state performance. However, the effectiveness of space utilization
5
in the conventional annular MR valve is very low because not all areas of the annular
channel can be utilized as the effective area. Therefore, any improvement effort on
the annular valve performance will typically tend to increase the valve size either in
length, by enlarging the effective area, or in diameter, by enlarging the electromagnetic
coil. Thus, in a constrained space device such as in the MR damper with internal valve
configuration, the desired performance improvements are sometimes difficult to be
achieved.
Due to the limitation of the annular MR valve, another type of valve, known as
the radial MR valve, was introduced by [21]. The radial MR valve, as a distinction from
annular MR valve, has radial flow channel inside the valve and utilize it as the effective
area. The utilization of radial channel as the effective area offers several benefits than
the effective area of the annular channel, especially in terms of area efficiency since
the radial channel can be made in multi-stage configuration. Therefore, with multistage capability of the radial MR valve, the performance improvement of radial valve
typically has lower implication to the valve size than the one in the annular MR valve.
As a result, the radial valve concept has been installed to serve several concepts of
large scale MR dampers [61, 62, 70, 71]. Recently, another concept of MR valve also
has been developed by combining both annular and radial valve concept in a single
valve [23, 24]. The combination of both annular and radial channel in an MR valve has
been proven effective to improve the performance of MR valve. It has been reported
by [72] that the MR valve with combination of annular and radial channel has higher
achievable pressure drop than annular valve with lower power consumption although
at the cost of lower valve ratio. The MR valve with combination of annular and radial
channel also has been implemented in MR mount design [42] and MR damper design
[73].
From these explanations, it can be observed that the technological advancement

of an MR valve has a significant impact to the advancement of other MR devices.
Therefore, particular explorations of the MR valve concept are necessary as a basis
to provide knowledge on how to improve the performance of MR devices in general.
The concept explorations is not limited to the geometrical and design arrangements of
the MR valve, but also in terms of behavioral characteristics of the MR valve such as
the identification of the MR valve hysteretic behavior. The hysteretic behavior, as well
as other complex characteristics, in generally in any MR devices is still considered a
challenging problem in terms of the modeling technique and controller design [14,74].
The hysteresis could be occurred due to magnetic field remnant in steel elements
and due to the viscoelastic properties of MR fluid itself. In terms of controllability,
6
hysteresis behavior is a disadvantage since the controller will face difficulties to track
the damper behavior. For example, according to Wang and Liao [74], tracking ability of
damping force is one of the highly important issues that should be considered in order
to get an accurate MR damper controller. However, a controller with such capability
will tend to be more complex, require more computational resources, be costly and
less robust. Therefore, innovation in the control design is also vital to support the
final implementation of MR devices. Innovation of the control algorithm will be more
difficult if the model that is used in the controller design phase is not able to simulate
the hysteresis phenomenon accurately. A simple and accurate model of an MR valve, in
particular, is needed in order to design an appropriate controller with good robustness,
stability and reliability. Therefore, the advancement of modeling technique that have
the ability to accommodate the hysteretic behavior of MR valve is as important as the
advancement of the MR valve concept.
1.3
Research Objectives
This study embarks on the following objectives:
(a)
To develop a new concept of MR valve with meandering flow path to improve

the achievable pressure drop.
(b)
To analyze the effect of gap size selection to the achievable pressure drop of MR
valve.
(c)
To assess the performance of MR valve using experimental work.
(d)
To model the hysteretic behavior of the MR valve.
1.4
Research Scope
In this research, a new concept of MR valve will be investigated. This study

focuses on the elaboration of a new concept of MR valve utilizing the combination of
multiple annular and radial gaps that formed a meandering flow path. The new concept
of MR valve is introduced to provide an adjustable pressure drop with a high on-state
limit. In order to examine the capability of the MR valve, the steady-state model of
the new MR valve concept is derived and the magnetic circuit performance of the
MR valve is simulated using Finite Element Method Magnetics (FEMM) software
7
package. The performance of MR valve, in this study, is only evaluated in terms of
the achievable pressure drop as a function of gap size of the flow channel, magnitude
of current input charged to the coil, and fluid flow rate. This research is also covering
the experimental evaluation of the MR valve using an MR valve testing cell in variable
flow rates, to reveal the hysteretic behavior, with constant current inputs. The measured
performance of the MR valve is also used to model the hysteretic behavior of the MR
valve, which is not covered in the steady-state model. However, the optimization of the
concept is not discussed in this research and the demonstration of control application
is only performed.
1.5
Significance of Research
The significance of this research is mainly in terms of general advancement

of MR devices and applications especially to answer the demand of smart, simple yet
high performance and reliable new MR devices. The new concept of MR valve with
meandering flow path is expected to provide a new method to improve the design of an
MR valve, which will highly influence the design of MR damper as well as other MR
based actuators. Moreover, the concept is expected to be performed as a demonstration
of a generic MR device that can suit various applications. Therefore, the concept can
be anticipated as a modular and re-sizable device so that the range of operation and the
capacity can be adjusted. The significances of this research are summarized as follows:
(a)
This study demonstrates a new concept of MR valve using a meandering flow

path structure.
(b)
This research provides knowledge of the effect of gap size selection to the
achievable pressure drop of the valve which will be further useful for valve sizing
process.
(c)
The hysteretic modeling process of the MR valve introduces a new modeling

approach of MR valve using modified LuGre hysteresis model.
1.6
Outline of Thesis
This thesis is organized in six chapters. Each respective chapter in this

thesis ends with a brief summary outlining the achievements and findings that were
8
established in the chapter. The outline of this thesis is organized as shown:
Chapter 2 covers the theoretical background, which includes the properties

and the working modes of MR fluids, the basic knowledge of MR valves, the recent
advancement of MR valves, as well as the applications of MR valves.
Chapter 3 explains the new concept of the MR valve with meandering flow
path, the design consideration for the performance assessment, the steady-state model
derivation, the magnetic simulation as well as the performance prediction of the new
MR valve with respect to various dependent variables.
Chapter 4 elaborates the experimental evaluation of the MR valve including the

description of the experimental setup, the experimental procedure and the analysis of
the experimental results.
Chapter 5 presents the development of two different hysteresis MR valve

models, the parameter estimation processes and the performance comparison of these
two models.
Chapter 6 concludes the work and presents the achieved contribution of the
research as well as recommends open problems for future work.
CHAPTER 2
LITERATURE REVIEW
2.1
Introduction
This chapter provides an overview of magnetorheological (MR) fluids, the

working modes of MR fluids, the MR valve design and modeling as well as the
application range of MR valves. The overview is presented in order to elaborate the
research background about the main idea presented in this research. The first part of
this chapter presents some of the main issues related to the MR fluids composition
and its working mode. The second part of the chapter elaborates the fundamentals and
classifications of the MR valves. The third part of this chapter discussed about the
modeling approach. The final part of the chapter provides the review of the current
application that involved MR valves in its working mechanism.
2.2
The MR fluid was first discovered in the Rabinows magnetic fluid clutch in
1948 [11] and has been classified as one of the class of smart materials [9]. MR fluid is
classified as smart material because of its adaptive behavior and their inherent ability to
provide a simple, fast and robust interface between electronic controls and mechanical
components. When the fluid is subjected to magnetic field, the iron particles start
aligning along the magnetic fluxes, as shown in Figure 2.1. Hence, the movement
of the fluid is restricted by particle chains thus increasing its apparent viscosity.
Consequently, the fluid changes its state from liquid to a viscoelastic solid depending
on the strength of the magnetic field. The advantages of MR fluid have created great
interest in MR based device development in a wide range of applications such as in
civil applications [60, 61, 63, 7580] medical prosthetic applications [8184], ground
vehicle applications [28,30,31,34,55,85], aerospace applications [8688] and military
10
applications [33, 8991]. Since the rheological properties of MR fluid can be modified
magnetically, an MR fluid based device requires no moving parts and therefore less
vulnerable to wear and tear.
Figure 2.1 Movement of magnetic particles in the MR fluid with and without magnetic
field
2.2.1
Composition of Magnetorheological Fluid
MR fluids are commonly referred to a two-phase fluid suspension made from

a mixture between magnetizable particle in a non-magnetizable liquid in such a way
so that it has very sensitive rheological properties to magnetic field [3]. Most of MR
fluids are using the carbonyl iron powders as the magnetic particles. The carbonyl
iron powders are normally acquired from the chemical vapor decomposition (CVD)
of pentacarbolyl iron, due to its large magnetization saturation and low coercivity
[92]. The carbonyl iron powders are also preferred in MR fluid synthesis because it
is chemically pure, naturally spherical in shape so that any anisotropy in magnetic
moment can be minimized [93]. In contrast with the particles, there are numerous kind
of carrier liquid that have been used in MR fluid such as water, mineral oil, silicon
oil, hydrocarbon oil, and polyesters [92]. In order to increase the strength of MR
fluid, the fraction between the volumetric amount of iron powder over the volumetric
amount of MR fluid is highly important. The shear stress of the fluid with the influence
of magnetic field, or known as the on-state MR effect, normally increases with the
increase of volume fraction of iron powder [94]. However, the trade-off appears in
the off-state viscosity value that also increases exponentially with the increase of
volume fraction of iron powder [95]. The trade-off has encouraged researchers to find
solutions to increase the strength of MR fluid and simultaneously suppress the off-state
viscosity value of the fluid such as to mix powders with different particle size and/or
size distributions [96] or to employ the functional composite particles that is coated
with magnetic nanoparticles layer using core-shell technique [97]. There are also other
11
approach of MR fluid synthesis which are mixing non-magnetizable particles dispersed
in magnetizable liquid (ferrofluid) that is known as the inverse ferrofluid [98] and the
Carbonnanotube (CNT)-magnetite based MR fluid [99].
MR fluid are often distincted from ferrofluid from its particle size. Ferrofluid,
similarly with MR fluid, is also a class of field sensitive fluid. However, the particle
size of MR fluid are often bigger (few microns) than the nano-sized particle ferrofluid
because it seek larger yield stress due to stonger dipole moment between the particles.
The yield stress of MR fluid represents the peak value of the stress versus strain curve,
since the gel-like structure will break when the stress reached this peak value [1].
However, since the particle size of the MR fluid is relatively large, it is susceptible to
settling and often caused unwanted abrasion [100]. The ferrofluid is less prone settling
because the nano-sized particles exhibit Brownian motion that keeps them dispersed
in the carrier fluid. The larger the size of the particle, the lesser the effect of Brownian
motion and the faster the tendency of settling. According to Bell et al. [101], the MR
fluids with particle size above 10 microns tend to settle in less than 20 minutes. That
is why, in order to inhibit sedimentation and particle aggregation, some approaches
such as the addition of thixotropic agents and surfactants [102, 103] or the attempt of
non-spherical particle shape implementations [104,105] were often given during to the
fluid synthesis process.
However, despite of several improvement efforts to the MR fluid composition,

there are generally no major modifications on the formulation technique of MR fluid
since the first discovery by Rabinow in the late 1940s. The strength of the very basic
fluid formulation can be considered adequate for most MR devices and the stability of
the fluid is also not a critical factor since most of MR devices such as dampers and
brakes can naturally act as a good mixing device [12].
2.2.2
Operational Modes of Magnetorheological Fluid
There are three known basic operational modes in any employment of MR fluid
in a device, shear mode, flow mode and squeeze mode [15, 92, 106, 107]. Recently,
another working mode was proposed by Goncalves and Carlson in [108], called the
magnetic gradient pinch mode. However, the working modes of MR fluid are not
strictly limited to these four since the MR devices can be operated in a combination of
these modes [109114].
12
Shear mode, also known as the clutch mode, occurs when MR fluid is exposed
to magnetic field between two parallel magnetic surfaces. One of the surfaces is
moving whilst the other is fixed, as shown in Figure 2.2. The fluid shear area where
MR fluid is exposed to magnetic field is called the effective area of shear mode. Shear
mode is commonly used in brakes and clutches but also appears in some particular
design of dampers, called the shear mode damper.
Figure 2.2 Shear mode
Flow mode, also known as the valve mode occurs, when MR fluid is exposed
to a magnetic field whilst the fluid flows between two fixed parallel magnetic surfaces.
The concept of flow mode is depicted in Figure 2.3. Similarly with the shear mode,
flow mode has the effective area, which is defined as the area where the flowing MR
fluid is exposed to magnetic field. Flow mode is commonly used in dampers and in
other applications in which the devices requires a valve to control fluid flow.
Despite the concept of squeeze mode has been used in several early rotor
vibration damper designs, the fundamental principle of this mode in MR fluid was
not discussed until later in several publications [115118]. Squeeze mode occurs when
MR fluid is exposed to magnetic field while at the same time being compressed or
decompressed, as shown in Figure 2.4. Since it involves very small displacement most
of its application appears in vibration suppression devices.
The latest working mode is called the magnetic gradient pinch mode. The
basic idea of this mode is similar to flow mode but with a different configuration of
13
Figure 2.3 Valve mode
Figure 2.4 Squeeze mode
magnetic circuit design. In the magnetic gradient pinch mode, as shown in Figure
2.5, the magnetic poles are arranged axially along the flow path and separated by a
non-magnetic material. This kind of poles arrangement will create elliptical magnetic
fibrils, which will block the flow of MR fluid in the valve gap. One of the unique
characteristics of this mode is the slope between pressure and velocity relationship in
magnetic gradient pinch mode will be significantly increased when the magnetic field
is increased [108]. This is unique, since in the conventional flow mode the slope tends
to remain constant under any magnetic field strength modifications. Another advantage
14
is the possibility to use MR fluid with coarser particles up to 100 microns, since a larger
orifice is feasible to be used with the magnetic gradient pinch mode [108].
Figure 2.5 Magnetic Gradient Pinch mode
2.3
Valve is the key components in almost any flow control mechanism. The
common practice in hydraulic system usually combines solenoid actuation valve
electronic devices in order to regulate fluid flow. However, conventional hydraulic
valve consists of several moving parts inside the valve, which makes it less responsive
while at the same time more vulnerable to wear and tear. Therefore, an MR fluid based
valve was introduced in order to improve the performance of valve. Moreover, since
working principles of most MR devices are based on the manipulation of fluid flow
rate, the key performance of MR device is determined by the performance of MR valve.
A successful improvement of MR valve performance can give a significant impact to
the development of other MR devices.
MR valve as one of the application branch of MR fluid typically consist of valve

body, valve core, electromagnetic coil, and fluid channel. The detail configuration
of each MR valve could be different depending on the specific design but the basic
principle is usually similar. The working principle of MR valve utilizes the sensitive
rheological properties of MR fluid. The change in MR fluid rheological properties due
to magnetic flux variation from the electromagnetic coil in the valve changes the fluid
flow resistance and therefore changes the pressure drop so that the fluid flow can be
slowed or even stopped [20].
15
Considering the importance of MR valve, many designs of MR valve have
been proposed. According to the structure of its flow path, MR valve can be divided
into three categories, the annular type MR valve, the radial type MR valve and the
combination of annular and radial MR valve.
2.3.1
Annular Magnetorheological Valve
The annular MR valve can be easily distincted as the MR valve with effective
area in the annular channel. Since the effective area of the valve mode is defined as
the area where the fluxes cross the fluid in perpendicular to the flow direction, the
fluxes in annular MR valve is always directed in radial direction. The simplicity of
mechanical structure of the annular MR valve has made it the most common MR valve
type employed in various MR devices. However, the length of annular channel that can
be exploited as the effective area is limited and consequently the magnetorheological
effect that can be generated is relatively narrow.
One of the earliest designs of stand-alone MR valve was proposed in Kordonski

[16] which later elaborated by Gorodkin et al. [17]. In the literatures, the initial
concept of MR valve, that is called the MR throttle valve, was presented to provide
a controllable MR fluid flow resistance. These designs basically comprise of two flat
annular throttling hydraulic channels, indicated by 1, that were connected via internal
channels, indicated by 2, as shown in Figure 2.6. The induction coil winding, 4, and
the core, 5, were made from magnetoconducting material and together are formed
the magnetoconducting body, 3. The hydraulic channel itself is used as a part of the
magnetic circuit, which the gap between the body and the core is acted as the zone
to induce the MR fluid with magnetic field. The design was also apparently prepared
to serve various flow connection pattern for this design can be used in both serial or
parallel connections depending on which flow port is connected to the flow conduit.
The valve is acted as serial valve when each channel, 1, is connected to inlet/outlet
port, 6, to the inlet/outlet pipe, 8. Meanwhile, the valve is served parallel configuration
when the pipe inlet, 10, is connected to the channel, 1, to the outlet, 8. The results of
performance testing of this design were shown the capability to provide pressure drop
of nearly 2.5 MPa at flow rate of 40 ml/s and 200 kA/m magnetic field intensity.
Another concept of annular MR valve was proposed by Yokota et al. [18]. The
design employed the very basic principles of valve mode using a C-shaped magnetic
16
Figure 2.6 MR throttle valve [17]
core attached to the flow channel as shown in Figure 2.7. The C-shaped magnetic
core act as the main flux guiders which conducted the magnetic flux loops to induce
the MR fluid flows in perpendicular direction. Although the experimental results were
only reported around 0.7 MPa of achievable pressure drop at similar flow rate and field
intensity with the previous design, the design supposed to be very efficient since flux
loss can be prevented to a very minimum value. However, the bulkiness of the design
would quite inconvenience for application and compared with the previous design,
this design is only equipped two flow port which made it only suitable for serial flow
connection.
Figure 2.7 C-shaped pressure control valve [18]
The work of Yokota et al. [18], was improved by Yoshida et al. [19] through the
17
introduction of a three-port annular MR valve as depicted in Figure 2.8. The three-port
annular MR valve design comprises of serially connected two-port MR valves similarly
with the one in Yokota et al. [18] concept. However, one supplementary port is added
in between these two-port to ease the application that requires parallel flow connection.
A NdFeB permanent magnet is added in between both coil to create a magnetic offset
as well as to regulate the direction of the fluid.
Figure 2.8 Three port MR valve [19]
Another group of researchers have focused to pursue an improvement of MR

valve in terms of size and performance by introducing the meso-scale (less than 25
mm outer diameter) annular MR valve [20]. As illustrated in Figure 2.9, the concept
was proposed using double coils with counter flux direction. The coils were wound in
the bobbin shaft which acted as the main flux core, while the valve casing was also
acted as the flux return. The efforts were focused in two aspects, the first one was the
analysis of plug thickness, , which was defined as the thickness of pre-yield region
and directly related to the relationship between magnetic field strength and the value
MR fluid yield stress, while the second one is related to the material selection of the
valve which influenced the induced magnetic field strength to the MR fluid. The results
demonstrated that the approach was successfully generate saturated pressure drop of 2
MPa at the flow rate of 40 ml/s.
18
Figure 2.9 Double-coil annular MR valve [20]
2.3.2
Radial Magnetorheological Valve
Although the advancement of annular MR valve were continuously explored,

the space limitation to exploit the effective area in the annular channel has attract
researchers to investigate the other types of MR valve known as the radial MR valve.
Radial MR valve is identified as the valve with effective area located in the radial flow
direction. In order to perpendicularly cross the radial fluid flow, the magnetic fluxes
are typically directed in annular direction. As a result, the length of channel required to
exploit the effective area is much smaller than the annular MR valve and therefore the
magnetorheological effect of the radial MR valve is potentially wider than the annular
MR valve.
The benefit of radial MR valve is introduced by Wang et al. [21] in the

discussion of radial MR valve for large-scale seismic damper application. The basic
structure of radial valve that was presented in the literature is shown in Figure 2.10. In
the literature, the benefit of radial MR valve was benchmarked using the case of large
scale damper applications. As a comparison, two large scale dampers with annular
MR valve were compared with a large scale MR damper with radial MR valve. The
results shown that the large scale MR damper with radial MR valve could provide
higher efficiency in shorter valve length. The other advantage of radial MR valve
structure, namely the capability of radial MR valve to be made in modular stage was
also highlighted in the discussion. The performance improvement of the MR valve was
shown to be linearly related with the number of addition of radial module in the MR
valve.
The work on radial MR valve was extended by the introduction of two-way

controllable MR valve by Aydar et al. [119] as illustrated in Figure 2.11. Similar
concept of radial MR valve was enhanced through the installation of permanent magnet
in adjacent to the electromagnets. The permanent magnet constantly provided initial
19
Figure 2.10 Basic structure of single stage radial MR valve [21]
magnetic flux to the radial gaps so that the off-state condition already has pre-yield
value. The two-way condition of the valve defined the ability of the valve to reduce the
pressure drop as well as to enhance the pressure drop depending on the resultant of the
flux directions. The presented design utilized the manipulation of flux similarly with
the one presented by Yoshida et al. [19]. The combination of permanent magnet and
electromagnet, which knows as the hybrid magnetic circuit, also provide additional
advantages such as the possibility to enhance the on-state performance, to reverse the
magnetorheological effect as well as to provide fail-safe features during power failure
to the electromagnet [120].
Figure 2.11 Two-way controllable radial MR valve [119]
20
2.3.3
Combination of Annular and Radial Magnetorheological Valve
The successful performance improvement on radial MR valve approach has

made researchers continuously explored the advancement of MR valve by combining
the annular and radial MR valve. Ai et al. [23] was among the first who explored
the gaps combination approach in MR valve, which later detailed by Wang et al. [24]
as shown in Figure 2.12. Basically, the main idea of the approach is maximizing the
engagement opportunities between the fluid and the flux by channeling the fluid in
annular and radial gaps so that the fluid can cross the magnetic flux perpendicularly.
With the combination of annular and radial gaps in an MR valve, the valve efficiency
can be increased in the meaning that the fluid flow blocking force can be increased
without any significant expenses to the valve size and energy consumption.
Figure 2.12 Annular-Radial MR Valve [24]
According to the study by Nguyen et al. [72] on the optimization of various

MR valve design, it was reported that the MR valve with annular and radial gaps
combination is capable to deliver higher achievable pressure drop than annular MR
valve. In the same value of generated pressure drop, the annular-radial MR valve also
has the smallest power consumption. However, in terms of the ratio of viscous pressure
drop the the field dependent pressure drop, the annular-radial MR valve has lower
performance than annular valves.
2.4
Experimental Assessment Method for Magnetorheological Valve
The experimental assessment of the MR valve normally aimed to measure the

pressure drop characteristics and validate the predicted performance of an MR valve
21
design. The characteristics of MR valve pressure drop are typically evaluated in terms
of response to the variation of flow rate and response to the variation of current input.
In general, the experimental assessment method of MR valve can be divided into two
types. The first type is the constant flow assessment while the second type is the
variable flow assessment.
The constant flow assessment evaluates the performance of MR valve by

measuring the pressure difference between both MR valve ports with subject to
steady fluid flow rate. This assessment arrangement has been used in several study
[18, 27, 121, 122] is capable to characterize the MR valve similarly with the results
from the quasi steady-state model where the pressure drop value varies in quasi steady
flow rate and yield stress. The constant flow assessment is also very useful to evaluate
the time constant of the MR valve, since it can maintain the constant flow rate value
while rapidly changing the current input. However, the testing arrangement of the
constant flow assessment method is a little bit complicated since it almost always
involves fluid pump in the arrangement such as shown in Figure 2.13. The complexity
of the measurement comes from the existence of iron particles in the MR fluid and its
abrasive characteristics that limit the option of pump selection.
Figure 2.13 Typical arrangement of constant flow assessment method [18]
The variable flow assessment evaluates the performance of MR valve by

measuring the pressure difference between both MR valve ports with subject to
variable fluid flow rate [20, 123, 124]. The fluid flow in variable flow assessment
is normally generated by a reciprocating movement of a hydraulic cylinder which
22
simultaneously generates pressure difference between cylinder chambers. The pressure
difference between hydraulic chambers induces alternating flow between those
chambers which the flow line passes through the MR valve. Since the flow is
alternating, the variable flow assessment is very ideal to observe the hysteresis behavior
of the MR valve. The result of variable flow assessment of an MR valve is also
very useful to further predict the dynamic behavior of a damper that utilizes the MR
valve. The measurement arrangement is also relatively simpler than the constant flow
assessment given that the dynamic test machine or fatigue test machine can be used
as the source of movement to the hydraulic cylinder such as shown in Figure 2.14.
However, the method cannot be used to measure the time constant of the MR valve
since the valve response to rapid change of current input in a constant flow rate cannot
be executed.
Figure 2.14 Typical arrangement of variable flow assessment method [124]
2.5
Modeling Approach for Magnetorheological Valve
In order to satisfy the requirement of control system design, the behavior of

MR based devices should be mathematically modeled. In general, the behavior of
any MR based devices can be described in two ways, the steady state behavior and
the dynamic behavior. The steady state behavior of the MR device is defined as the
behavior when the variation of variables is time-invariant. The dynamic behavior, on
23
the other hand, is defined as the behavior when the influenced variables are changing
over time. According to Snyder et al. [125], the linear steady state behavior model is
only sufficient to predict the energy dissipation, however, it cannot accurately portrays
the force response of the MR damper. The force response of the MR damper is known
as a highly nonlinear variable that is dependent to amplitude and frequency of motion
and it can only be modeled using nonlinear dynamic model [126129].
2.5.1
Steady-state Model
In general, any model of MR based devices would fundamentally refer to three

known models of MR fluid characteristic, Bingham plastic model, biviscous model,
and Herschel-Bulkley model [74]. Bingham plastic model is the simplest and the most
popular model in field controllable fluids such as Electrorheological (ER) fluid and
MR fluid. Bingham plastic model describes the rheological behavior of MR fluid in the
pre-yield region as a rigid solid body while in the post-yield region as Newtonian fluid
[130]. The Bingham plastic model can be described in the stress-strain relationship
equation as follows [74]:
= y (H) sgn ()
+
(2.1)
where is the MR fluid viscosity when no magnetic field applied, is the shear-strain
rate and sgn ()
is the signum function for shear-strain rate.
Biviscous model is particularly used for the analysis of MR based devices

operating in the squeeze flow mode [74], because it has an additional expression for
static yield stress. The equation for biviscous model can be expressed as follows [6]:
(H) + ,
y
po
=
,
pre
f or | | > ys
(2.2)
f or | | < ys
The expression of the biviscous model in Equation 2.2 has two states. For
absolute shear stress | | above static yield stress, ys , the expression is similar to
the Bingham plastic model with post-yield viscosity, po as the viscosity component
of the fluid. For shear stress below static yield stress, the expression is similar to a
conventional Newtonian fluid with pre-yield viscosity pre as the viscosity component
of the fluid. The relationship between dynamic yield stress y (H) and static yield stress
24
ys can be expressed as [6]:

po
y (H) = ys 1
pre
(2.3)
Wang and Gordaninejad [131] are among the first who generalized the use
of the Herschel-Bulkley model in MR fluid. The Herschel-Bulkley model can be
considered as the improvement of the Bingham plastic model since it has parameter
to anticipate shear thinning and shear thickening behavior, which only occurs at high
shear rates. The Herschel-Bulkley model can be expressed as [6]:

1
= y (H) + K ||
m sgn ()
(2.4)
In general the Herschel-Bulkley model is similar to the Bingham plastic model

except for the existence of two new parameters K and m. K and m are fluid parameters
known as the consistency parameter and the flow behavior index respectively. K is
similar to off-state fluid viscosity in Bingham plastic model while m is the index to
represent shear thinning or shear thickening effect. For m > 1, Equation 2.4 represents
a shear thinning fluid, while shear thickening fluids are described by m < 1. The
Herschel-Bulkley model can provide exactly the same results as the Bingham plastic
model when m = 1.
In the case of MR valve, most of its steady-state behavior are also modeled
based on Bingham plastic model equation, although in some conditions the more
sophisticated models, such as the Herschel-Bulkley model, are more preferred [117].
In particular, the MR valve behavior normally declared by the following equation:
P = Pviscous + Pyield
(2.5)
where the behavior of MR valve is described as the value pressure drop of an MR

valve that is consist of two parts, the pressure drop from viscous properties of the fluid
and pressure drop from the field dependent yield stress of the fluid. The most known
model of MR valve is the mathematical model based on the annular flow channel
configuration, which also became the common understanding of the flow mode of MR
25
fluid. The annular based MR valve model can be described as follows [23]:
Pviscous =
Pyield =
6QL
d3 R
(2.6)
c (H) L
d
(2.7)
where the pressure drop from viscous properties in Equation 2.6 is proportionally
related to the fluid base viscosity (), the volumetric flow rate (Q), and the annular
channel length of the valve (L) but inverse-cubical to the valve gaps (d) and inverse
proportional to the channel radius (R). Meanwhile, the field dependent pressure drop
in Equation 2.7 is proportionally associated with the field dependent yield stress value
( (H)) of MR fluid, annular channel length (L), and flow-velocity profile coefficient
(c) but inverse-proportionally related to the gap size (d). The coefficient c is normally
defined as the ratio between field dependent pressure drop and viscous pressure drop
which its value is in the range of 1 to 3.
Figure 2.15 Illustration of significant variables in an MR valve [132]
Note should be taken that the model shown in Equation 2.6, described the
viscous pressure drop using the constant of 6. In some other literature [22, 27], the
viscous pressure drop was described using the constant of 12. The reason behind the
difference is in the understanding of the annular channel length (L). In a common
annular valve design, there will always be two annular channel length (L1 and L2) that
26
are separated by the coil winding as shown in Figure 2.15. Some researchers, prefer
to define both length separately and therefore the pressure drop expression of each
annular channel length is defined using the constant of 6. Some others, assume both
channel length identical (L1 = L2 = L) and therefore since the length of L defined as
2L, the constant of 6 became 12.
The other form of annular MR valve model was also presented using an
as shown in the following
analytical terms called the nondimensional plug thickness, ,
equation [20, 133]:
P =
bd3
12QL

2
1
1+
(2.8)
The nondimensional plug thickness is defined as = 0 if the fluid flows as the

Newtonian fluid and defined as = 1 if the fluid is completely blocked by the valve
with infinite blocking pressure. However, the concept of plug thickness is difficult to be
implemented in the pressure drop prediction since the relation between the magnetic
strength and yield stress of the fluid to the pressure drop are not direcly expressed in
the equation. The plug thickness concept will be more suitable for reverse calculation
of yield stress using measured pressure drop data or to model the valve behavior using
steady-state equation.
The other steady-state model that has been developed is the radial MR valve
model. Wang et al. [21] proposed the radial valve model as described in the following
equation:
Pviscous
Pyield

Do
Di

i
(H)
2 Do D
2
=
d
6Q
ln
=
d3
(2.9)
(2.10)
where the valve gap size (d) in Equations 2.9 and 2.10 refers to the size of radial gaps
while the Do and Di refer to the outer radius of the radial gaps and the inner radius
of the radial gaps respectively. The field dependent yield stress value ( (H)) of MR
fluid in Equations 2.9 and 2.10 is also specifically refers to the yield stress value of in
the radial gaps. Similarly with the annular valve model, the gap size is also inversecubically related with the viscous pressure drop and inverse proportionally related with
the field dependent pressure drop.
27
In the same form with the Equation 2.9 and 2.10, the radial valve model was
also presented in the following form [24]:
Pviscous
6Q
=
ln
d3
Pyield =
Do
Di
c (H)
(Ro Ri )
d
(2.11)
(2.12)
where the main differences between these models are only in preference of using
diameter (D) and radius (R) of the radial gaps in the equation and also in the constant
selection of the field dependent pressure drop where in equation 2.10 the moderate
constant of 2 is used while in equation 2.12, the flow-velocity profile coefficient, c, is
chosen.
According to these equations, it can be summarized that the pressure drop

of an MR valve is determined by several factors. The fluid base viscosity, flow rate
are the parameters that proportionally related only to the viscous pressure drop. The
flow path length, on the other hand, is the only variable that influencing both viscous
pressure drop and yield pressure drop simultaneously. Meanwhile, the channel gap can
be considered as the most significant parameter, since, according to the equations, it
inverse-cubically determine the value of the viscous pressure drop as well as inverseproportionally influence the yield pressure drop. However, intrinsically, the influence
of channel gap in determining the pressure drop is even more significant since it also
determine the magnetic field strength. According to Wang et al. [24], the fluid channel
gap has magnetic resistance one thousand times than the resistance of the magnetic
materials of the valve core. Therefore, the alteration of gap size will dramatically
change the magnetic circuit performance which finally will influence the yield stress
value.
2.5.2
Dynamic Model
While the steady state modeling approaches for MR valve has been established,
the dynamic modeling technique of MR valve as a single component is rarely
disseminated. This can be understood, since most of the literatures were only focused
their modeling in the prediction of the valve peak performance. However, to fulfill
other purpose such as the control design process, the steady-state model only will
not adequately represent the device behavior. The necessity of dynamic modeling
of MR valve also can be considered higher than MR damper because the MR valve
28
and obviously the model can be used in the wider range of applications aside of
MR damper. For example, the knowledge of MR valve behavior and its virtual
representation can be used as a reference to design and predict the performance of
an MR damper or to help further development of new concept of actuators that require
MR fluids flow control [25, 128, 134].
In general, the dynamic modeling of an MR based device can be divided into

two approaches, the parametric modeling and non-parametric modeling. Parametric
modeling is the modeling approach with a collection of mechanical elements as
a hypothetic representation of the device so that the device behavior can be
mathematically modeled using the dynamic relationship between elements. Parametric
approach has the advantages due to the ability to provide a generalized form of the
model, which means that the same form can be used repeatedly in other devices.
Nevertheless, there are pitfalls in the accuracy of the results since the model ability
to follow the device behavior is limited and in order to mimic the characteristics
of a particular device in a good match, the involvement of a rigorous parameter
identification method is mandatory [135, 136]. On the contrary, the non-parametric
approach is able to avoid the problem in modeling accuracy. The non-parametric
modeling is approaching the modeling problem with the analytical expression to
describe the device characteristics based on the testing data [74]. However, there is
no generalized form since the model structure will be unique for each device and the
development is highly dependent on a specific testing data. Hence, the process flow to
generate the non-parametric model is typically more complicated than the parametric
model. Moreover, the validity of the generated model is only limited to a particular
device that is incorporated in the modeling process.
There are various kinds of parametric model that have been developed. For
example, Sakai et al. [137], proposed a simple parametric model based on the LuGre
friction operator as an inverse model of an MR damper as an actuator for vibration
control. The LuGre friction model also explored and modified by Jimenez and AvarezIcaza [138] to model an MR damper as a part of a semi-active control system. The other
type of parametric model, which also apparently the most popular parametric model,
known as the Bouc-Wen model has been extensively explored as the parametric model
of MR damper [136,139143]. The Bouc-Wen model approximate the device behavior
through a set of hypothetic spring and damper as shown in Figure 2.16.
Another example of parametric modeling approach for MR device is the class

of algebraic model, namely the hyperbolic tangent model, that was proposed by
29
Figure 2.16 Bouc-Wen model
Gavin et al. [144]. The method was then utilized by Kwok et al. [135] to model
an MR damper incorporating particle swarm optimization (PSO) as the parameter
identification method. Jiang and Christenson [145] also use similar model in a realtime hybrid simulation environment for a large scale MR damper. A performance
comparison study of various parametric modeling approach that has been conducted
by ahin et al. [146] reported that the class of algebraic model is able to provide better
performance fit than differential model. Although, according to comprehensive review
by Wang and Liao [74], the accuracy of the parametric model is not just rely on the
basic form of the model itself but also highly determined by the identification method
of the parameters. In general, Wang and Liao [74] inferred that model with higher
number of parameters and equipped with an appropriate identification method will
have better accuracy than the one with less parameters.
The other parametric method that is also often explored in the MR damper
model is the polynomial model. Choi et al. [147] introduced the approximation of the
force-velocity curve of an MR damper using polynomial equations. The method is
simple yet unique because there were two separated 6th order polynomial equations
for each positive and negative directions as shown in Figure 2.17. to compensate
the inequality of the force-velocity curve in positive and negative direction due to
hysteresis. The method was then applied by Ubaidillah et al. [148] in the modeling
approach of an MR damper for automotive semi-active suspension system. The
polynomial model is favored because it is convenient to use and easy to be adapted in
an open-loop system [149]. However, according to Sahin et al. [146], the polynomial
model has weakness in characterizing the device behavior in low velocity region due
to the absence of variables that characterize the pre-yield force.
30
Figure 2.17 Parametric hysteretic polynomial model [147]
Nevertheless, while the achievable accuracy of a parametric model is always

bounded by the model complexity, if the issue of model accuracy became more
stringent, the non-parametric model often chosen as a better option. There are several
methods that have been executed to model the MR damper behavior with nonparametric approach. One of the most extensively discussed is the artificial neural
network [150152]. The artificial neural network was often chosen as the base method
of non-parametric model due to its ability to automatically learn from the trend of
measurement data and generate an emulated functions afterward as depicted in Figure
2.18. The artificial neural network is also able to be combined with various learning
method such as lazy learning [153], and ANFIS [154].
2.6
Utilization of Magnetorheological Valve
Despite the lack of specific literature discussed about MR valve as a standalone device, in fact, it is the most vastly used MR devices. The MR valve has become
the most influential component in the MR damper, which is commercially the most
successful MR device. Recently, the development progress of new MR actuators also
stimulated the rapid development of MR valve. In this section, the utilization of MR
31
Figure 2.18 Artificial Neural Network model [151]
valve is elaborated in three different MR devices, namely the linear MR damper, the
rotary MR damper and the new MR actuator.
2.6.1
Linear Magnetorheological Damper
Linear MR damper is one of the most discussed MR devices so far. The range
of application of MR damper is widely ranged from automotive to civil applications.
According to the working modes, there are two basic contruction of MR damper. The
most common design is the valve mode MR damper with valve located in the piston as
shown in Figure 2.19. The other MR damper equipped with shear mode as depicted in
Figure 2.20.
Since both types of damper regulated flow using the narrow gaps in the piston,
these narrow gaps are functioned as the valve that is embedded in the damper piston.
Although these structures are the most common structure of MR damper due to its
compactness the fact that the coil should be wound in the piston has increased the
complexity of wire assembly as well as increasing the weight and thickness of the
piston head [14]. The internal coil placement also made the heat dissipation from the
coil more difficult and potentially increase the temperature of the MR fluid [58].
The alternatives of coil placement has been proposed by Chen et al. [155]
using external coil structure (see Figure 2.21). The construction of the damper not
just allow the heat dissipation to be directly released to the atmosphere but also made
it possible to add the self-powered and self-sensing features [157]. The coil also can
32
Figure 2.19 Valve mode MR damper [9]
Figure 2.20 Shear mode MR damper [75]
33
Figure 2.21 External coil MR damper [155]
Figure 2.22 MR damper with bifold valves [28]
be easily accessed for maintenance without having to interfere with the internal part of
the cylinder nor the fluid. However the basic construction of the external coil design
is quite complicated and the performance of the damper is apparently less than the
internal coil design due to difficulties in optimally routing the magnetic flux.
The other approach of coil placement are using the bifold valve design shown
in Figure 2.22 [28] where the coils are statically installed in both ends of the cylinder.
The bifold valve damper gain the advantage in design compactness similarly with the
damper with internal valve. Good damping performance was also reported especially
in high shock velocity. However, the design is also quite complicated and the valve
34
Figure 2.23 Bifold MR damper for high impulsive loads [156]
Figure 2.24 Bifold MR damper for shock vibration mitigation [158]
effective area is very limited and thus the performance expansion cannot be executed.
Since the design can provide advantage in high shock velocity, the design has been
explored for vibration mitigation application by Facey et al. [156] and Mao et al. [158]
as illustrated in Figure 2.23 and 2.24.
If the design simplicity and the ease of capacity expansion are the priority,
then the bypass structure can be a better option. The bypass structure where the valve
35
Figure 2.25 Basic structure of Bypass MR damper [8]
are separated from the piston and physically located outside the cylinder as depicted
in Figure 2.25 [8]. The separation between the cylinder and the valve has made the
assembly process of the bypass damper is far simpler than the other damper since
variety of commercially available hydraulic cylinder can be used. In the other words,
the customization requirement of damper components is very minimum. Though the
assembly process is simpler, the mounted valve outside the cylinder typically made the
size of bypass damper larger than the damper with internal valve. Nevertheless, most
of applications of bypass MR damper is on a large scale damping component such as
the seismic dampers shown in Figure 2.26 [60].
Figure 2.26 Bypass MR damper for large scale seismic application [60]
2.6.2
Rotary Magnetorheological Damper
The rotary MR damper is another type of MR damper that operates in angular

movement. For some applications that require large stroke, the rotary MR damper can
be a better option than linear damper since the risk of buckle is very minimum. The
36
rotary damper also more suitable in a hostile working environment since the damper
shaft in a rotary damper is sealed better than the damper rod in a linear damper.
However, the rotary MR damper that equipped with valve is so far is limited to the
vane type MR damper only. The other devices that can be classified as rotary MR
damper, e.g. MR brake, are typically not equipped with valve and operate mainly in
shear mode.
In terms of device advancement, unfortunately, vane type MR damper has not

been discussed in many publications. Only three written publications were found over
the last 10 years. The first one is the work of Zhang et al. [159], which adapted
the conventional structure of hydraulic vane damper as shown in Figure 2.27. In a
conventional hydraulic vane damper, the fluid would be pumped from one chamber
to another across gaps between the static vane and the rotary vane when the damper
rotates. In the proposed design, these gaps were utilized as valve, using the effective
area of shear mode by incorporating them with electromagnetic coil. In order to install
the electromagnetic coil, the static vane or in this case was called the clapboard, was
made in a hollow structure. Meanwhile, the effective area of shear mode of the valve
was made in an arc configuration and therefore it was called as an arc MR valve.
Figure 2.27 Vane type MR damper with arc valve [159]
Preventing leakages between chambers is the main challenge in the design of

a vane type damper. Therefore, the main strength of the proposed design in [159] is
its capability to use the potentially leaked gap between chambers into the flow control
zone. However, since the coil was located inside the damper, the generated heat from
the coil would be difficult to effectively dissipate. Maintenance of the damper is also
difficult, if there is a problem with the coil, the whole damper has to be dismantled.
Unfortunately, given that the intention of the publication was only to demonstrate
optimization analysis of the new vane type MR damper design, no disclosed results
37
on damper performance were presented.
The second publication of vane type MR damper was written by Giorgetti et

al. [32]. As shown in Figure 2.28, the second vane type MR damper, consisted of
moving vane, indicated by A, static valve body, B, the valve core, C, the vane damper
casing, D, and the electromagnetic coil seat, E. As with the previous design by Zhang
et al. in [159], it incorporated an internal valve design. However, in this design, the
valve was not located in the gaps between static and rotary vane, but was intentionally
made in the static vane body. The new configuration in [32] has changed the nature
of internal valve mechanism from shear mode to flow mode. Leakages that potentially
occurred at the gaps between the static and rotary vane still occur, but are neglected
since they are very small.
Figure 2.28 Vane type MR damper with outer coil valve [32]
Generally, the vane type MR damper design by Giorgetti et al. in [32] has
shown simplicity and compactness. The experimental results showed that the damper
could provide up to 200 Nm damping torque at 3 A current input in a feasible geometry
to be applied in passenger vehicle suspension. However, the incorporated internal valve
could have the same complexity in terms of coil maintenance as that described in [159].
Yang et al. [160] discussed a vane type MR damper design with flow type
internal valve. The design has shown similarities with the work done by Giorgetti et
al. in [32] in the context of annular flow mode valve design, where the effective areas
are located in line with the direction of fluid flow. The main difference was in the
placement of the coil, as shown in Figure 2.29. In Giorgetti et al. design [32], the coil
was located in the outer radial side of the fluid path, while in Yang et al. design [160],
the coil was located in the inner radial side of the fluid path.
38
Figure 2.29 Vane type MR damper with inner coil valve [160]
Since Yang et al. [160] put the coil in the inner radius, a lower resistance value
of the coil could be obtained leading to lower power consumption. The valves in the
designs of Yang et al. [160] and Giorgetti et al. [32] have similar dimension and number
of coil turns as well as the wire gauge, the coils in the Giorgetti et al. design [32] would
simply have higher resistance than the coils in the Yang et al. design [160]. Higher
resistance in coil means higher power consumption for the same current input, since
power relates proportionally to resistance and quadratically to current. Therefore, if it
is assumed that both valves have the same effective area and gap width, then both valve
would have the same performance but the Yang et al. design [160] would consume less
power.
The disclosed experimental results show that the achievable damping force
that can be generated by Yang et al. [160] could be up to 3 kN at 4A current input.
However, since the damping performance was evaluated in different measurement units
the results are difficult to be compare with those of Giorgetti et al. [32]. Giorgetti et
al. [32] demonstrated the damper performance in torque unit, while Yang et al. [160]
showed them in force unit. Although it is known that the achievable damping force
can be converted to torque by multiplying them with the length of rotating hub, the
length of rotating hub was not specifically mentioned in any part of their publications.
Nevertheless, logically, both designs should be in the same class in terms of achievable
damping torque, since both of them were designed for passenger vehicle suspension.
Both of them were also capable of providing almost the same ratio between on-state
and off-state damping at around 6.
39
2.6.3
New Magnetorheological-based Actuators
Aside from the device component, the MR valve also has been developed as a
stand-alone device. The stand-alone MR valve is an independent component that can
be used to retrofit the hydraulic valve or to complete a newly developed actuators.
There are several discussions about a new MR based actuator that can be functioned as
an active actuator similarly with a hydraulic or pneumatic actuator such as the bellowdriven motion controller shown in Figure 2.30. [19], the MR hydraulic power actuation
system in Figure 2.31 [25], the MR actuation system with embedded terfenol-D pump
in Figure 2.32 [26], and the MR based link manipulator in Figure 2.33 [134].
Figure 2.30 MR based bellow-driven motion control [19]
Figure 2.31 MR hydraulic power actuation system [25]
40
Figure 2.32 Actuation with embedded Terfenol-D pump [26]
Figure 2.33 MR based link manipulator [134]
One of the unique working application of the MR valve is the concept of 4/3
way directional MR valve proposed by Saloom and Samad [27] as depicted in Figure
2.34. The 4/3 way directional valve is one of the common functional hydraulic valve
that serve the hydraulic actuator and determine the direction of force actuation. The
4/3 way directional valve normally implemented in servo hydraulic actuation but in
order to operate in higher performance, a sophisticated valve actuation is required. The
design of 4/3 way directional MR valve was one of the proposed solution that utilizing
the rapid response of MR fluid to magnetic field. The proposed 4/3 way MR valve
concept was basically constructed of four single MR valves that each of the valve
element was arranged in between the valve ports. There are five ports in this valve
design, one inlet pressure port, two actuator ports, and two return ports. By activating
and adjusting the strength of magnetic field in different combination of annular valve,
the direction and rate of MR fluid flow can be regulated so that the actuation control of
force and direction of hydraulic can be achieved.
41
Figure 2.34 4/3 way directional MR valve [27]
2.7
The overview of MR fluids, MR valve structure, MR valve modeling technique

and the variety of MR valve applications have been presented. The explanations of MR
fluids have been elaborated based on its composition and its working modes. There
are many types of MR fluids but generally, researchers have made a clear distinction
of MR fluid from other types of magnetically responsive fluid from its micron-sized
particle. The MR valve explanations were conducted based on the categorization of
flow channel types, namely the annular MR valve, radial MR valve and combination
of annular and radial MR valve. The annular MR valve, as the most popular and
commonly used MR valve in the MR damper, has limitation in performance expansion
capability which promote the discussion of radial MR valve that has advantages in
multi-stage configuration. The most recent study that explored the MR valve with
combination of both annular and radial flow channel also has been covered in the
overview. The modeling overview of the MR valve was also presented and divided
into two parts, the steady-state and dynamic model. The steady-state model discussion
covered the mathematical approaches that often used to model the physical behavior of
the MR valve, while the dynamic model explored two general approach of hysteresis
modeling of MR devices, the parametric model and the non-parametric model. In
the final section of the chapter, various devices and applications that incorporates
MR valve as its main component were presented. According to the review, it can be
summarized that the current advancements of MR valve are still unable to improve the
achievable pressure drop in a compact valve size and low power consumption. Through
the review it was also shown that although the dynamic models of MR damper have
been widely studied, the dynamic models of MR valve, in particular, have never been
thoroughly discussed.
CHAPTER 3
MAGNETORHEOLOGICAL VALVE CONCEPT
3.1
Introduction
A clear understanding in MR valve behavior and characteristics are important

aspects in the design process of an MR valve. In this chapter, a methodology to design
a new concept of MR valves is carried out. The concept development process is divided
into two phases, the first phase is mainly related to the design consideration including
the magnetic circuit simulation while the second phase is related to the derivation of
the MR valve steady-state model to predict the performance of each valve design.
The Finite Element Method Magnetics (FEMM) version 4.2 is utilized to perform
the electromagnetic circuit design simulation for the MR valve, which results are
then combined with the steady-state mathematical model of MR valve to predict the
achievable pressure drop that can be generated by the valve. The results of performance
prediction are then used to select the valve design for experimental assessment.
3.2
Design of Magnetorheological Valve
The elaboration of MR valve design is needed to explain the technical

considerations that are taken to develop the concept. There are several aspects in
the design process of the MR valve that are critical such as the determination of the
material for the valve component, the strucural design consideration and the simulation
of the magnetic circuit of the MR valve.
43
3.2.1
Conceptual Design
There are two disclosed MR valve concepts that provided less than 50 mm
diameter MR valve with achievable pressure drop more than 1.5 MPa at 40 ml/s
flow rate. The first concept was proposed by Yoo and Wereley [20] using double coil
annular configuration while the second concept was introduced by Ai et al. [23] using
combination of annular and radial configuration which experimentally elaborated [24].
The double coil annular concept, as depicted previously in Figure 2.9, was proposed as
a high performance MR valve that can generate pressure drop up to 2 MPa at fluid flow
rate of 40 ml/s with outer diameter size of 25.4 mm and uses annular flow path channel
with internal coil attachment. The valve size still can be reduced up to 40% without any
performance degradation if high permeability material such as Hiperco alloy is used.
Meanwhile, the combination of annular and radial concept, shown in Figure 2.12, was
introduced as an effort to improve the performance of conventional MR valve design.
The first concept of MR valve with annular and radial flow path has outer diameter size
of 45 mm and capable to perform in similar pressure drop range with the double coil
annular MR valve.
Learning from these examples, the principles of pressure drop improvement in

the MR valve basically can be done by maximizing the yield stress of MR fluid in
a larger effective area. Effective area of MR fluid is the area where the rheological
properties of MR fluid can be effectively regulated by the activation of magnetic
field. Yield stress of MR fluid can be augmented by increasing the magnetic flux
density. However, magnetic flux density is highly related to the magnetic field intensity
from the coil and permeability of magnetic material that is used as a medium. If the
utilization of high permeability material is exempted, the common consequences of
increasing magnetic flux density are higher dimension of coil, which lead to larger size
of MR valve, and higher power consumption. The other way to increase yield stress of
MR fluid is by increasing the effective area by enlarging the valve dimension and/or
by extending the flow path of the MR fluid.
When the valve size matters, any improvement efforts which involve
enlargement of valve dimension are not preferable. For that reason, extending the
flow path of MR fluid is the feasible option. The combination of annular and radial
concept by Ai et al. [23] is a good example of effective area expansion by extending
the flow path. Another study by Nguyen et al. [72] has also confirmed that the MR
valve with combination of annular and radial gaps could provide higher achievable
pressure drop than other types of annular MR valve with the same outer radius and
44
power consumption. However, since the current design uses only a pair of annular
and radial gaps, the total effective area improvement is still very narrow; thus, the
enhancement of pressure drop that can be achieved is also limited. Therefore, in order
to expand the total effective area more, the length of flow path should be increased
using multiple combination of annular and radial gaps forming a meandering flow path.
The conducted study in this research will explore the effectiveness of the meandering
flow path concept to improve the pressure drop performance of the MR valve in the
sequence shown in Figure 3.1.
Figure 3.1 Concept assessment sequence of the new MR valve concept
The detailed structure of MR valve with meandering flow path that is explored
in this study is shown in Figure 3.2. The MR valve structure can be divided into
three components, the casing, the coil and the valve core. The casing resembles the
outer shell of the valve with two main functions. The first function is to secure the
whole structure of the valve together as well as to connect the valve with other devices
through its embedded fittings. The second function is to properly guide the magnetic
flux from the coil so that the loss of flux to the air, which reduce the efficiency, can be
minimized. The coil is the main component that generates and determines the magnetic
field strength of the MR valve. In this proposed MR valve structure, the coil is turned
in a coil bobbin that is also functioned as the wall of the flow channel. The core is the
component that formed the annular and radial flow channel, which can be divided into
four parts, the side core, the center core, the orifice core and the spacer.
45
Figure 3.2 Basic concept of MR valve with meandering flow path
3.2.2
Design Consideration
To facilitate the design process, the consideration of materials for each

component as well as the consideration of coil properties, valve size and fluid types
are needed. In order to select the component materials, the understanding of flux
routing concept is required. The flux path can be modified through the flux route
determination process by combining the materials with higher permeability (magnetic
materials) and the materials with lower permeability (non-magnetic materials) in such
a way so that the magnetic fluxes will penetrate the MR fluid channel. The common
understanding in magnetic circuit design for MR application usually recommends
penetration of magnetic fluxes in perpendicular direction to the direction of fluid flow.
Quantitatively, the flux routing process can be evaluated by calculating the magnetic
circuit performance for each combination of materials using analytical or numerical
analysis. However, in order to simplify the routing process, a qualitative evaluation of
the material selection and combination for the three main components of the proposed
MR valve is conducted as summarized in Table 3.1.
In Table 3.1, the combination summary of all possible material arrangement for
the three main components of the valve is shown. The main goal of these combinations
is to find the best material selection that capable to guide the fluxes passing through
46
No.
1
2
3
4
5
6
7
Table 3.1: Materials selection of valve component for valve routing

Valve components
Results
Casing
Coil bobbin
Core
Flux looped through the
Magnetic
Magnetic
Magnetic
coil bobbin and casing
Magnetic
Magnetic
Non-Magnetic
Flux looped through casing
and core with a high flux
Magnetic
Non-Magnetic
Magnetic
density in the fluid gaps
Magnetic
Non-Magnetic Non-Magnetic
Non-Magnetic
Magnetic
Magnetic
Non-Magnetic
Magnetic
Non-Magnetic
Flux looped through casing
and core with low flux
Non-Magnetic Non-Magnetic
Magnetic
density in the fluid gaps
Non-Magnetic Non-Magnetic Non-Magnetic coil bobbin and casing with
the lowest field intensity
the fluid gaps with sufficient magnetic field strength. From these eight possible
combinations, only two combinations show the possibility of flux to pass through the
fluid flow gaps. However, only the combination of magnetic casing and core with nonmagnetic coil bobbin is capable to provide a high magnetic field strength in the fluid
gaps. From the remarks in Table 3.1, it can be concluded that in order to successfully
guide the fluxes to pass through the fluid gaps, the permeability of the core component
should be higher than the permeability of the coil bobbin. However, to get a higher
magnetic field strength in the fluid gaps, the permeability of the casing should be at
least similar to the permeability of the core.
In this study, the selected material of the casing is the mild steel with properties
that compatible with the AISI 1010. The design of the casing consists of two identical
parts, which each part has four threaded holes as the holder for the locking bolts and
one female BSPP (British Standard Pipe Parallel) 1/4 fluid channel port. Since the
coil bobbin has to be made from the low permeability material, aluminum is chosen
for the material of the coil bobbin. The coil bobbin is also designed with a pair of
grooves for O-rings installation and has the threaded holes in pair to the holes in the
casing. Meanwhile, the valve core is separated into two side cores, two orifice cores, a
center core and six spacers. All parts in the valve core are also made from mild steel
47
that comply with AISI 1010 material except for the spacers, which are made from
aluminum.
The outer diameter size of the valve is determined of 50 mm with overall length
of 77 mm. On the other side, the inner diameter size, that is the inner diameter of
the coil bobbin that becomes the bore size of the flow channel is determined of 15
mm. The core parts are then arranged to form the annular and radial gaps alternately
in such a way it meanders the flow channel. As one of the subject of the study, the
size of annular and radial gaps are both varied from 0.5 to 2.0 mm. Since the valve
is designed to generate high pressure drop, the casing should be strong enough to
withstand such pressure and for that purpose the thickness of the casing is determined
of 5.0 mm. As a consequence, the space available for the coil winding that is made from
23 SWG copper wire is sufficient for around 545 turns with total resistance of around
3.03 . In this study, the current input is limited to 1.0 A therefore the maximum power
consumption of the valve is also limited to around 3.03 W.
3.2.3
Magnetic Simulation
In order to predict the effect of current input to the MR effect strength of

MR fluid in the effective area, the permeability (B-H curve) and the yield stress, ,
need to be determined. There are specific B-H curve and curve for each type of
MR fluid. In this study, the MR fluid that will be used is the Lord Corporationss
MRF-132DG, which curves was already approximated by Nguyen et al. in [72],
as depicted in Figure 3.3, using polynomial approximation as a function of B in the
following function:
(B) = 52.962B 4 176.51B 3 + 158.79B 2 + 13.708B + 0.1442
(3.1)
where B is the magnetic flux density in Tesla which is highly determined by current
input, coil turns and dimension, and total reluctance of the medium. The permeability
curve of the MRF-132DG, together with the other permeability value of the materials
used in the MR valve can be used in the calculation to determine the magnetic flux
density, B, at the effective area. Since the number of turns of the coil is already
determined and the permeability of magnetic material was assumed to follow the B-H
curves of AISI 1010, the magnitude of magnetic flux density can be determined by
the magnitude of current input supplied to the coil. However, due to the complexity
of magnetic structure and non-linear permeability of the materials, the magnitude of
48
magnetic flux density for each zone is difficult to be calculated analytically [111].
Thus, in order to determine the value of B for each corresponding current input, a
finite element method based software for magnetic simulation called FEMM is used.
The results from FEMM are shown in Figure 3.4 and 3.5.
Figure 3.3 Approximation of yield stress as a function of magnetic flux density, B
Figure 3.4 Two-dimensional axisymmetric model of the MR valve in FEMM
Figure 3.4 shows two dimensional axisymmetric meshed model in the FEMM
using triangular element with total elements number of 103771 and total nodes number
of 52281. Figure 3.5 shows the contour and flux lines of magnetic flux density for the
49
Figure 3.5 Flux lines and contour of magnetic field of the MR valve in FEMM
whole magnetic circuit. Notes should be taken that the flux that can influence yield
stress of MR fluid is only the fluxes that passed through MR fluid gaps. The MR fluid
gaps area that crossed by the magnetic flux is known as the effective area. According
to the result in Figure 3.5, it can be observed that the flux lines have passed across the
outer annular gap, outer radial gap and inner radial gap. Therefore, only MR fluid that
passed through these three gaps can be rheologically influenced by the changing of the
current input. Meanwhile, the inner annular and orifice gaps are not intensively exposed
to magnetic field and therefore the influence of magnetic field to yield stress of MR
fluid in these areas can be neglected. Yet, although the inner annular and orifice gaps
do not provide field-dependent pressure drop, these gaps still have the contribution to
the viscous pressure drop. Nevertheless, more focus should be taken at the flux lines
in the outer annular gap and outer radial gap since the structural configuration of outer
annular and outer radial gaps have made these gaps formed as a parallel channel for the
flux lines. Thus if flux-loss is neglected, the total number of magnetic flux that passed
through the outer annular and outer radial gaps are equal to the number of magnetic
flux that passed through the inner radial gap. This is why the inner radial gaps have
higher magnetic flux density than the outer annular gaps or the outer radial gaps. The
difference of magnetic flux density between the outer radial gaps and the inner radial
gaps is also the reason why the radial gaps are divided into the outer and inner zones.
The distribution of magnetic flux density value along the flow path channel for
0.5 mm gap size is shown in Figure 3.6 where the flux density is seen to be increased
to the augmentation of current input to the coil. In this study, the maximum current
50
Figure 3.6 Magnetic flux density along MR fluid flow path for 0.5 mm gap size with
respect to various current input
input is limited to 1.0 A while the maximum flux density that can be reached along the
flow path channel is around 0.7 Tesla at the inner radial gaps. As a result of parallel
magnetic channel, the second highest flux density is reached at the outer radial gaps
of around 0.5 Tesla while the flux density of the outer annular gaps is only around 0.2
Tesla. The magnetic flux density values for each corresponding current input is used to
predict the yield stress of MR fluid using Equation 3.1. Therefore, from these results,
it can be predicted that the pressure drop of the inner radial zone will be the most
significant contributor to the total MR valve pressure.
3.3
Steady-state Modeling of Magnetorheological Valve
The assessment of MR valve performance is done using both simulation

and experimental work. The simulation of electromagnetic circuit is one of the
most important phases in the simulation work and is done using Finite Element
Method Magnetics (FEMM) software, an open-source software package for magnetic
simulation. The derivation of quasi-steady phenomenological model of MR valve
design is also important in the first stage of simulation to predict the performance
of the MR valve. Typically, the quasi-steady model of the valve will be expressed in
terms of the achievable pressure drop, which is consist of two parts, pressure drop from
51
viscous properties of the fluid and pressure drop from the field dependent yield stress
of the fluid. The basic expression of pressure drop in an MR valve can be declared by
following quasi-steady equations [24]:
P = Pviscous + Pyield
Pviscous =
Pyield =
(3.2)
6QL
d3 R
(3.3)
c (B) L
d
(3.4)
where expressed that the pressure drop from viscous properties in Equation 3.3 is
proportionally related to the fluid base viscosity (), flow rate (Q), and annular channel
length of the valve (L) but inverse-cubical to the valve gaps (d) and inverse proportional
to the channel radius (R). Meanwhile, the field dependent pressure drop in Equation
3.4 is proportionally associated with the field dependent yield stress value ( ) of MR
fluid, annular channel length (L), and flow-velocity profile coefficient (c) but inverseproportionally related to the gap size (d). The coefficient c is obtained by calculating
the ratio between field dependent pressure drop and viscous pressure drop using the
approximation function as defined by [72] in the following equation:
c = 2.07 +
12Q
12Q + 0.8Rd2 (B)
(3.5)
However, the expressions in Equations 3.3 and 3.4 are only valid for MR valve
with annular gaps. For radial gaps, the viscous pressure drop and the yield pressure
drop can be expressed as [24]:
Pviscous
6Q
=
ln
d3
Pyield =
Ro
Ri
c (B)
(Ro Ri )
d
(3.6)
(3.7)
where the valve gap size (d) in Equations 3.6 and 3.7 refers to radial gaps while the
and refer to the outer radius of the radial gaps and the inner radius of the radial gaps
respectively.
The mathematical expression for the orifice gaps is slightly different, because
no field dependent yield stress is incorporated into the equation since the fluxes did not
pass through the orifice gaps. Therefore, the pressure drop equation for the orifice gaps
52
is only expressed by the equation of viscous resistance of the fluid [22].
P =
8QL
R 4
(3.8)
Since the proposed MR valve design has both annular and radial gaps, the
mathematical expressions of pressure drop should include all the expressions in
Equations 3.3 to 3.8. However, to make the derivation simpler, the valve gaps is
separated into five different zones, the outer annular gaps zone, the outer radial gaps
zone, the inner annular gaps zone, the inner radial gaps zone, and the orifice gaps zone
as shown in Figure 3.7. The valve needs to be clustered into five zones because each
zone was presumed to have different magnetic flux density. These five zones also could
also be divided into two categories, the zones that categorized as effective area and the
zones that have only viscous resistance. The zones that categorized as effective area
are the outer annular gaps zone, the outer radial gaps zone, and the inner radial gaps
zone while the zones that have only viscous resistance are the inner annular zone and
the orifice zone.
Figure 3.7 Gaps zone in MR valve with multiple annular and radial gaps
From Figure 3.7, it can be seen that there are two outer annular gaps, two
outer radial gaps, three inner annular gaps, four inner radial gaps and two orifice gaps.
Using the expressions of annular and radial pressure drop in Equation 3.3 to 3.8, the
quasi-steady pressure drop of MR valve with multiple annular and radial gaps can be
53
described in the following equations:
Pvalve = Pannular outer + Pradial outer + Pannular inner
+ Pradial inner + Porif ice

Pannular outer
ca a (B) Lao
6QLao
+
=2
3
dao Rao
dao

Pradial outer
6Q
=2
ln
d3ro
Pannular inner = 3
6Q
=4
ln
d3ri
Porif ice = 2
R1
R0 outer

cro ro (B)
+
(R1 R0 outer )
dro
6QLai
d3ai Rai

Pradial inner
(3.9)
(3.10)
(3.11)
(3.12)

R1
R0 inner

cri ri (B)
+
(R1 R0 inner )
dri
8QLo
R0 inner 4
(3.13)
(3.14)
The complete parameters for the MR valve with multiple annular and radial
gaps are shown in Table 3.2 while the dimensions and the specific location of each
corresponding variable of the proposed MR valve are shown in Figure 3.8.
Table 3.2: List of MR valve parameter

Parameter
Unit Value
(MRF-132DG) Pa s 0.092
Q
ml/s 40-80
dao
mm
0.5
dr = dr o = dr
mm
0.5
dai
mm
1.0
Lao
mm
10.0
Lai
mm 6.0-2dr
Lo
mm
5.0
R1
mm
6.5
mm
3.0
R0 outer
R0 inner
mm
2.5
54
Figure 3.8 Dimension and variables of MR valve
3.4
Simulation of Magnetorheological valve Performance
Figure 3.9 illustrates the results of maximum achievable pressure drop of the
multiple annular and radial MR valve at 40 ml/s flow rate with gap size of 0.5 mm
and current input 1.0 A is around 5.3 MPa. As a comparison, the achievable pressure
drop of the previous design of compact MR valve at 40 ml/s flow rate is only around 2
MPa [20,24]. In other words, the proposed MR valve design has shown the contribution
to improve the achievable pressure drop of a compact MR valve. Furthermore, it is
shown that the pressure drop from the inner radial gap zone is the highest contributor to
the total achievable pressure drop of the MR valve. Surprisingly, although the magnetic
flux density of the outer annular zone was already known to be the lowest from the
other effective area, the maximum pressure drop of the outer annular zone is higher
than the outer radial zone. Meanwhile, the total viscous pressure drop of the MR valve
is shown to be nearly 10 MPa which mostly contributed from the outer annular zone,
followed by the inner radial zone and the outer radial zone. Since the contribution of
viscous pressure drop from the inner annular zone and the orifice zone are relatively
small, these zones are not included in the graph in Figure 3.9. The percentage of
contribution from each zone to the MR valve pressure drop is illustrated in Figure
3.10.
55
Figure 3.9 Estimation of achievable pressure drop of MR valve with 0.5 mm gap size
The illustrations of pressure drop percentages from each zone in Figure 3.10
can be used to explain why the pressure drop at the outer annular zone can outperform
the outer radial zone although the magnetic flux density in the outer annular zone is
lower than the outer radial zone. The viscous pressure drop of outer annular zone has
been shown to be the highest contributor of viscous pressure drop of the MR valve. The
magnitude of viscous pressure drop at the outer annular zone has contributed more
than half of the total viscous pressure drop of MR valve while the outer radial and
the inner radial zone have only contributed around 11% and 28% respectively. The
channel length, channel radius, gap size and fluid flow rate, as shown in Equation 3.10,
(a)
(b)
Figure 3.10 Percentage of pressure drop contribution from each zone (a) viscous (b)
field-dependent at 1 A current input
56
are the variables that responsible in determining the magnitude of the viscous pressure
drop at the outer annular zone. However, among them, gap size is known as the most
significant variable since it inverse-cubically influences the viscous pressure drop.
As the most significant variable that determines the viscous pressure drop, the
effect of gap size variations on the pressure drop of MR valve should be evaluated.
Since the flow path has been divided into five zones, there are five different gaps to be
varied. However, to simplify the combinations, the gap size for both outer radial and
inner radial zones are determined to be equal (dro = dri ) and varied from 0.5 mm to
2.0 mm. The same variations and interval are also used at the outer annular gap size.
Meanwhile, for zones that have only viscous resistance such as the inner annular gap
and the orifice gap, the gap size is kept constant. The effect of each combination on the
viscous and field dependent pressure drop of MR valve are shown in Figure 3.11:
Figure 3.11a and 3.11b show that the gap size has a very significant effect
to the magnitude of viscous pressure drop as well as the magnitude of field-dependent
pressure drop. However, the most significant influence of the annular gap to the viscous
pressure drop of the MR valve is seen in the interval between 0.50 to 0.75 mm where
the viscous pressure drop of the MR valve could be reduced in nearly 0.4 MPa.
Obviously, the change of annular gap also has some effect to the magnitude of fielddependent pressure drop. Figure 3.11b shows that the increase of annular gap size
from 0.50 to 0.75 mm reduces the field-dependent pressure drop of around 0.8 MPa.
Meanwhile, the enlargement of radial gap seems to have more influential effect to the
reduction of field-dependent pressure drop magnitude than to the viscous pressure drop
magnitude. From Figure 3.11b, it can be observed that increasing radial gap size from
0.50 to 0.75 mm reduces the viscous pressure drop of only 0.25 MPa but responsible
to the decrease of field-dependent pressure drop for almost 2.5 MPa.
The results in Figure 3.11 also illustrate an interesting fact regarding

relationship between gap size and the magnitude of field-dependent pressure drop.
As shown in Equations 3.10, 3.11 and 3.13, the relationship rule between gap size and
field-dependent pressure drop are inverse-proportional while the relationship between
gap size and viscous pressure drop are inverse-cubical. Therefore, any modifications to
the gap size supposed to have more significant effect to the viscous pressure drop than
the field-dependent pressure drop. However, the results in the Figure 3.11 show that the
effects of gap changing to the pressure drop apparently are not consistently followed
the rule stated in Equations 3.10, 3.11 and 3.13. For example, the radial gap reduction
from 0.50 to 0.75 mm degrades the field dependent pressure drop around 40% but
57
(a)
(b)
Figure 3.11 Effect of gap size on the pressure drop (a) viscous (b) field-dependent at
1 A current input
58
only reduces viscous pressure drop about 27%. The reason of this can be explained
by considering the effect of gap size changing to the magnetic circuit performance.
Since the magnetic resistance of MR fluid in the gaps is a lot higher than the magnetic
resistance of magnetic material in casings and cores, small change in the gap size will
significantly change the magnetic circuit and influence the magnetic flux density [24].
Therefore, although in Equations 3.10, 3.11 and 3.13 the gap size seems to only
inverse-proportionally related to the field-dependent pressure drop, in actual, the gap
size also influences the field-dependent pressure drop by simultaneously changing the
magnitude of field-dependent yield-stress of MR fluid.
However, in terms of overall MR valve performance, high achievable fielddependent pressure drop only is not sufficient to show that the MR valve is working
in a decent performance. Since, MR valve with high field-dependent pressure drop
but also with high viscous pressure drop value means that the operational range of MR
valve is low. The operational range of MR valve can be determined using the following
expression.
=
Pyield
Pviscous
(3.15)
Equation 3.15 expressed the operational range in terms of amplification ratio of

pressure drop that can be generated using MR valve. For instance, it can be said that the
operational range is high if the field-dependent pressure drop is high and the viscous
pressure drop is low while the operational range is low if the field-dependent pressure
drop is low and the viscous pressure drop is high. The expression is an inverse equation
to the valve ratio term in [161] which defined as the ratio of the viscous pressure drop
to the field dependent pressure drop of the MR valve. As it is desirable to have a small
valve ratio, it is also desirable to have a large the operational range of MR valve. The
expression in Equation 3.15 is also similar with the non-dimensional parameter called
Bingham number which defined as the ratio of the dynamic field-dependent yield stress
to the viscous shear stress [133]. Bingham number is a very useful parameter and has
been used as a generic term to analyze the performance range of various MR devices
such as dampers and valves [162164]. However, in order to calculate the Bingham
number, specific value of yield stress, fluid velocity and gap size are required which
makes Bingham number calculation for MR valve with gap types combination more
difficult than by just comparing the achievable pressure drop.
From Figure 3.12, it can be observed that the operational range of the valve
59
Figure 3.12 Comparison of operational range between various gap configurations
consistently increased when the gap size at the outer annular zone is increased. The
most significant operational range improvement occurred when the outer annular gap
size is increased in the interval between 0.50 to 0.75 mm. Since the increase of
annular gap size provides more significant impact to the viscous pressure drop than
the field-dependent pressure drop, consequently, the occurrence of operational range
augmentation in this interval can be considered as a result of significant viscous
pressure drop reduction. In the contrary, it is clear that the increase of radial gap
size will have inverse relationship with the operational range since the increase of
radial gap size will cause significant degradation to the field-dependent pressure drop.
However, some exceptions are found for the combinations of 1.0, 1.5 and 2.0 mm
annular gap sizes with radial gap size in the interval between 0.50 to 0.75 mm. The
anomaly of trend in these intervals appears because the performance degradation of
the viscous pressure drop in these gap combinations are higher than the performance
degradation of the field-dependent pressure drop. The higher viscous pressure drop
degradation is also shown to be increased in the combination with higher annular gap
size. However, as the size of radial gap increased, the dominance of field-dependent
pressure drop degradation is started to overlap the degradation of viscous pressure
drop. Nevertheless, it still can be concluded that the increase of annular gap size
generally has positive effect to the operational range while the increase of radial gap
size tend to reduce the operational range. In this case, due to trend anomaly and
significant impact to the change of pressure drop value, the gap size interval in between
0.5 to 1.0 mm require more attention and will be further explored in the experimental
60
assessment.
As a benchmark with previous MR valve concept, Table 3.3 show a comparison

between the performance of the meandering flow path MR valve with the other types
of MR valve. The source of the comparison are combined from several literatures
[21,24,165, 166] and compared in terms of the ratio between the maximum achievable
pressure drop of each valve with its corresponding power consumption and gross valve
size (volume). The valve performance can be considered more efficient if the ratio
of pressure drop to power and volume are higher. According to the table it can be
concluded that the MR valve with meandering flow path structure has a very significant
improvement over the annular and radial types of MR valve in terms of pressure
drop per power value (more than 100 times improvement). The meandering flow path
concept also show enhancement over the annular-radial concept with nearly 150%
improvement of the pressure drop to volume ratio.
Table 3.3: Performance benchmarking between the proposed MR valve concept and
the counterparts
200 kN
400 kN
250 kN
AnnularProposed
Specification MR damper MR damper MR damper
Radial MR
MR valve
[165]
[166]
[21]
valve [24]
Valve type
Annular
Annular
Radial
Annular-Radial Meandering
Maximum
pressure
3.8
16.1
9.6
2.5
5.3
drop (MPa)
Maximum
Power (W)
262
1920
900
-NA3
Pressure
drop per
Power input
14.5
8.39
10.7
-NA1766
(kPa/W)
Gross
Volume
1
31
19
0.1
0.15
(dm3 )
Pressure
drop per
Volume
3.8
0.52
0.48
24.87
35.08
3
(kPa/cm )
61
3.5
A new concept for a MR valve with meandering flow path was presented in
this chapter. The new MR valve is presented by elaborating the conceptual design
of the valve followed by the explanation about the consideration of the design
and magnetic field simulation. The derivation of the MR valve steady-state model
and the performance prediction of the MR valve were also presented in separate
section. The novelty of the new MR valve concept is mainly justified by the unique
arrangement of the effective area which makes it capable to deliver higher pressure
drop capacity than its counterparts. The steady-state modeling and simulation work of
the achievable pressure drop in accordance to the viscous and field-dependent pressure
drop contributions have been conducted to predict the achievable performance of the
MR valve. The effect of gap size selection in the annular and radial channel to the
magnitude of viscous and field-dependent pressure drop also has been analyzed to
evaluate the significance of each channel to the overall performance of the MR valve.
According to the assessment results, the outer annular channel was shown as the most
influential region to the characteristics of viscous pressure drop. Meanwhile, the inner
radial channel was more influential to determine the characteristics of field-dependent
pressure drop. The assessments of the gap size effect in each regions were also
analyzed in terms of dynamic range of MR valve. The gap size of annular channel has
been reported to have positive correlation with the dynamic range of MR valve while
the gap size of radial channel were reported to have inverse effect. The knowledge of
these characteristics is important as a reference to perform the experimental testing
that will be discussed in the next chapter.
CHAPTER 4
EXPERIMENTAL ASSESSMENT
4.1
Introduction
The advantage of the meandering flow path concept has been showed in
Chapter 3 through the comparison between the achievable pressure drop of the
meandering valve concept with its predecessors, which typically has maximum
pressure drop up to 2.5 MPa. In this chapter, the concept will be experimentally
explored to prove and demonstrate its performance capability in improving pressure
drop. This chapter is arranged in three sections, which is started with the explanation
about the experimental apparatus, followed by the elaboration of the experimental setup, which explains the measurement methodology, and finally the discussion about
the experimental results. The chapter is discussed in four different subsections, the
off-state characteristics, the on-state characteristics, the analysis about the frequency
excitations effect, and the analysis about the current input effect. This chapter is ended
by the last section that concludes the chapter.
4.2
Experimental Apparatus
4.2.1
The specific MR fluid that is used in this study is the MRF-132DG, an MR

fluid made by Lord Corporation for general use in controllable, energy-dissipating
applications such as shocks, dampers and brakes [167]. The fluid consists of
micron-sized magnetizable particles in hydrocarbon fluid. According to [168], the
magnetizable particles contained in the fluid are between 1 to 20 microns with weight
ratio of 80.98%. The typical properties and material compatibilities of MRF-132DG
63
are listed in Table 4.1 while its magnetic properties, presented in the B-H curve, and
the curve are shown in Figure 4.1 and 4.2 respectively.
Table 4.1: Typical properties and material compatibility of MRF-132DG [167, 168]
FLUID PROPERTIES
Base Fluid
Hydrocarbon
Solid Content by Weight, %
80.98
3
Density, g/cm (lb/gal)
2.95-3.15 (24.6-26.3)
Operating Temperature, C ( F)
-40 to 130 (-40 to 266)
Flash Point, C ( F)
>150 (>302)
Viscosity, Pa-s @ 40 C (104 F)

0.092 0.015
calculated as slope 800-1200 sec1
MATERIAL COMPATIBILITIES
Buna N (Nitrile)
Good
Butyl
Poor
EPDM/EPR
Poor
Fluoroelastomer
Good
Natural Rubber
Poor
Neoprene
Good
Silicone
Fair
Iron
Good
Stainless Steel
Good
Aluminum
Good
Polyurethane
Good
Figure 4.1 B-H curve of the MRF-132DG [167]
64
Figure 4.2 Field induced yield stress of the MRF-132DG [167]
4.2.2
As mentioned earlier in Section 3.2.1, the MR valve structure is divided into

three main components, which are the casing, the coil and the valve core. The casing,
which should be made from magnetic material, is made from mild steel that compatible
with AISI 1010 and consists of two identical parts, which each of them also has a
fluid channel with female BSPP 1/4port and four holes for locking bolts. The coil
consists of copper wire windings and an aluminum coil bobbin, which has a groove
on each side for O-ring installation as a sealing mechanism. The coil bobbin also has
four M2 threaded holes in each side as a bolt-locking mechanism with the casing.
Viton is selected as the O-ring material rather than rubber due to poor compatibility of
rubber with hydrocarbon as the carrier liquid of the MRF-132DG (see Table 4.1). The
selection of Viton is referred to several references [50, 169], which utilized the same
material for their sealing for hydrocarbon based MR fluid. The last component is the
valve core which consists of several parts namely two mild steel side cores, two mild
steel orifice cores, a mild steel centre core and six aluminum spacers. The spacers are
used to maintain the clearance between core parts. The exploded view of the MR valve
design is illustrated in Figure 4.3.
In order to experimentally explore the achievable pressure drop of the MR valve

with meandering flow path, only four different gap size configurations of the MR valve
that are experimentally assessed in this work, namely the MR valve with with 0.5 mm
65
Figure 4.3 Exploded view of the MR valve prototype
radial and 0.5 mm annular gaps, the 0.5 mm radial and 1 mm annular gaps, the 1
mm radial and 0.5 annular gaps, and the 1 mm radial and 1 mm annular gaps. These
gap configurations, according to the simulation results in Section 3.4, were reported to
potentially have the highest achievable pressure drop.
Figure 4.4 Failure of the bolt-locking mechanism to withstand internal pressure
66
However, after some initial tests, it was found that the design has weakness
in the strength of bolt-locking mechanism between the casing and the coil bobbin.
As shown in Figure 4.4, the M2 female thread in the aluminum bobbin is apparently
not strong enough to hold the casings together under high internal fluid pressure which
caused failure. Therefore, the modification is made to the design, especially to improve
the strength of casings assembly joint. The modifications is made by changing the boltlocking mechanism from the casings to the aluminum bobbin to a directly-coupled
thread between casings as shown in Figure 4.5. With the modified design, better
strength of casing assembly can be achieved and less component is required in the
valve structure.
(a)
(b)
Figure 4.5 Modification and comparison of the MR valve prototype(a) Exploded view
of the MR valve design (b) Fabricated prototype of MR valve
67
4.2.3
Testing cell
MR fluid flow need to be induced acrossed the MR valve,in order to observe

the dynamic behavior. There are various methods that have been used by researchers
to induce flow of MR fluid such as demonstrated in [19, 24, 170], however, the method
chosen to induce fluid flow is similar with the method used in [20], where the MR valve
is installed in the bypass channel of a hydraulic cylinder, which is fully filled with
MR fluid. The fluid flow was then induced by introducing movement to the hydraulic
cylinder, which act as the MR valve testing cell, using dynamic test machine. In this
study, the testing cell is made from a double rod hydraulic cylinder with maximum
stroke length of 70 mm, bore size of 30 mm and rod diameter of 18 mm (the net
piston area is around 452.4 mm2 ). The double rod cylinder is used as the testing cell
to eliminate the requirement of accumulator since no volume compensation is needed
during operation. The prototype of MR valve with meandering flow path is installed in
the testing cell as shown in Figure 4.6.
Figure 4.6 MR valve installation in the testing cell
For instance, the arrangement of the MR valve in the testing cell has made the
device looks similar with a bypass linear damper. The main difference, in this case, is
that the dynamic force-velocity behavior of the testing cell is not the point of interests
since the behavior of the valve is mainly related to the valve pressure drop and the fluid
68
volumetric flow rate. In order to measure the valve pressure drop, two pressure sensors
are installed in the nearest point to both MR valve ports to ensure that the pressure
drop effect from the flow conduits does not alter the pressure measurements. The
volumetric flow rate is measured indirectly by multiplying the measured velocity of
the piston and the effective piston area. The velocity data is obtained by differentiating
the displacement data of the piston that is measured by the native displacement sensor
of the dynamic test machine.
4.3
Experimental Set-up
A simultaneous measurement of pressure drop generated by valve in

accordance to the variations of MR fluids flow rate and magnetic field is conducted
to measure the performance of the MR valve. In order to generate the flow, the testing
cell is actuated by the dynamic test machine and the variation of fluid flow can be
recorded through the measured stroke of the testing cell. Meanwhile, the variation of
magnetic field is represented by the changing of the current input to the coil.
The experimental setup for the MR valve performance measurement is shown

in Figure 4.7 where the pressure drop of the MR valve is measured as well as the
displacement of the testing cell. The measured displacement data then converted to
flow rate by differentiating and multiplying the data with the piston net cross-sectional
area. The realization of these setup is shown in Figure 4.8 where the MR valve is
installed in the testing cell with pressure sensors (Parker Hannifin PTDSB1001B1C1,
0-10 MPa) tapped in each port of the valve. An in-house steel platform powered by
a geared electric AC induction motor equipped with an adjustable crank to generate
sinusoidal movement in the platform. An inverter is used to control the rotational
speed of the electric motor while a wire displacement transducer (Celesco SP2-25)
measures the length of platform displacement. All measurement data from the sensors
are connected to the computer through the NI cDAQ-9174 and LabView.
The platform is used to provide the oscillatory movement to the testing cell
which will result in pressure changing inside the cylinder and will periodically pump
the MR fluid to flow through the MR valve. The pressure sensor measures the realtime changing of the fluid pressure in each port of the valve. The pressure difference
between the inlet port pressure and the outlet port pressure represents the MR valve
pressure drop. In order to form the relationship between pressure drop and flow
69
Figure 4.7 Experimental arrangement schematic for the MR valve testing
rate as well as the current input, the measurement was conducted in three different
inverter-controlled frequencies namely 5.0 Hz, 7.5 Hz, and 10 Hz, which, according
to the measurement, are approximately equal to peak flow rate of 37 ml/s, 55 ml/s
and 73 ml/s. The peak flow rate and its associated pressure drop will be plotted to
form the relation. The approach is similar with the common method in any damper
characterization process to form the force-velocity curve [154].
Since the force induced to the testing cell is also proportional to the pressure
drop of the valve, the larger the pressure drop is, the higher the force required to
move the testing cell. Unfortunately, the in-house dynamic test machine, which use
the inverter-controlled AC motor as the prime-mover of the platform, is not able to
provide sufficient force to move the testing cell in low frequency excitation. The
speed reduction of the motor with inverter control is proportional to the reduction
of motor power, which obviously decrease the torque of the motor. Therefore, even
when the platform is moving, it is difficult to keep the excited frequency in constant
value, especially during the increased pressure drop of the MR valve. As a result, the
measurement range of the in-house dynamic test machine is very limited.
Due to these limitations, the results of the in-house test machine will only
be used to evaluate the peak off-state pressure drop of the valve and the peak onstate pressure drop. The dynamic pressure drop, on the other hand, is evaluated with
70
Figure 4.8 MR valve testing set-up using in-house test machine
Shimadzu Hydraulically actuated Fatigue Dynamic Test Machine. The machine is

equipped with a 20 kN force sensor and a displacement sensor. The installation of
the testing cell in the Shimadzu Hydraulically-Actuated Fatigue Dynamic Machine
is depicted in Figure 4.9. Similarly with the installation arrangement in the in-house
test machine, the alternating pressure-induced flow is generated every time the rod is
compressed and extended by the dynamic machine. The magnitude of the flow rate is
assumed to be proportional to the velocity of the piston strokes so that the velocity of
the actuation can represent the flow rate variation.
During experimental testing with the Fatigue Dynamic Test Machine, the
testing cell is subjected to a sinusoidal wave, where one end of the testing cell is
fixed and the other end is actuated. Since the sinusoidal wave can be expressed in
u = A sin (2f t)), the amplitude, A, is fixed to +/- 25 mm, while the excitation
frequency, f , is varied from 0.50 to 1.00 Hz. In order to demonstrate the MR effect,
the magnetic field strength is also varied by the changing the magnitude of the current
input. The summary of variable arrangement for the experimental test is shown in Table
4.2.
71
Figure 4.9 Testing cell installation in the Shimadzu Fatigue Dynamic Test Machine
Table 4.2: The variable arrangement of experimental test using Shimadzu Fatigue
Dynamic Test Machine
Excitation Frequency Excitation Amplitude Current Input Cycles
(Hz)
(mm)
(Ampere)
0.50
25
0-1
25
0.75
(interval 0.1)
1.00
4.4
Experimental Results
4.4.1
Off-state and On-state Pressure Drop Characteristics
The off-state pressure drop can be defined as the pressure drop generated by the
MR valve when the electromagnet is charged with 0.0 A current. In the other words,
the off-state pressure drop is the pressure drop with only viscous contribution without
any magnetorheological effect from the MR fluid. The peak value of the measured off-
72
state pressure drop across the MR valve with meandering flow path as a function of
flow rate is shown in Figure 4.10.
Figure 4.10 Comparison of measured and theoretical off-state peak pressure drop at
various flow rates for 0.5-0.5 mm (annular-radial) gaps configuration
Figure 4.10 shows that generally the trend of the theoretical off-state pressure
drop value is in accordance with the measured one, in the range of flow rate
measurement between 37-73 ml/s, the pressure drop increased proportionally to the
flow rate. In specific, the exact value of the measured off-state pressure drop is slightly
lower than the theoretical off-state pressure drop from the model if the assumed
viscosity value is equal to 0.092 Pa s, which is the specified viscosity value of the
MRF-132DG from the manufacturer. However, since according to the documentation
from Lord Corp. [168], the exact viscosity value also can be varied up to +/- 0.015
Pa s from the specified value, thus the theoretical off-state pressure drop value also
can be dropped lower than the measured value if lower viscosity value is selected
in the calculation. If the range of viscosity value is taken into consideration, then the
theoretical off-state pressure drop as a function of flow rate will not form into a straightline curve, but will be more like a band that representing the range of predicted offstate pressure drop values. Figure 4.10, in this case, shows the agreement of the range
of predicted off-state pressure drop values with the measured off-state pressure drop
values in the measurement range between 37-73 ml/s.
73
The on-state pressure drop is the pressure drop that is generated when the
valve coil is charged with a current. The performance of the valve during the on-state
condition is measured in the range of current input between 0.2 to 1.0 A with 0.2 A
interval. The relationship between the measured on-state peak pressure drop and the
current input for three different flow rate variations, along with its comparison to the
predicted value from the model are shown in Figure 4.11.
Figure 4.11 Comparison of measured and theoretical on-state peak pressure drop at
various current inputs and flow rates for 0.5-0.5 mm (annular-radial) gaps configuration
In Figure 4.11, the trend line agreements of the pressure drop to the current
input curve between theoretical and measurement are shown. Due to the power
limitation of the test rig prime mover, the measurement data for 37 ml/s flow rate
are only able to be recorded up to 0.6 A current input. From Figure 4.11, the measured
pressure drop values are shown smaller than the theoretical value when the current
input is lower than 0.8 A. However, during and above 0.8 A, the measurement results
from the experiment show higher pressure drop than the predicted one. The reason
can be related to the explanation of the off-state pressure drop behavior where the
viscosity value plays an important role in the accuracy of the predicted value. Because
although the graph in Figure 4.11 is showing the on-state performance of the valve,
the off-state pressure drop is actually still exists as an initial pressure drop. However,
the percentage of off-state pressure drop to the total on-state pressure drop will be
74
varied to the magnitude of the current input. Typically, the percentage of the off-state
pressure drop is the highest at the lowest current input but will be getting lower as the
current input is increased. For example, at the 73 ml/s theoretical curve, the percentage
of the off-state pressure drop is around 62 % for the total on-state pressure drop at
0.2 A current input but dropped to 23 % at 1.0 A current input. Therefore, it can be
considered that the effect of the base viscosity value in the total pressure drop is still
quite significant in the measured pressure drop below 0.8 A.
However, the superiority of the measurement results over the predicted results
in the current input higher than 0.8 A is quite surprising. Normally, the measured onstate pressure drop will be lower than the estimated pressure drop due to unpredicted
flux loss and prediction error in the field strength value of the FEM software. The
reason is not clear now, but several causes can be the source of the phenomenon. The
first cause can be assumed as a result of the thickening effect of the fluid, which in
several literatures were mentioned to be occurred due to spalling of the particles inside
the MR fluid [12, 22]. The thickening effect of the fluid is currently not considered in
the model since the MR valve model is based on the Bingham fluid model. However,
the thickening effect can be modeled if the Herschel-Bulkley fluid model is used as
the basis of the MR valve modeling [74]. The second cause is assumed due to the
enhanced yield stress phenomenon due to aggregation process [4], which occurred
when several single-chain structure of magnetic particle joined together and formed
a thick column. As discussed by Tao [5], the yield stress enhancement of MR fluid
can be done by changing the microstructure arrangement of magnetic particle during
on-state condition. As already shown in Figure 4.2, The MRF-132DG that is used in
this study has saturated yield stress value in around 48 kPa. However, the relationship
curve between yield stress and the field strength is derived based on the assumption
that the particles are arranged in a single chain structure [94]. If the microstructure
arrangement of magnetic particle inside the MR fluid is no longer arranged into a
single chain structure but rather a multiple chain structure, the yield stress value of the
MR fluid can be potentially enhanced. However, the validity of this assumption need
further confirmation since the occurrence of multiple chain structure that was discussed
in [4] and [5] is, so far, only reported in compression mode, which is possible due to
fluid-particle separation phenomena [171]. The yield enhancement in valve mode, to
the knowledge of the author, has not been reported yet. Nevertheless, the MR valve
concept with meandering flow path is already proving an improvement by delivering
higher achievable pressure drop than the previous concept of MR valve.
75
4.4.2
Effect of Gap Size
The testing results for all gap combinations at flow rate of 37 ml/s are shown
in Figure 4.12. It is shown that the pressure drop of all gap size combinations is
consistently increasing with the increase of current input. The slope, however, is
slightly gradual in the condition where the current input is lower than 0.2 A. The
difference in the slope is actually expected considering the nonlinearity of the fluid
yield stress curve provided by the fluid manufacturer as shown in Figure 4.2. From the
curve, it is shown that the yield stress is only appeared when the flux density value is
above 0.08 Tesla. Meanwhile, from the previous Section 3.4, it has been shown that
the flux density of the MR valve at the effective area with the excitation current to the
coil below 0.2 A are below 0.1 Tesla. Due to that reason, the slope of the pressure
drop curve for the MR valve with all gap size combinations below 0.2 A of current
excitation is not as steep as the slope of the curve above 0.2 A.
Furthermore, it can be seen that the smaller the gap size, the higher the pressure
drops of the MR valve. This effect is obvious, because the smaller the gap, the higher
the effect of wall friction to the fluid, which increased the viscous pressure drop and
field-dependent pressure drop simultaneously. However, a specific note should be taken
that the change of the gap size of the radial channel provide more significant effect to
the pressure drop than the gap size of the annular channel. This is the proof of the
simulated performances that have been presented Section 3.4, which can be explained
in two reasons. The first reason is regarding the number of the radial channel, which
is twice the number of the annular channel, while the second reason is regarding the
higher flux density at the radial channel than the flux density at the annular channel.
Another important aspect that can be evaluated from the data in Figure 4.13 is
the dynamic range of the MR valve. The dynamic range of MR valve is defined as the
ratio between the on-state and the off-state pressure drops of the valve as described in
Equation 3.15. Using the expression in Equation 3.15, the dynamic ranges of the MR
valve for each gap size combination are shown in Figure 4.13. The dynamic range is
an important variable as an indication that the valve can be controlled in a wide range
of performance. Ideally, a good MR device is expected to provide the highest on-state
performance while capable to suppress the off-state to the lowest value.
Based on the dynamic range values shown in the Figure 4.13, it can be
concluded that among four types of gap size combinations, the MR valve with 1.0
mm annular gap size will always have a higher dynamic range than the MR valve with
76
7
0.5 mm annular 0.5 mm radial

Pressure Drop (MPa)
0
0
0.1
0.2
0.3
0.4
0.5
0.6
Current Input (A)
0.7
0.8
0.9
Figure 4.12 Comparison of measured peak pressure drop in various gap size
combinations
0.5 mm annular gap size. On the contrary, the increase of radial gap size, although not
as significant as the change in annular gap, has a weakening effect to the dynamic range
of the MR valve. Both of these effects are in agreement with the prediction shown in
Section 3.4. The knowledge of both factors, the achievable pressure drop value and the
dynamic range, will be useful to decide which gap size combination to be used when
designing an MR valve for a particular application.
4.4.3
Effect of Current Input Variation
Figures 4.14a to 4.14c detail the characteristics of valve with current input
variations at 0.50 Hz, 0.75 Hz and 1.00 Hz of frequency excitations. As predicted, the
MR valve with meandering flow path is consistently showing pressure drop adjustment
with the variation of the current input. As seen in Figure 4.14a, the peak pressure drop
is around 0.65 MPa at the current of 0.0 A with 0.50 Hz excitation frequency. When
the current increases in the interval of 0.1 A, the peak pressure drop are also steadily
increased with average increments of 0.58 MPa up to the maximum pressure drop of
around 6.42 MPa at the current of 1.0 A. Similarly, in Figure 4.14b, at the current of 0.0
77
Figure 4.13 Comparison of the MR valve dynamic range for each gap size
combinations
A with 0.75 Hz excitation frequency the maximum pressure drop that can be achieved
is around 0.95 MPa. However, with 0.1 A of current increment, the pressure drop are
normally increased in around 0.56 MPa until the maximum observed pressure drop
of 6.53 MPa is achieved at 1.0 A current. Lastly, at 1.00 Hz excitation frequency as
shown in Figure 4.14c, the maximum off-state pressure drop is around 1.35 MPa with
the average increments of pressure drop are seen around 0.55 MPa with each 0.1 A
increment of current so that the peak pressure drop at 1.0 A is measured at 6.86 MPa.
The relationships between peak pressure drop in each current increment for 0.50 Hz,
0.75 Hz and 1.00 Hz excitation frequency are depicted in Figure 4.15.
As shown in Figure 4.15, the rise of peak pressure drop is almost linear to
the increment of the current input. However, there is an interesting phenomenon that
can be observed from the curves where the slopes between 0.0 to 0.2 A and 0.8 to
1.0 A are slightly lower than the other regions. The phenomenon can be explained
using the curve of yield stress versus magnetic field strength of the MRF-132DG in
Figure 4.2 where the slope is leaner at the lower magnetic field strength value and also
leaner when it almost reached saturation in the higher magnetic field strength value.
78
8
1.0 A
0.9 A
0.8 A
0.7 A
0.6 A
0.5 A
Pressure Drop (MPa)
0.4 A
0.3 A
0.2 A
0.1 A
0.0 A
0
2
4
6
8
40
30
20
10
0
Flow Rate (ml/s)
10
20
30
40
(a)
8
1.0 A
0.9 A
0.8 A
0.7 A
0.6 A
Pressure Drop (MPa)
0.5 A
0.4 A
0.3 A
0.2 A
0.1 A
0.0 A
0
2
4
6
8
60
40
20
0
Flow Rate (ml/s)
20
40
60
(b)
8
1.0 A
0.9 A
0.8 A
0.7 A
0.6 A
0.5 A
Pressure Drop (MPa)
0.4 A
0.3 A
0.2 A
0.1 A
0.0 A
2
0
2
4
6
8
80
60
40
20
0
Flow Rate (ml/s)
20
40
60
80
(c)
Figure 4.14 The pressure dynamics of MR valve at various current input for 0.5-0.5
mm (annular-radial) gaps configuration, (a) 0.50 Hz (b) 0.75 Hz (c) 1.00 Hz
79
8
6
Pressure Drop (MPa)
4
2
0
2
4
6
8
0
0.50 Hz
0.75 Hz
1.00 Hz
0.1
0.2
0.3
0.4
0.5
Current Input (A)
0.6
0.7
0.8
0.9
Figure 4.15 The trend of peak pressure drop at various current input for 0.5-0.5 mm
(annular-radial) gaps configuration
Although, the declaration that the 1.0 A current input is already reached saturation
region would be too premature, since the magnetic simulation results in Section 3.2.3
showed that the peak magnetic flux density at 1.0 A current input is only around 0.85
Tesla. The saturation region of the yield stress, according to Figure 3.3 is started to
appear above 0.9 Tesla. However, the slope of the yield stress curve is already seen
to gradually decrease after the magnetic flux density reached 0.7 Tesla. In the other
words, it is valid to predict that the attempt to increase the current input above 1.0 A
will not give any significant improvement to the pressure drop performance of the MR
valve, due to the yield stress saturation of the MR fluid.
4.4.4
Effect of Excitation Frequency Variation
The relationship between valve pressure drop against flow rate of the MR fluids
at the coil current of 1.0 A at various frequencies is shown in Figure 4.16. The results
are the typical characteristics from the fifth cycle onwards since the characteristics of
the first until the fourth cycle are usually unstable and inconsistent as similarly reported
by Li et al. [172]. From Figure 4.16, it can be concluded that the peaks of the valve
pressure drop are increased with the increase of maximum flow rate. Similarly, the
shifting of peak pressure drop also shifts the width of the hysteretic region. In other
words, the width of the hysteretic region of pressure drop is highly dependent to the
value of peak flow rate as typically found in the characteristics of MR damper in its
analogous force-velocity pattern.
80
8
Pressure Drop (MPa)
8
80
0.50 Hz
0.75 Hz
1.00 Hz
60
40
20
0
Flow Rate (ml/s)
20
40
60
80
Figure 4.16 The pressure dynamics of MR valve at current input of 1A at various

frequency excitation for 0.5-0.5 mm (annular-radial) gaps configuration
There is an interesting note regarding the bend of the curve shape in each time it
passes through the 0 MPa of pressure drop. The bends of the curve were interpreted as
the pressure lag effect similarly with the ones reported in the behavior of MR dampers
by Yang [173] and Zhang et al [174] which caused by the uncompressed air pocket
that inevitably exists in the testing cell. The existence of air pocket inside the cylinder
creates force lag during the transition between the positive and negative direction of
the piston. In fact, these effect were not a unique case in the MR fluids device since its
appearance was also reported in the characteristics of a passive hydraulic damper [175].
In order to minimize the air pocket effect, a vacuum technique is often used in the
fluid filling process as well as the utilization of the accumulator to pre-compress the
air pocket in the damper assembly [123, 173]. However, since no report has strongly
mentioned about side effects of the air pocket to the overall damping characteristics
other than just the force lag, in this study, the pressure lag effect will not be taken into
consideration.
4.5
The elaboration of experimental assessment procedure and the performance

evaluation of MR valve with meandering flow path has been presented in this chapter.
The experimental assessment has been elaborated in terms of the materials and
equipments involved in the testing, the MR valve prototype and the testing method
to assess the MR valve performance. The performance of the MR valve is measured
using testing cell installed in an inhouse dynamic test platform and a commercial
81
type Fatigue Dynamic Test Machine. The experimental results are assessed in terms
of off-state pressure drop characteristics, on-state characteristics, the effect of current
variations and the effect of frequency excitation to the MR valve performance. The
measurement results of the off-state characteristics concluded that the prototype could
deliver off-state pressure drop in the range that is already predicted by the steadystate model, while the on-state characteristics of the prototype with both 0.5 mm
annular and radial gaps proved that the concept successfully increased the achievable
pressure drop of MR valve more than 6 MPa. The effect of flow rate and current
variations to the dynamic characteristics of the MR valve were also demonstrated
through the experimental assessment where the increment of flow rate and current
result in consistent increment of the pressure drop. In addition the hysteretic behavior,
which is also captured in the measurement, can be used as a source to develop the
dynamic model of the MR valve which will be covered in the next chapter.
CHAPTER 5
HYSTERESIS MODELING OF MAGNETORHEOLOGICAL VALVE
5.1
Introduction
The experimental results of MR valve have showed that the relationship

between pressure drop and fluid flow rate exhibit hysteretic behavior, which cannot be
covered in the steady-state model. In order to represent the hysteretic behavior, a new
model has to be derived based on the experimental data. In this chapter, two types of
parametric hysteresis model, the polynomial model and the modified LuGre hysteresis
model are evaluated. The performance of both model are assessed by comparing the
model results with the experimental data.
5.2
Polynomial-based Hysteresis Modeling Approach
Polynomial based modeling, as one of the parametric hysteresis modeling

approach for MR damper, is chosen as the first approach to model the hysteretic effect
of MR valve in this study. The polynomial model was first proposed by [147] to model
the hysteresis behavior of MR damper. The method also employed by [148] in the
modeling of MR damper to demonstrate the force tracking control of MR damper.
The polynomial modeling process is basically conducted by separating the

hysteresis curve into two different curves: the curve with positive acceleration (lower
curve) and the curve with negative acceleration (upper curve). Each curve is then
approximated using a unique polynomial equation, which can be expressed in the
following term:
F =
n
X
i=0
ai x i , n = order of polynomial
(5.1)
83
where ai is the coefficient that should be empirically determined using the curve fitting
of the experimental data. In both [147] and [148],the changing trend of each coefficient
as a variable to the magnitude of current input is then approximated using linear
regression so that
ai = bi + ci I , i = 0, 1, ..., n
(5.2)
where the b and c are the tracking coefficient for the a as a function of current input, I.
By substituting the equation 5.2 to equation 5.1, the damping force can be represented
as:
F =
n
X
(bi + ci I) x i , n = order of polynomial
(5.3)
i=0
.
x
MR Damper model
MR Valve model
DP
Figure 5.1 The difference between MR damper model and MR valve model excitation
Since the determination of the coefficient is conducted individually for the

upper curve and the lower curve in each current input, the number of unique sets of
coefficient for each curve will be dependent to the order of polynomial, n. However,
according to the work of Choi et al. [147], the order of polynomial should be at least
six or above in order to capture the hysteresis behavior well.
Adopting the method to the case of MR valve requires some changes in the
input and output variables, because there are some differences between the modeling
requirement of MR damper and MR valve as shown in Figure 5.1. Therefore, the
84
generalized form of the polynomial-based parametric model for the MR valve is as
follows:
P =
n
X
ai Qi , n = 6
(5.4)
i=0
where the P is the pressure drop of MR valve, similarly to the force F of MR damper
and Q is the volumetric flow rate of the MR fluid across the MR valve, similarly to the
piston velocity x of the MR damper. Sixth order of polynomial is also chosen to model
the MR valve with reference to [147] and [148].
In order to find the set of coefficient a, the experimental data from the frequency
excitation of 0.75 Hz is selected as a reference. There are 11 sets of curve that
representing the pressure drop response for current inputs of 0.0 to 1.0 A with interval
of 0.1 A. There are 1000 points of measurement for each curve which can be evenly
divided into two regions, the region with positive flow acceleration (lower loop) and
the region with negative flow acceleration (lower loop). Each region is then fitted with
the sixth order polynomial equation to find the specific sets of coefficient ai for each
curve. The results of curve fittings for both lower loop and upper loop are shown in
figure 5.2 and 5.3 respectively.
The trends of coefficients are presented in normalized values to enhance the

visibility of the variations. The method of presentation using normalized value also
has been used in [127, 146]. The first impression when noticing the trend of values
in figure 5.2 and 5.3 are that the linear regression approach, such as implemented in
[147] and [148], is no longer suitable to represent most of the relationship between
the coefficient a and the current input I. The fluctuations of coefficient values for both
lower and upper loops are mostly too large to be approximated using linear function.
The suitability of these coefficient trends to be represented by linear relationship also
can be tested using the Pearson correlation coefficient that is expressed by

I I (a a
)
r = qP
2 P
I I
(a a
)2
P
(5.5)
If the correlation coefficient, r is close enough to +1 or -1, the relationship can

be considered suitable to be represented by linear approximation. However, if the r
is equal to zero, then there are no correlation between the corresponding coefficient a
with the current input I. The results of correlation test for each coefficient with respect
85
25
9
a *2*10
6
20
a *108
15
a4*6*10
Coefficient Value
10
5
0
5
10
15
0
0.1
0.2
0.3
0.4
0.5
0.6
Current Input (A)
0.7
0.8
0.9
0.7
0.8
0.9
(a)
35
a *1*104
30
a *2*102
2
Coefficient Value
25
a *10
1
a *1
0
20
15
10
5
0
5
0
0.1
0.2
0.3
0.4
0.5
0.6
Current Input (A)
(b)
Figure 5.2 Trend of the normalized coefficient values at the positive flow acceleration
(lower loop) to the variations of current input (a) a6 , a5 and a4 (b) a3 , a2 , a1 and a0
to the current input variability are shown in table 5.1 It can be seen that most of the
coefficient has non-linear correlation with the current input and therefore cannot be
represented by linear function.
Table 5.1: Correlation test results between the model coefficient a and current input I
Correlation coefficient
Model coefficient
Lower loop Upper loop
a0
-0.9863
0.9906
a1
0.8396
0.8511
a2
0.5486
-0.8251
a3
0.0068
-0.1176
a4
-0.1181
0.5095
a5
-0.2821
-0.1492
a6
-0.0722
-0.3090
86
25
9
a *2*10
6
20
a *108
15
a4*6*10
Coefficient Value
10
5
0
5
10
15
0
0.1
0.2
0.3
0.4
0.5
0.6
Current Input (A)
0.7
0.8
0.9
0.7
0.8
0.9
(a)
45
Coefficient Value
40
a3*1*10
a *2*10
35
a1*10
30
25
20
15
10
5
0
5
0
0.1
0.2
0.3
0.4
0.5
0.6
Current Input (A)
(b)
Figure 5.3 Trend of the normalized coefficient values at the negative flow acceleration
(upper loop) to the variations of current input (a) a6 , a5 and a4 (b) a3 , a2 , a1 and a0
Since most of the coefficient cannot be represented by linear relationship,

another polynomial approximation is conducted to fit the coefficient trend. On the
basis of trial and error, it has been observed that 4th order of polynomial is decently
representing the relationship between ai and I, so that the relationship can be expressed
by
ai = bi I 4 + ci I 3 + di I 2 + ei I + fi , i = 0, 1, ..., 6
(5.6)
and the pressure drop can be represented by

P =
n
X
i=0

bi I 4 + ci I 3 + di I 2 + +ei I + fi Qi , n = 6
(5.7)
87
Using the same curve fitting process, the fourth order polynomial coefficient
determination is conducted for a6 , a5 , a4 , a3 , a2 , a1 and a0 with respect to current input
variations. The specific values of bi , ci , di , ei and fi are listed in table 5.2.
Table 5.2: List of coefficients for the polynomial-based parametric MR valve model
Curve
Order of
Coefficient Values
Region Polynomial
b
c
d
e
f
6
-3.337e-08 1.156e-08 1.641e-08 8.011e-10 6.406e-10
5
-1.149e-06 8.699e-07 2.292e-07 -1.139e-07 8.383e-09
Lower
4
1.209e-04 4.760e-06 -1.053e-04 -7.317e-06 -3.218e-06
Loop
3
5.091e-03 -3.820e-03 -9.721e-04 1.661e-04 -3.785e-05
2
-0.0197
-0.224
0.239
0.0151
4.698e-03
1
-3.2792
-1.2161
4.9471
0.5715
0.2141
0
-170.0049
335.074
-193.1929
-4.9548
-1.6389
6
4.634e-08 -4.068e-08 -3.928e-09 -3.063e-10 -4.239e-10
5
-2.286e-06 2.938e-06 -8.588e-07 5.549e-08 1.565e-09
Upper
4
-1.715e-04 1.138e-04 5.632e-05 3.617e-06 2.291e-06
Loop
3
9.737e-03
-0.0123
3.502e-03 -5.047e-04 -9.416e-06
2
0.0514
0.1260
-0.2001
-8.136e-03 -3.707e-03
1
-7.7430
7.0455
0.6173
1.1776
0.1862
0
199.877
-373.947
211.433
1.279
1.9134
5.3
Modified LuGre-based Hysteresis Modeling Approach
The second approach to parametrically model the hysteretic behavior of the

MR valve is based on the LuGre hysteresis model. The LuGre model is the hysteresis
model that initially developed to model the friction dynamics [176,177] and is selected
as the base model for the MR valve. The selection of LuGre model as the base model is
mainly because hypothetically the flow dynamics is easier to be represented as friction
dynamics rather than as a set of springs and dampers. The primary form of LuGre
hysteresis model for MR damper is as follows:
F (t) = 0 z + 1 z + 2 x
(5.8)
where the 0 , 1 , and 2 are the parameters of the model and z is the variable that can
be expressed as:
z = x
|x|
z
g (x)
(5.9)
88
where g (x)
is the additional function that depends on various factors such as material
properties and temperature.
Further implementations of the LuGre model have brought some modifications

to the primary form such as the modification proposed by Sakai et al. [137] that
expressed the LuGre model as:
F (t) = a z + 0 zv + 1 z + 2 x + b xv
(5.10)
z = x 0 a0 |x|
z
(5.11)
where a , 0 , 1 , 2 , b , and a0 are the parameters of the model while v is the voltage
input to the coil and the z is interpreted as the variable that expressed the average MR
fluids transient deformation generated when the direction of force is changing or also
known as the evolutionary variable.
The other modification to the LuGre model was also proposed by Jimenez and
Avarez-Icaza [138] in the following form:
F (t) = 0b zv + 1 z + 2a x
(5.12)
z = x 0b a0 |x|
z (1 + a1 v)
(5.13)
where 0b , 1 , 2a , , a0 and a1 are the parameters of the model, while v is similarly

the applied voltage to the coil as well as the z as the evolutionary variable.
It can be seen that both forms use the voltage, v, as the input variable that is
related to the coil magnetization strength. In another model, the magnetization strength
is expressed with the magnitude of the current, i, given to the coil [127]. Physically
both can be considered as equal term given that the change of coil resistance during
magnetization is neglected. However, both current and voltage are just the indirect
variables to express the units of magnetization strength. In other words, the value of
voltage or current that is expressed in the model will not be universally applicable since
the strength of magnetic field that influences the MR effect is also highly subjected to
coil turns and dimensions.
In similar reasons with the modifications of the polynomial-based parametric

model as shown in Figure 5.1, some modifications to the base form of the LuGre
hysteresis model are also needed to transform the form of damper model to the form
89
of valve model. The modifications of the base model are also needed to reduce the
parameter and simplify the model. Therefore, the generalized form of the modified
LuGre based MR valve model proposed in this paper is as follows:
P = Az + Bz + CQ
(5.14)
z = Q a0 |Q| z
(5.15)
thus, there are five independent parameters that need to be identified, and these set of
parameters are defined as follows:
= [A, B, C, , a0 ]
As a general form, these five parameters can be dependent or independent to

the flow rate and/or the current input depending on the selected method of parameter
identification. However, in common, at least one of these parameters should be
dependent to the current input since the current input applied, as one of the independent
variables, is not yet accommodated in the generalized form of the model.
In order to identify the values of model parameters, the characteristics of MR

valve with 0.5 mm annular and 0.5 mm radial gaps at frequency excitation of 0.75 Hz
is taken as a sample. The identification process of the model parameter is conducted
using the gradient descent method through the Parameter Estimation Tool (PET) in
MATLAB. The method is used to determine the optimum values of the five different
parameters to match the model predictions and the experimental data. Since the curve
is unique for different current input, there are also unique sets of parameter values for
each curve. The collection of each parameter values for different current input are then
retraced to find its trend line. The approximated functions of the trend line will be the
empirical function for to determine the parameter values with respect to the variation
of the current input.
Figure 5.4 present the trends of normalized estimated parameters at various

current inputs. There are similarities of parameter trends in these three different
frequency excitation where the A, B and tend to increase, and C tends to decrease.
The a0 is intentionally made invariant to the current input and proportionally variable
to the frequency excitations. From these figures it can be seen that although there are
some fluctuations in the estimated parameters, these fluctuations did not qualitatively
change the overall trend of the parameters.
90
14
A*20
B/5
C*2.5
*5
a0*5
12
10
8
6
4
2
0
0
0.2
0.4
0.6
Current Input (A)
0.8
Figure 5.4 Trend of estimated parameters with respect to current input
The trend of each parameter is used to generate an approximation function to

mimic the trend of each parameter, similarly with the approach demonstrated by Jiang
and Christenson [145]. The approximated function will be specific for each parameter
for a bonded range of the current input. The functions are needed so that the model,
although case specific, can be used in a more compact form of mathematical functions
and more practical to be used in control design process. The approximated functions
for each parameter are shown in Table 5.3
Table 5.3: List of approximated function for different parameters

Parameters
Approximated Functions
A
1.1612i3 + 1.397i2 + 0.3613i + 0.0391
B
22.33i5 + 181.31i4 324.92i3 + 192.26i2 + 7.37i + 2.035
C
2.268i3 + 0.6812i2 + 1.1267i + 0.0683
3.6414i4 9.1188i3 + 8.032i2 3.247i + 1.2009

a0
1.0
5.4
Model Performance Comparison
The simulation results of both polynomial-based hysteresis model and the

LuGre based hysteresis model in comparison with the experimental data for frequency
91
excitation of 0.75 Hz at current inputs of 0.3 A, 0.6 A and 0.9 A are shown in Figure
5.5a to 5.5c respectively.
Generally, in these figures, the results from both models are in a good
agreement with the experimental data, especially in terms of the peak pressure drop
value. Though there are some compromises when the curve is reaching the pressure
drop of 0 MPa due to the air pocket effect as mentioned in section 4.4.4. These air
pocket effects, unfortunately, failed to be represented in both models and apparently
have been tampering the parameter values during the parameter identification process.
Although in general both model are able to represent the hysteretic behavior of
the MR valve, in order to compare the model accuracy, a specific measurement need to
be taken. In this study, the relative error, that is adopted from [127], is used to measure
the level of accuracy of each model. The relative error is expressed in the following
equation:

Pn
P exp Pimodel
i
RE = i=1 Pn
exp
|
i=1 |Pi
(5.16)
where n is the number of measurement points, Piexp is the experimental data of

pressure drop at i-th point and Pimodel is the i-th pressure drop obtained from the
model. The comparison of relative error between the polynomial-based hysteresis
model and the LuGre based hysteresis model are presented in table 5.4
Table 5.4: Comparison of relative error at 0.75 Hz frequency excitation

Current Input
Relative Error
(A)
Polynomial-based model LuGre based model
0.0
0.0813
0.0805
0.1
0.0805
0.1048
0.2
0.0686
0.0566
0.3
0.0609
0.0662
0.4
0.0963
0.1027
0.5
0.0639
0.0522
0.6
0.0641
0.0612
0.7
0.0518
0.0418
0.8
0.0510
0.0556
0.9
0.0644
0.0944
1.0
0.0792
0.1186
Average
0.0693
0.0759
According to table 5.4, most of the results from both model are showing relative
92
3
Pressure Drop (MPa)
-1
-2
Experimental Data
Polynomial-based Hysteresis Model
LuGre based Hysteresis Model
-3
-60
-40
-20
0
Flow Rate (ml/s)
20
40
60
(a)
6
Pressure Drop (MPa)
-2
-4
Experimental Data
-6
-60
-40
-20
0
Flow Rate (ml/s)
20
40
60
(b)
8
Pressure Drop (MPa)
-2
-4
-6
-8
-60
Experimental Data
-40
-20
0
Flow Rate (ml/s)
20
40
60
(c)
Figure 5.5 Comparison between the test data and the model results for various current
input, (a) 0.3 A (b) 0.6 A (c) 0.9 A
93
error less than 10%, which can be considered good since the results from the model
developed by [127] were reported to have even higher relative error in around 15%.
In average, the relative error of the polynomial-based hysteresis model is also lower
than the relative error of the LuGre based hysteresis model, which is demonstrating
the advantage of the polynomial model in terms of accuracy as already explained in
chapter 2.
Pressure Drop (MPa)
-2
-4
-6
-8
-40
Experimental Data
-30
-20
-10
0
Flow Rate (ml/s)
10
20
30
40
(a)
8
Pressure Drop (MPa)
-2
-4
-6
-8
-80
Experimental Data
-60
-40
-20
0
Flow Rate (ml/s)
20
40
60
80
(b)
Figure 5.6 Comparison between the test data and the model results for current input
of 1.0 A at various frequency excitations, (a) 0.50 Hz (b) 1.00 Hz
However, while it is obvious that the models, that have been specifically
identified using the 0.75 Hz experimental data, can mimic the 0.75 Hz measurement
data well, the real challenge is actually to perform at the frequency excitations other
than 0.75 Hz. As an example. the performance comparisons of each model with the
experimental data at 0.50 Hz and 1.00 Hz frequency excitations for current input
94
of 1.0 A are depicted in Figure 5.6. The results in these figures have shown that
the deviations are apparent for both models in the 0.50 Hz and 1.00 Hz frequency
excitations. However, the deviations of the polynomial-based hysteresis model is seen
much larger than the deviations of the LuGre based hysteresis model, which are also
shown by the comparison of relative error in table 5.5. These large deviations can be
explained by considering that the polynomial equations, at some values, will reach its
extreme points. In this case the extreme points of the pressure drop are visible at the
1.00 Hz frequency excitations, which occurred at the flow rate slightly below +/- 55
ml/s.
Table 5.5: Comparison of relative error at 0.50 Hz and 1.00 Hz frequency excitations
Current
Relative Error
Input
Polynomial-based model LuGre based model
(A)
0.50 Hz
1.00 Hz
0.50 Hz 1.00 Hz
0.0
0.1268
0.5871
0.0767
0.0674
0.1
0.1240
0.5132
0.1053
0.0950
0.2
0.1582
0.5029
0.0685
0.0544
0.3
0.1670
0.6500
0.0894
0.0507
0.4
0.1891
0.8125
0.1274
0.0803
0.5
0.1550
0.9269
0.0603
0.0682
0.6
0.1503
1.0091
0.0669
0.1066
0.7
0.1405
0.9001
0.0614
0.0757
0.8
0.1794
0.6162
0.1098
0.0423
0.9
0.2073
0.3196
0.1424
0.0821
1.0
0.3033
0.8669
0.1427
0.1369
Average 0.1728
0.7004
0.0955
0.0781
According to the model verification results, it can be concluded that the

polynomial-based hysteresis model is actually providing better accuracy than the
LuGre-based hysteresis model but only as far as the model inputs are similar with
the inputs used in the model development. The LuGre based hysteresis model, on the
other hand, although cannot compete with the polynomial-based model in terms of
accuracy, is shown the capability to show a decent performance and smaller deviations
to the measurement data in wider range of model inputs. Additionally, the number of
parameters of the LuGre based hysteresis model is also smaller than the polynomialbased hysteresis model so that the parameter identification process is relatively easier.
Although it should be noted that generally the accuracy of the parametric model is
not just determined by the model form but also by the identification method of the
parameters [74]. The good agreement between the valve experimental data and the
model showed that the model can be used for developing further applications of the
95
MR valve including actuator design and damper design as well as the development of
control system for suspension and vibration isolation devices that involves MR fluids
flow control.
5.5
The hysteretic modeling of the MR valve has been discussed in this chapter.
The hysteretic modeling is presented in two types of parametric modeling approaches,
namely the polynomial-based hysteresis modeling approach and the modified LuGrebased hysteresis modeling approach. The performance of these two approaches in
modeling the hysteretic behavior of the MR valve is compared with the experimental
data. According to the performance comparison results, it can be concluded that
the polynomial-based hysteresis model has better accuracy than the LuGre-based
hysteresis model but only at the range of inputs that is recognized during the
identification process. In the wider range of inputs, the LuGre-based model show better
capability to consistently mimic the behavior of MR valve than the polynomial-based
model. Moreover, the LuGre based model has fewer parameters to identify than the
polynomial based model so that the identification process is relatively simpler.
CHAPTER 6
CONCLUSIONS AND RECOMMENDATIONS
6.1
Conclusions
Magnetorheological (MR) fluids based devices are offering advanced features

through the fluids sensitive rheological properties to magnetic field. The device has
penetrated various field of applications that requires energy dissipation component
such as damping and brakes. There are three known ways to utilize the unique
characteristics of MR fluids in a device. The first is known as the shear mode, which
often utilized in brakes, the second is known as the flow mode, which commonly used
in dampers and the last one is the squeeze mode, which mostly appeared in vibration
mountings. An MR valve, in particular, is the main component that responsible of
determining the performance of MR devices that work in flow mode. In this study,
a new concept of MR valve with the aim to achieve higher pressure drop without
any major requirement to enlarge the valve size was introduced. The concept was
presented in both simulation, with the aid of Finite Element Method Magnetics
(FEMM) software, and experimental work, using the dynamic test machines. The
mathematical models of the new concept of MR valve also have been developed and
assessed with the experimental data. The detailed conclusions of the study are listed in
the following:
6.1.1
The New Magnetorheological Valve Concept
In this study, the new concept of MR valve has been introduced using the
combination of multiple annular and radial gaps to form a meandering flow path. The
meandering flow path approach is used to increase the path length of MR fluid flow
inside the valve so that the opportunity of the MR fluid to be exposed with magnetic
field is higher. In order to predict the performance of the new MR valve concept, the
97
steady-state model of the MR valve with multiple annular and radial valve is derived.
According to the simulation results of the steady-state model, the new MR valve with
meandering flow path concept is able to reach pressure drop of around 6 MPa in a
comparable valve size and power consumption of a conventional 2.5 MPa annular MR
valve. The performance of the MR valve was also analyzed by assessing the effect of
current input variations, flow rate variations and gap size variations to the generated
pressure drop of the MR valve.
6.1.2
Gap Size Selection Effect
Among several variables that influencing the performance of MR valve, gap

size can be considered as the most significant variable to the overall performance of the
MR valve. The gap size has been known to have inverse-correlation with the viscous
pressure drop at the power of three. At the same time, the gap size is also capable to
influence the magnetic field strength at the effective area which eventually will relate to
the field-dependent pressure drop of the MR valve. According to the simulation results,
the reduction of gap size at the annular channels of the proposed MR valve concept has
a very significant effect to the increase of viscous pressure drop value of the MR valve.
Obviously, it also has similar effect to the improvement of the field-dependent pressure
drop but the impact was not as high as the viscous pressure drop. On the other hand,
the gap size reduction at the radial channels has greater impact to the improvement
of field-dependent pressure drop than the annular channel, but has lower effect to the
viscous pressure drop value. Further, the actual selection of the gap size will highly
determined by the designated applications of the MR valve.
6.1.3
Experimental Assessment of Magnetorheological Valve Performance
The predicted results of the steady-state model are well confirmed by the
experimental work with small deviations in the off-state pressure drop value. The
experimental work is conducted using a double rod hydraulic cylinder, as the testing
cell, actuated by the dynamic test machine. The experimental test was conducted in
three different frequency excitations with constant amplitude of 25 mm to emulate the
variations of peak fluid flow rate. In each frequency excitations the current input is
varied from 0.0 A to 1.0 A with 0.1 A increment. Using the reciprocating sinusoidal
movement from the dynamic test machine, the fluid flow rate across the MR valve
98
is also varied in sinusoidal value, which allow the observation of hysteretic pressure
drop behavior. According to the experimental results, the hysteretic pressure behavior
occurs when there are different measured pressure drop at the same flow rate with
non-zero flow acceleration at different flow direction.
6.1.4
Hysteretic Modeling of Magnetorheological Valve
The hysteretic behavior cannot be shown in the steady-state model. Therefore,

another model of MR valve, which can represent the hysteretic behavior, is needed. In
order to develop a model that can mimic the hysteretic behavior, the measured pressure
drop from the experimental test, which show the hysteretic behavior of the MR valve,
is used. Two hysteretic model are developed and compared with the experimental
data. The first model is developed using the polynomial-based parametric approach
and the second model is derived based on LuGre hysteresis parametric operator.
The comparison results have shown that the polynomial-based parametric model is
producing more accurate results than the LuGre-based parametric model for the same
inputs that are used in the parameter identification process. Meanwhile, at the other
range of inputs, the results from the LuGre-based parametric model is more consistent
and robust than the results from the polynomial-based parametric model.
Since the verified hystetic model is a very good representation of the MR valve,
the model can be used to design the control system for further applications of MR
valve. In this study, the process of designing a pressure tracking control system using
PI controller is presented as a demonstration on how the MR valve model can be used
to serve various applications. The assessment of performance of the pressure tracking
control is evaluated by introducing three different types of reference input, which
representing the class of continuous and discontionuous functions, to the controller.
The simulation results have shown that the controller is able to track these three signals
well. Further, the pressure tracking control system can be used as the embedded inner
loop controller of the MR valve that act to ensure that the generated pressure drop of
an MR valve can be accurately delivered as commanded by the outer loop controller to
serve the application demands.
99
6.2
Contributions of the Research
In this research, the new way to improve the achievable pressure drop of MR
valve has been demonstrated with the meandering flow path concept. The new concept
has been studied thoroughly by evaluating the MR valve behavior in steady-state and
dynamic environment. The results that were obtained from the study showed that the
new concept can be potentially used in various applications that requires flow control
as well as enhancing the performance of MR devices that utilize valve as one of its
main component. Nevertheless, the contributions of this research are listed as follows:
(a)
This research contributed to the introduction of the new concept of MR valve.

The novelty of the new MR valve concept is mainly justified by the unique
arrangement of the effective area, known as the meandering flow path structure,
which makes it capable to improve the achievable pressure drop of the MR
valve. The meandering flow path structure was formed by the use of multiple
combination of annular and radial flow channel in a single MR valve structure.
The performance improvement that can be achieved with the meandering
flow path structure has been proven in this research through simulation and
experimental assessment.
(b)
The second contribution of this research was related to the knowledge on the
effect of gap size selection to the achievable pressure drop of the valve. The
specific selections of the gap size have a very significant effect to the pressure
drop characteristics of the valve where the annular gaps have shown to be
responsible in the determination of off-state pressure drop while the radial gaps
were highly influencing the achievable on-state performance. In terms of valve
dynamic range, it has been confirmed in both simulation and experimental work
that the reduction of annular gap size is highly affecting the degradation of valve
dynamic range while in contrary occured with the reduction of radial gap size,
although the effect is not significant. Further, the knowledge on how to select
the size of both gaps are very important in the sizing process for particular
applications.
(c)
The third contribution was regarding the development of new parametric model
based hysteretic MR valve model. The hysteresis was identified during the
experimental assessment of MR valve. The model is the first parametric model
derived specifically to model the hysteretic behavior of MR valve.
100
6.3
Open Problems and Recommendations for Future Works
6.3.1
Pressure Tracking Control System
There are wide ranges of applications that can utilize the advance features of
MR valve, since most of actuators that rely on flow control mechanism will require
valve as their main control element. However, in order for the MR valve to be
well functioned as the final control element of an actuator, the pressure drop that
is generated by the valve has to be controllable accurately. Therefore, the pressure
tracking control system is needed as the valve inner-loop controller and integrated
module. As a case study for future reference, the explanation of the basic structure
of the pressure tracking controller and its simulated performance using the MR valve
hysteresis model in the case of tracking various types of input references are presented.
Desired
pressure
drop
+
Error
signal
PI
Controller
Control
signal
Current
Driver
Current
input
MR
Valve
Flow
rate
Actual
pressure
drop
Figure 6.1 Basic structure of pressure tracking control of MR valve
The basic structure of the pressure tracking control of the MR valve are shown
in figure 6.1, which depicts a closed-loop control system with PI controller. The PI
controller is chosen because it is simple, easy and widely recognized as the most
common controller in industrial application [178]. The input of the controller is the
error signal, which is acquired from the subtraction between the reference signal and
the feedback signal. The reference signal in the case of MR valve pressure tracking
control is the desired pressure drop and the feedback signal is the actual pressure drop.
Therefore, the formulation of the PI controller for the pressure tracking control can be
expressed as follows
Z
u (t) = Kp e (t) + Ki
e (t)
e (t) = Pd (t) Pa (t)
(6.1)
(6.2)
101
where the u(t) is the control signal which drives the current driver to charge the
electromagnetic coil of the MR valve with current, the Kp and Ki are the proportional
and integral gain respectively as well as the Pd and Pa as the desired pressure drop
and actual pressure drop.
The current driver is needed to adjust the value of the output signal from the
controller with the signal input that can drive the MR valve. Since the control signal
will not always satisfy the signal requirement of the MR valve. For example, the value
of Pa can be higher or lower than the value of the Pd . Therefore, the value of e(t)
can be positive or negative which concurrently cause the control signal u(t) to be in
positive or negative value. However, the positive of negative sign of the current input
to the electromagnetic coil of the MR valve in this study practically has no effect to the
magnitude of the pressure drop. Therefore, it is the task of the current driver to adjust
the sign, which simply can be done using absolute function.
In order to assess the performance of the pressure tracking control, the LuGrebased hysteresis MR valve model is used with several types of reference signal, namely
the sinusoidal signal, the pulse signal and the saw-tooth signal. The selection of these
signals as the reference input are based on the assumption that if the system can
track these signal well, then the system is considered robust to track various types
of continuous and discontinuous functions [148]. The simulation results of pressure
tracking control under various reference functions are shown in figure 6.2. The results
are the outcome of pressure tracking control with controller parameters of Kp = 0.08
and Ki = 4, which acquired from MATLAB auto-tuning tools, and Kp = 0.05 and
Ki = 2, which acquired from trial and error method. From figure 6.2, it can be
concluded that the pressure tracking control of the hysteresis model of MR valve is
able to track various types of continuous and discontinuous functions well.
Although the explanation of the pressure tracking control in this section only
involved a closed-loop PI controller, the further application of inner-loop controller
of the MR valve is not just limited to the PI controller. Various control strategy can
be developed and integrated with the MR valve. The demonstrated pressure tracking
control system using PI controller shown in this section is merely an illustration of
the kind of applications that the MR valve can serve. With the involvement of the
inner-loop controller, the MR valve can serve both semi-active and active application
accurately. Considering the necessity of the study, the further exploration of the
pressure tracking control system of the meandering type MR valve is recommended
for future works.
102
5.5
Desired Pressure Drop
Controlled Pressure Drop (PI Autotuning)
Controlled Pressure Drop (Trial and Error)
Pressure Drop (MPa)
4.5
3.5
2.5
2
6
Time (s)
10
(a)
5.5
Pressure Drop (MPa)
4.5
3.5
2.5
2
6
Time (s)
10
(b)
5.5
Pressure Drop (MPa)
4.5
3.5
2.5
2
6
Time (s)
10
(c)
Figure 6.2 Simulation results of pressure tracking control under various functions as
reference, (a) Sinusioidal (b) Pulse (c) Saw-tooth
103
6.3.2
Other Open Problems
There are also several problems encountered that were not able to be discussed
in this study and can be recommended for future works as a continuation of this study.
These problems are as the following:
(a)
The conceptual design of the MR valve with meandering flow path was not
considering any optimization on valve design. The gap sizes of annular and radial
flow channel, the coil properties, the MR fluid types and the overall valve size in
this study were selected merely to serve the observation purposes and therefore
not considering any optimum value. In the future, it will be necessary to put the
consideration on the optimization on these variables such as to conduct the study
on the critical thickness of the valve casing with respect to magnetic reluctance
and mechanical strength to withstand the internal pressure of the valve.
(b)
The fundamental knowledge of the flow behavior in radial valve, in particular,

was also found interesting by the author. The interesting part existed in the stress
behavior of the MR fluid when radially flowing in concave and convex direction
since the shear rate of the fluid is not constant even in a constant volumetric flow
rate. Due to these characteristics, it is make sense to suspect that the common
knowledge of MR fluid flow behavior in a parallel channel is no longer valid
to explain the flow behavior in a radial channel. Therefore, a thorough study is
necessary to reveal the flow behavior in a radial channel in the future, which in
particular, can contribute to improve the model of MR valve with meandering
flow path.
(c)
In terms of experimental assessment, in this study the MR valve has experienced

testing where the flow rate is constantly changing due to the reciprocating
movement of the testing cell. With these kind of assessment, the transient
behavior of the MR valve due to the variations of fluid flow rate are observable.
However, in these testing conditions, the current input remains constant and
therefore the transient response of the MR valve to the change of current input
has not been investigated yet. The knowledge of transient response of the MR
valve with respect to the current input is necessary to develop a dynamic model
of the MR valve with accurate time constant to serve the control design purposes.
(d)
The meandering flow path structure that consists of the combination of annular
and radial flow channel inevitably formed many sharp corner similarly with the
effect of elbow. The fluid flow in elbow, in general, has been widely known as a
complex phenomenon due to frictional and separation effects [179,180]. Though
104
for the flow case of non-Newtonian fluid the friction loss coefficient has been
shown lower than in the case of Newtonian fluid [181], the losses in a small flow
channel such as in an MR valve may become significantly large [124]. Therefore,
it may be necessary in the future to consider the elbow effect in the meandering
flow path structure in the pressure drop characteristics of MR valve to further
examine the concept.
(e)
The results of polynomial-based hysteresis model have revealed that the

polynomial model is actually cannot act as a stand-alone model. The limitation of
the model to track the characterization results can be solved by adding another
system that capable to continuously update the polynomial model parameters
when flow rate input is changing. The addition of such system will provide a
very promising hysteresis model since the polynomial model has been shown as
a very accurate model in a limited range of flow rate. The further study of on-line
parameter tuning for polynomial-based hysteresis model is an interesting topic
and will be beneficial in the field of modeling of hysteretic system.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
Bossis, G., Lacis, S., Meunier, A. and Volkova, O. Magnetorheological

fluids. Journal of Magnetism and Magnetic Materials, 2002. 252: 224228.
Bica, I., Liu, Y. D. and Choi, H. J.
Physical characteristics of
magnetorheological suspensions and their applications.
Journal of
Industrial and Engineering Chemistry, 2013. 19(2): 394406.
Jolly, M. R., Bender, J. W. and Carlson, J. D. Properties and applications
of commercial magnetorheological fluids. Journal of Intelligent Material
Systems and Structures, 1999. 10(1): 513.
Tang, X., Zhang, X., Tao, R. and Rong, Y. Structure-enhanced yield stress of
magnetorheological fluids. Journal of Applied Physics, 2000. 87(5): 2634.
Tao, R. Super-strong magnetorheological fluids. Journal of Physics:
Condensed Matter, 2001. 13(50): R979R999.
Choi, Y. T., Cho, J. U., Choi, S. B. and Wereley, N. M. Constitutive models
of electrorheological and magnetorheological fluids using viscometers.
Smart Materials and Structures, 2005. 14(5): 10251036.
Chen, S., Huang, J., Shu, H., Sun, T. and Jian, K. Analysis and testing
of chain characteristics and rheological properties for magnetorheological
fluid. Advances in Materials Science and Engineering, 2013. 2013(Ci): 1
6.
Bolter, R. and Janocha, H. Design rules for MR fluid actuators in different
working modes. Proceedings of SPIE Vol. 3045. SPIE. 1997, vol. 3045.
148159.
Carlson, J. and Jolly, M. R. MR fluid, foam and elastomer devices.
Mechatronics, 2000. 10(4-5): 555569.
Zipser, L., Richter, L. and Lange, U. Magnetorheologic fluids for actuators.
Sensors and Actuators A: Physical, 2001. 92(1-3): 318325.
Rabinow, J. The magnetic fluid clutch. Transactions of the American
Institute of Electrical Engineers, 1948. 67(2): 13081315.
Carlson, J. Critical factors for MR fluids in vehicle systems. International
Journal of Vehicle Design, 2003. 33(1/2/3): 207.
Milecki, A. and Hauke, M. Application of magnetorheological fluid in
industrial shock absorbers. Mechanical Systems and Signal Processing,
106
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
2012. 28: 528541.

Zhu, X., Jing, X. and Cheng, L. Magnetorheological fluid dampers: A
review on structure design and analysis. Journal of Intelligent Material
Wang, J. and Meng, G. Magnetorheological fluid devices: principles,
characteristics and applications in mechanical engineering. Proceedings
of the Institution of Mechanical Engineers, Part L: Journal of Materials:
Design and Applications, 2001. 215(3): 165174.
Kordonski, W., Gorodkin, S., Kolomentsev, A., Kuzmin, V., Lukianovich,
A., Protasevich, N., Prokhorov, I. and Shulman, Z. Magnetorheological
valve and devices incorporating magnetorheological elements, 1994.
Gorodkin, S., Lukianovich, A. and Kordonski, W. Magnetorheological
throttle valve in passive damping systems. Journal of Intelligent Material
Yokota, S., Yoshida, K. and Kondoh, Y. A pressure control valve using MR
fluid. Proceedings of the JFPS International Symposium on Fluid Power,
1999. 4: 377380.
Yoshida, K., Takahashi, H., Yokota, S., Kawachi, M. and Edamura, K. A
bellows-driven motion control system using a magneto-rheological fluid.
Proceedings of the JFPS International Symposium on Fluid Power, 2002.
(5-2): 403408.
Yoo, J.-H. and Wereley, N. M.
Design of a high-efficiency
magnetorheological valve. Journal of Intelligent Material Systems and
Structures, 2002. 13(10): 679685.
Wang, X., Gordaninejad, F., Hitchcock, G. H., Bangrakulur, K., Fuchs, A.,
Elkins, J., Evrensel, C. A., Dogruer, U., Ruan, S., Siino, M. and Kerns,
M. Q. A new modular magneto-rheological fluid valve for large-scale
seismic applications. Wang, K.-W., ed. Proc. SPIE 5386. Bellingham, WA:
SPIE. 2004, vol. 5386. 226237.
Grunwald, A. and Olabi, A. G. Design of magneto-rheological (MR) valve.
Sensors and Actuators A: Physical, 2008. 148(1): 211223.
Ai, H. X., Wang, D. H. and Liao, W. H. Design and modeling of a
magnetorheological valve with both annular and radial flow paths. Journal
of Intelligent Material Systems and Structures, 2006. 17(4): 327334.
Wang, D. H., Ai, H. X. and Liao, W. H. A magnetorheological valve with
both annular and radial fluid flow resistance gaps. Smart Materials and
Structures, 2009. 18(11): 115001.
Yoo, J.-H., Sirohi, J. and Wereley, N. M.
A nagnetorheological
piezohydraulic actuator. Journal of Intelligent Material Systems and
107
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
Structures, 2005. 16(11-12): 945953.

John, S., Chaudhuri, A. and Wereley, N. M. A magnetorheological actuation
system: test and model. Smart Materials and Structures, 2008. 17(2):
025023.
Salloom, M. and Samad, Z. Design and modeling magnetorheological
directional control valve. Journal of Intelligent Material Systems and
Structures, 2011. 23(2): 155167.
Lee, H.-S. and Choi, S.-B. Control and response characteristics of a
magneto-rheological fluid damper for passenger vehicles. Journal of
Intelligent Material Systems and Structures, 2000. 11(1): 8087.
Yao, G. Z., Yap, F. F., Chen, G., Li, W. H. and Yeo, S. H. MR damper
and its application for semi-active control of vehicle suspension system.
Mechatronics, 2002. 12(7): 963973.
Lee, H. G., Sung, K. G. and Choi, S. B. Ride comfort characteristics with
different tire pressure of passenger vehicle featuring MR damper. Journal
of Physics: Conference Series, 2009. 149: 012069.
Choi, S.-B., Seong, M.-S. and Ha, S.-H. Vibration control of an MR vehicle
suspension system considering both hysteretic behavior and parameter
variation. Smart Materials and Structures, 2009. 18(12): 125010.
Giorgetti, A., Baldanzini, N., Biasiotto, M. and Citti, P. Design and testing
of a MRF rotational damper for vehicle applications. Smart Materials and
Structures, 2010. 19(6): 065006.
Ha, S. H., Seong, M.-S. and Choi, S.-B. Design and vibration control of
military vehicle suspension system using magnetorheological damper and
disc spring. Smart Materials and Structures, 2013. 22(6): 065006.
Bai, X.-X., Hu, W. and Wereley, N. M. Magnetorheological damper utilizing
an inner bypass for ground vehicle suspensions. IEEE Transactions on
Magnetics, 2013. 49(7): 34223425.
Fischer, D. and Isermann, R. Mechatronic semi-active and active vehicle
suspensions. Control Engineering Practice, 2004. 12(11): 13531367.
Poussot-Vassal, C., Sename, O., Dugard, L., Gaspar, P., Szabo, Z. and Bokor,
J. A new semi-active suspension control strategy through LPV technique.
Control Engineering Practice, 2008. 16(12): 15191534.
Ubaidillah, Hudha, K. and Jamaluddin, H. Simulation and experimental
evaluation on a skyhook policy-based fuzzy logic control for semi-active
suspension system. International Journal of Structural Engineering, 2011.
2(3): 243.
Sam, Y. and Hudha, K. Modelling and force tracking control of hydraulic
actuator for an active suspension system. 2006 1ST IEEE Conference on
108
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
Industrial Electronics and Applications. IEEE. 2006. ISBN 0-7803-9513-1.

16.
Imaduddin, F., Hudha, K., Mohammad, J. I. and Jamaluddin, H. Simulation
and experimental investigation on adaptive multi-order proportionalintegral control for pneumatically actuated active suspension system using
knowledge-based fuzzy. International Journal of Modelling, Identification
and Control, 2011. 14(1/2): 7392.
Gysen, B., Paulides, J., Janssen, J. and Lomonova, E.
Active
electromagnetic suspension system for improved vehicle dynamics. IEEE
Transactions on Vehicular Technology, 2010. 59(3): 11561163.
Gysen, B. L. J., van der Sande, T. P. J., Paulides, J. J. H. and Lomonova, E. A.
Efficiency of a regenerative direct-drive electromagnetic active suspension.
IEEE Transactions on Vehicular Technology, 2011. 60(4): 13841393.
Nguyen, Q. H., Choi, S. B., Lee, Y. S. and Han, M. S. Optimal design
of high damping force engine mount featuring MR valve structure with both
annular and radial flow paths. Smart Materials and Structures, 2013. 22(11):
115024.
Phu, D. X., Shah, K. and Choi, S.-B. A new magnetorheological mount
featured by changeable damping gaps using a moved-plate valve structure.
Phu, D. X., Choi, S. B., Lee, Y. S. and Han, M. S. Design of a
new engine mount for vertical and horizontal vibration control using
magnetorheological fluid. Smart Materials and Structures, 2014. 23(11):
117001.
Phu, D. X. and Choi, S.-B. Vibration control of a ship engine system
using high-load magnetorheological mounts associated with a new indirect
fuzzy sliding mode controller. Smart Materials and Structures, 2015. 24(2):
025009.
Choi, S.-B., Nam, M.-H. and Lee, B.-K. Vibration control of a MR seat
damper for commercial vehicles. Journal of Intelligent Material Systems
and Structures, 2000. 11(12): 936944.
Mcmanus, S., St. Clair, K., Boileau, P., Boutin, J. and Rakheja, S. Evaluation
of vibration and shock attenuation performance of a suspension seat with
a semi-active magnetorheological fluid damper. Journal of Sound and
Vibration, 2002. 253(1): 313327.
Choi, S.-B. and Han, Y.-M. MR seat suspension for vibration control of a
commercial vehicle. International Journal of Vehicle Design, 2003. 31(2):
202.
Hiemenz, G. J., Hu, W. and Wereley, N. M. Semi-active magnetorheological
109
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
helicopter crew seat suspension for vibration isolation. Journal of Aircraft,

2008. 45(3): 945953.
Karakoc, K., Park, E. J. and Suleman, A. Design considerations for an
automotive magnetorheological brake. Mechatronics, 2008. 18(8): 434447.
Park, E. J., da Luz, L. F. a. and Suleman, A. Multidisciplinary design
optimization of an automotive magnetorheological brake design. Computers
& Structures, 2008. 86(3-5): 207216.
Nguyen, Q. H., Jeon, J. C. and Choi, S. B. Optimal design of a
hybrid magnetorheological brake for middle-sized motorcycles. Applied
Mechanics and Materials, 2011. 52-54: 371377.
Nguyen, Q. H. and Choi, S. B. Optimal design of a T-shaped drum-type
brake for motorcycle utilizing magnetorheological fluid. Mechanics Based
Design of Structures and Machines, 2012. 40(2): 153162.
Lee, U., Kim, D., Hur, N. and Jeon, D. Design analysis and experimental
evaluation of an MR fluid clutch. Journal of Intelligent Material Systems
and Structures, 1999. 10(9): 701707.
Neelakantan, V. A. and Washington, G. N. Modeling and reduction
of centrifuging in magnetorheological (MR) transmission clutches for
automotive applications. Journal of Intelligent Material Systems and
Structures, 2005. 16(9): 703711.
Kavlicoglu, B., Gordaninejad, F., Evrensel, C., Fuchs, A. and Korol, G. A
semi-active, high-torque, magnetorheological fluid limited slip differential
clutch. Journal of Vibration and Acoustics, 2006. 128(5): 604.
Wang, D., Tian, Z., Meng, Q. and Hou, Y. Development of a novel two-layer
multiplate magnetorheological clutch for high-power applications. Smart
Materials and Structures, 2013. 22(8): 085018.
Dogruoz, M. B., Wang, E. L., Gordaninejad, F. and Stipanovic, A. J.
Augmenting heat transfer from fail-safe magneto-rheological fluid dampers
using fins. Journal of Intelligent Material Systems and Structures, 2003.
14(2): 7986.
Kim, H.-C., Oh, J.-S. and Choi, S.-B. The field-dependent shock profiles
of a magnetorhelogical damper due to high impact: an experimental
investigation. Smart Materials and Structures, 2015. 24(2): 025008.
Sodeyama, H., Suzuki, K. and Sunakoda, K. Development of large capacity
semi-active seismic damper using magneto-rheological fluid. Journal of
Pressure Vessel Technology, 2004. 126(1): 105.
Gordaninejad, F., Wang, X., Hitchcock, G., Bangrakulur, K., Ruan, S. and
Siino, M. Modular high-force seismic magneto-rheological fluid damper.
Journal of Structural Engineering, 2010. 136(2): 135143.
110
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
Liao, C. R., Zhao, D. X., Xie, L. and Liu, Q. A design methodology for a
magnetorheological fluid damper based on a multi-stage radial flow mode.
Dyke, S., Spencer, B., Sain, M. and Carlson, J. Modeling and control
of magnetorheological dampers for seismic response reduction. Smart
Milecki, A. Investigation and control of magnetorheological fluid dampers.
International Journal of Machine Tools and Manufacture, 2001. 41(3): 379
391.
Li, W. H., Du, H. and Guo, N. Q. Design and testing of an MR
steering damper for motorcycles. The International Journal of Advanced
Manufacturing Technology, 2003. 22(3-4): 288294.
Milecki, Andrzej and Sedziak, Dariusz and Ortmann, Jaroslaw and Hauke,
M. Controllability of MR shock absorber for vehicles. International Journal
of Vehicle Design, 2005. 38(2/3): 222233.
Yu, M., Liao, C. R., Chen, W. M. and Huang, S. L. Study on MR semiactive suspension system and its road testing. Journal of Intelligent Material
Systems and Structures, 2006. 17(8-9): 801806.
Batterbee, D. C., Sims, N. D., Stanway, R. and Rennison, M.
Magnetorheological landing gear: 2. Validation using experimental data.
Huang, J., He, J. and Zhang, J. Viscoplastic flow of the MR fluid in a
cylindrical valve. Key Engineering Materials, 2004. 274-276: 969974.
Sahin, H., Liu, Y., Wang, X., Gordaninejad, F., Evrensel, C. and Fuchs,
A. Full-scale magnetorheological fluid dampers for heavy vehicle rollover.
Journal of Intelligent Material Systems and Structures, 2007. 18(12): 1161
1167.
Guo, P., Guan, X. and Ou, J. Physical modeling and design method of the
hysteretic behavior of magnetorheological dampers. Journal of Intelligent
Material Systems and Structures, 2014. 25(6): 680696.
Nguyen, Q., Choi, S. and Wereley, N. Optimal design of magnetorheological
valves via a finite element method considering control energy and a time
constant. Smart Materials and Structures, 2008. 17(2): 025024.
Bai, X.-X., Wang, D.-H. and Fu, H. Principle, modeling, and testing of an
annular-radial-duct magnetorheological damper. Sensors and Actuators A:
Physical, 2013. 201: 302309.
Wang, D. H. and Liao, W. H. Magnetorheological fluid dampers: a review of
parametric modelling. Smart Materials and Structures, 2011. 20(2): 023001.
Yang, G., Spencer, B. F., Carlson, J. D. and Sain, M. K. Large-scale MR fluid
111
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
dampers: modeling and dynamic performance considerations. Engineering

Structures, 2002. 24(3): 309323.
Yang, G., Spencer, B. F., Jung, H.-J. and Carlson, J. D. Dynamic modeling
of large-scale magnetorheological damper systems for civil engineering
applications. Journal of Engineering Mechanics, 2004. 130(9): 11071114.
Wang, J., Ni, Y., Ko, J. and Spencer, B. Magneto-rheological tuned liquid
column dampers (MR-TLCDs) for vibration mitigation of tall buildings:
modelling and analysis of open-loop control. Computers & Structures, 2005.
83(25-26): 20232034.
Bitaraf, M., Ozbulut, O. E., Hurlebaus, S. and Barroso, L. Application of
semi-active control strategies for seismic protection of buildings with MR
dampers. Engineering Structures, 2010. 32(10): 30403047.
Tu, J. W., Liu, J., Qu, W. L., Zhou, Q., Cheng, H. B. and Cheng, X. D. Design
and fabrication of 500-kN large-scale MR damper. Journal of Intelligent
Weber, F. BoucWen model-based real-time force tracking scheme for MR
dampers. Smart Materials and Structures, 2013. 22(4): 045012.
Carlson, J. D., Matthis, W. and Toscano, J. R. Smart prosthetics based on
magnetorheological fluids. Proceedings of SPIE. SPIE. 2001, vol. 4332.
308316.
Li, W. and Du, H. Development of an ankle physiotherapy device using
an MR damper. The International Journal of Advanced Manufacturing
Technology, 2004. 25(3-4): 205213.
Gudmundsson, K. H., Jonsdottir, F. and Thorsteinsson, F. A geometrical
optimization of a magneto-rheological rotary brake in a prosthetic knee.
Chen, J. Z. and Liao, W. H.
Design, testing and control of a
magnetorheological actuator for assistive knee braces. Smart Materials and
Structures, 2010. 19(3): 035029.
Wang, D. H. and Liao, W. H. Semi-active suspension systems for railway
vehicles using magnetorheological dampers. Part I: system integration and
modelling. Vehicle System Dynamics, 2009. 47(11): 13051325.
Choi, Y.-T. and Wereley, N. M. Vibration control of a landing gear system
featuring electrorheological/magnetorheological fluids. Journal of Aircraft,
2003. 40(3): 432439.
Hu, W., Wereley, N. M., Chemouni, L. and Chen, P. Semi-active linear
stroke magnetorheological fluid-elastic helicopter lag damper. Journal of
Guidance, Control, and Dynamics, 2007. 30(2): 565575.
Dong, X. M. and Xiong, G. W. Vibration attenuation of magnetorheological
112
[89]
[90]
[91]
[92]
[93]
[94]
[95]
[96]
[97]
[98]
[99]
[100]
[101]
[102]
landing gear system with human simulated intelligent control. Mathematical

Problems in Engineering, 2013. 2013: 113.
Li, Z. C. and Wang, J. A gun recoil system employing a magnetorheological
fluid damper. Smart Materials and Structures, 2012. 21(10): 105003.
Kozlowska, J. and Leonowicz, M.
Processing and properties of
magnetorheological fluids for prospective application in a passive armour.
IEEE Transactions on Magnetics, 2013. 49(8): 47214724.
Singh, H. J. and Wereley, N. M. Optimal control of gun recoil in direct fire
using magnetorheological absorbers. Smart Materials and Structures, 2014.
23(5): 055009.
de Vicente, J., Klingenberg, D. J. and Hidalgo-Alvarez, R. Magnetorheological fluids: a review. Soft Matter, 2011. 7(8): 3701.
Genc, S. and Phule, P. P. Rheological properties of magnetorheological
fluids. Smart Materials and Structures, 2002. 11(1): 140146.
Ginder, J. M. and Davis, L. C. Shear stresses in magnetorheological fluids:
Role of magnetic saturation. Applied Physics Letters, 1994. 65(26): 3410.
Ierardi, R. F. and Bombard, A. J. F. Off-state viscosity and yield
stress optimization of magneto-rheological fluids: A mixture design of
experiments approach. Journal of Physics: Conference Series, 2009. 149:
012037.
Chiriac, H. and Stoian, G. Influence of particle size distributions on
magnetorheological fluid performances. Journal of Physics: Conference
Series, 2010. 200(7): 072095.
Rodrguez-Lopez, J., Shum, H. C., Elvira, L., Montero de Espinosa, F. and
Weitz, D. A. Fabrication and manipulation of polymeric magnetic particles
with magnetorheological fluid. Journal of Magnetism and Magnetic
Materials, 2013. 326: 220224.
Skjeltorp, A. One and two-dimensional crystallization of magnetic holes.
Physical Review Letters, 1983. 51(25): 23062309.
Samouhos, S. and McKinley, G. Carbon nanotubemagnetite composites,
with applications to developing unique magnetorheological fluids. Journal
of Fluids Engineering, 2007. 129(4): 429.
Rosenfeld, N., Wereley, N. M., Radakrishnan, R. and Sudarshan,
T. S. Behavior of magnetorheological fluids utilizing nanopowder iron.
International Journal of Modern Physics B, 2002. 16(17-18): 23922398.
Bell, R. C., Miller, E. D., Karli, J. O., Vavreck, A. N. and Zimmerman, D. T.
Influence of particle shape in the properties of magnetorheological fluids.
International Journal of Modern Physics B, 2007. 21(28 & 29): 50185025.
Li, F. and Tao, C. Research on magneto-rheological technology and its
113
[103]
[104]
[105]
[106]
[107]
[108]
[109]
[110]
[111]
[112]
[113]
[114]
application. 2011 Chinese Control and Decision Conference (CCDC), 2011:

40724076.
Xu, Y., Gong, X., Xuan, S., Zhang, W. and Fan, Y. A high-performance
magnetorheological material: preparation, characterization and magneticmechanic coupling properties. Soft Matter, 2011. 7(11): 5246.
Upadhyay, R. V., Laherisheth, Z. and Shah, K. Rheological properties of
soft magnetic flake shaped iron particle based magnetorheological fluid in
dynamic mode. Smart Materials and Structures, 2014. 23(1): 015002.
Shah, K., Xuan Phu, D. and Choi, S.-B. Rheological properties of bidispersed magnetorheological fluids based on plate-like iron particles with
application to a small-sized damper. Journal of Applied Physics, 2014.
115(20): 203907.
Olabi, A. G. and Grunwald, A. Design and application of magnetorheological fluid. Materials & Design, 2007. 28(10): 26582664.
Mazlan, S. A., Ismail, I., Zamzuri, H. and Fatah, A. Y. A. Compressive and
tensile stress of magnetorheological fluids in squeeze mode. International
Journal of Applied Electromagnetics and Mechanics, 2011. 36: 327337.
Goncalves, F. D. and Carlson, J. D. An alternate operation mode for MR
fluidsmagnetic gradient pinch. Journal of Physics: Conference Series, 2009.
149: 012050.
Hong, S., Wereley, N., Choi, Y. and Choi, S.
Analytical and
experimental validation of a nondimensional Bingham model for mixedmode magnetorheological dampers. Journal of Sound and Vibration, 2008.
312(3): 399417.
El Wahed, A. K. and Mcewan, C. A. Design and performance evaluation
of magnetorheological fluids under single and mixed modes. Journal of
Yazid, I., Mazlan, S. A., Zamzuri, H., Mughni, M. and Chuprat, S.
Parameters consideration in designing a magnetorheological damper. Key
Engineering Materials, 2013. 543: 487490.
Yang, C., Fu, J., Yu, M., Zheng, X. and Ju, B. A new magnetorheological
elastomer isolator in shearcompression mixed mode. Journal of Intelligent
Yazid, I. I. M., Mazlan, S. A., Kikuchi, T., Zamzuri, H. and Imaduddin,
F. Design of magnetorheological damper with a combination of shear and
squeeze modes. Materials & Design, 2014. 54: 8795.
Mughni, M. J., Mazlan, S. A., Zamzuri, H., Yazid, I. I. M. and Rahman,
M. A. A. Simulation study of magnetorheological testing cell design by
incorporating all basic operating modes. Smart Structures and Systems,
114
[115]
[116]
[117]
[118]
[119]
[120]
[121]
[122]
[123]
[124]
[125]
[126]
[127]
2014. 14(5): 901916.

Mazlan, S. A., Ekreem, N. B. and Olabi, A. G. The performance of
magnetorheological fluid in squeeze mode. Smart Materials and Structures,
2007. 16(5): 16781682.
Mazlan, S. A., Issa, A., Chowdhury, H. A. and Olabi, A. G. Magnetic circuit
design for the squeeze mode experiments on magnetorheological fluids.
Materials & Design, 2009. 30(6): 19851993.
Farjoud, A., Ahmadian, M., Mahmoodi, N., Zhang, X. and Craft, M.
Nonlinear modeling and testing of magneto-rheological fluids in low shear
rate squeezing flows. Smart Materials and Structures, 2011. 20(8): 085013.
Wang, H., Bi, C., Kan, J., Gao, C. and Xiao, W. The mechanical property
of magnetorheological fluid under compression, elongation, and shearing.
816.
Aydar, G., Wang, X. and Gordaninejad, F. A novel two-way-controllable
magneto-rheological fluid damper. Smart Materials and Structures, 2010.
19(6): 065024.
Bose, H. and Ehrlich, J. Performance of magnetorheological fluids in a novel
damper with excellent fail-safe behavior. Journal of Intelligent Material
Kostamo, E., Kostamo, J., Kajaste, J. and Pietola, M. Magnetorheological
valve in servo applications. Journal of Intelligent Material Systems and
Structures, 2012. 23(9): 10011010.
Hu, G., Long, M., Yu, L. and Li, W. Design and performance evaluation
of a novel magnetorheological valve with a tunable resistance gap. Smart
Cook, E., Hu, W. and Wereley, N. M. Magnetorheological bypass damper
exploiting flow through a porous channel. Journal of Intelligent Material
Sahin, H., Wang, X. and Gordaninejad, F. Magneto-rheological fluid flow
through complex valve geometries. International Journal of Vehicle Design,
2013. 63(2/3): 241255.
Snyder, R. A., Kamath, G. M. and Wereley, N. M. Characterization and
analysis of magnetorheological damper behavior under sinusoidal loading.
AIAA Journal, 2001. 39(7): 12401253.
Spaggiari, A. and Dragoni, E.
Efficient dynamic modelling and
characterization of a magnetorheological damper. Meccanica, 2012. 47(8):
20412054.
Domnguez-Gonzlez, A., Stiharu, I. and Sedaghati, R. Practical hysteresis
115
[128]
[129]
[130]
[131]
[132]
[133]
[134]
[135]
[136]
[137]
[138]
model for magnetorheological dampers. Journal of Intelligent Material

Case, D., Taheri, B. and Richer, E. Dynamical modeling and experimental
study of a small-scale magnetorheological damper.
IEEE/ASME
Transactions on Mechatronics, 2014. 19(3): 10151024.
Balamurugan, L., Jancirani, J. and Eltantawie, M. A.
Generalized
magnetorheological (MR) damper model and its application in semi-active
control of vehicle suspension system. International Journal of Automotive
Technology, 2014. 15(3): 419427.
Susan-Resiga, D. A rheological model for magneto-rheological fluids.
1010.
Wang, X. and Gordaninejad, F. Flow analysis of field-controllable, electroand magneto-rheological fluids using Herschel-Bulkley model. Journal of
Li, W., Du, H. and Guo, N. Finite element analysis and simulation evaluation
of a magnetorheological valve. The International Journal of Advanced
Manufacturing Technology, 2003. 21(6): 438445.
Wereley, N. M. and Pang, L. Nondimensional analysis of semi-active
electrorheological and magnetorheological dampers using approximate
parallel plate models. Smart Materials and Structures, 1998. 7(5): 732743.
Yoshida, K., Soga, T., Kawachi, M., Edamura, K. and Yokota, S. Magnetorheological valve-integrated cylinder and its application. Proceedings of the
Institution of Mechanical Engineers, Part I: Journal of Systems and Control
Engineering, 2010. 224(1): 3140.
Kwok, N., Ha, Q., Nguyen, T., Li, J. and Samali, B. A novel hysteretic
model for magnetorheological fluid dampers and parameter identification
using particle swarm optimization. Sensors and Actuators A: Physical,
2006. 132(2): 441451.
Kwok, N. M., Ha, Q. P., Nguyen, M. T., Li, J. and Samali, B. Bouc-Wen
model parameter identification for a MR fluid damper using computationally
efficient GA. ISA transactions, 2007. 46(2): 16779.
Sakai, C., Ohmori, H. and Sano, A. Modeling of MR damper with hysteresis
for adaptive vibration control. Proceedings of 42nd IEEE International
Conference on Decision and Control. Maui, Hawaii: IEEE. 2003, December.
ISBN 0-7803-7924-1. 38403845.
Jimenez, R. and Avarez-Icaza, L.
LuGre friction model for a
magnetorheological damper. Structural Control and Health Monitoring,
2005. 12(1): 91116.
116
[139]
[140]
[141]
[142]
[143]
[144]
[145]
[146]
[147]
[148]
[149]
[150]
[151]
[152]
Ikhouane, F. and Rodellar, J. On the hysteretic BoucWen model part I:

forced limit cycle characterization. Nonlinear Dynamics, 2005. 42(1): 63
78.
Ikhouane, F. and Rodellar, J. On the hysteretic BoucWen model part II:
robust parametric identification. Nonlinear Dynamics, 2005. 42(1): 7995.
Ikhouane, F., Hurtado, J. E. and Rodellar, J. Variation of the hysteresis loop
with the BoucWen model parameters. Nonlinear Dynamics, 2006. 48(4):
361380.
Ikhouane, F. and Dyke, S. J. Modeling and identification of a shear mode
magnetorheological damper. Smart Materials and Structures, 2007. 16(3):
605616.
Rochdi, Y., Giri, F., Ikhouane, F., Chaoui, F. Z. and Rodellar, J. Parametric
identification of nonlinear hysteretic systems. Nonlinear Dynamics, 2009.
58(1-2): 393404.
Gavin, H. P. Multi-duct ER dampers. Journal of Intelligent Material Systems
and Structures, 2001. 12(5): 353366.
Jiang, Z. and Christenson, R. E. A fully dynamic magneto-rheological fluid
damper model. Smart Materials and Structures, 2012. 21(6): 065002.
Sahin, I., Engin, T. and Cesmeci, S. Comparison of some existing
parametric models for magnetorheological fluid dampers. Smart Materials
and Structures, 2010. 19(3): 035012.
Choi, S.-B., Lee, S.-K. and Park, Y.-P. A hysteresis model for the fielddependent damping force of a magnetorheological damper. Journal of
Sound and Vibration, 2001. 245(2): 375383.
Ubaidillah, Hudha, K. and Kadir, F. A. Modelling , characterisation and
force tracking control of a magnetorheological damper under harmonic
excitation. International Journal of Modelling, Identification and Control,
2011. 13(1/2): 921.
Du, H. P., Sze, K. Y. and Lam, J. Semi-active H-infinity control of
vehicle suspension with magneto-rheological dampers. Journal of Sound
and Vibration, 2005. 283(3-5): 981996.
Xia, P.-Q. An inverse model of MR damper using optimal neural network
and system identification. Journal of Sound and Vibration, 2003. 266(5):
10091023.
Wang, D. H. and Liao, W. H. Modeling and control of magnetorheological
fluid dampers using neural networks. Smart Materials and Structures, 2005.
14(1): 111126.
Tudon-Martnez, J. C., Lozoya-Santos, J. J., Morales-Menendez, R. and
Ramirez-Mendoza, R. A. An experimental artificial-neural-network-based
117
[153]
[154]
[155]
[156]
[157]
[158]
[159]
[160]
[161]
[162]
[163]
[164]
[165]
modeling of magneto-rheological fluid dampers. Smart Materials and

Structures, 2012. 21(8): 085007.
Boada, M., Calvo, J., Boada, B. and Daz, V.
Modeling of a
magnetorheological damper by recursive lazy learning. International
Journal of Non-Linear Mechanics, 2011. 46(3): 479485.
Zeinali, M., Mazlan, S. A., Abd Fatah, A. Y. and Zamzuri, H. A
phenomenological dynamic model of a magnetorheological damper using a
neuro-fuzzy system. Smart Materials and Structures, 2013. 22(12): 125013.
Chen, C. and Liao, W.-H. A self-powered, self-sensing magnetorheological
damper. 2010 IEEE International Conference on Mechatronics and
Automation. IEEE. 2010. ISBN 978-1-4244-5140-1. 13641369.
Facey, W. B., Rosenfeld, N. C., Choi, Y.-T., Wereley, N. M., Choi, S. B.
and Chen, P. Design and testing of a compact magnetorheological Damper
for high impulsive loads. International Journal of Modern Physics B, 2005.
19(7-8-9): 15491555.
Chen, C. and Liao, W. H. A self-sensing magnetorheological damper with
power generation. Smart Materials and Structures, 2012. 21(2): 025014.
Mao, M., Hu, W., Choi, Y.-T. and Wereley, N. M. A magnetorheological
damper with bifold valves for shock and vibration mitigation. Journal of
Zhang, J. Q., Feng, Z. Z. and Jing, Q. Optimization analysis of a new vane
MRF damper. Journal of Physics: Conference Series, 2009. 149: 012087.
Yang, L., Chen, S. Z., Zhang, B. and Feng, Z. Z.
A rotary
magnetorheological damper for a tracked vehicle. Advanced Materials
Research, 2011. 328-330: 11351138.
Nguyen, Q., Han, Y., Choi, S. and Wereley, N. Geometry optimization of
MR valves constrained in a specific volume using the finite element method.
Choi, Y.-T. and Wereley, N. M. Comparative analysis of the time response of
electrorheological and magnetorheological dampers using nondimensional
parameters. Journal of Intelligent Material Systems and Structures, 2002.
13(7-8): 443451.
Rosenfeld, N. C. and Wereley, N. M. Volume-constrained optimization
of magnetorheological and electrorheological valves and dampers. Smart
Hong, S., Choi, S., Choi, Y. and Wereley, N. Non-dimensional analysis and
design of a magnetorheological damper. Journal of Sound and Vibration,
2005. 288(4-5): 847863.
Sodeyama, H., Sunakoda, K., Fujitani, H., Soda, S., Iwata, N. and Hata,
118
[166]
[167]
[168]
[169]
[170]
[171]
[172]
[173]
[174]
[175]
[176]
[177]
[178]
K. Dynamic Tests and Simulation of Magneto-Rheological Dampers.

Computer-Aided Civil and Infrastructure Engineering, 2003. 18(1): 4557.
Fujitani, H., Sodeyama, H., Tomura, T., Hiwatashi, T., Shiozaki, Y., Hata,
K., Sunakoda, K., Morishita, S. and Soda, S. Development of 400kN
magnetorheological damper for a real base-isolated building. Proc. SPIE,
2003. 5052: 265276.
Lord Corp. Lord technical data: MRF-132DG magneto-rheological fluid,
2011.
Lord Corp. Lord product selector guide: Lord magneto-rheological fluids,
2008.
Tan, K., Stanway, R. and Bullough, W. Braking responses of inertia/load by
using an electro-rheological (ER) brake. Mechatronics, 2007. 17(6): 277
289.
Salloom, M. Y. and Samad, Z. Experimental test of magneto-rheological
directional control valve. Advanced Materials Research, 2011. 383-390:
54095413.
Ismail, I., Mazlan, S. A., Zamzuri, H. and Olabi, A. G. Fluid particle
separation of magnetorheological fluid in squeeze mode. Japanese Journal
of Applied Physics, 2012. 51: 067301.
Li, W. H., Yao, G. Z., Chen, G., Yeo, S. H. and Yap, F. F. Testing and steady
state modeling of a linear MR damper under sinusoidal loading. Smart
Yang, G. Large-scale magnetorheological fluid damper for vibration
mitigation: modeling, testing and control. Phd dissertation. University of
Notre Dame. 2001.
Zhang, H. H., Liao, C. R., Yu, M. and Huang, S. L. A study of an inner
bypass magneto-rheological damper with magnetic bias. Smart Materials
and Structures, 2007. 16(5): N40N46.
Yun, Y.-W., Lee, S.-M. and Park, M.-K. A study on the efficiency
improvement of a passive oil damper using an MR accumulator. Journal
of Mechanical Science and Technology, 2010. 24(11): 22972305.
Canudas de Wit, C., Olsson, H., Astrom, K. and Lischinsky, P. A new
model for control of systems with friction. IEEE Transactions on Automatic
Control, 1995. 40(3): 419425.
Lischinsky, P., Canudas-de Wit, C. and Morel, G. Friction compensation for
an industrial hydraulic robot. IEEE Control Systems Magazine, 1999. 19(1):
2532.
strom, K. and Hagglund, T. The future of PID control. Control
A
Engineering Practice, 2001. 9(11): 11631175.
119
[179]
[180]
[181]
Crawford, N. M., Cunningham, G. and Spedding, P. L. Prediction of

pressure drop for turbulent fluid flow in 90 bends. Proceedings of the
Institution of Mechanical Engineers, Part E: Journal of Process Mechanical
Engineering, 2003. 217(3): 153155.
Spedding, P., Benard, E. and Mcnally, G. Fluid Flow through 90 Degree
Bends. Developments in Chemical Engineering and Mineral Processing,
2004. 12(1-2): 107128.
Etemad, S. G. Turbulent flow friction loss coefficients of fittings for purely
viscous non-newtonian fluids. International Communications in Heat and
Mass Transfer, 2004. 31(5): 763771.
APPENDIX A
CAD DRAWINGS
Figure A.1 CAD drawing of coil bobbin
ISOMETRIC VIEW
LEFT VIEW
4X M4X0.7
1.09
Q.A
MFG
APPV'D
CHK'D
DRAWN
FI
NAME
38
FINISH:
DATE
WEIGHT:
MATERIAL:
SECTION B-B
SIGNATURE
UNLESS OTHERWISE SPECIFIED:

DIMENSIONS ARE IN MILLIMETERS
SURFACE FINISH:
TOLERANCES:
LINEAR: 0.01
ANGULAR: 0.01
16 + 0.02
22.64
16.98
3.5
O-ring Groove
DEBUR AND
BREAK SHARP
EDGES
10.0
15
SCALE:1:1
DWG NO.
TITLE:
REVISION
Coil Bobbin
SHEET 2 OF 2
A2
A4
3.5
40 - 0.02
4X M4X0.7
RIGHT VIEW
DO NOT SCALE DRAWING
16 + 0.02 THRU ALL
Alumunium
10
121
Figure A.2 CAD drawing of valve casing version 1
12
TOP VIEW
ISOMETRIC VIEW
4X M4X0.7 THRU ALL
3/8 inch
6 + 0.02 THRU ALL
15
12
Q.A
MFG
APPV'D
CHK'D
DRAWN
FI
NAME
DATE
WEIGHT:
MATERIAL:
SECTION A-A
FINISH:
SIGNATURE

SURFACE FINISH:
TOLERANCES:
LINEAR: 0.01
ANGULAR: 0.01
15
50
DEBUR AND
BREAK SHARP
EDGES
24
5 - 0.02
19 - 0.02
Mild Steel 1010
SCALE:1:1
DWG NO.
TITLE:
16 - 0.02
REVISION
SHEET 1 OF 2
A1
50
40 + 0.02
Valve Casing
BOTTOM VIEW
A4
122
49
39
50
123
22
A
50
15
6+
12.5
READ
11.5
0.0
EMA
LE TH
24
1/4 F
SPP
11 (B
10
44
40 + 0.1
14 + 0.02

SURFACE FINISH:
TOLERANCES:
LINEAR:
ANGULAR:
NAME
DRAWN
FI
DEBUR AND
BREAK SHARP
EDGES
FINISH:
SIGNATURE
16 - 0.02
CUT-FACE A-A
6 + 0.02
11.8
5 - 0.02
DATE
REVISION
TITLE:
Valve Casing Female
29/4/13
CHK'D
APPV'D
MFG
Q.A
MATERIAL:
WEIGHT:
Casing2_rev_ver2
DWG NO.
Mild Steel 1010
SCALE:1:1
SHEET 1 OF 1
Figure A.3 CAD drawing of valve casing version 2 - female
A4
49
124
50
22
A
46
50
50
.02
6+0
15
23
E THR
EAD
12

SURFACE FINISH:
TOLERANCES:
LINEAR:
ANGULAR:
NAME
FI
EMA
L
11.8
DEBUR AND
BREAK SHARP
EDGES
FINISH:
SIGNATURE
16 - 0.02
CUT-FACE A-A
6 + 0.02
5 - 0.02
DRAWN
1/4 F
11 (B
10
SPP
40 + 0.1
25 + 0.02
DATE
REVISION
TITLE:
Valve Casing Male
29/4/13
CHK'D
APPV'D
MFG
Q.A
MATERIAL:
DWG NO.
MILD STEEL 1010

WEIGHT:
Casing2_rev_ver2_2
SCALE:1:1
SHEET 1 OF 1
Figure A.4 CAD drawing of valve casing version 2 - male
A4
Figure A.5 CAD drawing of spacer - 0.5 mm
20
TOP VIEW
ISOMETRIC VIEW
70
Q.A
MFG
APPV'D
CHK'D
DRAWN
FI
NAME
FINISH:
DATE
WEIGHT:
MATERIAL:
SECTION A-A
SIGNATURE

SURFACE FINISH:
TOLERANCES:
LINEAR: 0.05
ANGULAR: 0.05
1 + 0.03
2.5 - 0.02
3 - 0.02
R6.5 + 0.03
R5.5
DEBUR AND
BREAK SHARP
EDGES
Alumunium
SCALE:3:1
DWG NO.
TITLE:
B1
REVISION
SHEET 1 OF 2
0.5
16 - 0.02
Bush 0.5mm
BOTTOM VIEW
A4
125
Figure A.6 CAD drawing of spacer - 1 mm
TOP VIEW
ISOMETRIC VIEW
70
20
Q.A
MFG
APPV'D
CHK'D
DRAWN
FI
NAME
FINISH:
DATE
WEIGHT:
MATERIAL:
SECTION B-B
SIGNATURE

SURFACE FINISH:
TOLERANCES:
LINEAR: 0.05
ANGULAR: 0.05
1 + 0.03
2 - 0.02
3 - 0.02
R6.5 + 0.03
R5.5
DEBUR AND
BREAK SHARP
EDGES
Alumunium
SCALE:3:1
DWG NO.
TITLE:
0.5
REVISION
SHEET 2 OF 2
B2
16 - 0.02
Bush 1mm
BOTTOM VIEW
A4
126
Figure A.7 CAD drawing of side core
ISOMETRIC VIEW
Q.A
MFG
APPV'D
CHK'D
DRAWN
FI
NAME
FINISH:
DATE
15.5
SIGNATURE

SURFACE FINISH:
TOLERANCES:
LINEAR: 0.05
ANGULAR: 0.05
5 - 0.02
WEIGHT:
MATERIAL:
DEBUR AND
BREAK SHARP
EDGES
10.5
Mild Steel 1010
SCALE:3:1
DWG NO.
TITLE:
REVISION
SHEET 1 OF 4
C1
5 - 0.02
13 - 0.02
Side Core 1
TOP VIEW
A4
127
Figure A.8 CAD drawing of orifice core
Q.A
MFG
APPV'D
CHK'D
DRAWN
FI
NAME
FINISH:
SIGNATURE

SURFACE FINISH:
TOLERANCES:
LINEAR:
ANGULAR:
5 - 0.02
DATE
WEIGHT:
MATERIAL:
DEBUR AND
BREAK SHARP
EDGES
Mild Steel 1010
SCALE:3:1
DWG NO.
TITLE:
REVISION
SHEET 1 OF 2
D1
5 THRU ALL
16 - 0.02
Orifice Core
A4
128
Figure A.9 CAD drawing of center core
Q.A
MFG
APPV'D
CHK'D
DRAWN
FI
NAME
FINISH:
SIGNATURE

SURFACE FINISH:
TOLERANCES:
LINEAR:
ANGULAR:
4 - 0.02
5 - 0.02
DATE
WEIGHT:
MATERIAL:
DEBUR AND
BREAK SHARP
EDGES
Mild Steel 1010
SCALE:3:1
DWG NO.
TITLE:
REVISION
13 - 0.02
13 - 0.02
SHEET 2 OF 2
D2
Middle Core
A4
129

Thesis Fitrian Imaduddin

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Thesis Fitrian Imaduddin

Încărcat de

Drepturi de autor:

Formate disponibile

A NOVEL MAGNETORHEOLOGICAL VALVE WITH MEANDERING FLOW

UNIVERSITI TEKNOLOGI MALAYSIA

We hereby declare that we have read this thesis and in our

Associate Professor Ir. Dr. Saiful Amri Mazlan

Dr. Hairi Zamzuri

A NOVEL MAGNETORHEOLOGICAL VALVE WITH MEANDERING FLOW

A thesis submitted in fulfilment of the

Malaysia-Japan International Institute of Technology

To my father, my mother, my wife and my brothers

Lastly, I would like to express my sincerest gratitude to my parents, Dr. Ahmad

The development of a new Magnetorheological (MR) valve with meandering

Pembangunan konsep baru injap reologi magnet (MR) dengan menggunakan

Combination of Annular and Radial

MAGNETORHEOLOGICAL VALVE CONCEPT

HYSTERESIS MODELING OF MAGNETORHEOLOGICAL VALVE

CONCLUSIONS AND RECOMMENDATIONS

MR based bellow-driven motion control

Comparison of measured and theoretical off-state peak

Magnetorheological (MR) fluid is one of the fluids in the class of field

Considering the importance of MR valve, many designs of MR valve have

The practice in the other type of MR damper, known as the bypass MR

Despite the advantages and disadvantages of each MR damper structure,

From these explanations, it can be observed that the technological advancement

This study embarks on the following objectives:

To develop a new concept of MR valve with meandering flow path to improve

To assess the performance of MR valve using experimental work.

To model the hysteretic behavior of the MR valve.

In this research, a new concept of MR valve will be investigated. This study

The significance of this research is mainly in terms of general advancement

This study demonstrates a new concept of MR valve using a meandering flow

The hysteretic modeling process of the MR valve introduces a new modeling

This thesis is organized in six chapters. Each respective chapter in this

Chapter 2 covers the theoretical background, which includes the properties

Chapter 4 elaborates the experimental evaluation of the MR valve including the

Chapter 5 presents the development of two different hysteresis MR valve

This chapter provides an overview of magnetorheological (MR) fluids, the

Composition of Magnetorheological Fluid

MR fluids are commonly referred to a two-phase fluid suspension made from

However, despite of several improvement efforts to the MR fluid composition,

Operational Modes of Magnetorheological Fluid

Figure 2.2 Shear mode

Figure 2.3 Valve mode

Figure 2.4 Squeeze mode

Figure 2.5 Magnetic Gradient Pinch mode

MR valve as one of the application branch of MR fluid typically consist of valve

Annular Magnetorheological Valve

One of the earliest designs of stand-alone MR valve was proposed in Kordonski

Figure 2.6 MR throttle valve [17]

Figure 2.7 C-shaped pressure control valve [18]

Figure 2.8 Three port MR valve [19]

Another group of researchers have focused to pursue an improvement of MR

Figure 2.9 Double-coil annular MR valve [20]

Radial Magnetorheological Valve

Although the advancement of annular MR valve were continuously explored,

The benefit of radial MR valve is introduced by Wang et al. [21] in the

The work on radial MR valve was extended by the introduction of two-way

Figure 2.10 Basic structure of single stage radial MR valve [21]

Figure 2.11 Two-way controllable radial MR valve [119]

Combination of Annular and Radial Magnetorheological Valve

The successful performance improvement on radial MR valve approach has

Figure 2.12 Annular-Radial MR Valve [24]

According to the study by Nguyen et al. [72] on the optimization of various

Experimental Assessment Method for Magnetorheological Valve

The experimental assessment of the MR valve normally aimed to measure the